Textbook of Biochemistry With Clinical Correlations, Devlin, 4th Ed.

February 2, 2017 | Author: Neraph Jarito | Category: N/A
Share Embed Donate


Short Description

Download Textbook of Biochemistry With Clinical Correlations, Devlin, 4th Ed....

Description

Textbook of Biochemistry with Clinical Correlations Fourth Edition

 

  

Abberviations in Biochemistry A (or Ade)

adenine

ACP

acyl carrier protein

ACTH

adrenocorticotropic hormone

acyl coA

acyl derivative of CoA

ADH

antidiuretic hormone

AdoMet

adenosylmethionine

Ala

alanine

ALA

aminolevulinic acid

AMP

adenosine monophosphate

cAMP

cyclic AMP

Arg

arginine

Asn

asparagine

Asp

aspartate

ATP

adenosine triphosphate

ATPase

adenosine triphosphatase

BMR

basal metabolic rate

BPG

D­2,3 hisphosphoglycerate

C (or Cyt)

cytosine

CDP

cytidine diphosphate

CMP

cytidine monophosphate

CTP

cytidine triphosphate

CoA or CoASH

coenzyme A

CoQ

coenzyme Q (ubiquinone)

cyclic AMP

adenosine 3 ,5 ­cyclic monophosphate

cyclic GMP

xuanosine 3 ,5 ­cyclic monophosphate

Cys

cysteine

d

2 ­deoxyriho

DNA

deoxyribonucleic acid

cDNA

complementary DNA

dopa

3,4­dihydroxyphenylalanine

EcoR1

EcoR1 restriction endonuclease

FAD

flavin adenine dinucleotide (oxidized form)

FADH2

flavin adenine dinucleotide (reduced form)

fMet

formylmethionine

FMN

flavin mononucleotide (oxidized form)

FMNH2

flavin mononucleotide (reduced form)

Fp

flavoprotein

G (or Gua)

guanine

GABA

g­aminobutyric acid

Gal

galactose

Glc

glucose

Gln

glutamine

Glu

glutamate

Gly

glycine

GDP

guanosine diphosphate

GMP

guanosine monophosphate

GTP

guanosine triphosphate

GSH

glutathione

Hb

hemoglobin

HbCO

carbon monoxide hemoglobin

HbO2

oxyhemoglobin

HDL

high density lipoprotein

HMG CoA

b ­hydroxy­ b ­methylglutaryl CoA

Hyp

hydroxyproline

IDL

intermediate density lipoprotein

IgG

immunoglobulin G

Ile

isoleucine

IP3

inositol 1,4,5 trisphosphate

ITP

inosine triphosphate

Km

Michaelis–Menten constant

kb

kilo base pair

LDL

low density lipoprotein

Leu

leucine

Lys

lysine

Mb

myoglobin

MbO2

oxymyoglobin

Met

methionine

MetHb

methemoglobin

NAD+

nicotinamide adenine dinucleotide (oxidized form)

NADH

nicotinamide adenine dinucleotide (reduced form)

NADP+

nicotinamide adenine dinucleotide phosphate (oxidized form)

NADPH

nicotinamide adenine dinucleotide phosphate (reduced form)

NANA

N­acetylneuraminic acid

PEP

phosphoenolpyruvate

Phe

phenylalanine

Pi

inorganic orthophosphate

PG

prostaglandin

PPi

inorganic pyrophosphate

Pro

proline

PRPP

phosphoribosylpyrophosphate

Q

ubiquinone (CoQ)

RNA

ribonucleic acid

mRNA

messenger RNA

rRNA

ribosomal RNA

tRNA

transfer RNA

RNase

ribonuclease

RQ

respiratory quotient (CO2 production/O2 consumption)

S

Svedberg unit

SAM

S­adenosylmethionine

Ser

serine

SH

sulfhydryl

T (or Thy)

thymine

TCA

Tricarhoxylic acid cycle (Krebs cycle)

TG

triacylglycerol

THF

tetrahydrofolic acid

Thr

threonine

TPP

thiamine pyrophosphate

Trp

tryptophan

TTP

thymidine triphosphate

Tyr

tyrosine

U (or Ura)

uracil

UDP

uridine diphosphate

UDP­galactose

uridine diphosphate galactose

UDP­glucose

uridine diphosphate glucose

UMP

uridine monophosphate

UTP

uridine triphosphate

Val

valine

VLDL

very low density lipoprotein

   

  

Page iii

Textbook of Biochemistry with Clinical Correlations:  Fourth Edition Edited by  Thomas M. Devlin, Ph.D. Professor Emeritus  Department of Biochemistry  MCP∙Hahnemann School of Medicine  Allegheny University of the Health Sciences  Philadelphia, Pennsylvania

 

  

Page iv

Address All Inquiries to the Publisher  Wiley­Liss, Inc., 605 Third Avenue, New York, NY 10158­0012 Copyright © 1997 Wiley­Liss, Inc. Printed in the United States of America. This text is printed on acid­free paper. Under the conditions stated below the owner of copyright for this book hereby grants permission to users to make photocopy reproductions of any part or all of its  contents for personal or internal organizational use, or for personal or internal use of specific clients. This consent is given on the condition that the copier pay the stated  per­copy fee through the Copyright Clearance Center, Incorporated, 27 Congress Street, Salem, MA 01970, as listed in the most current issue of "Permissions to  Photocopy" (Publisher's Fee list, distributed by CCC, Inc.), for copying beyond that permitted by sections 107 or 108 of the US Copyright Law. This consent does  not extend to other kinds of copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Cover Illustration: An artist's conception of the initiation of the DNA transcription mechanism catalyzed by RNA polymerase and involving protein transcription  factors. Subject Editor: Stephanie Diment  Design: Laura Ierardi  Senior Managing Editor: John Sollami  Marketing Managers: David Stier and David Steltenkamp Manufacturing Manager: Rick Mumma  Illustration Coordinator: Barbara Kennedy  Illustrations and Cover: Page Two This book was set in ITC Garamond Light by Bi­Comp Incorporated, and was printed and bound by Von Hoffmann Press. Library of Congress Cataloging­in­Publication Data Textbook of biochemistry: with clinical correlations/edited by  Thomas M. Devlin — 4th ed.  p.  cm.  Includes bibliographical references and index.  ISBN 0­471­15451­2  1. Biochemistry.  2. Clinical biochemistry.  I. Devlin, Thomas M.  [DNLM: 1. Biochemistry.  QU 4 T355 1997]  QP514.2.T4  1997                                                                                            97­1078 612'.015—dc21                                                                                                      CIP 10 9 8 7 6 5 4 3

 

  

Page v

To  Katie, Matthew, Ryan, and Laura

 

  

Page vii

Contributors Stelios Aktipis, Ph.D.  Professor  Department of Molecular and Cellular Biochemistry Stritch School of Medicine  Loyola University of Chicago  2160 S. First Avenue  Maywood, IL 60153 Carol N. Angstadt, Ph.D.  Professor  Department of Biomedical Sciences, M.S.# 456 Allegheny University of the Health Sciences  Broad and Vine Streets  Philadelphia, PA 19102­1192 email: angstadt@allegheny William Awad, JR., M.D., Ph.D.  Professor  Departments of Medicine and of Biochemistry University of Miami School of Medicine  P.O. Box 016960  Miami, FL 33101 email: [email protected] James Baggott, Ph.D.  Associate Professor  Department of Biochemistry  MCP∙Hahnemann School of Medicine  Allegheny University of the Health Sciences 2900 Queen Lane  Philadelphia, PA 19129 email: [email protected] Stephen G. Chaney, Ph.D.  Professor  Departments of Biochemistry and Biophysics and of Nutrition  Mary Ellen Jones Building  University of North Carolina at Chapel Hill School of Medicine CB# 7260 Chapel Hill, NC 27599­7260 email: schaney. [email protected] Marguerite W. Coomes, Ph.D.  Associate Professor  Department of Biochemistry and Molecular Biology Howard University College of Medicine  520 W Street, N.W.  Washington, DC 20059­0001 email: [email protected] Joseph G. Cory, Ph.D.  Professor and Chair  Department of Biochemistry  Brody Medical Sciences Building  East Carolina University School of Medicine Greenville, NC 27858­4354 David W. Crabb, M.D.  Professor  Departments of Medicine and of Biochemistry and Molecular Biology Emerson Hall 317  Indiana University School of Medicine  545 Barnhill Drive  Indianapolis, IN 46202­5124 email: dcrabb@­medicine.dmed.iupi.edu Thomas M. Devlin, Ph.D.  Professor Emeritus  Department of Biochemistry  MCP∙Hahnemann School of Medicine  Allegheny University of the Health Sciences Broad and Vine Streets  Philadelphia, PA 19102­1192 email: [email protected] John E. Donelson, Ph.D.  Professor  Howard Hughes Medical Institute and Department of Biochemistry University of Iowa College of Medicine  300 Eckstein Medical Research Building  Iowa City, IA 52242 email: [email protected]

 

  

Page viii

Robert H. Glew, Ph.D.  Professor and Chair  Department of Biochemistry  Basic Medical Science Building, Room 249 University of New Mexico  School of Medicine  915 Camino de Salud NE  Albuquerque, NM 87131 email: [email protected] Dohn G. Glitz, Ph.D.  Professor  Department of Biological Chemistry UCLA School of Medicine  Los Angeles, CA 90095­1737 email: [email protected] Robert A. Harris, Ph.D.  Showalter Professor and Chair  Department of Biochemistry and Molecular Biology Indiana University School of Medicine  635 Barnhill Drive  Indianapolis, IN 46202­5122 email: [email protected] Ulrich Hopfer, M.D., Ph.D.  Professor  Department of Physiology and Biophysics Case Western Reserve University  2109 Abington Road  Cleveland, OH 44106­4970 email: [email protected] Michael N. Liebman, Ph.D.  Director, Bioinformatics and Genomics VYSIS, Inc.  3100 Woodcreek Drive  Downers Grove, IL 60515 email: [email protected] Gerald Litwack, Ph.D.  Professor and Chair  Department of Biochemistry and Molecular Pharmacology Deputy Director Kimmel Cancer Institute  Jefferson Medical College  Thomas Jefferson University  233 South 10th Street  Philadelphia, PA 19107 email: [email protected] Bettie Sue Siler Masters, Ph.D.  Robert A. Welch Foundation Professor in Chemistry  Department of Biochemistry  University of Texas Health Science Center at San Antonio 7703 Floyd Curl Drive  San Antonio, TX 78284­7760 email: [email protected] Denis McGarry, Ph.D.  Professor  Departments of Internal Medicine and of Biochemistry  Bldg. G5, Room 210  University of Texas Southwestern Medical Center at Dallas 5323 Harry Hines Blvd  Dallas, TX 75235­9135 email: [email protected] Richard T. Okita, Ph.D.  Professor  Department of Pharmaceutical Science 105 Wegner Hall  College of Pharmacy  Washington State University  Pullman, WA 99164­6510 email: [email protected] Merle S. Olson, Ph.D.  Professor and Chair  Department of Biochemistry  University of Texas Health Science Center 7703 Floyd Curl Drive  San Antonio, TX 78284­7760 email: [email protected] Francis J. Schmidt, Ph.D.  Professor  Department of Biochemistry  M121 Medical Sciences  University of Missouri­Columbia Columbia, MO 65212­0001 email: bcfranks@muccmail. missouri.edu Thomas J. Schmidt, Ph.D.  Associate Professor  Department of Physiology and Biophysics 5­610 Bowen Science Building  University of Iowa, College of Medicine  Iowa City, IA 52242­1109 email: thomas­[email protected]

 

  

Page ix

Richard M. Schultz, Ph.D.  Professor and Chair  Department of Molecular and Cellular Biochemistry Stritch School of Medicine  Loyola University of Chicago  2160 South First Avenue  Maywood, IL 60153 email: [email protected] Nancy B. Schwartz, Ph.D.  Professor  Departments of Pediatrics and of Biochemistry and Molecular Biology University of Chicago, MC 5058  5841 S. Maryland Ave.  Chicago, IL 60637­1463 email: n­[email protected] Thomas E. Smith, Ph.D.  Professor and Chair  Department of Biochemistry and Molecular Biology College of Medicine  Howard University  520 W Street, N.W.  Washington, DC 20059­0001 email: [email protected] Gerald Soslau, Ph.D.  Professor  Department of Biochemistry and Director, IMS Program MCP∙Hahnemann School of Medicine, M.S. 344  Allegheny University of the Health Sciences  Broad and Vine Streets  Philadelphia, PA 19102­1192 email: [email protected] J. Lyndal York, Ph.D.  Professor  Department of Biochemistry and Molecular Biology College of Medicine  University of Arkansas for Medical Science  4301 W. Markham St.  Little Rock, AR 72205­7199 email: [email protected]

 

  

Page xi

Foreword These are very exciting times for biochemistry and especially for that part that pertains to human biology and human medicine. The much discussed Human Genome  Project is likely to be completed very early in the next millennium, by the time most users of Textbook of Biochemistry With Clinical Correlations have graduated.  The Human Genome Project should provide a blueprint of the 100,000 or so genes that the human genome is estimated to contain and lead to an explosion of amazing  proportions in knowledge on complex physiological processes and multigenic disorders. This mapping will reveal undreamed of interrelationships and elucidate control  mechanisms of the fundamental processes of development of the human organism and of their interactions with both milieus (the internal and external). Already, one  eukaryotic genome (that of brewer's yeast, comprising 14 million base pairs in 16 chromosomes) was completed just before I set out to write this Foreword, while  three microbial genomes (that of Mycoplasma genitalium—580,070 base pairs, Hemophilus influenzae—1.83 million base pairs, and Synechosystis—a  photosynthetic organism—3.57 million base pairs) have been completed within 3 to 18 months of isolation of their DNA. Work on the genomes of Mycobacterium  tuberculosis (4.5 million base pairs) and of Plasmodium falciparum—the malarial parasite (27 million base pairs in 14 chromosomes)—is now being undertaken and  should lead to knowledge that can produce novel approaches to the treatment and control of these two scourges of humankind. The theoretical and technical principles  involved in this type of work are clearly described in Chapters 14, 15, and 18 of Textbook of Biochemistry With Clinical Correlations, which will ensure that  readers will understand and appreciate future developments in the field. Discoveries on the molecular basis of human disease are also being reported at an unprecedented and dizzying rate, opening wider and wider the window to many less  frequent afflictions produced by mutated genes accumulating in the human gene pool. The era of molecular medicine has already arrived. Since the very first edition of  Textbook of Biochemistry With Clinical Correlations, the correlations have been a feature that has made the book truly unique. In this new edition, the correlations  are numerous, succinct, and integrated with, but also independent of, the text. They not only reflect current progress but indicate more than ever before how  biochemistry, molecular biology, and human genetics have become the foundation stones of all areas of modern medicine. These previously separate disciplines have  become so intimately and inextricably intertwined that little knowledge and understanding of one can occur without knowledge and understanding of others. One of the  many strengths of this book is that clear examples of the convergence and integration of biological disciplines can be found in the clinical correlations. In this fourth edition of Textbook of Biochemistry With Clinical Correlations, the contributors have provided an up­to­date and logical coverage of basic  biochemistry, molecular biology, and normal and abnormal aspects of physiological chemistry. This material is appropriate and relevant for medical and other health  science students, particularly as we approach the third millenium in the midst of amazing and pervasive progress in medical science and biotechnology. To enhance the  text, a completely new series of vivid illustrations has been added, which will undoubtedly further the readers' understanding of the complexity of many of the concepts.  Students of medical and health sciences should appreciate that the time and effort invested in learning the material presented here will be very well spent. This  knowledge will provide the framework within which further developments will be understood and applied as the readers begin to care for the physical and mental well­ being of those entrusted to them. Furthermore, the knowledge derived from this book will also provide satisfying insight into the processes that underlie human life and  the amazing power of the human mind to explore and understand it. As in previous editions, the fourth edition includes many multiple choice questions (and answers) at  the end of each chapter that should facilitate this learning while ensuring success in professional and other examinations. I am happy and privileged to have watched the growth of human biochemistry (because of my teaching and research responsibilities) since my medical student days  nearly half­a­century ago. It has been an amazing spectacle, full of thrills and exciting adventures into aspects of human cells that were previously shrouded in mystery  and ignorance. As my knowledge has increased, so has my sense of awe and wonder at the unfolding beauty of this marvelous display of nature's secrets. As the late Alberto Sols frequently said: "The Biochemistry of today is the Medicine of tomorrow." Textbook of Biochemistry With Clinical Correlations illustrates  the veracity of this insight. FRANK VELLA  UNIVERSITY OF SASKATCHEWAN

 

  

Page xiii

Preface The purposes of the fourth edition of the Textbook of Biochemistry With Clinical Correlations remain unchanged from the earlier editions: to present a clear  discussion of the biochemistry of mammalian cells; to relate the biochemical events at the cellular level to the physiological processes occurring in the whole animal; and  to cite examples of deviant biochemical processes in human disease. The continued rapid advances in knowledge, particularly due to the techniques of molecular biology, required a critical review and evaluation of the entire content of  the previous edition. Every chapter has been revised and updated. Significant additions of new material, clarifications, and some deletions were made throughout.  Amino acid metabolism was combined into a single chapter and DNA structure and function was divided into two chapters for better coverage of this rapidly  expanding field. Topics for inclusion were selected to cover the essential areas of both biochemistry and physiological chemistry for upper­level undergraduate,  graduate­level and especially professional school courses in biochemistry. Since the application of biochemistry is so important to human medicine, the text has an  overriding emphasis on the biochemistry of mammalian cells. The textbook is written such that any sequence considered most appropriate by an instructor can be presented. It is not formally divided into major sections, but  related topics are grouped together. After an introductory chapter on cell structure, Chapters 2 to 5 concern the Major Structural Components of Cells, that is,  proteins and their many functions, and cell membranes and their major roles. Metabolism is discussed in the following eight chapters, starting with the conservation of  energy, then the synthesis and degradation of the major cellular components, and concluding with a chapter on the integration of these pathways in humans. The next  section of six chapters covers Information Transfer and Its Control, describing the structure and synthesis of the major cellular macromolecules, that is, DNA,  RNA, and protein. A separate chapter on Biotechnology is included because information from this area has had such a significant impact on the development of our  current state of biochemical knowledge. The section concludes with a chapter on the Regulation of Gene Expression in which mechanisms in both prokaryotes and  eukaryotes are presented. The fourth major section represents Signal Transduction and Amplification and includes two chapters on hormones that emphasize their  biochemical functions as messengers and a chapter on Molecular Cell Biology describes four major mammalian signal transducing systems. The textbook concludes  with six chapters on topics that comprise Physiological Chemistry, including cytochrome P450 enzymes and xenobiotic metabolism, iron and heme metabolism, gas  transport and pH regulation, digestion and absorption, and human nutrition. A major addition from previous editions is the extensive use of color in the illustrations as a means to emphasize important points. All figures were reviewed and new  drawings were prepared to illustrate the narrative discussion. In many cases the adage ''A picture is worth a thousand words" is appropriate and the reader is  encouraged to study the illustrations because they are meant to illuminate often confusing aspects of a topic. In each chapter the relevancy of the topic to human life processes are presented in Clinical Correlations, which describe the aberrant biochemistry of disease states.  A number of new correlations have been included. The correlations are not intended to review all of the major diseases but rather to cite examples of disease  processes where the biochemical implications are well established. In addition, we specifically avoided presenting clinical case reports because it was considered more  significant to deal with the general clinical condition. References are included to facilitate exploration of the topic in more detail. In some cases similar clinical problems  are presented in different chapters, but each from a different perspective. All pertinent biochemical information is presented in the main text, and an understanding of  the material does not require a reading of the correlations. In a few cases, clinical discussions are part of the principal text because of the close relationship of some  topics to medical conditions. Each chapter concludes with a set of Questions and Answers; the multiple­choice format was retained as being valuable to students for self­assessment of their  knowledge. The question type was limited to the types now occurring in national examinations. All questions were reviewed and many new ones added. The questions  cover a range of topics in each chapter, and each has an annotated answer, with references to the page in the textbook covering the content of the question. The appendix, Review of Organic Chemistry, is designed as a ready reference for the nomenclature and structures of organic molecules encountered in  biochemistry and is not intended as a comprehensive review of organic chemistry. The material is presented in the Appendix rather than at the beginning of those  chapters dealing with the metabolism of each class of organic molecules. The reader might find it

 

  

Page xiv

valuable to become familiar with the content and then use the Appendix as a ready reference when reading related sections in the main text. We still believe that a multicontributor textbook is the best approach to achieve an accurate and current presentation of biochemistry. Each author is involved actively  in teaching biochemistry in a medical or graduate school and has an active research interest in the field in which he or she has written. Thus, each has the perspective of  the classroom instructor, with the experience to select the topics and determine the emphasis required for students in a course of biochemistry. Every contributor,  however, brings to the book an individual approach, leading to some differences in presentation. However, every chapter was critically edited and revised in order to  have a consistent writing style and to eliminate repetitions and redundancies. A limited repetition of some topics in different chapters was permitted when it was  considered that the repetition would facilitate the learning process. The individual contributors were requested to prepare their chapters for a teaching book. The book is not intended as a compendium of biochemical facts or a review  of the current literature, but each chapter contains sufficient detail on the subject to make it useful as a resource. Each contributor was requested not to refer to specific  researchers; our apologies to those many biochemists who rightfully should be acknowledged for their outstanding research contributions to the field of biochemistry.  Each chapter contains a Bibliography that can be used as an entry point to the research literature. In any project one person must accept the responsibility for the final product. The decisions concerning the selection of topics and format, reviewing the drafts, and  responsibility for the final checking of the book were entirely mine. I welcome comments, criticisms, and suggestions from the students, faculty, and professionals who  use this textbook. It is our hope that this work will be of value to those embarking on the exciting experience of learning biochemistry for the first time and to those who  are returning to a topic in which the information is expanding so rapidly. THOMAS M. DEVLIN

 

  

Page xv

Acknowledgments Without the encouragement and participation of many people, this project would never have been accomplished. My personal and very deep appreciation goes to  each of the contributors for accepting the challenge of preparing the chapters, for sharing their ideas and making recommendations to improve the book, for accepting  so readily suggestions to modify their contributions, and for cooperating throughout the period of preparation. To each I extend my sincerest thanks for a job well  done. The contributors received the support of associates and students in the preparation of their chapters, and, for fear of omitting someone, it was decided not to  acknowledge individuals by name. To everyone who gave time unselfishly and shared in the objective and critical evaluation of the text, we extend a sincere thank you.  In addition, every contributor has been influenced by former teachers and colleagues, various reference resources, and, of course, the research literature of  biochemistry; we are deeply indebted to these many sources of inspiration. I am particularly indebted to Dr. Frank Vella, Professor of Biochemistry at the University of Saskatchewan, Canada, who assisted me in editing the text. Dr. Vella is a  distinguished biochemist who has made a major personal effort to improve the teaching of biochemistry throughout the world. He read every chapter in draft form and  made significant suggestions for clarifying and improving the presentation. Dr. Vella also honored me by writing the Foreword to the fourth edition of this textbook. I  extend to him my deepest appreciation and thanks for his participation and friendship. A very special thanks to two friends and colleagues who again have been of immeasurable value to me during the preparation of this edition: My gratitude goes to Dr.  James Baggott, who patiently allowed me to use him as a sounding board for ideas and who unselfishly shared with me his suggestions and criticisms of the text, and to  Dr. Carol Angstadt, who reviewed many of the chapters and gave me valuable suggestions. To each I extend my deepest gratitude. I extend my sincerest appreciation and thanks to the members of the staff of the STM Division of John Wiley & Sons who participated in the preparation of this  edition. Special recognition and thanks go to Dr. Brian Crawford, Vice President and General Manager of Life Sciences and Medicine, who gave his unqualified  support to the preparation of the fourth edition. I am indebted to Joe Ingram, Publisher, Life Sciences, who conscientiously guided the planning of this edition. I am  very indebted to Dr. Stephanie Diment, Associate Editor, for always being available to answer questions and to make valuable suggestions, and who has patiently kept  me on track. She has been a constant support; thank you. My deepest appreciation is extended to John Sollami, Senior Managing Editor, who with constant good  humor meticulously oversaw the production. He kept the flow of activities reasonable, listened patiently to my suggestions and concerns, and kept us on schedule. It  has been a real pleasure to work with a really knowledgeable and conscientious professional and to him I extend a very special thanks. I extend to Louise Page, New  Media Editor, my deepest appreciation for her skillful organization of the CD containing the figures from the textbook. Credit for the design of the book goes to Laura  Ierardi, to whom I extend my appreciation. My thanks to Christina Della Bartolomea, copyeditor, and Maria Coughlin, indexer, both of whom did an excellent job. A  significant improvement in this edition is the addition of many original illustrations. My most heartfelt thanks go to Dean Gonzalez. STM Illustration Manager, and  Barbara Kennedy, Illustration Supervisor, at Wiley, who handled the details and flow of illustrations. A special recognition is extended to Dr. Lisa Gardner, Production  Manager and Editor of Page Two, and her staff who transformed the rough drawings of the contributors into meaningful illustrations. No book is successful without the  activities of a Marketing Department; special thanks are due to Reed Elfenbein, Vice President, Marketing and Sales, David Stier, Senior Marketing Manager, David  Steltenkamp, Associate Marketing Manager, and their colleagues at Wiley for their new ideas and efforts. Finally, a very special thanks to my loving, supportive, and considerate wife, Marjorie, who had the foresight to encourage me to undertake this project, who again  supported me during the days of intensive work, and who again created an environment in which I could devote the many hours required for the preparation of this  textbook. To her my deepest appreciation. THOMAS M. DEVLIN

 

  

Page xvii

Contents in Brief 1  Eukaryotic Cell Structure

1

2  Proteins I: Composition and Structure

23

3  Proteins II: Structure­Function Relationships in Protein Families

87

4  Enzymes: Classification, Kinetics, and Control

127

5  Biological Membranes: Structure and Membrane Transport

179

6  Bioenergetics and Oxidative Metabolism

217

7  Carbohydrate Metabolism I: Major Metabolic Pathways and their Control

267

8  Carbohydrate Metabolism II: Special Pathways

335

9  Lipid Metabolism I: Utilization and Storage of Energy in Lipid Form

361

10  Lipid Metabolism II: Pathways of Metabolism of Special Lipids

395

11  Amino Acid Metabolism

445

12  Purine and Pyrimidine Nucleotide Metabolism

489

13  Metabolic Interrelationships

525

14  DNA I: Structure and Conformation

563

15  DNA II: Repair, Synthesis, and Recombination

621

16  RNA: Structure, Transcription, and Processing

677

17  Protein Synthesis: Translation and Posttranslational Modifications

713

18  Recombinant DNA and Biotechnology

757

19  Regulation of Gene Expression

799

20  Biochemistry of Hormones I: Polypeptide Hormones

839

21  Biochemistry of Hormones II: Steroid Hormones

893

22  Molecular Cell Biology

919

23  Biotransformations: The Cytochromes P450

981

24  Iron and Heme Metabolism

1001

25  Gas Transport and pH Regulation

1025

26  Digestion and Absorption of Basic Nutritional Constituents

1055

27  Principles of Nutrition I: Macronutrients

1087

28  Principles of Nutrition II: Micronutrients

1107

Appendix  Review of Organic Chemistry

1137

Index

1149

   

  

Page xix

Contents 1  Eukaryotic Cell Structure Thomas M. Devlin

1

1.1 Overview: Cells and Cellular Compartments

2

1.2 Cellular Environment: Water and Solutes

4

1.3 Organization and Composition of Eukaryotic Cells

12

1.4 Functional Role of Subcellular Organelles and Membrane Systems

15

Clinical Correlations

 

1.1 Blood Bicarbonate Concentration in Metabolic Acidosis

12

1.2 Mitochondrial Diseases: Luft's Disease

16

1.3 Lysosomal Enzymes and Gout

18

1.4 Lysosomal Acid Lipase Deficiency

19

1.5 Zellweger Syndrome and the Absence of Functional Peroxisomes

20

2  Proteins I: Composition and Structure Richard M. Schultz and Michael N. Liebman

23

2.1 Functional Roles of Proteins in Humans

24

2.2 Amino Acid Composition of Proteins

25

2.3 Charge and Chemical Properties of Amino Acids and Proteins

30

2.4 Primary Structure of Proteins

39

2.5 Higher Levels of Protein Organization

42

2.6 Other Types of Proteins

49

2.7 Folding of Proteins from Randomized to Unique Structures: Protein  Stability

62

2.8 Dynamic Aspects of Protein Structure

68

2.9 Methods for Characterization, Purification, and Study of Protein  Structure and Organization

69

Clinical Correlations

 

2.1 Plasma Proteins in Diagnosis of Disease

37

2.2 Differences in Primary Structure of Insulins Used in Treatment of  Diabetes Mellitus

41

2.3 A Nonconservative Mutation Occurs in Sickle Cell Anemia

42

2.4 Symptoms of Diseases of Abnormal Collagen Synthesis

50

2.5 Hyperlipidemias

56

2.6 Hypolipoproteinemias

59

2.7 Glycosylated Hemoglobin, HbA1c

62

2.8 Use of Amino Acid Analysis in Diagnosis of Disease

74

3  Proteins II: Structure­Function Relationships in Protein Families Richard M. Schultz and Michael N. Liebman

87

3.1 Overview

88

3.2 Antibody Molecules: The Immunoglobulin Superfamily

88

3.3 Proteins with a Common Catalytic Mechanism: Serine Proteases

97

3.4 DNA­Binding Proteins

108

3.5 Hemoglobin and Myoglobin

114

Clinical Correlations

 

3.1 The Complement Proteins

91

3.2 Functions of Different Antibody Classes

92

3.3 Immunization

92

3.4 Fibrin Formation in a Myocardial Infarct and the Action of  Recombinant Tissue Plasminogen Activator (rt­PA)

98

3.5 Involvement of Serine Proteases in Tumor Cell Metastasis

99

4  Enzymes: Classification, Kinetics and Control J. Lyndal York 4.1 General Concepts

128

4.2 Classification of Enzymes

129

4.3 Kinetics

133

4.4 Coenzymes: Structure and Function

142

4.5 Inhibition of Enzymes

147

4.6 Allosteric Control of Enzyme Activity

151

4.7 Enzyme Specificity: The Active Site

155

4.8 Mechanism of Catalysis

159

4.9 Clinical Applications of Enzymes

166

4.10 Regulation of Enzyme Activity

174

Clinical Correlations 4.1 A Case of Gout Demonstrates Two Phases in the Mechanism of  Enzyme Action    

  

127

 

138

Page xx

4.2 The Physiological Effect of Changes in Enzyme Km  Values

139

4.3 Mutation of a Coenzyme­Binding Site Results in Clinical Disease

142

4.4 A Case of Gout Demonstrates the Difference between an Allosteric  and Substrate­Binding Site

152

4.5 Thermal Lability of Glucose­6­Phosphate Dehydrogenase Results in  Hemolytic Anemia

166

4.6 Alcohol Dehydrogenase Isoenzymes with Different pH Optima

167

4.7 Identification and Treatment of an Enzyme Deficiency

169

4.8 Ambiguity in the Assay of Mutated Enzymes

169

5  Biological Membranes: Structure and Membrane Transport Thomas M. Devlin

179

5.1 Overview

180

5.2 Chemical Composition of Membranes

180

5.3 Micelles and Liposomes

187

5.4 Structure of Biological Membranes

189

5.5 Movement of Molecules through Membranes

196

5.6 Channels and Pores

201

5.7 Passive Mediated Transport Systems

204

5.8 Active Mediated Transport Systems

206

5.9 Ionophores

211

Clinical Correlations

 

5.1 Liposomes As Carriers of Drugs and Enzymes

189

5.2 Abnormalities of Cell Membrane Fluidity in Disease States

195

5.3 Cystic Fibrosis and the Cl– Channel

202

5.4 Diseases Due to Loss of Membrane Transport Systems

212

6  Bioenergetics and Oxidative Metabolism Merle S. Olson

217

6.1 Energy­Producing and Energy­Utilizing Systems

218

6.2 Thermodynamic Relationships and Energy­Rich Components

220

6.3 Sources and Fates of Acetyl Coenzyme A

226

6.4 The Tricarboxylic Acid Cycle

231

6.5 Structure and Compartmentation by Mitochondrial Membranes

238

6.6 Electron Transfer

246

6.7 Oxidative Phosphorylation

261

Clinical Correlations

 

6.1 Pyruvate Dehydrogenase Deficiency

233

6.2 Fumarase Deficiency

237

6.3 Mitochondrial Myopathies

247

6.4 Subacute Necrotizing Encephalomyelopathy

258

6.5 Cyanide Poisoning

259

6.6 Hypoxic Injury

261

7  Carbohydrate Metabolism I: Major Metabolic Pathways and their Control Robert A. Harris

267

7.1 Overview

268

7.2 Glycolysis

269

7.3 The Glycolytic Pathway

272

7.4 Regulation of the Glycolytic Pathway

283

7.5 Gluconeogenesis

299

7.6 Glycogenolysis and Glycogenesis

312

Clinical Correlations

 

7.1 Alcohol and Barbiturates

281

7.2 Arsenic Poisoning

283

7.3 Fructose Intolerance

285

7.4 Diabetes Mellitus

287

7.5 Lactic Acidosis

291

7.6 Pickled Pigs and Malignant Hyperthermia

291

7.7 Angina Pectoris and Myocardial Infarction

292

7.8 Pyruvate Kinase Deficiency and Hemolytic Anemia

299

7.9 Hypoglycemia and Premature Infants

300

7.10 Hypoglycemia and Alcohol Intoxication

312

7.11 Glycogen Storage Diseases

317

8  Carbohydrate Metabolism II: Special Pathways Nancy B. Schwartz

335

8.1 Overview

336

8.2 Pentose Phosphate Pathway

336

8.3 Sugar Interconversions and Nucleotide Sugar Formation

341

8.4 Biosynthesis of Complex Carbohydrates

346

8.5 Glycoproteins

348

8.6 Proteoglycans

351

Clinical Correlations

   

  

 

8.1 Glucose 6­Phosphate Dehydrogenase: Genetic Deficiency or  Presence of Genetic Variants in Erythrocytes

338

8.2 Essential Fructosuria and Fructose Intolerance: Deficiency of  Fructokinase and Fructose 1­Phosphate Aldolase

342

8.3 Galactosemia: Inability to Transform Galactose into Glucose

343

8.4 Pentosuria: Deficiency of Xylitol Dehydrogenase

345

8.5 Glucuronic Acid: Physiological Significance of Glucuronide  Formation

346

8.6 Blood Group Substances

348

8.7 Aspartylglycosylaminuria: Absence of 4­L­Aspartylglycosamine  Amidohydrolase

349

8.8 Heparin Is an Anticoagulant

350

8.9 Mucopolysaccharidoses

352

Page xxi

9  Lipid Metabolism I: Utilization and Storage of Energy in Lipid Form J. Denis McGarry

361

9.1 Overview

362

9.2 Chemical Nature of Fatty Acids and Acylglycerols

363

9.3 Sources of Fatty Acids

365

9.4 Storage of Fatty Acids As Triacylglycerols

375

9.5 Methods of Interorgan Transport of Fatty Acids and their Primary  Products

378

9.6 Utilization of Fatty Acids for Energy Production

381

Clinical Correlations

 

9.1 Obesity

378

9.2 Leptin and Obesity

378

9.3 Genetic Abnormalities in Lipid­Energy Transport

380

9.4 Genetic Deficiencies in Carnitine Transport or Carnitine  Palmitoyltransferase

384

9.5 Genetic Deficiencies in the Acyl­CoA Dehydrogenases

385

9.6 Refsum's Disease

387

9.7 Diabetic Ketoacidosis

390

10  Lipid Metabolism II: Pathways of Metabolism of Special Lipids Robert H. Glew

395

10.1 Overview

396

10.2 Phospholipids

397

10.3 Cholesterol

409

10.4 Sphingolipids

420

10.5 Prostaglandins and Thromboxanes

431

10.6 Lipoxygenase and Oxy­Eicosatetraenoic Acids

436

Clinical Correlations

 

10.1 Respiratory Distress Syndrome

400

10.2 Treatment of Hypercholesterolemia

416

10.3 Atherosclerosis

417

10.4 Diagnosis of Gaucher's Disease in an Adult

430

11  Amino Acid Metabolism Marguerite W. Coomes

445

11.1 Overview

446

11.2 Incorporation of Nitrogen into Amino Acids

447

11.3 Transport of Nitrogen to Liver and Kidney

452

11.4 Urea Cycle

453

11.5 Synthesis and Degradation of Individual Amino Acids

456

Clinical Correlations

 

11.1 Carbamoyl Phosphate Synthetase and N­Acetylglutamate  Synthetase Deficiencies

456

11.2 Deficiencies of Urea Cycle Enzymes

457

11.3 Nonketotic Hyperglycinemia

461

11.4 Folic Acid Deficiency

463

11.5 Phenylketonuria

465

11.6 Disorders of Tyrosine Metabolism

467

11.7 Parkinson's Disease

467

11.8 Hyperhomocysteinemia and Atherogenesis

471

11.9 Other Diseases of Sulfur Amino Acids

471

11.10 Diseases of Metabolism of Branched­Chain Amino Acids

479

11.11 Diseases of Propionate and Methylmalonate Metabolism

480

11.12 Diseases Involving Lysine and Ornithine

481

11.13 Histidinemia

482

11.14 Diseases of Folate Metabolism

483

12  Purine and Pyrimidine Nucleotide Metabolism Joseph G. Cory

489

12.1 Overview

490

12.2 Metabolic Functions of Nucleotides

490

12.3 Chemistry of Nucleotides

492

12.4 Metabolism of Purine Nucleotides

493

12.5 Metabolism of Pyrimidine Nucleotides

503

12.6 Deoxyribonucleotide Formation

507

12.7 Nucleoside and Nucleotide Kinases

511

12.8 Nucleotide­Metabolizing Enzymes As a Function of the Cell Cycle and  Rate of Cell Division

511

12.9 Nucleotide Coenzyme Synthesis

514

12.10 Synthesis and Utilization of 5­Phosphoribosyl­1­Pyrophosphate

516

12.11 Compounds that Interfere with Cellular Purine and Pyrimidine  Nucleotide Metabolism: Chemotherapeutic Agents

517

Clinical Correlations

 

12.1 Gout

498

12.2 Lesch–Nyhan Syndrome

499

12.3 Immunodeficiency Diseases Associated with Defects in Purine  Nucleoside Degradation

503

12.4 Hereditary Orotic Aciduria

505

13  Metabolic Interrelationships Robert A. Harris and David W. Crabb

525

   

  

13.1 Overview

526

13.2 Starve–Feed Cycle

528

13.3 Mechanisms Involved in Switching the Metabolism of Liver between the  Well­Fed State and the Starved State

539

13.4 Metabolic Interrelationships of Tissues in Various Nutritional and  Hormonal States

547

Page xxii

Clinical Correlations

 

13.1 Obesity

526

13.2 Protein Malnutrition

527

13.3 Starvation

527

13.4 Reye's Syndrome

533

13.5 Hyperglycemic, Hyperosmolar Coma

537

13.6 Hyperglycemia and Protein Glycation

538

13.7 Noninsulin­Dependent Diabetes Mellitus

549

13.8 Insulin­Dependent Diabetes Mellitus

550

13.9 Complications of Diabetes and the Polyol Pathway

551

13.10 Cancer Cachexia

553

14  DNA I: Structure and Conformation Stelios Aktipis

563

14.1 Overview

564

14.2 Structure of DNA

565

14.3 Types of DNA Structure

584

14.4 DNA Structure and Function

609

Clinical Correlations

 

14.1 DNA Vaccines

565

14.2 Diagnostic Use of Probes in Medicine

583

14.3 Topoisomerases in Treatment of Cancer

594

14.4 Hereditary Persistence of Fetal Hemoglobin

600

14.5 Therapeutic Potential of Triplex DNA Formation

600

14.6 Expansion of DNA Triple Repeats and Human Disease

602

14.7 Mutations of Mitochondrial DNA: Aging and Degenerative  Diseases

617

15  DNA II: Repair, Synthesis, and Recombination Stelios Aktipis

621

15.1 Overview

622

15.2 Formation of the Phosphodiester Bond in Vivo

622

15.3 Mutation and Repair of DNA

627

15.4 DNA Replication

642

15.5 DNA Recombination

661

15.6 Sequencing of Nucleotides in DNA Clinical Correlations

671  

15.1 Mutations and the Etiology of Cancer

633

15.2 Defects in Nucleotide Excision Repair and Hereditary Diseases

638

15.3 DNA Ligase Activity and Bloom Syndrome

639

15.4 DNA Repair and Chemotherapy

639

15.5 Mismatch DNA Repair and Cancer

641

15.6 Telomerase Activity in Cancer and Aging

658

15.7 Inhibitors of Reverse Transcriptase in Treatment of AIDS

661

15.8 Immunoglobulin Genes Are Assembled by Recombination

663

15.9 Transposons and Development of Antibiotic Resistance

670

15.10 DNA Amplification and Development of Drug Resistance

671

15.11 Nucleotide Sequence of the Human Genome

672

16  RNA: Structure, Transcription, and Processing Francis J. Schmidt

677

16.1 Overview

678

16.2 Structure of RNA

679

16.3 Types of RNA

681

16.4 Mechanisms of Transcription

689

16.5 Posttranscriptional Processing

699

16.6 Nucleases and RNA Turnover

708

Clinical Correlations

 

16.1 Staphylococcal Resistance to Erythromycin

683

16.2 Antibiotics and Toxins that Target RNA Polymerase

692

16.3 Fragile X Syndrome: A Chromatin Disease?

697

16.4 Involvement of Transcriptional Factors in Carcinogenesis

701

16.5 Thalassemia Due to Defects in Messenger RNA Synthesis

705

16.6 Autoimmunity in Connective Tissue Disease

706

17  Protein Synthesis: Translation and Posttranslational Modifications Dohn Glitz

713

17.1 Overview

714

17.2 Components of the Translational Apparatus

714

17.3 Protein Biosynthesis

724

17.4 Protein Maturation: Modification, Secretion, and Targeting

735

17.5 Organelle Targeting and Biogenesis

739

17.6 Further Posttranslational Protein Modifications

743

17.7 Regulation of Translation

748

17.8 Protein Degradation and Turnover

750

Clinical Correlations

   

  

 

17.1 Missense Mutation: Hemoglobin

721

17.2 Disorders of Terminator Codons

722

17.3 Thalassemia

722

17.4 Mutation in Mitochondrial Ribosomal RNA Results in Antibiotic­ Induced Deafness

734

17.5 I­Cell Disease

740

17.6 Familial Hyperproinsulinemia

743

17.7 Absence of Posttranslational Modification: Multiple Sulfatase  Deficiency

746

17.8 Defects in Collagen Synthesis

749

Page xxiii

17.9 Deletion of a Codon, Incorrect Posttranslational Modification, and  Premature Protein Degradation: Cystic Fibrosis

752

18  Recombinant DNA and Biotechnology Gerald Soslau

757

18.1 Overview

758

18.2 Polymerase Chain Reaction

759

18.3 Restriction Endonuclease and Restriction Maps

760

18.4 DNA Sequencing

762

18.5 Recombinant DNA and Cloning

765

18.6 Selection of Specific Cloned DNA in Libraries

770

18.7 Techniques for Detection and Identification of Nucleic Acids

773

18.8 Complementary DNA and Complementary DNA Libraries

777

18.9 Bacteriophage, Cosmid, and Yeast Cloning Vectors

778

18.10 Techniques to further Analyze Long Stretches of DNA

781

18.11 Expression Vectors and Fusion Proteins

783

18.12 Expression Vectors in Eukaryotic Cells

784

18.13 Site­Directed Mutagenesis

786

18.14 Applications of Recombinant DNA Technologies

790

18.15 Concluding Remarks

795

Clinical Correlations

 

18.1 Polymerase Chain Reaction and Screening for Human  Immunodeficiency Virus

760

18.2 Restriction Mapping and Evolution

762

18.3 Direct Sequencing of DNA for Diagnosis of Genetic Disorders

766

18.4 Multiplex PCR Analysis of HGPRTase Gene Defects in Lesch– Nyhan Syndrome

770

18.5 Restriction Fragment Length Polymorphisms Determine the Clonal  Origin of Tumors

776

18.6 Site­Directed Mutagenesis of HSV I gD

789

18.7 Normal Genes Can Be Introduced into Cells with Defective Genes  in Gene Therapy

793

18.8 Transgenic Animal Models

795

19  Regulation of Gene Expression John E. Donelson

799

19.1 Overview

800

19.2 Unit of Transcription in Bacteria: The Operon

800

19.3 Lactose Operon of E. Coli

802

19.4 Tryptophan Operon of E. Coli

807

19.5 Other Bacterial Operons

813

19.6 Bacterial Transposons

816

19.7 Inversion of Genes in Salmonella

818

19.8 Organization of Genes in Mammalian DNA

820

19.9 Repetitive DNA Sequences in Eukaryotes

822

19.10 Genes for Globin Proteins

824

19.11 Genes for Human Growth Hormone­Like Proteins

829

19.12 Mitochondrial Genes

830

19.13 Bacterial Expression of Foreign Genes

832

19.14 Introduction of Rat Growth Hormone Gene into Mice Clinical Correlations

835  

19.1 Transmissible Multiple Drug Resistances

816

19.2 Duchenne/Becker Muscular Dystrophy and the Dystrophin Gene

822

19.3 Huntington's Disease and Trinucleotide Repeat Expansions

823

19.4 Prenatal Diagnosis of Sickle Cell Anemia

828

19.5 Prenatal Diagnosis of Thalassemia

829

19.6 Leber's Hereditary Optic Neuropathy (LHON)

831

20  Biochemistry of Hormones I: Polypeptide Hormones Gerald Litwack and Thomas J. Schmidt 20.1 Overview

840

20.2 Hormones and the Hormonal Cascade System

841

20.3 Major Polypeptide Hormones and their Actions

846

20.4 Genes and Formation of Polypeptide Hormones

849

20.5 Synthesis of Amino Acid­Derived Hormones

853

20.6 Inactivation and Degradation of Hormones

857

20.7 Cell Regulation and Hormone Secretion

859

20.8 Cyclic Hormonal Cascade Systems

866

20.9 Hormone–Receptor Interactions

871

20.10 Structure of Receptors: b ­Adrenergic Receptor

875

20.11 Internalization of Receptors

876

20.12 Intracellular Action: Protein Kinases

878

20.13 Oncogenes and Receptor Functions

888

Clinical Correlations

 

  

 

20.1 Testing Activity of the Anterior Pituitary

844

20.2 Hypopituitarism

846

20.3 Lithium Treatment of Manic–Depressive Illness: The  Phosphatidylinositol Cycle

863

21  Biochemistry of Hormones II: Steroid Hormones Gerald Litwack and Thomas J. Schmidt

 

839

893

21.1 Overview

894

21.2 Structures of Steroid Hormones

894

21.3 Biosynthesis of Steroid Hormones

896

21.4 Metabolic Inactivation of Steroid Hormones

901

21.5 Cell–Cell Communication and Control of Synthesis and Release of  Steroid Hormones

901

Page xxiv

21.6 Transport of Steroid Hormones in Blood

908

21.7 Steroid Hormone Receptors

909

21.8 Receptor Activation: Upregulation and Downregulation

914

21.9 A Specific Example of Steroid Hormone Action at Cell Level:  Programmed Death

915

Clinical Correlations

 

21.1 Oral Contraception

907

21.2 Apparent Mineralocorticoid Excess Syndrome

911

21.3 Programmed Cell Death in the Ovarian Cycle

916

22  Molecular Cell Biology Thomas E. Smith

919

22.1 Overview

920

22.2 Nervous Tissue: Metabolism and Function

920

22.3 The Eye: Metabolism and Vision

932

22.4 Muscle Contraction

946

22.5 Mechanism of Blood Coagulation Clinical Correlations

960  

22.1 Lambert–Eaton Myasthenic Syndrome

927

22.2 Myasthenia Gravis: A Neuromuscular Disorder

929

22.3 Macula Degeneration: Other Causes of Vision Loss

936

22.4 Niemann–Pick Disease and Retinitis Pigmentosa

938

22.5 Retinitis Pigmentosa Resulting from a De Novo Mutation in the  Gene Coding for Peripherin

940

22.6 Abnormalities in Color Perception

946

22.7 Troponin Subunits As Markers for Myocardial Infarction

954

22.8 Voltage­Gated Ion Channelopathies

956

22.9 Intrinsic Pathway Defects: Prekallikrein Deficiency

963

22.10 Classic Hemophilia

969

22.11 Thrombosis and Defects of the Protein C Pathway

971

23  Biotransformations: The Cytochromes P450 Richard T. Okita and Bettie Sue Siler Masters

981

23.1 Overview

982

23.2 Cytochromes P450: Nomenclature and Overall Reaction

982

23.3 Cytochromes P450: Multiple Forms

984

23.4 Inhibitors of Cytochromes P450

986

23.5 Cytochrome P450 Electron Transport Systems

987

23.6 Physiological Functions of Cytochromes P450

989

23.7 Other Hemoprotein­ and Flavoprotein­Mediated Oxygenations: The  Nitric Oxide Synthases

995

Clinical Correlations

 

23.1 Consequences of Induction of Drug­Metabolizing Enzymes

986

23.2 Genetic Polymorphisms of Drug­Metabolizing Enzymes

987

23.3 Deficiency of Cytochrome P450 Steroid 21­Hydroxylase  (CYP21A2)

992

23.4 Steroid Hormone Production during Pregnancy

993

23.5 Clinical Aspects of Nitric Oxide Production

996

24  Iron and Heme Metabolism William M. Awad, Jr.

1001

24.1 Iron Metabolism: Overview

1002

24.2 Iron­Containing Proteins

1003

24.3 Intestinal Absorption of Iron

1005

24.4 Molecular Regulation of Iron Utilization

1006

24.5 Iron Distribution and Kinetics

1007

24.6 Heme Biosynthesis

1009

24.7 Heme Catabolism

1017

Clinical Correlations

 

24.1 Iron Overload and Infection

1003

24.2 Duodenal Iron Absorption

1005

24.3 Mutant Iron­Responsive Element

1007

24.4 Ceruloplasmin Deficiency

1008

24.5 Iron­Deficiency Anemia

1009

24.6 Hemochromatosis: Molecular Genetics and the Issue of Iron­ Fortified Diets

1011

24.7 Acute Intermittent Porphyria

1013

24.8 Neonatal Isoimmune Hemolysis

1020

24.9 Bilirubin UDP­Glucuronosyltransferase Deficiency

1020

24.10 Elevation of Serum Conjugated Bilirubin

1021

25  Gas Transport and pH Regulation James Baggott

1025

25.1 Introduction to Gas Transport

1026

25.2 Need for a Carrier of Oxygen in Blood

1026

25.3 Hemoglobin and Allosterism: Effect of 2,3­Bisphosphoglycerate

1029

25.4 Other Hemoglobins

1030

25.5 Physical Factors that Affect Oxygen Binding

1031

25.6 Carbon Dioxide Transport

1031

25.7 Interrelationships among Hemoglobin, Oxygen, Carbon Dioxide,  Hydrogen Ion, and 2,3­Bisphosphoglycerate

1036

25.8 Introduction to pH Regulation

1036

25.9 Buffer Systems of Plasma, Interstitial Fluid, and Cells

1036

25.10 The Carbon Dioxide–Bicarbonate Buffer System

1038

25.11 Acid–Base Balance and its Maintenance

1041

25.12 Compensatory Mechanisms

1046

25.13 Alternative Measures of Acid–Base Imbalance

1049

25.14 The Significance of Na+ and Cl– in Acid–Base Imbalance Clinical Correlations

   

  

1050  

25.1 Diaspirin Hemoglobin

1026

25.2 Cyanosis

1028

Page xxv

25.3 Chemically Modified Hemoglobins: Methemoglobin and  Sulfhemoglobin

1030

25.4 Hemoglobins with Abnormal Oxygen Affinity

1032

25.5 The Case of the Variable Constant

1039

25.6 The Role of Bone in Acid–Base Homeostasis

1042

25.7 Acute Respiratory Alkalosis

1047

25.8 Chronic Respiratory Acidosis

1048

25.9 Salicylate Poisoning

1049

25.10 Evaluation of Clinical Acid–Base Data

1051

25.11 Metabolic Alkalosis

1052

26  Digestion and Absorption of Basic Nutritional Constituents Ulrich Hopfer

1055

26.1 Overview

1056

26.2 Digestion: General Considerations

1059

26.3 Epithelial Transport

1063

26.4 Digestion and Absorption of Proteins

1070

26.5 Digestion and Absorption of Carbohydrates

1073

26.6 Digestion and Absorption of Lipids

1077

26.7 Bile Acid Metabolism

1083

Clinical Correlations

 

26.1 Cystic Fibrosis

1067

26.2 Bacterial Toxigenic Diarrheas and Electrolyte Replacement  Therapy

1068

26.3 Neutral Amino Aciduria (Hartnup Disease)

1073

26.4 Disaccharidase Deficiency

1075

26.5 Cholesterol Stones

1081

26.6 A­ b ­Lipoproteinemia

1082

27  Principles of Nutrition I: Macronutrients Stephen G. Chaney

1087

27.1 Overview

1088

27.2 Energy Metabolism

1088

27.3 Protein Metabolism

1089

27.4 Protein–Energy Malnutrition

1093

27.5 Excess Protein–Energy Intake

1094

27.6 Carbohydrates

1095

27.7 Fats

1097

27.8 Fiber

1097

27.9 Composition of Macronutrients in the Diet

1098

Clinical Correlations

 

27.1 Vegetarian Diets and Protein–Energy Requirements

1091

27.2 Low­Protein Diets and Renal Disease

1092

27.3 Providing Adequate Protein and Calories for the Hospitalized  Patient

1093

27.4 Carbohydrate Loading and Athletic Endurance

1096

27.5 High­Carbohydrate Versus High­Fat Diets for Diabetics

1096

27.6 Polyunsaturated Fatty Acids and Risk Factors for Heart Disease

1099

27.7 Metabolic Adaptation: The Relationship between Carbohydrate  Intake and Serum Triacylglycerols

1100

28  Principles of Nutrition II: Micronutrients Stephen G. Chaney

1107

28.1 Overview

1108

28.2 Assessment of Malnutrition

1108

28.3 Recommended Dietary Allowances

1109

28.4 Fat­Soluble Vitamins

1109

28.5 Water­Soluble Vitamins

1118

28.6 Energy­Releasing Water­Soluble Vitamins

1119

28.7 Hematopoietic Water­Soluble Vitamins

1123

28.8 Other Water­Soluble Vitamins

1127

28.9 Macrominerals

1128

28.10 Trace Minerals

1130

28.11 The American Diet: Fact and Fallacy

1132

28.12 Assessment of Nutritional Status in Clinical Practice

1133

Clinical Correlations

 

28.1 Nutritional Considerations for Cystic Fibrosis

1112

28.2 Renal Osteodystrophy

1113

28.3 Nutritional Considerations in the Newborn

1117

28.4 Anticonvulsant Drugs and Vitamin Requirements

1118

28.5 Nutritional Considerations in the Alcoholic

1120

28.6 Vitamin B6 Requirements for Users of Oral Contraceptives

1124

28.7 Diet and Osteoporosis

1129

28.8 Nutritional Considerations for Vegetarians

1134

28.9 Nutritional Needs of Elderly Persons

1134

Appendix  Review of Organic Chemistry Carol N. Angstadt

1137

Index

1149

   

  

Page xxvii

Chapter Questions and Answers The questions at the end of each chapter are provided to help you test your knowledge and increase your understanding of biochemistry. Since they are intended to  help you strengthen your knowledge, their construction does not always conform to principles for assessing your retention of individual facts. Specifically, you will  sometimes be expected to draw on your knowledge of several areas to answer a single question, and some questions may take longer to analyze than the average time  allowed on certain national examinations. Occasionally, you may disagree with the answer. If this occurs, we hope that after you read the commentary that  accompanies the answer to the question, you will see the point and your insight into the biochemical problem will be increased. The question types conform to those currently used in objective examinations. They are: Type 1: Choose the one best answer Type 2: Match the numbered statement or phrase with one of the lettered options given above.

 

  

Page 1

Chapter 1—  Eukaryotic Cell Structure Thomas M. Devlin

1.1 Overview: Cells and Cellular Compartments

2

1.2 Cellular Environment: Water and Solutes

4

Hydrogen Bonds Form between Water Molecules

4

Water Has Unique Solvent Properties

5

Some Molecules Dissociate with Formation of Cations and Anions

5

Weak Electrolytes Dissociate Partially

6

Water Is a Weak Electrolyte

6

Many Biologically Important Molecules Are Acids or Bases

7

The Henderson–Hasselbalch Equation Defines the Relationship between  pH and Concentrations of Conjugate Acid and Base

9

Buffering Is Important to Control pH

10

1.3 Organization and Composition of Eukaryotic Cells

12

Chemical Composition of Cells

13

1.4 Functional Role of Subcellular Organelles and Membrane Systems Plasma Membrane Is the Limiting Boundary of a Cell

16

Nucleus Is Site of DNA and RNA Synthesis

16

Endoplasmic Reticulum Has a Role in Many Synthetic Pathways

16

The Golgi Apparatus Is Involved in Sequestering of Proteins

17

Mitochondria Supply Most Cell Needs for ATP

17

Lysosomes Are Required for Intracellular Digestion

17

Peroxisomes Contain Oxidative Enzymes Involving Hydrogen Peroxide

19

Cytoskeleton Organizes the Intracellular Contents

19

Cytosol Contains Soluble Cellular Components

20

Conclusion

20

Bibliography

20

Questions and Answers

21

Clinical Correlations

   

  

15

 

1.1 Blood Bicarbonate Concentration in Metabolic Acidosis

12

1.2 Mitochondrial Diseases: Luft's Disease

16

1.3 Lysosomal Enzymes and Gout

18

1.4 Lysosomal Acid Lipase Deficiency

19

1.5 Zellweger Syndrome and the Absence of Functional Peroxisomes

20

Page 2

1.1—  Overview:  Cells and Cellular Compartments Over three billion years ago, under conditions not entirely clear and in a time span difficult to comprehend, elements such as carbon, hydrogen, oxygen, nitrogen, sulfur,  and phosphorus formed simple chemical compounds. They combined, dispersed, and recombined to form a variety of larger molecules until a combination was  achieved that was capable of replicating itself. These macromolecules consisted of simpler molecules linked together by chemical bonds. With continued evolution and  formation of ever more complex molecules, the water environment around some of these self­replicating molecules became enclosed by a membrane. This  development gave these primordial structures the ability to control their own environment to some extent. A form of life had evolved and a unit of three­dimensional  space—a cell—had been established. With the passing of time a diversity of cells evolved, and their chemistry and structure became more complex. They could  extract nutrients from the environment, chemically converting these nutrients to sources of energy or to complex molecules, control chemical processes that they  catalyzed, and carry out cellular replication. Thus the vast diversity of life observed today began. The cell is the basic unit of life in all forms of living organisms, from the  smallest bacterium to the most complex animal. The limiting outer membrane of cells, the plasma membrane, delineates the space occupied by a cell and separates the variable and potentially hostile environment  outside from the relatively constant milieu within. It is the communication link between the cell and its surroundings. On the basis of microscopic and biochemical differences, living cells are divided into two major classes: prokaryotes, which include bacteria, blue­green algae, and  rickettsiae, and eukaryotes, which include yeasts, fungi, and plant and animal cells. Prokaryotes have a variety of shapes and sizes, in most cases being 1/1000 to  1/10,000 the size of eukaryotic cells. They lack intracellular membrane­bound structures that can be visualized by a microscope (Figure 1.1). The deoxyribonucleic  acid (DNA) of prokaryotes is often segregated into a discrete mass, the nucleoid region, that is not surrounded by a membrane or envelope. The plasma membrane is  often invaginated. In contrast, eukaryotic cells have a well­defined membrane surrounding a central nucleus and a variety of intracellular structures and organelles  (Figure 1.1b). Intracellular membrane systems establish distinct subcellular compartments, as described in Section 1.4, that permit a unique degree of subcellular  specialization. By compartmentalization different chemical reactions that require different environments can occur simultaneously. Many reactions occur in or on  specific membranes, thus creating an additional environment for the diverse functions of cells. Besides these structural variations between prokaryotic and eukaryotic cells (Figures 1a and 1b), there are differences in chemical composition and biochemical  activities. Prokaryotes lack histones, a class of proteins that complex with DNA in eukaryotes. There are major structural differences in the ribonucleic acid–protein  complexes involved in biosynthesis of proteins between the cell types, as well as differences in transport mechanisms across the plasma membrane and in enzyme  content. The many similarities, however, are equally striking. The emphasis throughout this book is on the chemistry of eukaryotes, particularly mammalian cells, but  much of our knowledge of the biochemistry of living cells has come from studies of prokaryotic and nonmammalian eukaryotic cells. The basic chemical components  and fundamental chemical reactions of all living cells are very similar. Availability of certain cell populations, for example, bacteria in contrast to human liver, has led to  much of our knowledge about some cells; in some areas our knowledge is derived nearly exclusively from studies of prokaryotes. The universality of many biochemical  phenomena, however, permits many extrapolations from bacteria to humans.

 

  

Page 3

Figure 1.1  Cellular organization of prokaryotic and eukaryotic cells.  (a) Electron micrograph of Escherichia coli, a representative  prokaryote; approximate magnification ×30,000. There is little  apparent intracellular organization and no cytoplasmic organelles.  Chromatin is condensed in a nuclear zone but not surrounded by a  membrane. Prokaryotic cells are much smaller than eukaryotic cells.  (b) Electron micrograph of a thin section of a liver cell (rat hepatocyte),  a representative eukaryotic cell; approximate magnification  ×7500. Note the distinct nuclear membrane, different  membrane­bound organelles or vesicles, and extensive membrane systems.  Various membranes create a variety of intracellular compartments.  Photograph (a) generously supplied by Dr. M. E. Bayer, Fox Chase Cancer  Institute, Philadelphia, PA; photograph  (b) reprinted with permission of  Dr. K. R. Porter, from Porter, K. R., and Bonneville, M. A. In:  Fine Structure of Cells and Tissues. Philadelphia: Lea & Febiger, 1972.

 

  

Page 4

Before we dissect the complexities of mammalian cells and tissues in the following chapters, it is appropriate to review some of the chemical and physical  characteristics of the environment in which the various biochemical phenomena occur. This environment places many constraints on the cell's activities. The concluding  section outlines the activities and roles of subcellular compartments.

Figure 1.2  Structure of a water molecule.  The H–O–H bond angle is 104.5°. Both hydrogen  atoms carry a partial positive charge  and the oxygen a partial negative  charge, creating a dipole.

1.2—  Cellular Environment: Water and Solutes All biological cells contain essentially the same building blocks and types of macromolecules. The general classes of substances in cells are presented in Table 1.1.  There are significant variations in concentration of specific components in different cell types and in organelles of eukaryotic cells. Microenvironments are also  created by macromolecules and membranes in which the composition differs from that of the surrounding milieu. Cells depend on the external environment for nutrients  required for replacement of components, growth, and energy needs. They have a variety of mechanisms to cope with variations in composition of the external  environment. Water is the one common component of all environments. It is the solvent in which the substances required for the cell's existence are dissolved or  suspended. The unique physicochemical properties of water make life on earth possible. Hydrogen Bonds Form between Water Molecules Two hydrogen atoms share their electrons with an unshared pair of electrons of an oxygen atom to form a water molecule. The oxygen nucleus has a stronger  attraction for shared electrons than hydrogen, and positively charged hydrogen nuclei are left with an unequal share of electrons, creating a partial positive charge on  each hydrogen and a partial negative charge on oxygen. The bond angle between hydrogens and oxygen is 104.5°, making the molecule electrically asymmetric and  producing an electric dipole (Figure 1.2). Water molecules interact because positively charged hydrogen atoms on one molecule are attracted to the negatively charged  oxygen atom on another, with formation of a weak bond between two water molecules (Figure 1.3a). This bond, indicated by a dashed line, is a hydrogen bond. A  detailed discussion of noncovalent interactions between molecules, including electrostatic, van der Waals, and hydrophobic, is presented on page 64. Five molecules of  water form a tetrahedral structure (Figure 1.3b), because each oxygen shares its electrons with four hydrogen atoms and each hydrogen with another oxygen. A  tetrahedral lattice structure is formed in ice and gives ice its crystalline structure. Some hydrogen bonds are broken as ice is transformed to liquid water. Each bond is  relatively

Figure 1.3  Hydrogen bonding.  (a) Hydrogen bonding,  indicated by dashed  lines, between two water  molecules.  (b) Tetrahedral  hydrogen bonding of five  water molecules. Water  molecules 1, 2, and 3  are in the plane of the  page, 4 is below, and  5 is above. TABLE 1.1 Chemical Components of Biological Cells Range of Molecular  Weights

Component

18

H2O Inorganic ions +

+



23–100 2–



2+

Na , K , Cl , SO4 , HCO3  Ca , 

 

Mg2+, etc. Small organic molecules Carbohydrates, amino acids, lipids,  nucleotides, peptides

100–1200

 

Macromolecules Proteins, polysaccharides, nucleic  acids

   

  

50,000–1,000,000,000

 

Page 5

weak compared to a covalent bond but the large number of hydrogen bonds between molecules in liquid water is the reason for the stability of water. Liquid water  actually has a definite structure due to hydrogen bonding that is in a dynamic state as these bonds break and reform. Hydrogen bonds in water have a half­life of less  than 1 × 10–10 s. Liquid water contains a significant number of hydrogen bonds even at 100°C, which accounts for its high heat of vaporization; in the transformation  from liquid to vapor state, hydrogen bonds are disrupted. Water molecules hydrogen bond to different chemical structures. Hydrogen bonding also occurs between other molecules and within a molecule wherever  electronegative oxygen or nitrogen comes in close proximity to hydrogen covalently bonded to another electronegative atom. Representative hydrogen bonds are  presented in Figure 1.4. Intramolecular hydrogen bonding occurs extensively in large macromolecules such as proteins and nucleic acids and is partially responsible for  their structural stability. Many models for the structure of liquid water have been proposed, but none adequately explains all its properties. Water Has Unique Solvent Properties The polar nature and ability to form hydrogen bonds are the basis for the unique solvent properties of water. Polar molecules are readily dispersed in water. Salts in  which a crystal lattice is held together by attraction of positive and negative groups dissolve in water because electrostatic forces in the crystal can be overcome by  attraction of charges to the dipole of water. NaCl is an example where electrostatic attraction of individual Na+ and Cl– atoms is overcome by interaction of Na+ with  the negative charge on oxygen atoms, and Cl– with positive charges on the hydrogen atoms. Thus a shell of water surrounds the individual ions. The number of weak  charge–charge interactions between water and Na+ and Cl– ions is sufficient to separate the two charged ions. Many organic molecules that contain nonionic but weakly polar groups are soluble in water because of attraction of these groups to molecules of water. Sugars and  alcohols are readily soluble in water for this reason. Amphipathic molecules, compounds that contain both polar and nonpolar groups, disperse in water if attraction of  the polar group for water can overcome hydrophobic interactions of nonpolar portions of the molecules. Very hydrophobic molecules, such as compounds that contain  long hydrocarbon chains, however, do not readily disperse as single molecules in water but interact with one another to exclude the polar water molecules.

Figure 1.4  Representative hydrogen   bonds of  importance in biological systems.

Some Molecules Dissociate with Formation of Cations and Anions Substances that dissociate in water into a cation (positively charged ion) and an anion (negatively charged ion) are classified as electrolytes. The presence of  charged ions facilitates conductance of an electrical current through an aqueous solution. Sugars or alcohols, which readily dissolve in water but do not carry a charge  or dissociate into charged species, are classified as nonelectrolytes.

Figure 1.5  Reactions that occur when sodium  lactate is dissolved in water.

Salts of alkali metals (e.g., Li, Na, and K), dissolved in water at low concentrations, dissociate completely; at high concentrations, however, there is increased potential  for interaction of anions and cations. With biological systems it is customary to consider such compounds as totally dissociated because their concentrations are low.  Salts of organic acids, for example, sodium lactate, also dissociate totally and are classified as electrolytes; the dissociated anion, lactate ion, reacts to a limited extent  with a proton to form undissociated acid (Figure 1.5). When such salts are dissolved in water, individual ions are present in solution rather than the undissociated salt.  If a solution has been prepared with

 

  

Page 6 +

+

several different salts (e.g., NaCl, K2SO4, and Na lactate), the original molecules do not exist as such in solution, only the ions (e.g., Na , K , SO42– and lactate–). Many acids, however, when dissolved in water do not totally dissociate but rather establish an equilibrium between undissociated and dissociated components. Thus  lactic acid, an important metabolic intermediate, partially dissociates into lactate anions and H+ as follows:

Because of their partial dissociation, however, such compounds have a lower capacity to carry an electrical charge on a molar basis when compared to those that  dissociate totally; they are termed weak electrolytes. Weak Electrolytes Dissociate Partially In partial dissociation of a weak electrolyte, represented by HA, the concentration of the various species can be determined from the equilibrium equation:

A– represents the dissociated anion and square brackets indicate concentration of each component in concentration units such as moles per liter (mol L–1) or millimol  L–1. The activity of each species rather than concentration should be employed in the equilibrium equation but since most compounds of interest in biological systems  are present in low concentration, the value for activity approaches that of concentration. Thus the equilibrium constant is indicated as   cannot be determined  because at equilibrium there is no remaining undissociated solute. Water Is a Weak Electrolyte Water dissociates as follows:

A proton that dissociates interacts with oxygen of another water molecule to form the hydronium ion, H3O+. For convenience, in this book the proton will be presented  as H+ rather than H3O+, even though the latter is the actual chemical species. At 25°C the value of   for dissociation of water is very small and is about 1.8 × 10–16:

With such a small   an insignificant number of water molecules actually dissociate relative to the number of undissociated molecules. Thus the concentration of water,  which is 55.5 M, is essentially unchanged. Equation 1.1 can be rewritten as follows:

 is a constant and is termed the ion product of water. Its value at 25°C is 1 × 10–14. In pure water the concentration of H+ equals OH–, and by  substituting [H+] for [OH–] in the equation above, [H+] is 1 × 10–7 M. Similarly,

 

  

Page 7 –

–7

+



[OH ] is also 1 × 10  M. The equilibrium of H2O, H , and OH  always exists in dilute solutions regardless of the presence of dissolved substances. If dissolved  material alters either the H+ or OH– concentration, as occurs on addition of an acid or base, a concomitant change in the other ion must occur in order to satisfy the  equilibrium relationship. By using the equation for the ion product, [H+] or [OH–] can be calculated if concentration of one of the ions is known. TABLE 1.2 Relationships Between [H+] and pH and  [OH–] and pOH [H +] (M)

pH

[OH–] (M)

pOH

1.0

  0

–14

1 × 10

14

0.1 (1 × 10–1)

  1

1 × 10–13

13

1 × 10–2

  2

1 × 10–12

12

1 × 10–3

  3

1 × 10–11

11

1 × 10–4

  4

1 × 10–10

10

1 × 10–5

  5

1 × 10–9

  9

1 × 10–6

  6

1 × 10–8

  8

1 × 10–7

  7

1 × 10–7

  7

1 × 10–8

  8

1 × 10–6

  6

1 × 10–9

  9

1 × 10–5

  5

1 × 10–10

10

1 × 10–4

  4

1 × 10–11

11

1 × 10–3

  3

1 × 10–12

12

1 × 10–2

  2

1 × 10–13

13

0.1 (1 × 10–1)

  1

1 × 10–14

14

1.0

  0

  The importance of hydrogen ions in biological systems will become apparent in subsequent chapters. For convenience [H+] is usually expressed in terms of pH,  calculated as follows:

In pure water [H+] and [OH–] are both 1 × 10–7 M, and pH = 7.0. [OH–] is expressed as the pOH. For the equation describing dissociation of water, 1 × 10–14 =  [H+][OH–]; taking negative logarithms of both sides, the equation becomes 14 = pH + pOH. Table 1.2 presents the relationship between pH and [H+]. The pH values of different biological fluids are presented in Table 1.3. In blood plasma, [H+] is 0.00000004 M or a pH of 7.4. Other cations are between 0.001 and  0.10 M, well over 10,000 times higher than [H+]. An increase in hydrogen ion to 0.0000001 M (pH 7.0) leads to serious medical consequences and is life threatening;  a detailed discussion of mechanisms by which the body maintains intra­ and extracellular pH is presented in Chapter 25. Many Biologically Important Molecules Are Acids or Bases The definitions of an acid and a base proposed by Lowry and Brønsted are most convenient in considering biological systems. An acid is a proton donor and a base  is a proton acceptor. Hydrochloric acid (HCl) and sulfuric acid (H2SO4) are strong acids because they dissociate totally, and OH– ion is a base because it accepts a  proton, shifting the equilibrium

When a strong acid and OH– are combined, H+ from the acid and OH– interact and are in equilibrium with H2O. Neutralization of H+ and OH– occurs because the ion  product for water is so small. Anions produced when strong acids dissociate totally, such as Cl– from HCl, are not bases because they do not associate with protons in solution. When an organic  acid, such as lactic acid, is dissolved in water it dissociates only partially, establishing an equilibrium between an acid (proton donor), an anion of the acid, and a proton  as follows:

Lactic acid is a weak acid. The anion is a base because it accepts a proton and reforms the acid. The weak acid and the base formed on dissociation are referred to  as a conjugate pair; other examples are presented in Table 1.4. Ammonium ion (NH4+) is an acid because it dissociates to yield H+ and ammonia (NH3), an  uncharged species, which is a conjugate base. Phosphoric acid (H3PO4) is an acid and PO43– is a base, but H2PO4– and HPO42– are either a base or acid depending  on whether the phosphate group is accepting or donating a proton. TABLE 1.3 pH of Some Biological Fluids Fluid

pH

Blood plasma

7.4

Interstitial fluid

7.4

Intracellular fluid

 

Cytosol (liver)

6.9

Lysosomal matrix

Below 5.0

Gastric juice

1.5–3.0

Pancreatic juice

7.8–8.0

Human milk

7.4

Saliva

6.4–7.0

Urine

5.0–8.0

  The tendency of a conjugate acid to dissociate H+ can be evaluated from the  A convenient method of stating the 

 

  

 is in the form of pK¢, as

 of 1 × 10–14 at 25°C.

Page 8 TABLE 1.4 Some Conjugate Acid–Base Pairs of Importance in Biological  Systems Proton Donor (Acid)

Proton Acceptor (Base)

   

CH3–CHOH–COOH

H+ + CH

 

(lactic acid)

(lactate)  

CH3–CO–COOH

H+ + CH3–CO–COO–

 

(pyruvic acid)

(pyruvate)  

HOOC–CH2–CH2–COOH

2H+ + –OOC–CH2–CH2–COO–

 

(succinic acid)

–CHOH–COO–

3

(succinate)  

H  + +H3N–CH2–COO–

H3PO4

 

H+ + H

H2PO4

 

H+ + HPO

 

H+ + PO43–

Glucose 6­PO3

 

H+ + glucose 6­PO32–

H2CO3

 

H+ + HCO3

+

H3NCH2–COOH

+

 

(glycine)

(glycinate)



HPO42– H–

NH4

PO4–

2

42–



+

H2O

 

H  + NH3

 

H+ + OH–

+

  Note the similarity of this definition with that of pH; as with pH and [H+], the relationship between pK  and  systems are presented in Table 1.5.

 and pK  for conjugate acids of importance in biological 

A special case of a weak acid important in medicine is carbonic acid (H2CO3). Carbon dioxide when dissolved in water is involved in the following equilibrium  reactions:

TABLE 1.5 Apparent Dissociation Constant and pK¢  of Some Compounds of Importance in Biochemistry Compound Acetic acid

 

  p K¢  (M)

(CH3—COOH)

Alanine

1.74 × 10–5

  4.76

4.57 × 10–3

  2.34 (COOH)   9.69 (NH3+)

2.04 × 10–10

  Citric acid

  Glutamic acid

  Glycine

8.12 × 10–4 1.77 × 10–5 3.89 × 10–6

  3.09   3.74   5.41

6.45 × 10–3 5.62 × 10–5

  2.19 (COOH)   4.25 (COOH)   9.67 (NH3+)

2.14 × 10–10 4.57 × 10–3 2.51 × 10–10

  2.34 (COOH)   9.60 (NH3+)

  Lactic acid

(CH3—CHOH—COOH)

1.38 × 10–4

  3.86

Pyruvic acid

(CH3—CO—COOH)

3.16 × 10–3

  2.50

Succinic acid

(HOOC—CH2—CH2—COOH)

6.46 × 10–5

  4.19

 

 

3.31 × 10–6

  5.48

Glucose 6­PO3H–

 

7.76 × 10–7

  6.11

H3PO4

 

1 × 10–2

  2.0

H2PO4–

 

2.0 × 10–7

  6.7

HPO4

2–

 

3.4 × 10–13

12.5

H2CO3

 

1.70 × 10–4

  3.77

NH4+

 

5.62 × 10–10

  9.25

H2O

 

1 × 10–14

   

  

14.0

Page 9

Carbonic acid is a relatively strong acid with a 

 of 3.77. The equilibrium equation for this reaction is

Carbonic acid is, however, in equilibrium with dissolved CO2 and the equilibrium equation for this reaction is

Solving Eq. 1.6 for H2CO3 and substituting for the H2CO3 in Eq. 1.5, the two equilibrium reactions are combined into one equation:

Rearranging to combine constants, including the concentration of H2O, simplifies the equation and yields a new combined constant, 

, as follows:

0009­06.gif It is common practice to refer to dissolved CO2 as a conjugate acid; it is the acid anhydride of H2CO3. The term   has a value of 7.95 × 10–7 and  . If the  aqueous system is in contact with an air phase, dissolved CO2 will also be in equilibrium with CO2 in the air phase. A decrease or increase of one component—that is,  CO2 (air), CO2 (dissolved), H2CO3, H+ or   —will cause a change in all the other components. 0009­07.gif

The Henderson–Hasselbalch Equation Defines the Relationship between pH and Concentrations of Conjugate Acid and Base A change in concentration of any component in an equilibrium reaction necessitates a concomitant change in every component. For example, an increase in [H+] will  decrease the concentration of conjugate base (e.g., lactate ion) with an equivalent increase in the conjugate acid (e.g., lactic acid). This relationship is conveniently  expressed by rearranging the equilibrium equation and solving for H+, as shown for the following dissociation:

Rearranging Eq. 1.9 by dividing through by both [H+] and 

 leads to

Taking the logarithm of both sides gives

Since pH = log 1/[H+] and 

 Eq. 1.11 becomes

Equation 1.12, developed by Henderson and Hasselbalch, is a convenient way of viewing the relationship between pH of a solution and relative amounts of conjugate  base and acid present. Analysis of Eq. 1.12 demonstrates that when the ratio of [base]/[acid] is 1 : 1, pH equals the pK  of the acid because log 1 = 0,

 

  

Page 10

Figure 1.6  Ratio of conjugate [base]/[acid] as a  function of the pH.  When the ratio of [base]/[acid] is 1, pH  equals pK  of weak acid.

and thus pH = pK . If pH is one unit less than pK , the [base]/[acid] ratio is 1 : 10, and if pH is one unit above pK , the [base]/[acid] ratio is 10 : 1. Figure 1.6 is a plot  of ratios of conjugate base to conjugate acid versus pH of several weak acids; note that ratios are presented on a logarithmic scale. Buffering Is Important to Control pH When NaOH is added to a solution of a weak acid such as lactic acid, the ratio of [conjugate base]/[conjugate acid] changes. NaOH dissociates totally and the OH–  formed is neutralized by existing H+ to form H2O. The decrease in [H+] will cause further dissociation of weak acid to comply with requirements of its equilibrium  reaction. The amount of weak acid dissociated will be so nearly equal to the amount of OH– added that it is considered to be equal. Thus the decrease in amount of  conjugate acid is equal to the amount of conjugate base that is formed. These series of events are represented in titration curves of two weak acids presented in Figure  1.7. When 0.5 equiv of OH– is added, 50% of the weak acid is dissociated and the [acid]/[base] ratio is 1.0; pH at this point is equal to pK  of the acid. Shapes of  individual titration curves are similar but displaced due to differences in pK  values. There is a rather steep rise in pH when only 0.1 equiv of OH– are added, but  between 0.1 and 0.9 equiv of added OH–, the pH change is only ~2. Thus a large amount of OH– is added with a relatively small change in pH. This is called  buffering and is defined as the ability of a solution to resist a change in pH when an acid or base is added. If weak acid were not present, the pH would be very high  with only a small amount of OH– because there would be no source of H+ to neutralize the OH–. The best buffering range for a conjugate pair is in the pH range near the pK  of the weak acid. Starting from a pH one unit below to a pH one unit above pK , ~82%  of a weak acid in solution will dissociate, and therefore an amount of base equivalent to about 82% of original acid can be neutralized with a change in pH of 2. The  maximum buffering range for a conjugate pair is considered to be between 1 pH unit above and below the pK . Lactic acid with pK  = 3.86 is an effective buffer in the  range of pH 3 to 5 but has no buffering capacity at pH = 7.0. The HPO42–/H2PO4– pair with pK  = 6.7, however, is an effective buffer at pH = 7.0. Thus at the pH of  the cell's cytosol (~7.0), the lactate–lactic acid pair is not an effective buffer but the phosphate system is.

Figure 1.7  Acid–base titration curves for lactic acid  (pK¢  3.86) and NH4+ (pK¢  9.25).  At pH equal to respective   pK  values, there will  be an equal amount of acid and base for  each conjugate pair.

Buffering capacity also depends on the concentrations of conjugate acid and base. The higher the concentration of conjugate base, the more added H+ with which it  can react. The more conjugate acid the more added OH– can be

 

  

Page 11 2–



neutralized by the dissociation of the acid. A case in point is blood plasma at pH 7.4. For HPO4 /H2PO4  the pK  of 6.7 would suggest that this conjugate pair would  be an effective buffer; the concentration of the phosphate pair, however, is low compared to that of the HCO3–/CO2 system with a pK  of 6.1, which is present at a  20­fold higher concentration and accounts for most of the buffering capacity. In considering the buffering capacity both the pK  and the concentration of the conjugate  pair must be taken into account. Most organic acids are relatively unimportant as buffers in cellular fluids because their pK  values are more than several pH units lower  than the pH of the cell, and their concentrations are too low in comparison to such buffers as HPO42–/H2PO4– and the HCO3–/CO2 system. The importance of pH and buffers in biochemistry and clinical medicine will become apparent, particularly in Chapters 2, 4, and 25. Figure 1.8 presents

Figure 1.8  Typical problems of pH and buffering.

 

  

Page 12

some typical problems using the Henderson–Hasselbalch equation and Clin. Corr. 1.1 is a representative problem encountered in clinical practice. CLINICAL CORRELATION 1.1  Blood Bicarbonate Concentration in Metabolic Acidosis Blood buffers in a normal adult control blood pH at about 7.40; if the pH should drop  below 7.35, the condition is referred to as an acidosis. A blood pH of near 7.0 could  lead to serious consequences and possibly death. Thus in acidosis, particularly that  caused by a metabolic change, it is important to monitor the acid–base parameters of a  patient's blood. Values of interest to a clinician include the pH and HCO3– and CO2  concentrations. Normal values for these are pH = 7.40, [HCO3–] = 24.0 mM, and [CO2]  = 1.20 mM. Blood values of a patient with a metabolic acidosis were pH = 7.03 and [CO2] = 1.10  mM. What is the patient's blood [HCO3–] and how much of the normal [HCO3–] has  been used in buffering the acid causing the condition? 1. The Henderson–Hasselbalch equation is

The pK  value for [HCO3–]/[CO2] is 6.10. 2. Substitute the given values in the equation.

or

The antilog of 0.93 is 8.5; thus

or

3. Since the normal value of [HCO3–] is 24 mM, there has been a decrease of 14.6 mmol  of HCO3– per liter of blood in this patient. If much more HCO3– is lost, a point would be  reached when this important buffer would be unavailable to buffer any more acid in the  blood and the pH would drop rapidly. In Chapter 25 there is a detailed discussion of the  causes and compensations that occur in such conditions.   1.3—  Organization and Composition of Eukaryotic Cells As described above, eukaryotic cells are organized into compartments, each delineated by a membrane (Figure 1.9). These are well­defined cellular organelles such as  nucleus, mitochondria, lysosomes, and peroxisomes. Membranes also form a tubule­like network throughout the cell enclosing an interconnecting space or cisternae,  as is the case of the endoplasmic reticulum or Golgi complex. As described in Section 1.4, these compartments have specific functions and activities. The semipermeable nature of cellular membranes prevents the ready diffusion of many molecules from one side to the other. Specific mechanisms in membranes for  translocation of large and small, charged and uncharged molecules allow membranes to modulate concentrations of substances in various compartments.  Macromolecules, such as proteins and nucleic acids, do not cross biological membranes unless there is a specific mechanism for their translocation or the membrane is  damaged. Thus the fluid matrix of various cellular compartments has a distinctive composition of inorganic ions, organic molecules, and macromolecules. Partitioning of  activities and components in membrane­enclosed compartments and organelles has a number of advantages for the economy of the cell. These include the sequestering  of substrates and cofactors where they are required, and adjustments of pH and ionic composition for maximum activity of biological processes. The activities and composition of cellular structures and organelles have been determined with intact cells by a variety of histochemical, immunological, and fluorescent  staining methods. Continuous observation in real time of cellular events in intact viable cells is possible. Examples are studies that involve changes of ionic calcium  concentration in the cytosol by the use of fluorescent calcium indicators. Individual organelles, membranes, and components of the cytosol can be isolated and analyzed  following disruption of the plasma membrane. Permeability of the plasma membrane can be altered to permit the release of subcellular components. Techniques for  disrupting membranes include use of detergents, osmotic shock, and homogenization of tissues, where shearing forces break down the plasma membrane. In an  appropriate isolation medium, cell organelles and membrane systems can be separated by centrifugation because of differences in size and density. Chromatographic  procedures have been employed for isolation of individual cellular fractions and components. These techniques have permitted isolation of cellular fractions from most  mammalian tissues. In addition, components of organelles such as nuclei and mitochondria can be isolated following disruption of the organelle membrane. In many instances the isolated structures and cellular fractions appear to retain the chemical and biochemical characteristics of the structure in situ. But biological  membrane systems are very sensitive structures, subject to damage even under very mild conditions, and alterations can occur during isolation, which can lead to  change in composition of the structure. The slightest damage to a membrane alters its permeability properties, allowing substances that would normally be excluded to  traverse the membrane barrier. In addition, many proteins are only loosely associated with membranes and easily dissociate when damage occurs (see p. 186). Not unexpectedly, there are differences in structure, composition, and activities of cells from different tissues due to the diverse functions of tissues. Major biochemical  activities of the cellular organelles and membrane systems, however, are fairly constant from tissue to tissue. Thus biochemical pathways in liver are often present in  other tissues. The differences between cell types are

 

  

Page 13

Figure 1.9  (a) Electron micrograph of a rat liver cell  labeled to indicate the major  structural components of eukaryotic cells and  (b) a schematic drawing of an animal cell.  Note the number and variety of subcellular organelles  and the network of interconnecting membranes enclosing channels, that is,  cisternae. All eukaryotic cells are not as complex in their appearance, but  most contain the major structures shown in the figure. ER, endoplasmic  reticulum; G, Golgi zone, Ly, lysosomes, P, peroxisomes; M, mitochondria.  Photograph (a) reprinted with permission of Dr. K. R. Porter from Porter,  K. R., and Bonneville, M. A. In: Fine Structure of Cells and Tissues.  Philadelphia: Lea & Febiger, 1972; schematic  (b) reprinted with permission from Voet, D., and Voet,  J. G. Biochemistry, 2nd ed. New York: Wiley, 1995.

usually in distinctive specialized activities. Even within one tissue, cells of different origin have qualitative and quantitative differences in cell organelle composition. Chemical Composition of Cells Each cellular compartment has an aqueous fluid or matrix that contains various ions, small molecular weight organic molecules, different proteins, and nucleic acids.  Localization of specific macromolecules, such as enzymes, has been

 

  

Page 14

Figure 1.10  Major chemical constituents of blood plasma  and cell fluid.  Height of left half of each column indicates  total concentration of cations; that of right half,  concentrations of anions. Both are expressed in  milliequivalents per liter (meq L–1) of fluid. Note that  chloride and sodium values in  cell fluid are questioned.  It is probable that, at least in muscle, the cytosol  contains some sodium but no chloride.  Adapted from Gregersen, M. I. In: P. Bard (Ed.),  Medical Physiology, 11th ed. St Louis, MO:  Mosby, 1961, p. 307.

determined but the exact ionic composition of the matrix of organelles is still uncertain. Each has a distinctly different ionic composition and pH. The overall ionic  composition of intracellular fluid, considered to represent the cytosol primarily, compared to blood plasma is presented in Figure 1.10. Na+ is the major extracellular  cation, with a concentration of ~140 meq L–1 (mM); very little Na+ is present in intracellular fluid. K+ is the major intracellular cation. Mg2+ is present in both extra­  and intracellular compartments at concentrations much lower than Na+ and K+. The major extracellular anions are Cl– and HCO3– with lower amounts of phosphate and  sulfate. Most proteins have a negative charge at pH 7.4 (Chapter 2), being anions at the pH of tissue fluids. Major intracellular anions are inorganic phosphate, organic phosphates, and  proteins. Other inorganic and organic anions and cations are present in concentrations well below the milliequivalent per liter (millimolar) level. Except for very small differences created by  membranes and leading to development of membrane potentials, the total anion concentration equals the total cation concentration in the different fluids.

Intracellular concentrations of most small molecular weight organic molecules, such as sugars, organic acids, amino acids, and phosphorylated intermediates, are in the  range of 0.01–1.0 mM but can have significantly lower concentrations. Coenzymes, organic molecules required for activity of some enzymes, are in the same range of  concentration. Substrates for enzymes are present in relatively low concentration in contrast to inorganic ions, but localization in a

 

  

Page 15

specific organelle or cellular microenvironment can increase their concentrations significantly. It is not very meaningful to determine the molar concentration of individual proteins in cells. In many cases they are localized with specific structures or in combination  with other proteins to create a functional unit. It is in a restricted compartment that individual proteins carry out their role, whether structural, catalytic, or regulatory. 1.4—  Functional Role of Subcellular Organelles and Membrane Systems The subcellular localization of various metabolic pathways will be described throughout this book. In some cases an entire pathway is located in a single compartment  but many are divided between two locations, with the intermediates in the pathway moving or being translocated from one compartment to another. In general,  organelles have very specific functions and the enzymatic activities involved are used to identify them during isolation. The following describes briefly some major roles of eukaryotic cell structures to indicate the complexity and organization of cells. A summary of functions and division  of labor within eukaryotic cells is presented in Table 1.6 and the structures are presented in Figure 1.9. TABLE 1.6 Summary of Eukaryotic Cell Compartments and Their Major Functions Compartment

Major Functions

Plasma membrane

Transport of ions and molecules

 

Recognition

 

Receptors for small and large molecules

 

Cell morphology and movement

Nucleus

DNA synthesis and repair

 

RNA synthesis

Nucleolus

RNA processing and ribosome synthesis

Endoplasmic reticulum

Membrane synthesis

 

Synthesis of proteins and lipids for some organelles  and for export

 

Lipid synthesis

 

Detoxication reactions

Golgi apparatus

Modification and sorting of proteins for  incorporation into organelles and for export

 

Export of proteins

Mitochondria

Energy conservation

 

Cellular respiration

 

Oxidation of carbohydrates and lipids

 

Urea and heme synthesis

Lysosomes

Cellular digestion: hydrolysis of proteins,  carbohydrates, lipids, and nucleic acids

Peroxisomes

Oxidative reactions involving O2

 

Utilization of H2O2

Microtubules and microfilaments

Cell cytoskeleton

 

Cell morphology

 

Cell motility

 

Intracellular movements

Cytosol

Metabolism of carbohydrates, lipids, amino acids,  and nucleotides

 

Protein synthesis

   

  

Page 16

CLINICAL CORRELATION 1.2  Mitochondrial Diseases: Luft's Disease A disease specifically involving mitochondrial energy transduction was first reported in  1962. A 30­year­old patient was described with general weakness, excessive  perspiration, a high caloric intake without increase in body weight, and an excessively  elevated basal metabolic rate (a measure of oxygen utilization). It was demonstrated that  the patient had a defect in the mechanism that controls mitochondrial oxygen utilization  (see Chapter 6). The condition is referred to as Luft's disease. Since that time, over 100  mitochondrial­based diseases have been identified, including those involving a variety of  enzymes and transport systems required for the proper maintenance and control of energy  conservation. Many involve skeletal muscle and the central nervous system. Replication of  mitochondria depends on the mitochondrial DNA (mtDNA) and inheritance of  mitochondria is by maternal transmission. Mutations of mtDNA as well as nuclear DNA  lead to genetic diseases. Mitochondrial damage may also occur due to free­radical  (superoxides) formation which can damage mtDNA. Thus age­related degenerative  diseases, such as Parkinson's and Alzheimer's, and cardiomyopathies may have a  component of mitochondrial damage. For details of specific diseases see Clin. Corr. 13.4  and 14.6. Luft, R. The development of mitochondrial medicine. Proc. Natl. Acad. Sci. USA  91:8731,1994.   Plasma Membrane Is the Limiting Boundary of a Cell The plasma membrane of every cell has a unique role in maintenance of that cell's integrity. One surface is in contact with a variable external environment and the  other with a relatively constant environment provided by the cell's cytoplasm. As will be discussed in Chapter 5, the two sides of the plasma membrane, and all  intracellular membranes, have different chemical compositions and functions. A major role of the plasma membrane is to permit entrance of some substances but  exclude many others. With cytoskeletal elements, the plasma membrane is involved in cell shape and movements. Through this membrane cells communicate; the  membrane contains many specific receptor sites for chemical signals, such as hormones (Chapter 20), released by other cells. The inner surface of plasma membranes  is the site for attachment of some enzymes involved in various metabolic pathways. Plasma membranes from a variety of cells have been isolated and studied  extensively; details of their structure and biochemistry and those of other membranes are presented in Chapter 5. Nucleus Is Site of DNA and RNA Synthesis Early microscopists divided the interior of cells into a nucleus, the largest membrane­bound compartment, and the cytoplasm. The nucleus is surrounded by two  membranes, termed the nuclear envelope, with the outer membrane being continuous with membranes of the endoplasmic reticulum. The nuclear envelope has  numerous pores about 90 Å in diameter that permit flow of all but the largest molecules between nuclear matrix and cytoplasm. The nucleus contains a  subcompartment, seen clearly in electron micrographs, the nucleolus. The vast amount of cellular deoxyribonucleic acid (DNA) is located in the nucleus as a DNA– protein complex, chromatin, that is organized into chromosomes. DNA is the repository of genetic information and the importance of the nucleus in cell division and  for controlling phenotypic expression of genetic information is well established. Biochemical reactions in the nucleus are replication of DNA during mitosis, repair of  DNA following damage (Chapter 15), and transcription of the information stored in DNA into a form that can be translated into cell proteins (Chapter 16).  Transcription of DNA involves synthesis of ribonucleic acid (RNA) that is processed into a variety of forms following synthesis. Part of this processing occurs in the  nucleolus, which is very rich in RNA. Endoplasmic Reticulum Has a Role in Many Synthetic Pathways The cytoplasm of most eukaryotic cells contains a network of interconnecting membranes that enclose channels, cisternae, that thread from the perinuclear envelope  to the plasma membrane. This extensive subcellular structure, termed endoplasmic reticulum, consists of membranes with a rough appearance in some areas and  smooth in other places. The rough appearance is due to the presence of ribonucleoprotein particles, that is, ribosomes, attached on the cytosolic side of the  membrane. Smooth endoplasmic reticulum does not contain bound ribosomes. During cell fractionation the endoplasmic reticulum network is disrupted, with the  membrane resealing into small vesicles called microsomes that can be isolated by differential centrifugation. Microsomes per se do not occur in cells. A major function of ribosomes on rough endoplasmic reticulum is biosynthesis of proteins for export to the outside of the cell and proteins for incorporation into  cellular organelles such as the endoplasmic reticulum, Golgi apparatus, plasma membrane, and lysosomes. Smooth endoplasmic reticulum is involved in membrane  lipid synthesis and contains an important class

 

  

Page 17

of enzymes termed cytochromes P450 that catalyze hydroxylation of a variety of endogenous and exogenous compounds. These enzymes are important in  biosynthesis of steroid hormones and removal of toxic substances (see Chapter 23). Endoplasmic reticulum with the Golgi apparatus has a role in formation of other  cellular organelles such as lysosomes and peroxisomes. The Golgi Apparatus Is Involved in Sequestering of Proteins The Golgi apparatus is a network of flattened smooth membranes and vesicles responsible for the secretion to the external environment of a variety of proteins  synthesized on the endoplasmic reticulum. Golgi membranes catalyze the transfer of carbohydrate and lipid precursors to proteins to form glycoproteins and  lipoproteins and is a major site of new membrane formation. Membrane vesicles are formed in the Golgi apparatus in which various proteins and enzymes are  encapsulated to be secreted from the cell after an appropriate signal. Digestive enzymes synthesized by the pancreas are stored in intracellular vesicles formed by the  Golgi apparatus and released when needed in the digestive process (see p. 1059). The role in membrane synthesis also includes the formation of intracellular organelles  such as lysosomes and peroxisomes. TABLE 1.7 Representative Lysosomal Enzymes and Their  Substrates Type of Substrate and Enzyme

Specific Substrate

POLYSACCHARIDE­

 

HYDROLYZING

 

ENZYMES

 

a­Glucosidase

Glycogen

a­Flucosidase

Membrane fucose

 

Galactosides

b­Galactosidase

Mannosides

a­Mannosidase

Glucuronides

b ­Glucuronidase Hyaluronidase

Hyaluronic acid and  chondroitin sulfates

Arylsulfatase

Organic sulfates

Lysozyme

Bacterial cell walls

PROTEIN­HYDROLYZING

 

ENZYMES

 

Cathepsins

Proteins

Collagenase

Collagen

Elastase

Elastin

Peptidases

Peptides

NUCLEIC ACID­

 

HYDROLYZING

 

ENZYMES

 

Ribonuclease Deoxyribonuclease

RNA DNA

LIPID­HYDROLYZING

 

ENZYMES

 

Lipases

Triglyceride and  cholesterol esters

Esterase

Fatty acid esters

Phospholipase PHOSPHATASES

Phospholipids

 

Phosphatase

Phospho­  monoesters

Phosphodiesterase

Phosphodiesters

SULFATASES

 

Heparan sulfate

 

Dermatan sulfate

 

  Mitochondria Supply Most Cell Needs for ATP Mitochondria appear as spheres, rods, or filamentous bodies that are usually about 0.5–1 mm in diameter and up to 7 mm in length. The internal matrix, the mitosol,  is surrounded by two membranes, distinctively different in appearance and biochemical function. The inner membrane convolutes into the matrix to form cristae and  contains numerous small spheres attached by stalks on the inner surface. Outer and inner membranes contain different enzymes. The components of the respiratory  chain and the mechanism for ATP synthesis are part of the inner membrane and are described in detail in Chapter 6. Major metabolic pathways involved in oxidation  of carbohydrates, lipids, and amino acids, and parts of special biosynthetic pathways involving urea and heme synthesis are located in the mitosol. The outer membrane  is relatively permeable but the inner membrane is highly selective and contains a variety of transmembrane transport systems. Mitochondria contain a specific DNA, with genetic information for some of the mitochondrial proteins, and the biochemical equipment for limited protein synthesis. The  presence of this biosynthetic capacity indicates the unique role that mitochondria have in their own destiny. See Clin. Corr. 1.2 for descriptions of diseases attributed to  deficits in mitochondrial function. Lysosomes Are Required for Intracellular Digestion Intracellular digestion of a variety of substances occurs inside structures designated as lysosomes. They have a single limiting membrane and maintain a pH lower in  the lysosomal matrix than that of the cytosol. Encapsulated in lysosomes is a group of glycoprotein enzymes—hydrolases—that catalyze hydrolytic cleavage of carbon­ oxygen, carbon–nitrogen, carbon–sulfur, and oxygen–phosphorus bonds in proteins, lipids, carbohydrates, and nucleic acids. A partial list of lysosomal enzymes is  presented in Table 1.7. As in gastrointestinal digestion, lysosomal enzymes split complex molecules into simple low molecular weight compounds that can be utilized by  metabolic pathways of the cell. Enzymes of the lysosome are characterized by being most active when the pH of the medium is acidic, that is, pH 5 and below. The  relationship between pH and enzyme activity is discussed in Chapter 4. The pH of the cytosol is close to pH 7.0 and lysosomal enzymes have little activity at this pH.

 

  

Page 18

CLINICAL CORRELATION 1.3  Lysosomal Enzymes and Gout Catabolism of purines, nitrogen­containing heterocyclic compounds found in nucleic  acids, leads to formation of uric acid, which is excreted in the urine (see Chapter 12 for  details). Gout is an abnormality in which there is excessive uric acid production with an  increase in uric acid in blood and deposition of urate crystals in joints. The consequences  are clinical manifestations in the joint, particularly the big toe, including inflammation, pain,  swelling, and increased warmth. Uric acid is not very soluble and some of the clinical  symptoms of gout can be attributed to damage done by urate crystals. Crystals are  phagocytosed by cells in the joint and accumulate in digestive vacuoles that contain  lysosomal enzymes. Crystals cause physical damage to the vacuoles, releasing lysosomal  enzymes into the cytosol. Even though the pH optima of lysosomal enzymes are lower  than the pH of the cytosol, they have some hydrolytic activity at the higher pH. This  activity causes digestion of cellular components, release of substances from the cell and  autolysis. Weissmann, G. Crystals, lysosomes and gout. Adv. Intern. Med. 19:239, 1974; and  Burt, H. M., Kalkman, P. H., and Mauldin, D. Membranolytic effects of crystalline  monosodium urate monohydrate. J. Rheumatol. 10:440, 1983.   The enzyme content of lysosomes in different tissues varies and depends on specific needs of individual tissues. The lysosomal membrane is impermeable to both small  and large molecules and specific protein mediators in the membrane are necessary for translocation of substances. Carefully isolated lysosomes do not catalyze  hydrolysis of substrates until this membrane is disrupted. The activities of lysosomal enzymes are termed ''latent." Membrane disruption in situ can lead to cellular  digestion, and various pathological conditions have been attributed to release of lysosomal enzymes, including arthritis, allergic responses, several muscle diseases, and  drug­induced tissue destruction (see Clin. Corr. 1.3). Lysosomes are involved in normal digestion of intra­ and extracellular substances that must be removed by a cell. Through endocytosis, external material is taken into  cells and encapsulated in membrane­bound vesicles (Figure 1.11). The plasma membrane invaginates around formed foreign substances, such as microorganisms, by  phagocytosis and takes up extracellular fluid containing suspended material by pinocytosis. Vesicles containing external material fuse with lysosomes to form  organelles that contain the materials to be digested and enzymes capable of carrying out the digestion. These vacuoles are identified microscopically by their size and  often by the presence of partially formed structures in the process of being digested. Lysosomes in which the

Figure 1.11  Diagrammatic representation of the role of lysosomes in intracellular digestion of  substances internalized by phagocytosis (heterophagy)  and of cellular components (autophagy).  In both processes substances to be digested are enclosed in a membrane vesicle,  followed by fusing with a primary lysosome to form a secondary lysosome.

 

  

Page 19

enzymes are not as yet involved in the digestive process are termed primary lysosomes, whereas those in which digestion of material is under way are secondary  lysosomes or digestive vacuoles that will vary in size and appearance. CLINICAL CORRELATION 1.4  Lysosomal Acid Lipase Deficiency Two phenotypic forms of a genetic deficiency of lysosomal acid lipase are known.  Wolman's disease occurs in infants and is usually fatal by age 1, while cholesterol ester­ storage disease usually is diagnosed in adulthood. Both are autosomal recessive  disorders. There is deposition of triacylglycerols and cholesterol esters in tissues,  particularly the liver. In the latter disease there is early onset of severe atherosclerosis.  Acid lipase catalyzes hydrolysis of mono­, di­, and triglycerols as well as cholesterol  esters. It is a critical enzyme in cholesterol metabolism, serving to make available free  cholesterol for cell needs. Hegele, R. A., Little, J. A., Vezina, C., et al., Hepatic lipase deficiency: clinical,  biochemical, and molecular genetic characteristics. Atherosclerosis and Thrombosis  13:720, 1993.   Cell constituents are synthesized and degraded continuously, and lysosomes function in digesting this cellular debris. The dynamic synthesis and degradation includes  proteins and nucleic acids, as well as structures such as mitochondria and endoplasmic reticulum. During this normal self­digestion process, that is, autolysis, cell  substances are encapsulated within a membrane vesicle that fuses with a lysosome to complete the degradation. The overall process is termed autophagy and is also  represented in Figure 1.11. Products of lysosomal digestion diffuse across lysosomal membranes and are reutilized by the cell. Indigestible material accumulates in vesicles referred to as residual  bodies, whose contents are removed from the cell by exocytosis. In some cases, residual bodies that contain a high concentration of lipid persist for long periods of  time. Lipid is oxidized and a pigmented substance, which is chemically heterogeneous and contains polyunsaturated fatty acids and proteins, accumulates in the cell.  This material, lipofuscin, is also called the "age pigment" or "wear and tear pigment" because it accumulates in cells of older individuals. It occurs in all cells but  particularly in neurons and muscle cells and has been implicated in the aging process. Under controlled conditions lysosomal enzymes are secreted from the cell for the digestion of extracellular material; an extracellular role for some lysosomal enzymes  has been demonstrated in connective tissue and prostate gland and in the process of embryogenesis. Thus they have a role in programmed cell death or apoptosis. Absence of specific lysosomal enzymes occurs in a number of genetic diseases in which there is accumulation in the cell of specific cellular components that cannot be  digested. Lysosomes of affected cells become enlarged with undigested material, which interferes with normal cell processes. Lysosomal storage diseases are  discussed in Chapter 10 (see p. 427); see Clin. Corr. 1.4 for a description of a deficiency of lysosomal lipase. Peroxisomes Contain Oxidative Enzymes Involving Hydrogen Peroxide Most eukaryotic cells of mammalian origin and those of protozoa and plants have organelles, designated peroxisomes or microbodies, which contain enzymes that  either produce or utilize hydrogen peroxide (H2O2). They are small (0.3–1.5 mm in diameter), spherical or oval in shape, with a granular matrix and in some cases a  crystalline inclusion termed a nucleoid. Peroxisomes contain enzymes that oxidize D­amino acids, uric acid, and various 2­hydroxy acids using molecular O2 with  formation of H2O2. Catalase, an enzyme present in peroxisomes, catalyzes the conversion of H2O2 to water and oxygen and oxidation by H2O2 of various compounds  (Figure 1.12). By having both peroxide­producing and peroxide­utilizing enzymes in one compartment, cells protect themselves from the toxicity of H2O2. Peroxisomes also contain enzymes involved in lipid metabolism, particularly oxidation of very long­chain fatty acids, and synthesis of glycerolipids and glycerol ether  lipids (plasmalogens) (see Chapter 10). See Clin. Corr. 1.5 for a discussion of Zellweger syndrome in which there is an absence of peroxisomes. Peroxisomes of different tissues contain different complements of enzymes, and the peroxisome content of cells can vary depending on cellular conditions.

Figure 1.12  Reactions catalyzed by catalase.

Cytoskeleton Organizes the Intracellular Contents Eukaryotic cells contain microtubules and actin filaments (microfilaments) as parts of the cytoskeletal network. The cytoskeleton has a role in maintenance

 

  

Page 20

of cellular morphology, intracellular transport, cell motility, and cell division. Microtubules, multimers of the protein tubulin, can be rapidly assembled and  disassembled depending on the cell's needs. Two very important cellular filaments, actin and myosin, occur in striated muscle and are responsible for muscular  contraction (see Chapter 22). Three mechanochemical proteins—myosin, dynein, and kinesin—convert chemical energy into mechanical energy for movement of  cellular components. These molecular motors are associated with the cytoskeleton; the actual mechanism for the energy conversion, however, has not been defined  completely. Dynein is involved in ciliary and flagellar movement, whereas kinesin is a driving force for the movement of vesicles and organelles along microtubules. CLINICAL CORRELATION 1.5  Zellweger Syndrome and the Absence of Functional Peroxisomes Zellweger syndrome is a rare, autosomal recessive disease characterized by abnormalities  of the liver, kidney, brain, and skeletal system. It usually results in death by age 6 months.  A number of seemingly unrelated biochemical abnormalities have been described  including decreased levels of glycerol­ether lipids (plasmalogens) and increased levels of  very long­chain fatty acids (C­24 and C­26) and cholestanoic acid derivatives  (precursors of bile acids). These abnormalities are due to the absence of functional  peroxisomes in the afflicted children. Peroxisomes are responsible for synthesis of  glycerol ethers, for shortening very long­chain fatty acids so that mitochondria can  completely oxidize them, and for oxidation of the side chain of cholesterol needed for bile  acid synthesis. Evidence indicates that there is a defect in the transport of peroxisomal  enzymes between the cytosol and the interior of peroxisomes during synthesis. Cells of  afflicted individuals contain empty ghosts of peroxisomes. The disease can be diagnosed  prenatally by assaying amniotic fluid cells for peroxisomal enzymes or analyzing the fatty  acids in the fluid. Datta, N. S., Wilson, G. N., and Hajra, A. K. Deficiency of enzymes catalyzing the  biosynthesis of glycerol­ether lipids in Zellweger syndrome. N. Engl. J. Med. 311:1080,  1984; Moser, A. E., Singh, I., Brown, F. R., Solish, G. I., Kelley, R. I., Benke, P. J.,  and Moser, H. W. The cerebrohepatorenal (Zellweger) syndrome. Increased levels and  impaired degradation of very long chain fatty acids and their use for prenatal diagnosis. N.  Engl. J. Med. 310:1141, 1984; and Wanders, R. J., Schutgens, R.B., and Barth, P. G.  Peroxisomal disorders: a review. J. Neuropathol. Exp. Neurol. 54:726, 1995.   Cytosol Contains Soluble Cellular Components The least complex in structure, but not in chemistry, is the organelle­free cell sap, or cytosol. It is here that many of the chemical reactions of metabolism occur and  where substrates and cofactors interact with various enzymes. Although there is no apparent structure to the cytosol, the high protein content precludes it from being a  truly homogeneous mixture of soluble components. Many reactions are localized in selected areas where substrate availability is more favorable. The actual  physicochemical state of the cytosol is poorly understood. A major role of the cytosol is to support synthesis of proteins on the rough endoplasmic reticulum by  supplying cofactors and energy. The cytosol also contains free ribosomes, often in a polysome form, for synthesis of intracellular proteins. Studies with isolated cytosol suggest that many reactions are catalyzed by soluble enzymes, but in the intact cell some of these enzymes may be loosely attached to one  of the many membrane structures or to cytoskeletal components and are readily released upon cell disruption. Conclusion A eukaryotic cell is a complex structure whose purpose is to replicate itself when necessary, maintain an intracellular environment to permit a myriad of complex  reactions to occur as efficiently as possible, and to protect itself from the hazards of its surrounding environment. Cells of multicellular organisms also participate in  maintaining the well­being of the whole organism by exerting influences on each other to maintain all tissue and cellular activities in balance. Thus, as we proceed to  study the separate chemical components and activities of cells in subsequent chapters, it is important to keep in mind the concurrent and surrounding activities,  constraints, and influences of the environment. Only by bringing together all the parts and activities of a cell, that is, reassembling the puzzle, will we appreciate the  wonder of living cells. Bibliography Water and Electrolytes Dick, D. A. T. Cell Water. Washington, DC: Butterworths, 1966. Eisenberg, D., and Kauzmann, W. The Structures and Properties of Water. Fairlawn, NJ: Oxford University Press, 1969. Morris, J. G. A Biologist's Physical Chemistry. Reading, MA: Addison­Wesley, 1968. Stillinger, F. H. Water revisited. Science 209:451, 1980. Westof, E., Water and Biological Macromolecules. Boca Raton, FL: CRC Press, 1993. Cell Structure Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. Molecular Biology of the Cell. New York: Garland, 1989. Becker, W. M., and Deamer, D. W. The World of the Living Cell, 2nd. ed. Redwood City, CA: Benjamin, 1991. DeDuve, C. Guided Tour of the Living Cell, Vols. 1 and 2. New York: Scientific American Books, 1984. Dingle, J. T., Dean, R. T., and Sly, W. S. (Eds.). Lysosomes in Biology and Pathology. Amsterdam: Elsevier (a serial publication covering all aspects of lysosomes).

 

  

Page 21

Fawcett, D. W. The Cell. Philadelphia: Saunders, 1981. Holtzman, E., and Novikoff, A. B. Cells and Organelles, 3rd ed. New York: Holt, Rinehart & Winston, 1984. Porter, K. R., and Bonneville, M. A. Fine Structure of Cells and Tissues. Philadelphia: Lea & Febiger, 1972. Vale, R. D. Intracellular transport using microtubule­based motors. Annu. Rev. Cell Biol. 3:347, 1987. Cell Organelles Attardi, G., and Chomyn, A. Mitochondrial Biogenesis and Genetics. San Diego, CA: Academic Press, 1995. Holtzman, E. Lysosomes. New York: Plenum Press, 1989. Latruffe, N., and Bugaut, M. Peroxisomes. New York: Springer­Verlag, 1994. Pavelka, M. Functional Morphology of the Golgi Apparatus. New York: Springer­Verlag, 1987. Preston, T. M., King, C. A., and Hyams, J. S. The Cytoskeleton and Cell Motility. New York: Chapman and Hall, 1990. Strauss, P. R., and Wilson, S. H. The Eukaryotic Nucleus: Molecular Biochemistry and Macromolecular Assemblies. Caldwell, NJ: Telford Press, 1990. Tyler, D. D. Mitochondrion in Health and Disease. New York: VCH, 1992. Tzagoloff, A. Mitochondria. New York: Plenum Press, 1982. Questions J. Baggot and C. N. Angstadt 1. Prokaryotic cells, but not eukaryotic cells, have: A. endoplasmic reticulum. B. histones. C. nucleoid. D. a nucleus. E. a plasma membrane. 2. Factors responsible for the polarity of the water molecule include: A. the similarity in electron affinity of hydrogen and oxygen. B. the tetrahedral structure of liquid water. C. the magnitude of the H–O–H bond angle. D. the ability of water to hydrogen bond to various chemical structures. E. the difference in bond strength between hydrogen bonds and covalent bonds. 3. Hydrogen bonds can be expected to form only between electronegative atoms such as oxygen or nitrogen and a hydrogen atom bonded to: A. carbon. B. an electronegative atom. C. hydrogen. D. iodine. E. sulfur. 4. Which of the following is least likely to be soluble in water? A. nonpolar compound B. weakly polar compound C. strongly polar compound D. weak electrolyte E. strong electrolyte 5. Which of the following is most likely to be partly associated in weak aqueous solution? A. alcohol B. lactic acid C. potassium sulfate (K2SO4) D. sodium chloride (NaCl) E. sodium lactate 6. The ion product of water: A. is independent of temperature. B. has a numerical value of 10–14 at 25°C. C. is the equilibrium constant for the reaction  D. requires that [H+] and [OH–] always be identical. E. is an approximation that fails to take into account the presence of the hydronium ion, H3O+. 7. Which of the following is both a Brønsted acid and a Brønsted base in water? A. H2PO4– B. H2CO3 C. NH3 D. NH4+ E. Cl– Refer to the following information for Questions 8 and 9. A. pyruvic acid

pK  = 2.50

B. acetoacetic acid

pK  = 3.6

C. lactic acid

pK  = 3.86

D. b ­hydroxybutyric acid

pK  = 4.7

E. propionic acid

pK  = 4.86

  8. Which weak acid will be 91% neutralized at pH 4.86? 9. Assuming that the sum of [weak acid] + [conjugate base] is identical for buffer systems based on the acids listed above, which has the greatest buffer capacity at pH  4.86? 10. All of the following subcellular structures can be isolated essentially intact EXCEPT: A. endoplasmic reticulum. B. lysosomes. C. mitochondria. D. nuclei. E. peroxisomes. 11. Biological membranes are associated with all of the following EXCEPT: A. prevent free diffusion of ionic solutes. B. release of proteins when damaged. C. contain specific systems for the transport of uncharged molecules. D. sites for biochemical reactions. E. proteins and nucleic acids cross freely. 12. Mitochondria are associated with all of the following EXCEPT: A. ATP synthesis. B. DNA synthesis. C. protein synthesis. D. hydrolysis of various macromolecules at low pH. E. two different membranes.

 

  

Page 22

13. Analysis of the composition of the major fluid compartments of the body shows that: A. the major blood plasma cation is K+. B. the major cell fluid cation is Na+. C. one of the major intracellular anions is Cl–. D. one of the major intracellular anions is phosphate. E. plasma and the cell fluid are all very similar in ionic composition. Refer to the following for Questions 14–17. A. peroxisome B. nucleus C. cytoskeleton D. endoplasmic reticulum E. Golgi apparatus 14. Consists of microtubules and actin fibers. 15. Oxidizes very long­chain fatty acids. 16. Connected to the plasma membrane by a network of membranous channels. 17. Transfers carbohydrate precursors to proteins during glycoprotein synthesis. Answers 1. C Prokaryotic DNA is organized into a structure that also contains RNA and protein, called nucleoid. A, B, and D are found in eukaryotic cells, and E is an element  of both prokaryotic and eukaryotic cells (p. 2). 2. C Water is a polar molecule because the bonding electrons are attracted more strongly to oxygen than to hydrogen. The bond angle gives rise to asymmetry of the  charge distribution; if water were linear, it would not be a dipole (p. 4). A: Hydrogen and oxygen have very different electron affinity. B and D are consequences of  water's structure, not factors responsible for it. 3. B Only hydrogen atoms bonded to one of the electronegative elements (O, N, F) can form hydrogen bonds (p. 5). A hydrogen atom participating in hydrogen  bonding must have an electronegative element on both sides of it. 4. A In general, compounds that interact with the water dipoles are more soluble than those that do not. Thus ionized compounds and polar compounds tend to be  soluble. Nonpolar compounds prefer to interact with one another rather than with polar solvents such as water (p. 5). 5. B Lactic acid is a weak acid, and weak acids dissociate only partially in aqueous solution (p. 6) A: Alcohol is fully associated. C–E: These are salts and are  considered to be fully dissociated under physiological conditions, although at high concentration some association occurs. 6. B The constant is a function of temperature and is numerically equal to the equilibrium constant for the dissociation of water divided by the molar concentration of  water (p. 6). D: [H+] = [OH–] in pure water, but not in solutions of solutes that contribute H+ or OH–. 7. A H2PO4– can donate a proton to become HPO42–. It can also accept a proton to become H3PO4. B and D are Brønsted acids; C is a Brønsted base. The Cl– ion in  water is neither (p. 8). 8. C If weak acid is 91% neutralized, 91 parts are present as conjugate base and 9 parts remain as the weak acid. Thus the conjugate base/acid ratio is 10 : 1.  Substituting into the Henderson–Hasselbalch equation, 4.86 = pK + log (10/1), and solving for pH gives the answer (p. 9). 9. E The buffer capacity of any system is maximal at pH = pK (p. 10). Buffer concentration also affects buffer capacity, but in this case concentrations are equal. 10. A Gentle disruption of cells will not destroy B–E. The tube­like endoplasmic reticulum, however, is disrupted and forms small vesicles. These vesicles, not the  original structure from which they were derived, may be isolated (pp. 12, 16). 11. E (p. 17). 12. D This is a lysosomal function (p. 17). Mitochondrial properties are described on p. 17. 13. D Phosphate and protein are the major intracellular anions. A, B, and E: Plasma and cell fluid are strikingly different. The Na+ ion is the major cation of plasma. C:  Most chloride is extracellular (p. 14, Figure 1.10). 14. C (p. 19). 15. A Fatty acid oxidation occurs in the mitochondria, but the oxidation of very long­chain fatty acids involves the peroxisomes (p. 19). 16. B This describes only the nucleus (p. 16). 17. E Lipids, too, are attached covalently to certain proteins in the Golgi apparatus (p. 17).

 

  

Page 23

Chapter 2—  Proteins I:  Composition and Structure Richard M. Schultz and Michael N. Liebman

2.1 Functional Roles of Proteins in Humans

24

2.2 Amino Acid Composition of Proteins

25

Proteins Are Polymers of a ­Amino Acids

25

Common Amino Acids Have a General Structure

25

Side Chains Define Chemical Nature and Structures of Different Amino  Acids

25

Amino Acids Have an Asymmetric Center

28

Amino Acids Are Polymerized into Peptides and Proteins

28

Cystine Is a Derived Amino Acid

30

2.3 Charge and Chemical Properties of Amino Acids and Proteins Ionizable Groups of Amino Acids and Proteins Are Critical for Biological  Function

30

Ionic Form of an Amino Acid or Protein Can Be Determined at a Given  pH

32

Titration of a Monoamino Monocarboxylic Acid: Determination of the  Isoelectric pH

32

Titration of a Monoamino Dicarboxylic Acid

33

General Relationship between Charge Properties of Amino Acids and  Proteins and pH

34

Amino Acids and Proteins Can Be Separated Based on pI Values

35

Amino Acid Side Chains Have Polar or Apolar Properties

38

Amino Acids Undergo a Variety of Chemical Reactions

38

2.4 Primary Structure of Proteins

39

2.5 Higher Levels of Protein Organization

42

Proteins Have a Secondary Structure

43

a ­Helical Structure

43

b ­Structure

44

Supersecondary Structures

44

Proteins Fold into a Three­Dimensional Structure Called the Tertiary  Structure

44

Homologous Three­Dimensional Domain Structures Are Often Formed  from Common Arrangements of Secondary Structures

47

A Quaternary Structure Occurs When Several Polypeptide Chains Form  a Specific Noncovalent Association

48

2.6 Other Types of Proteins Fibrous Proteins Include Collagen, Elastin, a ­Keratin, and  Tropomyosin

Amino Acid Composition of Collagen

50

Amino Acid Sequence of Collagen

50

Structure of Collagen

52

Formation of Covalent Cross­links in Collagen

54

Elastin Is a Fibrous Protein with Allysine­Generated Cross­links

54

a ­Keratin and Tropomyosin

55

Lipoproteins Are Complexes of Lipids with Proteins

56

Glycoproteins Contain Covalently Bound Carbohydrate

60

Types of Carbohydrate–Protein Covalent Linkages

61 62

The Protein Folding Problem: A Possible Pathway

62

Chaperone Proteins May Assist the Protein Folding Process

63

Noncovalent Forces Lead to Protein Folding and Contribute to a  Protein's Stability

63

Hydrophobic Interaction Forces

64

Hydrogen Bonds

66

Electrostatic Interactions

66

Van der Waals–London Dispersion Forces

66 67

2.8 Dynamic Aspects of Protein Structure

68

2.9 Methods for Characterization, Purification, and Study of Protein  Structure and Organization

69

Separation of Proteins on Basis of Charge Capillary Electrophoresis

 

50

50

Denaturation of Proteins Leads to Loss of Native Structure

  

49

Distribution of Collagen in Humans

2.7 Folding of Proteins from Randomized to Unique Structures: Protein  Stability

 

30

69 70

Separation of Proteins Based on Molecular Mass or Size

71

Ultracentrifugation: Definition of Svedberg Coefficient

71

Molecular Exclusion Chromatography

72

Page 24

Polyacrylamide Gel Electrophoresis in the Presence of a Detergent

72

HPLC Chromatographic Techniques Separate Amino Acids, Peptides,  and Proteins

72

Affinity Chromatography

73

General Approach to Protein Purification

73

Determination of Amino Acid Composition of a Protein

74

Techniques to Determine Amino Acid Sequence of a Protein

74

X­Ray Diffraction Techniques Are Used to Determine the Three­ Dimensional Structure of Proteins

76

Various Spectroscopic Methods Are Employed in Evaluating Protein  Structure and Function

79

Ultraviolet Light Spectroscopy

79

Fluorescence Spectroscopy

79

Optical Rotatory Dispersion and Circular Dichroism Spectroscopy

80

Nuclear Magnetic Resonance

81

Bibliography

82

Questions and Answers Clinical Correlations

83  

2.1 Plasma Proteins in Diagnosis of Disease

37

2.2 Differences in Primary Structure of Insulins used in Treatment of  Diabetes Mellitus

41

2.3 A Nonconservative Mutation Occurs in Sickle Cell Anemia

42

2.4 Symptoms of Diseases of Abnormal Collagen Synthesis

50

2.5 Hyperlipidemias

56

2.6 Hypolipoproteinemias

59

2.7 Glycosylated Hemoglobin, HbA1c

62

2.8 Use of Amino Acid Analysis in Diagnosis of Disease

74

  2.1—  Functional Roles of Proteins in Humans Proteins perform a surprising variety of essential functions in mammalian organisms. These may be grouped into dynamic and structural. Dynamic functions include  transport, metabolic control, contraction, and catalysis of chemical transformations. In their structural functions, proteins provide the matrix for bone and connective  tissue, giving structure and form to the human organism. An important class of dynamic proteins are the enzymes. They catalyze chemical reactions, converting a substrate to a product at the enzyme's active site. Almost all of  the thousands of chemical reactions that occur in living organisms require a specific enzyme catalyst to ensure that reactions occur at a rate compatible with life. The  character of any cell is based on its particular chemistry, which is determined by its specific enzyme composition. Genetic traits are expressed through synthesis of  enzymes, which catalyze reactions that establish the phenotype. Many genetic diseases result from altered levels of enzyme production or specific alterations to their  amino acid sequence. Transport is another major function for proteins. Particular examples discussed in greater detail in this text are hemoglobin and myoglobin, which  transport oxygen in blood and in muscle, respectively. Transferrin transports iron in blood. Transport proteins bind and carry steroid hormones in blood from their site  of synthesis to their site of action. Many drugs and toxic compounds are transported bound to proteins. Proteins participate in contractile mechanisms. Myosin and  actin function in muscle contraction. Proteins have a protective role through a combination of dynamic functions. Immunoglobulins and interferon are proteins that protect the human against bacterial or  viral infection. Fibrin stops the loss of blood on injury to the vascular system. Many hormones are proteins or peptides. Protein hormones include insulin, thyrotropin, somatotropin (growth hormone), luteinizing hormone, and follicle­stimulating  hormone. Many diverse polypeptide hormones have a low molecular weight ( 250

170

256

36–38

160–300

Catalytic core

165–180

125

215

36–38

125

Other subunits

70, 50, 60

  48

  55

None

35, 47

Activities 3    5  Exonuclease

No

Yes

Yes

No

Yes

Processivity

Low

  High

  High

Low

High

Fidelity

High

  High

  High

Low

High

Source: Adapted from Kornberg, A., and Baker, T. A. DNA Replication, 2nd ed. New  York: Freeman, 1992. a With the exception of polymerase g, which is a mitochondrial enzyme, all other 

polymerases are located in the cell nucleus.

   

  

Page 627

endows polymerase   with very high processivity and is also involved in eukaryotic DNA excision repair (see p. 636). In an overall sense DNA polymerases operate at a high level of fidelity, which is required of their function as DNA replicating and repair enzymes. Escherichia coli  polymerases have an overall error rate in base incorporation of 10–7 to 10–8. The experimentally observed accuracy for DNA replication in E. coli, however, is  substantially higher, with errors made at the rate of only one for every 109 to 1010 nucleotides incorporated. The discrepancy in these numbers is accounted for by the  operation of a DNA repair system that removes mismatched bases that have escaped the scrutiny of the proofreading activity of the polymerases. This repair system,  known as the mismatch repair system, is examined on page 638. The necessity to maintain high fidelity in replication is probably also the reason why polymerases synthesize polynucleotides only in the 5    3  direction. If  polynucleotide chains could be elongated in the 3    5  direction, the hypothetical growing 5  terminus, rather than the incoming nucleotide, would carry a triphosphate  that is unsuitable for further elongation by the synthetic activity of the polymerase. 15.3—  Mutation and Repair of DNA Mutations Are Stable Changes in DNA Structure One of the fundamental requirements for a structure that serves as a permanent depository of genetic information is high stability. Such stability is essential, at least in  those parts that code for the genetic information. The structure of the DNA bases, however, is not totally exempt from gradual change. Normally, changes occur  infrequently and they affect very few bases. Chemical and irradiation­induced reactions modify the structure of some bases, disrupt phosphodiester bonds, and sever  strands. Extensive chemical changes of the bases occur spontaneously. Errors also occur during replication and DNA recombination, leading to incorporation of one or  more erroneous bases. In almost every instance, however, a few cycles of DNA replication are required before a modification in the structure of a base can lead to  irreversible damage. In effect, DNA polymerases must use the polynucleotide initially damaged as a template for the synthesis of a complementary strand for the initial  change to become permanent. As Figure 15.5 suggests, use of the damaged strand as template extends the damage from a change of a single base to a change of a  complete base pair and subsequent replication perpetuates the change. Other sources of permanent modifications of DNA include changes resulting from insertion to  deletion from a DNA of short or longer nucleotide sequences during the process of DNA recombination (see p. 661). Intercalation of certain planar organic ring  structures can also lead to insertion of nucleotides (see p. 631). Finally, deletions may occur as a result of chemical modification of the bases.

Figure 15.5  Mutation perpetuated by replication.   Mutations introduced on a DNA strand, such as the  replacement of a cytosine by a uracil resulting from  deamination of cytosine, extend to both strands when the  damaged strand is used as a template during replication.  In the first round of replication uracil selects adenine as   omplementary base. In the second round of replication uracil is  replaced by thymine. Similar events occur when the other  bases are altered.

 

  

Page 628

Irreversible alteration of a few DNA base pairs can cause drastic changes in the organism. These changes, referred to as mutations, may be hidden or visible, that is,  phenotypically silent or expressed. Therefore a mutation is defined as a stable change in the DNA structure of a gene, which may be expressed as a phenotypic  change in the organism. Mutations may be classified into two categories: base substitutions and frameshift mutations. Base substitutions include transitions,  substitutions of one purine–pyrimidine pair by another, and transversions, substitutions of a purine–pyrimidine pair by a pyrimidine–purine pair. Frameshift mutations,  which are the most radical, are the result of either the insertion of a new base pair or the deletion of a base pair or a block of base pairs from the DNA base sequence  of the gene. These changes are illustrated in Figure 15.6. Chemical Modification of Bases Irradiation and certain chemical compounds are recognized as among the main mutagens. The incorporation of erroneous bases by DNA polymerase can also lead to  mutations. Other mutations occur spontaneously. Bases in DNA are sensitive to the action of numerous chemicals including nitrous acid (HNO2), hydroxylamine  (NH2OH), and various alkylating agents such as dimethyl sulfate and N­methyl­N8­nitro­N­nitrosoguanidine. Chemical modifications of bases, brought about by these  reagents, are shown in Figure 15.7. Conversion of guanine to xanthine by nitrous acid has no effect on the hydrogen­bonding properties since xanthine, the new base, can pair with cytosine, the normal  partner of guanine. However, the conversion of either adenine to hypoxanthine or the change from cytosine to uracil disrupts the normal hydrogen bonding of the  double helix, because neither hypoxanthine nor uracil can form complementary pairs with the base present in the initial double helix (Figure 15.8). Subsequent  replication of the DNA extends and perpetuates these base changes (Figure 15.5). Alkylating agents may affect the structure of the bases as well as disrupt  phosphodiester bonds so as to lead to the fragmentation of the strands. In addition, certain alkylating agents can interact covalently with both strands, creating  interstrand bridges. DNA undergoes spontaneous changes as a result of various physical perturbations, such as thermal fluctuations or reactions with reactive forms of oxygen.  Spontaneous deamination of cytosine in human DNA occurs at a rate of

Figure 15.6  Mutations.  Mutations are classified as transition,  transversion, and frameshift. Bases undergoing  mutation are shown in color.  (a) Transition: A purine–pyrimidine base pair is  replaced by another. This mutation occurs  spontaneously or can be induced chemically by  such compounds as 5­bromouracil or nitrous acid.  (b) Transversion: A purine–pyrimidine base pair  is replaced by a pyrimidine–purine pair.  This mutation occurs spontaneously and  is common in humans. About one­half of the  mutations in hemoglobin are of this type.  (c) Frameshift: This mutation results from  insertion or deletion of a base pair.  Some insertions can be caused by mutagens  such as acridines, proflavin, and ethidium bromide.  Deletions are often caused by deaminating  agents. Alteration of bases by these agents  prevents pairing.

 

  

Page 629

Figure 15.7  Reactions of various mutagens.  (a) Deamination by nitrous acid (HNO ) converts cytosine to uracil, adenine to  2

hypoxanthine, and guanine to xanthine.  (b) Reaction of bases with hydroxylamine (NH2OH) as illustrated by the action of this  reagent on cytosine.  (c) Alkylations of guanine by dimethyl sulfate (DMS). Formation of a quaternary  nitrogen destabilizes the deoxyriboside bond and releases deoxyribose. Among the  effective agents for methylation of bases are nitrosoguanidines such as  N­methyl­N8­nitro­N­nitrosoguanidine.

about 100 base pairs per genome per day and DNA depurination occurs at even higher rates of 5000 bases per genome per day (Figure 15.9) as a result of thermal  disruption of the N­glycosyl bonds of the bases. Some other changes that occur in DNA (as shown in Figure 15.10) can lead to either deletion of one or more base  pairs in the daughter DNA after DNA replication or to a base pair substitution. Radiation Damage Ultraviolet light, including sunlight, and X­ray irradiation are also effective means of producing mutations. Radiation energy absorbed by the DNA induces the  formation of minor amounts of the ionized forms of the bases. These ionized forms cannot pair with the normal partners of the base, but, instead, they engage in  atypical base pairing as shown in Figure 15.11. The presence of ionized base forms at the moment of DNA replication is therefore expected to increase the frequency  of mutation in the newly synthesized DNA strands. UV irradiation of DNA causes formation of dimers between adjacent pyrimidine bases. Activation of the ethylene  bond of these bases frequently leads to a photochemical

 

  

Page 630

dimerization of two adjacent pyrimidines, as shown in Figure 15.12. Thymine residues are particularly susceptible to this reaction, although cytosine dimers and  thymine–cytosine combinations are also produced.

Figure 15.8  Chemical modifications that alter  hydrogen­bonding properties of bases.  Hypoxanthine, obtained by deamination  of adenine, has different hydrogen­  bonding properties from adenine and  pairs with cytosine. Similarly, uracil  obtained from cytosine has a  different hydrogen­bonding  specificity than cytosine and pairs  with adenine. Alkylation of guanine  modifies hydrogen­bonding  properties of the base.

High­energy radiation (X­rays or gamma rays) brings about direct modifications in the structure of the bases. Intermediates produced by electron expulsion can be  rearranged, leading to the opening of the heterocyclic rings of the bases and the disruption of phosphodiester bonds. In the presence of oxygen additional reactions  take place, yielding a variety of oxidation products. DNA Polymerase Errors With the appropriate deoxyribonucleoside triphosphates, DNA polymerases function with a high degree of fidelity. Some mutations do occur during DNA replication,  but these changes are limited by the high synthetic fidelity of DNA polymerase and the "proofreading" exonuclease properties of this enzyme. The fidelity of DNA  replication is further enhanced postreplicatively by an excision repair process known as the mismatched repair system. This system recognizes and corrects  mismatches in newly replicated DNA by detecting distortions on the outside of the helix that are produced from poor fit between paired noncomplementary bases.  Clearly, accurate correction of mismatched bases requires that the mismatched repair system discriminate between preexisting

Figure 15.9  Spontaneous deamination of pyrimidines and depurination of polynucleotides.  DNA undergoes substantial structural modifications as a result of thermal  perturbations that include  (1) extensive hydrolysis of the N­glycosyl bonds that connect purines to the  deoxyribose residue and  (2) deamination of cytosine residues to uracil. In absence of repair mechanisms, these  changes would have disastrous consequences for cell survival because of  the high frequency of their occurrence.

 

  

Page 631

Figure 15.10  DNA sites subject to spontaneous chemical modifications.  Nucleotides are subject to various spontaneous chemical changes at sites indicated  by arrows including (1) hydrolytic attack, (2) oxidative damage, and (3) methylation.  The frequency and extent of chemical change vary from site to site.

and newly synthesized DNA strands. Such discrimination is feasible because certain adenine residues in DNA, which are part of a recurring GATC sequence, are  subject to methylation that occurs posttranscriptionally, but with some delay. Mismatched proofreading is carried out by a multienzyme complex that excises  mismatched nucleotides only from newly synthesized strands. The complex identifies these nucleotides by searching for unmethylated adenine residues in the GATC  sequences of each strand. The mechanism of mismatched repair is described later. DNA polymerases are unable to distinguish between the normal deoxyribonucleoside triphosphate substrates and other nucleotides with very similar structures, thus  leading to their incorporation and a mutation. Classic examples of such analogs are deoxyribonucleotides of 5­bromouracil (5­BrdU) and 2­aminopurine (2­AP) that  have been used experimentally for the introduction of mutations. Incorporation of 5­BrdU into DNA introduces, with a high frequency, a transition mutation in which a  pu­py pair is transformed to another pu­py. Specifically, 5­BrdU paired with A is changed to a C­G pair, which amounts to a TA   GC transition. The unusual  pairing properties of 5­BrdU appear to relate to the higher tendency of this base to be transformed to an ionized form, relative to T for which it is a substitute. This  occurs presumably because of the higher electronegative nature of the bromine atom in comparison to the corresponding methyl group in thymine.

Figure 15.11  Base pairing between the ionized  forms of the bases.  Adenine and cytosine are prone  to protonation especially at lower pH.  Also, an ionized form of thymine can be  generated by loss of a proton.  Reactions that give rise to ionized  forms of bases occur readily at near­  neutral pH, within certain nucleotide  sequence contexts. Whereas some of  the ionized complexes form with  Watson–Crick hydrogen bonding, as,  for instance, the T (ionized)–G pair,  other ionized bases form more unusual  types of H bonding. For example, the  A (ionized)–G(syn) base pair involves  H bonding between an A in the anti  position and a G in the syn configuration.

Stretching of the Double Helix Organic compounds characterized by planar aromatic ring structures of appropriate size and geometry can be inserted between base pairs in double­stranded DNA.  This process is referred to as intercalation. During intercalation neighboring base pairs in DNA are separated to allow for the insertion of the intercalating ring system,  causing an elongation of the double helix by stretching. In effect the double helix is locally unwound into a ladder­like structure in which the base pairs are transiently  arranged at 0.68 nm apart. This localized arrangement doubles the 0.34­nm distance characteristic of the double helix and generates sufficient space between base  pairs for the insertion of the intercalator. In effect, intercalation disrupts the continuity of the base sequences in DNA and the reading of the DNA template by the DNA  polymerase, producing a daughter strand with an additional base incorporated into DNA. The resulting mutation is referred to as a frameshift. Acridines, ethidium  bromide, and other intercalators are known to be effective frameshift mutagens (Figure 15.13). Clinical Correlation 15.1 discusses mutations and the etiology of  cancer.

 

  

Page 632

Figure 15.12  Dimerization of adjacent pyrimidines in irradiated DNA.  Thymine activated by absorption of UV light can react  with a second neighboring thymine and form a  thymine dimer.

Figure 15.13  Intercalation between base pairs of the  double helix.  (a) Insertion of planar ring system of intercalators  between two adjacent base pairs requires  stretching of the double helix  (b). During replication this stretching apparently  changes the frame used by DNA polymerase  for reading the sequence of nucleotides.  Consequently, newly synthesized DNA is  frameshifted.   (b­1) Original DNA helix;  (b­2) helix with intercalative binding of ligands.  Redrawn based on figure in Lippard, S. J.  Acct. Chem. Res. 11:211, 1978.  Copyright © 1978 by the American  Chemical Society.

 

  

Page 633

CLINICAL CORRELATION 15.1  Mutations and the Etiology of Cancer Considerable progress in understanding the etiology of cancer has been achieved in  recent years by the realization that long­term exposure to certain chemicals leads to  various forms of malignancy. It is now suggested that the great majority of cancers are  triggered by agents in the environment that modify underlying genetic predisposition  factors. Carcinogenic (cancer­causing) compounds are not only introduced into the environment  by the increasing use of new chemicals in industrial applications but are also present in the  form of natural products. For instance, the aflatoxins, produced by certain molds, and  benz[a]anthracene, present in cigarette smoke and charcoal­broiled foods, are  carcinogenic. Some carcinogens act directly, while others, such as benz[a]anthracene,  must undergo prior hydroxylation by arylhydroxylases, present mainly in the liver, before  their carcinogenic potential can be expressed.

The reactivity of many carcinogenic compounds toward guanine residues results in  modification of the guanine structure, usually by alkylation at the N­7 position and by  cleavage of the phosphodiester bond, events that upon replication lead to permanent  mutations. Chemical mutagens are generally carcinogenic and vice versa. Vulnerability of  DNA to alkylating agents and other chemicals underscores the concerns expressed by  many scientists about the ever­increasing exposure of our environment to new chemicals.  What is of concern is that the carcinogenic potential of new chemicals released into the  environment cannot be predicted with confidence even when they appear to be  chemically innocuous toward DNA. In the past, tests for carcinogenicity, that is, the ability of a substance to cause cancer,  required the use of many experimental animals treated with high doses of suspected  carcinogen over a long period of time. Such tests, which are time consuming as well as  expensive, are the only approach still available for testing carcinogenicity directly. A much  simpler and inexpensive indirect test for carcinogenicity is also available. This test, the  Ames Test, is based on the premise that carcinogenicity and mutagenicity are essentially  manifestations of the same underlying phenomenon—the structural modification of DNA.  The test measures the rate of mutation that bacteria undergo when exposed to chemicals  suspected to be carcinogens. A major criticism of this test is that the assumption of an equivalence between  mutagenicity and carcinogenicity is not always valid. Because of economic implications of  labeling a chemical with widespread use as a potential carcinogen, the scrutiny often  exercised in assessing the reliability of applicable tests for labeling a chemical as a  carcinogen is understandable. Certain exceptions notwithstanding, the great majority of  chemicals tested have shown that a good correlation exists between the tendency of a  chemical to produce bacterial mutations and animal cancer. Even the direct and very  costly tests for carcinogenicity have not completely escaped criticism. The reliability of  such tests has been questioned because of the relatively large doses of chemicals  employed, doses that are essential for shortening the long­term chemical exposure of the  animals to a practically manageable period of time. Another criticism of direct tests is that  they make projections from animals, usually rodents, to humans. This criticism has some  merit. During the past few years it has became apparent that rodents are less efficient than  humans at repairing certain types of damage in nontranscribed regions of their DNA.  Damage in nontranscribed DNA regions is more slowly repaired than damage within  transcribed genes, which have first priority for repair. Although damage in nontranscribed  DNA regions has few immediate consequences, it appears with time that this damage  leads to cancer. The relatively large doses of chemicals used for testing are likely to  exceed the capacity of rodent DNA repair systems, making the extrapolation of the  results obtained from rodents to humans unreliable. The enzymes that activate carcinogens are often members of the cytochrome P450 family  (Chapter 23) that can be induced by noncarcinogenic compounds such as ethanol; hence  alcohol can increase the potential risk of cancer development after exposure to  carcinogens. Ames, B., Dursto, W. E., Yamasaki, E., and Lee, F. D. Carcinogens are mutagens: a  simple test system combining liver homogenates for activation and bacteria for detection.  Proc. Natl. Acad. Sci. USA 70:2281, 1973; and Culotta, E., and Koshland, D. E. Jr.  DNA repair works its way to the top. Science 266:1926, 1994.   DNA Is Repaired Rather Than Degraded DNA is the only macromolecule that is repaired rather than degraded. The repair processes are very efficient with fewer than 1 out of 1000 accidental changes  resulting in mutations. The rest are corrected through various processes

 

  

Page 634

of DNA repair. Mutation rates can be estimated using two entirely different approaches, that is, from the frequency with which new mutants arise either in populations,  such as fruitflies, or in specific proteins in cells growing in tissue culture. These experiments provide estimates of mutation rates of 1 base pair change per 109 base  pairs for each cell per each generation. On this basis, for an average­sized protein, which contains about 1000 coding base pairs, a mutation may occur once in 106  cell generations. DNA repair is a high­priority process for maintaining cellular function. Germ cells must be protected against high rates of mutation to preserve the species, and somatic  mutation must be controlled in order to avoid uncontrolled cell growth and disease. Unchecked accumulation of damage can lead to accumulation of nonfunctional  proteins or unregulated growth characteristic of malignant cells. Commonly encountered DNA lesions are listed in Table 15.4. There are multiple DNA repair pathways and each specializes in a certain type of damage, although some repair pathways have a wider versatility than others.  Generally, repair mechanisms are applicable to both prokaryotic and eukaryotic DNA repair. Repairs may be carried out under rare circumstances as a direct reversal of the damage or, far more commonly, by the replacement of the damaged DNA section.  DNA repair depends on the existence of two complementary DNA strands except for postreplication repair of rare lesions and postreplication SOS repair.  Damage or imperfection on one DNA strand can be corrected since the complementary strand provides the necessary information for accurate repairs. Postreplication  repair is not a true repair mechanism but rather a stop­gap measure that allows for DNA replication to occur until damage can be repaired permanently.  Postreplication repair cannot use the complementary DNA strand for repairs because this strand is also altered by the replication that precedes the repair.  Postreplication repair depends, instead, on another process—DNA recombination. Recombination permits the use of homologous DNA strands, namely, DNA  strands with the same or almost the same sequence as the damaged strand, for carrying out the repair of the damaged DNA section. An intriguing feature of DNA  repair that has been appreciated recently is its apparent intimate coupling to other central processes in which DNA participates, such as recombination, transcription,  and control of the cell cycle. Enzymes involved in DNA repair participate in DNA replication, DNA recombination, and particularly DNA transcription. DNA  metabolism integrates important processes that are coordinated through the use of the same molecular tools to achieve different tasks. TABLE 15.4 DNA Lesions that Require Repair DNA Lesion

Cause

Missing base

Acid and heat remove purines (~104 purines  per day per cell in mammals)

Altered base

Ionizing radiation; alkylating agents

Incorrect base

Spontaneous deaminations: C   U, A  hypoxanthine

Deletion–insertion

Intercalating agents (e.g., acridine dyes)

Cyclobutyl dimer

UV irradiation

Strand breaks

Ionizing radiation; chemicals (bleomycin)

Cross­linking of strands

Psoralin derivatives (light­activated);  mitomycin C (antibiotic)

 

Source: From Kornberg, A. DNA Replication. San Francisco: Freeman,  1980, p. 608.

   

  

Page 635

Figure 15.14  Action of DNA ligase.  The enzyme catalyzes the joining of polynucleotide strands that are part of a  double­ stranded DNA. A single phosphodiester bond is formed between 3 ­OH  and 5 ­P ends of two strands. In E. coli cells, energy for formation of the  bond is derived from cleavage of the pyrophosphate bond of NAD+. In  eukaryotic cells and bacteriophage­infected cells, energy is provided by  hydrolysis of the a,b­pyrophosphate bond of ATP.

Excision Repair in E. coli Excision repair is catalyzed by different enzymatic systems tailored to specific types of damage. This repair mechanism is universal, occurring in all organisms  investigated. The mechanisms are characterized by four sequential steps: incision, excision, resynthesis, and ligation. Incision is the recognition step and is individualized  for the specific type of damage present. It is also the rate­controlling step in the process. During excision the damaged DNA section is excised, leaving a gap in the  DNA strand. In the resynthesis step the gap is filled by DNA polymerase I. This enzyme functions like DNA polymerase III in that it catalyzes the stepwise addition of  nucleotide triphosphates on a 3 ­OH generated by the preceding incision step. Polymerase I, however, differs from polymerase III in that it is less processive, tending  to dissociate from the DNA after incorporation of 10–12 nucleotides. At this stage the gap is reduced to the size of a single phosphodiester bond. Because of the  combined synthetic–nucleolytic action of polymerase I, the nick can move along the strand, undergoing repair until it is finally bridged during the ligation step by the  action of DNA ligase (Figure 15.14). The ligation step appears to be very similar for all types of excision repair.

Figure 15.15  Uracil DNA glycosylase repair of DNA.  Uracil DNA glycosylase removes uracil, formed  by accidental deamination of cytosine, by cutting  the glycosidic bond, leaving DNA with a missing  base. AP endonuclease subsequently cuts out the  sugar–phosphate remnant. Repair is completed by  DNA polymerase and ligase.

Base excision repair eliminates modified bases from DNA. The amino groups of cytosine, adenine, and guanine are susceptible to spontaneous elimination, and  various chemicals lead to modifications in the structures of purines, including methylation and ring opening. In addition, ring opening may result from exposure to  ionizing radiation. Bases that have been deaminated, methylated, or otherwise chemically modified are hydrolytically removed by enzymes referred to as DNA  glycosylases. Removal of deaminated cytosine (i.e., uracil) by the enzyme uracil DNA glycosylase is illustrated in Figure 15.15. This enzyme removes the damaged  cytosine, producing a deoxyribose residue with the base missing [apurinic–apyrimidinic (AP) site]. AP sites are also generated without the involvement of DNA  glycosylases, as in the case of spontaneous hydrolysis of purines (depurination) that occurs at very high rates in DNA. AP sites can also result from depyrimidination  but the greater stability of the purine–glycoside bond makes this reaction almost insignificant. Once an AP site has been created, the enzyme AP endonuclease nicks  the phosphodiester backbone at the depurinized site and excises the sugar–phosphate residue. The action of DNA polymerase I and ligase on this structure leads to  the restoration of the damaged strand. A second type of excision repair referred to as nucleotide excision repair is activated when DNA is damaged in a way that produces a ''bulky" lesion. This occurs  when DNA interacts with polycyclic aromatic hydrocarbons, such as benzo[a]pyrenes and dialkylbenzathracenes generated by smoking, thymine–psoralene adducts,  and guanine–cisplatin adducts formed by chemotherapeutic drugs. UV light­induced dimerization of adjacent pyrimidines also causes bulky lesions. Nucleotide excision  repair also corrects other lesions that do not distort the helix, such as the presence of methylated bases. Once the lesion has been located, an endonuclease activity  cleaves the modified strand on both sides of

 

  

Page 636

the distortion and the entire lesion is removed (Figure 15.16). Repair is initiated by recognition of the distortion of the DNA by an endonuclease system consisting of  the products of three E. coli genes uvrA, uvrB, and uvrC. A tetramer consisting of two UvrA and two UvrB proteins, which is formed on DNA during a series of  preincision steps, "melts" the DNA locally at the expense of ATP and locates the bulky lesion. The complex is subsequently subjected to incision at both sides of the  bulky lesion. First, UvrB makes a 3  incision and then UvrC makes a 5  incision, leading to the release of an oligonucleotide consisting of 12 or 13 residues that  includes the pyrimidine dimer. This nuclease activity, which is unique to DNA repair, has been christened excision nuclease or excinuclease to clearly distinguish it  from other endonucleases. For the remainder of the repair, E. coli makes use of the protein UvrD which, acting as a helicase, unwinds and releases the  oligonucleotide that was excised by UvrB and UvrC. The repair is completed by polymerase I and ligase.

Figure 15.16  Nucleotide excision repair in E. coli.  Nucleotide excision repair in E. coli and in  human DNA occurs in a series of analogous  steps. Initial damage in E. coli is recognized by  UvrA protein, which also serves as a  "molecular matchmaker" by recruiting, at the  damaged site, UvrB protein. UvrA binds to  the lesion, unwinds and kinks DNA. UvrA also  causes a conformational change in UvrB that  promotes strong binding of UrvB at the site  of the lesion. Subsequent dissociation of  UvrA from UvrB–DNA complex makes the  complex a target for UvrC. UvrB then makes a  3  cut that is followed by a 5  incision made by  UvrC. Helicase II (UvrD) releases the excised  oligonucleotide 12­mer and DNA polymerase  displaces UvrB and fills the excision gap  prior to ligation.  Redrawn based on figure in  Moran, L. A., Scrimgeour, K. G., Horton, H.  R., Ochs, R. S., and Rawn, J. D.  Biochemistry. Englewood Cliffs, NJ:  Neil Patterson/Prentice Hall, 1994.

Eukaryotic Excision Repair Excision repair in prokaryotes and eukaryotes is remarkably similar with the following distinctions. The exonuclease activity of human cells consists of a much larger  number of proteins (16–17 different polypeptides) as apposed to the four proteins (UvrA, B, C, and D) that constitute the exonuclease activity of E. coli. Some of the  protein constituents of human excinucleases are listed in Table 15.5. Proteins XPA to XPG have been identified as seven different genetic complementation groups  (A to G) of patients with xeroderma pigmentosum (XP), a condition characterized by UV sensitivity and corresponding deficiencies in DNA repair. The human  nucleotide repair genes are therefore referred to by an XP or ERCC (excision repair component) designation. Nucleotide excision repair of human DNA begins with  the binding of XPA to a dimer between XPF and ERCC1 (Figure 15.17). XPA recognizes and binds to the damaged site along with the replication protein HSSB. An  intriguing aspect of human DNA repair is involvement of an additional enzymic complex con­ TABLE 15.5 Excinuclease Activity of Human  DNA Human Gene

Protein Function

XPA

Damage recognition protein  (binds to damaged DNA)

XPB (ERCC3)

DNA helicase activity;  subunit of transcription  factor TFIIH

XPC

Interacts with general  transcription factor TFIIH

XPD (ERCC2)

DNA helicase activity;  subunit of transcription  factor TFIIH

XPF

Nuclease activity

XPG

Nuclease activity

ERCC1

Part of nuclease activity  (binds to XPF and to  replication protein RPA)

HSSB (RPA)

Binds to the XPF–ERCC1  complex and together with  XPA binds to the lesion site

   

  

Page 637

Figure 15.17  Nucleotide excision repair of human DNA.  In human DNA damage is recognized by the  XPA factor (abbreviated in the figure as A) that  recruits to the damaged site factors XPF and  ERCC1 (abbreviated as F and 1, respectively)  in the form of a dimer. XPF is an excinuclease  that is recruited to the damaged site early on  just as UvrB is recruited in the E. coli system.  The replication protein (HSSB) binds to XPA  and the lesion site. XPA also recruits to the  damaged site the general transcription factor  TFIIH, which, as it turns out, is also a repair  protein since two of its protein subunits  are repair factors XPB and XPD (abbreviated  as B and D). In analogy with UvrA, TFIIH may  be involved in kinking and unwinding of DNA  at the damaged site and in recruiting XPC and  XPG proteins, which are vested with helicase  activity. Excinuclease cuts are made at the 3 site by XPG, whereas XPF nicks at the 5  site  of the lesion, leading to the excision of a 23­mer  oligonucleotide. Gap repair is carried out by  polymerases   and   with PCNA and replication  protein RFC, followed by ligation.  Redrawn based on figure in Sancar, A.  Science 266: 1954, 1994. Copyright  © 1994 American Association for the  Advancement of Science.

 

  

Page 638

sisting of eight different protein subunits and known as the general transcription factor TFIIH. This factor is essential for transcription initiation and for nucleotide  excision repair. In fact, two of the eight subunits of TFIIH are the helicases XPB and XPD that evidently not only act in excision repair but also catalyze the opening of  DNA to initiate transcription. This intimate involvement of a transcription factor suggests that DNA repair and transcription are not fully separable processes and may  be coupled to each other. The TFIIH factor interacts with XPC and the entire complex is recruited to the damaged site by XPA, where it is joined by the  endonuclease XPG. The two recruited endonucleases, XPF and XPG, complete the excinuclease systems with the XPG making the 3  nick and the XPF, in the form  of a complex with ERCC1, making the 5  nick. The major XPG incision is made at the third phosphodiester bond 3  to the lesion, whereas the XPF–ERCC1 complex  incises primarily at the 25th phosphodiester bond 5  to the lesion. The role of TFIIH is presumably to unwind the double helix at the damaged site so as to enable the  endonucleases XPF and XPG to activate the excinuclease system. A protein associated with polymerase  , PCNA (proliferating cell nuclear antigen), releases the  excinuclease subunits and the excised oligomer, which is larger than the oligonucleotide released during E. coli repair (27–29 nucleotides versus 12–13 nucleotides in  E. coli). The gap is filled by polymerases   and   and the DNA is ligated. Excision repair also removes cross­links between complementary DNA strands, such as those introduced by the mustards and drugs used in cancer therapy (i.e.,  mitomycin D and platinum complexes). Error­free repair is not possible if the cross­link extends across directly opposing bases. Clinical Correlations 15.2 and 15.3  discuss defects in DNA repair that are associated with human disease; Clin. Corr. 15.4 examines the role of DNA repair in chemotherapy. Mismatch Repair Mismatch repair in both prokaryotic and eukaryotic cells deals with errors created during DNA replication. In effect, three serially operating mechanisms—base  selection, exonucleolytic proofreading, and postreplicative mismatch re­ CLINICAL CORRELATION 15.2  Defects in Nucleotide Excision Repair and Hereditary Diseases Defects in nucleotide excision repair are implicated in at least three rare hereditary  disorders, xeroderma pigmentosum (XP), Cockayne's syndrome (CS), and  trichothiodystrophy (TTD). XP patients exhibit sunlight­induced photodermatoses  characterized by severe skin reactions that range initially from excessive freckling and skin  ulcerations to the eventual development of skin cancers. Some forms are also  accompanied by neurological abnormalities. The symptoms exhibited by CS and TTD  patients are associated instead only with developmental abnormalities. CS syndrome is  characterized by growth and mental retardation, neurological deficiencies, and  photosensitivity but not an increased rate of cancer or skeletal abnormalities. TDD  patients, on the other hand, have scaly skin, brittle hair, short stature, and neuroskeletal  abnormalities. Xeroderma pigmentosum is a group of closely related abnormalities in excision repair.  About 80% of XP patients fall into one of seven complementation groups (different  syndromes). Each group carries a mutation in a different gene and is characterized by  varying levels of UV sensitivity caused by corresponding deficiencies in "excinuclease"  repair activity. The remainder fall in the XPV (V for variant) group. In this variant UV  irradiation produces different types of mutations compared to normal cells. During normal  DNA synthesis, whenever the DNA polymerase bypasses a pyrimidine dimer in the  template that has not yet been repaired, a purine (most often A) is incorporated into  nascent DNA but this preference is not maintained by XPV cells. It appears that the  mechanism of bypass by the DNA polymerase in XPV cells is altered possibly because  of changes in one or more of the subunits of the polymerase or possibly some other  protein factor that assists the polymerase to bypass the DNA lesions. The neurological  abnormalities that frequently accompany XP appear to result from both abnormal gene  expression and DNA deterioration caused by the accumulation of unrepaired DNA  damage. Cockayne's syndrome is associated with mutations in the CSB/ERCC6, XPD, and XPB  genes. Trichothiodystrophy is caused by mutations in XPB, XPD, and XPG genes and  perhaps in additional subunits of TFIIH or TFIIH­associated excision repair subunits.  Obviously, different mutations in the XPB and XPD genes are responsible for each  syndrome. Tanaka, K., and Wood, R. D. Xeroderma pigmentosum and nucleotide excision repair of  DNA. TIBS 9:83, 1994.    

  

Page 639

CLINICAL CORRELATION 15.3  DNA Ligase Activity and Bloom Syndrome Bloom syndrome is a rare genetic disease that is characterized by chromosomal  instability. Other chromosome breakage syndromes include Fanconi's anemia (FA),  ataxia telangiestasia (AT), Werner's syndrome (WS), and Gardner's syndrome (GS).  Deficiencies in the effective repair of DNA lesions, which can probably be attributed to  defective DNA ligation, are presumably responsible for many of these syndromes. These  repair deficiencies appear to increase the tendency to develop malignancies among those  affected with the syndromes. Bloom syndrome is a prototype of somatic mutation disease. The clinical features of  Bloom syndrome are small body size, a sun­sensitive skin with well­defined hyper­ and  hypopigmented skin lesions, and increased sensitivity to bacterial infections due to  immunodeficiency. Cancer, chronic lung disease, and diabetes are common  complications. Cells from Bloom syndrome patients have high rates of mutation, and the  excessive number of accumulated somatic mutations are responsible for many of the  clinical features of this syndrome. In patients suffering from Bloom syndrome,  hypermutability is responsible for the abolition of ligase I activity needed for completing  DNA repair and (perhaps) DNA recombination. German, J. Bloom syndrome. Dermatol. Clin. 13(1):7, 1995.   CLINICAL CORRELATION 15.4  DNA Repair and Chemotherapy Many anticancer drugs cause DNA damage. For example, cisplatin, used for treatment of  several forms of cancer and particularly effective against testicular tumors, forms two  intrastrand adducts with DNA. The major one, the 1,2­intrastrand d(GpG) cross­link, is  repaired by excision repair. DNA adducts are believed to be the primary cytotoxic lesion  and cells deficient in excision repair are very sensitive to this drug. The high mobility group  (HMG)­domain proteins "shield" and specifically inhibit DNA repair of this major  cisplatin–DNA adduct, thus increasing the cytotoxicity of cisplatin. The types and levels  of HMG­domain proteins in a given tumor may influence the responsiveness of that  cancer to cisplatin chemotherapy. This information may provide a basis for the  development of new platinum anticancer drugs that may have greater therapeutic  potential. Huang, J. C., Zamble, D. B., Reardon, J. T., Lippard, S. J., and Sancar, A. HMG­ domain proteins specifically inhibit the repair of the major DNA adduct of the anticancer  drug cisplatin by human excision nuclease. Proc. Natl. Acad. Sci. USA 91:10394, 1994.   pair­participate in ensuring fidelity of replication. The mismatch repair system recognizes and eliminates mispairing from newly synthesized DNA strands, improving the  fidelity of the synthesis. Base selection and proofreading act more effectively against transversion than transitions, whereas mismatch repair does the opposite. DNA  replication errors are difficult to recognize because mismatches consist of erroneous but unaltered base structures. The repair system relies on other signals within the  helix to identify the newly synthesized strand, which by definition harbors the replication error. Such signals are provided in E. coli by a methylation reaction catalyzed  by Dam methylase that modifies GATC sequences by introducing a methyl group at the N­6 position of adenines. Shortly after replication these GATC sequences  exist in an unmethylated state that betrays the newly synthesized nature of the DNA strand and permits strand discrimination by the mismatch repair system (Figure  15.18). The mismatch repair system in E. coli includes several different protein components, which repair mismatches in the vicinity of a GATC sequence according to  complementary rules dictated by the base sequence of the methylated (i.e., preexisting) parental strand. Proteins that catalyze the process of mismatch repair have  been named MutS, MutH, and MutL. Repair is initiated by binding of MutS to the mismatch followed by the addition of MutL. Formation of the MutS–MutL  complex activates a latent GATC endonuclease activity, vested in the MutH protein, that nicks the unmodified strand at a hemimethylated GATC site. The strand  break, which can occur on either side of the mismatch, will take place as long as the mismatched base is located within the general vicinity of the GATC site, which  means within a few hundred base pairs from the GATC sequence. This nick marks the strand that will be excised. When the mismatch is located on the 5  side of the  cleavage site the unmethylated strand is unwound, degraded, and replaced by new DNA synthesized in the 3    5  direction until the mismatch is reached and  excised. This reaction requires a DNA helicase II, referred to also as the MutU protein, a 3    5  exonuclease (exonuclease I), DNA polymerase III, and finally  DNA ligase to seal the repaired strand. If the mismatch is located on the 3  site of the cleavage, a series of completely analogous steps takes place, except that a 5     3  exonuclease (RecJ) replaces exonuclease I (an exonuclease with both 5    3  and 3    5  activity, exonuclease III can also substitute for RecJ in the latter repair).  This unusual bidirectional excision activity of the mismatch repair system suggests that this system "keeps track" of the side on which the mispair of the GATC  sequence signal is located. Analogous mismatch repair systems have been identified in eukaryotes. Both yeast and human cells code for proteins homologous to the bacterial

 

  

Page 640

Figure 15.18  Mismatch DNA repair.  Methylation of adenine in palindromic 5 ­GATC  sequences serves to distinguish parental strands from  newly synthesized strands that are methylated only  after some delay. Methylation directs the mismatch  repair system to repair mispaired bases. Methylated  GATC sequences are recognized by MutH, which is  also an endonuclease that cleaves the unmethylated  strand on the 5  site of the G in the GATC sequence,  whereas the mispaired site is recognized and bound by  the MutS protein. MutL, which is a molecular  matchmaker, links MutH and MutS together. The  segment of the unmethylated strand, which represents  newly synthesized DNA between the site cleaved by  MutH and a point just past the mismatched base, is  then removed by the action of helicase II,  exonuclease I, and SSB protein. The gap is repaired by  DNA polymerase III and ligase. A similar mechanism,  but based on the presence of nicks to identify newly  synthesized strands, is used by eukaryotes. The  eukaryotic mismatch repair system does not use  MutH and depends on MutL for the degradation  of newly synthesized strands that contain  base mismatches.

proteins MutS and MutL but lack the MutH protein. In eukaryotic mismatch repair the role of MutL is to scan nearby DNA for the presence of nicks. Upon finding a  nick, MutL degrades the nicked strand starting at the nick site and extending just past the site of the mismatched base pair. Replication errors are thereby selectively  removed. Clinical Correlation 15.5 describes the role of mismatch repair in the development of certain types of cancer. Mechanisms That Reverse Damage Formation of dimers can be directly reversed by the action of light. Photoreversal is catalyzed by deoxyribodipyrimidine photolyase, which disrupts the covalent

 

  

Page 641

CLINICAL CORRELATION 15.5  Mismatch DNA Repair and Cancer DNA is constantly being damaged. In the absence of efficient repair, this may be the  cause of as much as 90% of all human cancers. The importance of defective mismatch  repair in the development of certain types of human cancer has been demonstrated  recently. Tumors associated with hereditary nonpolyposis colorectal cancer (HNPCC),  which causes cancer predisposition and certain sporadic cancers, have been found to be  prone to mutation by as much as two orders of magnitude higher than normal human cells.  These high mutation rates have been found to be consistently associated with deficiencies  in mismatch repair. That loss of mismatch repair fidelity is a central step in the development of HNPCC  tumors has been concluded from the finding that the majority of these tumors are  attributable to defects at any one of four different human genome loci. These are the  hMSH2 gene, which codes for a protein homolog of bacterial MutS protein, and the  hMLH1, hPMS1, and hPMS2 genes, which specify three similar but distinct MutL  analogs. These findings demonstrate that the primary event in the development of  HNPCC tumors is the loss of critical mismatch repair activity. Inefficiencies in DNA  repair presumably lead to mutations that circumvent the regulatory systems controlling cell  proliferation. The link between mismatch repair and the development of colon cancer  provides support for the hypothesis that cancers are initiated when cells accumulate a  certain mutation load. A current emphasis in studies of cancer is the search for and study  of particular genes, the mutations of which appear to lead to cancer. The new findings,  which demonstrate the importance of mismatch repair defects in the development of  cancers, may now expand the search from simply attempting to decipher the role of  certain genes in carcinogenesis to also asking why and how some cells accumulate an  excessive number of mutations. Modrich, P. Mismatch repair, genetic stability and cancer. Science 266:1959, 1994.   bonds that hold together the pyrimidine molecules in the dimer. Photolyases are activated by light in the range of 300–600 nm. Photolyases are present in bacteria but  are not essential for DNA repair; humans lack the enzymes. Removal of a methyl or ethyl group from the 6 position of the enol form of a guanine residue reestablishes the normal structure of guanine. A specific protein accepts  alkyl groups and becomes alkylated. Postreplication Repair The repair processes reviewed so far deal with damage of bases on one of the two DNA strands and use of the second complementary strand as a template for repair.  Such repair occurs prior to replication of DNA that turns DNA damage into permanent mutation. For example, normal DNA replication with DNA polymerase III in  E. coli cannot proceed past most types of DNA lesions until such lesions are first repaired. These lesions cannot be excised because excision would leave breaks in  both strands that replication would perpetuate. Eventually, replication resumes past the site of the lesion with the polymerase skipping over a few of the damaged  bases. After synthesis the daughter strand is found to be missing a base that would normally be present across the damaged base. The lesion itself is eventually repaired  by borrowing template information from a homologous DNA strand. This type of repair is illustrated in Figure 15.19.

Figure 15.19  Postreplication repair.  Most DNA lesions in E. coli are  repaired prior to replication. If an  unrepaired lesion is encountered by the  replication complex near the replication  fork, replication is blocked at the site  and resumes only beyond the unrepaired  site. The gap, initially left behind in an  unreplicated single­stranded segment of  DNA, is eventually repaired by the  process of recombination. Recombination  allows the use of a complementary  strand from another DNA as template.

 

  

Page 642

Figure 15.20  SOS DNA repair.  Under normal conditions the SOS repair proteins are not expressed.  This is because a repressor protein, LexA, binds to promoter regions and inhibits the  transcription of many genes required for DNA repair and DNA recombination. LexA also  inhibits its own expression and the expression of another protein with multiple enzymatic  roles, RecA. DNA damage, identified by the presence of single­stranded DNA, inactivates  LexA. Inactivation of LexA is the result of proteolysis by the RecA protein, which when  bound to single­stranded DNA functions as a specific protease. In the absence of LexA,  genes that were previously inhibited by LexA can be expressed. After the damage of   DNA is repaired, LexA begin to accumulate again, repressing the expression of SOS genes.

SOS Postreplication Repair Many of the enzymes involved in DNA repair in E. coli, including the ABC excinuclease system, are inducible and regulated by proteins LexA and RecA that,  together with the genes coding for the inducible proteins, form the SOS repair system. Under normal conditions LexA binds tightly to the control region of genes that code for repair enzymes and several other proteins and prevents the expression. Genes  in the SOS response also induce the polB gene encoding a polymerization subunit of DNA polymerase required for error­prone translesion replication. The SOS  system is activated as a result of severe DNA damage. Activation can be described as the RecA­mediated cleavage and destruction of LexA in an autoproteolytic  manner (Figure 15.20). The fragmented LexA dissociates from the DNA, allowing the efficient expression of the SOS response genes. Some of the products of the  SOS response assemble at the lesion to form a specialized replication system that depends on DNA polymerase II for replicating past DNA lesions, which normally  block DNA polymerase III. This translesion replication is made possible because of the distinct properties of polymerase II. The signal that activates RecA is the binding of RecA onto exposed single­stranded DNA or damaged double­stranded DNA, when DNA replication is stalled  because of extensive DNA damage. The SOS response to heavy DNA damage is a process that converts a lesion at a replication error­prone site and allows  replication to be temporarily restored over the lesion. 15.4—  DNA Replication Complementary Strands Are Basic to the Mechanism of Replication The double­stranded structure of DNA permits each strand to serve as a template for the synthesis of a new strand identical to the other strand, as suggested in Figure  15.21. The correctness of this overall scheme of replication has solidly

 

  

Page 643

Figure 15.21  Each DNA strand serves  as template for synthesis of a new  complementary strand.  Replication of DNA proceeds by a  mechanism in which a new DNA  strand (indicated by a red line) is  synthesized that matches each of the  original strands (shown by green lines).

been established. Even some bacteriophages, which contain single­stranded instead of double­stranded DNA, have been shown to convert their DNA to a double­ stranded form before replication. The simplicity of the basic scheme for replication conceals a rather complex set of coordinated intricate processes. A multiplicity of  enzymes and protein factors participate in these processes. The enzymes involved in replication must also deal with a variety of topological problems. DNA­dependent  DNA polymerase can synthesize new strands by operating only along the 5    3  direction, and therefore it is unable to elongate the two antiparallel strands of the  helix in the same macroscopic direction. In addition, DNA polymerases are unable to start DNA synthesis in the absence of a preexisting primer and the replication  cannot proceed unless the complementary strands are separated at an early stage of the synthesis. Separation requires the commitment of energy for disrupting the  thermodynamically favorable double­helical arrangement and the unwinding of a highly twisted double helix at extremely rapid rates. Double­stranded DNA is normally  a topologically closed domain, which, unless properly modified, will not tolerate strand unwinding to any appreciable degree. Obviously, these multiple difficulties must  be dealt with before the replication of DNA can take place. Replication Is Semiconservative Three possibilities by which information transfer could take place during replication were initially visualized as indicated in Figure 15.22. Conservative replication could,  in principle, yield a product consisting of a double helix of the original two strands and a daughter DNA consisting of completely newly synthesized chains. A second  possibility, labeled dispersive, would have resulted if the nucleotides of the parental DNA were randomly scattered along the strands of the newly synthesized DNA.  The synthesis of DNA eventually proved to be a semiconservative process. After each round of replication, the structure of parental DNA is found to preserve one  of its own original strands combined with a newly synthesized complementary polynucleotide.

Figure 15.22  Three possible types of DNA replication.  Replication has been shown to occur  exclusively according to the semicon­  servative model; that is, after each  round of replication one of the parental  strands is maintained intact, and  it combines with one newly  synthesized complementary strand.  Circles represent the 5  terminals.

 

  

Page 644

Figure 15.23  Semiconservative replication of DNA.  Schematic representation of the experiment of Meselson and Stahl that  demonstrated semiconservative replication of DNA. This model of replication  requires that, if the parent molecule (dark red) contains 15N, each of the  molecules produced during the first generation contain 15N in one strand and  14N in the other. Furthermore, in the second generation two molecules  must contain only 14N, and two molecules must contain equal amounts  of 14N and 15N. The results of separating DNA molecules  from successive generations, shown on the right, are consistent with this model.

The semiconservative nature of replication was elegantly suggested by a classic experiment that allowed the physical separation and identification of the parental and  the newly synthesized strands. Escherichia coli was grown in a medium containing [15N]­ammonium chloride as the exclusive source of nitrogen. Several cell divisions  were allowed to occur, during which the naturally occurring 14N in the DNA of E. coli was, for all practical purposes, replaced by the heavier 15N isotope. The 14N­ containing nutrient was then added, and cells were removed at appropriate intervals. The DNA of these cells was extracted, and the ratios of 14N to 15N content were  determined by equilibrium density gradient centrifugation. The separation between [14N]DNA and [15N]DNA was achieved based on the lower density of DNA,  which contained the lighter isotope. In subsequent experiments, the newly synthesized DNA was thermally denatured and the individual strands were completely  separated. The results, shown in Figure 15.23, demonstrated that daughter DNA molecules consisted of two strands with different densities, corresponding to the  densities of single­stranded polynucleotides containing exclusively 14N or 15N. Conservative and dispersive replications are clearly inconsistent with these findings.

Figure 15.24  Synthesis of primer for DNA replication.  Primer (dashed line) is synthesized by  primase. A primer permits new DNA  (orange line) to be synthesized by DNA  polymerases. The primer is excised at  the completion of DNA synthesis.

A Primer Is Required The semiconservative nature of replication requires that each strand serve as a DNA polymerase template for the synthesis of a new complementary strand. Elongation  is catalyzed by polymerase III (Table 15.1), as distinguished from polymerase I, which is primarily involved in repair. Polymerase III, which is ATP­dependent, is  unable to asemble the first few nucleotides of a new strand and requires a primer. In E. coli primers are segments 10–60 nucleotides long. With few exceptions, the  primer is an oligonucleotide synthesized by other enzymes, as indicated in Figure 15.24. Primers are formed by primases, although in a few instances RNA  polymerases are known to synthesize a primer. In some bacterial systems and phages, the priming enzyme has activity characteristic of an RNA polymerase because  the ribonucleotides condense to form the primer. In other systems the primase does not discriminate between 5 ­ribonucleotides and 5 ­deoxyribonucleotides. As a  general rule, however, primases use ribonucleotides for incorporation into primers. Some enzymes that catalyze the synthesis of primers act exclusively as primases,  while others possess additional enzymatic activities. In mammalian cells primase activity is vested in DNA polymerase a , an enzyme that is also involved in DNA  strand

 

  

Page 645

elongation and in DNA repair. Once the primers have been synthesized, the DNA polymerase can move in and take over the process of synthesis. It is not clear what  signal causes a switchover from primase to DNA polymerase, although it has been suggested that a specialized ribonuclease (RNaseH) is involved. 15.6 RNA Primers Replicating System

RNA  Oligonucleotidea

Bacteriophage T4

pppAC (N)3

Bacteriophage T7

pppACCA

 

pppACCC

Mouse polyoma virus

pppA (N)9

 

pppG (N)9

Lymphoblastoid cells

pppA (N)8

 

pppG (N)8

a N stands for any ribonucleotide. The primer 

lengths for the mouse polyoma virus and the animal  cells are averages.

  If DNA polymerase were the enzyme that would begin DNA synthesis by laying down the very first nucleotide complementary to the template, the efficiency of DNA  synthesis would be severely reduced. Since the bases in a very short segment of a double helix have high configurational flexibility, the first nucleotide introduced into a  newly synthesized DNA strand would likely be mispaired and would immediately activate the proofreading activity of DNA polymerase. The outcome would be a  fruitless back­and­forth cycle of synthesis and proofreading by DNA polymerase with little net synthesis of new DNA. In contrast, primases, which have no  proofreading ability, can quickly and efficiently position primers that can be elongated with DNA polymerases without appreciable backtracking. The primases ignore  mismatches and produce an RNA chain long enough to allow the DNA polymerase to operate at the 3  end of a double­stranded structure that restricts newly  introduced nucleotides on the basis of strict complementary rules. The mismatches introduced by the primase are irrelevant because the characteristic RNA­like  structure of primers allows for their subsequent wholesale removal and replacement by DNA of an equivalent composition. Although primers are almost invariably short RNA or RNA­like segments (Table 15.6), RNA priming is not used universally. In the ''rolling circle" replication  mechanism of DNA, a 3 ­OH primer is generated by endonuclease digestion of parental DNA, and with parvoviruses a 3 ­OH primer is generated by the folding  back of an existing 3  terminus. A single deoxyribonucleotide can serve as primer in adenovirus. Such a nucleotide, with its 3 ­OH terminus free, is attached to the  end of a template strand through a virus­encoded specific protein (Figure 15.25).

Figure 15.25  An unusual primer used in the  replication of adenovirus DNA.  This primer is a single nucleotide attached, by its  5 ­terminal phosphate, to a serine residue of a  protein. Adenovirus DNA is synthesized by  extension of the 3  terminus of this nucleotide.

 

  

Page 646

Figure 15.26  Both DNA strands serve as templates for  DNA synthesis.  Each DNA strand must serve as a template  for DNA synthesis. The new DNA can be  synthesized only in the 5    3  direction.  If only a single initiation origin were  considered, the result of continuous  synthesis would be the formation of two  new nonidentical double­stranded DNA  molecules (one above and one below the  initiation origin). Also, the upper part of  strand A and the lower part of strand B  could not have been used as templates.  In fact, the synthesis occurs both  continuously and discontinuously.

Both Strands of DNA Serve As Templates Concurrently In the preceding section, the events leading to the synthesis of DNA by DNA polymerase were examined and attention was directed to one of the two parental DNA  strands used as template. In fact, synthetic events occur at both strands almost concurrently. This would appear to generate some problems of geometry. Specifically,  if a single initiation site is considered, and the synthesis continued in the 5    3  direction until each template is completely copied, the result of the synthesis would be  the creation of two new double­stranded molecules. Examination of Figure 15.26 indicates that, at least in the case of linear double­stranded DNA, neither of these  two hypothetical DNA molecules would be identical to the parental DNA. Such an outcome is not in agreement with the actual course of DNA replication. The discrepancy can be accounted for by recognizing that the microscopic synthesis of  the new strands does not proceed uninterrupted. In fact, the synthesis occurs in a discontinuous fashion and in a manner that permits the assembly of the synthesized  polynucleotide portions into appropriate complete DNA strands. Synthesis Is Discontinuous The overall process of DNA synthesis may now be considered past the immediate vicinity of initiation by examining a larger section of DNA. One of the two parts of  DNA that would be generated if the macromolecule were divided at the site of chain initiation is shown in Figure 15.27. In almost every instance the synthesis is  bidirectional, which means that the synthetic events occurring at the part of the molecule indicated by solid lines are of the same general nature as those occurring on  the other site and indicated with dashed lines. A prerequisite for the semiconservative mechanism of replication is that the two complementary strands of DNA gradually separate as the synthesis of new strands  takes place. The mechanics of this separation are addressed later, but it may be apparent that as a result of separating the strands at an interior position, two  topologically equivalent forks are created at the point of diversion of the two strands. Various lines of evidence have indicated that DNA polymerase acts in a discontinuous manner; that is, along each DNA molecule there are numerous initiation points  at which primers are formed. In eukaryotes primers may be formed at locations that are determined by nucleosome spacing. In the case of bacteriophage T7,  primosomes appear to recognize TGGT and GGGT through prepriming proteins. Once a site for primer initiation has been recognized,

Figure 15.27  Discontinuous synthesis ofDNA.  This figure emphasizes the synthetic events occurring at only one side of the  initiation site (dark red line). The two complementary strands of DNA separate as  the discontinuous synthesis of small DNA segments takes place on both strands  located at different sites on the DNA. After excision of the primers, the excised  parts are repaired, and the segments are joined together. Although segments are  clearly synthesized in opposite directions on the two strands, overall macroscopic  impression is that DNA grows in the single direction suggested by the solid red  arrow on the right.

 

  

Page 647

single­strand binding proteins (SSB), which interact with single­stranded polynucleotides, are displaced and the primase lays down a primer. After promoting  primer initiation at one point, prepriming proteins move along the template strand in order to synthesize the adjacent primer. At each one of these locations, DNA  polymerase III makes use of the assembled primers for the synthesis of DNA. When DNA polymerase reaches the end of the single­stranded template, it comes upon  the next primer annealed to the template. The polymerase, as indicated by its very high processivity, can overcome this hurdle by sliding over the intervening double­ stranded DNA–RNA hybrid and resuming replication at the 3  end of this new primer. The segments synthesized by DNA polymerase upon each primer, known as precursor (Okazaki) fragments or nascent DNA, vary in size from about 100 to 200  deoxyribonucleotides in eukaryotes to ten times as long in bacteria. Once these segments of the new DNA are synthesized on both strands of a fork (Figure 15.27),  the fork opens up further, and the same process of synthesis is repeated. Shortly after synthesis, the primer portions of the Okazaki fragments are excised by the 5     3  exonuclease activity of DNA polymerase I, which also synthesizes short segments of DNA. This discontinuous mechanism compensates for the inability of DNA polymerase to synthesize strands in the 3    5  direction. By synthesizing portions of DNA  strands only in the 5    3  direction on both antiparallel strands of the parental DNA, the polymerase is able to create the illusion, when the synthesis is experimentally  visualized by electron microscopy techniques, that both strands are concurrently elongated in the same macroscopic direction. In Figure 15.27 this direction is  indicated by a large solid arrow. It should be noted that the first strand synthesized, often referred to as the leading strand, is synthesized continuously. It is the other  strand, the lagging strand, that must be synthesized discontinuously. Macroscopic Synthesis Is As a Rule Bidirectional At the site of initiation of DNA synthesis two identical forks are created (Figure 15.27). Therefore two possibilities exist for the synthesis of DNA: the process may  occur at only one fork and proceed in a single direction, as shown by the thick solid arrow, or alternatively it may occur at both forks and in both directions away from  the starting point. The events occurring in the forks located below the starting line are simply a mirror image repetition of what occurs in the fork that is located above  the line. Bidirectional replication is the mechanism of DNA synthesis. The only known exceptions are in a small number of phages and plasmids that replicate  unidirectionally. In the case of a small linear chromosome (e.g., bacteriophage  ) each fork moves along, synthesizing new DNA, until the end of the chromosome is  reached. In a circular chromosome (e.g., E. coli) the two forks proceed in opposite directions until they meet at a predetermined site on the other side of the  chromosome, as depicted in Figure 15.28. As the two forks meet, a new copy of the parental DNA is completed and released. The average rate at which each fork  moves during replication is of the order of 60,000 bases per minute at 37°C. Upon completion, new DNA is released by the action of a type II topoisomerase as  illustrated in Figure 15.29. Strands Must Unwind and Separate Separation of the strands of the parental DNA prior to synthesis of new strands is a requirement because the bases of each template must be made accessible to the  complementary deoxyribonucleotides from which the new strands are constructed. The overall process of separation consists of a number of enzymatically catalyzed,  coordinated steps, including the local unwinding of the helix, and the nicking and rejoining of the strands necessary for continuation of the

 

  

Page 648

Figure 15.28  Bidirectional replication of a circular chromosome.  Replication starts at a fixed origin and proceeds at a constant rate in opposite  directions until the two replication forks meet. Newly synthesized strands are  indicated by dashed lines. After DNA synthesis is complete, two newly  synthesized circular DNA molecules are separated by action of topoisomerases.

unwinding process. Once the strands are unwound, they must be kept separate so that they can operate freely as templates. Specialized proteins accomplish rapid orderly unwinding of the strands. These proteins, helicases, separate DNA strands in advance of the moving replication fork  and just in front of DNA polymerase. In E. coli they are referred to as helicase II and rep protein. Helicases move unidirectionally along DNA and separate the  strands in advance of replication. They destabilize the interaction between complementary base pairs at the expense of ATP. Once the strands have been separated, the single­stranded regions are stabilized by specific proteins, the single­strand binding (SSB) proteins. The DNA single  strands are covered by the SSB proteins because of their high affinity for single­stranded DNA. As the helicase moves in advance of the replication fork, SSB proteins  go on and off the DNA, with protein molecules that are displaced from one site reassociating with another (Figure 15.30). SSB proteins do not consume ATP and do  not exhibit any enzymatic activities. Their role is only to keep the strands apart long enough for the priming process to occur. In E. coli DNA, it is calculated that the parental double helix must unwind at a rate of about 6000 turns per minute. These high rates would generate insurmountable  difficulties if strands were to separate over an appreciable length of DNA. The large free­energy requirements of bringing about the unwinding of large regions of DNA  can, however, be reduced to manageable levels by the nicking of one or both of the DNA strands near the replicating fork. Since the fork is a moving entity, the  nicking must be visualized as a reversible cut­and­rejoin process, which moves along with the fork. Nicking is indispensable for a topological reason as well.  Unwinding at one of the two forks requires that the parental double helix rotate in the opposite direction to that necessary for the unwinding of the opposite fork. In the  absence of a nick as the unwinding at one of the forks would progress, an increasing number of positive supercoils would have to be introduced into the double helix.  Once the limit of the helix to accommodate the supercoils were reached, unwinding and replication would have to cease.

Figure 15.29  Function of topoisomerases II in separating  interlocked DNA double helices.  Topoisomerase II attaches to both strands of  DNA through reversible covalent bonds, thus  forming an interrupted double helix with a  topoisomerase "gate." A second DNA helix can  pass through the portal using an  "open­and­shut­the­gate" mechanism,   leading to two separated DNA molecules. After  separation of the molecules topoisomerase  dissociates from DNA.

These topological restraints are overcome if DNA is maintained during replication in the negative superhelical form. This form could serve as a "sink" for the  positive supercoils that could potentially be generated during replication. In E. coli, this is apparently achieved by the action of gyrase, a topoisomerase type II,  which induces the formation of negative supercoils

 

  

Page 649

Figure 15.30  Model for DNA replication in E. coli.  The initial stages of replication are depicted. Primers are removed from newly  synthesized segments of DNA at the lagging strand, and the segments are joined.  Since replication is normally bidirectional, similar events take place  concurrently at the other side of the initiation origin.

at the expense of ATP. Topoisomerases type I may also be involved. The superhelicity of DNA may be negatively regulated through a balance between  topoisomerases of types I and II; that is, a diminishment of topoisomerase II activity may bring about a decrease in the amount of negative superhelicity that can be  created, whereas an inhibition of topoisomerase I activity may increase it. During replication the linking number between parental strands decreases from a large value  at the beginning of replication to zero at the end of a complete round of DNA synthesis. TABLE 15.7 Components of the Replisome Protein

Function

SSB

Single­strand binding

Protein i  (dnaT)  Protein n  Protein n Protein n dnaG

 

Primosome assembly  and function Primase

 

(primer synthesis)

Pol III  holoenzyme

Processive chain  elongation

Pol I

Gap filling and primer  excision

Ligase

Ligation

Gyrase

Supercoiling

gyrA

 

gyrB

 

rep

Helicase

Helicase II

  Helicase

dnaB dnaA  dnaC

Origin of replication

 

  Escherichia coli Provides Basic Model for Replication of DNA Extensive studies in E. coli and its phages have permitted the proposal of a replication model that depends on the action of a large number of proteins, some of which  are listed in Table 15.7. With the specific exceptions noted in the sections that follow, this model may also be viewed as a basic scheme for DNA replication in most  other cells. Initiation and Progression of DNA Synthesis Synthesis of DNA begins at a specific site of the chromosome referred to as the replication origin, which in E. coli is referred to as OriC (Figure 15.30). Initiation of  DNA synthesis involves participation of as many as 20–30 different proteins, many of which are needed to be present at the origin of replication in multiple copies.  OriC must be recognized by specific proteins, and the origin must unwind to allow helicase, primase, and DNA polymerase III to have access to each DNA strand.  OriC is a sequence of 245 base pairs that contains four sites (nucleotide 9­mers with a similar nucleotide sequence) at which dnaA, a tetramer consisting of four  identical subunits, can initiate the stepwise assembly of all the proteins and enzymes necessary to carry out replication (Figure 15.31). In addition, the origin contains 11  methylation sites recognized by Dam methylase and three AT­rich direct tandem repeats consisting of 13 base pairs each. This final assembly is called a replisome. Formation of a replisome begins with the binding of one dnaA molecule

 

  

Page 650

at each one of the 9­mers, provided that these binding sites are fully methylated. The dnaA apparently recognizes these 9­mers on the basis of their conformation,  which appears to be slightly curved with the double helix somewhat elongated relative to typical B­DNA. Several more additional dnaA molecules are then added via  a highly cooperative process to form a nucleosome­like structure. An additional factor, HU protein, participates in the formation of this complex. The dnaA and HU protein interact with the OriC in a manner that promotes the opening of the DNA strands in the AT­rich regions adjacent to the origin. Finally,  dnaA, with the aid of dnaC, adds dnaB in the complex. The dnaB, by virtue of its helicase activity, creates an initiation "bubble" consisting of a few hundred nucleotide  pairs. The energy for the formation of the "bubble" is provided by ATP in a reaction catalyzed by topoisomerase II, and the "bubble" is stabilized by SSB proteins. Synthesis of an RNA primer begins with the formation of a prepriming complex. The prepriming assembly consists of the dnaB–dnaC complex to which four other  proteins (polypeptides n, n , n", and i) have been added. Addition of primase, dnaG, converts the prepriming complex to a primosome

Figure 15.31  Model for initiation of replication in E. coli.  Step 1: Initiation of replication begins with binding of dnaA molecules to four sites  consisting of nine­nucleotide long sequences each. These sequences are present  at the origin of replication in E. coli (OriC).  Step 2: DNA­bound dnaA molecules subsequently coalesce and are joined by  additional dnaA molecules to form a nucleosome­like DNA­protein complex,  which promotes nearby "melting" of the double helix.  Step 3: The resulting opening of strands allows a dnaB–dnaC complex to become  attached to DNA so that helicase activity of dnaB can further unwind the DNA.  Unwinding is accompanied by a displacement of dnaA molecules.  Redrawn based on figure in Rawn, J. D., Biochemistry. Burlington, NC: Neil Patterson  Publishers, 1989.

 

  

Page 651

(Figure 15.32). The primosome interacts with a template, at each one of the two forks generated by the formation of a "bubble," and begins the synthesis of RNA  primers on the two leading strands. Assembly of the replisome is completed by addition to the primosome of DNA polymerase III and rep proteins.

Figure 15.32  Primosome of E. coli.  The primosome is formed by binding of  primase, together with a complex of  dnaB and dnaC proteins, at specific  sequences of DNA that serve as sites  for formation of RNA primers.  Additional factors, described as n  proteins, are specific primosomal  components that are responsible for  placing the primosome at the  appropriate sequences. In effect, the  primosome "searches" the DNA for  these sequences at the expense of  ATP. Once the correct destination of  the primosome is reached, RNA primer  synthesis is initiated.

Initiation can be regulated by either restricting the availability of dnaA­binding sites at OriC or by limiting the concentration of dnaA. Methylation provides a switch  for the availability of dnaA­binding sites. Once replication has been initiated, the dnaA near OriC binds to the plasma membrane and becomes unavailable to Dam  methylase. In addition, binding of DNA in the vicinity of OriC to the cellular membrane sequesters the dnaA gene, which is situated near OriC (only 40 kb away). As a  result, the synthesis of dnaA protein is inhibited and its cellular concentration is lowered. Initiation of the leading DNA strand at OriC by the primosome is more complex than the subsequent initiation of synthesis of Okazaki fragments on the lagging strand  initiated by primase at sites selected by the prepriming proteins. The initiation of the leading strand does not present the cell with serious topological problems, but for  continuation of synthesis helicase II and rep protein are essential. These enzymes unwind and separate the strands in each of the two forks created by the initiation  event. As the helicases move in advance of each fork, two single­stranded regions are generated on parental DNA. These regions are immediately covered by single­ strand binding protein that keeps the fork open and allows DNA polymerase III to take over the elongation of primers. A signal for initiation of the lagging strand,  uncovered on the template by the movement of helicase, leads to the binding of primase. Primase, the action of which is triggered by the prepriming proteins,  synthesizes a brief complementary segment of the strand. This segment serves as a primer for covalent extension of the strand synthesized by DNA polymerase III and  for formation of Okazaki fragments. DNA polymerase III complexes are endowed with similar but somewhat distinct properties, one tailored for the continuous  synthesis of the leading strand and the other for the discontinuous synthesis of the lagging strand. This polymerase assembly, which appears to combine primase activity  with nonidentical twin active sites for polynucleotide synthesis, allows for concurrent replication on both strands. In this scheme, looping of the lagging strand template  by 180° brings it to the same orientation as the leading strand template (Figure 15.33). Thus a primer synthesized at the lagging strand is drawn past it. When a nascent  (Okazaki) fragment reaches the 5  end of the previously synthesized Okazaki fragment, the lagging strand template is released and unlooped. Removal of the primer  portions at the 5  end of the Okazaki fragments by DNA polymerase I, repair by the same enzyme, and joining of the repaired fragments by DNA ligase produces  intact DNA strands. Termination of DNA Synthesis Termination occurs near the center of a 270­kb region across from OriC, the ter or t locus. This region incorporates five ter sequences, that is, loci with the core  sequence GTGTGTTGT that bind the Tus protein (terminator utilization substance) that promotes the termination of synthesis (Figure 15.34). Tus protein is a  contrahelicase in that it functions by literally interfering with the ATP­dependent and dnaB helicase­promoted unwinding of DNA rather than simply impeding the  propagation of this helicase along the double helix. The organization of the ter region is shown in Figure 15.34. Each Tus site has directional properties (asymmetry)  and it arrests only those replisomes that reach the Tus site from one specific direction. Replisomes arriving from the opposite direction apparently force the dissociation  of the Tus protein and thus

 

  

Page 652

Figure 15.33  Model for the simultaneous synthesis of leading and lagging DNAstrands by  DNA polymerase.  Two molecules of DNA polymerase operating in concert, and in the same rather  than the opposite direction, may be participating in the simultaneous synthesis of  DNA on both strands. In this model the replisome consists of a DNA polymerase  dimer associated with the primosome and helicases. The primer made by the  primosome is extended by the replisome as the lagging­strand template is looped  through it. The primer continues to be extended until the previously completed  Okazaki fragment is reached, at which point the loop is relaxed. The stretch of  unpaired lagging­strand template then loops back again to participate in the  formation of the next Okazaki fragment.  Redrawn based on figure in Kornberg, A. DNA Replication.  San Francisco: Freeman, 1992.

can proceed unimpeded past the Ter–Tus site. Because of the distribution and orientation of sites in the ter region, each replisome must first pass over all sites that are  oriented the opposite way before arriving at the Tus site that is oriented in a way that causes termination. This arrangement makes it inevitable that a replisome will not  dissociate from DNA until it actually collides with the replisome entering the ter region from the opposite direction. This ensures the complete replication of the  chromosome and prevents overreplication. The products of replication are two concatenated progeny chromosomes usually interwound by as many as 30 coils. The  newly synthesized DNA is untangled from the parental DNA apparently by the action of a topoisomerase II. Rolling Circle Model for Replication DNA synthesis directed by circular mtDNA, and in some instances by bacteria and viruses, gives rise to linear daughter DNA molecules that contain the base  sequence of parental DNA repeated numerous times. These repeated linear DNAs, which are known as concatemers, are essential for the bacterial mating and may  be involved in gene amplification. The synthesis of concatemer DNA occurs by a mechanism known as rolling circle replication.

 

  

Page 653

Figure 15.34  Termination of DNA replication in E. coli.  Termination region (ter) of E. coli incorporates five  asymmetric ter sites. Each ter site can interact with Tus  protein. TerB and terC are oriented in the same direction  and the remaining three ter sites are oriented in the  opposite direction. Because of the orientation of  Tus­bound ter sites, each replisome that reaches the ter  region must cross all the Tus–ter sites that are  oriented the opposite way before arriving at a site that  causes termination. A replisome moving in the direction  shown by the arrow must first cross terE, terD, and terA  before terminating replication at either the terC or terB  site. This arrangement ensures that each replisome  continues to synthesize DNA until it collides with a  replisome entering the ter region from the opposite  direction, leading to the dissociation of both replisomes  from DNA.  Adapted from Hidaka, M., Kobayashi, T., and  Horiuchi, T. J. Bacteriol. 173:381, 1991.

An example is the replication of certain circular single­stranded bacteriophages such as  X174. When the virus enters a host bacterium the single­stranded genome is  converted to a double­stranded DNA by action of primase and DNA polymerase III. The DNA strand complementary to the bacteriophage genome that is first  synthesized [labeled the (–) strand] serves as the template for the genomic DNA [the (+) strand]. The atypical characteristic of this replication scheme is that the (+)  strand is nicked at a specific site (by a phage­encoded endonuclease) so that it can serve as a primer for its own replication. The (+) strand is elongated from the 3 ­ hydroxyl end of the nick by DNA polymerase III by incrementally displacing segments of the (+) strand associated with the "helper" (–) strand (Figure 15.35). A second characteristic is that the circular template does not dissociate from the complementary strand during the synthesis. Instead the replication of the leading strand  goes on beyond the length of circle­generating linear concatemeric DNA. Appropriately sized DNA molecules are subsequently generated from concatemers by  specific endonuclease cleavage. Eukaryotic DNA Replication The DNA synthesis in eukaryotes appears to be a process that is fundamentally similar to that occurring in prokaryotes. Formation of a replication fork, primer

Figure 15.35  Replication by the rolling circle mechanism.  In ssDNA of certain bacteriophages, such  as  X174, the (+) strand is converted into  dsDNA upon injection into a host bacterium.  This transformation occurs by action of  primase and polymerase III upon ssDNA  that synthesizes a complementary (–) strand.  Replication of (+) strands begins with nicking  of (+) strand so that it can serve as a primer for  its own replication. The (+) strand is elongated  from the 3 ­hydroxyl end of the nick, as the  newly synthesized strand gradually displaces  from the helper­strand the original (+) strand.  Redrawn based on figure in Moran, L. A.,  Scrimgeour, K. G., Horton, H. R. Achs, R. S.,  and Rawn, S. D. Biochemistry. Englewood  Cliffs, NJ: Neil Patterson/Prentice Hall, 1994.

 

  

Page 654

synthesis, Okazaki fragments, primer removal, and gap bridging between newly synthesized DNA segments, all parallel the corresponding steps that occur in  prokaryotes, but the overall process is quite a bit more complex. Replication among eukaryotes, from yeasts to humans, shares similarities. As expected, differences are more pronounced between prokaryotes and eukaryotes. In rapidly growing prokaryotes, DNA is replicated through much of the cell  cycle and cell division occurs as soon as DNA synthesis has ceased. In contrast, eukaryotic DNA synthesis (and histone synthesis) is confined to only one part of the  cell cycle, specifically the synthetic (S) phase of the interphase. This phase is preceded and followed by two periods during which DNA is not synthesized (gap periods  G1 and G2). Cell division occurs at a different time within the interphase, referred to as the mitotic (M) period. Beyond this characteristic limitation of eukaryotic  replication to a certain period of the cell cycle, important differences in replication between prokaryotes and eukaryotes arise primarily from the larger size of  eukaryotic DNA (about 105–106 kb content) as compared to prokaryotic DNA (about 5 × 103 kb for E. coli), the distinct packaging of eukaryotic DNA in the form  of chromatin, and the slower rates of fork movement in eukaryotes. For DNA to become available to DNA polymerases, nucleosomes must disassemble, a step that  slows the rates of fork movement. DNA polymerase movement does not exceed 30,000 base pairs per minute, which is considerably slower than the rates observed  for E. coli. Based on the higher DNA content of animal cells, and the lower activities of DNA polymerases in comparison to bacteria, the replication cycle of  eukaryotic cells could be expected to take as long as a month to complete. In fact, however, the replication cycle is completed within hours, because compensating  factors are in operation. Eukaryotic cells contain a large number of DNA polymerase molecules (often in excess of 20,000) as compared to a few dozen in each E.  coli cell. DNA polymerase initiates bidirectional synthesis but at several origins of replication located anywhere between 5 and 300 kilobase pairs (kb) apart within the  chromosome, depending on species and cell type (Figure 15.36). DNA segments between two origins of replication are termed replicons. An average human  chromosome contains as many as 100 replicons and replication may proceed simultaneously at as many as 200 forks. More origins can be found in developmentally  active cells that carry out DNA synthesis at very rapid rates. During early embryogenesis the largest chromosome of Drosophila melanogaster contains as many as  6000 replicating forks, or one for every 10 kb. Role of Eukaryotic DNA Polymerases In prokaryotes synthesis is catalyzed by two similar but distinct subunits of DNA polymerase III. In eukaryotes, synthesis of the leading and lagging strands is carried  out by different enzymes (Table 15.2). DNA polymerase d , a polymerase of high processivity, catalyzes the synthesis of the leading strand. This enzyme consists of a  large subunit that is vested with 5    3  nucleotide polymerizing activity and a smaller subunit that has a 3    5  proofreading exonuclease activity. The high  processivity of DNA polymerase   is attributed to the presence of an accessory factor, the proliferating cell nuclear antigen (PCNA), that is found in large amounts  in the nuclei of proliferating cells. PCNA (mol wt 25,000) is a multimeric protein that can act as a "clamp" to keep the enzyme from disassociating off the leading DNA  strand. The "clamp" consists of three PCNA molecules, each containing two topologically identical domains that are tightly associated to form a closed ring. This  suggests that in eukaryotes PCNA is the functional equivalent of the b  subunit of E. coli polymerase III. Another accessory protein, the replication factor C (RFC),  also binds to polymerase   and probably assists with association between PCNA and DNA to form the "clamp." Alternatively, RFC may be involved in setting up a  link

 

  

Page 655

Figure 15.36  Replication of mammalian DNA.  Mammalian DNA replicates by using a very  large number of replicating forks  simultaneously. This mechanism accelerates  the process of replication, which in  mammalian systems is limited by rates of fork  movement that are considerably slower than  those characteristic of prokaryotes.  Redrawn based on figure in Huberman, J. A.,  and Riggs, A. D. J. Mol. Biol. 32:327, 1968.

between polymerase   and polymerase a . Therefore the role of RFC in DNA synthesis is analogous to the roles of the g complex and the   subunits of E. coli DNA  polymerase III. Synthesis of the lagging strand is catalyzed by DNA polymerase a . This polymerase has similar structure and properties in all eukaryotes. The large subunit (mol wt  ~ 180,000) of the tetrameric DNA polymerase a  is vested with the usual 5    3  nucleotide polymerizing activity. Polymerase a , isolated from some but not most  sources, also has a 3    5  exonuclease activity. Two of the other subunits of the enzyme are primases. The primary proofreading function in eukaryotes appears to  be carried out by polymerase  . Polymerase   improves the fidelity of replication by a factor of 102 and contributes in limiting the rates of overall error to 10–9 to 10–12. The relatively low processivity of DNA polymerase a  is typical for an enzyme involved in synthesis of the lagging strand that is assembled from segments of DNA that  are no larger than 100–200 bp. The size of these Okazaki fragments is approximately equal to the length of DNA wrapped around a nucleosome. This observation  suggests that eukaryotic DNA may be releasing one nucleosome at a time for priming of the lagging chain. The primase subunit of the enzyme synthesizes Okazaki  segments as a closely coordinated priming–synthesizing activity, by laying down RNA primers containing 5–15 nucleotides that are subsequently extended by the  synthetic activity of polymerase a . This polymerase catalyzes the synthesis of a polynucleotide chain at a rate of 50 nucleotides per second, which is about 1/20 the  rate of E. coli DNA polymerase III synthesis. Looping of the lagging strand allows a combined polymerase a ­polymerase   asymmetric dimer to assemble and  elongate both the leading and lagging strands in the same overall direction that corresponds to the direction of the fork movement. A third large monomeric protein,  polymerase e , is vested with a synthetic 5    3  polymerase activity and both a 3    5  proofreading exonuclease activity and a 5    3  exonuclease activity.  Polymer­

 

  

Page 656

ase   is mainly required for DNA repair and for filling the gaps between Okazaki fragments on the lagging strand. Eukaryotic DNA synthesis requires replication protein A (RPA), also known as replication factor A (RFA). This protein is the functional equivalent of prokaryotic  single strand binding (SSB) protein. While helicase activities are part of the prokaryotic chromosome, eukaryotic helicases do not appear to be associated with  primase activity. Eukaryotic helicase activity appears to be associated with DNA polymerase  . Initiation of Eukaryotic DNA Replication Origins of replication in eukaryotic cells have been identified in yeast (Saccharomyces) and are termed ARS for autonomously replicating sequence. ARSs are  about 100–120 bp long, each of which is characterized by an AT­rich central region. The 400 or so copies of the ARS in the yeast genome have highly conserved  nucleotide sequences within the central region with variations in the flanking sequences. The core sequences of ARS contain 11­bp elements known as the ARS  consensus sequence rich in AT pairs that appear to be analogous to the AT­rich 13­mers present in the OriC of E. coli. The flanking elements consist of overlapping  sequences that include variants of the core sequence. Protein binding to form a so­called origin of replication complex (ORC) promotes DNA strand unwinding  over the AT­rich sequences of the ARS cores. The unwound region is stabilized by single­strand­binding protein and RPA, and is extended by helicase. Polymerases  a  and  , RFC, and PCNA are thus introduced into the origin of replication and begin DNA synthesis. Weaker binding sites identified as B1, B2, and B3 are also present near the origin. B1 and B2 serve as sites for ORC formation, while B3 is associated with a protein  that promotes initiation of transcription. This observation highlights the close association between eukaryotic DNA replication and transcription. Controlled activation  of variant ARS­like subgroups, consisting of ARS­like sequences with different flanking elements, may determine the order of initiation of DNA synthesis in  eukaryotes. Sequences completely comparable to yeast ARS have not been identified in higher eukaryotes. In mammals it appears that initiation depends more on  chromosomal context than on specific sequences. Origins of initiation may be found within a broad section of the genome that also contains a small number of ''hot  spots," at which initiation is favored. In spite of these differences in the origins of replication between yeast and higher eukaryotes, the rest of the replication machinery  appears to be remarkably analogous. Eukaryotic genomes replicate in a definite order, and at definite times within the S phase, with some DNA regions replicating  early in the S phase and other DNA regions replicating later. Genes that replicate early are found in active segments of chromosomes, and genes that replicate later are  located in the inactive areas of chromosomes. This pattern of activation changes with development. Differences in the rate of replication are regulated by variations in  the duration of the S phase, which can be achieved either by controlling the number of replicons activated per unit length of chromosome or by slowing down the rate  of DNA unwinding and replication. Sequence elements similar to the ARS subgroups in yeast may control replicon activation in other eukaryotes through the  interaction of initiating proteins with these elements. Origins that are activated simultaneously are expected to share the same DNA sequences and bind to the same  control proteins. Since eukaryotic DNA is present in packaged form as chromatin, DNA replication is sandwiched between two additional steps, namely, a carefully ordered and  incomplete dissociation of the chromatin and reassociation of DNA with the histone octamers to form nucleosomes. Methylation at the 5 position of cytosine residues  by a DNA methyltransferase appears to function by loosening up the chromatin structure and allowing DNA access of proteins and enzymes needed for DNA  replication. The synthesis of new histones occurs

 

  

Page 657

mainly during the S phase simultaneously with DNA replication. Histone molecules appear to rarely leave the DNA to which they are bound. Instead transcription and  replication forks are apparently able to move past the parental nucleosomes as they synthesize mRNA or new DNA. One possibility is that each nucleosome  dissociates into two halves, thereby permitting DNA polymerase to replicate transiently uncoiled DNA. Newly synthesized DNA inherits some parental histones, which  it combines with an equal amount of new histones to complete the structure of nascent nucleosomes that are formed behind the moving replication forks. In coordinating the synthesis of DNA the eukaryotic cell copies millions of base pairs, distributed over numerous chromosomes, with remarkable accuracy and at just  the right time in the cycle of cell division. Copying start at hundreds of different origins, some of which are triggered early in the S phase of the cell cycle while others  are triggered late. Recent evidence indicates that the replication initiator, that is the ORC complex, does not act alone in controlling initiation. One or more additional  proteins bind to the initiation origins late in mitosis and remain attached until the S phase begins. These proteins are known as cyclin­dependent kinases (CDKs) and  operate in association with specific protein substrates (cyclins). Cyclins and CDKs may control the cell cycle; they push the cell to the S phase and initiation of DNA  synthesis. Cyclin­CDK pair also prevents DNA synthesis from being initiated a second time, so that only one S phase occurs per cell cycle. Degradation of CDKs  removes the signal that inhibits cell division and the cell cycle moves again to mitosis. This scheme suggests that DNA initiation depends upon the formation of a  prereplication complex by adding to or removing from the ORC cyclins and CDKs in a cyclical manner. This scheme in which the same enzyme first activates DNA  replication and then, once one round of DNA replication has begun, inhibits reformation of the prereplication complex provides an efficient arrangement for the  coordination of the initiation of DNA synthesis. DNA Replication at the End of Linear Chromosomes Linear chromosomes cannot be fully replicated in the absence of additional steps that provide for the replication of their terminals. As a replisome falls off from the end  of a linear chromosome, and the daughter DNA molecules separate, synthesis of DNA on the end of the lagging strand cannot be fully completed. A gap resulting from  removal of a primer that was used to start replication is generated on the lagging strand (Figure 15.37). The exact size of this gap

Figure 15.37  DNA replication at the ends of linear chromosomes.  In the absence of a special mechanism of replication operating at the ends of  chromosomes, the completion of DNA synthesis of linear dsDNA would leave gaps at  ends of newly synthesized strands. These gaps would result from removal of primers  used to start replication. Upon each subsequent round of replication the gaps would  be continuously expanded and accumulated because DNA polymerase requires a  primer and therefore it cannot fill such gaps.

 

  

Page 658

CLINICAL CORRELATION 15.6  Telomerase Activity in Cancer and Aging Telomerase activity maintains appropriate length of the telomere sequences of  chromosomes. Surprisingly, however, telomerase activity is absent from most somatic  cells. In such cells telomere repeats gradually decrease in number with aging, as repeated  cell divisions produce a substantial shortening of the telomere structure. Loss of  telomerase activity in protozoans, such as Tetrahymena, is responsible for a gradual  shortening of telomeres following each cell division, throughout the life of the cell. In  human cultured fibroblast cells a linear inverse relationship exists between the length of  telomeres and the age of the subject from which the cells are obtained. Eventual loss of  telomeres leads to chromosomal instability and cell senescence and it may be an  important factor that contributes to the process of aging. Specifically, telomere length  appears to serve as a mitotic clock that limits the replication potential of mammalian cells.  If it is true that the shortening of telomeres may be a contributing factor to the aging  process, then the natural life span of an individual may be determined by the length of its  telomere DNA. However, the possibility that telomere shortening may be the result,  rather than the cause, of aging cannot be excluded. In any event, many other factors are  also likely to contribute to the process of aging. Since telomere length may serve as a mitotic clock, telomerase activity may stimulate cell  division. The expression of telomerase may thus provide a selective advantage that allows  tumor cells to divide indefinitely. Current understanding of telomere biology is still modest  but as it improves telomerase may indeed become an important potential target for cancer  chemotherapy. Allsopp, R. C., Vaziri, H., Patterson, C. et al. Telomere length predicts replicative  capacity of human fibroblasts. Proc. Natl. Acad. Sci. USA 89:10114, 1992; and  Counter, C. M., Hirte, H. W., Bacchetti, S., and Harley, C. B. Telomerase activity in  human ovarian carcinoma. Proc. Natl. Acad. Sci. USA 9:2900, 1994.   depends on the location of the last Okazaki fragment synthesized. As a minimum, the daughter DNA synthesized would have an 8–12 base gap generated by removal  of the RNA primer for the Okazaki fragment. Without intervention this gap would be continuously regenerated and accumulated during each subsequent round of  replication because it cannot be filled by DNA polymerase that requires a primer. The products of DNA replication would become shorter relative to parental DNA,  leading to the gradual loss of DNA at the ends of human chromosomes. Cell senescence in humans and other mammals may be related to this chromosomal shortening  as described in Clin. Corr. 15.6. In human cells that carry information to daughter cells (gamete cells) and in the linear chromosomes of bacteria and viruses, however,  the integrity of DNA during replication cannot be compromised. Maintenance of intact chromosomal

Figure 15.38  Replication of adenovirus DNA.  The adenovirus uses a protein as a primer, the terminal protein (TP), for synthesis of  both strands of its DNA. TP, covalently associated with one dCMP, binds at the 3 end of each template chain and the dCMP residue provides a 3 ­OH for DNA  polymerase­catalyzed synthesis of a complementary strand. Since both strands of the  viral DNA are synthesized continuously in the 5    3  direction, DNA synthesis is  complete, leaving no gaps at the ends of the chromosome.  Redrawn based on figure in Wolfe, S. L. Molecular and Cellular Biology.  Belmont, CA: Wadsworth, 1993.

 

  

Page 659

structure requires a distinct mechanism for replication at the ends of DNA molecules. Prokaryotic Replication Different replication strategies have evolved to deal with the problem in viruses, plasmids, and organelle DNA. One approach is the use of a primer consisting of a  protein, referred to as terminal protein, TP, that binds covalently to the 5  ends of viral DNA molecules via a phosphodiester bond with the hydroxyl group of a  serine residue (Figure 15.38). Modified versions of TP that are distinct for different viruses also participate in replication. For instance, in the case of the mammalian  adenovirus, the TP contains covalently bound dCMP. In bacteriophage  29, the bound nucleotide is dAMP. These nucleotides pair with the terminal nucleotides at the  3  end of each strand and serve as primers for replication. A special polymerase coded by each virus recognizes the TP and copies the strands unidirectionally from  their 3  to 5  ends. With the priming limited to the ends of the parental DNA strands, both strands are replicated completely as if they both are leading strands. The TP  molecule is cleaved from the primer nucleotide and it is released upon completion of the synthesis. Other viruses form circular intermediates that are copied by a rolling  circle mechanism. Finally, some viruses, with identical sequences at the ends of their DNA, can hybridize their terminal sequences, forming linear repeats (linear  concatenates). These concatenates are cleaved postreplicatively to generate progeny virus of the proper size (Figure 15.39). Eukaryotic Replication: Telomerases Eukaryotes employ different strategies than prokaryotes and viruses for the replication of their chromosomal ends, known as telomeres. One approach that is used,  albeit rarely, is the lengthening of chromosomal ends by the transposition of DNA segments known as transposons. This approach is apparently used for maintaining  the chromosome ends in Drosophila. In most eukaryotes, however, telomere replication utilizes a specialized reverse transcriptase enzyme called telomerase.  Telomerase activity depends on the presence of an RNA molecule that constitutes part of the telomerase structure and serves as an "internal" template. Maintenance of  the chromosomal length depends on the action of telomerase on repetitive DNA sequences that constitute the telomeres of eukaryotic chromosomes (Figure 15.40).  These telomeric tandem repeats can be several thousand nucleotides long and they consist of multiple copies of short G­ and T­rich oligonucleotide sequences. Their  size varies extensively from 20 bp in length for some protozoa to 150 kb in mouse telomers. For humans and other vertebrates the repetitive DNA is constructed with  variants of the sequence TTAGGG. A short segment of single­stranded DNA ending in a 3 ­OH group caps the end. Telomerase recognizes the G­rich single­strand  at the 3  terminus and elongates it in the 5    3  direction, by adding telomere repeats at the end of the lagging chain. The RNA of telomerase, which has a sequence  of about 150 nucleotides complementary to the telomer repeats, provides a movable template that substitutes for the absence of a normal DNA template. Telomerase  provides in one package all that is needed for elongation of the strand that ends in a 3  terminus, namely, both template and enzymic activity. Extension of the telomeric  sequence elongates the 3  end of DNA by about 100 nucleotides. This is then used as template for synthesis of the complementary strand by DNA polymerase a .  Telomerase is then repositioned to repeat the process as illustrated in Figure 15.40. In this manner telomerase and polymerase a  serve to maintain chromosomal length  during repeated rounds of DNA replication. Maintenance is affected by such factors as telomerase processivity and its frequency of action on telomers as well as the  rate of degradation of telomeric DNA. Telomeres may grow, shrink, or stay fairly stable depending

Figure 15.39  Replication of bacteriophage T7 DNA.  Bacteriophage T7 DNA has repetitive  identical sequences at its chromosomal  termini so that, following replication, the  daughter molecules can hybridize end to end  to form dimers. During subsequent rounds of  replication the process is repeated until a  large linear DNA, a concatenate, is formed. A  specific nuclease then cleaves the large  concatenate into fully replicated genome­size  DNA segments.  Redrawn based on figure in  Mathews, C. K. and Van Holde, K. E.  Biochemistry. Redwood City, CA:  Benjamin/Cummings, 1990.

 

  

Page 660

Figure 15.40  Telomere replication.  Telomerase contains an RNA template that codes for the extension of the ends of  chromosomes and serves as a template for DNA polymerase. The DNA strand made  on the lagging side of a replication fork of a linear chromosome is incomplete. For this  strand to be completed, telomerase extends the 3  end on the complementary strand  at the leading side of the fork. Telomerase first binds to a TG primer at the 3  end  of this DNA strand. Binding is the result of base pairing between primer and RNA  template that is part of the telomerase complex. The enzyme adds more T and G  residues to the primer and repositions the RNA template so that more TG  repeats can be added to the end of the primer. The extended primer is eventually  recognized by DNA polymerase a, which proceeds to replicate the 5  end of the  DNA using the single­stranded 3  end as template. Primase activity is vested  in a subunit of DNA polymerase a.  Redrawn based on figure in Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K.,  and Watson, J. D. Molecular Biology of the Cell. New York: Garland, 1994.

on genetic or nutritional changes. For example, the size of yeast telomeres can vary from about 200 to 400 bp depending on conditions. DNA Can Be Synthesized Using an RNA Template For many years it had been assumed without reservation that the only direction in which genetic information can flow is from DNA to RNA. This dogma had to be  revised, however, when it was discovered that the genomes of certain viruses, such as the retroviruses, consist of RNA instead of DNA and that during viral infection  this genomic RNA is copied into DNA. The DNA that is obtained can either be transcribed to produce more viruses or it may be incorporated into the DNA of the  host. In the latter case the viral genome is replicated along the DNA of the host and often remains latent for many host chromosome generations. Enzymes that use RNA templates for DNA synthesis are called reverse transcriptases. Reverse transcriptases are often virally encoded but they are not limited to  viruses. Enzymes with reverse transcriptase activities are also found in uninfected cells and are involved in the formation of pseudogenes and in the replication of  transposable elements (see p. 669). Reverse transcriptases are the most error­prone type of DNA polymerases because they lack 3    5  exonuclease activities, thus  lacking a proofreading function. Inhibitors of reverse transcriptase are used for the treatment of AIDS as described in Clin. Corr. 15.7. DNA Replication, Repair, and Transcription Are Closely Coordinated It has become increasingly clear that DNA replication, transcription, and repair are not separable, as most DNA lesions block both replication and transcription. Thus  repair occurs with "expressed genes" as a priority, with the repair of dormant genes deferred. In addition, transcription and repair appear to cross paths at several  points, with certain repair proteins participating in the activation

 

  

Page 661

CLINICAL CORRELATION 15.7  Inhibitors of Reverse Transcriptase in Treatment of AIDS AIDS is caused by a retrovirus, the human immunodeficiency virus (HIV). Treatment of  AIDS is complicated by the high mutability of this virus, which reflects the low fidelity of  the HIV reverse transcriptase responsible for the synthesis of the viral genome. This  transcriptase is about one order of magnitude less accurate than other transcriptases and  produces one or more mutations per generation, which means that any two HIV DNA  molecules are almost never exactly the same in their nucleotide sequence. The first drug that was used with some success, and continues in use, in controlling the  rate of advancement of the disease is a structural analog of deoxythymidine, known as  AZT.

This drug is converted to the triphosphate by cell kinases and the triphosphate is  incorporated into the HIV genome in place of dTTP. AZT triphosphate competes  successfully with dTTP for incorporation into the viral genome because of the higher  binding affinity of AZT relative to dTTP toward the HIV reverse transcriptase. Since  AZT has a lower affinity for cellular DNA polymerases than dTTP, it is not incorporated  into cellular DNA. Incorporation of AZT triphosphate causes a premature termination of  viral DNA synthesis because it lacks a 3 ­OH site that is needed as the primer for  incorporation of additional nucleotides. Other nucleotide analogs, with similar reverse transcriptase­dependent mechanisms of  actions, have been included in the treatment of AIDS. These include dideoxyinosine (ddI)  dideoxycytidine (ddC), and azidothymidine (ZDV). Current approaches use ZDV or  combination therapies of ZDV and ddI or ZDV and ddC. Other compounds that are not  nucleotide analogs, referred to as nonnucleoside reverse transcriptase inhibitors  (NNRTI), and a diverse group of other agents, such as protease inhibitors and HIV  immune­based therapies, are currently under investigation for treatment of AIDS. A new  class of drugs that inhibit proteases essential for HIV replication, when used in  combination with reverse transcriptase inhibitors, is reported to reduce viral loads in  AIDS patients to undetectable levels and in many instances reverse rather than simply  arrest the symptoms of the disease. Finkelstein, D. M., and Shoenfeld, D. A. (Eds.). AIDS Clinical Trials. New York:  Wiley­Liss, 1995.   of initiation or elongation steps of transcription. For example, subunits of the TFIIH factor, which is essential for transcription, also participate in eukaryotic nucleotide  excision repair. Repair and replication appear also to be coupled at the level of the protein factor, HSSB. This protein binds single­stranded DNA with high affinity  during replication but it is also a repair protein required for the formation of the preincision complex. A protein induced as a result of DNA damage, the so­called  Gadd45 protein, has regulatory effects on both DNA repair and replication. Gadd45 appears to both stimulate excision repair and inhibit DNA replication. 15.5—  DNA Recombination DNA recombination refers to a number of distinct processes during which genetic material is rearranged by breaking and joining portions of the same DNA molecule  or portions of different DNA molecules. Recombination also takes place between the DNAs of different organisms to generate a new "composite" DNA. Both  prokaryotic and eukaryotic DNAs undergo recombination. Three well­characterized processes listed in Table 15.8 fall under this general description of genetic  recombination. Other DNA rearrangements have been noted whose mechanism and function are not well­understood and are referred to as illegitimate; these will not  be reviewed in this chapter. Recombination creates new combinations of genes on the chromosome, which increase the chance of survival of a population. This  increase of genetic diversity offers no advantage for individuals within a population. Individual survival partially

 

  

Page 662 TABLE 15.8 Characteristics of Different Types of Genetic Recombination Sequence  Homology

Heteroduplex  Sequences

Homologous

Extensive, but the homology  is DNA sequence  independent

Long

RecA, RecBCD, RuvAB,  RuvC, and DNA repair  enzymesa

Some

Site­specific

Short but specific DNA  sequences are required on  both DNAs

Short

Recombinases

Some

Transpositional

Homology is not required;  specific sequences needed on  one of the DNAs

None

Transposases

Minor  (only to fill  gaps)

Type

Proteins  Involved

DNA Synthesis

a Several additional protein factors including RecE (exonuclease VIII), RecF, RecG, RecJ, RecN, RecOR, RecQ, RecT, 

SbcCD, DNA polymerase I, DNA gyrase, DNA topoisomerase I, DNA ligase, and DNA helicases participate in  catalyzing homologous recombination.

  depends, instead, on the operation of DNA repair. However, certain types of DNA repair depend on DNA recombination and therefore it is possible that  recombination evolved as a mechanism of repair. Homologous genetic recombination produces an exchange between a pair of distinct DNA molecules, often two slightly variant copies of the same chromosome, or  two segments of DNA generated from the same DNA molecule. The main requirement for this process to occur is that the recombining DNAs are homologous. This  means that the two DNAs share very similar base sequences over an extended region that may contain several thousand bases. An important example of homologous  recombination in eukaryotes is the exchange of sections of homologous chromosomes during the early development of gametes (egg and sperm cells). In this manner  slightly different versions of the same gene (alleles) can evolve during meiosis. Gene "mixing and reassortment" by general recombination is also widespread in bacteria.  Homologous recombination is quite complex and involves a multistep mechanism catalyzed by a large number of different proteins. Prominent among them is the RecA  protein, which also participates in SOS DNA repair. Conservative site­specific recombination or site­specific recombination requires the presence of only short homologous DNA sequences. However, site­ specific recombinations occur only in specific DNA sequences present in both the participating DNA molecules. The process is catalyzed by enzymes known as  recombinases. Transpositional site­specific recombination, or simply transposition, differs from conservative site­specific recombination in that it does not require a specific  DNA sequence in the "target" chromosome. Transposition is catalyzed by transposases. Both transposases and recombinases recognize and act on specific DNA  sequences. Recombination of either type is responsible for the insertion of viruses, plasmids, and transposable elements (transposons) into chromosomal DNA.  Transposons are DNA elements that can move from location to location within a genome, in both bacteria and eukaryotes. Viruses are related to plasmids and  transposons but also differ from these genetic elements in that viruses can synthesize a protein coat that allows them more host­independent existence. Plasmids and  transposons are confined to replicate only within a specific cell and the progeny of that cell. The most common recombination is the homologous type. Site­specific recombination and transposition are relatively rare, but important, events in that they may  control replicative function in some viruses and certain aspects of development. Homologous recombination generates new combinations of genes that can lead to  genetic diversity. DNA mutation and recombination are

 

  

Page 663

the two principal approaches by which the cell creates variation that is required for evolution to occur. In addition, recombination events are involved in DNA repair.  In those instances in which DNA damage occurs across complementary DNA sites, DNA repair can occur only through recombination. A large variety of protein  structures used by the human immune system are produced by recombination as described in Clin. Corr. 15.8. Homologous Recombination Homologous recombination, which is accompanied by the formation of a heteroduplex DNA region, clearly requires breaking and rejoining of chromosomal DNA.  Recombination occurs via a fairly complex multistep mechanism. A scheme that explains the outcome of recombination is shown in Figure 15.41. This scheme gives a  minimal overview of recombination, in that each of the steps shown may represent more than one enzymatically catalyzed process. Numerous gene products are  involved in homologous recombination. Recombination may begin by introduction of a single­strand nick at a selected site of one of the DNA duplexes undergoing recombination. The resulting 3 ­ended  single­strand tail can then invade a homologous DNA duplex. Homologous DNA duplexes are chromosomes with the same linear arrangement of genes but with base  sequences that may differ between the two duplexes. The variance is usually minor and may consist of no more than one different base among the millions of base pairs  present in the chromosome. Single­strand invasion places the homologous DNA duplexes side by side in a process referred to as synapsis. Synapsis does not  necessarily involve contacts between homologous sequences and further movement of the DNAs with respect to each other may be necessary until homologous  sequences come into contact. This process is referred to as homologous alignment. Strand invasion is accompanied by strand displacement in the homologous  DNA duplex resulting in the formation of a so­called D­loop. The "D­loop" strand that has been displaced by strand invasion is now nicked and it pairs with its  complementary strand in the original duplex. The ends of exchanged strands are then ligated to form a stable cross­stranded intermediate known as Holliday  junction. The junction can migrate in either direction by unwinding and rewinding of the two CLINICAL CORRELATION 15.8  Immunoglobulin Genes Are Assembled by Recombination Immunoglobulins (antibodies) are molecules that recognize and specifically bind to any  substance that antibodies identify as foreign to the human body (see p. 88 for details).  Because of the immense variety of infectious agents, including millions of microorganisms  that are present in the environment, the human genome, which is equipped with only a  limited pool of probably no more than 100,000 genes, does not have the capacity to  directly produce an equivalent number of different antibodies necessary for specific  recognition of all infectious agents. This inherent limitation in the gene­coding potential of  the human genome is, however, overcome by recombination, which allows production,  from a limited amount of gene­coding DNA, of an almost unlimited number of distinct  antibodies. Human immunoglobulins consist of two heavy and two light chains with each chain having  a variable region, with a sequence that is characteristic for each immunoglobulin, and a  chain with constant amino acid sequence (see p. 89). Recombination leads to diversity in  the variable region of immunoglobulins. During the maturation of a bone marrow stem cell  into a B lymphocyte, one V segment and one J segment are brought together by site­ specific recombination. In the process the intervening DNA is deleted and a joint between  the two regions is established by an RNA­splicing reaction that occurs following  transcription. Since the V region consists of 300 segments and the J region of 4, at least  1200 different combinations can be generated by recombination. Similar considerations apply to the   light chains and the heavy chains, with the latter  being assembled in as many as 5000 distinct combinations. Because individual light and  individual heavy chains can subsequently be assembled in combination, at least 6 × 106  different IgG molecules can be produced. Furthermore, because some variations occur in  the exact location of the V­J junction, the actual number of IgG molecules is two to three  times higher than estimated above. Additional IgG diversity is produced during the  process of maturation of B lymphocytes by mutational processes.    

  

Page 664

Figure 15.41  Overview of homologous recombination.  Transformations that can lead to formation of recombinant  and nonre­combinant heteroduplexes, by participation of  two homologous DNA molecules in homologous recombination  are outlined. Each step indicated need not be the outcome  of a single, enzymatically catalyzed or well­understood  reaction. The sequence of steps shown is not necessarily  universally applicable.

duplexes to produce a further exchange of single strands between interacting chromosomes. This process, known as branch migration, results in strand exchange and  it produces heteroduplex regions of varying lengths. The resulting heteroduplex, shown in Figure 15.41, can also be presented in another form that is generated by  merely pulling the ends of the heteroduplex together (Figure 15.42). A twist of this structure produces an isomeric heteroduplex, which is called the Chi form. In order  to resolve the Chi form two additional single­strand nicks can be made, in either the horizontal direction or vertical

 

  

Page 665

Figure 15.42  Patch and splice recombinant heteroduplexes.

direction, leading to two distinct products. Gaps present in these structures are repaired and ligated, leading to either one of the two products. The manner in which  nicks are introduced in the horizontal and vertical directions is fundamentally different. In one case (horizontal direction) nicks are introduced again into the strands that  were initially nicked, although at different sites, producing two duplexes in which one strand of each remains intact. These duplexes contain heteroduplex regions,  generated by branch migration, that are misleadingly referred to as Patch recombinant heteroduplexes. These duplexes contain the same genes and in the same  linear order as the initial duplexes. In vertical direction nicks, the complementary strands that previously were left intact are nicked again (though at different sites),  producing two duplexes of true recombinant DNA, referred to as splice recombinant heteroduplexes. In these true recombinant heteroduplexes the linear order of  DNA sequences contained in the original duplexes is clearly rearranged. Support for this multistep recombination scheme has accumulated over the years based on genetic investigations, on electron microscopy of Holliday junctions, and by  isolation of proteins and enzymes that can catalyze many of the transformations described in this recombination scheme. Enzymes and Proteins That Catalyze Homologous Recombination Homologous recombination in E. coli requires about 25 enzymes for recombination. A partial list includes RecA protein, RecBCD enzyme (which is the product of  three distinct E. coli genes, recB, recC, and recD), RuvAB and RuvC proteins, DNA polymerase I, DNA gyrase, DNA topoisomerase I, DNA ligase, and DNA  helicases (Table 15.8). Proteins homologous to RecA have also been isolated from yeast and human cells. Homologous recombination in E. coli begins with RecBCD, which is a site­specific endonuclease and an ATP­dependent helicase (Figure 15.43).

 

  

Page 666

RecBCD can initiate recombination by unwinding DNA and, on occasion, cleaving one strand. The enzyme binds to one end of linear DNA and travels along the helix  at the expense of ATP, unwinding DNA as it moves and rewinding DNA behind it at a slower rate than unwinding. This produces a ''bubble" consisting of two single­ stranded loops that propagate on the DNA with the advance of the RecBCD. Escherichia coli DNA is characterized by the presence of about 1000 copies of the  sequence 5 ­GGTGGTGG­3  that, on average, occurs at intervals of 4–5 kb. These Chi sites are "hot spots" for recombination as they increase the frequency of  recombination. When the advancing RecBCD encounters a Chi site within a "bubble," it cleaves the DNA strand that incorporates the 5 ­GGTGGTGG­3  sequences  5–6 nucleotides to the 3  side of the Chi site. The helicase activity generates a 3  single­stranded tail of DNA that is progressively lengthened to several kilobases.

Figure 15.43  Activities of RecBCD protein.  RecBCD combines helicase and nuclease  activities and appears to be involved in  initiation of homologous genetic  recombination in E. coli. RecBCD, using its  helicase activity, enters the double helix  and, using energy derived from ATP  hydrolysis, travels along the helix until  it encounters a Chi site, which consists of  the sequence  5 ­GCTGGTGG­3 . RecBCD introduces a cut,  within the Chi site, that leads to displacement  of a 3 ­terminating single strand. This single  strand initiates recombination by pairing with a  homologous DNA double helix.  Redrawn based on figure in Liehninger, A. L.,  Nelson D. L., and Cox, M. M. Principles of  Biochemistry. New York: Worth, 1993.

This growing single­stranded tail can then initiate the strand invasion process with the assistance of RecA, which catalyzes a multiplicity of reactions in DNA  recombination (Figure 15.41). RecA interacts with single­stranded (ss) and double­stranded (ds) DNA and catalyzes pairing of homologous DNA sequences,  invasion of ssDNA into the homologous double helix, formation of the Holliday junction, and migration of this junction (branch migration). These activities of RecA  depend on the presence of a RecA site that recognizes ssDNA and promotes the cooperative binding of the protein to ssDNA. Formation of a long and relatively stiff  nucleofilament (Figure 15.44) prevents the

Figure 15.44  DNA strand exchange mediated by RecA.  Replacement of a complementary  strand in a DNA duplex by a single­stranded  DNA is catalyzed by RecA. RecA begins the  exchange by coating both ssDNA and dsDNA by  RecA (only coating of the single strand is  shown). The coating modifies the  conformation of both the single­stranded  and double­stranded polynucleotides and  catalyzes the invasion of the  single­stranded intermediate. Switches in  the base pairing between the strands, and  the accompanying rotation of the DNA, move  the three­stranded region from left to  right as one strand of the DNA duplex is  displaced by the identical, or nearly  identical, invading ssDNA. Continuing  branch migration leads to eventual  separation of the displaced strand.

 

  

Page 667

ssDNA tail from reassociating with the complementary strand within the DNA duplex, from which it originated, and prepares the single strand for invasion. In the  resulting nucleofilament that binds one RecA molecule per 3 bases, the polynucleotide is positioned within a deep groove of the RecA protein. A second site on RecA  recognizes and binds preferentially to dsDNA. In this nucleofilament each RecA monomer covers six nucleotides and each successive monomer binds to the opposite  site of the DNA helix. For the sake of simplicity the dsDNA in Figure 15.44 is shown as free from RecA. The RecA–ssDNA and RecA–dsDNA nucleofilaments  differ in their geometry from B­DNA, but both filaments represent partially unwound and unstacked helical structures that are extended lengthwise by 50% relative to  B­DNA. DNA unwinding in the RecA–dsDNA nucleofilament (to about 18.6 bp per turn) exposes H­bond donors and acceptors in the major groove of the double  helix, making them available for interaction with the ssDNA–RecA filament. Thus RecA contributes to the recognition of regions of homology between DNA strands.  Once homologous alignment is established, a fairly stable triple­stranded intermediate can be formed (Figure 15.45). In this structure the third strand is in contact  with the major groove of the duplex, aligned in a manner that permits RecA to flip the base pairing of the two identical strands.

Figure 15.45  Model for the triple­stranded  intermediate formed during DNA  recombination.  RecA catalyzes the formation of  a triple­stranded DNA intermediate  as a result of the association of a  dsDNA, the strands of which are  marked D, and an invading ssDNA,  marked S (shown in the middle).  Both dsDNA and ssDNA are  present in the form of complexes  with RecA. This protein catalyzes  unwinding of the strands of the  double helix and makes the matrix  of hydrogen­bond donors and  acceptors in the major groove of  the double helix available for  pairing with ssDNA. The ssDNA  is also unwound by RecA, providing  for proper alignment between dsDNA  and ssDNA.

The flipping of the base pairing and the resulting invasion of the RecA–ssDNA filament involve the exchange of two identical (or nearly identical) strands between  helical structures, which therefore requires an ordered rotation of two aligned strands. The polynucleotides are prepared for this exchange by "the extended"  conformation generated by RecA. Strand exchange can be extended by branch migration, which means that progression of the exchange requires both invasion and  branch migration. Branch migration may be described as a process in which an unpaired region of a single DNA strand displaces a DNA strand from a region of  homologous dsDNA and moves the branching point, without appreciably increasing the total number of disrupted base pairs. Migration is achieved by RecA­catalyzed  rotation of RecA­bound DNA strands involved in the exchange (Figure 15.44). The resulting "spooling" action, in which topoisomerases may be involved, moves the  branch as ATP is hydrolyzed. Branch migration also occurs at the Holliday junction that is subsequently formed. In this intermediate homologous DNA helices that were initially paired are held  together by mutual exchange of two of the four strands (Figure 15.46). Stereochemistry of the intermediate is determined by the juxtaposition of the grooves and the  phosphate backbones of the participating helices, and the point of exchange or actual junction can be moved back and forth along the helices. Migration of the junction  can proceed in the absence of RecA. This RecA­independent migration of the junction is catalyzed by a complex of RuvA and RuvB. RuvA binds to the junction and  acts as a specificity factor that targets RuvB, which is an ATPase, to the junction. The RuvAB complex promotes migration and increases the length of the  heteroduplex DNA at the expense of ATP. Finally, the Holliday junction is recognized and resolved into products by the RuvC endonuclease, a dimer of 19­kDa  subunits related to each other by a dyad axis of symmetry. The catalytic center of this resolvase lies at the bottom of a cleft that fits a DNA duplex. Only strands with  the same polarity are cleaved and produce two types of heteroduplex molecules, one type in which only single­strand segments are exchanged (patch recombinants)  and another type, a true recombinant, in which the ends of molecules have been exchanged (splice recombinants). Resolution is completed by DNA polymerase I,  DNA topoisomerase I, DNA gyrase, and DNA ligase. RecA also exhibits a highly specific protease activity that is activated by unpaired DNA strands and is directed at specific regulatory proteins. Thus RecA has unique  properties for coordinating regulation of a number of cellular

 

  

Page 668

Figure 15.46  Structure of the Holliday junction.  The Holliday intermediate is a four­way junction that adopts a right­handed  antiparallel X­shaped structure by pairwise coaxial stacking of the two double  helices. The junction consists of fully stacked base pairs with the participating  strands present as hydrogen­bonded DNA duplexes.  Redrawn based on figure in Moran, L. A., Scrimgeour, K. G., Horton, H. R.,  Ochs, R. S., and Rawn, J. D. Biochemistry. Englewood Cliffs, NJ:  Neil Patterson/Prentice Hall, 1994.

functions that occur when DNA damage, or the interruption of DNA replication, leads to the production of ssDNA segments. An example is the postreplication repair  of DNA damaged by UV light or other mutagens. Site­Specific Recombination This process separates and joins dsDNA molecules at specific sites. Site­specific recombination is limited to select regions of a genome and is driven by  recombinases that recognize short (20–200 bp) specific sequences on both recombination sites. When recombinase binds to both recombination sites on DNA  molecules it can produce an insertion of DNA. A well­studied example is provided by the integration of so­called temperate phages, such as E. coli bacteriophage  ,  into the host chromosome of the corresponding host (Figure 15.47). The circular   chromosome becomes integrated into a specific site in the E. coli chromosome  consisting of about 20 nucleotides, the so­called attP site. Integration requires the alignment of the   phage in a specific orientation with the E. coli chromosome. The  alignment is achieved by a specific recombinase known as integrase (Int) and the participation of a protein known as the integration host factor (IHF) encoded by  the bacterium. Integrase brings together the attB site of the bacterium with a corresponding specific site on the phage chromosome, which consists of 230 bp and is  known as the attP site. Int generates a precise wrapping of DNA to juxtapose specific nucleotide sequences for the splicing reactions that follow. Functioning as a  topoisomerase, Int unwinds the attP region and forms an Int–attP nucleoprotein. A corresponding nucleoprotein is also formed between Int and attB that brings the  attP and attB

 

  

Page 669

Figure 15.47  Site­specific recombination of l phage.  Site­specific recombination is carried out by integrase. The phage chromosome  undergoes recombination between the attP site and a corresponding site on the  bacterium, attB. Integration of the phage chromosome generates two new attachment  sites (attR and attL) that flank the integrated phage DNA. The reverse reaction,  excision of the integrated phage chromosome, requires the participation of protein  XIS produced by the bacteriophage and protein FIS encoded by the bacterium.

sites together. Integrase then generates a staggered cut, 7 base pairs apart within a core sequence of 15 bp, that is present in both the attP and attB sites and catalyzes  the exchange of strands at the position of the cut to form a Holliday intermediate. To complete the exchange, cutting and rejoining must be repeated at a second  point within each of the two recombination sites. Normally, limited branch migration is required prior to an Int­catalyzed second cleavage and strand exchange.  Following ligation by Int, the original sequence of the recombination site is regenerated but the DNA on either side of the site is recombined. Recombinases often act  in a reversible manner, restoring the sequences of original DNAs. Integrase also acts in a reversible manner so that the circular phage chromosome can be excised as  conditions change. The forward and reverse steps of the integration reaction are separately regulated, with the reverse step being dependent on the presence of  additional proteins: the XIS protein encoded by the phage and FIS encoded by the bacterium. Both reactions also require IHF. Transposition Transposition is a form of recombination catalyzed by recombinases called transposases. This type of recombination is best understood in bacteria but DNA of all  cells, including eukaryotes such as Drosophila, maize, and yeast, contains segments that can move, generally with very low frequencies of 10–5–10–7 per cell  generation, from a donor site to another target site within a chromosome. These segments are known as transposable elements (transposons). Transposition differs from homologous recombination in not requiring sequence homology between donor and target sites. Only the donor site, that is, the transposon,  has specific nucleotide sequences located on both sides of the transposon that serve as binding sites for transposases. Most bacterial transposons have short repeats of  about 15–25 bp at the two ends of the transposable DNA segment. In contrast, the target sites are not well defined and are not characterized by specific DNA  sequences. Heteroduplex joints are not formed as a result of transposition. Three classes, I, II, and III, of transposable elements are recognized. Class I transposons are called insertion sequences (IS) if they consist of a gene

 

  

Page 670

CLINICAL CORRELATION 15.9  Transposons and Development of Antibiotic Resistance Genes conferring to bacteria resistance to commonly used antibiotics such as pencillin or  tetracycline are usually carried on plasmids. The DNA sequences of these plasmids do  not have any homology with the chromosomal DNA sequences of the host. Yet, as a  result of transposition, antibiotic resistance genes can be transferred to the chromosome  of bacterial hosts. The existence of genes that can move from one chromosome to  another is of course of great importance in understanding the factors that produce  changes in the organization of genomes. From the clinical standpoint these "transposable"  genes are of critical significance for understanding how populations of antibiotic­resistant  bacteria arise with use of antibiotics in the treatment of bacterial infections in humans and  animals.   coding for transposase together, of course, with the repeats that normally flank the transposable element. IS elements vary in size between 800 and 1300 bp. When  Class I transposons also contain an additional gene, such as a gene conferring antibiotic resistance to bacteria, they are called composite transposons (Tn). Class II  transposons differ from Class I in that, in addition, they code for the gene of a second enzyme, resolvase. Typically, composite transposons and Class II  transposons are several thousand base pairs long. Finally, a small group of bacteriophages, such as bacteriophage Mu, that insert their chromosome into a host  chromosome are classified as Class III transposable elements. Transposition begins by a transposase­catalyzed introduction of a staggered cut at the target DNA sequence. Cuts are also made on each side of the transposon so  that it can be moved onto the target site. The relocation leaves a double­stranded break at the site from which the transposon is excised. At the target site the  transposon is spliced into the staggered cut as shown in Figure 15.48. Specifically, 3–12 bp at the target site are duplicated by DNA polymerase I, to form an  additional short repeat at each end of the inserted transposon, and the "tailored" transposon then is ligated within the target site. In Class II and III transposition, in  addition to duplication of the short repeats, the transposon itself is replicated and one copy of it remains at the donor site while the other copy is transferred to the  target site. This type of transposition, referred to as replicative transposition, requires the enzyme resolvase and therefore does not occur in Class I transposition.  Replicative transposition can reshape the structure of a chromosome beyond the simple act of relocating a transportable element from one site to another. Because this  type of transposition places two homologous sequences within the same chromosome, homologous recombination between these two sequences can produce either a  deletion or an insertion, depending on whether these sequences are oriented in the same or in opposite directions, as shown in Figure 15.49. Finally, transposition may inactivate a gene by mutation if a transposon is inserted into a coding sequence and interrupts it. Alternatively, insertion by transposition of a  promoter or a transcriptional activator next to a gene may activate the gene. Clinical Correlation 15.9 reviews the role of transposition and Clin. Corr. 15.10 the role of  DNA amplification in the development of drug resistance.

Figure 15.48  Direct repeats at the ends of transposons.  Transposons are inserted into gaps generated at a target  sequence by introduction of a staggered cut by a transposase.  Ligation of transposon to the protruding ends of target DNA  leaves gaps at both sides of the transposon. Repair of these  gaps is responsible for the presence of direct repeats that  flank transposons.  Redrawn based on figure from Mathews, C. K.  and Van Holde, K. E. Biochemistry. Redwood City, CA:  Benjamin/Cummings, 1990.

 

  

Page 671

Figure 15.49  Genomic rearrangements promoted by transposons.  In replicative transposition a transposable element is replicated, with one  copy of it remaining at the donor site and a new copy becoming inserted  within a different location. This produces two homologous sequences  within the same chromosome that can subsequently enter into homologous  recombination.  (a) When these homologous sequences are of the same polarity,  recombination can yield a deletion of DNA by a process that is  superficially analogous, but certainly not similar, to the reverse  reaction occurring in site­specific recombination.  (b) Inversion of DNA flanked by these transposons can result when  transposable elements are present in the chromosome oriented in  opposite direction.  Redrawn based on figure from Mathews, C. K. and Van Holde, K. E.  Biochemistry. Redwood City, CA: Benjamin/Cummings, 1990.

15.6—  Sequencing of Nucleotides in DNA Restriction Maps Give the Sequence of Segments of DNA The sequences of many genes and adjoining DNA segments have been determined for bacteria, viruses, plants, and humans. The determination of the sequence of a  large DNA molecule begins by cutting the DNA into pieces of a more manageable size with appropriate restriction endonucleases. Restriction digests permit the  construction of a characteristic restriction map for each DNA. One protocol depends on the generation of partial restriction digests of end­labeled DNA. Partial  digests are obtained by setting the conditions so that CLINICAL CORRELATION 15.10  DNA Amplification and Development of Drug Resistance An important limitation in the effectiveness of chemotoxic drugs in the treatment of cancer  is the development of drug resistance. Thus cancer cells become resistant to methotrexate  (see p. 520), an inhibitor of dihydrofolate reductase (DHFR). Drug resistance in cultured  cells results from the specific amplification of a large DNA segment that incorporates the  DHFR gene but the exact mechanism by which amplification occurs is not clear. It  appears likely that amplification results from recombination of identically oriented  homologous sequences that flank the amplified DNA. Amplification can occur by tandem  duplication of DNA that contains the DHFR gene or alternatively the DHFR­containing  segment can be excised (apparently by a recombination process), producing  extrachromosomal DNA (minichromosomes). The two mechanisms of DHFR gene  amplification are not mutually exclusive and, in fact, some resistant cells contain both  types of amplified DNA. Gene amplification is gradually reversed in the absence of methotrexate, first with the  disappearance of the extrachromosomal copies. Chromosomally amplified genes,  however, persist for several generations after removal of the drug. The amplification of  genes is a general phenomenon not limited to methotrexate or the development of cell  resistance toward other drugs. In fact, gene amplification and the accompanying  resistance extend to areas well beyond clinical medicine, as, for instance, in agriculture  with the development of pesticide­resistant insects.    

  

Page 672

CLINICAL CORRELATION 15.11  Nucleotide Sequence of the Human Genome The purpose of the Human Genome Project is to provide a detailed map of the human  genome and establish what DNA sequences determine human phenotypic characteristics  and guide human development. A corollary to this goal is to identify genes responsible for  human disease so that new approaches can be developed for diagnosis, prevention, and  therapy. The human genome is believed to consist of 70,000–100,000 different genes that  determine the genetic characteristics of every cell in the human individual. The human  genome consists of about three billion base­paired nucleotides that are assembled in the  form of 23 pairs of chromosomes. The availability of restriction endonucleases and the  development of effective physical mapping procedures for DNA, combined with the  increasing rapidity of contemporary nucleotide sequencing methods, have provided strong  impetus for the very ambitious undertaking of determining the nucleotide sequence of the  entire human genome. Extensive physical mapping has been completed. In addition, genetic mapping seeks to  locate over 500 known genetic markers on the human chromosomes. Cumulatively over  150 million base pair sequences, representing parts of the chromosome sequences of  both human DNA and that of other organisms, have been determined. Also, the  sequences of certain continuous stretches of DNA, ranging from one million to several  million base pairs in length, are being determined. Considering that the size of different  human chromosomes varies from 263 million to less than 50 million base pairs, the  determination to date of a total of about 150 million base pairs represents an important  accomplishment. It is conservatively estimated that complete sequencing of the genome  will take more than a decade and a half. Because of the routine nature of determining the nucleotide sequences involved, many  scientists have questioned the wisdom of diverting resources from perhaps more creative  scientific endeavor, to the effort required to sequence the human genome. Others have  pointed out that the project is fraught with technical uncertainties. Proponents point out  the great potential benefits of determining the imprint that controls the genetic properties  of the human cell at the highest possible level of resolution. Presently, as many as 4000  genetic diseases have been identified and many of them, namely, those inherited in  Mendelian fashion, are caused by a single mutant gene. Searching for the imprint of  human disease at the level of nucleotide sequences may permit understanding of all  disease states at the genomic level. Determination of the complete sequence appears to  be one of the prerequisites for understanding human disease at the molecular level. There  is little doubt that the sequencing of the human genome will present us with many new  challenges and opportunities in medicine. Grant Cooper, N. (Ed.). The Human Genome Project. Mill Valley, CA: University  Science Books, 1994.   the restriction endonuclease will not recognize all sites in every DNA molecule but will instead produce a digest that includes a collection of partial fragments. Double­ stranded DNA is end­labeled by treatment with alkaline phosphatase, which removes the phosphate residue at the 5  end, and then g­labeled with [32P]ATP and a  polynucleotide kinase, which incorporates the 32P into the two 5  termini of the DNA strands. Alternately, the 32P­label can be introduced at the 3  termini by the  incorporation of 32P­labeled deoxyribonucleotide triphosphates using DNA polymerase. End­labeling allows for each fragment to be identified on an electrophoresis  gel. The details of this procedure are presented on page 762. Thus, with a series of different site cuts, the fragments can be mapped directly relative to the labeled end.  Restriction maps are used for characterization of various DNAs and for ordering of smaller DNA fragments within a particular DNA sequence. Such ordering is  essential before the nucleotide sequence of large DNA molecules can be determined. Several methods have been developed for rapid sequencing of large poly­deoxyribonucleotides. They are impressively accurate. Digests obtained using different  restriction enzymes produce segments with overlapping lengths of nucleotide sequences. The accuracy of sequencing methods are increased by sequencing the  complementary strand. These procedures can also be used for sequencing of RNA molecules by prior conversion of the RNA sequence to a complementary DNA by  use of reverse transcriptase. Sequences up to 500 bp can be determined in a single automated operation and stretches of 10,000 bp, which correspond to the average  length of a gene, are now routinely determined. Clinical Correlation 15.11 discusses the application of these procedures for obtaining the sequence of the human  genome.

 

  

Page 673

Bibliography Adams, R. L. P. DNA Replication. Oxford, England: Oxford University Press; New York: IRL Press, 1991. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. Molecular Biology of the Cell, 3rd ed. New York: Garland, 1994. Bardwell, A. J., Bardwell, L., Wang, Z., Siede, W., et al. Recent insights on DNA repair: the mechanism of damaged nucleotide excision in eukaryotes and its  relationship to other cellular processes. Ann. N.Y. Acad. Sci. 726:281, 1994. Bell, S. P., and Stillman, B. ATP­dependent recognition of eukaryotic origins of DNA replication by a multiprotein complex. Nature 357:128, 1992. Bernstein, C. Aging, Sex and DNA Repair. San Diego, CA: Academic Press, 1991. Bhattacharyya, A., Murchie, A. I., von Kitzing, E., Diekmann, S., Kemper, B., and Lilley, D. M. J. A model for the interaction of DNA junctions and resolving  enzymes. J. Mol. Biol. 221:1191, 1991. Binden, R. R. DNA Structure and Function. San Diego, CA: Academic Press, 1994. Blackburn, E. H. Telomerases. Annu. Rev. Biochem. 61:113, 1992. Bootsma, D., Weeda, G., Vermeulen, W., van Vuuren, H., et al. Nucleotide excision repair syndromes: molecular basis and clinical symptoms. Philos. Trans. R. Soc.  Lond. B. Biol. Sci. 347:75, 1995. Calladine, C., and Drew, H. Understanding DNA—The Molecule and How It Works. San Diego, CA: Academic Press, 1992. Calsou, P., and Salles, B. Properties of damage­dependent DNA incision by nucleotide excision repair in human cell­free extracts. Nucleic Acids Res. 22:4937,  1994. Chiu, S. K., Rao, B. J., Story, R. M., and Radding, C. M. Interactions of three strands in joints made by RecA protein. Biochemistry 32:13146, 1993. Dianov, G., and Lindahl, T. Reconstitution of the DNA base excision­repair pathway. Curr. Biol. 4:1069, 1994. Duckett, D. R., Murchie, A. I., Giraud­Panis, M. J., Pohler, J. R., and Lilley, D. M. Structure of the four­way DNA junction and its interaction with proteins. Philos.  Trans. R. Soc. Lond. B Biol. Sci. 347:27, 1995. Echols, H., and Goodman, M. F. Fidelity mechanisms in DNA replication Annu. Rev. Biochem. 60:477, 1991. Fangman, W. L., and Brewer, B. J. Activation of replication origins in yeast chromosomes. Annu. Rev. Cell Biol. 7:375, 1991. Friedberg, E. C., Walker, G. C., and Siede, W. DNA Repair and Mutagenesis. Materials Park, OH: ASM Press, 1995. Hanawalt, P. C. Transcription­coupled repair and human disease. Science 266:1957, 1994. Hiasa, H., and Marians, K. J. Tus prevents overreplication of oriC plasmid DNA. J. Biol. Chem. 269:26959, 1994. Horiuchi, T., Fujimura, Y., Nishitani, H., Kobayashi, T., and Hidaka, M. The DNA replication fork blocked at the Ter site may be an entrance for the RecBCD  enzyme into duplex DNA. J. Bacteriol. 176:4656, 1994. Huang, J.­C., Svoboda, D. L., Reardon, J. T., and Sancar, A. Human nucleotide excision nuclease removes thymine dimers from DNA by incising the 22nd  phosphodiester bond 5  and the 6th phosphodiester bond 3  to the photodimer. Proc. Natl. Acad. Sci. USA 89:3664, 1992. Joyce, C. M., and Steitz, T. A. Function and structure relationships in DNA polymerases. Annu. Rev. Biochem. 63:777, 1994. Kamenetskii, F. Unraveling DNA. New York: VCH Publishers, 1993. Kim, M. G., Zhurkin, V. B., Jernigan, R. L., and Camerini­Otero, R. D. Probing the structure of a putative intermediate in homologous recombination: the third strand  in the parallel DNA triplex is in contact with the major groove of the duplex. J. Mol. Biol. 247:874, 1995. Kim, S. T., and Sancar, A. Photochemistry, photophysics, and mechanism of pyrimidine dimer repair by DNA photolyase. Photochem. Photobiol. 57:895, 1993. Kornberg, A. DNA Replication, 4th ed. New York: Wiley, 1992. Landy, A. Dynamic, structural and regulating aspects of site­specific recombination. Annu. Rev. Biochem. 58:913, 1989. Lewin, B. Genes V. New York: Oxford University Press, 1994. Lindsten, J., and Pettersson, U. Etiology of Human Disease at the DNA Level. New York: Raven Press, 1991. Luzio, J. P., and Thompson, R. J. In: Molecular Medical Biochemistry. New York: Cambridge University Press, 1990. Marians, K.J. Prokaryotic DNA replication. Annu. Rev. Biochem. 61:673, 1992. Matsunaga, T., Mu, D., Park, C.­H., Reardon, J. T., and Sancar, A. Human DNA repair excision nuclease. J. Biol. Chem. 270:20862, 1995. Modrich, P. Mismatch repair, genetic stability, and cancer. Science 266:1959, 1994. Parsons, C. A., Stasiak, A., Bennett, R. J., and West, B. C. Structure of a multisubunit complex that promotes DNA branch migration. Nature 374:375, 1995. Pederson, D. S. Transcription Factors and DNA Replication. Austin, TX: R. B. Landes, 1994. Richardson, C. C., and Lehman, I. R. Molecular Mechanisms in DNA Replication and Recombination. New York: Wiley, 1990. Sancar, A. Mechanisms of DNA excision repair. Science 266:1954, 1994. Schlotterer, C., and Tautz, D. Slippage synthesis of simple sequence DNA. Nucleic Acids Res. 20:211, 1992. Takahashi, M., and Nordien, B. Structure of RecA–DNA complex and mechanism of DNA strand exchange reaction in homologous recombination. Adv. Biophys.  30:1, 1994. Vos, J.­M. H. DNA Repair Mechanism: Impact on Human Disease and Cancer. New York: Demos Vermande, 1995. Wang, T. S. F. Eukaryotic DNA polymerases. Annu. Rev. Biochem. 60:513, 1991. Wolfe, S. L. Molecular and Cellular Biology. Belmont, CA: Wadsworth, 1993. Yang, S. S., Fliakas­Boltz, V., Bader, J. P., and Buckheit, R. W. Jr. Characteristics of a group of nonnucleoside reverse transcriptase inhibitors with structural  diversity and potent anti­human immunodeficiency virus activity. Leukemia 9:575, 1995. Questions C. N. Angstadt and J. Baggott 1. Which of the following statements about E. coli DNA polymerases is correct? A. All polymerases have both 3  

 5  and 5  

 3  exonuclease activity.

B. The primary role of polymerase III is in DNA repair. C. Polymerases I and III require both a primer and a template. D. Polymerase I tends to remain bound to the template until a large number of nucleotides have been added. E. The specificity of the polymerase reaction is inherent in the nature of polymerases. 2. Proofreading activity to maintain the fidelity of DNA synthesis: A. occurs after the synthesis has been completed. B. is a function of the 3  

 5  exonuclease activity of the DNA polymerases.

C. requires the presence of an enzyme separate from the DNA polymerases. D. occurs in prokaryotes but not eukaryotes. E. is independent of the polymerase activity in prokaryotes.

 

  

Page 674

3. Which of the following would result in a frameshift mutation? A. formation of ionized bases by radiation. B. substitution of a purine–pyrimidine pair by a pyrimidine–purine pair. C. intercalation of ethidium bromide into the nucleotide chain. D. deamination of cytosine to uracil. E. conversion of guanine to xanthine. 4. One way of introducing a transition mutation into DNA is by: A. using a structural analog of a base during synthesis. B. the action of an acridine dye. C. introducing a methyl group on the adenine of a GATC sequence. D. blocking the proofreading action of DNA polymerase. E. stretching the DNA helix. 5. Which of the following is (are) step(s) in excision repair mechanisms? A. excision. B. incision. C. ligation. D. all of the above. E. none of the above. 6. Base excision repair: A. is used only for bases that have been deaminated. B. uses enzymes called DNA glycosylases to generate an abasic sugar site. C. removes about 10–15 nucleotides. D. requires the action of DNA polymerase III (E. coli). E. recognizes a bulky lesion. 7. All of the following are true about nucleotide excision repair EXCEPT: A. it is deficient in the disease xeroderma pigmentosum. B. it removes thymine dimers generated by UV light. C. it involves the activity of excision nuclease, which is an endonuclease. D. it requires polymerase I (E. coli) and ligase. E. it occurs in prokaryotes but not in eukaryotes. 8. Mismatch repair: A. recognizes and removes mismatched bases during the process of replication. B. occurs only if the mismatch is on a strand containing methylated bases. C. in E. coli, recognizes mismatches within a few hundred base pairs of a GATC sequence. D. looks for a distortion where the base structure has been altered. E. is characterized by all of the above being correct. 9. Both strands of DNA serve as templates concurrently in: A. replication. B. excision repair. C. mismatch repair. D. repair catalyzed by photolyase. E. all of the above. 10. Replication: A. is semiconservative. B. requires only proteins with DNA polymerase activity. C. uses 5  

 3  polymerase activity to synthesize one strand and 3  

 5  polymerase activity to synthesize the complementary strand.

D. requires a primer in eukaryotes but not in prokaryotes. E. must begin with an incision step. 11. The discontinuous nature of DNA synthesis: A. requires that DNA polymerase III dissociate from the template when it reaches the end of each single­stranded region. B. is necessary only because synthesis is bidirectional from the initiation point. C. leads to the formation of Okazaki fragments. D. means that synthesis occurs on the second strand of DNA only after synthesis on the first strand is completed. E. means that both 3  

 5  and 5    3  polymerases are used.

12. All of the following are factors in the unwinding and separation of DNA strands for replication EXCEPT: A. the tendency of negative superhelices to partially unwind. B. destabilization of complementary base pairs by helicases. C. the action of topoisomerases. D. the enzymatic activity of SSB proteins. E. energy in the form of ATP. 13. Initiation of replication in E. coli: A. begins with dnaA binding at the OriC site if certain bases are methylated. B. results in the formation of several "bubbles," each consisting of a few nucleotide pairs. C. forms a primosome, which then uses a topoisomerase to open a replication fork. D. requires the action of helicase to initiate synthesis on the leading strands. E. begins with the formation of the replisome, followed by the formation of the primosome to begin replication. 14. In eukaryotic DNA replication: A. only one replisome forms because there is a single origin of replication. B. the leading and lagging strands are synthesized by the same enzyme. C. helicase dissociates from DNA as soon as the initiation bubble forms. D. at least one DNA polymerase has a 3  

 5  exonuclease activity.

E. the process occurs throughout the cell cycle. 15. All of the following statements about telomerase are correct EXCEPT: A. the RNA component acts as a template for the synthesis of a segment of DNA. B. it adds telomeres to the 5  ends of the DNA strands. C. it provides a mechanism for replicating the ends of linear chromosomes in most eukaryotes. D. telomerase recognizes a G­rich single strand of DNA. E. it is a reverse transcriptase.

 

  

Page 675

16. Homologous recombination: A. occurs only between two segments from the same DNA molecule. B. requires that a specific DNA sequence be present. C. requires that one of the duplexes undergoing recombination be nicked in both strands. D. may result in strand exchange by branch migration. E. is catalyzed by transposases. 17. All of the following are true about transpositions EXCEPT: A. transposons move from one location to a different one within a chromosome. B. both the donor and target sites must be homologous. C. composite transposons contain an additional gene that is not present in an insertion sequence (IS). D. transposase introduces a staggered cut in the target DNA sequence. E. transposition may either activate or inactivate a gene. Answers 1. C The primer is the initial 3  terminus of an existing strand and the template is the free portion of the complementary stand. B: Polymerase III functions in synthesis.  D: Polymerase I has low processivity because it dissociates after only a few nucleotides add. E: Specificity is a function of complementary hydrogen­bonding between  the base being added and the template (pp. 622–623). 2. B This activity removes a newly added base if there is a mismatch with the template. A: This is called repair. C: The polymerases are multifunctional enzymes. D:  Not all eukaryotic polymerases have 3    5  exonuclease activity but some do. E: The polymerase active site seems to be the one that detects the mismatch and  directs the 3  terminus to the proofreading site (p. 624). 3. C Since the bases are read in groups of three, insertion of an additional base would shift the reading frame (p. 628). Intercalation stretches the DNA so when DNA  is replicated an additional base is inserted near the intercalation site (p. 633). A: Ionized bases show atypical base pairing. B and D are both examples of base  substitution type of mutations. E: This change probably wouldn't make any difference (p. 628). 4. A 5­Bromouracil and 2­aminopurine are used to deliberately introduce mutations for research. B and E: Intercalating agents stretch the helix, allowing the insertion  of an extra base—a frameshift mutation. C: The mismatched repair system uses methylated adenine to distinguish between the old and newly synthesized strands. D:  This could lead to a mutation but not necessarily a transition (p. 628). 5. D The other step in the process is resynthesis to fill in the gap left by the actions of A and B (p. 635). 6. B These catalyze the first step of the process. A: Methylated and other chemically modified bases can also be removed. C and E: These are characteristics of a  different repair system. D: Polymerase I is the repair enzyme (p. 635). 7. E It is common to both systems. A: This is a genetic disease which requires more proteins than the prokaryotic system. B: Thymine dimers are only one cause of  bulky lesions. C and D: The excision nuclease is a complex of proteins needed to unwind the DNA and remove the lesion. The polymerase and ligase fill in the gap  (pp. 635–636). 8. C Methylated adenine in this sequence is a postreplicative event and signals the correct strand (i.e., unmethylated strand is newly synthesized). A: This is the function  of proofreading; mismatch repair is postreplicative. B: Unmethylated sequences shortly after replication denote the newly synthesized strand. D: The bases are  unaltered (p. 639). 9. A This allows for the synthesis of two identical DNA molecules. B and C: In both of these the damaged segment is removed so both strands are not available. D:  This simply disrupts the inappropriate covalent bond of thymine dimers; no synthesis is involved (pp. 642 and 646). 10. A B and D: Replication requires a primer, usually synthesized by a primase. Ligases, helicases, and other proteins are required as well. C: Replication involves  Okazaki fragments because synthesis occurs only in the 5    3  direction. E: Incision is the recognition step for DNA repair (p. 643). 11. C These are the segments of DNA built upon the primer. A: DNA polymerase remains bound to the template and slides over the next primer to continue synthesis.  B and E: This mechanism compensates for the inability to synthesize 3    5  and would be necessary even if synthesis were unidirectional. D: Both strands are  synthesized concurrently (pp. 646–647). 12. D SSB proteins stabilize the single strands after separation but have no enzymatic activity. A: This is especially true in regions of high AT pairs. B and E: This helps  in the original unwinding at the expense of ATP. C: Topoisomerases nick and reseal one of the strands to prevent the introduction of an increasing number of positive  supercoils (p. 648). 13. A Methylation seems to be a key in recognition of the OriC site. B: Escherichia coli forms only one bubble, a few hundred nucleotide pairs in size. C: The forks  form and are stabilized before primase adds. D: The negative superhelicity favors initiation but helicase is necessary for the continuation of synthesis. E: The replisome  is the final assembly and includes DNA polymerase III and rep proteins (pp. 649–651). 14. D Polymerase a  shows this activity that provides proofreading during synthesis.   and   have this for proofreading. A: There are multiple initiation sites. The DNA  segments between two initiation points are called replicons (p. 654). B: In prokaryotes, the DNA polymerase III does both; in eukaryotes   synthesizes the leading  strand and a  the lagging strand, at least for the initiation process. C: Helicase activity is also necessary for the continuation of synthesis, that is, the opening of the forks  (p. 655). E: Replication is confined to the S phase (p. 654). 15. B It is the 3  end of each strand that cannot be conventionally replicated. A and C: Telomerase both positions itself at the 3  ends of the DNA and provides the  template for extending that end (p. 660, Figure 15.40). D: This is a characteristic of the 3 ­end. E: It is using an RNA template to synthesize DNA (p. 659). 16. D This is just one of the events in this complex process. A and B: It may occur between two distinct DNA molecules; the

 

  

Page 676

requirement is that the two sequences be homologous but not that they be specific sequences. C: The nicks are usually on a single strand. E: These are the enzymes of  transpositional site­specific recombination (p. 663). 17. B Only the donor site requires a specific nucleotide sequence; homology is not required. A: This is the definition. C: The IS contains the gene for transposon plus  the flanking sequences; composite transposons have an additional gene—for example, one that confers antibiotic resistance in bacteria. D: This permits the transposon  to be inserted. E: Insertion into the middle of a gene would inactivate it; insertion of a promoter next to a gene may activate it (pp. 670–671).

 

  

Page 677

Chapter 16—  RNA:  Structure, Transcription, and Processing Francis J. Schmidt

16.1 Overview

678

16.2 Structure of RNA

679

RNA Is a Polymer of Ribonucleoside 5 ­Monophosphates

679

Secondary Structure of RNA Involves Intramolecular Base Pairing

680

RNA Molecules Have Tertiary Structures

680

16.3 Types of RNA

681

Transfer RNA Has Two Roles: Activating Amino Acids and Recognizing  Codons in mRNA

682

Ribosomal RNA Is Part of the Protein Synthesis Apparatus

683

Messenger RNAs Carry the Information for the Primary Structure of  Proteins

684

Mitochondria Contain Unique RNA Species

685

RNA in Ribonucleoprotein Particles

686

Some RNAs Have Catalytic Activity

686

RNAs Can Form Binding Sites for Other Molecules

688

16.4 Mechanisms of Transcription

689

The Initial Process of RNA Synthesis Is Transcription

689

The Template for RNA Synthesis Is DNA

690

RNA Polymerase Catalyzes the Transcription Process

690

The Steps of Transcription in Prokaryotes Have Been Determined

692

Initiation

695

Elongation

695

Termination

696

Transcription in Eukaryotes Involves Many Additional Molecular  Events

696

The Nature of Active Chromatin

697

Enhancers

697

Transcription of Ribosomal RNA Genes

697

Transcription by RNA Polymerase II

698

Promoters for mRNA Synthesis

699

Transcription by RNA Polymerase III

699

16.5 Posttranscriptional Processing

699

Transfer RNA Precursors Are Modified by Cleavage, Additions, and  Base Modification Cleavage

700

Additions

700

Modified Nucleosides

701

Ribosomal RNA Processing Releases the Various RNAs from a Longer  Precursor

702

Messenger RNA Processing Requires Maintenance of the Coding  Sequence

703

Blocking of the 5  Terminus and Poly(A) Synthesis

704

Removal of Introns from mRNA Precursors

705

Mutations in Splicing Signals Cause Human Diseases

706

Alternate pre­mRNA Splicing Can Lead to Multiple Proteins Being  Made from a Single DNA Coding Sequence

706

16.6 Nucleases and RNA Turnover

708

Bibliography

709

Questions and Answers

710

Clinical Correlations

   

  

700

 

16.1 Staphylococcal Resistance to Erythromycin

683

16.2 Antibiotics and Toxins That Target RNA Polymerase

692

16.3 Fragile X Syndrome: A Chromatin Disease?

697

16.4 Involvement of Transcriptional Factors in Carcinogenesis

701

16.5 Thalassemia Due to Defects in Messenger RNA Synthesis

705

16.6 Autoimmunity in Connective Tissue Disease

706

Page 678

16.1—  Overview The primary information store of a cell is its genetic complement, that is, its DNA. DNA information is exactly analogous to the master copies of a computer program  or any database: It is the source of cellular information and therefore must be kept as error­free as possible. Chapter 14 has detailed some of the elaborate  mechanisms that are employed to keep DNA information intact from one cell generation to the next. This chapter describes another type of information transfer that  helps to ensure the integrity of genomic information. Just as a careful computer programmer makes working copies of a program or data set, a cell makes  macromolecular copies of the information in DNA. These macromolecules, ribonucleic acids (RNAs), are linear polymers of ribonucleoside monophosphates. The  sequence of DNA is copied exactly into RNA. The process by which RNA copies of selected DNA sequences are made is termed transcription. The primary role of  RNA within the cell is its involvement in protein synthesis, that is, translation. The overall process of information transfer in the cell is therefore given by the so­called central dogma of molecular biology:

TABLE 16.1 Characteristics of Cellular RNAs Abbreviation

Messenger RNA  Cytoplasmic

mRNA

Transfer of genetic  Depends on size of  information from  protein 1000–10,000  nucleus to cytoplasm, or  nucleotides from gene to ribosome

Nucleoplasm

Blocked 5  end; poly(A) tail  on 3  end; nontranslated  sequences before and after  coding regions; few base  pairs and methylations

mt mRNA

 

Mitochondria

 

tRNA

Transfer of amino acids  65–110 nucleotides  4S to mRNA–ribosome  complex and correct  sequence insertion

Nucleoplasm

Highly base paired; many  modified nucleotides;  common specific structure

Mitochondrial Transfer RNA  Cytoplasmic

9S–40S

mt tRNA

 

Ribosomal RNA  Cytoplasmic

rRNA

Structural framework for  28S, 5400 nucleotides  18S, 2100 nucleotides  ribosomes 5.8S, 158 nucleotides  5S, 120 nucleotides

Mitochondrial

mt rRNA

 

Heterogeneous  nuclear RNA

hnRNA

Small nuclear RNA

Small cytoplasmic  RNA  [7S(L) RNA]

Mitochondrial

   

  

Function

Size and Sedimentation  Coefficient

Type of RNA

Site of Synthesis

Structural Features

Mitochondria

 

Nucleolus  Nucleolus  Nucleolus  Nucleoplasm

5.8S and 5S highly base  paired; 28S and 18S have  some base paired regions and  some methylated nucleotides

16S, 1650 nucleotides  12S, 1100 nucleotides

Mitochondria

 

Some are precursors to  mRNA and other RNAs

Extremely variable 30S– 100S

Nucleoplasm

mRNA precursors may have  blocked 5  ends and 3 ­poly (A) tails; many have base  paired loops

snRNA

Structural and  regulatory RNAs in  chromatin

100–300 nucleotides

Nucleoplasm

 

scRNA

Selection of proteins for  129 nucleotides export

3.2S–4S

Cytosol and rough  Associated with proteins as  endoplasmic  part of signal recognition  reticulum particle

Page 679

RNA information is occasionally reverse transcribed into DNA, a process important in the life cycle of infectious retroviruses such as the human immunodeficiency  virus (HIV), which causes the acquired immunodeficiency syndrome (AIDS). Reverse translation of protein sequence into nucleic acid sequence information, however,  does not occur in nature. RNA molecules are classified according to the roles they play in information transfer processes (Table 16.1). In prokaryotes, transcription and translation occur close  together; in fact, ribosomes can begin translating a mRNA while it is still being synthesized. In eukaryotes, these processes are spatially separated: transcription occurs  in the nucleus and translation in the cytoplasmic portions of the cell. Messenger RNAs (mRNA) serve as templates for the synthesis of protein; they carry  information from the DNA to the cellular protein synthetic machinery. Here a number of other RNA species contribute to the synthesis of the peptide bond. The molecules that transfer specific amino acids from soluble amino acid pools to ribosomes, and ensure the alignment of these amino acids in the proper sequence  prior to peptide bond formation, are transfer RNAs (tRNA). All tRNA molecules are approximately the same size and shape. The assembly site, or factory, for  peptide synthesis involves ribosomes. These complex subcellular particles contain three or four ribosomal RNA (rRNA) molecules and 70–80 ribosomal proteins. Protein synthesis requires a close interdependent relationship between mRNA, the informational template, tRNA, the amino acid adaptor molecule, and rRNA, part of  the synthetic machinery. In order for protein synthesis to occur at the correct time in a cell's life, the syntheses of mRNA, tRNA, and rRNA must be coordinated with  the cell's response to the intra­ and extracellular environments. All cellular RNA is synthesized on a DNA template and reflects a portion of the DNA base sequence. Therefore all RNA is associated with DNA at some time.  Although DNA is the more prevalent genetic store of information, RNA can also carry genetic information. Genomic RNA is found in the RNA tumor viruses and the  other small RNA viruses, such as poliovirus and reovirus. 16.2—  Structure of RNA RNA Is a Polymer of Ribonucleoside 5¢ ­Monophosphates Chemically, RNA is similar to DNA. Although RNA is one of the more stable components within a cell, it is not as stable as DNA. The presence of the adjacent 2 ­ hydroxyl group makes the RNA phosphodiester bond more susceptible to chemical and enzymatic hydrolysis than its DNA counterpart. Some RNAs, such as  bacterial mRNA, are synthesized, used, and degraded within minutes, whereas others, such as rRNA, are more stable metabolically. RNA is an unbranched linear polymer of ribonucleoside monophosphates. The purines found in RNA are adenine and guanine; the pyrimidines are cytosine and  uracil. Except for uracil, which replaces thymine, these are the same bases found in DNA. A, C, G, and U nucleotides are incorporated into RNA during transcription. Many RNAs also contain modified nucleotides, which are synthesized after  transcription. Modified nucleotides are especially characteristic of stable RNA species (i.e., tRNA and rRNA); however, some methylated nucleotides are also present  in eukaryotic mRNA. For the most part, the functions of the modified nucleotides in RNA have not been identified. Where known, the function of nucleotide  modification seems to involve ''fine tuning" rather than an indispensable role in the cell. The 3 ,5 ­phosphodiester bonds of RNA form a chain or backbone from

 

  

Page 680

which the bases extend (Figure 16.1). Eukaryotic RNAs vary from approximately 65 nucleotides long to more than 200,000 nucleotides long. RNA sequences are  complementary to the base sequences of specific portions of only one strand of DNA. Thus, unlike the base composition of DNA, molar ratios of A + U and G + C in  RNA are not equal. All cellular RNA so far examined is linear and single stranded, but double­stranded RNA is present in some viral genomes.

Figure 16.1  Structure of the 3¢ ,5¢ ­phosphodiester  bonds between ribonucleotides forming  a single strand of RNA.  The phosphate joins the 3 ­OH group of one  ribose with the 5 ­OH group of the next ribose.  This linkage produces a polyribonucleotide  having a sugar–phosphate "backbone." The  purine and pyrimidine bases extend away  from the axis of the backbone and may pair  with complementary bases to form double  helical base paired regions.

Secondary Structure of RNA Involves Intramolecular Base Pairing RNA, being single stranded rather than double stranded, does not usually form an extensive double helix. Rather, the structure in an RNA molecule arises from  relatively short regions of intramolecular base pairing. Considerable helical structure exists in RNA even in the absence of extensive base pairing, for example, in the  portions of an RNA that do not form intramolecular Watson–Crick base pairs. This helical structure is due to the strong base­stacking forces between A, G, and C  residues. Base stacking is more important than simple hydrogen bonding in determining inter­ and intramolecular interactions. These forces act to restrict the possible  conformations of an RNA molecule (Figure 16.2). RNA helical structures generally are of the "A type" with 11 nucleotides per turn in a double helix. Double helical regions in RNA are often called "hairpins." There are considerable variations in the fine structural details of "hairpin" structures, including the length of  base paired regions and the size and number of unpaired loops (Figure 16.3). Transfer RNAs are excellent examples of base stacking and hydrogen bonding in a  single­stranded molecule (Figure 16.4a). About 60% of the bases are paired in four double helical stems. In addition, the unpaired regions have the capability to form  base pairs with free bases in the same or other looped regions, thereby contributing to the molecule's tertiary structure. The anticodon region in tRNA is an unpaired,  base­stacked, loop of seven nucleotides. The partial helix caused by base stacking in this loop binds, by specific base pairing, to a complementary codon in mRNA so  that translation (peptide bond formation) can occur. RNA Molecules Have Tertiary Structures The actual functioning structures of RNA molecules are more complex than the base­stacked and hydrogen­bonded helices mentioned above. RNAs in vivo are

Figure 16.2  Helical structure of tRNA.  Models indicating a helical structure due to  (a) base stacking in the CCA terminus of  tRNA and  (b) the lack of an ordered helix  when no stacking occurs in this non­base  paired region.  Redrawn from Sprinzl, M., and Cramer,  F. Prog. Nucl. Res. Mol. Biol. 22:9,  1979.

 

  

Page 681

Figure 16.3  Proposed base pairing regions in the mRNA for mouse immunoglobulin light chain.  Base paired structures shown have free energies of at least –5 kcal. Note the variance in loop size and length  of paired regions.  Redrawn from Hamlyn, P. H., Browniee, G. G., Cheng, C. C., Gait, M. J., and Milstein, C. Cell 15:1067, 1978.

dynamic molecules that undergo changes in conformation during synthesis, processing, and functioning. Proteins associated with RNA molecules often lend stability to  the RNA structure; in fact, it is perhaps more correct to think of RNA–protein complexes rather than naked RNA molecules as functioning components of the cell. In  addition to the secondary, base paired structure, RNA molecules also form other hydrogen bonds to form the tertiary structure of the molecule. Again, the structure  of tRNA provides a number of examples. In solution, tRNA is folded into a compact "L­shaped" conformation (Figure 16.4b). The arms and loops are folded in  specific conformations held in position not only by Watson–Crick base pairing, but also by base interactions involving more than two nucleotides. Bases can donate  hydrogen atoms to bond with the phosphodiester backbone. The 2 ­OH of the ribose is an important donor and acceptor of hydrogens. All these interactions  contribute to the folded shape of an RNA molecule. 16.3—  Types of RNA RNA molecules are traditionally classified as transfer, ribosomal, and messenger RNAs according to their usual function; however, we now know that RNA molecules  perform or facilitate a variety of other functions in a cell.

 

  

Page 682

Figure 16.4  Coverleaf structure of tRNA  (a) Cloverleaf diagram of the two­dimensional structure and nucleotide sequence of yeast tRNAPhe. Red lines connecting nucleotides  indicate hydrogen­bonded residues. Insertion of nucleotides in the D loop occurs at positions a  and b  for different tRNAs.  (b) Tertiary folding of the cloverleaf structure in (a). Hydrogen bonds are indicated by cross rungs.  Redrawn with permission from Quigley, G. J., and Rich, A. Science 194:797, 1976.  Copyright © 1976 by the American Association for the Advancement of Science.

Transfer RNA Has Two Roles:  Activating Amino Acids and Recognizing Codons in mRNA About 15% of the total cellular RNA is tRNA. Transfer RNA has two functions that are essential for its cellular role as an "adapter" of nucleic acid to protein  information. First, tRNA molecules activate amino acids for protein synthesis so that formation of peptide bonds is energetically favored. The activated amino acid is  transported to the polyribosome where it is transferred to the growing peptide chain (hence tRNA's name). The second function of tRNA is to recognize codons in  mRNA to ensure that the correct amino acid is incorporated into the growing peptide chain. These two functions are reflected in the fact that tRNAs have two primary  active sites, the 3 ­OH terminal CCA, to which specific amino acids are attached enzymatically, and the anticodon triplet, which base pairs with mRNA codons. Each tRNA can transfer only a single amino acid. Although there are only 20 amino acids used in protein synthesis, free­living organisms synthesize a larger set of  tRNAs. For example, analysis of the recently determined genomic sequence of Haemophilus influenzae identified genes for 54 tRNA species. Mitochondria  synthesize a much smaller number of tRNAs. Transfer RNAs that accept the same amino acid are called isoacceptors. A tRNA that accepts phenylalanine would be  written as tRNAPhe, whereas one accepting tyrosine would be written tRNATyr.

 

  

Page 683

CLINICAL CORRELATION 16.1  Staphylococcal Resistance to Erythromycin Bacteria exposed to antibiotics in a clinical or agricultural setting often develop resistance  to the drugs. This resistance can arise from a mutation in the target cell's DNA, which  gives rise to resistant descendants. An alternative and clinically more serious mode of  resistance arises when plasmids coding for antibiotic resistance proliferate through the  bacterial population. These plasmids may carry multiple resistance determinants and  render several antibiotics useless at the same time. Erythromycin inhibits protein synthesis by binding to the large ribosomal subunit.  Staphylococcus aureus can become resistant to erythromycin and similar antibiotics as a  result of a plasmid­borne RNA methylase that converts a single adenosine in 23S rRNA  to N6­dimethyladenosine. Since the same ribosomal site binds lincomycin and  clindamycin, the plasmid causes cross­resistance to these antibiotics as well. Synthesis of  the methylase is induced by erythromycin. The microorganism that produces an antibiotic must also be immune to it or else it would  be inhibited by its own toxic product. The producer of erythromycin, Streptomyces  erythreus, itself possesses an rRNA methylase that acts at the same ribosomal site as the  one from S. aureus. Which came first? It is likely that many of the resistance genes in target organisms evolved  from those of producer organisms. In several cases, DNA sequences from resistance  genes of the same specificity are conserved between producer and target organisms. We  may therefore look on plasmid­borne antibiotic resistance as a case of "natural genetic  engineering," whereby DNA from one organism (e.g., the Streptomyces producer) is  appropriated and expressed in another (e.g., the Staphylococcus target). Cundliffe, E. How antibiotic­producing microorganisms avoid suicide. Annu. Rev.  Microbiol. 43:207, 1989.   Transfer RNAs range from 65 to 110 nucleotides in length, corresponding to a molecular weight range of 22,000–37,000. The sequences of all tRNA molecules (over  1000 are known) can be arranged into a common secondary structure that has the appearance of a cloverleaf. The cloverleaf structure is determined by  complementary Watson–Crick base pairs forming three stem and loop or hairpin structures. The anticodon triplet sequence is at one "leaf" of the cloverleaf while the  CCA acceptor stem is at the "stem" (see Figure 16.4). This arrangement where the two active sites of a tRNA are spatially separated is preserved in the tertiary  structure of tRNAPhe shown in Figure 16.4. Additional, non­Watson–Crick, hydrogen bonds form in the L­shaped molecule. The nucleotide sequence and structure of the tRNAPhe molecule depicted in Figure 16.4 show that tRNAs have several modified nucleotides. The modified nucleotides  affect tRNA structure and stability but are not required for the formation or maintenance of tertiary conformation. For example, a modified base in the anticodon loop  makes codon recognition more efficient but a tRNA without this modification can still be read correctly by the ribosome. Many structural features are common to all tRNA molecules. Seven base pairs are present in the amino acid acceptor stem, which terminates with the nucleotide triplet  CCA. This CCA triplet is not base paired. The dihydrouracil or "D" stem has three or four base pairs, while the anticodon and T stems have five base pairs each. Both  the anticodon loop and T loop contain seven nucleotides. Differences in the number of nucleotides in different tRNAs are accounted for by the variable loop. Thus  80% of tRNAs have small variable loops of 4–5 nucleotides, while others have larger loops of 13–21 nucleotides. The positions of some nucleotides are constant in all  tRNAs (see Figure 16.4a). Ribosomal RNA Is Part of the Protein Synthesis Apparatus Protein synthesis takes place on ribosomes. These complex assemblies are composed in eukaryotes of four RNA molecules, representing about two­thirds of the  particle mass, and 82 proteins. The smaller subunit, the 40S particle, contains one 18S RNA and 33 proteins. The larger subunit, the 60S particle, contains the 28S,  the 5.8S, and the 5S rRNAs and 49 proteins. The total assembly is called the 80S ribosome. Prokaryotic ribosomes are somewhat smaller: the 30S subunit contains  a single 16S rRNA and 21 proteins, while the larger subunit (70S) contains 5S and 23S rRNAs as well as 34 ribosomal proteins. The rRNAs account for 80% of the total cellular RNA and are metabolically stable. This stability, required for repeated functioning of the ribosome, is enhanced by  close association with the ribosomal proteins. The 28S (4718 nucleotides), 18S (1874 nucleotides), and 5.8S (160 nucleotides) rRNAs are synthesized in the  nucleolar region of the nucleus. The 5S rRNA (120 nucleotides) is not transcribed in the nucleolus but rather from separate genes within the nucleoplasm (Figure 16.5).  Processing of the rRNAs (see Section 16.5) includes cleavage to the functional size, internal base pairing, modification of particular nucleotides, and association with  ribosomal proteins to form a stable tertiary conformation. The larger rRNAs contain most of the altered nucleotides found in rRNA. These are primarily methylations on the 2  position of the ribose, yielding 2 ­O­ methylribose. Methylation of rRNA has been directly related to bacterial antibiotic resistance in a pathogenic species (see Clin. Corr. 16.1). A small number of N6­ dimethyladenines are present in 18S rRNA. The 28S rRNA has about 45 methyl groups and the 18S rRNA has 30 methyl groups. Biochemical studies of ribosome function indicate that rRNA molecules are more than macromolecular scaffolds for enzymatic proteins. The exact extent to which  rRNA participates in protein biosynthetic reactions is the subject of current investigation. Several lines of evidence indicate that the actual formation of a peptide bond  may be catalyzed by the large RNA subunit of the ribosome.

 

  

Page 684

Figure 16.5  Secondary, base paired, structure proposed for 5S rRNA.  Arrows indicate regions protected by proteins in the large ribosomal subunit.  Combined information from Fox, G. E., and Woese, C. R.  Nature 256:505, 1975; and R. A. Garrett and P. N. Gray.

Messenger RNAs Carry the Information for the Primary Structure of Proteins The mRNAs are the direct carriers of genetic information from genomes to the ribosomes. Each eukaryotic mRNA is monocistronic; that is, it contains information  for only one polypeptide chain. In prokaryotes, mRNA species often encode more than one protein in a polycistronic molecule. A cell's phenotype and functional  state are related directly to its mRNA content. In the cytoplasm mRNAs have relatively short life spans. Some mRNAs are known to be synthesized and stored in an inactive or dormant state in the cytoplasm,  ready for a quick protein synthetic response. An example of this is the unfertilized egg of the African clawed toad, Xenopus laevis. Immediately upon fertilization the  egg undergoes rapid protein synthesis in the absence of transcription, indicating the presence of preformed mRNA. Eukaryotic mRNAs have unique structural features not found in rRNA or tRNA (see Figure 16.6). Since the information within mRNA lies in the linear sequence of  the nucleotides, the integrity of this sequence is extremely im­

Figure 16.6  General structure for a eukaryotic mRNA.  There is a "blocked" 5  terminus (cap) followed by the nontranslated leader containing a  promoter sequence. The coding region usually begins with the initiator codon AUG and  continues to the translation termination sequence UAG, UAA, or UGA. This is followed  by the nontranslated trailer and a poly(A) tail on the 3  end.

 

  

Page 685

Figure 16.7  Diagram of the "cap" structure or blocked 5¢  terminus in mRNA.  The 7­methylguanosine is inverted to form a 5 ­phosphate to 5 ­phosphate linkage  with the first nucleotide of the mRNA. This nucleotide is often a methylated purine.

portant. Any loss or change of nucleotides could alter the protein being translated. The translation of mRNA on the ribosomes must also begin and end at specific  sequences. Structurally, starting from the 5  terminus, eukaryotic mRNA is capped with an inverted methylated base attached via 5¢­phosphate–5¢­phosphate bonds  rather than the usual 3 ,5 ­phosphodiester linkages. The cap is attached to the first transcribed nucleotide, usually a purine, methylated on the 2 ­OH of the ribose (see  Figure 16.7). The cap is followed by a nontranslated or "leader" sequence to the 5  side of the coding region. Following the leader sequence are the initiation  sequence or codon, most often AUG, and the translatable coding region of the molecule. At the end of the coding sequence is a termination sequence signaling  termination of polypeptide formation and release from the ribosome. A second nontranslated or "trailer" sequence follows, terminated by a string of 20–200 adenine  nucleotides, called a poly(A) tail, which makes up the 3  terminus of the mRNA. The 5  cap has a positive effect on the initiation of message translation. In the initiation of translation of a mRNA, the cap structure is recognized by a single ribosomal  protein, an initiation factor (see Chapter 17). The poly(A) sequence is correlated with the stability of the mRNA molecule; for example, histone mRNA molecules  lack a poly(A) tail and are also present in the cell only transiently. Mitochondria Contain Unique RNA Species Mitochondria (mt) have their own protein­synthesizing apparatus, including ribosomes, tRNAs, and mRNAs. The mt rRNAs, 12S and 16S, are transcribed from the  mitochondrial DNA (mt DNA), as are 22 specific tRNAs and 13 mRNAs, most of which encode proteins of the electron transport chain and ATP synthetase. Note  that there are fewer mt tRNAs than prokaryotic or cytoplasmic tRNA species; there is only one mt tRNA species per amino acid. The mt RNAs account for 4% of  the total cellular RNA. They are transcribed by a mitochondrial­specific RNA polymerase and are processed from a pair of mt RNA precursors. Each precursor is an  exact copy of the entire mitochondrial genome, complementary to either the heavy (H) or light (L) strand of mt DNA. Genes for 12 tRNAs are located on the heavy  mt DNA strand and 7 on the light strand. Some of the mRNAs have eukaryotic characteristics, such as 3 ­poly(A) tails. A large degree of coordination exists between  the nuclear and mitochondrial genomes. Most of the aminoacylating enzymes for the mt tRNAs and all of the mitochondrial ribosomal proteins are specified by nuclear  genes, translated in the cytoplasm

 

  

Page 686

and transported into the mitochondria. The modified bases in mt tRNA species are synthesized by enzymes encoded in nuclear DNA. RNA in Ribonucleoprotein Particles Besides tRNA, rRNA, and mRNA, small, stable RNA species can be found in the nucleus, cytoplasm, and mitochondria. These small RNA species function as  ribonucleoprotein particles (RNPs), with one or more protein subunits attached. Different RNP species have been implicated in RNA processing, splicing, transport,  and control of translation, as well as in the recognition of proteins due to be exported. The actual roles of these species, where known, are described more fully in the  discussion of specific metabolic events. Some RNAs Have Catalytic Activity RNA can be an enzyme. In several cases the RNA component of a ribonucleoprotein particle has been shown to be the catalytically active subunit of the enzyme. In  other cases, in vitro catalytic reactions can be carried out by RNA in the absence of any protein. Enzymes whose RNA subunits carry out catalytic reactions are  called ribozymes. There are four classes of ribozyme. Three of these RNA species carry out self­processing reactions while the fourth, ribonuclease P (RNase P),  is a true catalyst. In the ciliated protozoan Tetrahymena thermophila, an intron in the rRNA precursor is removed by a multistep reaction (Figure 16.8). A guanosine nucleoside or  nucleotide reacts with the intron–exon phosphodiester linkage to displace the donor exon from the intron. This reaction, a transesterification, is promoted by the folded  intron itself. The free donor exon then similarly attacks the intron–exon phosphodiester bond at the acceptor end of the intron. Introns of this type (Group I introns)  have been found in a variety of genes in fungal mitochondria and in the bacteriophage T4. Although these introns are not true enzymes in vivo because they only work  for one reaction cycle, they can be made to carry out catalytic reactions under specialized conditions. Group II self­splicing introns are found in the mitochondrial RNA precursors of yeasts and other fungi. The self­splicing of these introns proceeds through a lariat  intermediate similar to the lariat intermediate in the splicing of nuclear mRNA precursors (see below). Since this reaction is carried out by a ribozyme the catalytic  activity of the small nucleus ribonucleoproteins (snRNPs) involved in nuclear mRNA splicing may also reside in the RNA component. A third class of self­cleaving RNAs is found in the genomic RNAs of several plant viruses. These RNAs self­cleave during the generation of single genomic RNA  molecules from large multimeric precursors. The three­dimensional structure of the hammerhead ribozyme, a member of this third class, has recently been  determined (Figure 16.9). Catalysis is carried out by a bound Mg2+ ion positioned near the bond to be cleaved in the folded ribozyme structure. The phosphate of the  cleaved bond is left at the 3  hydroxyl position of the RNA product. A self­cleaving RNA is found in a small satellite virus, hepatitis delta virus, that is implicated in  severe cases of human infectious hepatitis. All of the above self­processing RNAs can be made to act as true catalysts (i.e., exhibiting multiple turnover) in vitro and in  vivo. Ribonuclease P contains both a protein and an RNA component. It acts as a true enzyme in the cell, cleaving tRNA precursors to generate the mature 5  end of the  tRNA molecule. RNase P recognizes constant structures associated with tRNA precursors (e.g., the acceptor stem and CCA sequence) rather than using extensive  base pairing to bind the substrate RNA to the ribozyme. The product of cleavage contains a 5  phosphate in contrast to the products of hammerhead and similar  RNAs. In all of these events the structure of the catalytic RNA is essential for intramolecular or enzyme catalysis.

 

  

Page 687

Figure 16.8  Mechanism of self­splicing of the rRNA precursor of Tetrahymena.  The two exons of the rRNA are denoted by dark blue.  Catalytic functions reside in the intron, which is purple. This  splicing function requires an added guanosine nucleoside or  nucleotide.  Reproduced from Cech, T. R. JAMA 260:308, 1988.

 

  

Page 688

Figure 16.9  "Hammerhead" structure of viral RNA  (a) The hammerhead structure of a self­cleaving  viral RNA. This artificial molecule is formed by  the base pairing of two separate RNAs. The  cleavage of the RNA sequence at the site indicated  by the arrow in the top strand requires its base  pairing with the sequence at the bottom of the  molecule. The boxed nucleotides are a consensus  sequence found in self­cleaving viral RNAs.  (b) The three­dimensional folding of the hammerhead  catalytic RNA. The star indicates the position of  the cleaved bond while M indicates a binding site  for a metal ion. Helices II and III stack to form  an apparently continuous helix while  non­Watson–Crick interactions position the  noncomplementary bases in the hammerhead into a  "uridine turn" structure identical to that found  in tRNA.  Part (a) redrawn from Sampson, J. R.,  Sullivan, F. X., Behlen, L. S., DiRenzo, A. B.,  and Uhlenbeck, O. C. Cold Spring Harbor Symp.  Quant. Biol. 52:267, 1987;  part (b) redrawn from Pley, H. W., Flaherty,  K. M., and McKay, D. B. Nature 372:68 1994.

The discovery of RNA catalysis has greatly altered our concepts of biochemical evolution and the range of allowable cellular chemistry. First, we now recognize that  RNA can serve as both a catalyst and a carrier of genetic information. This has raised the possibility that the earliest living organisms were based entirely on RNA and  that DNA and proteins evolved later. This model is sometimes referred to as the "RNA world." Second, we know that many viruses, including human pathogens, use  RNA genetic information; some of these RNAs have been shown to be catalytic. Thus catalytic RNA presents opportunities for the discovery of RNA­based  pharmaceuticals. Third, many of the information processing events in protein synthesis and mRNA splicing require RNA components. These RNAs may also be  fulfilling a catalytic function. RNAs Can Form Binding Sites for Other Molecules Consideration of the RNA world has led to a new type of biological chemistry based on the large number of potential sequences (4N) that would be made if A, C, G,  or U were inserted randomly in each of N positions in a nucleic acid. A set of chemically synthesized, randomized, nucleic acid molecules 25 nucleotides long would  contain 425 = 1015 potential members. Individual molecules within this large collection of RNAs would be expected to fold into a similarly large collection of shapes.  The large number of molecular shapes implies that some member of this collection will be capable of strong, specific binding to any ligand, much as group I introns bind  guanosine nucleotides specifically. Though a single molecule would be too rare to study within the original population, the RNA capable of binding can be selected and  preferentially replicated in vitro. In one case, for example, an RNA capable of distinguishing theophylline from caffeine was selected from a complex population  (see

 

  

Page 689

Figure 16.10  Structures of theophylline and caffeine.  Although these compounds differ  only by a single methyl group, a specific  synthetic RNA can bind to theophylline  10,000­fold more tightly than to caffeine.

Figure 16.10). Theophylline is used in the treatment of chronic asthma but the level must be carefully controlled to avoid side effects. The monitoring of theophylline by  conventional antibody­based clinical chemistry is difficult because caffeine and theophylline differ only by a single methyl group. Therefore anti­theophylline antibodies  show considerable cross­reaction with caffeine. RNA molecules have been found that bind theophylline 10,000­fold more tightly than caffeine. Other extensions of the technology have used selection procedures to identify new, synthetic ribozymes and potential therapeutic RNAs. 16.4—  Mechanisms of Transcription The Initial Process of RNA Synthesis Is Transcription The process by which RNA chains are made from DNA templates is called transcription. All known transcription reactions take the following form:

Enzymes that catalyze this reaction are designated RNA polymerases; it is important to recognize that they are absolutely template dependent. In contrast to DNA  polymerases, however, RNA polymerases do not require a primer molecule. The energetics favoring the RNA polymerase reaction are twofold: first, the 5  a ­ nucleotide phosphate of the ribonucleoside triphosphate is converted from a phosphate anhydride to a phosphodiester bond with a change in free energy ( G ) of  approximately 3 kcal (12.5 kJ) mol–1 under standard conditions; second, the released pyrophosphate, PPi, can be cleaved into two phosphates by pyrophosphatase  so that its concentration is low and phosphodiester bond formation is more favored relative to standard conditions (see Chapter 6 for a fuller discussion of metabolic  coupling). Since a DNA template is required for RNA synthesis, eukaryotic transcription takes place in the cell nucleus or mitochondrial matrix. Within the nucleus, the nucleolus  is the site of rRNA synthesis, whereas mRNA and tRNA are synthesized in the nucleoplasm. Prokaryotic transcription is accomplished on the cell's DNA, which is  located in a relatively small region of the cell. In the case of prokaryotic plasmids, the DNA template need not be associated with the chromosome. Structural changes occur in DNA during its transcription. In the polytene chromosomes of Drosophila, transcriptionally active genes are visualized in the light  microscope as puffs distinct from the condensed, inactive chromatin. Furthermore, the nucleosome patterns of active genes are disrupted so that active chromatin is  more accessible to, for example, DNase attack. In prokaryotes and eukaryotes, the DNA double helix is transiently opened (unwound) as the transcription complex  proceeds down the DNA. These openings and unwindings are a manifestation of a topological necessity. If the RNA chain were copied off DNA without this unwinding, the transcription  complex and growing end of the RNA chain would have to wind around the double helix once every 10 base pairs as they travel from the beginning of the gene to its  end. Such a process would wrap the newly synthesized RNA chain around the DNA double helix. Local opening and unwinding of the DNA solves this problem  before it occurs by allowing transcription to proceed on a single face or side of the DNA. In addition, the opening of DNA base pairs during transcription allows  Watson–Crick base pairing between template DNA and the bases in the newly synthesized RNA. The process of transcription is divided into three parts: initiation refers to the recognition of an active gene starting point by RNA polymerase and the beginning of the  bond formation process; Elongation is the actual synthesis of the RNA chain and is followed by chain termination and release.

 

  

Page 690

Figure 16.11  Determination of a consensus sequence for prokaryotic promoters.  A portion of the data set used for the identification of the consensus sequence for E. coli promoter activity.  The –10 region (sometimes called the Pribnow box) is shaded in red and the –35 region nucleotides are colored.  Note that none of the individual promoters has the entire consensus sequence.  Modified from Rosenberg, M., and Court, D. Ann. Rev. Genet. 13:319, 1979.

The Template for RNA Synthesis Is DNA Each cycle of transcription begins and ends with the recognition of specific sites in the DNA template. The DNA sequencing of a large number of transcription start  regions, called promoters, has shown that certain conserved sequences occur in promoters with great regularity. An example is shown in Figure 16.11. Similar considerations demonstrate that termination occurs at different conserved sequences. In addition, sites within a transcript  may allow premature termination of transcription. These sites can act as molecular switches affecting the continuation of synthesis of an RNA molecule. Conserved sequences near the transcription start are found for both prokaryotic and some eukaryotic promoters. In addition, eukaryotic transcription has been shown  in some cases to be affected by internal promoter elements and other sequences called enhancers. Enhancers are gene­specific sequences that positively affect  transcription. Enhancer sequences can stimulate transcription whether they are located at the beginning, in the middle, or at the end of a gene. An enhancer sequence  must be on the same DNA strand as the transcribed gene (genetically in a cis position) but can function in either orientation. Cellular protein factors are known that  specifically bind different enhancers. The most likely hypothesis is that protein factors bound to enhancers cause a structural change in the DNA template, allowing  protein–protein interaction with other factors or with RNA polymerase itself. This interaction facilitates transcription. RNA Polymerase Catalyzes the Transcription Process RNA polymerases all synthesize RNA in the 5    3  direction using a DNA template; in this respect, they are similar to template­dependent DNA polymerases  discussed in Chapter 15. Unlike DNA polymerases, however, RNA polymerases initiate polymerization at a promoter sequence without the need of a DNA

 

  

Page 691

or RNA primer. Cellular RNA polymerases, both prokaryotic and eukaryotic, are large multisubunit enzymes whose mechanisms are only partially understood. The most intensely studied prokaryotic RNA polymerase is that from Escherichia coli, which consists of five subunits having an aggregate molecular weight of over  500,000 (Table 16.2). Two a  subunits, one b  subunit, and one   subunit constitute the core enzyme, which is capable of faithful transcription but not of specific (i.e.,  promoter­initiated) RNA synthesis. The addition of a fifth protein subunit, designated s, results in the holoenzyme that is capable of specific RNA synthesis in vitro  and in vivo. The logical conclusion, that   is involved in the specific recognition of promoters, has been borne out by a variety of biochemical studies and is discussed  below. Specific s factors can recognize different classes of genes. For example, a specific   factor recognizes promoters for genes that are induced as a result of heat  shock. In sporulating bacteria, specific   factors recognize genes induced during sporulation. Some bacteriophage synthesize   factors that allow the appropriation of  the cell's RNA polymerase for transcription of the viral DNA. The common prokaryotic RNA polymerases are inhibited by the antibiotic rifampicin (used in treating tuberculosis), which binds to the b  subunit (see Clin. Corr.  16.2). Eukaryotic nuclear RNA polymerases are inhibited differentially by the compound a ­amanitin, which is synthesized by the poisonous mushroom Amanita  phalloides. Three nuclear RNA polymerase classes can be distinguished by these experiments. Very low concentrations of a ­amanitin inhibit the synthesis of mRNA  and some small nuclear RNAs (snRNAs); higher concentrations inhibit the synthesis of tRNA and other snRNAs, whereas rRNA synthesis is not inhibited at these  concentrations of drug. Messenger RNA synthesis is the function of RNA polymerase II. Synthesis of transfer RNA, 5sRNA, and some snRNAs are carried out by  RNA polymerase III. Ribosomal RNA genes are transcribed by RNA polymerase I, which is concentrated in the nucleolus. (The numbers refer to the order of  elution of the enzymes from a chromatography column.) Each enzyme is highly complex structurally (Table 16.2). In addition, a mitochondrial RNA polymerase is responsible for the synthesis of this organelle's mRNA, tRNA, and rRNA species. This enzyme, like bacterial RNA  polymerase, is inhibited by rifampicin. TABLE 16.2 Comparative Properties of Some RNA Polymerases Nuclear

  I (A)

 

 

II (B)

III (C)

 

Mitochondrial

E. coli

High MW subunitsa

195–197

240–214

155

65

160 ( )

 

117–126

140

138

 

150 ( )

Low MW subunits

  61–51

  41–34

  89

 

86 ( )

 

  49–44

  29–25

  70

 

40 ( )

 

  29–25

  27–20

  53

 

10 ( )

 

  19–16.5

  19.5

  49

 

 

 

 

  19

  41

 

 

 

 

  16.5

  32

 

 

 

 

 

  29

 

 

 

 

 

 

 

  19

Variable forms

2–3 types

3–4 types

2–4 types

  1

Specialization

Nucleolar; rRNA

mRNA

tRNA

All mtRNA

 

 

Viral RNA

5S rRNA

 

 

Inhibition by a­ amanitin

Insensitive  (>1 mg mL–1)

Very sensitive  (10–9–10–8 M)

Sensitive  (10–5–10–4 M)

Insensitive, but  sensitive to  rifampicin

  Rifampicin  sensitive

a

 Molecular weight × 10–3.

   

  

1 None

Page 692

CLINICAL CORRELATION 16.2  Antibiotics and Toxins That Target RNA Polymerase RNA polymerase is obviously an essential enzyme for life since transcription is the first  step of gene expression. No RNA polymerase means no enzymes. Two natural products  point out this principle; in both cases inhibition of RNA polymerase leads to death of the  organism. The ''death cap" or "destroying angel" mushroom, Amanita phalloides, is highly  poisonous and still causes several deaths each year despite widespread warnings to  amateur mushroom hunters (it is reputed to taste delicious, incidentally). The most lethal  toxin, a ­amanitin, inhibits the largest subunit of eukaryotic RNA polymerase II, thereby  inhibiting mRNA synthesis. The course of the poisoning is twofold: initial, relatively mild,  gastrointestinal symptoms are followed about 48 h later by massive liver failure as  essential mRNAs and their proteins are degraded but not replaced by newly synthesized  molecules. The only therapy is supportive, including liver transplantation; but this latter  course is clearly a desperate measure of unproven efficacy. More benign (at least from the point of view of our own species) is the action of the  antibiotic rifampicin to inhibit the RNA polymerases of a variety of bacteria, most notably  in the treatment of tuberculosis. Mycobacterium tuberculosis, the causative agent, is  insensitive to many commonly used antibiotics, but it is sensitive to rifampicin, the product  of a soil streptomycetes. Since mammalian RNA polymerase is so different from the  prokaryotic variety, inhibition of the latter enzyme is possible without great toxicity to the  host. This consideration implies a good therapeutic index for the drug, that is, the ability to  treat a disease without causing undue harm to the patient. Together with improved public  health measures, antibiotic therapy with rifampicin and isoniazid (an anti­metabolite) has  greatly reduced the morbidity due to tuberculosis in industrialized countries.  Unfortunately, the disease is still endemic in impoverished populations in the United States  and in other countries. Furthermore, in increasing numbers, immunocompromised  individuals, especially AIDS patients, have active tuberculosis. Mitchel, D. H. Amanita mushroom poisoning. Annu. Rev. Med. 31:51, 1980; Gilman,  A. G., Rall, T. W., Nies, A. S., and Taylor, P. (Eds.). The Pharmacological Basis of  Therapeutics, 8th ed. New York: Pergamon Press, 1990, pp. 129–130; DeCock, K.  M., Soro, B., Colibaly, I. M., Lucas, S. B. Tuberculosis and HIV infection in sub­ Saharan Africa. JAMA 268:1581, 1992.   The Steps of Transcription in Prokaryotes Have Been Determined Transcription is a strand­selective process; most double helical DNA is transcribed in only one direction. This is illustrated as follows:

The DNA strand that serves as the template for RNA synthesis is sometimes called the sense strand because it is complementary to the RNA transcript.  Conventionally, the sense strand is usually the "bottom" strand of a double­stranded DNA as written. The other strand, the "top" strand, has the same direction as the  transcript when read in the 5    3  direction; this strand is sometimes (confusingly) called the antisense strand. When only a single DNA sequence is given in this  book, the antisense strand is represented. Its sequence can be converted to the RNA transcript of a gene by simply substituting U (uracil) for T (thymine) bases.  Prokaryotic transcription begins with the binding of RNA polymerase to a gene's promoter (Figures 16.11 and 16.12). RNA polymerase holoenzyme binds to one  face of the DNA extending 45 bp or so upstream and 10 bp downstream from the RNA initiation site. Two short oligonucleotide sequences in this region are highly  conserved. One sequence that is located about 10 bp upstream from the transcription start is the consensus sequence (sometimes called a Pribnow box):

The positions marked with an asterisk are the most conserved; indeed, the last T residue is always found in E. coli promoters. A second consensus sequence is located upstream from the Pribnow or "–10" box. This "–35 sequence"

 

  

Page 693

Figure 16.12  Early events in prokaryotic transcription.  (a) Recognition: RNA polymerase (drawn smaller than scale) with "sigma" factor  binds to a DNA promoter region in a "closed" conformation.  (b) Initiation: The complex is converted to an "open" conformation and the first  nucleoside triphosphate aligns with the DNA.  (c) Bond formation: The first phosphodiester bond is formed and the "sigma''  factor released.  (d) Elongation: Synthesis of nascent RNA proceeds with movement  of the RNA polymerase along the DNA. The double helix reforms.

 

  

Page 694

is centered about 35 bp upstream from the transcription start; the nucleotides with asterisks are most conserved. The spacing between the –35 and –10 sequences is  crucial with 17 bp being highly conserved. As shown in Figure 16.13, the TTGACA and TATAAT sequences are asymmetrical, that is, they do not have the same  sequence if the complementary sequence is read. Thus the promoter sequence itself determines that transcription will proceed in only one direction. What difference do  the consensus sequences make to a gene? Measurements of RNA polymerase binding affinity and initiation efficiency to

Figure 16.13  Biosynthesis of RNA showing asymmetry in transcription.  Nucleoside 5 ­triphosphates align with complementary bases on one DNA strand,  the template. RNA polymerase catalyzes the formation of the  3 ,5 ­phosphodiester links by attaching the 5 ­phosphate of the incoming  nucleotide to the 3 ­OH group of the growing nascent RNA releasing P .  i

The new RNA is synthesized from its 5  end toward the 3  end.

 

  

Page 695

various promoter sequences have shown that the most active promoters fit the consensus sequences most closely. Statistical measurements of promoter homology  conform closely to the measured "strength" of a promoter, that is, its kinetic ability to initiate transcription with –35 purified RNA polymerase. Bases flanking the –35 and –10 sequences, bases near the transcription start, and bases located near the –16 position are weakly conserved. In some of these weakly  conserved regions, RNA polymerase may require that a particular nucleotide not be present or that local variations in DNA helical structure be present. Promoters for E. coli heat shock genes have different consensus sequences at the –35 and –10 homologies. This is consistent with their being recognized by a different   factor. An RNA transcript usually starts with a purine riboside triphosphate; that is, pppG . . . or pppA . . ., but pyrimidine starts are also known (Figures 16.11 and 16.12).  The position of transcription initiation differs slightly among various promoters but usually is from five to eight base pairs downstream from the invariant T of the  Pribnow box. Initiation Two kinetically distinct steps are required for RNA polymerase to initiate the synthesis of an RNA transcript. In the first step, RNA polymerase holoenzyme binds to  the promoter DNA to form a "closed complex." In the second step, the holoenzyme forms a more tightly bound "open complex," which is characterized by a local  opening of about 10 bp of the DNA double helix. Since the consensus Pribnow box is A­T rich, it can facilitate this local unwinding. As discussed in Chapter 14,  opening 10 bp of DNA is topologically equivalent to the relaxation of a single negative supercoil. As might be predicted from this observation, the activity of some  promoters depends on the superhelical state of the DNA template; some promoters are more active on highly supercoiled DNA while others are more active when the  superhelical density of the template is lower. The unwound DNA binds the initiating triphosphate and RNA polymerase then forms the first phosphodiester bond. The  enzyme translocates to the next position (this is the rifampicin­inhibited step) and continues synthesis. At or a short time after the initial bond formation,   factor is  released and the enzyme is considered to be in an elongation mode. Other RNA polymerase molecules can now bind to the promoter so that a gene can be  transcribed many times (Figure 16.14).

Figure 16.14  Simultaneous transcription of a gene by many  RNA polymerases, depicting the increasing  length of nascent RNA molecules.  Courtesy of Dr. O. L. Miller, University of  Virginia. Reproduced with permission from  Miller, O. L., and Beatty, B. R. J.  Cell Physiol. 74:225, 1969.

Elongation RNA polymerase continues the binding–bond formation–translocation cycle at a rate of about 40 nucleotides per second. This rate is only an average, however, and  there are many examples known for which RNA polymerase pauses or slows down at particular sequences, usually inverted repeats (palindrome sequence of  nucleotides). As will be discussed below, these pauses can bring about transcription termination. As RNA polymerase continues down the double helix, it continues to separate the two strands of the DNA template. As seen in Figure 16.12, this process allows the  template (sense) strand of the DNA to base pair with the growing RNA chain. Thus a single mechanism of information transfer (Watson–Crick base pairing) serves  several processes: DNA replication, DNA repair, and transcription of genetic information into RNA. (As will be seen in Chapter 17, base pairing is essential for  translation as well.) The process of unwinding and restoring the DNA double helix is aided by DNA topoisomerases I and II, which are components of the  transcription complex. Changes in the transcription complex during the elongation phase can affect subsequent termination events. These changes depend on the binding of another cellular  protein (nusA protein) to core RNA polymerase. Failure to

 

  

Page 696

Figure 16.15  The stem–loop structure of the  RNA transcript that determines  rho­independent transcriptional  termination.  Note the two components of the  structure: the G + C­rich stem and  loop,followed by a sequence of U  residues.

bind sometimes results in an increased frequency of termination and, consequently, a reduced level of gene expression. Termination The RNA polymerase complex also recognizes the ends of genes (Figure 16.15). Transcription termination can occur in either of two modes, depending on whether or  not it is dependent on the protein factor rho. Terminators are thus classified as rho independent or rho dependent. Rho­independent terminators are better characterized (Figure 16.15). A consensus­type sequence is involved here: a G­C rich palindrome (inverted repeat)  precedes a sequence of 6–7 U residues in the RNA chain. As a result the RNA chain forms a stem and loop structure preceding the U residues. The secondary  structure of the stem and loop is crucial for termination; base change mutations in the stem and loop that disrupt pairing also reduce termination. Furthermore, the most  efficient terminators are the most G­C rich and therefore most stable. The terminator stem and loop stabilize prokaryotic mRNA against nucleolytic degradation. Rho­dependent terminators are less well defined. Rho factor is a hexameric protein possessing an essential RNA­dependent ATPase activity. The sequences of  rho­dependent termination sites feature regularly spaced C residues within a relatively unstructured length of the transcript. The nascent RNA is thought to wrap  around rho factor while ATP hydrolysis leads to dissociation of the transcript from the template. Prokaryotic ribosomes usually attach to the nascent mRNA while it is being transcribed. This coupling between transcription and translation is important in gene control  by attenuation, which is discussed in Chapter 19. Transcription in Eukaryotes Involves Many Additional Molecular Events Eukaryotic transcription is considerably more complex than the process in prokaryotes. While the information specifying a promoter is still carried in a DNA sequence,  several molecular events besides RNA polymerase binding are required for transcription initiation. First, chromatin containing the promoter sequence must be spatially  accessible to the transcription machinery. Second, protein transcription factors distinct from RNA polymerase must bind to sequences in the promoter region for a  gene to be active. Third, other sequences located some distance away from the promoter affect transcription; these sequences are termed enhancers and they, too,  bind protein factors to stimulate transcription. Finally, recall that the eukaryotic RNA polymerase consists of three distinct enzyme forms, each specific form capable of  transcribing only a

Figure 16.16  DNase­hypersensitive (DH) sites upstream of the promoter for  the chick lysozyme gene, a typical eukaryotic transcriptional unit.  Hypersensitive sites, that is, sequences around the lysozyme gene  where nucleosomes are not bound to the DNA, are indicated by  arrows. Note that some hypersensitive sites are found in the  lysozyme promoter whether the oviduct is synthesizing or not  synthesizing lysozyme; the synthesis of lysozyme is accompanied  by the opening up of a new hypersensitive site in mature  oviduct. In contrast, no hypersensitive sites are present in  nucleated erythrocytes that never synthesize lysozyme.  Adapted from Elgin, S. C. R. J. Biol. Chem.  263:1925, 1988.

 

  

Page 697

CLINICAL CORRELATION 16.3  Fragile X Syndrome: A Chromatin Disease? Fragile X syndrome is the single most common form of inherited mental retardation,  affecting 1/1250 males and 1/2000 females. A variety of anatomical and neurological  symptoms result from the inactivation of the FMR1 gene, located on the X chromosome.  The genetics of the syndrome are complex due to the molecular mechanism of the Fragile  X mutation. The Fragile X condition results from the expansion of a trinucleotide repeat sequence,  CGG, found at the 5 ­untranslated region of the FRM1 gene. Normally, this repeat is  present in 30 copies, although normal individuals can have up to 200 copies of the repeat.  In individuals with Fragile X syndrome, the FMR1 gene contains many more copies, from  200 to thousands, of the CGG repeat. The complex genetics of the disease result from  the potential of the CGG repeat sequence to expand from generation to generation. The presence of an abnormally high number of CGG repeats induces extensive DNA  methylation of the entire promoter region of FMR1. Methylated DNA is transcriptionally  inactive, so FMR1 mRNA is not synthesized. The absence of FMR1 protein leads to the  pathology of the disease. FMR1 protein normally is located in the cytoplasm in all tissues of the early fetus and,  later, especially in the fetal brain. Its sequence has some characteristics of an RNA­ binding protein. One hypothesis is that the protein aids in the translation of brain­specific  mRNAs during development. Warren, S. L., and Nelson, D. L. Advances in molecular analysis of Fragile X syndrome.  JAMA 271:536, 1994; and Caskey, C. T. Triple repeat mutations in human disease.  Science 256:784, 1992.   single class of cellular RNA. By contrast, transcription in prokaryotes requires, in the simplest case, only an appropriate sequence of DNA, RNA polymerase  holoenzyme, and nucleoside triphosphate substrates. The Nature of Active Chromatin The structural organization of eukaryotic chromosomes was discussed in Chapter 14. Although chromatin is organized into nucleosomes whether or not it is capable  of being transcribed, an active gene has a generally "looser" configuration than does transcriptionally inactive chromatin. This difference is most striking in the promoter  sequences, parts of which are not organized into nucleosomes at all (Figure 16.16). The lack of nucleosomes is manifested experimentally by the enhanced sensitivity  of promoter sequences to external reagents that cleave DNA, such as the enzyme DNase I. This enhanced accessibility of promoter sequences (termed DNase I  hypersensitivity) ensures that transcriptional factors will be able to bind to appropriate regulatory sequences. In addition, although the transcribed parts of a gene  may be organized into nucleosomes, the nucleosomes are less tightly bound than those in an inactive gene. Finally, DNA may be transcriptionally inactivated by  methylation (see Clin. Corr. 16.3). The overall theme is one of partially unfolded chromatin being necessary but not sufficient for transcription. Enhancers Enhancer sequences increase (enhance) the expression of a gene about 100­fold, hence the name. They function only when located on the same DNA molecule  (chromosome) as the promoter whose activity they affect. They can function when located in either the 5  or 3  direction and as much as 1000 bp away from the  relevant promoter. Protein factors bind to enhancer DNA and are necessary for enhancer function. Transcription of Ribosomal RNA Genes Recall that rRNA genes are located in a specialized nuclear structure, the nucleolus. There are several hundred copies of each rRNA gene in a eukaryotic cell,  tandemly repeated in the DNA of a specific region of one chromosome, the nucleolar organizer. The repeat units contain a copy of each RNA sequence (28S, 5.8S,  and 18S) and are separated from each other by nontranscribed spacer regions. Figure 16.17 is a diagram of this arrangement. Each repeat unit is transcribed as a  unit, yielding a primary transcript containing one copy each of the 28S, 5.8S, and 18S sequences, ensuring synthesis of equimolar amounts of these three RNAs. The  primary transcript is then processed by ribonucleases and modifying enzymes to the three mature rRNA species (see Section 16.5). Termination of transcription occurs  within the nontranscribed spacer region before RNA polymerase I reaches the promoter of the next repeat unit. The promoter recognized by RNA polymerase I is located within the non­transcribed spacer, from about positions –40 to +10 and from –150 to –110. A  transcription factor binds to the promoter and thereby directs RNA polymerase

Figure 16.17  Structure of a rRNA transcription unit.  Ribosomal RNA genes are arranged with many copies one after another. Each copy is  transcribed separately and each transcript is processed into three separate RNA  species. Promoter and enhancer sequences are located in the nontranscribed  regions of the tandemly repeated sequences.

 

  

Page 698

recognition of the promoter sequence. In addition, an enhancer element is located about 250 bp upstream from the promoter in human ribosomal DNA. The size of the  nontranscribed spacer varies considerably from one organism to the next, as does the position of the enhancer element. Transcription of rRNA can be very rapid; this reflects the fact that synthesis of ribosomes is rate­limiting for cell growth. Phosphorylation of RNA polymerase I may  activate especially rapid transcription of rRNA, for example, during embryonic growth or liver regeneration. Transcription by RNA Polymerase II RNA polymerase II is responsible for the synthesis of mRNA in the nucleus. Three common themes have emerged from research on a large number of genes (Figure  16.18). (1) The DNA sequences controlling transcription are complex; a single gene may be controlled by as many as six or eight DNA sequence elements in addition  to the promoter (RNA polymerase binding region) itself. The controlling sequence elements function in combination to give a finely tuned pattern of control. (2) The  effect of the controlling sequences on transcription is mediated by the binding of protein molecules to each sequence element. These transcription factors recognize the  nucleotide sequence of the appropriate controlling sequence element. (3) Bound transcription factors bind with each other and with RNA polymerase to activate  transcription. The DNA binding and activation activities of the factors reside in separate domains of the proteins.

Figure 16.18  Interaction of transcription factors with promoters.  A large number of transcriptional factors interact with eukaryotic promoter regions.  (a) A hypothetical array of factors that interact with specific DNA sequences near the promoter.  This includes a factor, TFIID, which binds to the TATA box and the Jun and Fos proteins,  which are proto­oncogenes (Clin. Corr. 16.4). The figure is not meant to imply that all of the DNA  binding factors bind to the promoter simultaneously.  (b) One way in which the DNA binding factors are hypothesized to bind to each other and  to RNA polymerase. Although this model is not completely proved, it is known that  proteins that bind to distant DNA sequences make protein–protein contacts with each other.  Reprinted with permission from Mitchell, P. J., and Tjian, R. Science 245:371, 1989.

 

  

Page 699

Promoters for mRNA Synthesis In contrast to prokaryotic RNA polymerase which recognizes only a single promoter sequence, RNA polymerase II can initiate transcription by recognizing several  classes of consensus sequences upstream from the mRNA start site. The first and most prominent of these, sometimes called the TATA box, has the sequence

The TATA box is centered about 25 bp upstream from the transcription unit. Experiments in which it was deleted suggest that it is required for efficient transcription,  although some promoters may lack it entirely. A second region of homology is located further upstream, in which the CAAT box sequence

is found. This sequence is not as highly conserved as the TATA box, and some active promoters may not possess it. Other sequences, described in Figure 16.18, may  also promote transcription. The CAAT and TATA boxes, as well as the other sequences shown in Figure 16.15, do not contact RNA polymerase II directly. Rather,  they require the binding of specific transcription factors to function. The current model for the activation of genes in this manner is shown in Figure 16.18. Note how  protein factors bind not only to their recognition sequences but also to each other and to RNA polymerase, itself a very large and complex enzyme. Despite the  complexities of the detailed interactions, the three principles elaborated above account for the known mechanisms of all class II transcription factors. Mutated forms of  several of these transcription factors function as nuclear oncogenes (see Clin. Corr. 16.4). Transcription by RNA Polymerase III The themes elaborated above for the transcription of class I and class II promoters hold for the transcription of 5S RNA and tRNA by RNA polymerase III.  Transcription factors bind to DNA and direct the action of RNA polymerase. One unusual feature of RNA polymerase III action in the transcription of 5S RNA is the  location of the factor­binding sequence; it can be located within the DNA sequence encoding the RNA. The DNA in the region that would normally be thought of as a  promoter, that is, the sequence immediately 5  to the transcribed region of the gene, has no specific sequence and can be substituted by other sequences without a  substantial effect on transcription. Figure 16.19 diagrams this unusual sequence arrangement. In other cases, for example, tRNA transcription, the factor­binding  sequence is located more conventionally at the 5  region of the gene, that is, preceding the transcribed sequences. 16.5—  Posttranscriptional Processing The immediate product of transcription is a precursor RNA molecule, called the primary transcript, which is modified to a mature, functional molecule. The  reactions of RNA processing can include removal of extra nucleotides, base modification, addition of nucleotides, and separation of different RNA sequences by the  action of specific nucleases. Finally, in eukaryotes, RNAs must be exported from the nucleus.

 

  

Page 700

Figure 16.19  Transcription factor for a class III eukaryotic gene.  The transcription factor TFIIIA binds to a sequence located within the Xenopus gene for  5S rRNA. The RNA polymerase III then binds to the factor and initiates transcription of the  5S sequence. No specific sequence in the DNA is required other than the factor binding  sequence.

Transfer RNA Precursors Are Modified by Cleavage, Additions, and Base Modification Cleavage The primary transcript of a tRNA gene contains extra nucleotide sequences both 5  and 3  to the tRNA sequence. In some cases these primary transcripts contain  introns in the anticodon region of the tRNA also. Processing reactions occur in a closely defined but not necessarily rigid temporal order. First, the primary transcript is  trimmed in a relatively nonspecific manner to yield a precursor molecule with shorter 5  and 3  extensions. Then ribonuclease P, a ribozyme (see above), removes the  5  extension by endonucleolytic cleavage. The 3  end is trimmed exonucleolytically, followed by synthesis of the CCA terminus. Synthesis of the modified nucleotides  occurs in any order relative to the nucleolytic trimming. Intron removal is dictated by the secondary structure of the precursor (see Figure 16.20, p. 702) and is  carried out by a soluble, two­component enzyme system; one enzyme removes the intron and the other reseals the nucleotide chain. Additions Each functional tRNA has the sequence CCA at its 3  terminus. In most instances this sequence is added sequentially by the enzyme tRNA nucleotidyltransfer­

 

  

Page 701

CLINICAL CORRELATION 16.4  Involvement of Transcriptional Factors in Carcinogenesis The conversion of a normally well­regulated cell into a cancerous one requires a number  of independent steps whose end result is a transformed cell capable of uncontrolled  growth and metastasis. Insights into this process have come from recombinant DNA  studies of the genes whose mutated or overexpressed products contribute to  carcinogenesis. These genes are termed oncogenes. Oncogenes were first identified as  products of DNA or RNA tumor viruses but normal cells have copies of these genes as  well. The normal, nonmutated cellular analogs of oncogenes are termed proto­oncogenes.  The products of proto­oncogenes are components of the many pathways that regulate  growth and differentiation of a normal cell; mutation into an oncogenic form involves a  change that makes the regulatory product less responsive to normal control. Some proto­oncogenic products are involved in the transduction of hormonal signals or  the recognition of cellular growth factors and act cytoplasmically. Other proto­oncogenes  have a nuclear site of action; their gene products are often associated with the  transcriptional apparatus and they are synthesized in response to growth stimuli. It is easy  to visualize how the overproduction or permanent activation of such a positive  transcription factor could aid the transformation of a cell to malignancy: genes normally  transcribed at a low or controlled level would be overexpressed by such a deranged  control mechanism. A more subtle genetic effect predisposing to cancer is exemplified by the human tumor  suppressor protein p53. This protein is the product of a dominant oncogene. A single  copy of the mutant gene causes Li–Fraumeni syndrome, an inherited condition  predisposing to carcinomas of the breast and adrenal cortex, sarcomas, leukemia, and  brain tumors. Somatic mutations in p53 can be identified in about half of all human cancers. Mutations  represent a loss of function, affecting either the stability or DNA­binding ability of p53.  Thus wild­type p53 functions as a tumor suppressor. The wild­type protein helps to  control the checkpoint between the G1 and S phases of the cell cycle, activates DNA  repair, and, in other circumstances, leads to programmed cell death (apoptosis). Thus the  biochemical actions of p53 serve to keep cell growth regulated, maintain the information  content of the genome, and, finally, eliminate damaged cells. All of these functions would  counteract neoplastic transformation of a cell. These varied roles are a function of p53's action as a transcription factor, inhibiting some  genes and activating others. For example, p53 inhibits transcription of genes with TATA  sequences, perhaps by binding to the complex formed between transcription factors and  the TATA sequence. Alternatively, p53 is a site­specific DNA­binding protein and  promotes transcription of some other genes, for example, those for DNA repair. The three­dimensional structure of p53 has been determined. Mutations found in p53  from tumors affect the DNA­binding domain of the protein. For example, nearly 20% of  all mutated residues involve mutations at two positions in p53. The crystal structure of the  protein–DNA complex shows that these two amino acids, both arginines, form hydrogen  bonds with DNA. Arginine 248 forms hydrogen bonds in the minor groove of the DNA  helix with a thymine oxygen and with a ring nitrogen of adenine. Mutation disrupts this H­ bonded network and therefore the ability of p53 to regulate transcription. Weinberg, R. A. Oncogenes, antioncogenes, and the molecular basis of multistep  carcinogenesis. Cancer Res. 49:3713, 1989; Cho, Y., Gorina, S., Jeffrey, P. D., and  Pavletich, N. P. Crystal structure of a p53 tumor suppressor–DNA complex:  understanding tumorigenic mutations. Science 265:346, 1994; Friend, S. p53: A glimpse  at the puppet behind the shadow play. Science 265:334, 1994; and Harris, C. C., and  Hollstein, M. Clinical implications of the p53 tumor­suppressor gene. N. Engl. J. Med.  329:1318, 1993.   ase. Nucleotidyltransferase uses ATP and CTP as substrates and always incorporates them into tRNA at a ratio of 2C/1A. The CCA ends are found on both  cytoplasmic and mitochondrial tRNAs. Modified Nucleosides Transfer RNA nucleotides are the most highly modified of all nucleic acids. More than 60 different modifications to the bases and ribose, requiring well over 100  different enzymatic reactions, have been found in tRNA. Many are simple, one­step methylations, but others involve multistep synthesis. Two derivatives,  pseudouridine and queuosine (7–4, 5­cis­dihydroxy­1­cyclopenten­3­ylamino methyl­7­deazaguanosine), actually require severing of the b ­glycosidic bond of the  altered nucleotide. One enzyme or set of enzymes produces a single site­specific modification in more than one species of tRNA molecule. Separate enzymes or sets  of enzymes produce the same modifications at more than one location in tRNA. In other words, most modification enzymes are site or nucleotide sequence specific,  not tRNA specific. Most modifications are completed before the tRNA precursors have been cleaved to mature tRNA size.

 

  

Page 702

Figure 16.20  Scheme for processing a eukaryotic tRNA.  The primary transcript is cleaved by RNase P and a 3 ­exonuclease, and the  terminal CCA is synthesized by tRNA nucleotidyltransferase before the intron  is removed, if necessary.

Ribosomal RNA Processing Releases the Various RNAs from a Longer Precursor The primary product of rRNA transcription is a long RNA, termed 45S RNA, which contains the sequences of 28S, 5.8S, and 18S rRNAs. Processing of 45S RNA  occurs in the nucleolus. Like the processing of mRNA precursors (see below), processing of the rRNA precursors is carried out by large multisubunit  ribonucleoprotein assemblies. At least three RNA species are required for processing. These all function as small nucleolar ribonucleoprotein complexes  (snoRNPs). Processing of the rRNAs follows a sequential order (Figure 16.21).

 

  

Page 703

Figure 16.21  Schemes for transcription and processing of rRNAs.  Redrawn from Perry, R. Annu. Rev. Biochem. 45:611, 1976.  Copyright © 1976 by Annual Reviews, Inc.

Processing of pre­rRNA in prokaryotes also involves cleavage of high molecular weight precursors to smaller molecules (see Figure 16.21). Some of the bases are  modified by methylation on the ring nitrogens of the bases rather than the ribose and by the formation of pseudouridine. The E. coli genome has seven rRNA  transcriptional units dispersed throughout the DNA. Each contains one 16S, one 23S, and one 5S rRNA or tRNA sequence. Processing of the rRNA is coupled  directly to transcription, so that cleavage of a large precursor primary transcript rapidly yields pre­16S, pre­23S, pre­5S, and pre­tRNAs. These precursors are  slightly larger than the functional molecules and only require trimming for maturation. Messenger RNA Processing Requires Maintenance of the Coding Sequence Most eukaryotic mRNAs have distinctive structural features added in the nucleus by enzyme systems other than RNA polymerase. These include the 3 ­terminal poly (A) tail, methylated internal nucleotides, and the cap 5  terminus. Cytoplasmic mRNAs are shorter than their primary transcripts, which can contain additional terminal  and internal sequences. Noncoding sequences present within pre­mRNA molecules, but not present in mature mRNAs, are called intervening sequences or introns. The expressed or retained sequences are called exons. The general pattern for mRNA processing is depicted in Figure 16.22. Incompletely processed mRNAs  make up a large part of the heterogeneous nuclear RNA (hnRNA). Processing of eukaryotic pre­mRNA involves a number of molecular reactions, all of which must be carried out with exact fidelity. This principle is most clear in the  removal of introns from an mRNA transcript. An extra nucleotide in the coding sequence of mature mRNA would cause the reading frame of

 

  

Page 704

Figure 16.22  Scheme for processing mRNA.  The points for initiation and termination of transcription are indicated on the DNA.  Arrows indicate cleavage points. The many proteins associated with  the RNA and tertiary conformations are not shown.

that message to be shifted and the resulting protein will almost certainly be nonfunctional. Indeed, mutations in the b ­globin gene that interfere with intron removal are a  major cause of the genetic disease b ­thalassemia (see Clin. Corr. 16.5). The task for the cell becomes even more daunting when seen in the light of the structure of  some important human genes that consist of over 90% intron sequences. The complex reactions to remove introns are accomplished by multicomponent enzyme  systems that act in the nucleus; after these reactions are completed the mRNA is exported to the cytoplasm where it interacts with ribosomes to initiate translation. Blocking of the 5¢ Terminus and Poly(A) Synthesis Addition of the cap structures occurs during transcription by RNA polymerase II (Figure 16.22). As the transcription complex moves along the DNA, the capping  enzyme complex modifies the 5  end of the nascent mRNA. This is the only eukaryotic pre­mRNA processing event that is known to occur cotranscriptionally, that is,  while RNA polymerase is still transcribing the downstream portions of the gene. After initiation and cap synthesis, RNA polymerase continues transcribing the gene until a polyadenylation signal sequence is reached (Figure 16.23). This  sequence, which has the consensus AAUAAA, appears in the mature mRNA but usually does not form part of its coding region. Rather, it signals cleavage of the  nascent mRNA precursor about 20 or so nucleotides downstream. The poly(A) sequence is then added by a soluble polymerase to the free 3  end produced by this  cleavage. Note that polyadenylation does not require a template. Somewhat paradoxically, RNA polymerase II continues transcription for as many as 1000  nucleotides beyond the point at which the transcript is released from chromatin. Nucleotides incorporated into RNA by this process are apparently turned over and  never appear in any cytoplasmic RNA species.

 

  

Page 705

CLINICAL CORRELATION 16.5  Thalassemia Due to Defects in Messenger RNA Synthesis The thalassemias are genetic defects in the coordinated synthesis of a ­ and b ­globin  peptide chains; a deficiency of b  chains is termed b ­thalassemia while a deficiency of a   chains is termed a ­thalassemia. Patients suffering from either of these conditions present  with anemia at about 6 months of age as HbF synthesis ceases and HbA synthesis would  become predominant. The severity of symptoms leads to the classification of the disease  into either thalassemia major, where a severe deficiency of globin synthesis occurs, or  thalassemia minor, representing a less severe imbalance. Occasionally, an intermediate  form is seen. Therapy for thalassemia major involves frequent transfusions, leading to a  risk of complications from iron overload. Unless chelation therapy is successful, the  deposition of iron in peripheral tissues, termed hemosiderosis, can lead to death before  adulthood. Carriers of the disease usually have thalassemia minor, involving mild anemia.  Ethnographically, the disease is common in persons of Mediterranean, Arabian, and East  Asian descent. As is the case for sickle cell anemia (HbS) and glucose 6­phosphate  dehydrogenase deficiency, the abnormality of the carriers' erythrocytes affords some  protection from malaria. Maps of the regions where one or another of these diseases is  frequent in the native population superimpose over the areas of the world where malaria is  endemic.

a ­Thalassemia is usually due to a genetic deletion, which can occur because the a ­globin  genes are duplicated; unequal crossing over between adjacent a  alleles apparently has  led to the loss of one or more loci. In contrast, b ­thalassemia can result from a wide  variety of mutations. Known events include mutations leading to frameshifts in the b ­ globin coding sequence, as well as mutations leading to premature termination of peptide  synthesis. Many b ­thalassemias result from mutations affecting the biosynthesis of b ­ globin mRNA. Genetic defects are known that affect the promoter of the gene, leading to  inefficient transcription. Other mutations result in aberrant processing of the nascent  transcript, either during splicing out of the two introns from the transcript or during  polyadenylation of the mRNA precursor. Examples where the molecular defect illustrates  a general principle of mRNA synthesis are discussed in the text. Orkin, S. H. Disorders of hemoglobin synthesis: the thalassemias. In: G.  Stamatoyannopoulis, A. W. Nienhuis, P. Leder, and P. W. Majerus (Eds.). The  Molecular Basis of Blood Diseases Philadelphia: Saunders, 1987; and Weatherall, D. J.,  Clegg, J. B., Higgs, D. R., and Wood, W. G. The hemoglobinopathies. In: C. R. Scriver,  A. L. Beaudet, W. S. Sly, and D. Valle (Eds.). The Metabolic and Molecular Bases of  Inherited Disease, 7th ed. New York: McGraw­Hill, 1995.   Removal of Introns from mRNA Precursors As pre­RNA is extruded from the RNA polymerase complex, it is rapidly bound by small nuclear ribonucleoproteins, snRNPs (snurps), which carry out the dual  steps of RNA splicing: (1) breakage of the intron at the 5  donor site and (2) joining the upstream and downstream exon sequences together. All introns begin with a  GU sequence and end with AG; these are termed the donor and acceptor intron–exon junctions, respectively. Not all GU or AG sequences are spliced out of RNA,  however. How does the cell know which GU sequences are in introns (and therefore must be removed) and which are destined to remain in mature mRNA? This  discrimination is accomplished by the formation of base pairs between U1 RNA and the sequence of the mRNA precursor surrounding the donor GU sequence (see  Clin. Corr. 16.6). See Figure 16.24 for an illustration of this process. Another snRNP, containing U2 RNA, recognizes

Figure 16.23  Cleavage and polyadenylation of eukaryotic mRNA precursors.  The 3  termini of eukaryotic mRNA species are derived by processing.  The sequence AAUAAA in the mRNA specifies the cleavage of the  mRNA precursor. The free 3 ­OH end of the mRNA is a primer for  poly(A) synthesis.  Adapted from Proudfoot, N. J. Trends Biochem. Sci. 14:105, 1989

 

  

Page 706

CLINICAL CORRELATION 16.6  Autoimmunity in Connective Tissue Disease Humoral antibodies in sera of patients with various connective tissue diseases recognize a  variety of ribonucleoprotein complexes. Patients with systemic lupus erythematosus  exhibit a serum antibody activity designated Sm, and those with mixed connective tissue  disease exhibit an antibody designated RNP. Each antibody recognizes a distinct site on  the same RNA–protein complex, U1 RNP, that is involved in mRNA processing in  mammalian cells. The U1–RNP complex contains U1 RNA, a 165­nucleotide sequence  highly conserved among eukaryotes, that at its 5  terminus includes a sequence  complementary to intron–exon splice junctions. Addition of this antibody to in vitro  splicing assays inhibits splicing, presumably by removal of the U1 RNP from the reaction.  Sera from patients with other connective tissue diseases recognize different nuclear  antigens, nucleolar proteins, and/or chromosomal centromeres. Sera of patients with  myositis have been shown to recognize cytoplasmic antigens such as aminoacyl­tRNA  synthetases. Although humoral antibodies have been reported to enter cells via Fc  receptors, there is no evidence that this is part of the mechanism of autoimmune disease.   important sequences at the 3  acceptor end of the intron. Still other snRNP species, among them U5 and U6, then bind to the RNA precursor, forming a large  complex termed a spliceosome (by analogy with the large ribonucleoprotein assembly involved in protein synthesis, the ribosome). The spliceosome uses ATP energy  to carry out the accurate removal of the intron. First, the phosphodiester bond between the exon and the donor GU sequence is broken, leaving a free 3 ­OH group at  the end of the first exon and a 5  phosphate on the donor G of the intron. This pG is then used to form an unusual linkage with the 2 ­OH group of an adenosine within  the intron to form a branched or lariat RNA structure, as shown in Figure 16.25. After the lariat is formed, the second step of splicing occurs. The phosphodiester  bond immediately following the AG is cleaved and the two exon sequences are ligated together. In pre­mRNAs containing a large number of introns, splicing occurs  roughly in order from the 5  to the 3  end of the mRNA precursor. However, this is not a hard and fast rule as there is no singly preferred order for removal. The end  result of processing is a fully functional coding mRNA, all introns removed, and ready to direct protein synthesis. Mutations in Splicing Signals Cause Human Diseases Messenger RNA splicing is an intricate process dependent on many molecular events. If these events are not carried out with precision, functional mRNA is not  produced. This principle is illustrated in the human thalassemias, which affect the balanced synthesis of a  and b ­globin chains (see Clin. Corr. 16.5). Some of the  mutations leading to b ­thalassemia interfere with the splicing of b ­globin mRNA precursors. For example, we know that all intron sequences begin with the  dinucleotide GU. Mutation of the G in this sequence to an A means that the splicing machinery will no longer recognize this dinucleotide as a donor site. Splicing will  ''pass by" the correct exon–intron junction. This could lead to two results: extra sequences that would normally be spliced out will appear in the b ­globin mRNA, or,  alternatively, sequences could be deleted from the mRNA product (Figure 16.26). In either event, functional b ­globin will be made in reduced amounts and the anemia  characteristic of the disease will result. Alternate pre­mRNA Splicing Can Lead to Multiple Proteins Being Made from a Single DNA Coding Sequence The existence of intron sequences is paradoxical. Introns must be removed precisely so that the mRNA can accurately encode a protein. As we have seen above, a  single base mutation can drastically interfere with splicing and cause a serious disease. Furthermore, the presence of intron sequences in a gene means that its overall  sequence is much larger than is required to encode its

Figure 16.24  Mechanism of splice junction recognition.  The recognition of the 5  splice junction involves base pairing between the intron–exon  junction and the U1 RNA snRNP. This base pairing targets the intron for removal.  Adapted from Sharp, P. A. JAMA 260:3035, 1988.

 

  

Page 707

Figure 16.25  Proposed scheme for mRNA splicing to include the lariat structure.  A messenger RNA is depicted with two exons (in dark blue) and an intervening intron  (in light blue). A 2 ­OH group of the intron sequence reacts with the 5 ­phosphate  of the intron's 5 ­terminal nucleotide producing a 2 –5  linkage and the lariat  structure. Simultaneously, the exon 1–intron phosphodiester bond is broken, leaving  a 3 ­OH terminus on this exon free to react with the 5 ­phosphate of the exon 2,  displacing the intron and creating the spliced mRNA. The released intron lariat is  subsequently digested by cellular nucleases.

protein product. A large gene is a target for more mutagenic events than is a small one. Indeed, common human genetic diseases like Duchenne muscular dystrophy  occur in genes that encompass millions of base pairs of DNA information. Why has nature not removed introns completely over the long time scale of eukaryotic  evolution? There are no clear answers to questions of this type but some introns do have beneficial effects.

Figure 16.26  Nucleotide change at an intron–exon  junction of the human b ­globin gene, which  leads to aberrant splicing and b ­thalassemia.  This figure shows the splicing pattern of a  mutated transcript containing a change of  G­U to A­U at the first two nucleotides of  the first intron. Loss of this invariant  sequence  means that the correct splice  junction cannot be used; therefore transcript  sequences that base pair with the U1 snRNA  less well than the correct sequence junction  are used as splice donors. The diagonal lines  indicate the portions spliced together in  mutant transcripts. Note that some of the  mutant mRNA precursor molecules are  spliced so that portions of the first  intron (denoted as a white box) appear in  the processed product. In other instances  the donor junction lies within the first  exon and portions of the first exon are  deleted. In no case is wild­type globin  mRNA produced.  Adapted from Orkin, S. H. In: G.  Stamatoyannopoulis et al. (Eds.).  The Molecular Basis of Blood  Diseases. Philadelphia: Saunders, 1987.

 

  

Page 708

Figure 16.27  Alternate splicing of tropomyosin gene transcripts results in a family  of tissue­specific tropomyosin proteins.  Redrawn from Breitbart, R. E., Andreadis,  A., and Nadal­Ginard, B. Annu. Rev. Biochem. 56:467, 1986.

Tropomyosin proteins are essential components of the contractile apparatus in the three types of muscle (see, p. 948) and each contractile cell type contains a specific  tropomyosin type. This diversity arises from a single gene that is transcribed into a primary transcript. The transcript is then processed as diagramed in Figure 16.27.  All cells containing tropomyosin make the same primary transcript but each cell type processes this transcript in a characteristic fashion. The resulting mRNA species  then are translated to yield the tropomyosins characteristic of each cell type. About 40 examples are well documented of tissue­specific splicing. Thus the existence of  introns supplies the organism with still another method of generating protein diversity. 16.6—  Nucleases and RNA Turnover The different roles of RNA and DNA in genetic expression are reflected in their metabolic fates. A cell's information store (DNA) must be preserved, thus the myriad  DNA repair and editing systems in the nucleus. Although individual stretches of nucleotides in DNA may turn over, the molecule as a whole is metabolically inert when  not replicating. The various RNA molecules, on the other hand, are individually dispensable and can be replaced by newly synthe­

 

  

Page 709

sized species of the same specificity. It is therefore no surprise that RNA repair systems are not known. Instead, defective RNAs are removed from the cell by  degradation into nucleotides, which then are repolymerized into new RNA species. This principle is clearest for mRNA species, which are classified as unstable. However, even the so­called stable RNAs turn over; for example, the half­life of tRNA  species in liver is on the order of 5 days. A fairly long half­life for a mammalian mRNA would be 30 h. Removal of RNAs from the cytoplasm is accomplished by  cellular ribonucleases. Messenger RNAs are initially degraded in the cytoplasm. The rates vary for different mRNA species, raising the possibility of control by  differential degradation. Two examples of the role of RNA stability in gene control illustrate how the stability of mRNA influences gene expression. Tubulin is the major component of the  microtubules found in many cell types as part of the cytoskeleton. When there is an excess of tubulin in the cell, the monomeric protein binds to and promotes the  degradation of tubulin mRNA, thereby reducing tubulin synthesis. A second example is provided by herpes simplex viruses (HSV), the agent causing cold sores and  some genital infections. An early event in the establishment of HSV infection is the ability of the virus to destabilize all the cellular mRNA molecules, thereby reducing  the competition for free ribosomes. Thus the viral proteins are more efficiently translated. Nucleases are of several types and specificities. The most useful distinction is between exonucleases, which degrade RNA from either the 5  or 3  end, and  endonucleases, which cleave phosphodiester bonds within a molecule. The products of RNase action contain either 3  or 5  terminal phosphates, and both endo­ and  exonucleases can be further characterized by the position (5  or 3 ) at which the monophosphate created by the cleavage is located. The structure of RNA also affects nuclease action. Most ribonucleases are less efficient on regions of highly ordered RNA structure. Thus tRNAs are preferentially  cleaved in unpaired regions of the sequence. On the other hand, many RNases involved in RNA processing require a defined three­dimensional structure for enzyme  activity. These enzymes are discussed more fully above in the consideration of RNA processing pathways. Bibliography Bradshaw, R. A. (Ed.). Transcription. Special issue of Trends Biochem. Sci. 16, November, 1991. Breitbart, R. E., Andreadis, A., and Nadal­Ginard, B. Alternative splicing: A ubiquitous mechanism for the generation of multiple protein isoforms from single genes.  Annu. Rev. Biochem. 56:467, 1987. Caskey, C.T. Triple repeat mutations in human disease. Science 256:784, 1992. Cundliffe, E. How antibiotic­producing microorganisms avoid suicide. Annu. Rev. Microbiol. 43:207, 1989. Das, A. Control of transcription by RNA­binding proteins. Annu. Rev. Biochem. 62:893, 1993. Friend, S. p53: A glimpse at the puppet behind the shadow play. Science 265:334, 1994. Gesteland, R. F., and Atkins, J. F. (Eds.). The RNA World. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 1993. Gold, L., Polisky, B., Uhlenbeck, O., and Yarus, M. Diversity of oligonucleotide functions. Annu. Rev. Biochem. 64:763, 1995. Koleske, A. J., and Young, R. A. The RNA polymerase II holoenzyme and its implications for gene regulation. Trends Biochem. Sci. 20:113, 1995. Lai, M. M. C. The molecular biology of hepatitis delta virus. Annu. Rev. Biochem. 64:259, 1995. Larson, D. E., Zahradka, P., and Sells, B. H. Control points in eucaryotic ribosome biogenesis. Biochem. Cell. Biol. 69:5, 1991. Orkin, S. H. Disorders of hemoglobin synthesis: The thalassemias. In: G. Stamatoyannopoulis, A. W. Nienhuis, P. Leder, and P. W. Majerus (Eds.), The Molecular  Basis of Blood Diseases. Philadelphia: Saunders, 1987, pp. 106–126. Pace, N. R., and Brown, J. W. Evolutionary perspective on the structure and function of ribonuclease P, a ribozyme. J. Bacteriol. 177:1919, 1995. Rosenberg, M., and Court, D. Regulatory sequences involved in the promotion and termination of RNA transcription. Annu. Rev. Genet. 12:319, 1979. Soll, D., and Rajbhandary, U. L. (Eds.). tRNA: Structure, Biosynthesis and Function. Washington DC: American Society for Microbiology, 1994. Weatherall, D.J., Clegg, J.B., Higgs, D.R., and Wood, W.G. The hemoglobinopathies. In: C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (Eds.), The  Metabolic and Molecular Bases of Inherited Disease, 7th ed. New York: McGraw­Hill, 1995. Weinberg, R. A. Oncogenes, antioncogenes and the molecular basis of multistep carcinogenesis. Cancer Res. 49:3713, 1989. Wise, J. A. Guides to the heart of the spliceosome. Science 262:1978, 1993.

 

  

Page 710

Questions C. N. Angstadt and J. Baggott 1. RNA: A. incorporates both modified and unmodified purine and pyrimidine bases during transcription. B. does not exhibit any double helical structure. C. structures exhibit base stacking and hydrogen­bonded base pairing. D. usually contains about 65–100 nucleotides. E. does not exhibit Watson–Crick base pairing. Refer to the following for Questions 2–4. A. HnRNA B. mRNA C. rRNA D. snRNA E. tRNA 2. Has the highest percentage of modified bases of any RNA. 3. Stable RNA representing the largest percentage by weight of cellular RNA. 4. Contains both a 7­methylguanosine triphosphate cap and a poly­adenylate segment. 5. Ribozymes: A. are any ribonucleoprotein particles. B. are enzymes whose catalytic function resides in RNA subunits. C. carry out self­processing reactions but cannot be considered true catalysts. D. bind to the mRNA precursor to recognize the 5 ­splice site for intron removal. E. function only in the processing of mRNA. 6. In eukaryotic transcription: A. RNA polymerase does not require a template. B. all RNA is synthesized in the nucleolus. C. consensus sequences are the only known promoter elements. D. phosphodiester bond formation is favored, in part, because it is followed by pyrophosphate hydrolysis. E. RNA polymerase requires a primer. 7. An enhancer: A. is a consensus sequence in DNA located where RNA polymerase first binds. B. may be located in various places in different genes. C. may be located on a separate chromosome from the gene it regulates. D. functions by binding RNA polymerase. E. stimulates transcription in both prokaryotes and eukaryotes. 8. The sigma ( ) subunit of prokaryotic RNA polymerase: A. is part of the core enzyme. B. binds the antibiotic rifampicin. C. is inhibited by a amanitin. D. must be present for transcription to occur. E. specifically recognizes promoter sites. Use this schematic representation of a prokaryotic gene to answer Questions 9–11. Numbers refer to positions of base pairs relative to the beginning of transcription.

9. Sigma ( ) factor might be released from RNA polymerase. 10. An "open complex" should form in this region. 11. Events beyond this region should be catalyzed by core enzyme. 12. Termination of a prokaryotic transcript: A. is a random process. B. requires the presence of the rho subunit of the holoenzyme. C. does not require rho factor if the end of the gene contains a G­C rich palindrome. D. is most efficient if there is an A­T rich segment at the end of the gene. E. requires an ATPase in addition to rho factor. 13. Eukaryotic transcription: A. is independent of the presence of consensus sequences upstream from the start of transcription. B. may involve a promoter located within the region transcribed rather than upstream. C. requires a separate promoter region for each of the three ribosomal RNAs transcribed. D. requires that the entire gene be in the nucleosome form of chromatin. E. is affected by enhancer sequences only if they are adjacent to the promoter. 14. All of the following are correct about a primary transcript in eukaryotes EXCEPT it: A. is usually longer than the functional RNA. B. may contain nucleotide sequences that are not present in functional RNA. C. will contain no modified bases. D. usually contains information for more than one RNA molecule. E. contains a TATA box. 15. The processing of transfer RNA involves all of the following EXCEPT: A. addition of a methylated guanosine at the 5  end. B. cleavage of extra bases from both the 3  and 5  ends. C. nucleotide sequence­specific methylation of bases. D. addition of the sequence CCA by a nucleotidyl transferase. E. sometimes, removal of intron from the anticodon region.

 

  

Page 711

16. Cleavage and splicing: A. are features of ribosomal RNA processing. B. always occur in the same way for a given primary transcript. C. remove noninformational sequences occurring anywhere within a primary transcript. D. are usually the first events in mRNA processing. E. are catalyzed by enzymes that recognize and remove specific introns. 17. In the cellular degradation of RNA: A. any of the nucleotides released may be recycled. B. regions of extensive base pairing are more susceptible to cleavage. C. endonucleases may cleave the molecule starting at either the 5  or 3  end. D. the products are nucleotides with a phosphate at either the 3 ­ or 5 ­OH group. E. all species except rRNA are cleaved. Answers 1. C Stacking stabilizes the single­stranded helix. A: Only the four bases A, G, U, and C are incorporated during transcription. B and C: Although single stranded,  RNA exhibits considerable secondary and tertiary structure. D: Only tRNA would be this small; sizes can range to more than 6000 nucleotides. E: This occurs in the  intrachain helical regions (pp. 679–680). 2. E Modified bases seem to be very important in the three­dimensional structure of tRNA (p. 683). 3. C Stability of rRNA is necessary for repeated functioning of ribosomes (p. 683). 4. B These are important additions during processing that yield a functional eukaryotic mRNA (p. 685, Table 16.1). 5. B A: Ribozymes are a very specific type of particle. C: One of the four classes, RNase P, catalyzes a cleavage reaction. D: This is the function of one of the  snRNPs, several of which binding to mRNA result in a spliceosome. E: Ribozymes have been implicated in the processing of ribosomal and tRNAs (p. 686). 6. D This is an important mechanism for driving reactions. A and B: Transcription is directed by the genetic code, generating rRNA precursors in the nucleolus and  mRNA and tRNA precursors in nucleoplasm. C: Eukaryotic transcription may have internal promoter regions as well as enhancers. E: This is a difference from DNA  polymerase (p. 689). 7. B B and C: Enhancer sequences seem to work whether they are at the beginning or end of the gene, but they must be on the same DNA strand as the transcribed  gene. D: They seem to function by binding proteins which themselves bind RNA polymerase (p. 697). 8. E A, D, and E: Sigma factor is required for correct initiation and dissociates from the core enzyme after the first bonds have been formed. Core enzyme can  transcribe but cannot correctly initiate transcription. B and C: Rifampicin binds to the b  subunit, and a ­amanitin is an inhibitor of eukaryotic polymerases (p. 691). 9. D Sigma factor is released when, or a short time after, the initial bond is formed. 10. C The high A­T content of the Pribnow box is believed to facilitate initial unwinding. 11. E Elongation, which requires only the core enzyme, is well underway in this region (p. 695, Figure 16.2). 12. C A, B, and E: There is a rho­dependent as well as a rho­independent process. Rho is a separate protein from RNA polymerase and appears to possess ATPase  activity (p. 696). C and D: Rho­independent termination involves secondary structure, which is stabilized by high G­C content. 13. B RNA polymerase III uses an internal promoter. A: RNA polymerase II activity involves the TATA and CAAT boxes. C: RNA polymerase I produces one  transcript, which is later processed to yield three rRNAs. D: Parts of the promoter are not in a nucleosome. E: Enhancers may be as much as 1000 bp away (pp. 696– 697). 14. E The TATA box is part of the promoter, which is not transcribed. A–D: Modification of bases, cleavage, and splicing are all important events in  posttranscriptional processing to form functional molecules (pp. 700–707). 15. A Capping is a feature of mRNA. B: The primary transcript is longer than the functional molecule. C: The same modifications, catalyzed by a certain (set of)  enzyme(s), occurs at more than one location. D: This is a posttranscriptional modification (pp. 700–701). 16. C A: Cleavage occurs, but splicing does not. B: Alternate splicing leads to different proteins from a single gene. D: Splicing occurs after other events. E: Specificity  of cleavage is related to specific sequences at the intron­exon junctions, not to the sequence of the intron itself (pp. 702–707). 17. D A: Modified bases cannot be recycled. B: Although some enzymes of maturation may require an ordered structure, degradative enzymes are less efficient on an  ordered structure. C: An endonuclease cleaves an interior phosphodiester bond. E: Even rRNA turns over although it is more stable than the other species (p. 709).

 

  

Page 713

Chapter 17—  Protein Synthesis:  Translation and Posttranslational Modifications Dohn Glitz

17.1 Overview

714

17.2 Components of the Translational Apparatus

714

Messenger RNA Is the Carrier of Information Present in DNA

714

Ribosomes Are Workbenches for Protein Biosynthesis

715

Transfer RNA Acts As a Bilingual Translator Molecule

717

The Genetic Code Uses a Four­Letter Alphabet of Nucleotides

718

Codons in mRNA Are Three­Letter Words

718

Punctuation

719

Codon–Anticodon Interactions Permit Reading of mRNA

719

"Breaking" the Genetic Code

720

Mutations

721

Aminoacylation of Transfer RNA Activates Amino Acids for Protein  Synthesis

721

Specificity and Fidelity of Aminoacylation Reactions

722

17.3 Protein Biosynthesis

724

Translation Is Directional and Colinear with mRNA

724

Initiation of Protein Synthesis Is a Complex Process

725

Elongation Is the Stepwise Formation of Peptide Bonds

727

Termination of Polypeptide Synthesis Requires a Stop Codon

733

Translation Has Significant Energy Cost

733

Protein Synthesis in Mitochondria Differs Slightly

733

Some Antibiotics and Toxins Inhibit Protein Biosynthesis

733

17.4 Protein Maturation: Modification, Secretion, and Targeting

735

Proteins for Export Follow the Secretory Pathway

735

Glycosylation of Proteins Occurs in the Endoplasmic Reticulum and Golgi  Apparatus

736

17.5 Organelle Targeting and Biogenesis

739

Sorting of Proteins Targeted for Lysosomes Occurs in the Secretory  Pathway

739

Import of Proteins by Mitochondria Requires Specific Signals

742

Targeting to Other Organelles Requires Specific Signals

742

17.6 Further Posttranslational Protein Modifications Insulin Biosynthesis Involves Partial Proteolysis

743

Proteolysis Leads to Zymogen Activation

743

Amino Acids Can Be Modified after Incorporation into Proteins

744

Collagen Biosynthesis Requires Many Posttranslational Modifications

746

Procollagen Formation in the Endoplasmic Reticulum and Golgi  Apparatus

746

Collagen Maturation

747

17.7 Regulation of Translation

748

17.8 Protein Degradation and Turnover

750

Intracellular Digestion of Some Proteins Occurs in Lysosomes

750

Ubiquitin Is a Marker in ATP­Dependent Proteolysis

751

Bibliography

753

Questions and Answers

754

Clinical Correlations

   

  

743

 

17.1 Missense Mutation: Hemoglobin

721

17.2 Disorders of Terminator Codons

722

17.3 Thalassemia

722

17.4 Mutation in Mitochondrial Ribosomal RNA Results in Antibiotic­ Induced Deafness

734

17.5 I­Cell Disease

740

17.6 Familial Hyperproinsulinemia

743

17.7 Absence of Posttranslational Modification: Multiple Sulfatase  Deficiency

746

17.8 Defects in Collagen Synthesis

749

17.9 Deletion of a Codon, Incorrect Posttranslational Modification, and  Premature Protein Degradation: Cystic Fibrosis

752

Page 714

17.1—  Overview Protein biosynthesis is also called translation since it involves the biochemical translation of information from the four­letter language and structure of nucleic acids into  the 20­letter language and structure of proteins. This process has many requirements: an informational messenger RNA molecule that is exported from the nucleus,  several "bilingual" transfer RNA species that read the message, ribosomes that serve as catalytic and organizational centers, a variety of protein factors, and energy.  Polypeptides are formed by the sequential addition of amino acids in the specific order determined by the information carried in the nucleotide sequence of the mRNA.  The protein is often then matured or processed by a variety of modifications. These may target it to a specific intracellular location or for secretion from the cell, or they  may modulate its activity or function. These complex processes are carried out with considerable speed and extreme precision. Levels of translation are regulated, both  globally and for specific proteins. Finally, when a protein becomes nonfunctional or is no longer needed, it is degraded and its amino acids are catabolized or recycled  into new proteins. Cells vary in their need and ability to synthesize proteins. At one extreme, terminally differentiated red blood cells have a life span of about 120 days, have no nuclei,  do not divide, and do not synthesize proteins because they lack the components of the biosynthetic apparatus. Nondividing cells need to maintain levels of enzymes  and other proteins and carry out limited protein synthesis. Growing and dividing cells must synthesize much larger amounts of protein. Finally, some cells synthesize  proteins for export as well as for their own use. For example, liver cells synthesize large numbers of enzymes needed for their many metabolic pathways as well as  proteins for export, including serum albumin, the major protein of blood plasma or serum. Liver cells are protein factories that are particularly rich in the machinery for  synthesis of proteins. 17.2—  Components of the Translational Apparatus Messenger RNA Is the Carrier of Information Present in DNA Genetic information is stored and transmitted in the nucleotide sequences of DNA. Selective expression of this information requires its transcription into mRNA that  carries specific and precise messages from the nuclear "data bank" to the cytoplasmic sites of protein synthesis. In eukaryotes, the messengers, mRNAs, are usually  synthesized as significantly larger precursor molecules that are processed prior to export from the nucleus. Eukaryotic mRNA in the cytosol has several identifying  characteristics. It is almost always monocistronic, that is, encoding a single polypeptide. The 5  end is capped with a specific structure consisting of 7­ methylguanosine linked through a 5 ­triphosphate bridge to the 5  end of the messenger sequence (see p. 704). A 5 ­nontranslated region, which may be short or up  to a few hundred nucleotides in length, separates the cap from the translational initiation signal, an AUG codon. Usually, but not always, this is the first AUG  sequence encountered as the message is read 5    3 . Uninterrupted sequences that specify a unique polypeptide sequence follow the initiation signal until a specific  translation termination signal is reached. This is followed by a 3 ­untranslated sequence, usually about 100 nucleotides in length, before the mRNA is terminated by a  100­ to 200­nucleotide long polyadenylate tail. Prokaryotic mRNA differs from eukaryotic mRNA in that the 5  terminus is not capped but retains a terminal triphosphate from initiation of its synthesis by RNA  polymerase. Also, most messengers are polycistronic, that is, encoding several polypeptides, and include more than one initiation AUG sequence. A

 

  

Page 715

ribosome­positioning sequence is located about 10 nucleotides upstream of a valid AUG initiation signal. An untranslated sequence follows the termination signal, but  there is no polyadenylate tail. Ribosomes Are Workbenches for Protein Biosynthesis Proteins are assembled on particles called ribosomes. These have two dissimilar subunits, each of which contains RNA and many proteins. With one exception, each  protein is present in a single copy per ribosome, as is each RNA species. The composition of major ribosome types is shown in Table 17.1, and characteristics of their  RNAs are given in Table 16.1. Ribosome architecture has been conserved in evolution. The similarities between ribosomes and subunits from different sources are more obvious than the differences,  and functional roles for each subunit are well defined. Details of ribosome structure and its relationship to function have been learned using many techniques. Overall  size and shape can be determined by electron microscopy. The location of many ribosomal proteins, some elements of the RNA, and functional sites on each subunit  have been determined by electron microscopy of subunits that are complexed with antibodies against a single ribosomal component. The antibody molecule serves as a  physical pointer to the site on the ribosome. Further structural information has been obtained from chemical cross­linking, which identifies near neighbors within the  structure, and from neutron diffraction measurements, which quantitate the distances between pairs of proteins. Ribosomes have been crystallized and X­ray structural  determination is under way. Sequence comparisons and chemical, immunological, and enzymatic probes give information about RNA conformation. Correlations of  structural data with functional measurements in protein synthesis have allowed development of models, such as that in Figure 17.1, that link ribosome morphology to  various functions in translation. Each subunit has an RNA core, folded into a specific three­dimensional structure, upon which proteins are positioned through protein– RNA and protein–protein interactions. Many of these experiments were possible because prokaryotic ribosomes can self­assemble; that is, the native structures can be reconstituted from mixtures of  purified individual proteins and RNAs. Reconstitution of subunits TABLE 17.1 Ribosome Classification and Composition

 

Subunits

  Monomer Size

Ribosome Source Eukaryotes

 

Cytosol

Small

  80S

 

40S:

 

 

34 proteins

 

 

18S RNA

Mitochondria

  55S–60S

Animals

 

 

 

 

40–45S: 16S RNA   70–100 proteins

40S:

60S:

19S RNA

25S, 5S RNAs 70–75 proteins

70S

Chloroplasts

30S:

 

 

20–24 proteins

 

 

16S RNA

Prokaryotes

  70S

Escherichia coli

28S, 5.8S, 5S RNAs

 

12S RNA

77S–80S

 

50 proteins

30S–35S:

  Higher plants

60S:

 

 

 

Large

50S: 34–38 proteins 23S, 5S, 4.5S RNAs

 

 

30S:

50S:

 

 

21 proteins

34 proteins

 

 

16S RNA

23S, 5S RNAs

   

  

Page 716

Figure 17.1  Ribosome structure and functional sites.  Top row shows the faces of each subunit that interact in the functional ribosome.  In (a) the large subunit is shown; note that sites of peptide bond formation and of   binding of the elongation factors are on opposite sides of the bulbous "central  protuberance. " The arm­like structure is somewhat flexible or mobile and is seldom  visualized in complete ribosomes.  In (b) the small subunit is shown with a "platform" or ledge protruding toward the  reader. mRNA and tRNA interact in a "decoding site," deep in the cleft between the  platform and subunit body. The orientation of mRNA and tRNA is depicted, although  their interaction in the decoding site is obscured by the platform.  In (c) the large subunit has been rotated 90° and the arm projects into the page. The exit  site near the base of the subunit is where newly synthesized protein emerges from the  subunit. This area of the subunit is in contact with membranes in the "bound'' ribosomes  of rough endoplasmic reticulum. The site of peptide bond formation, the  peptidyltransferase center, is distant from the exit site; the growing peptide  passes through a groove or tunnel in the ribosome to reach the exit site.  In (d) the small subunit has been rotated 90° such that the platform projects toward the  dish­like face of the large subunit and the cleft is apparent.  In (e) subunits have been brought together to show their relative orientation in the  ribosome. Note that tRNA bound by the small subunit is oriented so that the  aminoacyl acceptor end is near the peptidyltransferase while the translocational  domain (where EF­1a and EF­2 bind) is near the decoding region and the area in  which mRNA enters the complex. Drawings are based on electron microscopy of stained  and unstained, frozen ribosomes. The latter technique preserves native structure and,  perhaps along with X­ray crystallography, should lead to a more detailed and complete  model of the ribosome.

from mixtures in which a single component is omitted or modified can show, for example, if a given protein is required for assembly of the subunit or for some specific  function. An assembly map for large ribosomal subunits of Escherichia coli is shown in Figure 17.2. Total reconstitution of subunits from eukaryotes has not yet been  achieved but the general conclusions about how ribosomes function, although determined using bacterial ribosomes, are fully applicable to eukaryotic systems. Ribosomes are organized in two additional ways. First, several ribosomes often translate a single mRNA molecule simultaneously. Purified mRNA­linked

 

  

Page 717

Figure 17.2  Assembly map of the large ribosomal subunit of E. coli.  The heavy bar at the top represents the 23S rRNA, and the individual ribosomal  proteins are identified by numbers in circles. Arrows that connect components  indicate their interaction. Red arrows from RNA to protein indicate that  the protein binds directly and strongly to RNA, while black arrows indicate a  weaker interaction. Similarly, red arrows between proteins show a strong binding  dependence and black arrows show a lesser dependency. For example, protein L4  binds RNA strongly; it then strongly stimulates binding of proteins L2, L22, and  L29. Protein L2 in turn stimulates binding of proteins L5 and L15. Proteins L5,  L15, and L18 are essential for binding 5S RNA. Proteins within the boxes are  required for a conformational transition that occurs during assembly. Diagram  shows both orderly progression of the assembly process and interdependence of  the components and their specific reactions with other components during the  assembly of the subunit.  Adapted from M. Herold and K. Nierbaus, J. Biol.  Chem. 262–8826, 1987. A similar assembly map for the small subunit was elucidated  earlier. (M. Nomura, Cold Spring Harbor Symp. Quant. Biol. 52:653, 1987.)

polysomes can be visualized by electron microscopy (Figure 17.3). Second, in eukaryotic cells some ribosomes occur free in the cytosol, but many are bound to  membranes of the rough endoplasmic reticulum. In general, free ribosomes synthesize proteins that remain within the cell cytosol or become targeted to the nucleus,  mitochondria, or some of the other organelles. Membrane­bound ribosomes synthesize proteins that will be secreted from the cell or sequestered and function in  other cellular membranes or vesicles. In cell homogenates, membrane fragments and the bound ribosomes constitute the microsome fraction; detergents that disrupt  membranes release these ribosomes.

Figure 17.3  Electron micrographs of polysomes.  (a) Reticulocyte polyribosomes shadowed  with platinum are seen in clusters of  three to six ribosomes, a number consistent  with the size of mRNA for a globin chain.  (b) Uranyl acetate staining in addition to  visualization at a higher magnification shows  polysomes in which parts of the mRNA  are visible.  Courtesy of Dr. Alex Rich, MIT.

Transfer RNA Acts As a Bilingual Translator Molecule All tRNA molecules have several common structural characteristics including the 3 ­terminal CCA sequence to which amino acids are bound, a highly conserved  cloverleaf secondary structure, and an L­shaped three­dimensional structure (see p. 682). But each of the many molecular species has a unique nucleotide sequence,  giving it individual characteristics that allow great specificity in inter­

 

  

Page 718

actions with mRNA and with the aminoacyl­tRNA synthetase that couples one specific amino acid to it. The Genetic Code Uses a Four­Letter Alphabet of Nucleotides Information in the cell is stored in the form of linear sequences of nucleotides in DNA, in a manner that is analogous to the linear sequence of letters of the alphabet in  the words you are now reading. The DNA language uses a simple four­letter alphabet that comprises the two purines, A and G (adenine and guanine), and the two  pyrimidines, C and T (cytosine and thymine). In mRNA the information is encoded in a similar four­letter alphabet, but U (uracil) replaces T. The language of RNA is  thus a dialect of the genetic language of DNA. Genetic information is expressed predominantly in the form of proteins that derive their properties from their linear  sequence of amino acids and to a much lesser extent as RNA species such as tRNA and rRNA. Thus, during protein biosynthesis, the four­letter language of nucleic  acids is translated into the 20­letter language of proteins. Implicit in the analogy to language is the directionality of these sequences. By convention, nucleic acid  sequences are written in a 5    3  direction, and protein sequences from the amino terminus to the carboxy terminus. These directions in mRNA and protein  correspond in both their reading and biosynthetic senses. Codons in mRNA Are Three­Letter Words A 1:1 correspondence of nucleotides to amino acids would only permit mRNA to encode four amino acids, while a 2:1 correspondence would encode 42 = 16 amino  acids. Neither is sufficient since 20 amino acids occur in most proteins. The actual three­letter genetic code has 43 = 64 permutations or words, which is also sufficient  to encode start and stop signals, equivalent to punctuation. The three­base words are called codons and they are customarily shown in the form of Table 17.2. Only  two amino acids are designated by

 

  

Page 719 TABLE 17.3 Nonuniversal Codon Usage in Mammalian  Mitochondria Codon

Usual Code

Mitochondrial Code

UGA

Termination

Tryptophan

AUA

Isoleucine

Methionine

AGA

Arginine

Termination

AGG

Arginine

Termination

  single codons: methionine as AUG and tryptophan as UGG. The rest are designated by two, three, four, or six codons. Multiple codons for a single amino acid  represent degeneracy in the code. The genetic code is nearly universal. The same code words are used in all living organisms, prokaryotic and eukaryotic. An  exception to universality occurs in mitochondria, in which a few codons have a different meaning than in the cytosol of the same organism (Table 17.3). Punctuation Four codons function partly or totally as punctuation, signaling the start and stop of protein synthesis. The start signal, AUG, also specifies methionine. An AUG at an  appropriate site and within an acceptable sequence in mRNA signifies methionine as the initial, amino­terminal residue. AUG codons elsewhere in the message specify  methionine residues within the protein. Three codons, UAG, UAA, and UGA, are stop signals; they specify no amino acid and are known as termination codons  or, less appropriately, as nonsense codons. Codon–Anticodon Interactions Permit Reading of mRNA

Figure 17.4  Codon–anticodon interactions.  Shown are interactions between  (a) the AUG (methionine) codon  and its CAU anticodon and  (b) the CAG (glutamine)  codon and a CUG anticodon.  Note that these interactions  involve antiparallel pairing of  mRNA with tRNA.

Translation of the codons of mRNA involves their direct interaction with complementary anticodon sequences in tRNA. Each tRNA species carries a unique amino  acid, and each has a specific three­base anticodon sequence. Codon–anticodon base pairing is antiparallel, as shown in Figure 17.4. The anticodon is far from the  amino acid­acceptor stem in both the tRNA cloverleaf and the L­shaped three­dimensional structure of all tRNA molecules. (See Chapter 16, p. 682.) Location of  the anticodon and amino acid residue at opposite extremes of the molecule permits the tRNA to conceptually and physically bridge the gap between the nucleotide  sequence of the ribosome­bound mRNA and the site of protein assembly on the ribosome. Since 61 codons designate an amino acid, it might seem necessary to have 61 different tRNA species. This is not the case. Variances from standard base pairing are  common in codon–anticodon interactions. Many amino acids can be carried by more than one tRNA species, and degenerate codons can be read by more than one  tRNA (but always one carrying the correct amino acid). Much of this complexity is explained by the "wobble" hypothesis, which permits less stringent base pairing  between the third position of a codon and the first position of its anticodon. Thus the first two positions of a codon predominate in tRNA selection and the degenerate (third) position is less important. A second modulator of codon–anticodon interactions is the presence of modified nucleotides at or beside the first nucleotide of the  anticodon in many tRNA species. A frequent anticodon nucleotide is inosinic acid (I), the nucleotide of hypoxanthine, which base pairs with U, C, or A. Wobble  base pairing rules are shown in Table 17.4. TABLE 17.4 Wobble Base Pairing Rules 3¢  Codon Base

5¢  Anticodon Bases  Possible

A

U or I

C

G or I

G

C or U

U

A or G or I

  If the wobble rules are followed, the 61 nonpunctuation codons could be read by as few as 31 tRNA molecules, but most cells have 50 or more tRNA

 

  

Page 720

species. Some codons are read more efficiently by one anticodon than another. Not all codons are used equally, some being used very rarely. Examination of many  mRNA sequences has allowed construction of "codon usage" tables that show that different organisms preferentially use different codons to generate similar  polypeptide sequences. "Breaking" the Genetic Code The genetic code (Table 17.2) was determined before methods were developed to sequence natural mRNA. These code­breaking experiments provide insight into  how proteins are synthesized. Important experiments used simple artificial mRNAs or chemically synthesized trinucleotide codons. Polynucleotide phosphorylase catalyzes the template­independent and readily reversible reaction:

where NDP is any nucleoside 5 ­diphosphate or a mixture of two or more. If the nucleoside diphosphate is UDP, a polymer of U designated poly(U) is formed. Under  nonphysiological conditions protein synthesis can occur in vitro without the initiation components that are normally required. With poly(U) as mRNA, the "protein"  polyphenylalanine is made. Similarly, poly(A) encodes polylysine and poly(C) polyproline. An mRNA with a random sequence of only U and C produces  polypeptides that contain not only proline and phenylalanine, as predicted, but also serine (from UCU and UCC) and leucine (from CUU and CUC). Because of  degeneracy in the code and the complexity of the products, experiments with random sequence mRNAs were difficult to interpret, and so synthetic messengers of  defined sequence were transcribed from simple repeating DNA sequences by RNA polymerase. Thus poly(AU), transcribed from a repeating poly(dAT), produces  only a repeating copolymer of Ile­Tyr­Ile­Tyr, read from successive triplets AUA UAU AUA UAU and so on. A synthetic poly(CUG) has possible codons CUG for  Leu, UGC for Cys, and GCU for Ala, each repeating itself once the reading frame has been selected. Since selection of the initiation codon is random in these in  vitro experiments, three different homopolypeptides are produced: polyleucine, polycysteine, and polyalanine. A perfect poly(CUCG) produces a polypeptide with  the sequence (­Leu­Ala­Arg­Ser­) whatever the initiation point. These relationships are summarized in Table 17.5; they show codons to be triplets read in exact  sequence, without overlap or omission. Other experiments used chemically synthesized trinucleotide codons as minimal messages. No proteins were made, but the  binding of only one amino acid (conjugated to an appropriate tRNA) by the ribosome was stimulated by a given codon. It was thus possible to decipher the meaning  of each possible codon and to identify termination codons. All of these conclusions were later verified by the determination of mRNA sequences.

 

  

Page 721

CLINICAL CORRELATION 17.1  Missense Mutation: Hemoglobin Clinically, the most important missense mutation known is the change from A to U in  either the GAA or GAG codon for glutamate to give a GUA or GUG codon for valine in  the sixth position of the b  chain for hemoglobin. An estimated 1 in 10 African­Americans  are carriers of this mutation, which in its homozygous state is the basis for sickle cell  disease, the most common of all hemoglobinopathies (see Clin. Corr. 2.3 for the effects  of this substitution on the polymerization of deoxygenated hemoglobin). The second most  common hemoglobinopathy is hemoglobin C disease, in which a change from G to A in  either the GAA or GAG codon for glutamate results in an AAA or AAG codon for lysine  in the sixth position of the b  chain. Over 600 other hemoglobin missense mutations are  now known. Methods for diagnosis of these and other genetic disorders are discussed in  Clin. Corr. 16.2. A recent advance in therapy of sickle cell anemia uses hydroxyurea  treatment to stimulate synthesis of gchains and thus increase fetal hemoglobin production  in affected adults. This decreases the tendency of the HbS in erythrocytes to form linear  multimers that result in cell shape distortion—that is, sickling—when the oxygen tension  decreases. Charache, S. et al. Effect of hydroxyurea on the frequency of painful crises in sickle cell  anemia. N. Engl. J. Med. 332:1317–1322, 1995.   Mutations An understanding of the genetic code and how it is read provides a basis for understanding the nature of mutations. A mutation is simply a change in a gene. Point  mutations involve a change in a single base pair in the DNA, and thus a single base in the corresponding mRNA. Sometimes this change occurs in the third position of  a degenerate codon and there is no change in the amino acid specified (e.g., UCC to UCA still codes for serine). Such silent mutations are only detected by gene  sequence determination. They are commonly seen during comparison of genes for similar proteins, for example, hemoglobins from different species. Missense  mutations arise from a base change that causes incorporation of a different amino acid in the encoded protein (see Clin. Corr. 17.1). Point mutations can also form or  destroy a termination codon and thus change the length of a protein. Formation of a termination codon from one that encodes an amino acid (see Clin. Corr. 17.2) is  often called a nonsense mutation; it results in premature termination and a truncated protein. Mutation of a termination codon to one for an amino acid allows the  message to be "read through" until another stop codon is encountered. The result is a larger than normal protein. This phenomenon is the basis of several disorders (see  Table 17.6 and Clin. Corr. 17.3). Insertion or deletion of a single nucleotide within the coding region of a gene results in a frameshift mutation. The reading frame is altered at that point and  subsequent codons are read in the new context until a termination codon is reached. Table 17.7 illustrates this phenomenon with the mutant hemoglobin Wayne. The  significance of reading frame selection is underscored by a phenomenon in some viruses in which a single segment of DNA encodes different polypeptides that are  translated using different reading frames. An example is the tumor­causing simian virus SV40 (Figure 17.5), whose small size physically limits the amount of DNA that  can be packaged within it. Aminoacylation of Transfer RNA Activates Amino Acids for Protein Synthesis In order to be incorporated into proteins, amino acids must first be "activated" by linkage to their appropriate tRNA carriers. This is a two­step process that requires  energy and is catalyzed by one of a family of aminoacyl­tRNA synthetases, each of which is specific for a single amino acid and its appropriate tRNA species. The  reactions are normally written as follows:

The brackets surrounding the aminoacyl­AMP–enzyme complex indicate that it is a transient, enzyme­bound intermediate. The "squiggle" (~) linkage of amino acid to  AMP identifies the aminoacyl­adenylate as a high­energy intermediate, a mixed acid anhydride with carboxyl and phosphoryl components. The aminoacyl ester linkage  in tRNA is lower in energy than the aminoacyl­adenylate, but still higher than that of the carboxyl group of the free amino acid. The

 

  

Page 722

CLINICAL CORRELATION 17.2  Disorders of Terminator Codons In hemoglobin McKees Rocks the UAU or UAC codon normally designating tyrosine in  position 145 of the b  chain has mutated to the terminator codon UAA or UAG. This  results in shortening of the b  chain from its normal 146 residues to 144 residues. This  change gives the hemoglobin molecule an unusually high oxygen affinity since the normal  C­terminal sequence involved in binding 2,3­bisphosphoglycerate is modified. The  response to decreased oxygen delivery is secretion of erythropoietin by the kidney and  increased red blood cell production that produces a polycythemic phenotype (see Clin.  Corr. 22.2). Another illness that results from a terminator mutation is a variety of b ­thalassemia.  Thalassemias are a group of disorders characterized at the molecular level by an  imbalance in the stoichiometry of a ­ and b ­globin synthesis. In  0­thalassemia no b ­ globin is synthesized. As a result, a ­globin, unable to associate with b ­globin to form  hemoglobin, accumulates and precipitates in erythroid cells. The precipitation damages  cell membranes, causing hemolytic anemia and stimulation of erythropoiesis. One variety  of  0­thalassemia, common in Southeast Asia, results from a terminator mutation at codon  17 of the b ­globin; the normal codon AAG that designates a lysyl residue at b ­17  becomes the stop codon UAG. In contrast to hemoglobin McKees Rocks, in which the  terminator mutation occurs late in the b ­globin message, the mutation occurs so early in  the mRNA that no useful b ­globin sequence can be synthesized, and b ­globin is absent.  This leads to anemia and aggregation of unused a ­globin in the red cell precursors. In  addition, b ­globin mRNA levels are depressed, probably because premature termination  of translation leads to instability of the mRNA. Winslow, R. M., Swenberg, M., Gross, E., et al. Hemoglobin McKees Rocks  . A human nonsense mutation leading to a shortened b  chain. J. Clin.  Invest. 57:772, 1976. Chang, J. C., and Kan, Y. W. b ­Thalassemia: a nonsense  mutation in man. Proc. Natl. Acad. Sci. USA 76:2886, 1979.   CLINICAL CORRELATION 17.3  Thalassemia There are two expressed a ­globin genes on each chromosome 16. Many instances of a ­ thalassemia arise from the deletion of two, three, or all four copies of the a ­globin gene.  The clinical severity increases with the number of genes deleted. In contrast, the disorders  summarized in Table 17.6 are forms of a ­thalassemia that arise from abnormally long a ­ globin molecules, which replace normal a ­globin, and are present only in small amounts.  These small amounts of a ­globin result from a decreased rate of synthesis or more likely  from an increased rate of breakdown of the abnormally elongated a ­globin. The normal  stop codon, UAA, for a ­globin mutates to any of four sense codons with resultant  placement of four different amino acids at position 142. Normal a ­globin is only 141  residues in length, but the four abnormal a ­globins are 172 residues in length, presumably  because a triplet of nucleotides in the normally untranslated region of the mRNA becomes  a terminator codon in the abnormal position 173. Elongated globin chains can also result  from frameshift mutations or insertions. Weatherall, D. J., and Clegg, J. B. The a ­chain termination mutants and their relationship  to the a ­thalassemias. Philos. Trans. R. Soc. Lond. 271:411, 1975.   reactions are written to show their reversibility. In reality, pyrophosphatases cleave the pyrophosphate released and the equilibrium is strongly shifted toward  formation of aminoacyl­tRNA. From the viewpoint of precision in translation, the amino acid, which had only its side chain (R group) to distinguish it, becomes linked  to a large, complex, and easily recognized carrrier. Specificity and Fidelity of Aminoacylation Reactions Cells contain 20 different aminoacyl­tRNA synthetases, each specific for one amino acid, and at most a small family of carrier tRNAs for that amino acid. In  translation, codon–anticodon interactions define the amino acid to be incorporated. If an incorrect amino acid is carried by the tRNA, it will be incorporated into the  protein. Correct selection of both tRNA and amino acid by the synthetase is necessary to avoid such mistakes. Accuracy of these enzymes is central to the fidelity of  protein synthesis. Aminoacyl­tRNA synthetases share a common mechanism and many are physically associated with one another in the cell. Nevertheless, they are a diverse group of  proteins that may contain one, two, or four identical subunits or pairs of dissimilar subunits. Detailed studies indicate that separate structural domains are involved in  aminoacyl­adenylate formation, tRNA recognition, and, if it occurs, subunit interactions. In spite of their structural diversity, each enzyme is capable of almost error­ free formation of correct aminoacyl­tRNA combinations. TABLE 17.6 "Read Through" Mutation in Termination Codons  Produce Abnormally Long a ­Globin Chains Hemoglobin A

a­ Codon  142

Amino Acid  142

a­Globin Length  (Residues)

UAA

 

141

Constant Spring

CAA

Glutamine

172

Icaria

AAA

Lysine

172

Seal Rock

GAA

Glutamate

172

Koya Dora

UCA

Serine

172

   

  

Page 723

Selection and incorporation of a correct amino acid require great discrimination on the part of some synthetases. While some amino acids be may easily recognized by  their bulk (e.g., tryptophan) or lack of bulk (glycine), or by positive or negative charges on the side chains (e.g., lysine and glutamate), others are much more difficult to  discriminate. Recognition of valine rather than threonine or isoleucine by the valyl­tRNA synthetase is difficult since the side chains differ by either an added hydroxyl or  single methylene group. The amino acid­recognition and ­activation sites of each enzyme have great specificity, as is characteristic of many enzymes. Nevertheless,  misrecognition does occur. An additional "proofreading" or "editing" step increases discrimination. This most often occurs through hydrolysis of the aminoacyl­ adenylate intermediate, with the release of amino acid and AMP. Valyl­tRNA synthetase efficiently hydrolyzes threonyl­adenylate and it hydrolyzes isoleucyl­adenylate  in the presence of bound (but not aminoacylated) tRNAVal. In other cases a misacylated tRNA is recognized and deacylated. Valyl­ and phenylalanyl­tRNA  synthetases deacylate tRNAs that have been mischarged with threonine and tyrosine, respectively. This proofreading is analogous to editing of misincorporated  nucleotides by the 3    5  exonuclease activity of DNA polymerases (Chapter 16). Editing is performed by many but not all aminoacyl­tRNA synthetases. The net  result is an average level of misacylation of one in 104 to 105.

Figure 17.5  Map of genome of simian virus 40 (SV40).  DNA of SV40, shown in red, is  a double­stranded circle of slightly more  than 5000 base pairs that encodes all  information needed by the virus for its  survival and replication within a host  cell. It is an example of extremely  efficient use of the information­coding  potential of a small genome. Proteins  VP1, VP2, and VP3 are structural proteins  of the virus; VP2 and VP3 are translated  from different initiation points to the  same carboxyl terminus. VP1 is translated  in a different reading frame so that its  amino­terminal section overlaps the VP2  and VP3 genes but its amino acid sequence  in the overlapping segment is different  from that of VP2 and VP3. Two additional  proteins, the large T and small t tumor  antigens, which promote transformation of  infected cells, have identical  amino­terminal sequences. The  carboxyl­terminal segment of small t  protein is encoded by a segment of mRNA  that is spliced out of the large T  message, and the carboxyl­terminal  sequence of large T is encoded by DNA  that follows termination of small t. This  occurs through differential processing of  a common mRNA precursor. The single site  of origin of DNA replication (ori) is  outside all coding regions of the genome.

Each synthetase must correctly recognize one to several tRNA species that correctly serve to carry the same amino acid, while rejecting incorrect tRNA species.  Given the complexity of tRNA molecules, this should be simpler than selection of a single amino acid. However, recall the conformational similarity and common  sequence elements of all tRNAs (p. 682). Different synthetases recognize different elements of tRNA structure. One logical element of tRNA recognition by the  synthetase is the anticodon, specific to one amino acid. For example, in the case of tRNAMet, changing the anticodon also alters recognition by the synthetase. In other  instances, this is at least partly true. Sometimes the anticodon is not a determinant of synthetase­tRNA recognition. Consider, for example, suppressor mutations that  "suppress" the expression of classes of chain termination (nonsense) mutations. A point mutation in a glutamine (CAG) codon produces a termination (UAG) codon,  which causes the premature termination of the encoded protein. A second suppressor mutation in the anticodon of a tRNATyr, in which the normal GUA anticodon is  changed to CUA, allows "read through" of the termination codon. The initial mutation is suppressed as a nearly normal protein is made, with the affected glutamine  replaced by tyrosine. Aminoacylation of the mutant tRNAtyr with tyrosine shows that in this case the anticodon does not determine synthetase specificity. In E. coli  tRNAAla, the primary recognition characteristic is a G3­U70 base pair in the acceptor stem; even if no other changes in the tRNAAla occur, any variation at this position  destroys its acceptor ability with alanine­tRNAAla synthetase. Incorporation of a G3­U70 base pair in tRNACys makes it an alanine acceptor, and even the isolated

 

  

Page 724

Figure 17.6  Interaction of a tRNA with its cognate aminoacyl­tRNA synthetase.  Figure shows sugar–phosphate backbone of E. coli glutaminyl  tRNA in green and the peptide backbone of the glutamine  tRNAGln synthetase in multiple colors. Note the  strong interactions of the synthetase with both the  partially unwound acceptor stem and the anticodon loop of  the tRNA, and placement of ATP, shown in red, within a few  angstroms of the 3  end of tRNA. Space­filling models  of the enzyme and tRNA would show both molecules to be  solid objects with several sites of direct contact.  Adapted from J. Perona, M. Rould, and T. Steitz,  Biochemistry 32:8758, 1993.

acceptor stem of tRNAAla can be aminoacylated. Other tRNA identification features include additional elements of the acceptor stem and sometimes parts of the  variable loop or the D­stem/loop. Usually multiple structural elements contribute to recognition, but many are not absolute determinants. The X­ray structure of the  glutaminyl synthetase–tRNA complex shown in Figure 17.6 shows binding at the concave tRNA surface, which is typical and compatible with the biochemical  observations. 17.3—  Protein Biosynthesis Translation Is Directional and Colinear with mRNA In the English language words are read from left to right and not from right to left. Similarly, mRNA sequences are written 5    3  and in the translation process they  are read in the same direction. Amino acid sequences are both written and biosynthesized from the amino­terminal residue to the carboxy terminus. This was first  demonstrated by following the incorporation of radioactive amino acids into specific sites in hemoglobin as a function of time. Only full length, complete globin chains  were isolated and analyzed. Completed chains that incorporated radioactive amino acids during the shortest exposures to the radioactive precursor were near to being  finished at the time of the pulse and were found to have radioactive amino acids only in the carboxy­terminal segments. Longer pulses with radioactive amino acids  resulted also in labeling of central segments of the protein, and the longest pulse time, still corresponding to less than that needed to synthesize a full­length polypeptide,  showed radioactivity approaching the amino­terminal segments. Again, this amino­ to carboxy­terminal directionality became obvious as details of translation were  clarified.

 

  

Page 725

The existence of stable polysomes and the directional nature of translation imply that each ribosome remains bound to an mRNA molecule and moves along the length  of the mRNA until it is fully read. Comparison of mRNA sequences with sequences of the proteins they encode shows a perfect, colinear, gap­free correspondence of  the mRNA coding sequence and that of the synthesized polypeptide. In fact, it is common to deduce the sequence of a protein solely from the nucleotide sequence of  its mRNA or the DNA of the gene encoding it. However, the deduced sequence may differ from the genuine protein because of posttranslational events and  modifications. Initiation of Protein Synthesis Is a Complex Process A good novel can be analyzed in terms of its beginning, its development or middle section, and its satisfactory ending. Protein biosynthesis will be described in a similar  conceptual and mechanical framework: initiation of the process, elongation during which the great bulk of the protein is formed, and termination of synthesis and release  of the finished polypeptide. We will then examine the posttranslational modifications that a protein may undergo. Initiation requires bringing together a small (40S) ribosomal subunit, the mRNA, and a tRNA complex of the amino­terminal amino acid, all in a proper orientation.  This is followed by association of the large (60S) subunit to form a completed initiation complex on an 80S ribosome. The ordered process is shown in Figure 17.7; it  also requires a complex group of proteins, known as initiation factors, that participate only in initiation. They are not ribosomal proteins, although many of them bind  transiently to ribosomes during initiation steps. There are many eukaryotic initiation factors and the specific functions of some remain unclear; prokaryotic protein  synthesis provides a useful and less complex model for comparison. As a first step, eukaryotic initiation factor 2a (eIF­2a) binds to GTP and one species of tRNAMet, designated 

 is recognized by prokaryotic IF­2.

The second step in initiation requires 40S ribosomal subunits associated with a very complex protein, eIF­3. Mammalian eIF­3 includes eight different polypeptides  and has a mass of 600–650 kDa. In electron micrographs eIF­3 is seen bound to the 40S subunit surface that will contact the larger 60S subunit, thus physically  blocking association of 40S and 60S subunits. Hence eIF­3 is also called a ribosome anti­association factor, as is eIF­6, which binds to 60S subunits. A complex  that includes eIF­2a ∙   ∙ GTP ternary complex, correctly oriented mRNA, and several protein factors.

 

  

Page 726

Figure 17.7  Initiation of translation in eukaryotes.  Details are given in the text. Ternary complex (step 1) first combines with  small ribosomal subunit to place the initiator tRNA (step 2). Figure shows interaction with a naked mRNA molecule to form a  preinitiation complex (step 3); additional small subunits later complex with the same mRNA as polysomes are formed. Formation of  the initiation complex (is shown in step 4). The different shape of eIF­2a in complexes with GTP and GDP indicates that  conformational change in the protein occurs upon hydrolysis of triphosphate.

 

  

Page 727

Formation of the complete initiation complex now proceeds with involvement of a 60S subunit and an additional factor, eIF­5. Protein eIF­5 first interacts with the  preinitiation complex; GTP is hydrolyzed to GDP and Pi, and eIF­2a ∙ GDP, eIF­3, and other factors are released. The 40S ∙   ∙ mRNA complex interacts  with a 60S subunit and initiation factor eIF­4d to generate an 80S ribosome with the mRNA and initiator tRNA correctly positioned on the ribosome. The eIF­2a ∙  GDP that is released interacts with the guanine nucleotide exchange factor eIF­2b and GTP to regenerate eIF­2a ∙ GTP for another round of initiation. Prokaryotes use fewer nonribosomal factors and a slightly different order of interaction. Their 30S subunits complexed with a simpler IF­3 first bind mRNA.  Orientation of the mRNA relies in part on base pairing between a pyrimidine­rich sequence of eight nucleotides in 16S rRNA and a purine­rich ''Shine–Dalgarno"  sequence (named for its discoverers) about 10 nucleotides upstream of the initiator AUG codon. Complementarity between rRNA and the message­positioning  sequence of an mRNA may include several mismatches but, as a first approximation, the better the complementary pairing the more efficient initiation at that AUG will  be. It is interesting that eukaryotes do not utilize an mRNA–rRNA base pairing mechanism, but instead use many protein factors to position mRNA correctly. After the  mRNA is bound by a 30S subunit, a ternary complex of IF­2,  , and GTP is bound. A third initiation factor, IF­1, also participates in formation of the  preinitiation complex. A 50S subunit is now bound; in the process, GTP is hydrolyzed to GDP and Pi, and the initiation factors are released. Elongation Is the Stepwise Formation of Peptide Bonds Protein synthesis now occurs by stepwise elongation to form a polypeptide chain. At each step ribosomal peptidyltransferase transfers the growing peptide (or in the  first step the initiating methionine residue) from its carrier tRNA to the a ­amino group of the amino acid residue of the aminoacyl­tRNA specified by the next codon.  Efficiency and fidelity are enhanced by nonribosomal protein elongation factors that utilize the energy released by GTP hydrolysis to ensure selection of the proper  aminoacyl­tRNA species and to move the mRNA and associated tRNAs through the decoding region of the ribosome. Elongation is illustrated in Figure 17.8. At a given moment, up to three different tRNA molecules may be bound at specific sites that span both ribosomal subunits. The initiating methionyl­tRNA is placed in  position so that its methionyl residue may be transferred (or donated) to the free a ­amino group of the incoming aminoacyl­tRNA; it thus occupies the donor site, also  called the peptidyl site or P site of the ribosome. The aminoacyl­tRNA specified by the next codon of the message is bound at the acceptor site, also called the  aminoacyl site or A site of the ribosome. Selection of the correct aminoacyl­tRNA is enhanced by elongation factor 1 (EF­1); a component of EF­1, EF­1a , first  forms a ternary complex with aminoacyl­tRNA and GTP. The EF­1a  ∙ aminoacyl­tRNA ∙ GTP complex binds to the ribosome and if codon–anticodon interactions  are correct, the aminoacyl­tRNA is placed at the A site, GTP is hydrolyzed to GDP and Pi, and the EF­1a  ∙ GDP complex dissociates. The initiating methionyl­tRNA  and the incoming amino­acyl­tRNA are now juxtaposed on the ribosome. Their anticodons are paired with successive codons of the mRNA in the decoding region  of the small subunit, and their amino acids are beside one another at the peptidyltransferase site of the large subunit. Peptide bond formation now occurs.  Peptidyltransferase catalyzes the attack of the a ­amino group of the aminoacyl­tRNA onto the carbonyl carbon of the methionyl­tRNA. The result is transfer of the  methionine to the amino group of the aminoacyl­tRNA, which then occupies a "hybrid"

 

  

Page 728

Figure 17.8  Elongation steps in eukaryotic protein synthesis.  (a) First cycle of elongation is shown. Step 1 shows completed initiation complex with methionyl       in 80S P site. At step 2 an aminoacyl­ tRNA has been placed in the ribosomal A site with participation of EF­1a.  Change in shape of EF­1a shows its conformational change upon GTP hydrolysis. At step 3 the first peptide bond has  been formed, new peptidyl tRNA occupies a hybrid (A/P) site on the ribosome, and the deacylated acceptor stem  of the   is displaced to the E site of the large subunit. At step 4 mRNA–peptidyl tRNA complex has been fully   translocated to the P site while deacylated initiator tRNA is moved to the E site.

 

  

Page 729

 (b) Further rounds of elongation are depicted. Binding of aminoacyl­tRNA probably causes concomitant release of deacylated tRNA  from the E site, resulting in complex at step 5. Formation of the next peptide bond again results in the new peptidyl RNA occupying a  hybrid A/P site on the ribosome (step 6), and translocation moves mRNA and new peptidyl tRNA in register into the P site (step 7).  Additional amino acids are added by successive repetitions of the cycle. For further details see text.

 

  

Page 730

position on the ribosome. The anticodon remains in the 40S A site, while the acceptor end and the attached peptide are in the 60S P site. The anticodon of the  deacylated tRNA remains in the 40S P site, and its acceptor end is located in the 60S exit or E site. The mRNA and the dipeptidyl­tRNA at the 40S A site must now be repositioned to permit another elongation cycle to begin. This is done by elongation factor 2  (EF­2), also called translocase. EF­2 moves the messenger and dipeptidyl­tRNA, in codon–anticodon register, from the 40S A site to the P site. In the process,  GTP is hydrolyzed to GDP plus Pi, providing energy for the movement, and the A site is fully vacated. As the dipeptidyl­tRNA is moved to the P site, the deacylated  donor (methionine) tRNA is also moved to the E site, which only exists on the 60S subunit. The ribosome can now enter a new cycle. The next aminoacyl­tRNA  specified by the mRNA is delivered by EF­1a  to the A site and the deacylated tRNA in the E site is probably released. Peptide transfer again occurs. Successive  cycles of binding of aminoacyl­tRNA, peptide bond formation, and translocation result in the stepwise elongation of the polypeptide toward its eventual carboxyl  terminus. Note that whatever the length of the growing chain, peptide bond formation always occurs through attack of the a ­amino group of the incoming aminoacyl­ tRNA on the peptide carboxyl­tRNA linkage; hence the geometric arrangement of the reacting molecules at the peptidyltransferase site remains constant. Peptide bond formation does not require any additional energy source such as ATP or GTP. The energy of the methionyl (or peptidyl) ester linkage to tRNA drives  the reaction toward peptide bond formation; recall that ATP is used to form each aminoacyl­tRNA and that these reactions are reversible. Isolated 60S subunits can  catalyze peptidyltransferase activity, and nonribosomal factors are not involved in the reaction. Yet peptidyltransferase has never been dissociated from the large  subunit or identified as a specific ribosomal protein. Reconstitution of E. coli peptidyltransferase activity requires only five to six different large subunit proteins and the  rRNA. Omission or significant modification of the rRNA or any of these proteins causes the loss of peptidyltransferase activity, while other proteins can be deleted  with little or no effect. The discovery of catalytic RNA molecules (Chapter 16) led to speculation that the primordial ribosome was an RNA particle in which peptide  bond formation was catalyzed by the RNA. Experiments with very conformationally "stable" large subunit RNA from a thermophilic bacterium suggest that the rRNA  may be the catalytic component of peptidyltransferase, while the proteins serve to stabilize RNA folding; however, this hypothesis remains controversial and not fully  proved. As determined with their prokaryotic equivalents, the role of GTP in the action of EF­1a  and EF­2 probably relates to conformational changes in these proteins.  Crystallographic studies have shown that a large rearrangement of domains with movements of several angstroms occurs upon GTP hydrolysis in EF­Tu, the  prokaryotic equivalent of EF­1a . Both EF­1a  and EF­2 bind ribosomes tightly as GTP complexes, while GDP complexes dissociate from the ribosome more easily.  Viewed another way, GTP stabilizes a protein conformation that confers upon EF­1a  high affinity toward aminoacyl­tRNA and the ribosome, while GDP stabilizes a  conformation with lower affinity for aminoacyl­tRNA and ribosome, thus allowing tRNA delivery and factor dissociation. Restoration of the higher affinity GTP­ associated conformation of EF­1a  requires participation EF­1b g (Figure 17.9). This protein displaces GDP from EF­1a , forming an EF­1a  ∙ EF­1  complex. GTP  then displaces EF­1 , forming an EF­1a  ∙ GTP complex that can successively bind an aminoacyl­tRNA and then a ribosome. Prokaryotes use a similar mechanism  in which EF­Tu binds GTP and aminoacyl­tRNA and EF­Ts displaces GDP and helps recycle the carrier molecule. Prokaryotes also utilize a GTP­dependent  translocase, equivalent to EF­2 but called EF­G or G factor.

 

  

Page 731

Figure 17.9  EF­1 in elongation cycle.  EF­1a   GTP   aminoacyl­tRNA complex (step 1) binds the ribosome (step 2) and transfers aminoacyl­tRNA to the ribosome  (step 2a) with concomitant hydrolysis of GTP and a change in conformation of EF­1a (step 3) that reduces its affinity for  tRNA and ribosome. The GDP is then displaced from EF­1a by EF­1 , resulting in the complex at step 4. Binding of GTP then  displaces EF­1  (step 5) and allows binding of an aminoacyl­tRNA by EF­1a in its higher affinity conformation (step 1). In  prokaryotes a similar cycle exists; EF­Tu functions as the carrier of aminoacyl­tRNA and EF­Ts is guanine nucleotide exchange factor. ∙

 

  



Page 732

Figure 17.10  Model of termination of protein biosynthesis.  When a termination codon (UAG, UAA, or UGA) in mRNA occupies the ribosomal  A site, binding of release factor–GTP complex occurs (step 1), probably with  concomitant release of deacylated tRNA from the ribosomal E site. In step 2  peptidyltransferase now functions as a hydrolase; protein is released by hydrolysis  of the ester bond linking it to tRNA, and acceptor end of deacylated tRNA is  probably displaced. GTP is hydrolyzed to GDP and P , presumably altering the  i

conformation of the release factor. Complex is now dissociated and  components can enter additional rounds of protein synthesis.

 

  

Page 733

Termination of Polypeptide Synthesis Requires a Stop Codon A chain­terminating UAG, UAA, or UGA codon in the A site does not promote binding of any tRNA species. Instead, another complex nonribosomal protein,  release factor (eRF), binds the ribosome as an eRF ∙ GTP complex (Figure 17.10). The peptide–tRNA ester linkage is cleaved through the action of peptidyl  transferase, acting here as a hydrolase, and the completed polypeptide is released from its carrier tRNA and the ribosome. Dissociation of eRF from the ribosome  requires hydrolysis of the GTP and frees the ribosome to dissociate into subunits and then reenter the protein synthesis cycle at the initiation stage. In prokaryotes three  release factors, RF­1, RF­2, and RF­3, carry out the termination function. The factor RF­1 acts in response to UAG or UAA codons, RF­2 acts in response to UGA  or UAA codons, and RF­3 is a GTPase that activates RF­1 and RF­2. Translation Has Significant Energy Cost There is a considerable use of energy in synthesis of a polypeptide. Amino acid activation converts an ATP to AMP and pyrophosphate, which is normally hydrolyzed  to Pi; the net cost is two high­energy phosphates. Two more high­energy bonds are hydrolyzed in the actions of EF­1a  and EF­2, for a total of four per peptide bond  formed. Posttranslational modifications may add to the energy cost, and of course energy is needed for biosynthesis of the multi­use mRNA, tRNAs, ribosomes, and  protein factors, but these costs are distributed among the proteins formed during their lifetime. Protein Synthesis in Mitochondria Differs Slightly Many characteristics of mitochondria suggest that they are descendants of aerobic prokaryotes that invaded and set up a symbiotic relationship within a eukaryotic cell.  Some of their independence and prokaryotic character are retained. Human mitochondria have a circular DNA genome of 16,569 base pairs that encodes 13  proteins, 22 tRNA species, and two mitochondrion­specific rRNA species. Their independent apparatus for protein synthesis includes RNA polymerase, aminoacyl­ tRNA synthetases, tRNAs, and ribosomes. Although the course of protein biosynthesis in mitochondria is like that in the cytosol, some details are different. The  synthetic components, tRNAs, aminoacyl­tRNA synthetases, and ribosomes, are unique to the mitochondrion. The number of tRNA species is small and the genetic  code is slightly different (see Table 17.3). Mitochondrial ribosomes are smaller and the rRNAs are shorter than those of either the eukaryotic cytosol or of prokaryotes  (see Table 17.1). An initiator  . Most mitochondrial proteins are encoded in nuclear DNA and synthesized in the cytosol, but mitochondrial protein  synthesis is clearly important (see Clin. Corr. 17.4). Cells must also coordinate protein synthesis within mitochondria with the cytosolic synthesis of proteins destined  for import into mitochondria. Some Antibiotics and Toxins Inhibit Protein Biosynthesis Protein biosynthesis is central to the continuing life and reproduction of cells. An organism can gain a biological advantage by interfering in the ability of its competitors  to synthesize proteins, and many antibiotics and toxins function in this way. Some are selective for prokaryotic rather than eukaryotic protein synthesis and so are  extremely useful in clinical practice. Examples of antibiotic action are listed in Table 17.8. Several mechanisms of interfering in ribosome subunit–tRNA interactions are utilized by different antibiotics. Streptomycin binds the small subunit of

 

  

Page 734

CLINICAL CORRELATION 17.4  Mutation in Mitochondrial Ribosomal RNA Results in Antibiotic­Induced  Deafness In some regions of China a significant percentage of irreversible cases of deafness has  been linked to use of normally safe and effective amounts of aminoglycoside antibiotics  such as streptomycin and gentamicin. The unusual sensitivity to aminoglycosides is  transmitted only through women. This maternal transmission suggests a mitochondrial  locus, since sperm do not contribute mitochondria to the zygote. Aminoglycosides are  normally targeted to bacterial ribosomes, so the mitochondrial ribosome is a logical place  to look for a mutation site. A single A   G point mutation at nucleotide 1555 of the gene on mitochondrial DNA for  the rDNA of the large subunit has been identified in three families with this susceptibility to  aminoglycosides. The mutation site is in a highly conserved region of the rRNA sequence  that is known to be involved in aminoglycoside binding; some mutations in the same  region confer resistance to the antibiotics, and the RNA region is part of the ribosomal A  site. It is hypothesized that the mutation makes the region more "prokaryote­like,"  increasing its affinity for aminoglycosides and the ability of the antibiotic to interfere in  protein synthesis in the mitochondrion. Proteins synthesized in the mitochondrion are  needed to form the enzyme complexes of the oxidative phosphorylation system, so  affected cells are starved of ATP. Aminoglycosides accumulate in the cochlea, making  this a particularly sensitive target and leading to sensorineural deafness. Fischel­Ghodsian, N., Prezant, T., Bu., X., and Öztas, S. Mitochondrial ribosomal RNA  gene mutation in a patient with sporadic aminoglycoside ototoxicity. Am. J. Otolaryngol.  14:399, 1993. Prezant, T., Agapian, J., Bohlman. M., et al. Mitochondrial ribosomal  RNA mutation associated with both antibiotic­induced and non­syndromic deafness.  Nature Genetics 4:289, 1993.   prokaryotic ribosomes, interferes with the initiation of protein synthesis, and causes misreading of mRNA. Although streptomycin does not directly bind ribosomal  protein S12 of the small subunit, mutations in this protein or in the small subunit rRNA can confer resistance to or even dependence on streptomycin. Protein S12 is  involved in tRNA binding, and streptomycin alters the interactions of tRNA with the ribosomal subunit and mRNA, probably by affecting subunit conformation. Other  aminoglycoside antibiotics, such as the neomycins or gentamicins, also cause mistranslation; they interact with the small ribosomal subunit, but at sites that differ  from that for streptomycin. The aminoglycoside kasugamycin binds small subunits and inhibits the initiation of translation. Kasugamycin sensitivity depends on base  methylation that normally occurs on two adjacent adenine moieties of small subunit rRNA. Tetracyclines bind directly to ribosomes and interfere in aminoacyl­tRNA  binding. Other antibiotics interfere with elongation. Puromycin (Figure 17.11) resembles an aminoacyl­tRNA; it binds at the ribosomal A site and acts as an acceptor in the  peptidyltransferase reaction. However, since it does not interact with mRNA it cannot be translocated, and since its aminoacyl derivative is not in an ester linkage to  the nucleoside it cannot serve as a peptide donor. Thus puromycin prematurely terminates translation, leading to release of peptidyl­puromycin. Chloramphenicol  directly inhibits peptidyltransferase by binding the transferase center; no transfer occurs, and peptidyl­tRNA remains associated TABLE 17.8 Some Inhibitors of Protein Biosynthesis Inhibitor

Processes Affected

Site of Action

Streptomycin

Initiation, elongation

Prokaryotes: 30S subunit

Neomycins

Translation

Prokaryotes: multiple sites

Tetracyclines

Aminoacyl­tRNA binding

30S or 40S subunits

Puromycin

Peptide transfer

70S or 80S ribosomes

Erythromycin

Translocation

Prokaryotes: 50S subunit

Fusidic acid

Translocation

Prokaryotes: EF­G

Cycloheximide

Elongation

Eukaryotes: 80S ribosomes

Ricin

Multiple

Eukaryotes: 60S subunit

   

  

Page 735

with the ribosome. The translocation step is also a potential target. Erythromycin, a macrolide antibiotic, interferes with translocation on prokaryotic ribosomes.  Eukaryotic translocation is inhibited by diphtheria toxin, a protein toxin produced by Corynebacterium diphtheriae, the toxin binds at the cell membrane and a  subunit enters the cytoplasm and catalyzes the ADP­ribosylation and inactivation of EF­2, as represented in the reaction:

ADP­ribose is attached to EF­2 at a posttranslationally modified histidine residue known as diphthamide. Posttranslational events are discussed in the next section. A third group of toxins attack the rRNA. Ricin (from castor beans) and related toxins are N­glycosidases that cleave a single adenine from the large subunit rRNA  backbone. The ribosome is inactivated by this apparently minor damage. A fungal toxin, a ­sarcin, cleaves large subunit rRNA at a single site and similarly inactivates  the ribosome. Some E. coli strains make extracellular toxins that affect other bacteria. One of these, colicin E3, is a ribonuclease that cleaves 16S RNA near the  mRNA­binding sequence and decoding region; it thus inactivates the small subunit and halts protein synthesis in competitors of the colicin­producing cell. 17.4—  Protein Maturation:  Modification, Secretion, and Targeting Some proteins emerge from the ribosome ready to function, while others undergo a variety of posttranslational modifications. These alterations may result in  conversion to a functional form, direction to a specific subcellular compartment, secretion from the cell, or an alteration in activity or stability. Information that  determines the posttranslational fate of a protein resides in its structure: that is, the amino acid sequence and conformation of the polypeptide determine whether a  protein will be a substrate for a modifying enzyme and/or identify it for direction to a subcellular or extracellular location. Proteins for Export Follow the Secretory Pathway Proteins destined for export are synthesized on membrane­bound ribosomes of the rough endoplasmic reticulum (ER) (Figure 17.12). A ribosome has no means of  classifying the polypeptide it is about to synthesize, so initiation and elongation begin on free cytosolic ribosomes. Proteins of the secretory pathway have a  hydrophobic signal peptide, usually at or near their amino terminus. There is no unique signal peptide sequence, but its characteristics include a positively charged N  terminus, a core of 8–12 hydrophobic amino acids, and a more polar C­terminal segment that eventually serves as a cleavage site for excision of the signal peptide. The signal peptide of 15–30 amino acids emerges from the ribosome early during polypeptide synthesis. As it appears it is bound by a cytosolic signal recognition  particle (SRP) (see Figure 17.13). The SRP is an elongated particle made up of six different proteins plus a small (7S) RNA molecule that serves as a backbone.  Binding to SRP halts protein synthesis and the ribosome moves to the ER. SRP recognizes and binds to an SRP receptor or "docking protein," localized at the  cytosolic surface of the ER membrane, in a reaction that requires GTP hydrolysis and presumably involves conformational changes in the SRP and/or the receptor. The  ribosome is transferred to a "translocon," a ribosome receptor on the membrane that serves as a passageway through the membrane. Both SRP and docking protein  are freed to direct other ribosomes to the ER,

Figure 17.11  Puromycin (right) interferes  with protein synthesis by  functioning as an analog  of aminoacyl­tRNA, here  tyrosyl­tRNA (left) in  peptidyltransferase reaction.

 

  

Page 736

and the translational block caused by SRP binding is relieved. The hydrophobic signal sequence, probably complexed by a receptor protein, is inserted into the  membrane, further anchoring the ribosome to the ER. Translation and extrusion into or through the membrane are now coupled. Translocon proteins form a pore or  channel through which the growing polypeptide passes; even very hydrophilic or ionic segments are directed through the hydrophobic membrane into the ER lumen and  folding into secondary and tertiary structures begins.

Figure 17.12  Rough endoplasmic reticulum of a plasma cell.  Three parallel arrows indicate three ribosomes  among the many attached to the extensive  membranes. Single arrow indicates a  mitochondrion for comparison.  Courtesy of Dr. U. Jarlfors,  University of Miami.

The completed export­destined protein within the ER lumen will probably be anchored to the membrane by the signal peptide. A cleavage site on the protein is  hydrolyzed by signal peptidase, an integral membrane protein located at the luminal surface of the ER. The protein completes folding into a three­dimensional  conformation, disulfide bonds can form, and components of multisubunit proteins may assemble. Other steps may include proteolytic processing and glycosylation that  occur within the ER lumen and during transit of the protein through the Golgi apparatus and into secretory vesicles. Glycosylation of Proteins Occurs in the Endoplasmic Reticulum and Golgi Apparatus Glycosylation of proteins to form glycoproteins (see p. 60) is important for two reasons. Glycosylation alters the properties of proteins, changing their stability,  solubility, and physical bulk. In addition, carbohydrates of glycoproteins act as recognition signals that are central to aspects of protein targeting and for cellular  recognition of proteins and other cells. Glycosylation can involve addition of a few carbohydrate residues or the formation of large branched oligosaccharide chains.  Sites and types of glycosylation are determined by the presence on a protein of appropriate amino acids and sequences, and by availability of enzymes and substrates  to carry out the glycosylation reactions.

Figure 17.13  Secretory pathway: signal peptide recognition.  At step A a hydrophobic signal peptide emerges from the exit site of a free ribosome  in the cytosol. Signal recognition particle (SRP) recognizes and binds the peptide and  peptide elongation is temporarily halted (step B). The ribosome moves to the ER  membrane where docking protein binds to SRP (step C). In step D the ribosome is  transferred to a ribosome receptor or translocon, protein biosynthesis is resumed,   and newly synthesized protein is extruded through the membrane into the ER lumen.

 

  

Page 737 TABLE 17.9 Glycosyltransferases in Eukaryotic Cells Sugar Transferred

Abbreviation

Donors

Glycosyltransferase

Mannose

Man

GDP­Man

Mannosyltransferase

 

 

Dolichol­Man

 

Galactose

Gal

UDP­Gal

Galactosyltransferase

Glucose

Glc

UDP­Glc

Glucosyltransferase

 

 

Dolichol­Glc

 

Fucose

Fuc

GDP­Fuc

Fucosyltransferase

N­Acetylgalactosamine

GalNAc

UDP­GalNac

N­acetylgalactosaminyltransferase

N­Acetylglucosamine

GlcNAc

UDP­GlcNAc

N­acetylglucosaminyltransferase

N­Acetylneuraminic acid (or  sialic acid)

NANA or NeuNAc  CMP­NANA  SA CMP­SA

N­Acetylneuraminyltransferase  (sialyltransferase)

  Glycosylation involves many glycosyltransferases, classes of which are summarized in Table 17.9. Up to 100 different enzymes each carry out a similar basic  reaction in which a sugar is transferred from an activated donor substrate to an acceptor, usually another sugar residue that is part of an oligosaccharide under  construction. The enzymes show three kinds of specificity: for the monosaccharide that is transferred, for structure and sequence of the acceptor molecule, and for the  site and configuration of the anomeric linkage formed. One class of glycoproteins has sugars linked through the amide nitrogen of asparagine residues in the process of N­linked glycosylation. The antibiotic tunicamycin,  which prevents N­glycosylation, has been valuable in elucidating the biosynthetic pathway. Formation of N­linked oligosaccharides begins in the ER lumen and  continues after transport of the protein to the Golgi apparatus. A specific sequence, Asn­X­Thr (or Ser) in which X may be any amino acid except proline or aspartic  acid, is required for N­glycosylation. Not all Asn­X­Thr/Ser sequences are glycosylated because some may be unavailable due to protein conformation. Biosynthesis of N­linked oligosaccharides begins with the synthesis of a lipid­linked intermediate (Figure 17.14). Dolichol phosphate (structure on p. 350) at the  cytoplasmic surface of the ER membrane serves as glycosyl acceptor of N­acetylglucosamine. The GlcNAc­pyrophosphoryldolichol is an acceptor for stepwise  glycosylation and formation of a branched (Man)5(GlcNAc)2­pyro­phosphoryldolichol on the cytosolic side of the membrane. This intermediate is then reoriented to  the luminal surface of the ER membrane, and four additional mannose and then three glucose residues are sequentially added to complete the structure. The complete  oligosaccharide is then transferred from its dolichol carrier to an asparagine residue of the polypeptide as it emerges into the ER lumen. Thus N­glycosylation is  cotranslational, that is, occurs as the protein is being synthesized, hence it can affect protein folding. Processing or modification of the oligosaccharide by glycosidases involves removal of some sugar residues from the newly transferred structure. The glucose  residues, which were required for transfer of the oligosaccharide from the dolichol carrier, are sequentially removed, as is one mannose. These alterations mark the  glycoprotein for transport to the Golgi apparatus where further trimming by glycosidases may occur. Additional sugars may also be added by a variety of  glycosyltransferases. The resulting N­linked oligosaccharides are diverse, but two classes are distinguishable. Each has a common core region (GlcNAc2Man3) linked  to asparagine and originating from the dolichol­linked intermediate. The high­mannose type includes mannose residues in a variety of linkages and shows less  processing from the dolichol­linked intermediate. The complex type is more highly processed and diverse, with a larger variety of sugars and linkages. Examples of  mature oligosaccharides are shown in Figure 17.15.

 

  

Page 738

The second major class of glycoproteins have sugars that are bound through either serine or threonine hydroxyl groups. Such O­linked glycosylation occurs only after  the protein has reached the Golgi apparatus, hence O­glycosylation is posttranslational and occurs only on fully folded proteins. O­linked carbohydrates always  involve N­acetylgalactosamine attachment to a serine or threonine residue of the protein. There is no defined amino acid sequence in which the

Figure 17.14  Biosynthesis of N­linked oligosaccharides at the surface of the endoplasmic reticulum.  Synthesis is initiated on the cytoplasmic face of the ER  membrane by transfer of N­acetylglucosamine phosphate to a dolichol acceptor  (step A) followed by formation of the first glycosidic bond upon transfer of a  second residue of N­acetylglucosamine (step B). Five residues of mannose are then  added sequentially (step C) from a GDP mannose carrier. At this stage  lipid­linked oligosaccharide is reoriented to the luminal face of the membrane,  and additional mannose (step D) and glucose (step E) residues are transferred  from dolichol­linked intermediates. Dolichol sugars are generated from cytosol  nucleoside diphosphate sugars. The completed oligosaccharide is finally  transferred to a protein in the process of being synthesized at the membrane  surface; signal peptide may have already been cleaved at this point.

 

  

Page 739

Figure 17.15  Structure of N­linked oligosaccharides.  Basic structures of both types of N­linked oligosaccharides are shown. In each  case structure is derived from that of the initial dolichol­linked oligosaccharide  through action of glycosidases and glycosyltransferases. Note the variety of  glycosidic linkages involved in these structures.

serine or threonine must occur, but only residues whose side chains are in an appropriate environment on the protein surface serve as acceptors for the GalNAc­ transferase. Sequential addition of sugars to the GalNAc acceptor follows, using the same glycosyltransferases that modified N­linked oligosaccharides in the Golgi apparatus. The  structures synthesized depend on types and amounts of glycosyltransferases in a given cell. If an acceptor is a substrate for more than one transferase, the amount of  each transferase controls the competition between them. Some oligosaccharides may be formed that are not acceptors for any glycosyltransferase present, hence no  further growth of the chain occurs. Other structures may be excellent acceptors that continue to grow until completed by one of a number of nonacceptor termination  sequences. These processes can lead to many different oligosaccharide structures on otherwise identical proteins, so heterogeneity in glycoproteins is common.  Examples are shown in Figure 17.16. 17.5—  Organelle Targeting and Biogenesis Sorting of Proteins Targeted for Lysosomes Occurs in the Secretory Pathway Protein transport from ER to Golgi apparatus occurs through carrier vesicles that bud from the ER. This transport requires GTP; inhibitors of oxidative phosphorylation  cause proteins to accumulate in the ER and vesicles. Sorting of proteins for their ultimate destinations occurs in conjunction with their glycosylation and proteolytic  trimming as they pass through the cis, medial, and trans elements of the Golgi apparatus.

 

  

Page 740

Figure 17.16  Examples of oligosaccharide structure.  Structures 1–3 are typical N­linked oligosaccharides of high­mannose (1) and complex types  (2, 3); note the common core structure from the protein asparagine residue through the first  branch point. Structures 4–8 are common O­linked oligosaccharides that may be quite simple  or highly complex. Note that although the core structure (GalNAc­Ser/Thr) is unlike that of  N­linked oligosaccharides, the termini can be quite similar (e.g., structures 2 and 6, 3, and 7).  Abbreviations: Man = mannose; Gal = galactose; Fuc = fucose; GlcNAc = N­acetylglucosamine;  GalNAc = N­acetylgalactosamine; NANA = N­acetylneuraminic acid (sialic acid).  Adapted from J. Paulson, Trends Biochem. Sci. 14:272, 1989.

CLINICAL CORRELATION 17.5  I­Cell Disease I­cell disease (mucolipidosis II) and pseudo­Hurler polydystrophy (mucolipidosis III) are  related diseases that arise from defects in lysosomal enzyme targeting because of a  deficiency in the enzyme that transfers N­acetylglucosamine phosphate to the high  mannose­type oligosaccharides of proteins destined for the lysosome. Fibroblasts from  affected individuals show dense inclusion bodies (hence I­cells) and are defective in  multiple lysosomal enzymes that are found secreted into the medium. Patients have  abnormally high levels of lysosomal enzymes in their sera and other body fluids. The  disease is characterized by severe psychomotor retardation, many skeletal abnormalities,  coarse facial features, and restricted joint movement. Symptoms are usually observable at  birth and progress until death, usually by age 8. Pseudo­Hurler polydystrophy is a much  milder form of the disease. Onset is usually delayed until the age of 2–4 years, the disease  progresses more slowly, and patients survive into adulthood. Prenatal diagnosis of both  diseases is possible, but there is as yet no definitive treatment. For a review of lysosomal enzyme trafficking, see Kornfeld, S. J. Clin. Invest. 77:1,  1986. For a comprehensive review of these diseases, see Kornfeld, S., and Sly, W. S.,  I­cEll disease and pseudo­Hurler polydystrophy. In: C. R. Scriver, A. L. Beaudet, W. S.  Sly, and D. Valle (Eds.), The Molecular and Metabolic Basis of Inherited Disease, 7th  ed. New York: McGraw­Hill, 1995, pp. 2495–2508.   The best understood sorting process is targeting of specific glycoproteins to lysosomes. In the cis Golgi some aspect of tertiary structure allows lysosomal proteins to  be recognized by a glycosyltransferase that attaches N­acetylglucosamine phosphate (GlcNAc­P) to high­mannose type oligosaccharides. A glycosidase then removes  the GlcNAc, forming an oligosaccharide that contains mannose 6­phosphate (Figure 17.17) that is recognized by a receptor protein responsible for  compartmentation and vesicular transport of these proteins to lysosomes. Other oligosaccharide chains on the proteins may be further processed to form complex type  structures, but the mannose 6­phosphate determines the lysosomal destination of these proteins. Patients with I­cell disease lack the GlcNAc­P glycosyltransferase  and cannot correctly mark lysosomal enzymes for their destination. Thus the enzymes are secreted from the cell (see Clin. Corr. 17.5). Other sorting signals are reasonably well understood. Proteins are retained in the ER lumen in response to a C­terminal KDEL (Lys­Asp­Glu­Leu) sequence, and a  different sequence in an exposed C terminus signals retention in the ER membrane. Transmembrane domains have been identified that result in reten­

 

  

Page 741

Figure 17.17  Targeting of enzymes to lysosomes.  Completed N­linked glycoprotein is released from ER membrane, and  during transport to and through the Golgi apparatus the  oligosaccharide is modified by glycosidases that remove glucose  residues (step 1). Some mannose residues may also be removed. An  element of protein structure is then recognized by a  glycosyltransferase that transfers one or sometimes two  N­acetylglucosamine phosphate residues to the oligosaccharide  (step 2). A glycosidase removes N­acetylglucosamine, leaving one  or two mannose 6­phosphate residues on the oligosaccharide  (step 3). The protein is then recognized by a mannose 6­phosphate  receptor and directed to lysosomes.  Adapted from R. Kornfeld and S. Kornfeld, Annu. Rev. Biochem.  54:631, 1985.

tion in the Golgi. Polypeptide­specific glycosylation and sulfation of some glycoprotein hormones in the anterior pituitary mediate their sorting into storage granules.  Polysialic acid modification of a neural cell adhesion protein appears to be both specific to the protein and regulated developmentally. Many other sorting signals must  still be deciphered to explain fully how the Golgi apparatus directs proteins to its own subcompartments, various storage and secretory granules, and specific elements  of the plasma membrane. The secretory pathway directs proteins to lysosomes, the plasma membrane, or outside the cell. Proteins of the ER and Golgi apparatus are targeted through partial  use of the pathway. For example, localization of proteins on either side of or spanning the ER membrane can utilize the signal recogni­

 

  

Page 742

Figure 17.18  Topology of proteins at membranes of endoplasmic reticulum.  Proteins are shown in several orientations with respect to the membrane.  In (a) the protein is anchored to the luminal surface of the membrane by an  uncleaved signal peptide.  In (b) the signal sequence is not near the N terminus; a domain of the protein was  synthesized before emergence of signal peptide. Insertion of the internal signal  sequence, followed by completion of translation, resulted in a protein with a  cytoplasmic N­terminal domain, a membrane­spanning central segment, and a  C­terminal domain in the ER lumen.  Diagram (c) shows a protein with the opposite orientation: an N­terminal  signal sequence, which might also have been cleaved by signal peptidase,  resulted in extrusion of a segment of protein into the ER lumen. A second  hydrophobic anchoring sequence remained membrane associated and prevented  passage of the rest of the protein through the membrane, thus allowing formation  of a C­terminal cytoplasmic domain.  In (d), several internal signal and anchoring sequences allow various segments of the  protein to be oriented on each side of the membrane.

tion particle in slightly different ways (Figure 17.18). If the signal sequence is downstream from the amino terminus of the protein, the amino end may not be inserted  into the membrane and may remain on the cytoplasmic surface. Internal hydrophobic anchoring sequences within a protein can allow much of the sequence either to  remain on the cytoplasmic surface or to be retained, anchored on the luminal surface of the ER membrane. Multiple anchoring sequences in a single polypeptide can  cause it to span the membrane several times and thus be largely buried in it. Such hydrophobic sequences are separated by polar loops whose orientation is  determined by positively charged flanking residues that predominate on the cytoplasmic side of the membrane. Import of Proteins by Mitochondria Requires Specific Signals Mitochondria provide a particularly complex targeting problem since specific proteins are located in the mitochondrial matrix, inner or outer membrane, or  intermembrane space. Most of these proteins are synthesized in the cytosol on free ribosomes and imported into the mitochondrion, and most are synthesized as larger  preproteins; N­terminal presequences mark the protein not only for the mitochondrion but also for a specific subcompartment. The mitochondrial matrix targeting  signal is not a specific sequence, but rather a positively charged amphiphilic a ­helix. With the aid of a protein chaperone, it is recognized by a mitochondrial  receptor and the protein is translocated across both membranes and into the mitochondrial matrix in an energy­dependent reaction. Passage occurs at adhesion sites  where the inner and outer membranes are close together. Proteases remove the matrix targeting signal but may leave other sequences that further sort the protein within  the mitochondrion. For example, a clipped precursor of cytochrome­b2 is moved back across the inner membrane in response to a hydrophobic signal sequence.  Further proteolysis frees the protein in the intermembrane space. In contrast, cytochrome­c apoprotein (without heme) binds at the outer membrane and is passed into  the intermembrane space. There it acquires its heme and undergoes a conformational change that prevents return to the cytosol. Outer membrane localization can utilize  the matrix targeting mechanism to translocate part of the protein, but a large apolar sequence blocks full transfer and leaves a membrane­bound protein with a C­ terminal domain on the surface of the mitochondrion. Targeting to Other Organelles Requires Specific Signals Nuclei must import many proteins involved in their own structure and for DNA replication, transcription, and ribosome biogenesis. Nuclear pores permit the

 

  

Page 743

passage of small proteins, but larger proteins are targeted by nuclear localization signals that include clusters of basic amino acids. Some nuclear proteins may be  retained in the nucleus by forming complexes within the organelle. Peroxisomes contain a limited array of enzymes. One targeting signal is a carboxy­terminal tripeptide,  Ser­Lys­Leu (SKL). An N­terminal targeting signal also exists, and others may yet be discovered. A different targeting problem exists for proteins that reside in more than one subcellular compartment. Sometimes gene duplication and divergence have resulted in  different targeting signals on closely related mature polypeptides. Alternative transcription initiation sites or pre­mRNA splicing can generate different messages  from a single gene. An example of the latter is seen in a calcium–calmodulin­dependent protein kinase; alternatively spliced mRNAs differ with respect to an internal  segment that encodes a nuclear localization signal. Without this segment, the protein remains in the cytosol. Alternative translation initiation sites lead to two forms  of rat liver fumarase, one of which includes a mitochondrial targeting sequence while the other does not and remains in the cytosol. A suboptimal localization signal can  lead to inefficient targeting and a dual location, as is seen in the partial secretion of an inhibitor of the plasminogen activator. Finally, some proteins contain more than  one targeting signal, which must compete with each other. 17.6—  Further Posttranslational Protein Modifications Several additional maturation events may modify newly synthesized polypeptides to help generate their final, functional structures. Many of these events are very  common, while others are specialized to one or a few known instances. CLINICAL CORRELATION 17.6  Familial Hyperproinsulinemia Familial hyperproinsulinemia, an autosomal dominant condition, results in approximately  equal amounts of insulin and an abnormally processed proinsulin being released into the  circulation. Although affected individuals have high levels of proinsulin in their blood, they  are apparently normal in terms of glucose metabolism, being neither diabetic nor  hypoglycemic. The defect was originally thought to result from a deficiency of one of the  proteases that process proinsulin. Three enzymes process proinsulin: endopeptidases that  cleave the Arg31–Arg32 and Lys64–Arg65 peptide bonds, and a carboxypeptidase. In  several families the defect is the substitution of Arg65 by His or Leu, which prevents  cleavage between the C­peptide and the A chain of insulin, resulting in secretion of a  partially processed proinsulin. In one family a point mutation (His10 Asp10) causes the  hyperproinsulinemia, but how this mutation interferes with processing is not known. Steiner, D. F., Tager, H. S., Naujo, K., Chan, S. J., and Rubenstein, A. H. Familial  syndromes of hyperproinsulinemia with mild diabetes. In: C. R. Scriver, A. L. Beaudet,  W. S. Sly, and D. Valle (Eds.), The Molecular and Metabolic Basis of Inherited  Disease, 7th ed. New York: McGraw­Hill, 1995, pp. 897–904.   Insulin Biosynthesis Involves Partial Proteolysis Partial proteolysis of proteins is a common maturation step. Sequences can be removed from either end or from within the protein. Proteolysis in the ER and Golgi  apparatus helps to mature the protein hormone insulin (Figure 17.19). Preproinsulin encoded by mRNA is inserted into the ER lumen. Signal peptidase cleaves the  signal peptide to generate proinsulin, which folds to form the correct disulfide linkages. Proinsulin is transported to the Golgi apparatus where it is packaged into  secretory granules. An internal connecting peptide (C peptide) is removed by proteolysis, and mature insulin is secreted. In familial hyperproinsulinemia, processing is  incomplete (see Clin. Corr. 17.6). This pathway for insulin biosynthesis has advantages over synthesis and binding of two separate polypeptides. First, it ensures production of equal amounts of A and B  chains without coordination of two translational activities. Second, proinsulin folds into a three­dimensional structure in which the cysteine residues are placed for  correct disulfide bond formation. Proinsulin can be reduced and denatured but refolds correctly to form proinsulin. Renaturation of reduced and denatured insulin is  less efficient, and incorrect disulfide linkages are also formed. Correct formation of insulin from separately synthesized chains might have required evolution of a helper  protein or molecular chaperone. Proteolysis Leads to Zymogen Activation Precursor protein cleavage is a common means of enzyme activation. Digestive proteases are classic examples of this phenomenon (see p. 1059). Inactive zymogen  precursors are packaged in storage granules and activated by proteol­

 

  

Page 744

Figure 17.19  Maturation of human proinsulin.  After cleavage at two sites indicated by arrows, the arginine residues 31, 32, and 65  and lysine residue 64 are removed to produce insulin and C­peptide.  Redrawn from G. I. Bell, W. F. Swain, R. Pictet,  B. Cordell, H. M. Goodman, and W. J. Putter, Nature 282:525, 1979.

ysis upon secretion. Thus trypsinogen is cleaved to give an amino­terminal peptide plus trypsin, and chymotrypsinogen is cleaved to form chymotrypsin and two  peptides. Amino Acids Can Be Modified after Incorporation into Proteins Only 20 amino acids are encoded genetically and incorporated during translation. Posttranslational modification of proteins, however, leads to formation of 100 or  more different amino acid derivatives in proteins. Modification may be permanent or highly reversible. The amounts of modified amino acids may be small, but they  often play a major functional role in proteins. Examples are listed in Table 17.10. Protein amino termini are frequently modified. Protein synthesis is initiated using methionine, but in the majority of proteins the amino­terminal residue is not methionine;  proteolysis has occurred. The amino terminus is then sometimes modified by, for example, acetylation or myristoylation. Amino­terminal glutamine residues  spontaneously cyclize; one possible result is the stabilization of the protein. Amino terminal sequences are occasionally lengthened by the addition of an amino acid (see  Section 17.8, Protein Degradation and Turnover). Posttranslational disulfide bond formation is catalyzed by a disulfide isomerase. The cystine­containing protein is conformationally stabilized. Disulfide formation can  prevent unfolding of proteins and their passage across membranes, so it also becomes a means of localization. As seen in the case of insulin, disulfide bonds can  covalently link separate polypeptides and be necessary

 

  

Page 745

for biological function. Cysteine modification also occurs; multiple sulfatase deficiency arises from reduced ability to carry out a posttranslational modification (see Clin.  Corr 17.7). Methylation of lysine  ­amino groups occurs in histone proteins and may modulate their interactions with DNA. A fraction of the H2A histone is also modified  through isopeptide linkage of a small protein, ubiquitin, from its C­terminal glycine to a lysine  ­amino group on the histone. A role in DNA interactions is postulated.  Biotin is also linked to proteins through amide linkages to lysine. Serine and threonine hydroxyl groups are major sites of glycosylation and of reversible phosphorylation by protein kinases and protein phosphatases. A classic  example of phosphorylation of a serine residue is glycogen phosphorylase, which is modified by phosphorylase kinase (see p. 322). Tyrosine kinase activity is a  property of many growth factor receptors; growth factor binding stimulates cell division and the proliferation of specific cell types. Oncogenes, responsible in part for  the proliferation of tumor cells, often have tyrosine kinase activity and show strong homology with normal growth factor receptors Dozens of other examples exist;  together the protein kinases and protein phosphatases control the activity of many proteins that are central to normal and abnormal cellular development.

Figure 17.20  Diphthamide (left) is a posttranslational  modification of a specific residue of  histidine (right) in EF­2.

ADP­ribosylation of EF­2 at a modified histidine residue represents a doubling of posttranslational modifications. First, a specific EP­2 histidine residue is modified to  generate the diphthamide derivative (Figure 17.20) of the functional protein. This modification is probably not absolutely required since yeast mutants that cannot make  diphthamide survive. ADP­ribosylation of the diphtham­ TABLE 17.10 Modified Amino Acids in Proteinsa Amino Acid

Modifications Found

Amino terminus

Formylation, acetylation, aminoacylation, myristoylation, glycosylation

Carboxyl terminus

Methylation, glycosyl­phosphatidylinositol anchor formation, ADP­  ribosylation

Arginine

N­Methylation, ADP­ribosylation

Asparagine

N­Glycosylation, N­methylation, deamidation

Aspartic acid

Methylation, phosphorylation, hydroxylation

Cysteine

Cystine formation, selenocysteine formation, palmitoylation, linkage to  heme, S­glycosylation, prenylation

Glutamic acid

Methylation, g­carboxylation, ADP­ribosylation

Glutamine

Deamidation, cross­linking, pyroglutamate formation

Histidine

Methylation, phosphorylation, diphthamide formation, ADP­ ribosylation

Lysine

N­acetylation, N­methylation, oxidation, hydroxylation, cross­linking,  ubiquitination, allysine formation

Methionine

Sulfoxide formation

Phenylalanine

b­Hydroxylation and glycosylation

Proline

Hydroxylation, glycosylation

Serine

Phosphorylation, glycosylation, acetylation

Threonine

Phosphorylation, glycosylation, methylation

Tryptophan

b­Hydroxylation, dione formation

Tyrosine

Phosphorylation, iodination, adenylation, sulfonylation, hydroxylation

Source: Adapted from R. G. Krishna and F. Wold, Post­translational modification of proteins. In:  A. Meister (Ed.), Advances in Enzymology, Vol. 67. New York: Wiley­Interscience, 1993, pp. 265– 298. a The listing is not comprehensive and some of the modifications are very rare. Note that no 

derivatives of alanine, glycine, isoleucine, and valine have been identified in proteins.

   

  

Page 746

CLINICAL CORRELATION 17.7  Absence of Posttranslational Modification: Multiple Sulfatase Deficiency A variety of biological molecules are sulfated; examples include glycosaminoglycans,  steroids, and glycolipids. Ineffective sulfation of the glycosaminoglycans chondroitin  sulfate and keratan sulfate of cartilage results in major skeletal deformities. Degradation of  sulfated molecules depends on the activity of a group of related sulfatases, most of which  are located in lysosomes. Multiple sulfatase deficiency is a rare lysosomal storage  disorder that combines features of metachromatic leukodystrophy and  mucopolysaccharidosis. Affected individuals develop slowly and from their second year  of life lose the abilities to stand, sit, or speak; physical deformities and neurological  deficiencies develop and death before age 10 is usual. Biochemically, multiple sulfatase  deficiency is characterized by severe lack of all the sulfatases. In contrast, deficiencies in  individual sulfatases are also known, and several distinct diseases are linked to single  enzyme defects. The molecular defect in multiple sulfatase deficiency arises from a deficiency in a  posttranslational modification that is common to all sulfatase enzymes and is necessary for  their enzymatic activity. In each case a cysteine residue of the enzyme is normally  converted to 2­amino­3­oxopropionic acid; the –CH2SH side chain of cysteine becomes  a –CHO (aldehyde) group, which may itself react with amino or hydroxyl groups of the  enzyme, a cofactor, and so on. Fibroblasts from individuals with multiple sulfatase  deficiency catalyze this modification with significantly lowered efficiency, and the  unmodified sulfatases are catalytically inactive. Schmidt, B., Selmer, T., Ingendoh, A, and von Figura, K. A novel amino acid  modification in sulfatases that is deficient in multiple sulfatase deficiency. Cell 82:271– 278, 1995.   ide by diphtheria toxin then inhibits EF­2 activity. Other instances of physiological ADP­ribosylation not mediated by bacterial toxins are reversible. Formation of g­carboxyglutamate from glutamic acid residues occurs in several blood­clotting proteins including prothrombin and factors VII, IX, and X. The g­ carboxyglutamate residues chelate calcium ion, which is required for normal blood clotting (see p. 963). In each case the modification requires vitamin K and can be  blocked by coumarin derivatives, which antagonize vitamin K. As a result, the rate of coagulation is greatly decreased. Collagen Biosynthesis Requires Many Posttranslational Modifications Collagen, the most abundant protein (or family of related proteins) in the human body, is a fibrous protein that provides the structural framework for tissues and organs.  It undergoes a wide variety of posttranslational modifications that directly affect its structure and function, and defects in its modification result in serious diseases.  Collagen is an excellent example of the importance of posttranslational modification. Different species of collagen, designated types I, II, III, IV, and so on (see Table 2.11) are encoded on several chromosomes and expressed in different tissues. Their  amino acid sequences differ, but their overall structural similarity suggests a common evolutionary origin. Each collagen polypeptide, designated an a  chain, has a  repeating sequence Gly­X­Y that is about 1000 residues long. Every third residue is glycine, about one­third of the X positions are occupied by proline and a similar  number of Y positions are 4­hydroxyproline, a posttranslationally modified form of proline. Proline and hydroxyproline residues impart considerable rigidity to the  structure, which exists as a polyproline type II helix (Figure 17.21; see also p. 52). A collagen molecule includes three a  chains intertwined in a collagen triple helix in  which the glycine residues occupy the center of the structure. Procollagen Formation in the Endoplasmic Reticulum and Golgi Apparatus Collagen a  chain synthesis starts in the cytosol, where the amino­terminal signal sequences bind signal recognition particles. Precursor forms, designated, for example,  prepro a 1(I), are extruded into the ER lumen and the signal peptides are cleaved. Hydroxylation of proline and lysine residues occurs cotranslationally, before  assembly of a triple helix. Prolyl 4­hydroxylase requires an ­X­Pro­

 

  

Page 747

Figure 17.21  Collagen structure, illustrating the regularity of the primary  sequence, the left­handed helix, the right­handed triple helix, the 300­nm  molecule, and the organization of molecules in a typical fibril, within which  collagen molecules are cross­linked.

Gly­ sequence (hence 4­hydroxyproline is found only at Y positions in the ­Gly­X­Y­ sequence). Also present in the ER is a prolyl 3­hydroxylase, which modifies a  smaller number of proline residues, and a lysyl hydroxylase, which modifies some of the Y­position lysine residues. These hydroxylases require Fe2+ and ascorbic acid,  the extent of modification depending on the specific a ­chain type. Proline hydroxylation stabilizes collagen and lysine hydroxylation provides sites for interchain cross­ linking and for glycosylation by specific glycosyl transferases of the ER. Asparagine residues are also glycosylated at this point, eventually leading to high mannose­type  oligosaccharides. Triple helix assembly occurs after the polypeptide chains have been completed. Carboxy­terminal globular proprotein domains fold and disulfide bonds are formed.  Interaction of these domains initiates winding of the triple helix from the carboxyl end toward the amino terminus. The completed triple helix, with globular proprotein  domains at each end, moves to the Golgi apparatus where oligosaccharides are processed and matured. Sometimes tyrosine residues are modified by sulfation and  some serines are phosphorylated. The completed procollagen is then released from the cell via secretory vesicles. Collagen Maturation Conversion of procollagen to collagen occurs extracellularly. The amino­terminal and carboxyl­terminal propeptides are cleaved by separate proteases that may also  be type specific. Concurrently, the triple helices assemble into fibrils

 

  

Page 748 TABLE 17.11 Selected Disorders in Collagen Biosynthesis and Structure Disorder

Collagen Defect

Clinical Manifestations

Osteogenesis imperfecta  Decreased synthesis of type I 1

Long bone fractures prior to  puberty

Osteogenesis imperfecta  Point mutations and exon  2 rearrangements in triple helical  regions

Perinatal lethality; malformed and  soft, fragile bones

Ehlers–Danlos IV

Poor secretion, premature  degradation of type III

Translucent skin, easy bruising,  arterial and colon rupture

Ehlers–Danlos VI

Decreased hydroxylysine in types I  Hyperextensive skin, joint  and III hypermobility

Ehlers–Danlos VII

Type I procollagen accumulation:  N­terminal propeptide not cleaved

Cutis laxa (occipital horn  Decreased hydroxylysine due to  syndrome) poor Cu distribution

Joint hypermobility and  dislocation Lax, soft skin; occipital horn  formation in adolescents

  and the collagen is stabilized by extensive cross­linking (see Figure 2.39). Lysyl oxidase converts some lysine or hydroxylysine to the reactive aldehydes, allysine, or  hydroxyallysine. These residues condense with each other or with lysine or hydroxylysine residues in adjacent chains to form Schiff's base and aldol cross­links.  Further and less well­characterized reactions can involve other residues including histidines and can link three a  chains. Defects at many of these steps are known.  Some of the best characterized are listed in Table 17.11 and described in Clin. Corr. 17.8. 17.7—  Regulation of Translation Translation requires considerable energy, and the formation of functioning proteins has significant consequences for the cell. It is logical that the process is carefully  controlled, both globally and for specific proteins. The most efficient and common mechanism of regulation is at the initiation stage. The best understood means of overall regulation of translation involves the reversible phosphorylation of eIF­2a. Under conditions that include nutrient starvation, heat  shock, and viral infection, eIF­2a is phosphorylated by a specific kinase. Phosphorylated eIF­2a ∙ GDP binds tightly to eIF­2b, the guanine nucleotide exchange factor,  which is present in limiting amounts. Since eIF­2b is unavailable for nucleotide exchange, no eIF­2a ∙ GTP is available for initiation. Phosphorylation can be catalyzed  by a heme­regulated inhibitor kinase, which, in the absence of heme, is activated by autophosphorylation. This kinase is present in many cells but is best studied in  reticulocytes that synthesize hemoglobin. Deficiencies in energy supply or any heme precursor activate the kinase. A related double­stranded RNA­dependent  kinase is autophosphorylated and activated in response to binding of ds­RNA that results from many viral infections. Production of this kinase is also induced by  interferon. Initiation factor eIF­4e (a component of the cap binding protein eIF­4f) is activated by phosphorylation in response to, for example, growth factors and is  inactivated by a protein phosphatase following, for example, viral infection. These effects may be greatest in the translation of mRNAs with long, highly structured  leader sequences that need to be unwound to allow identification of a translational start site. Regulation of translation of specific genes also occurs. A clear example is the regulation by iron of synthesis of the iron­binding protein, ferritin. In

 

  

Page 749

CLINICAL CORRELATION 17.8  Defects in Collagen Synthesis Ehlers–Danlos Syndrome, Type IV Ehlers–Danlos syndrome is a group of at least ten disorders that are clinically, genetically,  and biochemically distinguishable, but that share manifestations of structural weaknesses  in connective tissue. The usual problems are fragility and hyperextensibility of skin and  hypermobility of the joints. The weaknesses result from defects in collagen structure. For  example, type IV Ehlers–Danlos syndrome is caused by defects in type III collagen,  which is particularly important in skin, arteries, and hollow organs. Characteristics include  thin, translucent skin through which veins are easily seen, marked bruising, and sometimes  an appearance of aging in the hands and skin. Clinical problems arise from arterial  rupture, intestinal perforation, and rupture of the uterus during pregnancy or labor.  Surgical repair is difficult because of tissue fragility. The basic defects in type IV Ehlers– Danlos appear to be due to changes in the primary structure of type III chains. These  arise from point mutations that result in replacement of glycine residues and thus  disruption of the collagen triple helix, and from exon­skipping, which shortens the  polypeptide and can result in inefficient secretion and decreased thermal stability of the  collagen, and in abnormal formation of type III collagen fibrils. In some cases type III  collagen is accumulated in the rough ER, overmodified, and degraded very slowly. Superti­Furga, A., Gugler, E., Gitzelmann, R., and Steinmann, B. Ehlers–Danlos  syndrome type IV: a multi­exon deletion in one of the two COL 3A1 alleles affecting  structure, stability, and processing of type III procollagen. J. Biol. Chem. 263:6226,  1988. Osteogenesis Imperfecta Osteogenesis imperfecta is a group of at least four clinically, genetically, and  biochemically distinguishable disorders, all characterized by multiple fractures with  resultant bone deformities. Several variants result from mutations producing modified a (I)  chains. In the clearest example a deletion mutation causes absence of 84 amino acids in  the a 1(I) chain. The shortened a 1(I) chains are synthesized, because the mutation leaves  the reading frame in register. The short a 1(I) chains associate with normal a 1(I) and a 2 (I) chains, thereby preventing normal collagen triple helix formation, with resultant  degradation of all the chains, a phenomenon aptly named ''protein suicide." Three­fourths  of all the collagen molecules formed have at least one short (defective) a 1(I) chain, an  amplification of the effect of a heterozygous gene defect. Other forms of osteogenesis  imperfecta result from point mutations that substitute another amino acid for one of the  glycines. Since glycine has to fit into the interior of the collagen triple helix, these  substitutions destabilize that helix. Barsh, G. S., Roush, C. L., Bonadio, J., Byers, P. H., and Gelinas, R. E. Intron mediated  recombination causes an a (I) collagen deletion in a lethal form of osteogenesis  imperfecta. Proc. Natl. Acad. Sci. USA 82:2870, 1985. Scurvy and Hydroxyproline Synthesis Scurvy results from dietary deficiency of ascorbic acid. Most animals can synthesize  ascorbic acid from glucose but humans have lost this enzymatic mechanism. Among other  problems, ascorbic acid deficiency causes decreased hydroxyproline synthesis because  prolyl hydroxylase requires ascorbic acid. The hydroxyproline provides additional  hydrogen­bonding atoms that stabilize the collagen triple helix. Collagen containing  insufficient hydroxyproline loses temperature stability, becoming less stable than normal  collagen at body temperature. The resultant clinical manifestations are distinctive and  understandable: suppression of the orderly growth process of bone in children, poor  wound healing, and increased capillary fragility with resultant hemorrhage, particularly in  the skin. Severe ascorbic acid deficiency leads secondarily to a decreased rate of  procollagen synthesis. Crandon, J. H., Lund, C. C., and Dill, D. B. Experimental human scurvy. N. Engl. J.  Med. 223:353, 1940. Deficiency of Lysyl Hydroxylase In type VI Ehlers–Danlos syndrome lysyl hydroxylase is deficient. As a result type I and  III collagens in skin are synthesized with decreased hydroxylysine content, and  subsequent cross­linking of collagen fibrils is less stable. Some cross­linking between  lysine and allysine occurs but these are not as stable and do not mature as readily as do  hydroxylysine­containing cross­links. In addition, carbohydrates add to the hydroxylysine  residues but the function of this carbohydrate is unknown. The clinical features include  marked hyperextensibility of the skin and joints, poor wound healing, and musculoskeletal  deformities. Some patients with this form of Ehlers–Danlos syndrome have a mutant form  of lysyl hydroxylase with a higher Michaelis constant for ascorbic acid than the normal  enzyme. Accordingly, they respond to high doses of ascorbic acid. Pinnell, S. R., Krane, S. M., Kenzora, J. E., and Glimcher, M. J. A heritable disorder of  connective tissue: hydroxylysine­deficient collagen disease. N. Engl. J. Med. 286:1013,  1972. Ehlers–Danlos Syndrome, Type VII In Ehlers–Danlos syndrome, type VII, skin bruises easily and is hyperextensible, but the  major manifestations are dislocations of major joints, such as hips and knees. Laxity of  ligaments is caused by incomplete removal of the amino­terminal propeptide of the  procollagen chains. One variant of the disease results from deficiency of procollagen N­ protease. A similar deficiency occurs in the autosomal recessive disease called  dermatosparaxis of cattle, sheep, and cats, in which skin fragility is so extreme as to be  lethal. In other variants the proa 1(I) and proa 2(I) chains lack amino acids at the  cleavage site because of skipping of one exon in the genes. This prevents normal cleavage  by procollagen N­protease. Cole, W. G., Chan, W., Chambers, G. W., Walker, I. D., and Bateman, J. F. Deletion  of 24 amino acids from the proa (I) chain of type I procollagen in a patient with the  Ehlers–Danlos syndrome type VII. J. Biol. Chem. 261:5496, 1986. Occipital Horn Syndrome In type IX Ehlers–Danlos syndrome and in Menke's (kinky­hair) syndrome there is  thought to be a deficiency in lysyl oxidase activity. In type IX Ehlers–Danlos syndrome  there are consequent cross­linking defects manifested in lax, soft skin and in the  appearance during adolescence of bony occipital horns. Copper­deficient animals have  deficient cross­linking of elastin and collagen, apparently because of the requirement for  cuprous ion by lysyl oxidase.   (continued)

 

  

Page 750

In Menke's (kinky­hair) syndrome there is a defect in intracellular copper transport that  results in low activity of lysyl oxidase, and in occipital horn syndrome there is also a  defect in intracellular copper distribution. A woman taking high doses of the copper­ chelating drug, d­penicillamine, gave birth to an infant with an acquired Ehlers–Danlos­ like syndrome, which subsequently cleared. Side effects of d­penicillamine therapy  include poor wound healing and hyperextensible skin. Peltonen, L., Kuivaniemi, H., Palotie, A., et al. Alterations of copper and collagen  metabolism in the Menkes syndrome and a new subtype of Ehlers–Danlos syndrome.  Biochemistry 22:6156, 1983. For a detailed overview of collagen disorders see: Byers,  P. H. Disorders of collagen biosynthesis and structure. In: C. R. Scriver, A. L. Beaudet,  W. S. Sly, and D. Valle (Eds.) The Metabolic and Molecular Basis of Inherited  Disease, Vol. III, 7th ed. New York: McGraw­Hill, 1995, pp. 4029–4077.   the absence of iron, a repressor protein binds to the iron­responsive element (IRE), a stem­loop structure in the 5  leader sequence of ferritin mRNA. This mRNA  is sequestered for future use.  ­Aminolevulinic acid synthase, an enzyme of heme biosynthesis, is also regulated by a 5 ­IRE in its mRNA. In contrast, more ferritin  receptor mRNA is needed if iron is limited; it has IREs in its 3 ­nontranslated region. Binding of the repressor protein stabilizes the mRNA and prolongs its useful  lifetime. Many growth­regulated mRNAs, including those for ribosomal proteins, have a polypyrimidine tract in their leader sequence. A polypyrimidine­binding protein  helps regulate their translation. 17.8—  Protein Degradation and Turnover Proteins have finite lifetimes. They are subject to environmental damage such as oxidation, proteolysis, conformational denaturation, or other irreversible modifications.  Equally important, cells need to change their protein complements in order to respond to different needs and situations. Specific proteins have very different lifetimes.  Cells of the eye lens are not replaced and their proteins are not recycled. Hemoglobin in red blood cells lasts the life of these cells, about 120 days. Other proteins  have lifetimes measured in days, hours, or even minutes. Some blood­clotting proteins survive for only a few days, so hemophiliacs are only protected for a short  period by transfusions or injections of required factors. Diabetics require insulin injections regularly since the hormone is metabolized. Metabolic enzymes vary  quantitatively depending on need; for example, urea cycle enzyme levels change in response to diet. Most amino acids produced by protein degradation are recycled  to synthesize new proteins but some degradation products will be excreted. In either case, proteolysis first reduces the proteins in question to peptides and eventually  amino acids. Several proteolytic systems accomplish this end. Intracellular Digestion of Some Proteins Occurs in Lysosomes Digestive proteases such as pepsin, trypsin, chymotrypsin, and elastase hydrolyze dietary protein and have no part in intracellular protein turnover within an organism  (see Chapter 25). Intracellular digestion of proteins from the extracellular environment occurs within lysosomes. Material that is impermeable to the plasma membrane  is imported by endocytosis. In pinocytosis large particles, molecular aggregates, or other molecules present in the extracellular fluid are ingested by engulfment.  Macrophages ingest bacteria and dead cells by this mechanism. Receptor­mediated endocytosis uses cell surface receptors to bind specific molecules. Endocytosis  occurs at pits in the cell surface that are coated internally with the multisubunit protein clathrin. Uptake is by invagination of the plasma membrane and the receptors to  form intracellular coated vesicles. One fate of such vesicles is fusion with a lysosome and degradation of the contents. Some intracellular protein turnover may also  occur within

 

  

Page 751

lysosomes, and under some conditions significant amounts of cellular material can be mobilized via lysosomes. For example, serum starvation of fibroblasts in culture or  starvation of rats leads to the lysosomal degradation of a subpopulation of cellular proteins. Recognition of a specific peptide sequence is involved, indicating that the  lifetime of a protein is ultimately encoded in its sequence. This concept will be more apparent in the next section on ubiquitin­dependent proteolysis. Although lysosomal degradation of cellular proteins occurs, it is not the main route of protein turnover. Calcium­dependent proteases, also called calpains, are present  in most cells. Activators and inhibitors of these enzymes are also present, and calpains are logical candidates for enzymes involved in protein turnover. However, their  role in these processes is not quantitatively established. Golgi and ER proteases degrade peptide fragments that arise during maturation of proteins in the secretory  pathway. They could also be involved in turnover of ER proteins. Apoptosis, programmed cell death, requires several proteases. It is likely that other uncharacterized  mechanisms exist in both the cytosol and in the mitochondrion. Ubiquitin Is a Marker in ATP­Dependent Proteolysis One well­described proteolytic pathway requires ATP hydrolysis and the participation of ubiquitin, a highly conserved protein containing 76 amino acids. One  function of ubiquitin is to mark proteins for degradation. Ubiquitin has other roles; as an example, linkage of ubiquitin to histones H2A and H2B is unrelated to turnover  since the proteins are stable, but modification may affect chromatin structure or transcription. The ubiquitin­dependent proteolytic cycle is shown in Figure 17.22. Ubiquitin is activated by enzyme E1 to form a thioester; ATP is required and a transient AMP– ubiquitin complex is involved. The ubiquitin is then passed to enzyme E2, and finally via one of a group of E3 enzymes it is coupled to a targeted protein. Linkage of  ubiquitin is through isopeptide bonds between  ­amino groups of lysine residues of the protein and the carboxyl­terminal glycine residues of ubiquitin. Several ubiquitin  molecules may be attached to the protein and to each other. ATP­dependent proteases then degrade the tagged protein and free the ubiquitin for further degradation  cycles. Ubiquitin­dependent proteolysis plays a major role in the regulation of cellular events. Cyclins are involved in control of progress through the cell cycle. The ubiquitin­ dependent destruction of a cyclin allows cells to pass from the M phase into G1. Other proteins known to be degraded by ubiquitin­dependent proteolysis include  transcription factors, the p53 tumor suppressor and other oncoproteins, a protein kinase, and immune system and other cell surface receptors. Damaged or mutant proteins are rapidly degraded via the ubiquitin pathway. In cystic fibrosis a mutation that results in deletion of one amino acid greatly alters the  stability of a protein (see Clin. Corr. 17.9), but it is not always clear how native proteins are identified for degradation. Selectivity occurs at the level of the E3 enzyme,  but most specific recognition signals are obscure. One determinant is simply the identity of the amino­terminal amino acid. Otherwise identical b ­galactosidase proteins  with different amino­terminal residues are degraded at widely differing rates. Amino termini may be modified to alter the lifetime of the protein, and some residues serve  as aminoacyl acceptors for a destabilizing residue from an aminoacyl­tRNA. Internal sequences and conformation are also likely to be important; destabilizing PEST  sequences (rich in Pro, Glu, Ser, and Thr) have been identified in several short­lived proteins. The ATP­dependent degradation of ubiquitin­marked proteins occurs in a 26S organelle called the proteasome. Proteasomes are dumbbell­shaped complexes of  about 25 polypeptides; a proteolytically active 20S cylindrical

Figure 17.22  ATP and ubiquitin­dependent  protein degradation.  Ubiquitin is first activated in a  two­step reaction involving formation  of a transient mixed anhydride of  AMP and the carboxy terminus of  ubiquitin (step 1a), followed by  generation of a thioester with enzyme  E1 (step 1b). Enzyme E2 can now  form a thioester with ubiquitin (step 2)  and serve as a donor in E3­catalyzed  transfer of ubiquitin to a targeted  protein (step 3). Several ubiquitin  molecules are usually attached to  different lysine residues of a targeted  protein at this stage. Ubiquitinylated   protein is now degraded by  ATP­dependent proteolysis (step 4);  ubiquitin is not degraded and can  reenter the process at step 1.

 

  

Page 752

Figure 17.23  Model of proteasome.  A 20S central  segment is made up of four  stacked heptameric rings of  two types. The core is hollow  and includes 12–15  different polypeptides;  several proteases with  different specificities are  localized within the rings.  V­shaped segments at each end  cap the cylinder and are  responsible for ATP­dependent  substrate recognition,  unfolding, and translocation  into the proteolytic core.  Upper cap structure is also  in contact with the central  segment but it is shown  displaced from it in order to  illustrate the hollow core of  the cylinder.  Adapted from D.  Rubin and D. Finley,  Curr. Biol. 5:854,  1995; and J.­M. Peters,  Trends Biochem. Sci.  19:377, 1994.

core is capped at each end by V­shaped complexes that bestow ATP dependence (Figure 17.23). It is speculated that the cap structure is involved in recognizing and  unfolding polypeptides and transporting them to the proteolytic core. The complex E. coli proteases Lon and Clp and similar enzymes in other microorganisms (and in  mitochondria) also require ATP hydrolysis for their action, but ubiquitin is absent in prokaryotes and the means of identification of proteins for degradation is still  obscure. It is likely that protein degradation will turn out to be as complex and important a problem as protein biosynthesis. CLINICAL CORRELATION 17.9  Deletion of a Codon, Incorrect Posttranslational Modification, and Premature  Protein Degradation: Cystic Fibrosis Cystic fibrosis (CF) is the most common autosomal recessive disease in Caucasians, with  a frequency of almost 1 per 2000. The CF gene is 230 kb in length and includes 27  exons encoding a protein of 1480 amino acids. The protein known as the cystic fibrosis  transmembrane conductance regulator or CFTR is a member of a family of ATP­ dependent transport proteins and it includes two membrane­spanning domains, two  nucleotide­binding domains that interact with ATP, and one regulatory domain that  includes several phosphorylation sites. CFTR functions as a cyclic AMP­regulated  chloride channel. CF epithelia are characterized by defective electrolyte transport. The  organs most strongly affected include the lungs, pancreas, and liver, and the most life­ threatening effects involve thick mucous secretions that lead to chronic obstructive lung  disease and persistent infections of lungs. In about 70% of affected individuals the problem is traced to a three­nucleotide deletion  that results in deletion of a single amino acid, phenylalanine 508, normally located in ATP­ binding domain 1 on the cytoplasmic side of the plasma membrane. As with several other  CF mutations, the Phe 508 deletion protein is not properly glycosylated or transported to  the cell surface. Instead, it is only partially glycosylated, and it is degraded within the  endoplasmic reticulum. It is postulated that the mutant protein does not fold properly and  is marked for degradation rather than movement to the plasma membrane. Ward, C., Omura, S., and Kopito, R. Degradation of CFTR by the ubiquitin– proteasome pathway. Cell 83:121, 1995.    

  

Page 753

Bibliography Ribosomes and Transfer RNA Filonenko, V. V., and Deutscher, M. P. Evidence for similar structural organization of the multienzyme aminoacyl­tRNA complex in vivo and in vitro. J. Biol. Chem.  269:17375, 1994. Frank, J., Zhu, J., Penczek, P., et al. A model of protein synthesis based on a new cryo­electron microscopy reconstruction of the E. coli ribosome. Nature 376:441,  1995. Freist, W. Mechanisms of aminoacyl­tRNA synthetases: a critical consideration of recent results. Biochemistry 28:6787, 1989. Gesteland, R. F., and Atkins, J. F. (Eds.). The RNA World. Cold Spring Harbor, NY: Cold Spring Harbor Press, 1993. Herold, M., and Nierhaus, K. H. Incorporation of six additional proteins to complete the assembly map of the 50S subunit from Escherichia coli ribosomes. J. Biol.  Chem. 262:8826, 1987. Hou, Y.­M., Francklyn, C., and Schimmel, P. Molecular dissection of a transfer RNA and the basis for its identity. Trends Biochem. Sci. 14:233, 1989. Jakubowski, H., and Goldman, E. Editing errors in selection of amino acids for protein synthesis. Microbiol. Rev. 56:412, 1992. Nierhaus, K. H., Franceschi, F., Subramanian, A. R., Erdmann, V. A., and Wittmann­Liebold, B. (Eds.). The Translational Apparatus: Structure, Function,  Regulation, Evolution. New York: Plenum Press, 1993. Nomura, M. The role of RNA and protein in ribosome function: a review of early reconstitution studies and prospects for future studies. Cold Spring Harbor Symp.  Quant. Biol. 52:653, 1987. Perona, J., Rould, M., and Steitz, T. Structural basis for transfer RNA aminoacylation by Escherichia coli glutaminyl­tRNA synthetase. Biochemstry 32:8758, 1993. Stark, H., Mueller, F., Orleva, E. V., et al. The 70S Escherichia coli ribosome at 23Å resolution: fitting the ribosomal RNA. Structure 3:815, 1995. Verschoor, A., and Frank, J. Three­dimensional structure of the mammalian cytoplasmic ribosome. J. Mol. Biol. 214:737, 1990. Protein Biosynthesis and its Regulation Barrell, B., Anderson, S., Bankier, A. T., et al. Different pattern of codon recognition by mammalian mitochondrial tRNAs. Proc. Natl. Acad. Sci. USA 77:3164,  1980. Berchtold, H., Reshetnikova, L., Reiser, C. O., et al. Crystal structure of active elongation factor Tu reveals major domain rearrangements. Nature 365:126, 1993. Gnirke, A., Geigenmuller, U., Rheinberger, H., and Nierhaus, K. The allosteric three­site model for the ribosomal elongation cycle. J. Biol. Chem. 264:7291, 1989. Merrick, W. C. Mechanism and regulation of eukaryotic protein synthesis. Microbiol. Rev. 56:291, 1992. Merrick, W. C. Eukaryotic protein synthesis: an in vitro analysis. Biochimie 76:822, 1994. Rhoads, R. E. Regulation of eukaryotic protein biosynthesis by initiation factors. J. Biol. Chem. 268:3017, 1993. Samuel, C. E. The eIF2a protein kinases as regulators of protein synthesis in eukaryotes from yeasts to humans. J. Biol. Chem. 268:7603, 1993. Zhouravleva, G., Frolova, L., LeGoff, X., et al. Termination of translation in eukaryotes is governed by two interacting polypeptide chain release factors, eRF1 and  eRF3. EMBO J 14:4065, 1995. Ziff, E. B. Transcription and RNA processing by the DNA tumour viruses. Nature 287:491, 1980. Protein Targeting and Posttranslational Modification Dahms, N. M., Lobel, P., and Kornfeld, S. Mannose 6­phosphate receptors and lysosomal enzyme targeting. J. Biol. Chem. 264:12115, 1989. Danpure, C. J. How can products of a single gene be localized to more than one subcellular compartment? Trends Cell Biol. 5:230, 1995. Kornfeld, R., and Kornfeld, S. Assembly of asparagine­linked oligosaccharides. Annu. Rev. Biochem. 54:631, 1985. Krishna, R. G., and Wold, F. Post­translational modification of proteins. Adv. Enzymol. 67:265, 1993. Kuhn, K. The classical collagens. In: R. Mayne and R. E. Burgeson (Eds.), Structure and Function of Collagen Types. Orlando, FL: Academic Press, 1987, pp.  1–42. Nothwehr, S. F., and Stevens, T. H. Sorting of membrane proteins in the yeast secretory pathway. J. Biol. Chem. 269:10185, 1994. Paulson, J. C. Glycoproteins: what are the sugar chains for? Trends Biochem. Sci. 14:272, 1989. Paulson, J. C., and Colley, K. J. Glycosyltransferases: structure, localization, and control of cell type­specific glycosylation. J. Biol. Chem. 264:17615, 1989. Pfanner, N., Craig, E., and Meijer, M. The protein import machinery of the mitochondrial inner membrane. Trends Biochem. Sci. 19:368, 1994. Pfeffer, S. R., and Rothman, J. E. Biosynthetic protein transport and sorting by the endoplasmic reticulum and Golgi. Annu. Rev. Biochem. 56:829, 1987. Rachubinski, R. A., and Subramani, S. How proteins penetrate peroxisomes. Cell 83:525, 1995. Rogenkamp, R. Targeting signals for protein import into peroxisomes. Cell Biochem. Function 10:193, 1992. Rudd, P. M., et al. The effects of variable glycosylation on the functional activities of ribonuclease, plasminogen and tissue plasminogen activator. Biochim. Biophys.  Acta 1248:1, 1995. von Heijne, G. Signals for protein targeting into and across membranes. In: A. H. Maddy and J. R. Harris (Eds.), Subcellular Biochemistry, Vol. 22: Membrane  Biochemistry. New York: Plenum Press, 1994, pp. 1–19. Walter, P., and Johnson, A. E. Signal sequence recognition and protein targeting to the endoplasmic reticulum membrane. Annu. Rev. Cell Biol. 10:87, 1994. Wolin, S. From the elephant to E. coli: SRP­dependent protein targeting. Cell 77:787, 1994. Protein Turnover and Proteasomes Ciechanover, A. The ubiquitin–proteasome proteolytic pathway. Cell 79:13, 1994. Dice, J. F. Molecular determinants of protein half­lives in eukaryotic cells. FASEB J. 1:349, 1987. Gonda, D. K., Bachmair, A., Wünning, I., et al. Universality and structure of the N­end rule. J. Biol. Chem. 264:16700, 1989. Jentsch, S. The ubiquitin­conjugation system. Annu. Rev. Genet. 26:179, 1992. Kessel, M., Maurizi, M. R., Kim, B., et al. Homology in structural organization between E. coli C1pAP protease and the eukaryotic 26S proteasome. J. Mol. Biol.  250:587, 1995. Löwe, J., Stock, D., Jap, B., et al. Crystal structure of the 20S proteasome from the archaeon T. acidophilum at 3.4A resolution. Science 268:533, 1995. Rechsteiner, M. Natural substrates of the ubiquitin proteolytic pathway. Trends Biochem. Sci. 66:615, 1991. Rock, K., Gramm, C., Rothstein, L., et al. Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC  class II molecules. Cell 78:761, 1994. Rogers, S., Wells, R., and Rechsteiner, M. Amino acid sequences common to rapidly degraded proteins: the PEST hypothesis. Science 234:364, 1986. Wlodawer, A. Proteasome: a complex protease with a new fold and a distinct mechanism. Structure 3:417, 1995.

 

  

Page 754

Questions J. Baggott and C. N. Angstadt 1. Degeneracy of the genetic code denotes the existence of: A. multiple codons for a single amino acid. B. codons consisting of only two bases. C. base triplets that do not code for any amino acid. D. different protein synthesis systems in which a given triplet codes for different amino acids. E. codons that include one or more of the "unusual" bases. 2. Deletion of a single base from a coding sequence of mRNA may result in a polypeptide product with any of the following EXCEPT: A. a sequence of amino acids that differs from the sequence found in the normal polypeptide. B. more amino acids. C. fewer amino acids. D. a single amino acid replaced by another amino acid. 3. During initiation of protein synthesis. A. methionyl­tRNA appears at the A site of the 80S initiation complex. B. eIF­3 and the 40S ribosomal subunit participate in forming a preinitiation complex. C. eIF­2 is phosphorylated by GTP. D. the same methionyl­tRNA is used as is used during elongation. E. a complex consisting of mRNA, the 60S ribosomal subunit, and certain initiation factors is formed. 4. Requirements for eukaryotic protein synthesis include all of the following EXCEPT: A. mRNA. B. ribosomes. C. GTP. D. 20 different amino acids in the form of aminoacyl­tRNAs. E. 

.

5. During the elongation stage of eukaryotic protein synthesis: A. the incoming aminoacyl­tRNA binds to the P site. B. a new peptide bond is synthesized by peptidyl transferase site of the large ribosomal subunit in a GTP­requiring reaction. C. the peptide, still bound to a tRNA molecule, is transloated to a different site on the ribosome. D. streptomycin can cause premature release of the incomplete peptide. E. peptide bond formation occurs by the attack of the carboxyl group of the incoming aminoacyl­tRNA on the amino group of the growing peptide chain. 6. Diphtheria toxin: A. acts catalytically. B. releases incomplete polypeptide chains from the ribosome. C. inhibits translocase. D. prevents release factor from recognizing termination signals. E. attacks the RNA of the large subunit. 7. How many high­energy bonds are expended in the formation of one peptide bond? A. 1 B. 2 C. 3 D. 4 E. 5 8. Formation of mature insulin includes all of the following EXCEPT: A. removal of a signal peptide. B. folding into a three­dimensional structure. C. disulfide bond formation. D. removal of a peptide from an internal region. E. g­carboxylation of glutamate residues. 9. 4­Hydroxylation of specific prolyl residues during collagen synthesis requires all of the following EXCEPT: A. Fe2+. B. a specific amino acid sequence. C. ascorbic acid. D. succinate. E. individual a ­chains, not yet assembled into a triple helix. 10. In the formation of an aminoacyl­tRNA: A. ADP and Pi are products of the reaction. B. aminoacyl adenylate appears in solution as a free intermediate. C. the aminoacyl­tRNA synthetase is believed to recognize and hydrolyze incorrect aminoacyl­tRNAs it may have produced. D. there is a separate aminoacyl­tRNA synthetase for every amino acid appearing in the final, functional protein. E. there is a separate aminoacyl­tRNA synthetase for every tRNA species. 11. During collagen synthesis, events that occur extracellularly include all of the following EXCEPT: A. modification of prolyl residues. B. amino­terminal peptide cleavage. C. carboxyl­terminal peptide cleavage. D. modification of lysyl residues. E. covalent cross­linking. 12. In the functions of ubiquitin all of the following are true EXCEPT: A. ATP is required for activation of ubiquitin. B. ubiquitin­dependent degradation of proteins occurs in the lysosomes. C. linkage of a protein to ubiquitin does not always mark it for degradation. D. the identity of the N­terminal amino acid is one determinant of selection for degradation. E. ATP is required by the protease that degrades the tagged protein.

 

  

Page 755

Match each of the following numbered markers with the appropriate lettered target site. A. export from the cell B. lysosomes C. mitochondria D. nucleus E. peroxisomes 13. Clusters of lysine and arginine amino acid residues. 14. Mannose 6­phosphate. 15. Positively charged amphiphilic a ­helix. 16. Ser­Lys­Leu (SKL). Answers 1. A This is the definition of degeneracy (p. 719). B and E are not known to occur, although sometimes tRNA reads only the first two bases of a triplet (wobble), and  sometimes unusual bases occur in anticodons (p. 719). C denotes the stop (nonsense) codons (p. 719). D is a deviation from universality of the code, as found in  mitochondria (p. 719). 2. D Deletion of a single base causes a frameshift mutation (p. 721). The frameshift would destroy the original stop codon; another one would be generated before or  after the original location. In contrast, replacement of one base by another would cause replacement of one amino acid (missense mutation), unless a stop codon is  thereby generated (p. 721). 3. B A:  4. E 

 is used internally. E: mRNA associates first with the 40S subunit (p. 725).  is involved in initiation of protein synthesis in prokaryote (p. 725).

5. C A: The incoming aminoacyl­tRNA binds to the A site. B: Peptide bond formation requires no energy source other than the aminoacyl­tRNA (pp. 727 and 730).  D: Streptomycin inhibits formation of the prokaryotic 70S initiation complex (analogous to the eukaryotic 80S complex) and causes misreading of the genetic code  when the initiation complex is already formed (p. 734). E: The electron pair of the amino group carries out a nucleophilic attack on the carbonyl carbon. 6. A This toxin catalyzes the formation of an ADP ribosyl derivative of translocase, which irreversibly inactivates the translocase (p. 735). 7. D One ATP is converted to AMP during activation of an amino acid (p. 721), and two GTP are converted to GDP during elongation (pp. 727 and 730). The ATP   AMP counts as two high­energy bonds expended. 8. E See p. 743. g­Carboxylation is of special importance in several blood clotting proteins (p. 746). 9. D See pp. 746–747. 10. C. C. Bonds between a tRNA and an incorrect smaller amino acid may form but are rapidly hydrolyzed (p. 723). A and B: ATP and the amino acid react to form  an enzyme­bound aminoacyl adenylate; PPi is released into the medium (p. 721). D: Some amino acids, such as hydroxyproline and hydroxylysine, arise by co­ or  posttranslational modification (p. 747). E: An aminoacyl­tRNA synthetase may recognize any of several tRNAs specific for a given amino acid (p. 722). 11. A See p. 747. Some modification of lysyl residues also occurs intracellularly (p. 747). 12. E A–D: True (see p. 751). C: Linkage to histones does not result in their degradation. 13. D (see p. 743). 14. B (see p. 740). 15. C (see p. 742). 16. E (see p. 743). This tripeptide must occur at the carboxyl terminal.

 

  

Page 757

Chapter 18—  Recombinant DNA and Biotechnology Gerald Soslau

18.1 Overview

758

18.2 Polymerase Chain Reaction

759

18.3 Restriction Endonuclease and Restriction Maps

760

Restriction Endonucleases Permit Selective Hydrolysis of DNA to  generate Restriction Maps

760

Restriction Maps Permit the Routine Preparation of Defined Segments of  DNA

761

18.4 DNA Sequencing Chemical Cleavage Method: Maxam–Gilbert Procedure

762

Interrupted Enzymatic Cleavage Method: Sanger Procedure

763

18.5 Recombinant DNA and Cloning

765

Recombinant DNA Vectors Can Be Produced in Significant Quantities  by Cloning

766

DNA Can Be Inserted into Vector DNA in a Specific Direction:  Directional Cloning

766

Bacteria Can Be Transformed with Recombinant DNA

767

It Is Necessary to Be Able to Select Transformed Bacteria

768

Recombinant DNA Molecules in a Gene Library

768

PCR May Circumvent the Need to Clone DNA

768

18.6 Selection of Specific Cloned DNA in Libraries

770

Loss of Antibiotic Resistance Is Used to Select Transformed Bacteria

770

a ­Complementation for Selecting Bacteria Carrying Recombinant  Plasmids

772

773

Southern Blot Technique Is Useful for Identifying DNA Fragments

774

Single­Strand Conformation Polymorphism

775 777

mRNA Is Used As a Template for DNA Synthesis Using Reverse  Transcriptase

777

Desired mRNA in a Sample Can Be Enriched by Separation  Techniques

777

Complementary DNA Synthesis

777

Total Cellular RNA May Be Used As a Template for DNA Synthesis  Using RT­PCR

778

18.9 Bacteriophage, Cosmid, and Yeast Cloning Vectors

778

Bacteriophage As Cloning Vectors

778

Screening Bacteriophage Libraries

779

Cloning DNA Fragments into Cosmid and Yeast Artificial Chromosome  Vectors

780

18.10 Techniques to Further Analyze Long Stretches of DNA

781

Subcloning Permits Definition of Large Segments of DNA

781

Chromosome Walking Is a Technique to Define Gene Arrangement in  Long Stretches of DNA

781

18.11 Expression Vectors and Fusion Proteins Foreign Genes Can Be Expressed in Bacteria Allowing Synthesis of  Their Encoded Proteins 18.12 Expression Vectors in Eukaryotic Cells

 

773

Nucleic Acids Can Serve As Probes for Specific DNA or RNA  Sequences

18.8 Complementary DNA and Complementary DNA Libraries

  

765

DNA from Different Sources Can Be Ligated to Form a New DNA  Species: Recombinant DNA

18.7 Techniques for Detection and Identification of Nucleic Acids

 

762

783 783

784

DNA Elements Required for Expression of Vectors in Mammalian Cells

785

Transfected Eukaryotic Cells Can Be Selected by Utilizing Mutant Cells  That Require Specific Nutrients

785

Foreign Genes Can Be Expressed in Eukaryotic Cells by Utilizing Virus  Transformed Cells

786

Page 758

18.13 Site­Directed Mutagenesis

786

Role of Flanking Regions in DNA Can Be Evaluated by Deletion and  Insertion Mutations

786

Site­Directed Mutagenesis of a Single Nucleotide

787

18.14 Applications of Recombinant DNA Technologies

790

Antisense Nucleic Acids Hold Promise As Research Tools and in  Therapy

790

Normal Genes Can Be Introduced into Cells with a Defective Gene in  Gene Therapy

793

Transgenic Animals

794

Recombinant DNA in Agriculture Will Have Significant Commercial  Impact

794

18.15 Concluding Remarks

795

Bibliography

795

Questions and Answers

796

Clinical Correlations

 

18.1 Polymerase Chain Reaction and Screening for Human  Immunodeficiency Virus

760

18.2 Restriction Mapping and Evolution

762

18.3 Direct Sequencing of DNA for Diagnosis of Genetic Disorders

766

18.4 Multiplex PCR Analysis of HGPRTase Gene Defects in Lesch– Nyhan Syndrome

770

18.5 Restriction Fragment Length Polymorphisms Determine the Clonal  Origin of Tumors

776

18.6 Site­Directed Mutagenesis of HSV I gD

789

18.7 Normal Genes Can Be Introduced into Cells with Defective Genes  in Gene Therapy

793

18.8 Transgenic Animal Models

795

  18.1—  Overview By 1970, the stage was set for modern molecular biology based on studies of numerous scientists in the previous 30 years, during which ignorance of what biochemical  entity orchestrated the replication of life forms with such fidelity gave way to a state where sequencing and manipulating the expression of genes would be feasible. The  relentless march toward a full understanding of gene regulation under normal and pathological conditions has moved with increasing rapidity since the 1970s.  Deoxyribonucleic acid, composed of only four different nucleotides covalently linked by a sugar–phosphate backbone, is deceptively complex. Complexity is  conferred on the DNA molecule by the nonrandom sequence of its bases, multiple conformations that exist in equilibrium in the biological environment, and specific  proteins that recognize and associate with selected regions. By the 1970s biochemical knowledge of the cellular processes and their macromolecular components had  established several facts required for the surge forward. It was clear that gene expression was highly regulated. Enzymes involved in DNA replication and RNA  transcription had been purified and their function in the synthetic process defined. The genetic code had been broken. Genetic maps of prokaryotic chromosomes had  been established based on gene linkage studies with thousands of different mutants. Finally, RNA species could be purified, enzymatically hydrolyzed into discrete  pieces, and laboriously sequenced. It was evident that further progress in the understanding of gene regulation would require techniques to selectively cut DNA into  homogeneous pieces. Even small, highly purified viral DNA genomes were too complex to decipher. The thought of tackling the human genome with more than 3 ×  109 base pairs was all the more onerous. Identification, purification, and characterization of restriction endonucleases that faithfully hydrolyze DNA molecules at specific sequences permitted the development  of recombinant DNA methodologies. Development of DNA sequencing opened the previously tightly locked molecular biology gates to the secrets held within the  organization of diverse biological genomes. Genes could finally be sequenced, but perhaps more importantly so could the flanking regions that regulate their expression.  Sequencing regulatory regions of numerous genes defined consensus sequences such as those found in promoters, enhancers, and many binding sites for regulatory  proteins (see Chapter 19). Each gene contains an upstream promoter where a DNA­dependent RNA polymerase binds prior to initiation of transcription. While some  DNA regulatory sites lie just upstream of the transcription initiation site, other regulatory regions are hundreds to thousands of bases removed and still others are  downstream.

 

  

Page 759

This chapter presents many of the sophisticated techniques, developed in the past 25 years, that allow for the dissection of complex genomes into defined fragments  with the complete analysis of the nucleotide sequence and function of these DNA regions. The modification and manipulation of genes, that is, genetic engineering,  facilitates the introduction and expression of genes in both prokaryotic and eukaryotic cells. Many methodological approaches in genetic engineering have been greatly  simplified by employment of a method that rapidly amplifies selected regions of DNA—the polymerase chain reaction (PCR). Proteins for experimental and clinical  uses are readily produced by these procedures and it is anticipated that in the not too distant future these methods will allow for the rapid increase of treatment  modalities of genetic diseases with gene replacement therapy. Current and potential uses of recombinant DNA technologies are also described. The significance to our  society of advancements in the understanding of genetic macromolecules and their manipulation cannot be overstated. 18.2—  Polymerase Chain Reaction The rapid production of large quantities of a specific DNA sequence took a leap forward with the development of the polymerase chain reaction (PCR). The PCR  requires two nucleotide oligomers that hybridize to the complementary DNA strands in a region of interest. The oligomers serve as primers for a DNA polymerase that  extends each strand. Repeated cycling of the PCR yields large amounts of each DNA molecule of interest in a matter of hours as opposed to days and weeks  associated with cloning techniques. The PCR amplification of a specific DNA sequence can be accomplished with a purified DNA sample or a small region within a complex mixture of DNA. The  principles of the reaction are shown in Figure 18.1. The nucleotide sequence of the DNA to be amplified must be known or it must be cloned in a vector (see p. 778)  where the sequence of the flanking DNA has been established. The product of PCR is a double­stranded DNA molecule and the reaction is completed in each cycle  when all of the template molecules have been copied. In order to initiate a new round of replication the sample is heated to melt the double­stranded DNA and, in the  presence of excess oligonucleotide primers, cooled to permit hybridization of the single­stranded template with free oligomers. A new cycle of DNA replication will  initiate in the presence of DNA polymerase and all four dNTPs. Heating to about 95°C as required for melting DNA inactivates most DNA polymerases, but a heat  stable polymerase,

Figure 18.1  Polymerase chain reaction (PCR).  A DNA fragment of unknown sequence is inserted  into a vector of known sequence by normal  recombinant methodologies. The recombinant DNA  of interest does not need to be purified from  contaminating DNA species. The DNA is heated to  90°C to dissociate the double strands  and cooled in the presence of excess amounts of  two different complementary oligomers that  hybridize to the known vector DNA sequences  flanking the foreign DNA insert. Only  recombinant single­stranded DNA species can  serve as templates for DNA replication,  yielding double­stranded DNA fragments of  foreign DNA bounded by the oligomer DNA  sequences. The heating–replication cycle  is repeated many times to rapidly produce  greatly amplified amounts of the original  foreign DNA. The DNA fragment of interest can  be purified from the polymerase chain reaction  mixture by cleaving it with the original  restriction endonuclease (RE), electrophoresing  the DNA mixture through an agarose gel, and  eluting the band of interest from the gel.

 

  

Page 760

CLINICAL CORRELATION 18.1  Polymerase Chain Reaction and Screening for Human Immunodeficiency Virus Use of the polymerase chain reaction (PCR) to amplify minute quantities of DNA has  revolutionized the ability to detect and analyze DNA species. With PCR it is possible to  synthesize sufficient DNA for analysis. Conventional methods for detection and  identification of the human immunodeficiency virus (HIV), such as Southern blot–DNA  hybridization and antigen analysis, are labor intensive, expensive, and have low sensitivity.  An infected individual, with no sign of AIDS (acquired immunodeficiency syndrome), may  test false negative for HIV by these procedures. Early detection of HIV infections in these  individuals is crucial to initiate treatment and/or monitor the progression of their disease.  In addition, a sensitive method is required to be certain that contributed blood from  donors does not contain HIV. PCR amplification of potential HIV DNA sequences within  DNA isolated from an individual's white blood cells permits the identification of viral  infections prior to the presence of antibodies, the so­called seronegative state. Current  methodologies are too costly to apply this testing to large­scale screening of donor blood  samples. PCR can also be used to increase the sensitivity to detect and characterize  DNA sequences of any other human infectious pathogen. Kwok, S., and Sninsky, J. J. Application of PCR to the detection of human infectious  diseases. In: H. A. Erlich (Ed.), PCR Technology. New York: Stockton Press, 1989, p.  235.   termed Taq DNA polymerase isolated from Thermus aquaticus, is now employed, obviating the need for fresh polymerase after each cycle. This has permitted the  automation of PCR with each DNA molecule capable of being amplified one million­fold. When the DNA to be amplified is present in very low concentrations relative to the total DNA in the sample, it is possible to amplify the DNA region of interest along  with other spurious sequences. In this situation the specificity of the amplification reaction can be enhanced by nested PCR. After conducting the first PCR with one  set of primers for 10–20 cycles, a small aliquot is removed for a second PCR. However, the second PCR is conducted with a new set of primers that are  complementary to the template DNA just downstream of the first set of primers, or ''nested" between the original set of primers. This process amplifies the DNA  region of interest twice with a greatly enhanced specificity. PCR has many applications including gene diagnosis, forensic investigations where only a drop of dried blood or a single hair is available, and evolutionary studies with  preserved biological material. Use of PCR for screening for human immunodeficiency virus is presented in Clin. Corr. 18.1. 18.3—  Restriction Endonuclease and Restriction Maps Restriction Endonucleases Permit Selective Hydrolysis of DNA to Generate Restriction Maps Nature possesses a diverse set of tools, the restriction endonucleases, capable of selectively dissecting DNA molecules of all sizes and origin into smaller fragments.  These enzymes confer some protection on bacteria against invading viruses, that is, bacteriophage. The bacterial DNA sequences normally recognized by the  restriction endonuclease may be protected from cleavage in the host cell by methylation of bases within the enzyme recognized palindrome while the unmethylated viral  DNA is recognized as foreign and is hydrolyzed. Numerous Type II restriction endonucleases, with differing sequence specificities, have been identified and purified;  many are now commercially available (see p. 609 for discussion of restriction endonuclease activities).

 

  

Page 761

Restriction endonuclease permits construction of a new type of genetic map, the restriction map, in which the site of enzyme cleavage within the DNA is identified.  Purified DNA species that contain restriction endonuclease sequences are subjected to restriction endonuclease cleavage. By regulating the time of exposure of the  purified DNA molecules to restriction endonuclease cleavage, a population of DNA fragments that are partially to fully hydrolyzed can be generated. Separation of  these enzyme­generated fragments by agarose gel electrophoresis allows for the construction of restriction maps; an example of this procedure with circular DNA is  presented in Figure 18.2. Analysis of a DNA completely hydrolyzed by a restriction endonuclease establishes how many sites the restriction endonuclease recognizes  within the molecule and what size fragments are generated. The size distribution of composite fragments generated by the partial enzymatic cleavage of the DNA  molecules demonstrates linkage of all potential fragments. The sequential use of different restriction endonucleases has permitted a detailed restriction map of numerous  circular DNA species including bacterial plasmids, viruses, and mitochondrial DNA. The method is also equally amenable to linear DNA fragments that have been  purified to homogeneity. Restriction Maps Permit the Routine Preparation of Defined Segments of DNA Restriction maps may yield little information as to the genes or regulatory elements within the various DNA fragments. They have been used to demonstrate sequence  diversity of organelle DNA, such as mitochondrial DNA, within species (see Clin. Corr. 18.2). Restriction maps can also be used to detect deletion mutations where a  defined DNA fragment from the parental strain

Figure 18.2  Restriction endonuclease mapping of DNA.  Purified DNA is subjected to restriction endonuclease digestion for varying times,  which generates partially to fully cleaved DNA fragments. The DNA fragments are  separated by agarose gel electrophoresis and stained with ethidium bromide. The  DNA bands are visualized with a UV light source and photographed. The size  of the DNA fragments is determined by the relative migration through the gel as  compared to co­electrophoresed DNA standards. The relative arrangement of  each fragment within the DNA molecule can be deduced from the size of the  incompletely hydrolyzed fragments.

 

  

Page 762

CLINICAL CORRELATION 18.2  Restriction Mapping and Evolution In the past, evolutionary studies of species have depended solely on anatomical changes  observed in fossil records and on carbon dating. More recently, these studies are being  supported by molecular analysis of the sequence and size of selected genes or whole  DNA molecules. Evolutionary alterations of a selected DNA molecule from different  species can be rapidly assessed by restriction endonuclease mapping. Generation of  restriction endonuclease maps requires a pure preparation of DNA. Mammalian  mitochondria contain a covalently closed circular DNA molecule of approximately  16,000 base pairs that can rapidly be purified from cells. The mitochondrial DNA  (mtDNA) can be employed directly for the study of evolutionary changes in DNA without  the need of cloning a specific gene. Mitochondrial DNA has been purified from the Guinea baboon, rhesus macaque, guenon,  and human and cleaved with 11 different restriction endonucleases. Restriction maps  were constructed for each species. The maps were all aligned relative to the direction and  nucleotide site where DNA replication is initiated. A comparison of shared and altered  restriction endonuclease sites allowed for calculation of the degree of divergence in  nucleotide sequence between species. It was found that the rate of base substitution  (calculated from the degree of divergence versus the time of divergence) has been about  tenfold greater than changes in the nuclear genome. This high rate of mutation of the  readily purified mtDNA molecule makes it an excellent model to study evolutionary  relationships between species. Brown, W. M., George, M. Jr., and Wilson, A. C. Rapid evolution of animal  mitochondrial DNA. Proc. Natl. Acad. Sci. USA 76:1967, 1979.   migrates as a smaller fragment in the mutated strain. Most importantly, the enzymatic microscissors used to generate restriction maps cut DNA into defined  homogeneous fragments that can be readily purified. These maps are crucial for cloning and for sequencing genes and their flanking DNA regions. 18.4—  DNA Sequencing To determine the complexities of regulation of gene expression and to seek the basis for genetic diseases, techniques were necessary to determine the exact sequence  of bases in DNA. In the late 1970s two different sequencing techniques were developed, one by A. Maxam and W. Gilbert, the chemical cleavage approach, and the  other by F. Sanger, the enzymatic approach. Both procedures may employ the labeling of a terminal nucleotide, followed by the separation and detection of generated  oligonucleotides. Chemical Cleavage Method:  Maxam–Gilbert Procedure Requirements for this procedure include (1) labeling of the terminal nucleotide, (2) selective hydrolysis of the phosphodiester bond for each nucleotide separately to  produce fragments with 1, 2, 3, or more bases, (3) quantitative separation of the hydrolyzed fragments, and (4) a qualitative determination of the label added in Step 1.  The following describes one approach of the Maxam–Gilbert procedure. The overall approach is presented in Figure 18.3. One end of each strand of DNA can be selectively radiolabeled with 32P. This is accomplished when a purified double helix DNA fragment contains restriction  endonuclease sites on either side of the region to be sequenced. Hydrolysis of the DNA with two different restriction endonucleases then results in different staggered  ends, each with a different base in the first position of the single­stranded region. Labeling of the 3  end of each strand is accomplished with addition of the next  nucleotide as directed by the corresponding base sequence on the complementary DNA strand. A fragment of E. coli DNA polymerase I, termed the Klenow  fragment, will catalyze this reaction. The Klenow fragment, produced by partial proteolysis of the polymerase holoenzyme, lacks 5    3  exonuclease activity but  retains the 3    5  exonuclease and polymerase activity. Each strand can therefore be selectively labeled in separate experiments. The complementary unlabeled  strand will not be detectable when analyzing the sequence of the labeled strand. The hydrolysis of the labeled DNA into different lengths is accomplished by first selectively destroying one or two bases of the four nucleotides. The procedure used  exposes the phosphodiester bond connecting adjoining bases and permits selective cleavage of the DNA at the altered base. In separate chemical treatments, samples  of labeled DNA are treated to alter purines and pyrimidines without disrupting the sugar–phosphate backbone; a method is not currently available to specifically alter  adenine or thymine. Conditions for base modification are selected such that only one or a few bases are destroyed randomly within any one molecule. The four  separate DNA samples are then reacted with piperidine, which chemically breaks the sugar–phosphate back­bone at sites where a base has been destroyed,  generating fragments of different sizes. Since labeling is specific at the end while the chemical alteration of the base is random and not total, some of the fragments will  be end labeled. For example, wherever a cytosine residue had been randomly destroyed in the appropriate reaction tube a break will be introduced into the DNA  fragment. The series of chemically generated, end­labeled DNA fragments from each of the four tubes are electrophoresed through a polyacrylamide gel. Bases  destroyed near the end­labeled nucleotide will generate fragments that migrate faster through the gel, as low molecular weight species, while fragments derived

 

  

Page 763

Figure 18.3  Maxam–Gilbert chemical method to sequence DNA.  A double­stranded DNA fragment to be sequenced is obtained by restriction  endonuclease cleavage and purified. Both strands are sequenced by selectively  labeling the ends of each DNA strand. One strand of DNA is end­labeled with  [32P]dGTP in reaction tube 1 while the other is end­labeled with  [32P]dCTP in reaction tube 2. The end­labeled DNA is then  subdivided into four fractions where the different bases are chemically  destroyed at random positions within the single­stranded DNA molecule. The  less selective chemical destruction of adenine simultaneously destroys G and  the destruction of thymine destroys the C bases. The single­stranded DNA is  cleaved at the sites of the destroyed bases. This generates end­labeled  fragments of all possible lengths corresponding to the distance from the end  to the sites of base destruction. Labeled DNA fragments are separated  according to size by electrophoresis. The DNA sequence can then be determined  from the electrophoretic patterns detected on autoradiograms.

from bases destroyed more distant from the end will migrate through the gel more slowly as higher molecular weight molecules. The gel is then exposed to X­ray film,  which detects the 32P, and the radioactively labeled bands within the gel can be visualized. The sequence can be read manually or by automated methods directly from  the X­ray autoradiograph beginning at the bottom (smaller fragments) and proceeding toward the top of the film (larger fragments). Sequencing the complementary  strand checks the correctness of the sequence. Interrupted Enzymatic Cleavage Method:  Sanger Procedure

Figure 18.4  Structure of deoxynucleoside triphosphate  and dideoxynucleoside triphosphate.  The 3 ­OH group is lacking on the ribose  component of the dideoxynucleoside  triphosphate (ddNTP). This molecule can  be incorporated into a growing DNA  molecule through a phosphodiester bond  with its 5 ­phosphates. Once incorporated,  the ddNTP blocks further synthesis of the  DNA molecule since it lacks the 3 ­OH  acceptor group for an incoming nucleotide.

The Sanger procedure of DNA sequencing is based on the random termination of a DNA chain during enzymatic synthesis. The technique is possible because the  dideoxynucleotide analog of each of the four normal nucleotides (Figure 18.4) can be incorporated into a growing DNA chain by DNA polymerase. The ribose of the  dideoxynucleotide triphosphate (ddNTP) has the OH group at both the 2  and 3  positions replaced with a proton, whereas dNTP has only a single OH group  replaced by a proton at the 2  position. Thus the ddNTP incorporated into the growing chain is unable to form a phosphodiester bond with another dNTP because the  3  position of the ribose does not contain an OH group. The growing DNA molecule can be terminated at random points, from the first nucleotide incorporated to the  last, by including in the reaction system both the normal nucleotide and the ddNTP (e.g., dATP and ddATP) at concentrations such that the two nucleotides compete  for incorporation. Identification of DNA fragments requires labeling of the 5  end of the DNA molecules or the incorporation of labeled nucleotides during synthesis. The technique,  outlined in Figure 18.5, is best conducted with pure single­stranded DNA; however, denatured double­stranded DNA can be used. Today, the DNA

 

  

Page 764

Figure 18.5  Sanger dideoxynucleoside triphosphate method  to sequence DNA.  The DNA region of interest is inserted into  bacteriophage DNA molecule. Replicating  bacteriophage produces a single­stranded  recombinant DNA molecule that is readily  purified. The known sequence of the  bacteriophage DNA downstream of the  DNA insert serves as a hybridization site for an  end­labeled oligomer with a complementary  sequence, a universal primer. Extension of this  primer is catalyzed with a DNA polymerase in  the presence of all four deoxynucleoside  triphosphates plus one dideoxynucleoside  triphosphate, for example, ddGTP. Synthesis  stops whenever a dideoxynucleoside triphosphate  is incorporated into the growing molecule. Note  that the dideoxynucleotide competes for  incorporation with the deoxynucleotide. This  generates end­labeled DNA fragments of all  possible lengths that are separated by  electrophoresis. The DNA sequence can then be  determined from the electrophoretic patterns.

to be sequenced is frequently isolated from a recombinant single­stranded bacteriophage (see p. 778) where a region flanking the DNA of interest contains a sequence  that is complementary to a universal primer. The primer can be labeled with either 32P or 35S nucleotide. Primer extension is accomplished with one of several different  available DNA polymerases; one with great versatility is a genetically engineered form of the bacteriophage T7 DNA polymerase. The reaction mixture, composed of  the target DNA, labeled primer, and all four deoxynucleoside triphosphates, is divided into four tubes, each containing a different dideoxynucleoside triphosphate. The  ddNTPs are randomly incorporated during the enzymatic synthesis of DNA and cause termination of the chain.

 

  

Page 765

Since the ddNTP is present in the reaction tube at a low level, relative to the corresponding dNTP, termination of DNA synthesis occurs randomly at all possible  complementary sites to the DNA template. This yields DNA molecules of varying sizes, labeled at the 5  end, that can be separated by polyacrylamide gel  electrophoresis. The labeled species are detected by X­ray autoradiography and the sequence is read. Initially, this method required a single­stranded DNA template, production of a specific complementary oligonucleotide primer, and the need for a relatively pure  preparation of the Klenow fragment of E. coli DNA polymerase I. These difficulties have been overcome and modifications have simplified the approach. The Sanger  method can rapidly sequence as many as 400 bases while the Maxam–Gilbert method is limited to about 250 bases. The PCR and Sanger methods can be combined for direct sequencing of small DNA regions of interest. The double­stranded PCR product is employed directly as  template. Conditions are set such that one strand of melted DNA (template) anneals with the primer in preference to reannealing of template with its complementary  strand, which would reform the original double­stranded DNA. Sequencing then follows the standard dideoxy chain termination reaction (typically with Sequenase in  lieu of the Klenow polymerase) with synthesis of random­length chains occurring as extensions of the PCR primer. This method has been employed successfully for the  diagnosis of genetic disorders (see Clin. Corr. 18.3). 18.5—  Recombinant DNA and Cloning DNA from Different Sources Can Be Ligated to Form a New DNA Species:  Recombinant DNA

Figure 18.6  Formation of recombinant DNA from restriction  endonuclease ­generated fragments containing  cohesive ends.  Many restriction endonucleases  hydrolyze DNA in a staggered fashion, yielding  fragments with single­stranded regions at their  5  and 3  ends. DNA fragments generated from  different molecules with the same restriction  endonuclease have complementary  single­stranded ends that can be annealed and  covalently linked together with a DNA ligase.  All different combinations are possible in a  mixture. When two DNA fragments of different  origin combine it results in a recombinant  DNA molecule.

The ability to selectively hydrolyze a population of DNA molecules with a battery of restriction endonucleases led to the development of a technique for joining two  different DNA molecules termed recombinant DNA. This technique combined with the various techniques for replication, separation, and identification permits the  production of large quantities of purified DNA fragments. The combined techniques, referred to as recombinant DNA technologies, allow the removal of a piece of  DNA out of a larger complex molecule, such as the genome of a virus or human, and amplification of the DNA fragment. Recombinant DNAs have been prepared  with DNA fragments from bacteria combined with fragments from humans, viruses with viruses, and so on. The ability to join two different pieces of DNA together at  specific sites within the molecules is achieved with two enzymes, a restriction endonuclease and a DNA ligase. There are a number of different restriction  endonucleases, varying in their nucleotide sequence specificity, that can be used (Section 18.3). Some hydrolyze the two strands of DNA in a staggered fashion,  producing "sticky or cohesive" ends (Figure 18.6), while others cut both strands symmetrically, producing a blunt end. A specific restriction enzyme cuts DNA at  exactly the same nucleotide sequence site regardless of the source of the DNA (bacteria, plant, mammal, etc.). A DNA molecule may have one, several, hundreds,  thousands, or no recognition sites for a particular restriction endonuclease. The staggered cut results in a fragmented DNA molecule with ends that are single stranded.  When different DNA fragments generated by the same restriction endonuclease are mixed, their single­stranded ends can hybridize, that is, anneal together. In the  presence of DNA ligase the two fragments are connected covalently, producing a recombinant DNA molecule. The DNA fragments produced from restriction endonuclease that form blunt ends can also be ligated but with much lower efficiency. The efficiency can be increased  by enzymatically adding a poly(dA) tail to one species of DNA and a poly(dT) tail to the ends of the second species of DNA. The DNA fragments

 

  

Page 766

CLINICAL CORRELATION 18.3  Direct Sequencing of DNA for Diagnosis of Genetic Disorders The X­linked recessive hemorrhagic disorder hemophilia B is caused by a coagulation  factor IX deficiency. The factor IX gene has been cloned and sequenced and contains 8  exons spanning 34 kb that encode a glyco­protein secreted by the liver. Over 300  mutations of the gene have been discovered of which about 85% are single base  substitutions and the rest are complete or partial gene deletions. Several methods have  been employed to identify carriers of a defective gene copy and for prenatal diagnoses.  Unfortunately, these methods were costly, time consuming, and all too often inaccurate.  Direct sequencing of PCR amplified genomic DNA has been employed to circumvent  these diagnostic shortcomings. Between 0.1 and 1  g of genomic DNA can readily be  isolated from patient blood samples and each factor IX exon can be PCR amplified with  appropriate primers. The amplified DNA can then be used for direct sequencing to  determine if a mutation in the gene exists that would be diagnostic of one of the forms of  hemophilia B. For example, a patient with a moderate hemophilia B (London 6) had an A   G transition at position 10442 that led to a substitution of Asp 64 by Gly. Green, P. M., Bentley, D. R., Mibashan, R. S., Nilsson, I. M., and Gianelli, F. Molecular  pathology of hemophilia b. EMBO J. 8:1067, 1989.   with complementary tails can be annealed and ligated in the same manner as fragments with restriction enzyme­generated cohesive ends. Recombinant DNA Vectors Can Be Produced in Significant Quantities by Cloning Synthesis of a recombinant DNA opens the way for production of significant quantities of interesting DNA fragments. By incorporating a recombinant DNA into a  cellular system that allows replication of recombinant DNA, amplification of DNA of interest can be achieved. A carrier DNA, termed a cloning vector, is employed.  Bacterial plasmids are ideally suited as recombinant DNA vectors. Many bacteria contain a single circular chromosome of approximately 4 million base pairs and  minicircular DNA molecules called plasmids. Plasmids are usually composed of only a few thousand base pairs and are rarely associated with the large chromosomal  molecule. Genes within the plasmid have various functions; one of the most useful is the ability to confer antibiotic resistance to the bacterium, an attribute useful in  selecting specific colonies of the bacteria. Plasmids replicate independently of replication of the main bacterial chromosome. One type of plasmid, the relaxed­control  plasmids, may be present in tens to hundreds of copies per bacterium, and replication is dependent solely on host enzymes that have long half­lives. Therefore  replication of "relaxed" plasmids can occur in the presence of a protein synthesis inhibitor. Bacteria can accumulate several thousand plasmid copies per cell under  these conditions. Other plasmid types are subjected to stringent control and their replication is dependent on the continued synthesis of plasmid­encoded proteins.  These plasmids replicate at about the same rate as the large bacterial chromosome, and only a low number of copies occur per cell. The former plasmid type is  routinely used for recombinant DNA studies. The first practical recombinant DNA molecule that could be cloned involved as a vector the E. coli plasmid pSC101, which contains a single EcoRI restriction  endonuclease site and a gene that encodes for a protein that confers antibiotic resistance to the bacteria. This plasmid contains an origin of replication and associated  DNA regulatory sequences that are referred to as a replicon. This vector, however, suffers from a number of limiting factors. The single restriction endonuclease site  limits the DNA fragments that can be cloned and the one antibiotic­resistance selectable marker reduces the convenience in selection; in addition, it replicates poorly. Plasmid vectors with broad versatilities have been constructed using recombinant DNA technology. The desirable features of a plasmid vector include a relatively low  molecular weight (3–5 kb) to accommodate larger fragments; several different restriction endonuclease sites useful in cloning a variety of restriction enzyme­generated  fragments; multiple selectable markers to aid in selecting bacteria with recombinant DNA molecules; and a high rate of replication. The first plasmid constructed  (Figure 18.7) to satisfy these requirements was pBR322 and this plasmid has been used for the subsequent generation of newer vectors in use today. Most currently  employed vectors contain an inserted sequence of DNA termed polylinker, restriction site bank, or polycloning site, which contains numerous restriction  endonuclease sites unique to the plasmid. DNA Can Be Inserted into Vector DNA in a Specific Direction:  Directional Cloning Directional cloning reduces the number of variable "recombinants" and enhances the probability of selection of the desired recombinant. Insertion of foreign DNA,  with a defined polarity, into a plasmid vector in the absence of the plasmid resealing itself can be accomplished by employing two restriction

 

  

Page 767

Figure 18.7  The pBR322 plasmid constructed in the laboratory to contain  features that facilitate cloning foreign DNA fragments.  By convention, the numbering of the nucleotides begins with the first T in the  unique EcoRI recognition sequence (GAATTC) and the positions on the map  refer to the 5  base of the various restriction endonuclease­recognition  sequences. Only a few of the unique restriction sites within the antibiotic  resistance genes and none of the numerous sites where an enzyme cuts more  than once within the plasmid are shown.

endonucleases to cleave the plasmids (Figure 18.8); vectors with polylinkers are ideally suited for this purpose. The use of two enzymes yields DNA fragments and  linearized plasmids with different "sticky" ends. Under these conditions the plasmid is unable to reanneal with itself. In addition, the foreign DNA can be inserted into  the vector in only one orientation. This is extremely important when one clones a potentially functional gene downstream from the promoter­regulatory elements in  expression vectors (see p. 778). Bacteria Can Be Transformed with Recombinant DNA The process of artificially introducing DNA into bacteria is referred to as transformation. It is accomplished by briefly exposing the cells to divalent cations that make  them transiently permeable to small DNA molecules. Recombinant plasmid molecules, containing foreign DNA, can be introduced into bacteria where it would  replicate normally.

Figure 18.8  Directional cloning of foreign DNA into vectors with a specified orientation.  Insertion of a foreign DNA fragment into a vector with a  specified orientation requires two different annealing sequences at each end  of the fragment and the corresponding complementary sequence at the two ends  generated in the vector. A polylinker with numerous unique restriction  endonuclease sites within the vector facilitates directional cloning.  Knowledge of the restriction map for the DNA of interest allows for selection  of appropriate restriction endonucleases to generate specific DNA fragments  that can be cloned in a vector.

 

  

Page 768

It Is Necessary to Be Able to Select Transformed Bacteria Once the plasmid has been introduced into the bacterium, both can replicate. Methods are available to select those bacteria that carry the recombinant DNA  molecules. In the recombinant process some bacteria may not be transformed or may be transformed with a vector not carrying foreign DNA; in preparing the vector  some may reanneal without inclusion of the DNA of interest. In some experimental conditions one can generate DNA fragments that can be readily purified for  recombinant studies. Such fragments can be generated from small, highly purified DNA species, for example, some DNA viruses. More typically, however, a single  restriction endonuclease will generate hundreds to hundreds of thousands of DNA fragments, depending on the size and complexity of DNA being studied. Individual  fragments cannot be isolated from these samples to be individually incorporated into the plasmid. Methods have therefore been developed to select those bacteria  containing the desired DNA. Restriction endonucleases do not necessarily hydrolyze DNA into fragments containing intact genes. If the fragment contains an entire gene it may not contain the  required flanking regulatory sequences, such as the promoter region. If the foreign gene is of mammalian origin, its regulatory sequences would not be recognized by  the bacterial synthetic machinery. The primary gene transcript (pre­mRNA) can also contain introns that cannot be processed by the bacteria. Recombinant DNA Molecules in a Gene Library When a complex mixture of thousands of different genes, arranged on different chromosomes, as in the human genome, is subjected to hydrolysis with a single  restriction endonuclease, thousands of DNA fragments are generated. These DNA fragments are annealed with a plasmid vector that has been cleaved to a linear  molecule with the same restriction endonuclease. By adjusting the ratio of plasmid to foreign DNA the probability of joining at least one copy of each DNA fragment  within a cyclized recombinant­plasmid DNA approaches one. Usually, only one out of the multiple DNA fragments is inserted into each plasmid vector. Bacteria are  transformed with the recombinant molecules such that only one plasmid is taken up by a single bacterium. Each recombinant molecule can now be replicated within the  bacterium and the bacterium will give rise to progeny, each carrying multiple copies of the recombinant DNA. The total population of bacteria now contain fragments  of DNA that may represent the entire human genome. This is termed a gene library. As in any library containing thousands of volumes, a selection system must be  available to retrieve the book or gene of interest. Plasmids are commonly employed to clone DNA fragments generated from molecules of limited size and complexity, such as viruses, and to subclone large DNA  fragments previously cloned in other vectors. Genomic DNA fragments are usually cloned from other vectors capable of carrying larger foreign DNA fragments than  plasmids (see p. 780). PCR May Circumvent the Need to Clone DNA Cloning and amplification of a DNA fragment carried within a vector may be employed for subcloning, mutagenesis, and sequencing. The PCR has, in many instances,  replaced the need to amplify recombinant DNA in a replicating biological system, greatly reducing the time and preparative steps required. It is not necessary to know  the sequence of the DNA insert (up to 6 kb) to amplify it by the PCR, since the sequence of the vector DNA flanking the insert is known. In some instances the PCR completely circumvents the need to clone the DNA of interest. For instance, a gene that has previously been cloned and sequenced can  readily be analyzed in patient DNA for the detection of mutations

 

  

Page 769

Figure 18.9  A multiplex PCR strategy to analyze a DNA region  of interest for mutated alterations.  A region of DNA within a complex DNA molecule, derived from  any source, can be amplified by the PCR with specific primers that are  complementary to sequences flanking the DNA region of interest  (Step 1). After multiple PCR cycles the amplified DNA (PCR  product) can then be used as a template simultaneously for  multiple pairs of primers (Step 2a) that are complementary  throughout the DNA (here they cover three segments—a,  b, and c). This procedure requires prior knowledge of the  sequence of the normal DNA/gene. Step 2a is repeated for DNA  derived from a patient with potential mutation(s) in the DNA  region of interest (Step 2b). The amplified DNA products from  the multiplex PCR step (Steps 2a and 2b) are then analyzed by  agarose gel electrophoresis to ascertain if the patient sample  contains a mutation (Step 3).

within this gene by a multiplex PCR strategy. DNA is isolated from patient blood cells and multiple pairs of oligonucleotide primers are synthesized to amplify the  entire gene or selected regions within the gene (Figure 18.9). Analysis of the amplified DNA fragments by agarose gel electrophoresis would

 

  

Page 770

CLINICAL CORRELATION 18.4  Multiplex PCR Analysis of HGPRTase Gene Defects in Lesch–Nyhan  Syndrome Lesch–Nyhan syndrome, as described in Clin. Corr. 12.2, results from a deficiency in  hypoxanthine–guanine phosphoribosyl­transferase (HGPRTase) activity. Several variant  forms of HGPRTase defects have been detected. Multiplex PCR amplification of the  HGPRTgene locus has been employed to analyze this gene in cells derived from Lesch– Nyhan patients and results account for the variability of the HGPRTase. The gene,  comprised of 9 exons, can be multiplex amplified using 16 different primers in a single  PCR. The products can be separated by agarose gel electrophoresis. Analysis of the  HGPRTgene locus by multiplex amplification of DNA derived from cells of several  patients detected great variations in deletions of different exons to total absence of the  exons. Rossiter, B. J. F., et al. In: M. J. McPherson, P. Quirke, and G. R. Taylor (Eds.), PCR.  A Practical Approach, Vol. 1. Oxford, England: Oxford University Press, 1994, p. 67.   allow one to detect any potential deletion mutation as compared to the normal gene products. Direct sequencing of multiple PCR products can be employed to detect  point mutations in the patient gene. Multiplex PCR has been used to detect various defects in the HGPRTase gene in Lesch–Nyhan patients (see Clin. Corr. 18.4). 18.6—  Selection of Specific Cloned DNA in Libraries Loss of Antibiotic Resistance Is Used to Select Transformed Bacteria When a single transformed bacterium carrying a recombinant molecule multiplies, its progeny are all genetically the same. If the transformed bacterium carries a  recombinant DNA, all progeny will carry copies of the same recombinant plasmid. The foreign DNA has been amplified and is derived from a single cloned DNA  fragment. The problem is how to identify the one colony containing the desired plasmid in a field of thousands to millions of different bacterial colonies. The plasmid  construct pBR322 and its descendants carry two genes that confer antibiotic resistance. Within these antibiotic­resistant genes are DNA sequences sensitive to  restriction endonuclease. When a fragment of foreign DNA is inserted into a restriction site within the gene for antibiotic resistance, the gene becomes nonfunctional.  Bacteria carrying this recombinant plasmid are sensitive to the antibiotic (Figure 18.10). The second antibiotic resistance gene within the plasmid, however, remains  intact and the bacteria will be resistant to this antibiotic. This technique of insertional inactivation of plasmid gene products affords a method to select bacteria that  carry recombinant plasmids. pBR322 contains genes that confer resistance to ampicillin (ampr) and tetracycline (tetr). A gene library with cellular DNA fragments inserted within the tetr gene can  be selected and screened in two stages (Figure 18.10). First, the bacteria are grown in an ampicillin­containing growth medium. Bacteria that are not transformed by a  plasmid (they lack a normal or recombinant plasmid) during the construction of the gene library will not grow in the presence of the antibiotic, thus eliminating this  population of bacteria. This, however, does not indicate which of the remaining viable bacteria carry a recombinant plasmid vector versus a plasmid with no DNA  insert. The second step is to identify bacteria carrying recombinant vectors with nonfunctional tetr genes, which are therefore sensitive to tetracycline. Bacteria insensitive to ampicillin are plated and grown on agar plates containing ampicillin (Figure 18.10). Replica plates can be made by touching the colonies on the  original agar plate with a filter and then touching additional sterile plates with the filter. All the plates will contain portions of each original colony at identifiable positions  on the plates. The replica plate can contain tetracycline, which will not support the growth of bacteria harboring recombinant plasmids with their tetr gene disrupted.  Comparison of replica plates with and without tetracycline will indicate which colonies on the original ampicillin plate contain recombinant plasmids. Thus individual  colonies containing the recombinant DNA can be selected, cultured, and analyzed. Either DNA or RNA probes (see pp. 583 and 773) can be utilized to identify the DNA of interest. Ampicillin­resistant bacterial colonies on agar can be replica  plated onto a nitrocellulose filter and adhering cells from each colony can be lysed with NaOH (Figure 18.10). DNA within the lysed bacteria is also denatured by the  NaOH and becomes firmly bound to the filter. A labeled DNA or RNA probe that is complementary to the DNA of interest can be hybridized to the nitrocellulose­ bound DNA. The filter is exposed to X­ray autoradiography. Any colony carrying the cloned DNA of interest will appear as a developed signal on the X­ray film.  These spots would then correspond to the colony on the

 

  

Page 771

Figure 18.10  Insertional inactivation of recombinant plasmids and detection of  transformed bacteria carrying a cloned DNA of interest.  When the insertion of a foreign DNA fragment into a vector disrupts a functional gene  sequence, the resulting recombinant DNA does not express the gene. The gene that  codes for antibiotic resistance to tetracycline (tetr) is destroyed by DNA insertion  while the ampicillin resistance gene (ampr) remains functional. Destruction  of one antibiotic resistance gene and retention of a second antibiotic  resistance gene allow for the detection of bacterial colonies carrying the  foreign DNA of interest within the replicating recombinant vector.

 

  

Page 772

original agar plate that can then be grown in a large­scale culture for further manipulation. Cloned and amplified DNA fragments usually do not contain a complete gene and are not expressed. The DNA inserts, however, can readily be purified for  sequencing or used as probes to detect genes within a mixture of genomic DNA, transcription levels of mRNA, and pathological conditions via clinical diagnostic tests.

a ­Complementation for Selecting Bacteria Carrying Recombinant Plasmids Other selection techniques can identify bacteria carrying recombinant DNA molecules. Vectors have been constructed (the pUC series) such that selected bacteria  transformed with these vectors carrying foreign DNA inserts can be identified visually (Figure 18.11). The pUC plasmids contain the regulatory sequences and a  portion of the 5 ­end coding sequence (N­terminal 146 amino acids) for the b ­galactosidase gene (lacZ gene) of the lac operon (Chapter 19, p. 802). The translated  N­terminal 146 amino acid fragment of b ­galactosidase is an inactive polypeptide. Mutant E. coli that code for the missing inactive carboxy­terminal portion of b ­ galactosidase can be transformed with the pUC plasmids. The translation of the host cell and plasmid portions of the b ­galactosidase in response to an inducer,  isopropylthio­ b ­D­galactoside, complement each other, yielding an active enzyme. The process is referred to as a ­complementation. When these transformed  bacteria are grown in the presence of a chromogenic substrate (5­bromo­4­chloro­3­indolyl­ b ­D­galactoside[X­gall) for b ­galactosidase they form blue colonies. If,  however, a foreign DNA fragment is inserted into the base sequence for the N­terminal portion of b ­galactosidase,

Figure 18.11  a ­Complementation for detection of transformed bacteria.  A vector has been constructed (pUC 18) that expresses the N­terminal coding  sequence for the enzyme b­galactosidase of the lac operon.  Bacterial mutants  coding for the C­terminal portion of b­galactosidase are transformed with pUC 18.  These transformed bacteria, grown in the presence of a special substrate for the  intact enzyme (X­gal), result in blue colonies because they contain the enzyme to  react with substrate. The functional N­terminal and C­terminal coding sequences  for the gene complement each other to yield a functional enzyme. If, however, a  foreign DNA fragment insert disrupts the pUC 18 N­terminal coding sequence for  b­galactosidase, bacteria transformed with this recombinant molecule will not  produce a functional enzyme. Bacterial colonies carrying these recombinant vectors  can then be visually detected as white colonies.

 

  

Page 773

the active enzyme cannot be formed. Bacteria transformed with these recombinant plasmids and grown on X­gal will yield white colonies and can be selected visually  from nontransformed blue colonies. 18.7—  Techniques for Detection and Identification of Nucleic Acids Nucleic Acids Can Serve As Probes for Specific DNA or RNA Sequences Selection of bacteria harboring recombinant DNAs of interest, analysis of mRNAs expressed in a cell, or identification of the presence of DNA sequences within a  genome require sensitive and specific detection methods. DNA and RNA probes meet these requirements. These probes contain nucleotide sequences  complementary to the target nucleic acid and will thus hybridize with the nucleic acid of interest. The degree of complementarity of a probe with the DNA under  investigation determines the tightness of binding of the probe. The probe does not need to contain the entire complementary sequence of the DNA. The probe, RNA  or DNA, can be labeled, usually with 32P. Nonradioactive labels are also employed that depend on enzyme substrates coupled to nucleotides, which when  incorporated into the nucleic acid can be detected by an enzyme­catalyzed reaction.

Figure 18.12  Nick translation to label DNA probes.  Purified DNA molecules can be  radioactively labeled and used to detect, by  hybridization, the presence of complementary  RNA or DNA in experimental samples.  (1) Nicking step: introduces random  single­stranded breaks in the DNA.  (2) Translation step: (a) E. coli DNA  polymerase (pol I) has 5    3  exonucleolytic  activity that hydrolyzes nucleotides from  the 5  end of the nick; (b) pol I simultaneously  fills in the single­stranded gap with  radioactively labeled nucleotides using the  3  end as a primer.

Labeled probes can be produced by nick translation of double­stranded DNA. Nick translation (Figure 18.12) involves the random enzymatic hydrolysis of a  phosphodiester bond in the backbone of one strand of DNA by DNase I; the enzymatic breaks in the DNA backbone are referred to as nicks. A second enzyme, E.  coli DNA polymerase I, with its 5    3  exonucleolytic activity and its DNA polymerase activity, creates single­strand gaps by hydrolyzing nucleotides from the 5   side of the nick and then filling in the gaps with its polymerase activity. The polymerase reaction is usually carried out in the presence of one a ­ 32P­labeled  deoxynucleotide triphosphate and three unlabeled deoxynucleotide triphosphates. The DNA employed in this method is usually purified and is derived from cloned  DNA, viral DNA, or cDNA. Another method to label DNA probes, random primer labeling of DNA, has distinct advantages over the nick translation method. The random primer method  typically requires only 25 ng of DNA as opposed to 1–2  g of DNA for nick translation and results in labeled probes with a specific activity (>109 cmp  g–1)  approximately ten times higher. This method generally produces longer labeled DNA probes. The double­stranded probe is melted and hybridized with a mixture of  random hexanucleotides containing all possible sequences (ACTCGG, ACTCGA, ACTCGC, etc.). The hybridized hexanucleotides serve as primers for DNA  synthesis with a DNA polymerase, such as the Klenow enzyme, in the presence of one or more radioactively labeled deoxynucleoside triphosphates. Labeled RNA probes have advantages over DNA probes. For one, relatively large amounts of RNA can be transcribed from a template, which may be available in  very limited quantities. A double­stranded DNA (dsDNA) probe must be denatured prior to hybridization with the target DNA and rehybridization with itself  competes for hybridization with the DNA of interest. No similar competition occurs with the single­stranded RNA probes that hybridize with complementary DNA or  RNA molecules. Synthesis of an RNA probe requires DNA as a template. To be transcribed the template DNA must be covalently linked to an upstream promoter  that can be recognized by a DNA­dependent RNA polymerase. Vectors have been constructed that are well suited for this technique. A labeled DNA or RNA probe can be hybridized to nitrocellulose­bound nucleic acids and identified by the detection of the labeled probe. The nucleic

 

  

Page 774

acids of interest can be transferred to nitrocellulose from bacterial colonies grown on agar or from agarose gels where the nucleic acid species have been  electrophoretically separated by size. Southern Blot Technique Is Useful for Identifying DNA Fragments A technique to transfer DNA species, separated by agarose gel electrophoresis, to a filter for analysis was developed in the 1970s, and it is an indispensable tool. The  method, developed by E. M. Southern, is referred to as the Southern blot technique (Figure 18.13). A DNA mixture of discrete restriction endonuclease­generated  fragments from any source and complexity can be separated according to size by electrophoresis through an agarose gel. The DNA is dena­

Figure 18.13  Southern blot to transfer DNA from  agarose gels to nitrocellulose.  Transfer of DNA to nitrocellulose, as single­stranded  molecules, allows for the detection of specific  DNA sequences within a complex mixture of DNA.  Hybridization with nick translated labeled  probes can demonstrate if a DNA sequence of  interest is present in the same or different  regions of the genome.

 

  

Page 775

tured by soaking the gel in alkali. The gel is then placed on absorbent paper and a nitrocellulose filter placed directly on top of the gel. Several layers of absorbent  paper are placed on top of the nitrocellulose filter. The absorbent paper under the gel is kept wet with a concentrated salt solution that by capillary action is pulled up  through the gel, the nitrocellulose, and into the absorbent paper layers above. The DNA is eluted from the gel by the upward movement of the high salt solution onto  the nitrocellulose filter directly above, where it becomes bound. The position of the DNA bound to the nitrocellulose filter is exactly that which was present in the  agarose gel. In its single­stranded membrane­bound form, the DNA can be analyzed with labeled probes. The Southern blot technique is invaluable in analytical procedures for detection of the presence and determination of the number of copies of particular sequences in  complex genomic DNA, confirming DNA cloning results, and demonstrating the polymorphic DNA arrangements of the human genome that correspond to  pathological states. An example of the use of Southern blots is shown in Figure 18.13. Here whole human genomic DNA, isolated from three individuals, was digested  with a restriction endonuclease, generating thousands of fragments. These fragments were distributed throughout the agarose gel according to size in an electric field.  The DNA was transferred (blotted) to a nitrocellulose filter and hybridized with a 32P­labeled DNA or RNA probe that represents a portion of a gene of interest. The  probe detected two bands in all three individuals, indicating that the gene of interest is cleaved at one site within its sequence. Individuals A and B presented a normal  pattern while patient C had one normal band and one lower molecular weight band. This is an example of altered DNA within different individuals of a single species,  restriction fragment length polymorphism (RFLP), and implies a deletion in a segment of the gene that may be associated with a pathological state. The gene from  this patient can be cloned, sequenced, and fully analyzed to characterize the altered nature of the DNA (see Clin. Corr. 18.5). Other techniques that employ the principles of Southern blot are the transfer of RNA (Northern blots) and of proteins (Western blots) to nitrocellulose filters or nylon  membranes. Single­Strand Conformation Polymorphism Southern blot analysis and detection of base changes in DNA from different individuals by RFLP analysis is dependent on alteration of a restriction endonuclease site.  Often a base substitution, deletion, or insertion does not occur within a restriction endonuclease site. However, these modifications can readily be detected by single­ strand conformation polymorphism (SSCP). This technique takes advantage of the fact that single­stranded DNA, smaller than 400 bases long, subjected to  electrophoresis through a polyacrylamide gel migrates with a mobility partially dependent on its conformation. A single base alteration usually modifies the DNA  conformation sufficiently to be detected as a mobility shift upon electrophoresis through a nondenaturing polyacrylamide gel. The analysis of a small region of genomic  DNA or cDNA for SSCP can be accomplished by PCR amplification of the region of interest. Sense and antisense oligonucleotide primers are synthesized that flank  the region of interest and this DNA is amplified by PCR in the presence of radiolabeled nucleotide(s). The resulting purified radiolabeled double­stranded PCR  product is then heat denatured in 80% formamide and immediately loaded onto a nondenaturing polyacrylamide gel. The mobilities of control products are compared  to samples derived from experimental/patient samples. Detection of mutations in patient samples can identify genetic lesions. This method depends on prior knowledge  of the sequence of the gene/gene fragment of interest, while analysis by RFLP requires only restriction map analysis of DNA.

 

  

Page 776

CLINICAL CORRELATION 18.5  Restriction Fragment Length Polymorphisms Determine the Clonal Origin of  Tumors It is generally assumed that most tumors are monoclonal in origin; that is, a rare event  alters a single somatic cell genome in such a fashion that the cells grow abnormally into a  tumor mass with all­daughter cells carrying the identically altered genome. Proof that a  tumor is of monoclonal origin versus polyclonal in origin can help to distinguish  hyperplasia (increased production and growth of normal cells) from neoplasia (growth of  new or tumor cells). The detection of restriction fragment length polymorphisms (RFLPs)  of Southern blotted DNA samples allows one to define the clonal origin of human tumors.  If tumor cells were collectively derived from different parental cells they should contain a  mixture of DNA markers characteristic of each cell of origin. However, an identical DNA  marker in all tumor cells would indicate a monoclonal origin. The analysis is limited to  females where one can take advantage of the fact that each cell carries only one active X  chromosome of either paternal or maternal origin with the second X chromosome being  inactivated. Activation occurs randomly during embryogenesis and is faithfully maintained  in all­daughter cells with one­half the cells carrying an activated maternal X chromosome  and the other one­half an activated paternal X chromosome. Analysis of the clonal nature of a human tumor depends on the fact that activation of an X  chromosome involves changes in the methylation of selected cytosine (C) residues within  the DNA molecule. Several restriction endonucleases, such as Hha I, which cleaves  DNA at GCGC sites, will not cleave DNA at their recognition sequences if a C is  methylated within this site. Therefore the methylated state (activated versus inactivated) of  the X chromosome can be probed with restriction endonucleases. Furthermore, the  paternal X chromosome can be distinguished from the maternal X chromosome in a  significant number of individuals based on differences in the electrophoretic migration of  restriction endonuclease generated fragments derived from selected regions of the  chromosome. These DNA fragments are identified on a Southern blot by hybridization  with a DNA probe that is complementary to this region of the X chromosome. An X­ linked gene that is amenable to these studies is the hypoxanthine  phosphoribosyltransferase (HPRTase) gene. The HPRTase gene consistently has two  BamHI restriction endonuclease sites (B1 and B3 in figure), but in many individuals a third  site (B2) is also present (see figure). The presence of site B2 in only one parental X chromosome HPRT allows for the  detection of restriction enzyme­generated polymorphisms. Therefore a female cell may  carry one X chromosome with the HPRT gene possessing two BamHI sites (results in a  single detectable DNA fragment of 24 kb) or three BamHI sites (results in a single  detectable DNA fragment of 12 kb). This figure depicts the expected results for the  analysis of tumor cell DNA to determine its monoclonal or polyclonal origin. As  expected, three human tumors examined by this method were shown to be of monoclonal  origin. Vogelstein, B., Fearon, E. R., Hamilton, S. R., and Feinberg, A. B. Use of restriction  fragment length polymorphism to determine the clonal origin of tumors. Science 227:642,  1985.

Analysis of genomic DNA to determine the  clonal origin of tumors.  (a) The X chromosome­linked HPRTase gene contains two  invariant BamHI restriction endonuclease sites  (B1 and B3) while in some individuals a third site,  B , is also present. The HPRTase gene also contains  2

several HhaI sites; however, all of these sites,  except H1, are usually methylated in the active  X chromosome. Therefore only the H1 site  would be available for cleavage by HhaI in the active  X chromosome. A cloned, labeled probe, pPB1.7, is  employed to determine which form of the HPRTase  gene is present in a tumor and if it is present  on an active X chromosome.  (b) Restriction endonuclease patterns predicted for monoclonal  versus polyclonal tumors are as follows:  (1) Cleaved with BamHI alone; 24­kb fragment derived    from HPRTase gene containing only B1 and B3 sites and 12­kb fragment derived from  HPRTase gene containing extra B  site. Pattern is  2

characteristic for heterozygous individual.  (2) Cleaved with BamHI plus HhaI, monoclonal tumor  with the 12­kb fragment derived from an  active X chromosome (methylated).  (3) Cleaved with BamHI plus HhaI; monoclonal tumor  with the 24­kb fragment derived from an active X chromosome  (methylated). (4) Cleaved with BamHI plus HhaI;  polyclonal tumor. All tumors studied displayed patterns  as in Lane 2 or Lane 3.

   

  

Page 777

18.8—  Complementary DNA and Complementary DNA Libraries The insertion of specific functional eukaryotic genes into vectors that can be expressed in a prokaryotic cell could produce large amounts of ''genetically engineered"  proteins with significant medical, agricultural, and experimental potential. Hormones and enzymes are currently produced by these methods, including insulin,  erythropoietin, thrombopoietin, interleukins, interferons, and tissue plasminogen activator. Unfortunately, it is impossible, except in rare instances, to clone functional  genes from genomic DNA. One reason for this is that most genes within the mammalian genome yield transcripts that contain introns that must be spliced out of the  primary mRNA transcript. Prokaryotic systems cannot splice out the introns to yield functional mRNA transcripts. This problem can be circumvented by synthesizing  complementary DNA (cDNA) from functional eukaryotic mRNA. mRNA Is Used As a Template for DNA Synthesis Using Reverse Transcriptase Messenger RNA can be reverse transcribed to cDNA and the cDNA inserted into a vector for amplification, identification, and expression. Mammalian cells normally  contain 10,000–30,000 different species of mRNA molecules at any time during the cell cycle. In some cases, however, a specific mRNA species may approach 90%  of the total mRNA, such as mRNA for globin in reticulocytes. Many mRNAs are normally present at only a few (1–14) copies per cell. A cDNA library can be  constructed from the total cellular mRNA but if only a few copies per cell of mRNA of interest are present, the cDNA may be very difficult to identify. Methods that  enrich the population of mRNAs or their corresponding cDNAs permit reduction of the number of different cDNA species within a cDNA library and greatly enhance  the probability of identifying the clone of interest. Desired mRNA in a Sample Can Be Enriched by Separation Techniques Messenger RNA can be separated by size by gel electrophoresis or centrifugation. Utilization of mRNA in a specific molecular size range will enrich several­fold an  mRNA of interest. Knowledge of the molecular weight of the protein encoded by the gene of interest gives a clue to the approximate size of the mRNA transcript or its  cDNA; variability in the predicted size, however, will arise from differences in the length of the untranslated regions of the mRNAs. Enrichment of a specific mRNA molecule can also be accomplished by immunological procedures but requires the availability of antibodies against the protein encoded  by the gene of interest. Antibodies added to an in vitro protein synthesis mixture will react with the growing polypeptide chain associated with the polysome and  precipitate it. The mRNA can be purified from the immunoprecipitated polysomal fraction. Complementary DNA Synthesis

Figure 18.14  Synthesis of cDNA from mRNA.  The 3  poly(A) tail of  mRNA is hybridized with an oligomer of dT  [oligo(dT)12–18] that serves as a  primer for reverse transcriptase, which  catalyzes the synthesis of the complementary  DNA (cDNA) strand in the presence of all  four deoxynucleotide triphosphates (dNTPs).  The resulting cDNA–mRNA hybrid is  separated into single­stranded cDNA by  melting with heat or hydrolyzing the mRNA  with alkali. The 3  end of  the cDNA molecule forms a hairpin loop that  serves as a primer for the synthesis of the  second DNA strand catalyzed by the Klenow  fragment of E. coli DNA polymerase. The  single­stranded unpaired DNA loop is  hydrolyzed by S  nuclease to  1

yield a double­stranded DNA molecule.

An isolated mRNA mixture is used as a template to synthesize a complementary strand of DNA using RNA­dependent DNA polymerase, reverse transcriptase  (Figure 18.14). A primer is required for the reaction; advantage is taken of the poly(A) tail at the 3  terminus of eukaryotic mRNA. An oligo(dT) with 12–18 bases is  employed as the primer that will hybridize with the poly(A) sequence. After cDNA synthesis, the hybrid is denatured or the mRNA hydrolyzed in alkali in order to  obtain the single­stranded cDNA. The 3  termini of single­stranded cDNAs form a hairpin loop that serves as a primer for the synthesis of the second strand of the  cDNA. Either the Klenow fragment or a reverse transcriptase can be used for this step. The resulting double­stranded cDNA contains a single­

 

  

Page 778

stranded loop that is selectively recognized and digested by S1 nuclease. The ends of the cDNA must be modified prior to cloning in a vector. One method involves  incubating blunt­ended cDNA molecules with linker molecules and a bacteriophage T4 DNA ligase that catalyzes the ligation of blunt­ended molecules (Figure 18.15).  The synthetic linker molecules contain restriction endonuclease sites that can now be hydrolyzed with the appropriate enzyme for insertion of the cDNA into a  compatibly cut vector.

Figure 18.15  Modification of cDNA for cloning.  The procedure begins with  double­stranded DNA containing a  hairpin loop. A linker DNA containing  a restriction endonuclease site  (RE1) is added to the free  end of the cDNA by blunt­end  ligation. The single­stranded hairpin  loop is next hydrolyzed with  S1 nuclease. A second  linker with a different restriction  endonuclease site within  (RE ) is blunt­end ligated  2

to the newly created free cDNA. The  second linker will probably bind to  both ends but will not interfere with  the first restriction endonuclease  site. The modified DNA is hydrolyzed  with the two restriction  endonucleases and can be inserted  into a plasmid or bacteriophage DNA  by directional cloning.

Bacteriophage DNA (see p. 779) is the most convenient and efficient vector to create cDNA libraries because they can readily be amplified and stored indefinitely.  Two bacteriophage vectors, lgt10 and lgt11, and their newer constructs have been employed to produce cDNA libraries. The cDNA libraries in  gt10 can be  screened only with labeled nucleic acid probes, whereas those in  gt11, an expression vector, can also be screened with antibody for the production of the protein or  antigen of interest. Total Cellular RNA May Be Used As a Template for DNA Synthesis Using RT­PCR Alternative methods to construct cDNA libraries employ a reverse transcriptase–PCR (RT­PCR) technique and obviate the need to purify mRNA. One such  strategy is depicted in Figure 18.16 and begins with the reverse transcriptase production of a DNA–mRNA hybrid. The method then adds a dG homopolymer tail to  the 3  end catalyzed by terminal transferase and the subsequent hydrolysis of the mRNA. PCR primers are synthesized to hybridize with the dG, dA tails and terminate  with two different restriction endonuclease sequences. The resulting PCR­amplified cDNA can then be hydrolyzed with the two different restriction endonucleases for  directional cloning (see p. 765, Section 18.5) into an appropriate vector. 18.9—  Bacteriophage, Cosmid, and Yeast Cloning Vectors Detection of noncoding sequences in most eukaryotic genes and distant regulatory regions flanking the genes necessitated new cloning strategies to package larger  DNA fragments than could be cloned in plasmids. Plasmids can accommodate foreign DNA inserts with a maximum length in the range of 5–10 kb (kilobases).  Portions of recombinant DNA fragments larger than this are randomly deleted during replication of the plasmid within the bacterium. Thus alternate vectors have been  developed. Bacteriophage As Cloning Vectors Bacteriophage l (l phage)—a virus that infects and replicates in bacteria—is an ideal vector for DNA inserts of approximately 15­kb lengths. The   phage  selectively infects bacteria and can replicate by either a lytic or nonlytic (lysogenic) pathway. The   phage contains a self­complementary 12­base single­stranded tail  (cohesive termini) at both ends of its 50­kb double­stranded DNA molecule. Upon infection of the bacteria the cohesive termini (cos sites) of a single   phage DNA  molecule self­anneal and the ends are covalently linked with the host cell DNA ligase. The circular DNA molecule serves as a template for transcription and  replication. The   phage, with restriction endonuclease­generated fragments representing a cell's whole genomic DNA inserted into it, is used to infect bacteria.  Recombinant bacteriophages, released from the lysed cells, are collected and constitute a genomic library in   phage. The phage library can be screened more rapidly  than a plasmid library due to the increased size of the DNA inserts. Numerous   phage vectors have been constructed for different cloning strategies. For the sake of simplicity only a generic   phage vector will be

 

  

Page 779

Figure 18.16  Generation of cDNA by reverse transcriptase–PCR (RT­PCR).  Total cellular RNA or mRNA can be used to generate cDNA by  RT­PCR. The mRNA with an oligo rA tail is reverse transcribed  with an oligo dT primer. An oligo dG tail is added to the 3  ends  of the RNA and DNA strands and the RNA strand is subsequently  hydrolyzed with NaOH. Sense and antisense primers, modified  with restriction site sequences, are then employed to amplify  the cDNA by the PCR. The products can be hydrolyzed with the  specific restriction endonucleases (RE  and RE )  1

2

for cloning and subsequent studies.

described here. Cloning large fragments of DNA in   phage takes advantage of the fact that a 15–25 kb segment of the phage DNA can be replaced without impairing  its replication in E. coli (Figure 18.17). Packaging of phage DNA into the virus particle is constrained by its total length, which must be approximately 50 kb. The  linear phage   DNA can be digested with specific restriction endonucleases that generate small terminal fragments with their cos sites (arms), which are separated  from the larger intervening fragments. Cellular genomic DNA is partially digested with the appropriate restriction enzymes to permit annealing and ligation with the  phage arms. Genomic DNA is not enzymatically hydrolyzed completely in order to randomly generate fragments that can be properly packaged into phage particles.  The DNA fragments that are smaller or larger than 15–25 kb can hybridize with the cos arms but are excluded from being packaged into infectious bacteriophage  particles. All of the information required for phage infection and replication in bacteria is carried within the cos arms. The recombinant phage DNA is mixed with    phage proteins in vitro, which assemble into infectious virions. The infectious recombinant   phage particles are then propagated in an appropriate E. coli strain to  yield a l phage library. Many different E. coli strains have been genetically altered to sustain replication of specific recombinant virions. Screening Bacteriophage Libraries The bacteriophage library can be screened by plating the virus on a continuous layer of bacteria (a bacterial lawn) grown on agar plates (Figure 18.18). The individual  phage will infect, replicate, and lyse one cell. The progeny virions will then infect and subsequently lyse bacteria immediately adjacent to the site of the first infected cell,  creating a clear region or plaque in the opaque bacterial

 

  

Page 780

Figure 18.17  Cloning genomic DNA in bacteriophage l.  Whole genomic DNA is incompletely digested with a restriction endonuclease  (e.g., EcoRI). This results in DNA of random size fragments with single­stranded  sticky ends. DNA fragments, cos arms, are generated with the same restriction  endonuclease from bacteriophage   DNA. The purified cos arm  fragments carry sequence signals required for packaging DNA into a  bacteriophage virion. The genomic fragments are mixed with the cos arms,  annealed, and ligated, forming linear concatenated DNA arrays. The in vitro  packaging with bacteriophage   proteins occurs only with genomic DNA  fragments of allowed lengths (15–25 kb) bounded by cos arms.

field. Phage, within each plaque, can be picked up on a nitrocellulose filter (as for replica plating) and the DNA fixed to the filter with NaOH. The location of cloned  DNA fragments of interest is determined by hybridizing the filter­bound DNA with a labeled DNA or RNA probe followed by autoradiography. Bacteriophages in the  plaque corresponding to the labeled filter­bound hybrid are picked up and amplified in bacteria for further analysis. Complementary DNA libraries in bacteriophage  are also constructed that contain the phage cos arms. If the cDNA is recombined with phage DNA that permits expression of the gene, such as  gt11, then plaques  can be screened immunologically with antibodies specific for the antigen of interest.

Figure 18.18  Screening genomic libraries in bacteriophage l.  Competent E. coli are grown to confluence on  an agar plate and then overlayed with the  recombinant bacteriophage. Plaques develop  where bacteria are infected and subsequently  lysed by the phage  . Replicas of the plate can  be made by touching the plate with a   nitrocellulose filter. The DNA is denatured  and fixed to the nitrocellulose with NaOH. The  fixed DNA is hybridized with a 32P­labeled probe  and exposed to X­ray film. The autoradiograph  identifies the plaque(s) with recombinant  DNA of interest.

Cloning DNA Fragments into Cosmid and Yeast Artificial Chromosome Vectors Even though   phage are the most commonly used vectors to construct genomic DNA libraries, the lengths of many genes exceed the maximum size of the

 

  

Page 781

DNA that can be inserted between the phage arms. A cosmid vector can accommodate foreign DNA inserts of approximately 45 kb. Yeast artificial  chromosomes (YACs) have been developed to clone DNA fragments of 200–500 kb lengths. While cosmid and yeast artificial chromosome vectors are difficult to  work with, their libraries permit the cloning of large genes with their flanking regulatory sequences, as well as families of genes or contiguous genes. Cosmid vectors are a cross between plasmid and bacteriophage vectors. Cosmids contain an antibiotic­resistance gene for selection of recombinant DNA molecules,  an origin of replication for propagation in bacteria, and a cos site for packaging of recombinant molecules in bacteriophage particles. The bacteriophage with  recombinant cosmid DNA can infect E. coli and inject its DNA into the cell. Cosmid vectors contain only approximately 5 kb of the 50­kb bacteriophage DNA and  therefore cannot direct replication and assembly of new infectious phage particles. Instead, the recombinant cosmid DNA circularizes and replicates as a large plasmid.  Bacterial colonies with recombinants of interest can be selected and amplified by methods similar to those described for plasmids. Standard cloning procedures and some novel methods are employed to construct YACs. Very large foreign DNA fragments are joined to yeast DNA sequences, one  that functions as a telomere (distal extremity of chromosome arm) and another that functions as a centromere and as an origin of replication. The recombinant YAC  DNA is introduced into the yeast by transformation. The YAC constructs are designed so that yeast transformed with recombinant chromosomes grow as visually  distinguishable colonies. This facilitates selection and analysis of cloned DNA fragments. 18.10—  Techniques to Further Analyze Long Stretches of DNA Subcloning Permits Definition of Large Segments of DNA Complete analysis of functional elements in a cloned DNA fragment requires sequencing of the entire molecule. Current techniques can sequence 200–400 bases in a  DNA fragment, yet cloned DNA inserts are frequently much larger. Restriction maps of the initial DNA clone are essential for cleaving the DNA into smaller pieces to  be recloned, or subcloned for further analysis. The sequences of each of the small subcloned DNA fragments can be determined. Overlapping regions of the  subcloned DNA properly align and confirm the entire sequence of the original DNA clone. Sequencing can often be accomplished without subcloning. Antisense primers can be synthesized that are complementary to the initially sequenced 3  ends of the  cloned DNA. This process is repeated until the full length of the cloned DNA has been sequenced. This method obviates the need to prepare subclones but it requires  synthesis/purchase of numerous primers. On the other hand, the subcloned DNA is always inserted back into the same region of the plasmid. Therefore one set of  primers complementary to the plasmid DNA sequences flanking the inserted DNA can be used for all of the sequencing reactions with subcloned DNA. Chromosome Walking Is a Technique to Define Gene Arrangement in Long Stretches of DNA Knowledge of how genes and their regulatory elements are arranged in a chromosome should lead to an understanding of how sets of genes may be coordinately  regulated. Currently, it is difficult to clone DNA fragments large enough to identify contiguous genes. The combination of several techniques allows for the analysis of  very long stretches of DNA (50–100 kb). The method, chromosome walking, is possible because   phage or cosmid libraries contain

 

  

Page 782

Figure 18.19  Chromosome walking to analyze contiguous DNA segments in a genome.  Initially, a DNA fragment is labeled by nick translation to screen a library for  recombinant   phage carrying a gene of interest. The amplified DNA is mapped  with a battery of restriction endonucleases to select a new region (1a) within the  original cloned DNA that can be recloned (subcloned). The subcloned DNA (1a) is  used to identify other DNA fragments within the original library that would overlap  the initially amplified DNA region. The process can be repeated many times to  identify contiguous DNA regions upstream and downstream of the initial DNA  (gene 1) of interest.

partially cleaved genomic DNA cut at specific restriction endonuclease sites. The cloned fragments will contain overlapping sequences with other cloned fragments.  Overlapping regions are identified by restriction mapping, subcloning, screening   phage or cosmid libraries, and sequencing procedures. The overall procedure of chromosome walking is shown in Figure 18.19. Initially the   phage library is screened for a sequence of interest with a DNA or RNA  probe. The cloned DNA is restriction mapped and a small segment is subcloned in a plasmid, amplified, purified, and labeled by nick translation. This labeled probe is  then used to rescreen the   phage library for complementary sequences, which are then cloned. The newly identified overlapping cloned DNA is then treated in the  same fashion as the initial DNA clone to search for other overlapping sequences. Caution must be taken that the subcloned DNA does not contain a sequence  common to the large numbers of repeating DNA

 

  

Page 783

sequences in higher eukaryotic genomes. If a subcloned DNA probe contains a repeat sequence it hybridizes to numerous bacteriophage plaques and prevents the  identification of a specific overlapping clone. 18.11—  Expression Vectors and Fusion Proteins Recombinant DNA methodology described to this point has dealt primarily with screening, amplification, and purification of cloned DNA species. An important goal of  recombinant DNA studies, as stated earlier, is to have a foreign gene expressed in bacteria with the product in a biologically active form. Sequencing the DNA of  many bacterial genes and their flanking regions has identified the spatial arrangement of regulatory sequences required for expression of genes. A promoter and other  regulatory elements upstream of the gene are required to transcribe a gene (Chapter 19, Section 19.3). mRNA transcript of a recombinant eukaryotic gene, however,  is not translated in a bacterial system because it lacks the bacterial recognition sequence, the Shine–Dalgarno sequence, required to properly orient it with a functional  bacterial ribosome. Vectors that facilitate the functional transcription of DNA inserts, termed expression vectors, have been constructed such that a foreign gene can  be inserted into the vector downstream of a regulated promoter but within a bacterial gene, commonly the lacZ gene. The mRNA transcript of the recombinant DNA  contains the lacZ Shine–Dalgarno sequence, codons for a portion of the 3  end of the lacZ gene protein, followed by the codons of the complete foreign gene of  interest. The protein product is a fusion protein that contains a few N­terminal amino acids of the lacZ gene protein and the complete amino acid sequence of the  foreign gene product. Foreign Genes Can Be Expressed in Bacteria Allowing Synthesis of Their Encoded Proteins Many plasmid and bacteriophage vectors have been constructed to permit expression of eukaryotic genes in bacterial cells. Rapidly replicating bacteria can serve as a  biological factory to produce large amounts of specific proteins, which have research, clinical, and commercial value. As an example, human protein hormones are  produced by recombinant technologies, which serve as replacement or supplemental hormones in patients with aberrant or missing hormone production. Figure 18.20  depicts a generalized plasmid vector for the expression of a mammalian gene. Recall that the inserted foreign gene must be in the form of cDNA from its corresponding  mRNA since the bacterial system cannot remove the introns in the pre­mRNA transcript. The DNA must be inserted in register with the codons of the 3 ­terminal  codons of the bacterial protein when creating a fusion protein. That is, insertion must occur after a triplet codon of the bacterial protein and at the beginning of a triplet  codon of the eukaryotic gene protein to ensure proper translation. Finally, the foreign gene must be inserted in the proper orientation relative to the promoter to yield a  functional transcript. This can be achieved by directional cloning. Eukaryotic proteins synthesized within bacteria are often unstable and are degraded by intracellular proteases. Fusion protein products, however, are usually stable.  The fusion protein amino acids encoded by the prokaryotic genome may be cleaved from the purified protein of interest by enzymatic or chemical procedures. An  alternative cloning strategy to circumvent the intracellular instability of some proteins is to produce a foreign protein that is secreted. This requires cloning the foreign  gene in a vector such that the fusion protein synthesized contains a signal peptide that can be recognized by the bacterial signal peptidase that properly processes the  protein for secretion.

 

  

Page 784

Figure 18.20  Construction of a bacterial expression vector.  A cDNA coding region of a protein of interest is inserted downstream of bacterial  regulatory sequences (promoter, P) for the lacZ gene, the coding sequence for the  mRNA Shine–Dalgarno sequence, the AUG codon, and a few codons for the  N­terminal amino acids of the lacZ gene protein. The mRNA produced from this  expression vector will therefore direct synthesis of a foreign protein in the bacterium  with a few of its N­terminal amino acids of bacterial protein origin (a fusion protein).

18.12—  Expression Vectors in Eukaryotic Cells Mammalian genetic diseases result from missing or defective intracellular proteins. To utilize recombinant techniques to treat these diseases, vectors have to be  constructed that can be incorporated into mammalian cells. In addition, these vectors have to be selective for the tissue or cells containing the aberrant protein.  Numerous vectors permit the expression of foreign DNA genes in mammalian cells grown in tissue culture. These vectors have been used extensively for elucidation of  the posttranslational processing and synthesis of proteins in cultured eukaryotic cells. Unfortunately, the goal to selectively express genes in specific tissues or at  specific developmental stages within an animal has met with very limited success. Several types of expression vectors have been developed that allow the replication, transcription, and translation of foreign genes in eukaryotic cells grown in vitro,  including both RNA and DNA viral vectors that contain a foreign DNA insert. These viral vectors are able to infect and then replicate in a host cell. Experimentally  constructed vectors that contain essential DNA elements, usually derived from a viral genome, permit expression of foreign gene inserts. Shuttle vectors contain both  bacterial and eukaryotic replication signals, thus permitting replication of the vector in both bacteria and mammalian cells. A shuttle vector allows a gene to be cloned  and purified in large quantities from a bacterial system and then the same recombinant vector can be expressed in a mammalian cell. Some expression vectors become  integrated into the host cell genome while others remain as extrachromosomal entities (episomes) with stable expression of their recombinant gene in the daughter cells.  Other expression vectors remain as episomal DNA, permitting only transient expression of their foreign gene prior to cell death. Foreign DNA, such as viral expression vectors, may be introduced into the cultured eukaryotic cells by transfection, a process that is analogous to transformation of  DNA into bacterial cells. The most commonly employed

 

  

Page 785

transfection methods involve the formation of a complex of DNA with calcium phosphate or diethylaminoethyl (DEAE)­dextran, which is then taken up by the cell by  endocytosis. The DNA is subsequently transferred from the cytoplasm to the nucleus, where it is replicated and expressed. The details of the mechanism of  transfection are not known. Both methods are employed to establish transiently expressed vectors while the calcium phosphate procedure is also used for permanently  expressed foreign genes. Typically, 10–20% of the cells in culture can be transfected by these procedures. DNA Elements Required for Expression of Vectors in Mammalian Cells Expression of recombinant genes in mammalian cells requires the presence of DNA­controlling elements within the vector that are not necessary in the bacterial  system. To be expressed in a eukaryotic cell the cloned gene is inserted in the vector in the proper orientation relative to control elements, including a promoter,  polyadenylation signals, and an enhancer sequence. Expression may be improved by the inclusion of an intron. Some or all of these DNA elements may be present in  the recombinant gene if whole genomic DNA is used for cloning. A particular cloned fragment generated by restriction endonuclease cleavage, however, may not  contain the required controlling elements. A cDNA would not possess these required DNA elements. It is therefore necessary that the expression vector to be used in  mammalian cells be constructed such that it contains all of the required controlling elements. An expression vector can be constructed by insertion of required DNA­controlling elements into the vector by recombinant technologies. Enhancer and promoter  elements, engineered into an expression vector, should be recognized by a broad spectrum of cells in culture for the greatest applicability of the vector. Controlling  elements derived from viruses with a broad host range are used for this purpose and are usually derived from the papovavirus, simian virus 40 (SV40), Rous  sarcoma virus, or the human cytomegalovirus. The vector must replicate so as to increase the number of copies within each cell or to maintain copies in daughter cells. The vector therefore is constructed to contain  DNA sequences that promote its replication in the eukaryotic cell. This DNA region is usually derived from a virus and is referred to as the origin of replication (Ori).  Specific protein factors, encoded by genes engineered into the vector or previously introduced into the host genome, recognize and interact with the ori sequences to  initiate DNA replication. Transfected Eukaryotic Cells Can Be Selected by Utilizing Mutant Cells That Require Specific Nutrients It is important to have a means of selectively growing the transfected cells since they often represent only 10–20% of the cell population. As was the case for the  bacterial plasmid, a gene can be incorporated into the vector that encodes an enzyme that confers resistance to a drug or confers selective growth capability to the cells  carrying the vector. Constructing vectors that express both a selectable marker and a foreign gene is difficult. Cotransfection circumvents this problem. Two different  vectors are efficiently taken up by those cells capable of being transfected. In most cases greater than 90% of transfected cells carry both vectors, one with the  selectable marker and the second carrying the gene of interest. Two of the more commonly employed selectable markers are the thymidine kinase (tk) and the dihydrofolate reductase gene. The tk gene product, thymidine kinase,  is expressed in most mammalian cells and participates in the salvage pathway for thymidine. Several mutant cell lines have been isolated that lack a functional thymidine  kinase gene (tk–) and in growth medium containing hypoxanthine, aminopterin, and thymidine these cells will not survive. Only

 

  

Page 786 –

those tk  mutant cells cotransfected with a vector carrying a tk gene, usually of herpes simplex virus origin, will grow in the medium. In most instances, these cells have  been cotransfected with the gene of interest. The dihydrofolate reductase gene (dhfr) is required to maintain cellular concentrations of tetrahydrofolate for nucleotide biosynthesis (see Chapter 13). Cells lacking  this enzyme will only survive in media containing thymidine, glycine, and purines. Mutant cells (dhfr–), which are transfected with the dhfr gene, can therefore be  selectively grown in a medium lacking these supplements. Expressing foreign genes in mutant cells, cotransfected with selectable markers, is limited to cell types that  can be isolated with the required gene defect. Normal cells, however, transfected with a vector carrying the dhfr gene, are also resistant to methotrexate, an inhibitor  of dihydrofolate reductase, and these cells can be selected for by growth in methotrexate. Another approach for selecting nonmutated cells involves the use of a bacterial gene coding for aminoglycoside 3 ­phosphotransferase (APH) for co­transfection. Cells  expressing APH are resistant to aminoglycoside antibiotics such as neomycin and kanamycin, which inhibits protein synthesis in both prokaryotes and eukaryotes.  Vectors carrying an APH gene can therefore be used as a selectable marker in both bacterial and mammalian cells. Foreign Genes Can Be Expressed in Eukaryotic Cells by Utilizing Virus Transformed Cells Figure 18.21 depicts the transient expression of a transfected gene in COS cells, a commonly used system to express foreign eukaryotic genes. The COS cells are  permanently cultured simian cells, transformed with an origin­defective SV40 genome. The defective viral genome has integrated into the host cell genome and  constantly expresses viral proteins. Infectious viruses, which are normally lytic to infected cells, are not produced because the viral origin of replication is defective. The  SV40 proteins expressed by the transformed COS cell will recognize and interact with a normal SV40 ori carried in a vector transfected into these cells. These SV40  proteins will therefore promote the repeated replication of the vector. A transfected vector containing both an SV40 ori and a gene of interest may reach a copy  number in excess of 105 molecules/cell. Transfected COS cells die after 3–4 days, possibly due to a toxic overload of the episomal vector DNA.

Figure 18.21  Expression of foreign genes in the eukaryotic  COS cell.  CV1, an established tissue culture  cell line of simian origin, can be infected and  supports the lytic replication of the simian  DNA virus, SV40. Cells are infected with an  origin (ori)­defective mutant of SV40 whose  DNA permanently integrates into the host  CV1 cell genome. The defective viral DNA  continuously codes for proteins that can  associate with a normal SV40 ori to regulate  replication. Due to its defective ori, the  integrated viral DNA will not produce  viruses. The SV40 proteins synthesized in  the permanently altered CV1 cell line,  COS­1, can, however, induce the replication  of recombinant plasmids carrying a wild­type  SV40 ori to a high copy number (as high as  105 molecules per cell). The foreign protein  synthesized in the transfected cells may be  detected immunologically or enzymatically.

18.13—  Site­Directed Mutagenesis By mutating selected regions or single nucleotides within cloned DNA, it is possible to define the role of DNA sequences in gene regulation and amino acid sequences  in protein function. Site­directed mutagenesis is the controlled alteration of selected regions of a DNA molecule. It may involve the insertion or deletion of selected  DNA sequences or the replacement of a specific nucleotide with a different base. A variety of chemical methods mutate DNA in vitro and in vivo usually at random  sites within the molecule. Role of Flanking Regions in DNA Can Be Evaluated by Deletion and Insertion Mutations Site­directed mutagenesis can be carried out in various regions of a DNA sequence including the gene itself or the flanking regions. Figure 18.22 depicts a simple  deletion mutation strategy where the sequence of interest is selectively cleaved with restriction endonuclease, the specific sequence removed, and the altered  recombinant vector recircularized with DNA ligase. The role of the

 

  

Page 787

deleted sequence can be determined by comparing the level of expression (translation) of the gene product, measured immunologically or enzymatically, to the  unaltered recombinant expression vector. A similar technique is used to insert new sequences at the site of cleavage. Deletion of a DNA sequence within the flanking  region of a cloned gene can help to define its regulatory role in gene expression. The presence or absence of a regulatory sequence may not be sufficient to evaluate its  role in controlling expression. The spatial arrangement of regulatory elements to one another, to the gene, and to its promoter may be important in the regulation of  gene expression (see Chapter 19). Analysis of potential regulatory sequences is conveniently conducted by inserting the sequence of interest upstream of a reporter gene in an expression vector. A  reporter gene, usually of prokaryotic origin, encodes for a gene product that can readily be distinguished from proteins normally present in the nontransfected cell and  for which there is a convenient and rapid assay. A commonly used reporter gene is the chloramphenicol acetyltransferase (CAT) gene of bacteria. The gene product  catalyzes the acetylation and inactivation of chloramphenicol, a protein synthesis inhibitor of prokaryotic cells. The ability of a regulatory element to enhance or  suppress expression of the CAT gene can be determined by assaying the level of acetylation of chloramphenicol in extracts prepared from transfected cells. The  regulatory element can be mutated prior to insertion into the vector carrying the reporter gene to determine its spatial and sequence requirements as a regulator of gene  expression. A difficulty encountered in analysis of regulatory elements is the lack of restriction endonuclease sites at useful positions within the cloned DNA. Deletion mutations  can be made, in the absence of appropriately positioned restriction endonuclease sites, by linearizing cloned DNA with a restriction endonuclease downstream of the  potential regulatory sequence of interest. The DNA can then be systematically truncated with an exonuclease, which hydrolyzes nucleotides from the free end of both  strands of the linearized DNA. Increasing times of digestion generates smaller DNA fragments. Figure 18.23 demonstrates how larger deletion mutations (yielding  smaller fragments) can be tested for functional activity. The enzymatic hydrolysis of the double strand of DNA occurs at both ends of the linearized recombinant  vector, destroying the original restriction endonuclease site (RE2). A unique restriction endonuclease site is reestablished to recircularize the truncated DNA molecule  for further manipulations to evaluate the function of the deleted sequence. This is accomplished by ligating the blunt ends with a linker DNA, a synthetic oligonucleotide  containing one or more restriction endonuclease sites. The ligated linkers are cut with the appropriate enzyme permitting recircularization and ligation of the DNA. Site­Directed Mutagenesis of a Single Nucleotide The previously discussed procedures can elucidate the functional role of small to large DNA sequences. Frequently, however, one wants to evaluate the role of a single  nucleotide at selected sites within the DNA molecule. A single base change permits evaluation of the role of specific amino acids in a protein (see Clin. Corr. 18.6).  This method also allows one to create or destroy a restriction endonuclease site at specific locations within a DNA sequence. The site­directed mutagenesis of a  specific nucleotide is a multistep process that begins with cloning the normal type gene in a bacteriophage (Figure 18.24). The M13 series of recombinant  bacteriophage vectors are commonly employed for these studies. M13 is a filamentous bacteriophage that specifically infects male E. coli that express sex pili encoded  for by a plasmid (F factor). M13 bacteriophage contains DNA in a single­stranded or replicative form, which is replicated to double­stranded DNA within an infected  cell. The double­stranded form of the

Figure 18.22  Use of expression vectors to study DNA  regulatory sequences.  The gene of interest  along with upstream and/or downstream DNA  flanking regions is inserted and cloned in an  expression vector and the baseline expression  of the gene in an appropriate cell is determined.  Defined regions of potential regulatory sequences  can be removed by restriction endonuclease  cleavage and the truncated recombinant DNA  vector can be recircularized, ligated, and  transfected into an appropriate host cell.  The level of gene expression in the absence  of the potential regulator is determined and  compared to controls to ascertain the regulatory  role of the deleted flanking DNA sequence.

 

  

Page 788

Figure 18.23  Enzymatic modification of potential  DNA regulatory sequences.  A purified recombinant DNA molecule with a  suspected gene regulatory element within flanking  DNA regions is cleaved with a restriction  endonuclease (RE2). The linearized recombinant  DNA is digested for varying time periods with  the exonuclease, Bal31, reducing the size of  the DNA flanking the potential regulatory  element. The resulting recombinant DNA  molecules of varying reduced sizes have small  DNA oligomers (linkers) containing a  restriction endonuclease sequence for  RE  ligated to their ends. The  2

linker­modified DNA is hydrolyzed with  RE , creating complementary  2

single­stranded sticky ends that permit  recircularization of recombinant vectors. The  potential regulatory element, bounded by  various reduced­sized flanking DNA sequences,  can be amplified, purified, sequenced, and  inserted upstream of a competent gene in an  expression vector. Modification of expression  of the gene in an appropriate transfected cell  can then be monitored to evaluate the role of  the potential regulatory element placed at  varying distances from the gene.

DNA is isolated from infected cells and used for cloning the gene to be mutated. The plaques of interest can be visually identified by a ­complementation (see p. 772). The M13 carrying the cloned gene of interest is used to infect susceptible E. coli. The progeny bacteriophages are released into the growth medium and contain single­ stranded DNA. An oligonucleotide (18–30 nucleotides long) is synthesized that is complementary to a region of interest except for the nucleotide to be mutated. This  oligomer, with one mismatched base, will hybridize to the single­stranded gene cloned in the M13 DNA and serves as a primer. Primer extension is accomplished with  the bacteriophage T4 DNA polymerase and the resulting double­stranded DNA can be transformed into susceptible E. coli, where the mutated DNA strand serves as  a template to replicate new (+) strands now carrying the mutated nucleotide. The bacteriophage plaques, containing the mutated DNA, are screened by hybridizing with a labeled probe of the original oligonucleotide. By adjusting the wash  temperature of the hybridized probe only the perfectly matched hybrid will remain complexed while the wild­type DNA–oligomer with mismatched nucleotide will  dissociate. The M13 carrying the mutated gene is then replicated in bacteria, the DNA purified, and the mutated region of the gene sequenced

 

  

Page 789

Figure 18.24  Site­directed mutagenesis of a single nucleotide and detection of the mutated DNA.  The figure is a simplified overview of the method. This process involves the  insertion of an amplified pure DNA fragment into a modified bacteriophage vector,  M13. Susceptible E. coli, transformed with the recombinant M13 DNA, synthesize the  (+) strand DNA packaged within the viron bacteriophage proteins. The bacteriophages  are isolated from the growth medium and the single­stranded recombinant M13 DNA  is purified. The recombinant M13 DNA serves as a template for DNA replication  in the presence of DNA polymerase, deoxynucleoside triphosphates (dNTPs),  DNA ligase, and a special primer The DNA primer (mismatched oligomer) is  synthesized to be exactly complementary to a region of the DNA (gene) of interest  except for the one base intended to be altered (mutated). The newly synthesized  M13 DNA therefore contains a specifically mutated base, which when reintroduced  into susceptible E. coli will be faithfully replicated. The transformed E. coli are grown  on agar plates with replicas of the resulting colonies picked up on a nitrocellulose  ilter. DNA associated with each colony is denatured and fixed to the filter with  NaOH and the filter­bound DNA is hybridized with a 32P­labeled mismatched DNA  oligomer probe. The putative mutants are then identified by exposing the filter to  X­ray film.

CLINICAL CORRELATION 18.6  Site­Directed Mutagenesis of HSV I gD The structural and functional roles of a carbohydrate moiety covalently linked to a protein  can be studied by site­directed mutagenesis. The gene that codes for a glycoprotein  whose asparagine residue(s) is normally glycosylated (N­linked) must first be cloned. The  herpes simplex virus type I (HSV I) glycoprotein D (gD) may contain as many as three  N­linked carbohydrate groups. The envelope bound HSV I gD appears to play a central  role in virus absorption and penetration. Carbohydrate groups may play a role in these  processes. The cloned HSV I gD gene has been modified by site­directed mutagenesis to alter  codons for the asparagine residue at the three potential glycosylation sites. These mutated  genes, cloned within an expression vector, were transfected into eukaryotic cells (COS­ 1), where the gD protein was transiently expressed. The mutated HSV I gD, lacking one  or all of its normal carbohydrate groups, can be analyzed with a variety of available  monoclonal anti­gD antibodies to determine if immunological epitopes (specific sites on a  protein recognized by an antibody) have been altered. Altered epitopes would indicate  that the missing carbohydrate moiety is directly associated with the normal recognition site  or played a role in the protein's native conformation. An altered protein conformation can  impact on immunogenicity (e.g., for vaccines) and protein processing (movement of the  protein from the endoplasmic reticulum, where it is synthesized, to the membrane, where  it is normally bound). Mutations at two of the glycosylation sites altered the native  conformation of the protein such that it was less reactive with selected monoclonal  antibodies. Alteration at a third site had no apparent effect on protein structure, and loss  of the carbohydrate chain at all three sites did not prevent normal processing of the  protein. Sodora, D. L., Cohen, G. H., and Eisenberg, R. J. Influence of asparagine­linked  oligosaccharides on antigenicity, processing, and cell surface expression of herpes simplex  virus type I glycoprotein D. J. Virol. 63:5184, 1989.    

  

Page 790

to confirm the identity of the mutation. Many modifications have been developed to improve the efficiency of site­directed mutagenesis of a single nucleotide including a  method to selectively replicate the mutated strand. M13 bacteriophage, replicated in a mutant E. coli, incorporates some uracil residues into its DNA in place of  thymine due to a metabolic defect in the synthesis of dTTP from dUTP and the lack of an enzyme that normally removes uracil residues from DNA. The purified  single­stranded M13 uracil­containing DNA is hybridized with a complementary oligomer containing a mismatched base at the nucleotide to be mutated. The oligomer  serves as the primer for DNA replication in vitro with the template (+) strand containing uracils and the new (–) strand containing thymines. When this double­ stranded M13 DNA is transformed into a wild­type E. coli, the uracil­containing strand is destroyed and the mutated (–) strand serves as the template for the progeny  bacteriophages, most of which will carry the mutation of interest. The polymerase chain reaction can also be employed for site­directed mutagenesis. Strategies have readily been developed to incorporate a mismatched base into one  of the oligonucleotides that primes the PCR. Some of these procedures employ M13 bacteriophage and follow the principles described in Figure 18.24. A variation of  these PCR methods, inverse PCR mutagenesis, has been applied to small recombinant plasmids (4–5 kb) (Figure 18.25). The method is very rapid with 50–100%  of the generated colonies containing the mutant sequence. The two primers are synthesized so that they anneal back­to­back with one primer carrying the mismatched  base. 18.14—  Applications of Recombinant DNA Technologies The practical uses of recombinant DNA methods in biological systems are limited only by one's imagination. Recombinant DNA methods are applicable to numerous  biological disciplines including agriculture, studies of evolution, forensic biology, and clinical medicine. Genetic engineering can introduce new or altered proteins into  crops (e.g., corn), so that they contain amino acids essential to humans but often lacking in plant proteins. Toxins that are lethal to specific insects but harmless to  humans can be introduced into crops to protect plants without the use of environmentally destructive pesticides. The DNA isolated from cells in the amniotic fluid of a  pregnant woman can be analyzed for the presence or absence of genetic defects in the fetus. Minuscule quantities of DNA can be isolated from biological samples that  have been preserved in ancient tar pits or frozen tundra and can be amplified and sequenced for evolutionary studies at the molecular level. The DNA from a single  hair, a drop of blood, or sperm from a rape victim can be isolated, amplified, and mapped to aid in identifying felons. Current technologies in conjunction with future  invented methods should permit the selective introduction of genes into cells with defective or absent genes. Developing methodologies are also likely to become  available to introduce nucleic acid sequences into cells to selectively turn off the expression of detrimental genes. Antisense Nucleic Acids Hold Promise As Research Tools and in therapy Recently, a new tool, antisense nucleic acids, has been introduced to study the intracellular expression and function of specific proteins. Natural and synthetic  antisense nucleic acids that are complementary to mRNAs will hybridize within the cell, inactivate the mRNA, and block translation. The introduction of antisense  nucleic acids into cells has opened new avenues to explore how proteins, whose expression has been selectively repressed in a cell, function within that cell. This  method also holds great promise in control of diseased processes

 

  

Page 791

Figure 18.25  Inverse PCR mutagenesis.  A single base can be mutated in recombinant DNA plasmids by  inverse PCR. Two primers are synthesized with their antiparallel 5 ends complementary to adjacent bases on the two strands of DNA.  One of the two primers carries a specific mismatched base that  is faithfully copied during the PCR amplification steps, yielding  ultimately a recombinant plasmid with a single mutated base.

 

  

Page 792

such as viral infections. Antisense technology, along with site­directed mutagenesis, are part of a new approach termed reverse genetics. Reverse genetics (from  gene to phenotype) selectively modifies a gene to evaluate its function, as opposed to classical genetics, which depends on the isolation and analysis of cells carrying  random mutations that can be identified. A second use of the term reverse genetics refers to the mapping and ultimate cloning of a human gene associated with a  disease where no prior knowledge of the molecular agents causing the disease exists. The use of the term ''reverse genetics" in this latter case is likely to be modified. Antisense RNA can be introduced into a cell by common cloning techniques. Figure 18.26 demonstrates one method. A gene of interest is cloned in an expression  vector in the wrong orientation. That is, the sense or coding strand that is normally inserted into the expression vector downstream of a promoter is intentionally  inserted in the opposite direction. This now places the complementary or antisense strand of the DNA under the control of the promoter with expression or  transcription yielding antisense RNA. Transfection of cells with the antisense expression vector introduces antisense RNA that is capable of hybridizing with normal  cellular mRNA. The mRNA–antisense RNA complex is not translated due to a number of reasons, such as its inability to bind to ribosomes, blockage of normal  processing, and rapid enzymatic degradation. DNA oligonucleotides have also been synthesized that are complementary to the known sequences of mRNAs of selected genes. Introduction of specific DNA  oligomers to cells in culture have inhibited viral infections including infections by the human immunodeficiency virus (HIV). It is conceivable that one day bone marrow  cells will be removed from AIDS patients and antisense HIV nucleic acids will be introduced into their cells in culture. These "protected" cells can then be reintroduced  into the AIDS patient's bone marrow (autologous bone marrow transplantation) and replace those cells normally destroyed by

Figure 18.26  Production of antisense RNA.  A gene, or a portion of it, is inserted into a vector by directional cloning downstream of a promoter and in the  reversed orientation to that normally found in the cell of origin. Transfection of this recombinant DNA into the  parental cell carrying the normal gene results in the transcription of RNA (antisense RNA) from the cloned  reversed­polarity DNA along with a normal cellular mRNA (sense RNA) transcript. The two anti­parallel  complementary RNAs hybridize within the cell, resulting in blocked expression (translation)of the  normal mRNA transcript.

 

  

Page 793

the virus. Experimental progress is also being made with antisense nucleic acids that can regulate the expression of oncogenes, genes involved in the cancer­forming  process. Harnessing antisense technologies holds great promise for treatment of human diseases. Normal Genes Can Be Introduced into Cells with a Defective Gene in Gene Therapy It is sometimes desirable for the transfected recombinant DNA to replicate to high copy numbers independent of the cell cycle. In other situations it is preferable for  only one or few copies to integrate into the host genome with its replication regulated by the cell cycle. Individuals who possess a defective gene resulting in a  debilitating or fatal condition could theoretically be treated by supplying their cells with a normal gene. Gene therapy is in its infancy; however, the successful transfer  of a normal gene to humans has been accomplished employing retroviral vectors (see Clin. Corr 18.7). The success of gene transfer depends, in part, on integration of  the gene into the host genome. This is directed by the retroviral integration system. Integration, however, is normally a random event that could result in deleterious  sequelae. Exciting studies are in progress that indicate that the viral integration machinery can be selectively tethered to specific target sequences within the host DNA  by protein–protein interactions to obviate these potential problems. CLINICAL CORRELATION 18.7  Normal Genes Can Be Introduced into Cells with Defective Genes in Gene  Therapy More than 4000 different genetic diseases are known, many of which are debilitating or  fatal. Most are currently incurable. With the advent of new technologies in molecular  biology, the clinical application of gene transfer and gene therapy is becoming a reality.  Adenosine deaminase (ADA) deficiency and Gaucher's disease are but two of many  genetic diseases that may readily be cured by gene therapy. ADA is important in purine salvage, catalyzing the conversion of adenosine to inosine or  deoxyadenosine to deoxyinosine. It is a protein of 363 amino acids with highest activity in  thymus and other lymphoid tissues. A defect in the ADA gene is inherited as an autosomal  recessive disorder. Over 30 mutations are associated with the disease. ADA deficiency  causes a severe combined immunodeficiency disease (SCID), by an unknown  mechanism. These immune­compromised children usually die in the first few years of life  due to overwhelming infections. The first authorized gene therapy in humans began on  September 14, 1990 with the treatment of a four­year­old girl with ADA deficiency. The  patient's peripheral blood T cells were expanded in tissue culture with appropriate growth  factors. The ADA gene was introduced within these cells by retroviral mediated gene  transfer. A modified retrovirus was constructed to contain the human ADA gene such that  it would be expressed in human cells without virus replication. (These viruses that cannot  replicate are first propagated in a cell line that contains a helper virus to produce  "infectious" viruses. The "infectious" viruses with foreign genetic information can now  infect and transfer information to cells without helper virus functions and, therefore, cannot  replicate.) Transfer of the ADA gene to the patient's T cells was mediated by retroviral  infection. Modified T cells carrying a normal ADA gene were then reintroduced to the  patient by autologous transfusion. Levels of ADA as low as 10% of normal are sufficient  to normalize the patient. Gaucher's disease is an autosomal recessive lysosomal storage disorder caused by a  deficiency of lysosomal glucocerebrosidase (GC). Clinical problems include  hepatosplenomegaly, pancytopenia, and bone deterioration. The enzyme is a lysosomal  membrane glycoprotein that contains 497 amino acids. Over 36 mutations, mostly  missense, that decrease catalytic activity are associated with the disease. The disorder  can be treated with enzyme replacement; however, this is very expensive and the patient  must be subjected to intravenous therapy throughout life. Viral constructs, similar to the  ADA protocol, have been made that carry the GC gene and have been successfully  transduced into a Gaucher patient's hematopoietic cells in culture with very high  efficiencies. These studies indicate that Gaucher patients may be normalized by gene  therapy in the near future. The genetically altered cells would become endogenous  factories capable of continuously synthesizing GC, thus obviating the need for intravenous  delivery of the missing enzyme. Blaese, R.M. Progress toward gene therapy. Clin. Immunol. Immunopathol 61:574,  1991; Mitani, K., Wakamiya, M., and Caskey, C. T. Long­term expression of retroviral­ transduced adenosine deaminase in human primitive hematopoietic progenitors. Hum.  Gene Ther. 4:9, 1993; and Xu, L., Stahl, S. K., Dave, H. P., Schiffman, R., Correll, P.  H., Kessler, S., and Karlsson, S. Correction of the enzyme deficiency in hematopoietic  cells of Gaucher patients using a clinically acceptable retroviral supernatant transduction  protocol. Exp. Hematol. 22:223, 1994.    

  

Page 794

Transgenic Animals Recombinant DNA methods allow production of large amounts of foreign gene products in bacteria and cultured cells. These methods also facilitate evaluation of the  role of a specific gene product in cell structure or function. In order to investigate the role of a selected gene product in the growth and development of a whole animal,  the gene must be introduced into the fertilized egg. Foreign genes can be inserted into the genome of a fertilized egg. Animals that develop from a fertilized egg with a  foreign gene insert carry that gene in every cell and are referred to as transgenic animals.

Figure 18.27  Production of transgenic animals.  Cloned, amplified, and purified functional  genes are microinjected into several fertilized  mouse egg pronuclei in vitro. The eggs are  implanted into a foster mother. DNA is isolated  from a small piece of each offspring pup's  tail and hybridized with a labeled probe to  identify animals carrying the foreign gene  (transgenic mouse). The transgenic mice can  be mated to establish a new strain of mice.  Cell lines can also be established from tissues  of transgenic mice to study gene regulation  and the structure/function of the foreign  gene product.

The most commonly employed method to create transgenic animals is outlined in Figure 18.27. The gene of interest is usually a cloned recombinant DNA molecule that  includes its own promoter or is cloned in a construct with a different promoter that can be selectively regulated. Multiple copies of the foreign gene are microinjected  into the pronucleus of the fertilized egg. The foreign DNA inserts randomly within the chromosomal DNA. If the insert disrupts a critical cellular gene the embryo will  die. Usually, nonlethal mutagenic events result from the insertion of the foreign DNA into the chromosome. Transgenic animals are currently being used to study several different aspects of the foreign gene, including the analysis of DNA regulatory elements, expression of  proteins during differentiation, tissue specificity, and the potential role of oncogene products on growth, differentiation, and induction of tumorigenesis. Eventually, it is  expected that these and related technologies will allow for methods to replace defective genes in the developing embryo (see Clin. Corr. 18.8). Recombinant DNA in Agriculture Will Have Significant Commercial Impact Perhaps the greatest gain to all humanity would be the practical use of recombinant technologies to improve our agricultural crops. Genes must be identified and  isolated that code for properties that include higher crop yield, rapid plant growth, resistance to adverse conditions such as arid conditions or cold periods, and plant  size. New genes, not common to plants, may be engineered into plants that confer resistance to insects, fungi, or bacteria. Finally, genes encoding existing structural  proteins can be modified to contain essential amino acids not normally present in the plant, without modifying the protein function. The potential to produce plants with  new genetic properties depends on the ability to introduce genes into plant cells that can differentiate into whole plants. New genetic information carried in crown gall plasmids can be introduced into plants infected with soil bacteria known as agrobacteria. Agrobacteria naturally  contain a crown gall or Ti (tumor­inducing) plasmid whose genes integrate into an infected cell's chromosome. The plasmid genes direct the host plant cell to produce  new amino acid species that are required for bacterial growth. A crown gall, or tumor mass of undifferentiated plant cells, develops at the site of bacterial infection.  New genes can be engineered into the Ti plasmid, and the recombinant plasmid introduced into plant cells upon infection with the agrobacteria. Transformed plant cells  can then be grown in culture and under proper conditions can be induced to redifferentiate into whole plants. Every cell would contain the new genetic information and  would represent a transgenic plant. Some limitations in producing plants with improved genetic properties must be overcome before significant advances in our world food supply can be realized. Clearly,  proper genes must yet be identified and isolated for desired characteristics. Also, important crops such as corn and wheat cannot be transformed by Ti plasmids;  therefore other vectors must be identified. However, significant success has been achieved in recent years in designing crop plants

 

  

Page 795

CLINICAL CORRELATION 18.8  Transgenic Animal Models Transgenic animal model systems hold promise for future methodologies to correct  genetic diseases early in fetal development. These animals are used to study the regulation  of expression and function of specific gene products in a whole animal and have the  potential for creating new breeds of commercially valuable animals. Transgenic mice have  been developed from fertilized mouse eggs with rat growth hormone (GH) genes  microinjected into their male pronuclei (see p. 835). The rat GH gene DNA, fused to the  mouse metallothionein­I (MT­I) promoter region, was purified from the plasmid in which  it had been cloned. Approximately 600 copies of the promoter–gene complex were  introduced into each egg, which was then inserted into the reproductive tract of a foster  mother mouse. The resulting transgenic mouse was shown to carry the rat GH gene within  its genome by hybridizing a labeled DNA probe to mouse DNA that had been purified  from a slice of the tail, restriction endonuclease digested, electrophoresed, and Southern  blotted. The diet of the animals was supplemented with ZnSO4 at 33 days postparturition.  The ZnSO4 presumably can activate the mouse MT­I promoter to initiate transcription of  the rat GH gene. The continuous overexpression of rat GH in some transgenic animals  produced mice nearly twice the size of littermates that did not carry the rat GH gene. A  transgenic mouse transmitted the rat GH gene to one­half of its offspring, indicating that  the gene stably integrated into the germ cell genome and that new breeds of animals can  be created. Palmiter, R. D., Brinster, R. L., Hammer, R. E., et al. Dramatic growth of mice that  develop from eggs microinjected with metallothionein­growth hormone fusion genes.  Nature 300:611, 1982.   with resistance to insects and viruses. Of equal importance is the very recent genetic engineering feat of inserting a foreign gene into pea plants that now produce a  protein that inhibits the feeding of weevil larvae on the pea seeds. Peas and other legume seeds will be able to be stored without the need of protective chemical  fumigants (currently Brazilian farmers lose 20–40% of their stored beans to pests). 18.15—  Concluding Remarks The old cliché, so close and yet so far away, seems appropriate for our current juncture in molecular biology. The eukaryotic yeast genome, which consists of  approximately 14 million base pairs of DNA distributed among 16 chromosomes, will be entirely sequenced by the mid­1990s. Equally impressive is the fact that the  entire human genome will likely be sequenced in the next decade or so (see Clin. Corr. 15.11). Two human chromosomes, 16 and 19, have been fully mapped and it  is anticipated they will be the first chromosomes to be fully sequenced. More than 100,000 cDNA clones are available for sequencing, which ultimately will provide  landmarks of the huge human genetic map. More than 100 clinical trials in gene therapy have been initiated since the apparent success with ADA. Genetic diseases  now identified and to be identified should eventually be curable by gene replacement therapy when the technical roadblocks are surmounted. If one looks at the  enormous advances made in molecular biology in just the past two decades it is reasonable to believe the "when" will not be that far off. Bibliography Askari, F. K., and McDonnell, W. M. Molecular medicine: Antisense oligonucleotide therapy. N. Engl. J. Med. 334:316, 1996. Brown, W. M., George, M. Jr., and Wilson, A. C. Rapid evolution of animal mitochondrial DNA. Proc. Natl. Acad. Sci. USA 76:1967, 1979. Bushman, F. Targeting retroviral integration. Science 267:1443, 1995. Davis, L. G., Kuehl, W. M., and Battey, J. F. Basic Methods in Molecular Biology, 2nd ed. Norwalk, CT: Appleton & Lange, 1994. Erlich, H. A. (Ed.). PCR Technology. Principles and Applications for DNA Amplification. New York: Stockton Press, 1989. Feinberg, A., and Vogelstein, B. Addendum: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem. 137:266,  1984. Jaenisch, R. Transgenic animals. Science 240:1468, 1988. Kreeger, K. Y. Influential consortium's cDNA clones praised as genome research time­saver. The Scientist 9:1, 1995. Kunkel, T. A. Rapid and efficient site­specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82:488, 1985. Marshall, E. Gene therapy's growing pains. Science 269:1050, 1995.

 

  

Page 796

Maxam, A. M., and Gilbert, W. A new method of sequencing DNA. Proc. Natl. Acad. Sci. USA 74:560, 1977. McPherson, M. J., Quirke, P., and Taylor, G. R. (Eds.). PCR. A Practical Approach, Vol. 1. Oxford, England: Oxford University Press, 1994. Mulligan, R. C. The basic science of gene therapy. Science 260:926, 1993. Palmiter, R. D., Brinster, R. L., Hammer, R. E., et al. Dramatic growth of mice that develop from eggs microinjected with metallothionein–growth hormone fusion  genes. Nature 300:611, 1982. Rigby, P. W. J., Dieckmann, M., Rhodes, C., and Berg, P. Labelled deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I.  J. Mol. Biol. 113:237, 1977. Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning. A Laboratory Manual, 2nd ed. New York: Cold Spring Harbor Laboratory Press, 1989. Sanger, F., Nicklen, S., and Coulson, A. R. DNA sequencing with chain­terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463, 1977. Southern, E. M. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503, 1975. Vogelstein, B., Fearon, E. R., Hamilton, S. R., and Feinberg, A. P. Use of restriction fragment length polymorphism to determine the clonal origin of human tumors.  Science 227:642, 1985. Watson, J. D., Tooze, J., and Kurtz, D. T. Recombinant DNA: A Short Course San Francisco: Scientific American Books/Freeman, 1983. Weintraub, H. M. Antisense RNA and DNA. Sci. Am. 262:40, 1990. Williams, N. Closing in on the complete yeast genome sequence. Science 268:1560, 1995. Zhang, Y., and Yunis, J. J. Improved blood RNA extraction micro­technique for RT­PCR. Biotechniques 18:788, 1995. Questions J. Baggott and C. N. Angstadt 1. Development of recombinant DNA methodologies is based on discovery of: A. the polymerase chain reaction (PCR). B. restriction endonucleases. C. plasmids. D. complementary DNA (cDNA). E. yeast artificial chromosomes (YACs). 2. The essential property of the DNA polymerase employed in the polymerase chain reaction (PCR) is that it: A. does not require a primer. B. is unusually active. C. is thermostable. D. replicates double­stranded DNA. E. can replicate both eukaryotic and prokaryotic DNA. 3. Construction of a restriction map of DNA requires all of the following EXCEPT: A. partial hydrolysis of DNA. B. complete hydrolysis of DNA. C. electrophoretic separation of fragments on a gel. D. staining of an electrophoretic gel to locate DNA. E. cyclic heating and cooling of the reaction mixture. 4. In the Maxam–Gilbert method of DNA sequencing: A. cleavage of the DNA backbone occurs randomly at only some of the sites where the base had been destroyed. B. all nucleotides produced during cleavage of the DNA backbone are detected by radioautography. C. electrophoretic separation of DNA fragments is due to differences in both size and charge. D. the sequence of bands in the four lanes of the autoradiogram contains the base sequence information. E. dideoxynucleoside triphosphates are used. 5. The Sanger and Maxam–Gilbert methods of DNA sequencing differ in that: A. the Maxam–Gilbert method involves labeling the 5  end, while the Sanger method requires labeling the 3  end of the DNA. B. only the Maxam–Gilbert method involves electrophoresing a mixture of fragments of different sizes. C. the Sanger method employs DNA cleavage, while the Maxam–Gilbert method employs interrupted DNA synthesis. D. only the Maxam–Gilbert method uses radioautography to detect fragments in which one of the termini is radioactively labeled. E. in the Maxam–Gilbert method, a complete DNA chain is cleaved, while in the Sanger method, synthesis of the chain is interrupted at different points. 6. Preparation of recombinant DNA requires: A. restriction endonucleases that cut in a staggered fashion. B. restriction endonucleases that cleave to yield blunt­ended fragments. C. poly (dT). D. DNA ligase. E. cDNA. 7. In the selection of colonies of bacteria that carry cloned DNA in plasmids, such as pBR322, that contain two antibiotic resistance genes: A. one antibiotic resistance gene is nonfunctional in the desired bacterial colonies. B. untransformed bacteria are antibiotic resistant. C. both antibiotic resistance genes are functional in the desired bacterial colonies. D. radiolabeled DNA or RNA probes play a role. E. none of the above. 8. A technique for defining gene arrangement in very long stretches of DNA (50–100 kb) is: A. RFLP. B. chromosome walking. C. nick translation. D. Southern blotting. E. SSCP. 9. Which of the following pairs of vectors and DNA insert sizes is correct? A. plasmids 5–10 kb B. cosmids 15 kb C YACs 2000–5000 kb D. bacteriophage   45 kb E. none of the above

 

  

Page 797

10. Expression of a eukaryotic gene in prokaryotes involves which of the following: A. a SD sequence in mRNA. B. absence of introns. C. regulatory elements upstream of the gene. D. a fusion protein. E. all of the above. Refer to the following for Questions 11–15. A. antisense nucleic acid B. polymerase chain reaction C. site­directed mutagenesis D. shuttle vector E. transfecton 11. Contains both bacterial and eukaryotic replication signals. 12. Complementary to mRNA and will hybridize to it, thus blocking translation. 13. Can rapidly produce large quantities of a specific DNA. 14. Oligomer with one mismatched base is used as a primer. 15. A process that introduces foreign DNA into a eukaryotic genome. Answers 1. B The ability to cleave DNA predictably at specific sites is essential to recombinant DNA technology (p. 760). 2. C PCR requires cycling between low temperatures, where hybridization of template DNA and oligomer primers occurs, and high temperatures, where DNA melts  (p. 759). The Taq DNA polymerase, isolated from a thermophilic organism discovered in a hot spring on federal land, is stable at high temperatures and makes the  cycling possible with no addition of fresh polymerase after each cycle. The lucrative commercialization of this publicly owned natural resource, with no royalties  accruing to the public (i.e., taxpayers') coffers, has evoked criticism from some observers. 3. E Cyclic heating and cooling are part of the PCR process, not of restriction mapping (p. 761). A and B: Restriction mapping involves all degrees of hydrolysis.  Partial hydrolysis gives fragments of varying sizes, and complete hydrolysis gives the smallest possible fragments. C and D: Fragments are electrophoretically separated  by size on agarose gel, which is stained to reveal the DNA. 4 D The relative positions of G are given by the bands in the lane corresponding to the destruction of G; of A by the bands in the AG lane that are not duplicated in the  G lane; of C by the bands in the C lane; of T by the bands in the CT lane that are not duplicated in the C lane. A: Cleavage occurs at all such sites. Limited destruction  of the bases is random (p. 762). B: Only the nucleotides that contain the labeled 5  terminal are detected. Other nucleotides are produced but are not detected by the  method and do not contribute information to the analysis (p. 762). C: Although charge is, of course, required to produce movement of a particle in a field, the  separation of these fragments is not due to charge differences, but to size differences, with the smallest fragments migrating farthest (pp. 762–763). E: This is part of the  Sanger method (p. 763). 5. D They both use radioautography to detect fragments in which one of the termini is radioactively labeled. A: The Sanger method involves a labeled 5  end. With the  Maxam–Gilbert method either end could be labeled. Here we show labeling of the 3  end. B: Both methods do this. C: This statement reverses the methodologies. E:  See pp. 762–765. 6. D DNA ligase covalently connects fragments held together by interaction of cohesive ends (p. 765). A: This is the most desirable type of restriction endonuclease to  use, but it is not essential. B: Restriction nucleases that make blunt cuts can also be used if necessary. C: This is used in conjunction with poly (dA) if restriction  endonucleases that make blunt cuts are employed, but it is not essential to all of recombinant DNA preparation. 7. A The foreign DNA is inserted into one antibiotic resistance gene, thus destroying it (p. 770). B: Resistance is due to the plasmids. C: See the comment for A  above. D: Radiolabeling detects the DNA of interest, not the colonies that contain cloned DNA (p. 770). 8. B A: Restriction fragment length polymorphism (RFLP) is a characteristic of DNA, not a technique (p. 775). C: Nick translation is used to label DNA during  chromosome walking (p. 773). D: Southern blotting is a method for analyzing DNA (p. 774). E: Single­strand conformation polymorphism (SSCP) is a method for  detecting base changes in DNA that do not alter restriction endonuclease sites. 9. A B: Cosmids will accept a 45­kb insert (p. 781). C: YACs will accept a 200–500 kb insert (p. 781). D: Bacteriophage   will accept a 15­kb insert (p. 779). 10. E A: The SD sequence is necessary for the bacterial ribosome to recognize the mRNA. B: Bacteria do not have the intracellular machinery to remove introns from  mRNA. C: Appropriate regulatory elements are necessary to allow the DNA to be transcribed. D: A fusion protein may be a product of the reaction (p. 783). 11. D (see p. 784). 12. A (see p. 790). 13. B (see p. 759). 14. C (see p. 788). 15. E (see p. 784).

 

  

Page 799

Chapter 19—  Regulation of Gene Expression John E. Donelson

19.1 Overview

800

19.2 Unit of Transcription in Bacteria: The Operon

800

19.3 Lactose Operon of E. Coli

802

Repressor of the Lactose Operon Is a Diffusible Protein

803

Operator Sequence of the Lactose Operon Is Contiguous on DNA with  a Promoter and Three Structural Genes

805

Promoter Sequence of Lactose Operon Contains Recognition Sites for  RNA Polymerase and a Regulator Protein

806

Catabolite Activator Protein Binds at a Site on the Lactose Promoter

807

19.4 Tryptophan Operon of E. Coli

807

Tryptophan Operon Is Controlled by a Repressor Protein

808

Tryptophan Operon Has a Second Control Site: The Attenuator Site

810

Transcription Attenuation Is a Mechanism of Control in Operons for  Amino Acid Biosynthesis

813

19.5 Other Bacterial Operons

813

Synthesis of Ribosomal Proteins Is Regulated in a Coordinated Manner

813

Stringent Response Controls Synthesis of rRNAs and tRNAs

815

19.6 Bacterial Transposons

816

Transposons Are Mobile Segments of DNA

816

Tn3 Transposon Contains Three Structural Genes

816

19.7 Inversion of Genes in Salmonella

818

19.8 Organization of Genes in Mammalian DNA

820

Only a Small Fraction of Eukaryotic DNA Codes for Proteins

820

Eukaryotic Genes Usually Contain Intervening Sequences (Introns)

820

19.9 Repetitive DNA Sequences in Eukaryotes

822

Importance of Highly Repetitive Sequences Is Unknown

822

A Variety of Repeating Units Are Defined as Moderately Repetitive  Sequences

822

19.10 Genes for Globin Proteins Recombinant DNA Technology Has Been Used to Clone Genes for  Many Eukaryotic Proteins

824

Sickle Cell Anemia Is Due to a Single Base Pair Change

828

Thalassemias Are Caused by Mutations in Genes for the a  or b  Subunits  of Globin

828

19.11 Genes for Human Growth Hormone­Like Proteins

829

19.12 Mitochondrial Genes

830

19.13 Bacterial Expression of Foreign Genes

832

Recombinant Bacteria Can Synthesize Human Insulin

832

Recombinant Bacteria Can Synthesize Human Growth Hormone

834

19.14 Introduction of Rat Growth Hormone Gene into Mice

835

Bibliography

836

Questions and Answers

836

Clinical Correlations

   

  

824

 

19.1 Transmissible Multiple Drug Resistances

816

19.2 Duchenne/Becker Muscular Dystrophy and the Dystrophin Gene

822

19.3 Huntington's Disease and Trinucleotide Repeat Expansions

823

19.4 Prenatal Diagnosis of Sickle Cell Anemia

828

19.5 Prenatal Diagnosis of Thalassemia

829

19.6 Leber's Hereditary Optic Neuropathy (LHON)

831

Page 800

19.1—  Overview To survive, a living cell must be able to respond to changes in its environment. One of many ways in which cells adjust to changes is to alter expression of specific  genes, which, in turn, affects the number of corresponding protein molecules in the cell. This chapter focuses on some of the molecular mechanisms that determine  when a given gene will be expressed and to what extent. The attempt to understand how expression of genes is regulated is one of the most active areas of biochemical  research today. It makes sense for a cell to vary the amount of a given gene product available under different conditions. For example, the bacterium Escherichia coli (E. coli)  contains genes for about 3000 different proteins, but it does not need to synthesize all of these proteins at the same time. Therefore it regulates the number of molecules  of these proteins that are made. The classic illustration of this phenomenon is the regulation of the number of b ­galactosidase molecules in the cell. This enzyme  converts the disaccharide lactose into the monosaccharides, glucose and galactose. When E. coli is growing in a medium containing glucose as the carbon source, b ­ galactosidase is not required and only about five molecules of the enzyme are present in the cell. When lactose is the sole carbon source, however, 5000 or more  molecules of b ­galactosidase occur in the cell. Clearly, the bacteria respond to the need to metabolize lactose by increasing the synthesis of b ­galactosidase molecules.  If lactose is removed from the medium, the synthesis of this enzyme stops as rapidly as it began. The complexity of eukaryotic cells means that they have even more extensive mechanisms of gene regulation than do prokaryotic cells. The differentiated cells of higher  organisms have a much more complicated physical structure and often a more specialized biological function that is determined, again, by the expression of their genes.  For example, insulin is synthesized in b  cells of the pancreas and not in kidney cells even though the nuclei of all cells of the body contain the insulin genes. Molecular  regulatory mechanisms facilitate the expression of insulin in pancreas and prevent its synthesis in kidney and other cells. In addition, during development of the organism  appearance or disappearance of proteins in specific cell types is tightly controlled with respect to timing and sequence of developmental events. As expected from the differences in complexities, far more is understood about the regulation of genes in prokaryotes than in eukaryotes. However, studies on the  control of gene expression in prokaryotes often provide exciting new ideas that can be tested in eukaryotic systems. Sometimes, discoveries about eukaryotic gene  structure and regulation alter the interpretation of data on the control of prokaryotic genes. Several of the best studied examples of gene regulation in bacteria will be discussed, followed by some illustrations of the organization and regulation of related genes in  the human genome. Finally, the use of recombinant DNA techniques to express some human genes of clinical interest will be presented. 19.2—  Unit of Transcription in Bacteria: The Operon The single E. coli chromosome is a circular double­stranded DNA molecule of about four million base pairs. Most of the approximately 3000 E. coli genes are not  distributed randomly throughout this DNA; instead, the genes that code for the enzymes of a specific metabolic pathway are clustered in one region of the DNA. In  addition, genes for associated structural proteins, such as the 70 or so proteins that comprise the ribosome, are frequently adjacent to one another. Members of a set  of clustered genes are usually coordinately regulated; they are transcribed together to form a "polycistronic" mRNA species that contains the coding sequences for  several proteins. The term operon describes the

 

  

Page 801

complete regulatory unit of a set of clustered genes. An operon includes the adjacent structural genes that code for the related enzymes or associated proteins, a  regulatory gene or genes that code for regulator protein(s), and control elements that are sites on the DNA near the structural genes at which regulator proteins act.  Figure 19.1 shows a partial genetic map of the E. coli chromosome that gives locations of structural genes of some of the different operons. When transcription of the structural genes of an operon increases in response to the presence of a specific substrate in the medium, the effect is known as induction.  The increase in transcription of the b ­galactosidase gene when lactose is the sole carbon source is an example of induction. Bacteria also respond to nutritional  changes by quickly turning off the synthesis of enzymes that are no longer needed. As will be described below, E. coli synthesizes the amino acid tryptophan as the  end product of a specific biosynthetic pathway. However, if tryptophan is supplied in the medium, the bacteria do not need to make it themselves, and synthesis of  enzymes for this metabolic pathway is stopped. This process is called repression. It permits the bacteria to avoid using their energy for making unnecessary and even  harmful proteins. Induction and repression are manifestations of the same phenomenon. In one case the bacterium changes its enzyme composition so that it can utilize a specific  substrate in the medium; in the other it reduces the number of enzyme

Figure 19.1  Partial genetic map of E. coli.  The locations of only a few of the genes identified and mapped in E. coli are shown  here. Three operons discussed in this chapter are indicated.  Reproduced with permission from Stent, G. S., and Calendar, R. Molecular Genetics,  An Introductory Narrative. San Francisco: Freeman, 1978, p. 289; modified from  Bachmann, B. J., Low, K. B., and Taylor, A. L. Bacteriol Rev. 40:116, 1976.

 

  

Page 802

molecules so that it does not overproduce a specific metabolic product. The signal for each type of regulation is the small molecule that is a substrate for the metabolic  pathway or a product of the pathway, respectively. These small molecules are called inducers when they stimulate induction and corepressors when they cause  repression to occur. Section 19.3 will describe in detail the lactose operon, the best studied example of a set of inducible genes. Section 19.4 will present the tryptophan operon, an  example of a repressible operon. Sections 19.5–19.7 will briefly describe some other operons as well as some gene systems in which physical movement of the genes  themselves within the DNA (i.e., gene rearrangements) plays a role in their regulation. 19.3—  Lactose Operon of E. Coli The lactose operon contains three adjacent structural genes as shown in Figure 19.2. LacZ codes for the enzyme b ­galactosidase, which is composed of four identical  subunits of 1021 amino acids. LacY codes for a permease, which is a 275­amino acid protein that occurs in the cell membrane and participates in the transport of  sugars, including lactose, across the membrane. The third gene, lacA, codes for b ­galactoside transacetylase, a 275­amino acid enzyme that transfers an acetyl group  from acetyl CoA to b ­galactoside. Of these three proteins, only b ­galactosidase actually participates in a known metabolic pathway. However, the permease is clearly  important in the utilization of lactose since it is involved in transporting lactose into the cell. The acetylation reaction may be associated with detoxification and excretion  reactions of nonmetabolized analogs of b ­galactosides. Mutations in lacZ or lacY that destroy the function of b ­galactosidase or permease prevent cells from cleaving lactose or acquiring it from the medium, respectively.  Mutations in lacA that destroy transacetylase activity do not seem to have an identifiable effect on cell growth and division. Perhaps there are other related enzymes in  the cell that serve as backups for this enzyme, or perhaps it has an unknown function that is required only under certain conditions. A single mRNA species containing the coding sequences of all three structural genes is transcribed from a promoter that occurs just upstream from the lacZ gene.  Induction of these three genes occurs during initiation of their

Figure 19.2  Lactose operon of E. coli.  The lactose operon is composed of the lacI gene, which codes for a repressor, the control  elements of CAP, lacP, and lacO, and three structural genes, lacZ, lacY, and lacA, which code for  b­galactosidase, a permease, and a transacetylase, respectively. The lacI gene is transcribed from its  own promotor. Three structural genes are transcribed from the promoter, lacP, to form a  polycistronic mRNA from which the three proteins are translated.

 

  

Page 803

transcription. Without the inducer, transcription of the gene cluster occurs only at a very low level. In the presence of the inducer, transcription begins at the promoter,  called lacP, and goes through all three genes to a transcription terminator located slightly beyond the end of lacA. Therefore the genes are coordinately expressed;  either all three are transcribed in unison or none is transcribed. The presence of three coding sequences on the same mRNA molecule suggests that the relative amounts of the three proteins are always the same under varying  conditions of induction. An inducer that causes a high rate of transcription will result in a high level of all three proteins; an inducer that stimulates only a little  transcription of the operon will result in a low level of the proteins. The inducer can be thought of as a molecular switch that influences synthesis of the single mRNA  species for all three genes. The number of molecules of each protein in the cell may be different, but this does not reflect differences in transcription; it reflects  differences in translation rates of the coding sequences or in degradation of the proteins themselves. The mRNA induced by lactose is very unstable; it is degraded with a half­life of about 3 min. Therefore expression of the operon can be altered very quickly.  Transcription ceases as soon as inducer is no longer present, existing mRNA molecules disappear within a few minutes, and cells stop making the proteins. Repressor of the Lactose Operon Is a Diffusible Protein The regulatory gene of the lactose operon, lacI, codes for a protein whose only function is to control the transcription initiation of the three lac structural genes. This  regulator protein is called the lac repressor. The lacI gene is located just in front of the controlling elements for the lacZYA gene cluster. However, it is not obligatory  that a regulatory gene be physically close to the gene cluster it regulates. In some of the other operons it is not. Transcription of lacI is not regulated; instead, this single  gene is always transcribed from its own promoter at a low rate that is relatively independent of the cell's status. Therefore affinity of the lacI promoter for RNA  polymerase seems to be the only factor involved in its transcription initiation. The lac repressor is initially synthesized as a monomer of 360 amino acids and four monomers associate to form a tetramer, the active form of the repressor. Usually  there are about 10 tetramers per cell. The repressor has a strong affinity for a specific DNA sequence that lies between lacP and the start of lacZ. This sequence is  called the operator and is designated lacO. The operator overlaps the promoter somewhat so that presence of repressor bound to the operator physically prevents  RNA polymerase from binding to the promoter and initiating transcription. In addition to recognizing and binding to the lac operator DNA sequence, the repressor also has a strong affinity for the inducer molecules of the lac operon. Each  monomer has a binding site for an inducer molecule. Binding of inducer to the monomers causes an allosteric change in the repressor that greatly lowers its affinity for  the operator sequence (Figure 19.3). In other words, when inducer molecules are bound to their sites on the repressor, a conformational change in the repressor  occurs that alters the binding site for the operator. The result is that repressor no longer binds to the operator so that RNA polymerase, in turn, can begin transcription  from the promoter. A repressor molecule that is already bound to the operator when the inducer becomes available can still bind to inducer so that the repressor– inducer complex immediately disassociates from the operator. A study of the lactose operon has been greatly facilitated by the discovery that some small molecules fortuitously serve as inducers but are not metabolized by b ­ galactosidase. Isopropylthiogalactoside (IPTG) is one of several thiogalac­

 

  

Page 804

Figure 19.3  Control of lac operon.  (a) Repressor tetramer binds to operator and prevents transcription of structural genes.  (b) Inducer binds to repressor tetramer, which prevents repressor from binding to  operator. Transcription of three structural genes can occur from the promoter.

tosides with this property. They are called gratuitous inducers. They bind to inducer sites on the repressor molecule causing the conformational change but are not  cleaved by the induced b ­galactosidase. Therefore they affect the system without themselves being altered (metabolized) by it. If it were not possible to manipulate  experimentally the system with these gratuitous inducers, it would have been much more difficult to reach our current understanding of the lactose operon in particular  and bacterial gene regulation in general. The product of the lacI gene, the repressor protein, acts in trans; that is, it is a diffusible product that moves through the cell to its site of action. Therefore mutations in  the lacI gene can exert an effect on the expression of other genes located far away or even on genes located on different DNA molecules. LacI mutations can be of  several types. One class of mutations changes or deletes amino acids of the repressor that are located in the binding site for the inducer. These changes interfere with  interaction between the inducer and the repressor but do not affect the affinity of repressor for the operator. Therefore the repressor is always bound to the operator,  even in the presence of inducer, and the lacZYA genes are never transcribed above a very low basal level. Another class of lacI mutations changes the amino acids in  the operator­binding site of the repressor. Most of these mutations lessen the affinity of the repressor for the operator. This means that repressor does not bind to the  operator and lacZYA genes are always being transcribed. These mutations are called repressor­constitutive mutations because lac genes are permanently turned  on. Interestingly, a few rare lacI mutants actually increase the affinity of repressor for the operator over that of wild­type repressor. In these cases inducer molecules  can still bind to repressor, but they are less effective in releasing repressor from the operator. Repressor­constitutive mutants illustrate the features of a negative control system. An active repressor, in the absence of an inducer, shuts off the expres­

 

  

Page 805

sion of the lac structural genes. An inactive repressor results in the constitutive, unregulated, expression of these genes. It is possible, using the recombinant DNA  techniques described in Chapter 18, to introduce into constitutive lacI mutant cells a recombinant plasmid containing the wild­type lacI gene (but not the rest of the lac  operon). Therefore these cells have one wild­type and one mutant lacI gene and will synthesize both active and inactive repressor molecules. Under these conditions,  normal wild­type regulation of the lactose operon occurs. In genetic terms, the wild­type induction is dominant over the mutant constitutivity. This property is the main  feature of a negative control system. Operator Sequence of the Lactose Operon Is Contiguous on DNA with a Promoter and Three Structural Genes The known control elements in front of the structural genes of the lactose operon are the operator and promoter. The operator was originally identified, like the lacI  gene, by mutations that affected the transcription of the lacZYA region. Some of these mutations also result in the constitutive synthesis of lac mRNA; that is, they are  operator­constitutive mutations. In these cases the operator DNA sequence has undergone one or more base pair changes so that the repressor no longer binds as  tightly to the sequence. Thus the repressor is less effective in preventing RNA polymerase from initiating transcription. In contrast to mutations in the lacI gene that affect the diffusible repressor, mutations in the operator do not affect a diffusible product. They exert their influence on the  transcription of only the three lac genes that lie immediately downstream of the operator on the same DNA molecule. This means that if a second lac operon is  introduced into a bacterium on a recombinant plasmid, the operator of one operon does not influence action on the other operon. Therefore an operon with a wild­ type operator will be repressed under the usual conditions, whereas in the same bacterium a second operon that has an operator­constitutive mutation will be  transcribed continuously. Operator mutations are frequently referred to as cis­dominant to emphasize that these mutations affect only adjacent genes on the same DNA molecule and that they  are not influenced by the presence in the cell of other copies of the unmutated sequence. Cis­dominant mutations occur in DNA sequences that are recognized by  proteins rather than in DNA sequences that code for the diffusible proteins. Trans­dominant mutations occur in genes that specify the diffusible products. Therefore  cis­dominant mutations also occur in promoter and transcription termination sequences, whereas trans­dominant mutations also occur in the genes for the subunit  proteins of RNA polymerase, the ribosomes, and so on. Figure 19.4 shows the sequence of both the lac operator and promoter. The operator sequence has an axis of dyad symmetry. The sequence of the upper strand on  the left side of the operator is nearly identical to the lower strand on the right side; only three differences occur between these inverted

Figure 19.4  Nucleotide sequence of control elements of lactose operon.  The end of the lacI gene (coding for the lactose repressor) and the beginning of the lacZ gene (coding for b­galactosidase)  are also shown. Lines above and below the sequence indicate symmetrical sequences within the CAP site and operator.

 

  

Page 806

DNA repeats. This symmetry in the DNA recognition sequence reflects symmetry in the tetrameric repressor. It probably facilitates the tight binding of the subunits of  the repressor to the operator, although this has not been definitively demonstrated. A common feature of many protein­binding or recognition sites on double­stranded  DNA, including most recognition sites for restriction enzymes, is a dyad symmetry in the nucleotide sequence. The 30 bp that constitute the lac operator are an extremely small fraction of the total E. coli genome of 4 × 106 bp and occupy an even smaller fraction of the total  volume of the cell. Therefore it would seem that the approximately 10 tetrameric repressors in a cell might have trouble finding the lac operator if they just randomly  diffuse about the cell. Although this remains a puzzling consideration, there are factors that confine the repressor to a much smaller space than the entire volume of the  cell. First, it probably helps that the repressor gene is very close to the lac operator. This means that the repressor does not have far to diffuse if its translation begins  before its mRNA is fully synthesized. Second, and more importantly, the repressor possesses a low general affinity for all DNA sequences. When the inducer binds to  the repressor, its affinity for the operator is reduced about a 1000­fold, but its low affinity for random DNA sequences is unaltered. Therefore all of the lac repressors  of the cell probably spend the majority of the time in loose association with the DNA. As the binding of the inducer releases a represser molecule from the operator, it  quickly reassociates with another nearby region of the DNA. Therefore induction redistributes the repressor on the DNA rather than generates freely diffusing  repressor molecules. This confines the repressor to a smaller volume within the cell. Another question is how does lactose enter a lac­repressed cell in the first place if the lacY gene product, the permease, is repressed yet is required for lactose  transport across the cell membrane? The answer is that even in the fully repressed state, there is a very low basal level of transcription of the lac operon that provides  five or six molecules of the permease per cell. Perhaps this is just enough to get a few molecules of lactose inside the cell and begin the process. An even more curious observation is that, in fact, lactose is not the natural inducer of the lactose operon as we would expect. When the repressor is isolated from fully  induced cells, the small molecule bound to each repressor monomer is allolactose, not lactose. Allolactose, like lactose, is composed of galactose and glucose, but  the linkage between the two sugars is different. It turns out that a side reaction of b ­galactosidase (which normally breaks down lactose to galactose and glucose)  converts these two products to allolactose. Therefore it appears that a few molecules of lactose are taken up and converted by b ­galactosidase to allolactose, which  then binds to the repressor and induces the operon. Further confirmation that lactose itself is not the real inducer comes from experiments indicating that lactose binding  to the purified repressor slightly increases the repressor's affinity for the operator. Therefore, in the induced state, a small amount of allolactose must be present in the  cell to overcome this ''anti­inducer" effect of the lactose substrate. Promoter Sequence of Lactose Operon Contains Recognition Sites for RNA Polymerase and a Regulator Protein Immediately in front of the lac operator sequence is the promoter sequence. This sequence contains the recognition sites for two different proteins, RNA polymerase  and the CAP­binding protein (Figure 19.4). The site at which RNA polymerase interacts with the DNA to initiate transcription has been identified using several  different genetic and biochemical approaches. Point mutations in this region frequently affect the affinity to which RNA polymerase will bind the DNA. Deletions (or  insertions) that extend into this region also dramatically affect the binding of RNA polymerase to the DNA. The end points of the sequence to which RNA polymerase  binds were identified by DNase protection

 

  

Page 807

experiments. Purified RNA polymerase was bound to the lac promoter region cloned in a bacteriophage DNA or a plasmid, and this protein–DNA complex was  digested with DNase I. The DNA segment protected from degradation by DNase was recovered and its sequence determined. The ends of this protected segment  varied slightly with different DNA molecules but corresponded closely to the boundaries of the RNA polymerase interaction site shown in Figure 19.4. The sequence of the RNA polymerase interaction site is not composed of symmetrical elements similar to those described for the operator sequence. This is not  surprising since RNA polymerase must associate with the DNA in an asymmetrical fashion for RNA synthesis to be initiated in only one direction from the binding site.  However, that portion of the promoter sequence recognized by the CAP­binding protein does contain some symmetry. A DNA–protein interaction at this region  enhances transcription of the lac operon as described in the next section. Catabolite Activator Protein Binds at a Site on the Lactose Promoter

Figure 19.5  Lack of synthesis of b ­galactosidase in E. coli  when glucose is present.  The bacteria are growing in a medium  containing initially 0.4 mg mL–1 of glucose and  2 mg mL–1 lactose. The left­hand ordinate  indicates the optical density of the growing  culture, an indicator of the number of bacterial  cells. The right­hand ordinate indicates the  units of b­galactosidase per milliliter. Note  that the appearance of b­galactosidase is  delayed until the glucose is depleted.  Redrawn from Epstein, W., Naono, S., and  Gros, F. Biochem. Biophys. Res. Commun.  24:588, 1966.

Escherichia coli prefers to use glucose instead of other sugars as a carbon source. For example, if the concentrations of glucose and lactose in the medium are the  same, the bacteria will selectively metabolize the glucose and not utilize the lactose. This phenomenon is illustrated in Figure 19.5, which shows that the appearance of  b ­galactosidase, the lacZ product, is delayed until all of the glucose in the medium is depleted. Only then can lactose be used as the carbon source. This delay  indicates that glucose interferes with the induction of the lactose operon. This effect is called catabolite repression because it occurs during the catabolism of glucose  and may be due to a catabolite of glucose rather than glucose itself. An identical effect is exerted on a number of other inducible operons, including the arabinose and  galactose operons, which code for enzymes involved in the utilization of various substances as energy sources. It probably is a general coordinating system for turning  off synthesis of unwanted enzymes whenever the preferred substrate, glucose, is present. Catabolite repression begins in the cell when glucose lowers the concentration of intracellular cyclic AMP (cAMP). The exact mechanism by which this reduction in  the cAMP level is accomplished is not known. Perhaps glucose influences either the rate of synthesis or degradation of cAMP. At any rate, cAMP can bind to another  regulatory protein, which has not been discussed yet, called CAP (for catabolite activator protein) or CRP (for cAMP receptor protein). CAP is an allosteric  protein, and when it is combined with cAMP, it is capable of binding to the CAP regulatory site that is at the promoter of the lac (and other) operons. The CAP– cAMP complex exerts positive control on the transcription of these operons. Its binding to the CAP site on the DNA facilitates the binding of RNA polymerase to the  promoter (Figure 19.6). Alternatively, if the CAP site is not occupied, RNA polymerase has more difficulty binding to the promoter, and transcription of the operon  occurs much less efficiently. Therefore, when glucose is present, the cAMP level drops, the CAP–cAMP complex does not form, and the positive influence on RNA  polymerase does not occur. Conversely, if glucose is absent, the cAMP level is high, a CAP–cAMP complex binds to the CAP site, and transcription is enhanced. 19.4—  Tryptophan Operon of E. Coli Tryptophan is essential for bacterial growth; it is needed for the synthesis of all proteins that contain tryptophan. Therefore, if tryptophan is not present in sufficient  amount by the medium, the cell must make it. In contrast, lactose is not absolutely required for the cell's growth; many other sugars can substitute for it, and, in fact, as  we saw in the previous section, the bacterium prefers to

 

  

Page 808

Figure 19.6  Control of lacP by cAMP.  A CAP–cAMP complex binds to the CAP site and enhances transcription at  lacP. Catabolite repression occurs when glucose lowers the intracellular  concentration of cAMP. This reduces the amount of the CAP–cAMP complex and  decreases transcription from lacP and from the promoters of several other operons.

use some of these other sugars for the carbon source. As a result, synthesis of the tryptophan biosynthetic enzymes is regulated differently than synthesis of the proteins  encoded by the lactose operon. Tryptophan Operon Is Controlled by a Repressor Protein In E. coli tryptophan is synthesized from chorismic acid in a five­step pathway that is catalyzed by three different enzymes as shown in Figure 19.7. The tryptophan  operon contains the five structural genes that code for these three enzymes (two of which have two different subunits). Upstream from this gene cluster is a promoter  where transcription begins and an operator to which binds a repressor protein encoded by the unlinked trpR gene. Transcription of the lactose operon is generally  "turned off" unless it is induced by the small molecule inducer. The tryptophan operon, on the other hand, is always "turned on" unless it is repressed by the presence of  a small molecule corepressor (a term used to distinguish it from the repressor protein). Hence the lac operon is inducible, whereas the trp operon is repressible.  When the trp operon is being actively transcribed, it is said to be derepressed; that is, the trp repressor is not preventing RNA polymerase from binding. This is  mechanistically the same as an induced lactose operon in which the lac repressor is not interfering with RNA polymerase. The biosynthetic pathway for tryptophan synthesis is regulated by mechanisms that affect both the synthesis and activity of the enzymes that catalyze the pathway. For  example, anthranilate synthetase, which catalyzes the first step of the pathway, is encoded by the trpE and trpD genes of the trp operon. The number of molecules of  this enzyme that is present in the cell is determined by the transcriptional regulation of the trpoperon. However, the catalytic activity of the existing molecules of the  enzyme is regulated by feedback inhibition. This is a common short­term means of regulating the first committed step in a metabolic pathway. In this case,  tryptophan, the end product of the pathway, can bind to an allosteric site on the anthranilate synthetase and interfere with its catalytic activity at another site. Therefore,  as the concentration of tryptophan

 

  

Page 809

Figure 19.7  Genes of tryptophan operon of E. coli.  Regulatory elements are the primary promoter (trpP), operator (trpO), attenuator (trp a), secondary internal promoter (trpP2),  and terminator (trp t) Direction of mRNA synthesis is indicated on the wavy lines representing mRNAs. Col  and CoII  signify  2

2

components I and II, respectively, of the anthranilate synthetase (ASase) complex; PR­anthranilate is N­5 ­phosphoribosyl­anthranilate;   CdRP is 1­(o­carboxy­phenylamino)­1­deoxyribulose­5­phosphate; InGP is indole­3­glycerol phosphate; PRPP is  5­phosphoribosyl­1­pyrophosphate; and TSase is tryptophan synthetase.  Redrawn from Platt, T. The tryptophan operon. In: J. H. Miller and W. Reznikoff (Eds.), The Operon. Cold Spring Harbor, NY:  Cold Spring Harbor Laboratory Press, 1978, p. 263.

builds up in the cell, it begins to bind to anthranilate synthetase and immediately decreases its activity on the substrate, chorismic acid. In addition, tryptophan also acts  as a corepressor to shut down the synthesis of new enzyme molecules from the trp operon. Thus feedback inhibition is a short­term control that has an immediate  effect on the pathway, whereas repression takes a little longer but has the more permanent effect of reducing the number of enzyme molecules. The trp repressor is a tetramer of four identical subunits of about 100 amino acids each. Under normal conditions about 20 molecules of the repressor tetramer are  present in the cell. The repressor by itself does not bind to the trp operator. It must be complexed with tryptophan in order to bind to the operator and therefore acts  in vivo only in the presence of tryptophan. This is exactly the opposite of the lac repressor, which binds to its operator only in the absence of its small molecule  inducer. Interestingly, trp repressor also regulates transcription of trpR, its own gene. As trp repressor accumulates in cells, the repressor­tryptophan complex binds  to a region upstream of this gene, turning off its transcription and maintaining the equilibrium of 20 repressors per cell. Another difference from the lac operon is that  the trp operator occurs entirely within the trp promoter rather than adjacent to it, as shown in Figure 19.8. The operator sequence is a region of dyad symmetry, and  the mechanism

Figure 19.8  Nucleotide sequence of control elements of the tryptophan operon.  Lines above and below sequence indicate symmetrical sequences within operator.

 

  

Page 810

of preventing transcription is the same as in the lac operon. Binding of the repressor–corepressor complex to the operator physically blocks the binding of RNA  polymerase to the promoter. Repression results in about a 70­fold decrease in the rate of transcription initiation at the trp promoter. (In contrast, the basal level of lac gene products is about 1000­ fold lower than the induced level.) However, the trp operon contains additional regulatory elements that impose further control on the extent of its transcription. One of  these additional control sites is a secondary promoter, designated trpP2, which is located within the coding sequence of the trpD gene (shown in Figure 19.7). This  promoter is not regulated by the trp repressor. Transcription from it occurs constitutively at a relatively low rate and is terminated at the same location as transcription  from the regulated promoter for the whole operon, trpP. The resulting transcription product from trpP2 is an mRNA that contains the coding sequences for trpCBA,  the last three genes of the operon. Therefore two polycistronic mRNAs are derived from the trp operon, one containing all five structural genes and one possessing  only the last three genes. Under conditions of maximum repression the basal level of mRNA coding sequence for the last three genes is about five times higher than the  basal mRNA level for the first two genes. The reason for a second internal promoter is unclear. Perhaps the best alternative comes from the observation that three of the five proteins do not contain tryptophan;  only the trpB and trpC genes contain the single codon that specifies tryptophan. Therefore, under extreme tryptophan starvation, these two proteins would not be  synthesized, which would prevent the pathway from being activated. However, since both of these genes lie downstream of the unregulated second promoter, their  protein products will always be present at the basal level necessary to maintain the pathway. Tryptophan Operon Has a Second Control Site:  The Attenuator Site Another important control element of the trp operon not present in the lac operon is the attenuator site (Figure 19.9). It lies within 162 nucleotides between the start  of transcription from trpP and the initiator codon of the trpE gene. Its existence was first deduced by the identification of mutations that mapped in this region and  increased transcription of all five structural genes. Within the 162 nucleotides, called the leader sequence, are 14 adjacent codons that begin with a methionine codon  and end with an in­phase termination codon. These codons are preceded by a canonical ribosome­binding site and

Figure 19.9  Nucleotide sequence of leader RNA from trp operon.  The 14 amino acids of the putative leader peptide are indicated over their codons.  Redrawn with  permission from Oxender, D. L., Zurawski, G., and Yanofsky, C.  Proc. Natl. Acad. Sci. USA 76:5524, 1979.

 

  

Page 811

could potentially specify a 14­residue leader peptide. This peptide has never been detected in bacterial cells, perhaps because it is degraded very rapidly. The  ribosome­binding site does function properly when its corresponding DNA sequence is ligated upstream of a structural gene using recombinant DNA techniques. The attenuator region provides RNA polymerase with a second chance to stop transcription if the trp enzymes are not needed by the cell. In the presence of  tryptophan, it acts like a rho­independent transcription termination site to produce a short 140­nucleotide transcript. In the absence of tryptophan, it has no effect on  transcription, and the entire polycistronic mRNA of the five structural genes is synthesized. Therefore, at both the operator and attenuator, tryptophan exerts the same  general influence. At the operator it participates in repressing transcription, and at the attenuator it participates in stopping transcription by those RNA polymerases that  have escaped repression. It has been estimated that attenuation has about a 10­fold effect on transcription of the trp structural genes. When multiplied by the 70­fold  effect of derepression at the operator, about a 700­fold range exists in the level at which the trp operon can be transcribed. The molecular mechanism by which transcription is terminated at the attenuator site is a marvelous example of cooperative interaction between bacterial transcription  and translation to achieve desired levels of a given mRNA. The first hints that ribosomes were involved in the mechanism of attenuation came from the observation that  mutations in the gene for tRNATrp synthetase (the enzyme that charges the tRNA with tryptophan) or the gene for an enzyme that modifies some bases in the tRNA  prevent attenuation. Therefore a functional tRNATrp must participate in the process. The leader peptide (Figure 19.9) of 14 residues contains two adjacent tryptophans in positions 10 and 11. This is unusual because tryptophan is a relatively rare amino  acid in E. coli. It also provides a clue about the involvement of tRNATrp in attenuation. If the tryptophan in the cell is low, the amount of charged tRNATrp will also be  low and the ribosomes may be unable to translate through the two trp codons of the leader peptide region. Therefore they will stall at this place in the leader RNA  sequence.

Figure 19.10  Schematic diagram showing the proposed  secondary structures in trp leader RNA from  E. coli.  Four regions can base pair to form  three stem­and­loop structures. These are  shown as 1–2, 2–3, and 3–4.  Reproduced with permission from Oxender,  D. L., Zurawski, G., and Yanofsky, C. Proc.  Natl. Acad. Sci. USA 76:5524, 1979.

It turns out that the RNA sequence of the attenuator region can adopt several possible secondary structures (Figure 19.10). The position of the ribosome within the  leader peptide­coding sequence determines the secondary structure that will form. This secondary structure, in turn, is recognized (or sensed) by the RNA polymerase  that has just transcribed through the attenuator coding region and is now located a small distance downstream. The RNA secondary structure that forms when a  ribosome is not stalled at the trp codons is a termination signal for the RNA polymerase. Under these conditions the cell does not need to make tryptophan, and  transcription stops after the synthesis of a 140­nucleotide transcript, which is quickly degraded. On the other hand, the secondary structure that results when the  ribosomes are stalled at the trp codons is not recognized as a termination signal, and the RNA polymerase continues on into the trpE gene. Figure 19.11 shows these  different secondary structures in detail. The structure in Figure 19.11a shows the situation when a ribosome does not stall at the two tandem trp codons, UGG­UGG, near the beginning of region 1, but  instead moves on to region 2. When the ribosome is in region 2, regions 1 and 2 cannot base pair but regions 3 and 4 can form base pairs, resulting in a hairpin loop  followed by eight U residues, a structure common to sequences that signal transcription termination. Thus when the leader RNA sequence is being synthesized in the  presence of sufficient tryptophan (and charged tryptophanyl­tRNATrp), it is likely that a loop between regions 3 and 4 will occur and be recognized as a signal for  termination by the RNA polymerase. A different structure occurs if the ribosome is stalled at the trp codons and

 

  

Page 812

Figure 19.11  Schematic diagram showing the model for attenuation in the trp operon of E. coli.  (a) Under conditions of excess tryptophan, the ribosome (green sphere) translating  the newly transcribed leader RNA will synthesize the complete leader peptide.  During this synthesis the ribosome will bind to regions 1 and 2 of the RNA and prevent  formation of stem and loop 1–2 or 2–3. Stem and loop 3–4 will be free to form and  signal the RNA polymerase molecule (not shown) to terminate transcription.  (b) Under conditions of tryptophan starvation, tryptophanyl­tRNATrp will be limiting,  and the ribosome will stall at the adjacent trp codons at the beginning of region 1 in  the leader peptide­coding region. Because region 1 is bound to the ribosomes, stem  and loop 2–3 will form, excluding formation of stem and loop 3–4, which is required as  the signal for transcription termination. Therefore RNA polymerase will continue  transcription into the structural genes.  (c) Under conditions in which the leader peptide is not translated, stem and loop 1–2  will form, preventing formation of stem and loop 2–3, and thereby permit formation  of stem and loop 3–4. This will signal transcription termination.  Reproduced with permission from Oxender, D. L., Zurawski, G., and Yanofsky,  C. Proc. Natl. Acad. Sci. USA 76:5524, 1979.

region 1 is prevented from base pairing with region 2 (Figure 19.11b). Under these circumstances, region 2 now can base pair with region 3. This region 2 and 3  hairpin ties up the sequence complementary to region 4, so that region 4 remains single stranded. Therefore the region 3 and 4 hairpin loop that serves as the  termination signal does not form, and the RNA polymerase continues on with its transcription. Thus for transcription to proceed past the attenuator, region 1 must be  prevented from pairing with region 2. This is accomplished if the ribosome stalls in region 1 due to an insufficient amount of charged tryptophan­tRNA for translation of  the leader peptide to continue beyond two trp codons. When this happens, region 1 is bound within the ribosome and cannot pair with region 2. Since regions 2 and 3  are synthesized before region 4, they, in turn, will base pair before region 4 appears in the newly transcribed RNA. Therefore region 4 remains single stranded, the  termination hairpin does not form, and RNA polymerase continues transcription into the structural genes. Since the two trp codons occur in region 1, if the ribosome happens to stall at an earlier codon in the leader sequence, it will have little effect on attenuation. For  example, starvation for lysine, valine, or glycine would be expected to reduce the amount of the corresponding charged tRNA and stall the ribosome at that codon, but  a deficiency in these amino acids has no effect on transcription of the trp operon. An exception is arginine whose codon occurs immediately after the trp codons.  Starving for arginine does attenuate transcription termination somewhat, probably because of ribosome stalling at this codon, but to less of an extent than a deficiency  in tryptophan. Cis­acting mutations in the attenuator region support this alternate hairpin model. Most of these mutations result in increased transcription because they disrupt base  pairing in the double­stranded portion of the termination hairpin

 

  

Page 813

Figure 19.12  Leader peptide sequences specified by biosynthetic operons of E. coli.  All contain multiple copies of amino acid(s) synthesized by enzymes coded for by operon.

and render it less stable. Some mutations, however, increase termination at the attenuator. One of these interferes with base pairing between regions 2 and 3, allowing  region 3 to be available for pairing with region 4 even when region 1 is bound to a stalled ribosome. Another mutation occurs in the AUG initiator codon for the leader  peptide so that the ribosome cannot begin its synthesis. Transcription Attenuation Is a Mechanism of Control in Operons for Amino Acid Biosynthesis Attenuation is a common phenomenon in bacterial gene expression; it occurs in at least six other operons that code for enzymes catalyzing amino acid biosynthetic  pathways. Figure 19.12 shows the corresponding leader peptide sequences specified by each of these operons. In each case, the leader peptide contains several  codons for the amino acid product of the biosynthetic pathway. The most extreme case is the 16­residue leader peptide of the histidine operon that contains seven  contiguous histidines. Starvation for histidine results in a decrease in the amount of histidinyl­tRNAHis and a dramatic increase in transcription of the his operon. As with  the trp operon, this effect is diminished by mutations that interfere with the level of charged histidinyl­tRNAHis. Furthermore, the nucleotide sequence of the attenuator  region suggests that ribosome stalling at the histidine codons also influences the formation of alternate hairpin loops, one of which resembles a termination hairpin  followed by several U residues. In contrast to the trp operon, transcription of the his operon is regulated entirely by attenuation; it does not possess an operator that is  recognized by a repressor protein. Instead, the ribosome acts rather like a positive regulator protein, similar to the cAMP–CAP complex discussed with the lac  operon. If the ribosome is bound to (i.e., stalled at) the attenuator site, then transcription of the downstream structural genes is enhanced. If the ribosome is not bound,  then transcription of these genes is greatly reduced. Transcription of the other operons shown in Figure 19.12 can be attenuated by more than one amino acid. For example, the thr operon is attenuated by either  threonine or isoleucine; the ilv operon is attenuated by leucine, valine, or isoleucine. This effect can be explained in each case by stalling of the ribosome at the  corresponding codon, which, in turn, interferes with the formation of a termination hairpin. Although not proved, it is possible that in the cases of the longer leader  peptides, stalling at more than one codon is necessary to achieve maximal transcription through the attenuation region. 19.5—  Other Bacterial Operons Synthesis of Ribosomal Proteins Is Regulated in a Coordinated Manner Many other bacterial operons have been studied and found to possess the same general regulatory mechanisms as the lac, trp, and his operons, as discussed

 

  

Page 814 Operon

Regulator  protein

Proteins specified by the operon

Spc

S8

L14­L24­L5­S14­S8­L6­L18­S5­L15­L30

S10

L4

S10­L3­L2­L4­L23­S19­L22­S3­S17­L16­L29

str

S7

S12­S7­EF•G­EF•Tu

a

S4

S13­S11­S4­a­L17

L11

L1

L11­L1

rif

L10

L10­L7­b­

  Figure 19.13  Operons containing genes for ribosomal proteins E. coli.  Genes for the protein components of the small (S) and large (L) ribosomal  subunits of E. coli are clustered on several operons. Some of these operons also contain genes for RNA polymerase subunits a, b, and  , and protein  synthesis factors EF∙G and EF∙Tu. At least one of the protein products  of each operon usually regulates expression of that operon (see text).

in Section 19.4. However, each operon has evolved its own distinctive quirks. For example, one interesting group of operons are those containing the structural genes  for the 70 or more proteins that comprise the ribosome (Figure 19.13). Each ribosome contains one copy of each ribosomal protein (except for protein L7­L12,  which is probably present in four copies). Therefore all 70 proteins are required in equimolar amounts, and it makes sense that their synthesis is regulated in a  coordinated fashion. Characterization of this set of operons is not yet complete, but six operons, containing about one­half of the ribosomal protein genes, occur in two  major gene clusters. One cluster contains four adjacent operons (str, Spc, S10, and a), and the other two operons are near each other elsewhere in the E. coli  chromosome. There is no obvious pattern to distribution of these genes among different operons. Some operons contain genes for proteins of just one ribosomal  subunit; others code for proteins of both subunits. In addition to structural genes for ribosomal proteins, these operons also contain genes for other (related) proteins.  For example, str operon contains genes for the two soluble translation elongation factors, EF∙Tu and EF∙G, as well as genes for some proteins in the 30S  ribosomal subunit. The a  operon has genes for proteins of both 30S and 50S ribosomal subunits plus a gene for one of the subunits of RNA polymerase. The rif  operon has genes for two other protein subunits of RNA polymerase and genes for ribosomal proteins. A common theme among the six ribosomal operons is that their expression is regulated by one of their own structural gene products; that is, they are self­regulated.  The precise mechanism of this self­regulation varies considerably with each operon and is not yet understood in detail. However, in some cases the regulation occurs at  the level of translation, not transcription as discussed for the lac and trp operons. After the polycistronic mRNA is made, the "regulatory" ribosomal protein binds to  this mRNA and determines which regions, if any, are translated. In general, the ribosomal protein that regulates expression of its own operon, or part of its own  operon, is a protein that is associated with one of the ribosomal RNAs (rRNAs) in the intact ribosome. This ribosomal protein has a high affinity for the rRNA and a  lower affinity for one or more regions of its own mRNA. Therefore a competition between the rRNA and the operon's mRNA for binding with the ribosomal protein  occurs. As the ribosomal protein accumulates to a higher level than the free rRNA, it binds to its own mRNA and prevents the initiation of protein synthesis at one or  more of the coding sequences on this mRNA (Figure 19.14). As more ribosomes are formed, the excess of this particular ribosomal protein is used up and translation  of its coding sequence on the mRNA can begin again.

 

  

Page 815

Figure 19.14  Self­regulation of ribosomal protein synthesis.  If free rRNA is not available for assembly of new ribosomal subunits, individual  ribosomal proteins bind to polycistronic mRNA from their own operon,  blocking further translation.

Stringent Response Controls Synthesis of rRNAs and tRNAs Bacteria have several ways in which to respond molecularly to emergency situations; that is, times of extreme general stress. One of these situations is when the  bacterium does not have a sufficient pool of amino acids to maintain protein synthesis. Under these conditions the cell invokes what is called the stringent response, a  mechanism that reduces the synthesis of the rRNAs and tRNAs about 20­fold. This places many of the activities within the cell on hold until conditions improve. The  mRNAs are less affected, but there is also about a three­fold decrease in their synthesis.

Figure 19.15  Stringent control of protein synthesis in E. coli.  During extreme amino acid starvation, an  uncharged tRNA in the A site of the ribosome  activates the relA protein to synthesize ppGpp  and pppGpp, which, in turn, are involved in  decreasing transcription of the genes coding  for rRNAs and tRNAs.

The stringent response is triggered by the presence of an uncharged tRNA in the A site of the ribosome. This occurs when the concentration of the corresponding  charged tRNA is very low. The first result, of course, is that further peptide elongation by the ribosome stops. This event causes a protein called the stringent factor,  the product of the relA gene, to synthesize guanosine tetraphosphate (ppGpp) and guanosine pentaphosphate (pppGpp), from ATP and GTP or GDP as shown  in Figure 19.15. Stringent factor is loosely associated with a few, but not all, ribosomes of the cell. Perhaps a conformational change in the ribosome is induced by  occupation of the A site by an uncharged tRNA, which, in turn, activates the associated stringent factor. The exact functions of ppGpp and pppGpp are unknown.  However, they seem to inhibit transcription initiation of the rRNA and tRNA genes. In addition they affect transcription of some operons more than others.

 

  

Page 816

CLINICAL CORRELATION 19.1  Transmissible Multiple Drug Resistances Pathogenic bacteria are becoming increasingly resistant to a large number of antibiotics,  which is viewed with alarm by many physicians. Many cases have been documented in  which a bacterial strain in a patient being treated with one antibiotic suddenly became  resistant to that antibiotic and, simultaneously, to several other antibiotics even though the  bacterial strain had never been previously exposed to these other antibiotics. This occurs  when the bacteria suddenly acquire from another bacterial strain a plasmid that contains  several different transposons, each containing one or more antibiotic­resistance genes.  Examples include the genes encoding b ­lactamase, which inactivates penicillins and  cephalosporins, chloramphenicol acetyltransferase, which inactivates chloramphenicol,  and phosphotransferases, which modify aminoglycosides such as neomycin and  gentamycin. Neu, H. C. The crisis in antibiotic resistance. Science 257: 1064, 1992.   19.6—  Bacterial Transposons Transposons Are Mobile Segments of DNA

Figure 19.16  General structure of transposons.  Transposons are relatively rare mobile  segments of DNA that contain genes  coding for their own rearangement and  (usually) genes that specify resistance to  various antibiotics.

So far we have only discussed the regulation of bacterial genes whose locations are fixed in the chromosome. Their positions relative to the neighboring genes do not  change. The vast majority of bacterial genes are of this type. In fact, genetic maps of E. coli and Salmonella typhimurium are quite similar, indicating the lack of  much evolutionary movement of most genes within the bacterial chromosome. There is a class of bacterial genes, however, in which newly duplicated gene copies  ''jump" to another genomic site with a frequency of about 10–7 per generation, the same rate as spontaneous point mutations occur. The mobile segments of DNA  containing these genes are called transposable elements or transposons (Figure 19.16). Transposons were first detected as rare insertions of foreign DNA into  structural genes of bacterial operons. Usually, these insertions interfere with the expression of the structural gene into which they have inserted and all downstream  genes of the operon. This is not surprising since they can potentially destroy the translation reading frame, introduce transcription termination signals, affect the mRNA  stability, and so on. Many transposons and the sites into which they insert have been isolated using recombinant DNA techniques and have been extensively  characterized. These studies have revealed many interesting features about the mechanisms of transposition and the nature of genes located within transposons. Transposons vary tremendously in length. Some are a few thousand base pairs and contain only two or three genes; others are many thousands of base pairs long,  containing several genes. Several small transposons can occur within a large transposon. All active transposons contain at least one gene that codes for a transposase, an enzyme required for the transposition event. Often they contain genes that code for resistance to antibiotics or heavy metals. Most transpositions involve generation  of an addition copy of the transposon and insertion of this copy into another location. The original transposon copy is the same after the duplication as before; that is,  the donor copy is unaffected by insertion of its duplicate into the recipient site. Transposons contain short inverted terminal repeat sequences that are essential for  the insertion mechanism, and in fact these inverted repeats are often used to define the two boundaries of a transposon. The multiple target sites into which most  transposons can insert seem to be fairly random in sequence; other transposons have a propensity for insertion at specific "hot spots." The duplicated transposon can  be located in a different DNA molecule than its donor. Frequently, transposons are found on plasmids that pass from one bacterial strain to another and are the source  of a suddenly acquired resistance to one or more antibiotics by a bacterium (Clin. Corr. 19.1). As with bacterial operons, each transposon or set of transposons has its own distinctive characteristics. The well­characterized transposon Tn3 will be discussed as an  example of their general properties. Tn3 Transposon Contains Three Structural Genes The transposon Tn3 has been cloned using recombinant DNA techniques and its complete sequence determined. It contains 4957 base pairs including 38 base pairs  at one end that occur as an inverted repeat at the other end (Figure 19.17). Three genes are present in Tn3. One gene codes for the enzyme b ­lactamase, which  hydrolyzes ampicillin and renders the cell resistant to this antibiotic. The other two genes, tnpA and tnpR, code for a transposase and a repressor protein, respectively.  The transposase has 1021 amino acids and binds to single­stranded DNA. Little else is known about its action, but it is thought

 

  

Page 817

Figure 19.17  Functional components of the transposon Tn3.  Genetic analysis shows there are at least four kinds of  regions: the inverted repeat termini; a gene for the enzyme  b­lactamase, which confers resistance to ampicillin and related  antibiotics; a gene encoding an enzyme required for  transposition (transposase); and a gene for a repressor  protein that controls transcription of genes for transposase  and for repressor itself. The horizontal arrows indicate direction  in which DNA of various regions is transcribed.  Redrawn from Cohen, S. N., and  Shapiro, J. A. Sci. Am. 242:40, 1980. W. H.  Freeman and Company, Copyright © 1980.

to recognize the repetitive ends of the transposon and to participate in the cleavage of the recipient site into which the new transposon copy inserts. The tnpR gene  product is a protein of 185 amino acids. In its role as a repressor it controls transcription of both the transposase gene and its own gene. The tnpA and tnpR genes are  transcribed divergently from a 163 base pair control region located between the two genes that is recognized by the repressor. The tnpR product also participates in  the recombination process that results in the insertion of the new transposon. Transcription of the ampicillin­resistance gene is not affected by the tnpR gene product. Mutations in the transposase gene generally decrease the frequency of Tn3 transposition, demonstrating its direct role in the transposition process. Mutations that  destroy the repressor function of the tnpR product cause an increased frequency of transposition. These mutations derepress the tnpA gene, resulting in more  molecules of the transposase, which increases the formation of more transposons. They also derepress the tnpR gene but, since the repressor is inactive, this has no  effect on the system. When a transposon, containing its terminal inverted repeats, inserts into a new site, it generates short (5–10 bp) direct repeats of the sequences at the recipient site that  flank the new transposon. This is due to the mechanism of recombination that occurs during the insertion process (Figure 19.18). The first step is the generation of  staggered nicks at the recipient sequence. These staggered single­strand, protruding 5  ends then join covalently to the inverted repeat ends of the transposon. The  resulting intermediate resembles two replicating forks pointing toward each other and separated by the length of the transposon. The replication machinery of the cell  fills in the gaps and continues the divergent elongation of the two primers through the transposon region. This ultimately results in two copies of the transposon  sequence. Reciprocal recombination within the two copies regenerates the original transposon copy at its original position and completes the process of forming a new  copy at the recipient site that is flanked by direct repeats of the recipient sequence. The practical importance of transposons located on plasmids has taken on increased significance for the use of antibiotics in treatment of bacterial infections. Plasmids  that have not been altered for experimental use in the laboratory usually contain genes that facilitate their transfer from one bacterium to another. As the plasmids  transfer (e.g., between different infecting bacterial strains), their transposons containing antibiotic­resistance genes are moved into new bacterial strains. Once inside  a new bacterium, the transposon can be duplicated onto the chromosome and become permanently established in that cell's lineage.

 

  

Page 818

Figure 19.18  Proposed molecular pathway for transposition and chromosome rearrangements.  Donor DNA, including the transposon, shown in red, recipient DNA contains  small light green. The pathway has four steps, beginning with staggered,  single­strand cleavages (Step 1a) at each end of the transposable element and  at each end of the "target" nucleotide sequence to be duplicated. The  cleavages expose (Step 1b) the DNA strand ends involved in the next step:  the joining of DNA strands from donor and recipient molecules in such a  way that the double­stranded transposable element has a DNA replication fork  at each end (Step 2). DNA synthesis (Step 3) replicates transposon (red bars)  and target sequence (light green squares), accounting for the observed  duplication. This step forms two new complete double­stranded molecules;  each copy of the transposable element joins a segment of the donor molecule  and a segment of the recipient molecules. (Copies of the element serve as  linkers for the recombination of two unrelated DNA molecules.) In the final  Step 4, reciprocal recombination between copies of the transposable  element inserts the element at a new genetic site and regenerates the  donor molecule.  Redrawn from Cohen, S. N., and Shapiro, J. A. Sci. Am. 252:40, 1980.  W. H. Freeman and Company, Copyright © 1980.

The result is that more and more pathogenic bacterial strains become resistant to an increasing number of antibiotics. 19.7—  Inversion of Genes in Salmonella A different mechanism of differential gene regulation has been discovered for one set of genes in Salmonella. Similar control mechanisms exist for the expression of  other genes in other prokaryotes (e.g., a bacteriophage called  ). Bacteria move by waving their flagella that are composed predominantly of subunits of a protein called flagellin. Many Salmonella species possess two different  flagellin genes and express only one of these genes at a time. Bacteria are said to be in phase 1 if they are expressing the H1 flagellin gene and in phase 2 if they are  expressing the H2 flagellin gene. A bacterial clone in one phase switches to the other phase about once every 1000 divisions. This switch is called phase variation,  and its occurrence is controlled at the level of transcription of H1 and H2 genes. Organization of the flagellin genes and their regulatory elements are shown in Figure 19.19. A 995­bp segment of DNA flanked by 14­bp repeats is adjacent to the H2 gene and a rhl gene that codes for a repressor of H1. The H2 and

 

  

Page 819

Figure 19.19  Organization of the flagellin genes of Salmonella.  Orientation of a 995­bp DNA segment flanked by 14­bp inverted repeats  (IRL and IRR) controls the expression of H1 and H2 flagellin genes. In phase 2,  transcription initiates at promoter P within the invertible segment and  continues through H2 and rh1 genes. In phase 1, the orientation is reversed  so that transcription of H2 and rh1 genes does not occur.

rhl genes are coordinately transcribed. Therefore, when H2 is expressed, the repressor is also made and turns off H1 expression. When H2 protein and the repressor  are not made, the H1 gene is derepressed and H1 synthesis occurs. The promoter for the operon containing H2 and rhl lies near one end of the 995­bp segment, just inside one copy of the 14­bp repeats. This segment can undergo  inversions between the 14­bp repeats. In one orientation of the segment, the promoter is upstream of the H2–rhl transcription unit; in the other orientation it points  toward the opposite direction so that H2 and rhl are not transcribed. In addition to containing this promoter, the invertible segment of DNA possesses the hin gene  whose product is an enzyme that catalyzes the inversion event itself. The hin gene seems to be transcribed constitutively at a low rate. Mutations in hin reduce the rate  of inversion by 10,000­fold. Therefore phase variation is controlled by physical inversion of the segment of DNA that removes a promoter from its position in front of  the H2–rhl operon. When the promoter is in the opposite direction, it presumably still initiates transcription, but the fate of that RNA is unknown. It does not initiate  transcription of the H1 that maps in this direction. That gene apparently has its own promoter controlled directly by the rhl repressor. Inversion of the hin segment probably occurs via recombination between the 14­bp inverted repeats that is similar to recombination events involved in the transposition  of a transposon. In fact, transposons do invert relative to their flanking sequences in a fashion exactly analogous to the hin inversion. Furthermore, the amino acid  sequence of the hin product shows considerable similarity to that of the tnpR product of the Tn3 transposon, which participates in the integration of the transposon  into a new site. Thus it is possible, and even likely, that the two processes are evolutionarily related.

 

  

Page 820

19.8—  Organization of Genes in Mammalian DNA The past 20 years have seen a virtual explosion of new information about the organization, structure, and regulation of genes in eukaryotic organisms. The reason for  this enormous increase in our knowledge about eukaryotic genes has been the concurrent development of recombinant DNA techniques and DNA sequencing  techniques (Chapter 18). Experiments undreamed of a few years ago are now routine accomplishments. The human haploid genome contains 3 × 109 bp of DNA, about 1000 times more DNA than the E. coli chromosome. All available evidence suggests that each of the  23 haploid chromosomes in the human genome has a single unique DNA molecule. Since the distance between two adjacent base pairs is 3.4 × 10–10 meters (3.4 Å),  if these 23 human chromosomal DNA molecules were stretched out end­to­end, they would extend about 1 meter. Each mammalian cell contains virtually a complete  copy of this genome, and all except the haploid germline cells contain two copies. Different types of mammalian cells express widely different proteins even though each contains the same complement of genes. In addition, widely different patterns of  protein synthesis occur at different developmental stages of the same type of cells. Therefore extremely intricate and complicated mechanisms of regulation for these  genes must exist, and, in fact, these mechanisms are not understood for even one mammalian gene to the extent that they are understood for many bacterial operons.  Despite the great advances of the past 20 years, our understanding of gene regulation in mammals, and indeed all eukaryotes, remains fragmentary at best and  probably is still very naive. Only a Small Fraction of Eukaryotic DNA Codes for Proteins It was appreciated even before the advent of recombinant DNA methodology that eukaryotic cells, including mammalian cells, contain far more DNA than seems  necessary to code for all of the required proteins. Furthermore, organisms that appear rather similar in complexity can have a several­fold difference in cellular DNA  content. A housefly, for example, has about six times the cellular DNA content of a fruitfly. Some plant cells have almost ten times more DNA than human cells.  Therefore DNA content does not always correlate with the complexity and diversity of functions of the organism. It is difficult to obtain an accurate estimate of the number of different proteins, and therefore genes, in a mammalian cell or in the entire mammalian organism. However,  nucleic acid hybridization procedures indicate that a maximum of 5000–10,000 different mRNA species may be present in a mammalian cell at a given time. Most of  these mRNAs code for proteins that are common to many cell types. Therefore a generous estimate is that there are approximately 100,000 genes for the entire  mammalian genome. If the average coding sequence is 1500 nucleotides (specifying a 500 amino acid protein), this accounts for 5% of the mammalian genome. DNA  regulatory elements, repetitive genes for rRNAs, and so on may account for another 5–10%. However, as much as 85–90% of the mammalian genome may not have  a direct genetic function. This remarkable conclusion is in contrast to the bacterial genome in which virtually all of the DNA is consumed by genes and their regulatory  elements. Eukaryotic Genes Usually Contain Intervening Sequences (Introns) As discussed in Chapter 16, coding sequences (exons) of eukaryotic genes are frequently interrupted by intervening sequences or introns that do not code for a  product. These introns are transcribed into a precursor RNA species found in the nucleus and are removed by RNA splicing events during the processing of the  nuclear precursor RNA to the mature mRNA in the cytoplasm.

 

  

Page 821

The number and length of the introns in a gene can vary tremendously. Histone genes and interferon genes lack introns; they contain a continuous coding sequence for  the protein as do bacterial genes. The mammalian collagen gene, on the other hand, has more than 50 different introns that collectively consume 90% of the gene. The  largest human gene discovered to date is 2400 kb, or more than half the size of the entire E. coli genome of 4000 kb. This gene contains 79 introns of about 30­kb  average size and encodes a 427­kDa muscle protein called dystrophin (Figure 19.20). Despite the fact that dystrophin is a very large protein, the dystrophin gene's  introns consume more than 99% of the gene's length. Mutations in this huge dystrophin gene are responsible for Duchenne/Becker muscular dystrophy (see Clin.  Corr. 19.2). On the basis of the many mammalian genes analyzed to date, it appears that most have three or four introns and that the presence of 50 or more introns in  a single gene represents an extreme case. Nevertheless, introns of genes clearly account for some of the "excess" DNA present in eukaryotic genomes. The significance of introns and their potential biological functions, if any, are the subject of much speculation and experimentation. In a few genes, including those for  the a ­ and b ­globin subunits of hemoglobin (see below), introns separate the coding regions for functional domains of the protein. In many other genes, however, no  obvious correlation exists between the intron positions of a gene and the three­dimensional domains of its encoded protein. In fact, the number of introns in a given  gene sometimes is not the same in different mammalian species, or even within a single species. For example, the rat haploid genome has two insulin genes, one with  two introns and one with a single intron. The haploid genomes of other rodents have a single insulin gene with two introns. One widely quoted hypothesis for the possible function of introns is that they may have served to facilitate the mixing and matching of exons during the course of  evolution so that occasionally new protein­encoding genes are created, which provide a selective advantage for the organism. Some circumstantial evidence exists to  support this possibility. For example, chicken collagen has a larger number of repeating Gly­X­Y triplets and most of the exons in its genes are multiples of 9 bp (i.e.,  45, 54, 99, 108, or 162 bp per exon) beginning with a glycine codon and ending with a Y codon. Thus the collagen gene may have evolved via multiple duplications of  an exon–intron unit. Genes of unicellular lower eukaryotes, such as yeast, have either no introns or a small number of introns that tend to be short compared to introns  of higher eukaryotes. Perhaps these lower eukaryotes, which reproduce much faster than do higher organisms, have to be more efficient in their DNA and RNA  metabolism and

Figure 19.20  Human dystrophin gene and its protein.  (a) The 79 exons (dark thin vertical lines) of human dystrophin gene span 2.4 × 106 bp  (2400 kb), more than one­half the length of the E. coli genome. The average dystrophin  exon is 140 bp and the average dystrophin intron (light gray background regions) is  more than 30,000 bp.  (b) Dystrophin (427 kDa) has 3685 amino acids. It contains an actin­binding domain  blue, 24 tandem repeats of about 109 amino acids that likely form a rod­like domain  (green), a cysteine­rich domain (purple), and a C terminus that may associate with the  membrane (red).  Redrawn from Ahn, A. H., and Kunkel, L. M. Nature Genetics 3:283, 1993.

 

  

Page 822

CLINICAL CORRELATION 19.2  Duchenne/Becker Muscular Dystrophy and the Dystrophin Gene Both Duchenne muscular dystrophy (DMD) and the milder Becker muscular dystrophy  (BMD) are inherited as X­linked recessive diseases. They result in degenerative disorders  of skeletal muscle and are the most common of all lethal neuromuscular genetic diseases,  affecting 1 in 3500 males. They are associated with abnormally high levels of serum  creatine kinase levels from birth. Although most afflicted males inherit the defect from their  unaffected, heterozygous mother, 30% of the cases exhibit no previous family history and  appear to be "spontaneous" new mutations in the germline of either the mother or her  parents. Both forms of muscular dystrophy are caused by defects in the dystrophin gene  on the X chromosome. This gene is huge and complicated. It has 79 exons and spans 2.4  × 106 bp and encodes a membrane­associated cytoskeleton protein. Its expression is  regulated in a cell­specific and developmentally controlled manner from at least five  different promoters. Many mutations responsible for DMD and BMD are large deletions  that remove one or more of the 79 exons, but the size of the deletion does not necessarily  correlate with the severity of the disorder. In DMD patients, dystrophin is undetectable or  absent, whereas in BMD patients, it is reduced or altered. Genetic, biochemical, and  anatomical studies suggest that dystrophin may serve diverse roles in many other tissues  besides muscle. It is hoped that future studies of dystrophin may lead to an understanding  of the cause and perhaps a rational treatment for muscular dystrophy. Ahn, A. H., and Kunkel, L. M. The structural and functional diversity of dystrophin.  Nature Genetics 3:283, 1993.   cannot tolerate large numbers of large introns. In many ways, however, introns remain as big an enigma as when first discovered. 19.9—  Repetitive DNA Sequences in Eukaryotes Another curiosity about mammalian DNA, and the DNA of most higher organisms, is that, in contrast to bacterial DNA, it contains repetitive sequences in addition to  single copy sequences. This repetitive DNA falls into two general classes—highly repetitive simple sequences and moderately repetitive longer sequences of  several hundred to several thousand base pairs. Importance of Highly Repetitive Sequences Is Unknown The highly repetitive sequences range from 5 to about 300 bp and occur in tandem. Their contribution to the total genomic size is extremely variable, but in most  organisms they are repeated millions of times and in a few organisms they consume 50% or more of the total DNA. These highly repetitive sequences are sometimes  called satellite DNAs because when total DNA isolated from a eukaryote is sheared slightly and centrifuged in a CsCl gradient, they can be separated as "satellites"  of the bulk of the DNA on the basis of their differing buoyant densities. They are concentrated primarily at the centromeres and to a lesser extent at telomeres (i.e.,  ends of chromosomes). Figure 19.21 shows the three main repeat units of the highly repetitive sequences at the chromosomal centromeres of the fruitfly, Drosophila  virilis. Repeats of these three sequences of 7 bp comprise 41% of the organism's DNA. They are obviously related evolutionarily since two of the repeats can be  derived from the third by a single base pair change. Relatively little transcription occurs from the highly repetitive sequences, and their biological importance remains,  for the most part, a mystery (see Clin. Corr. 19.3). Those repetitive sequences that occur near the telomeres are probably required for the replication of the ends of  the linear DNA molecules. The ones at the centromeres might play a structural role since these regions attach to the microtubules of the mitotic spindle during  chromosome pairing and segregation in mitosis and meiosis. Highly repetitive sequences occur in human DNA at both centromeres and telomeres but their repeat units  at centromeres are longer and more variable in sequence than those of Drosophila virilis shown in Figure 19.21. A Variety of Repeating Units Are Defined as Moderately Repetitive Sequences The moderately repetitive sequences consist of a large number of different sequences repeated to such different extents that it is somewhat misleading to group them  under one heading. Some are clustered in one region of the genome; Genome  (%)

Number of  copies in  genome

Predominant  sequence

25

1 ×107

5   ­ACAAACT­  3 3   ­TGTTTGA­  5

8

3.6 ×106

8

3.6 ×106

   

  Figure 19.21  Main repeat units of repetitive sequences of the  fruitfly Drosophila virilis.  Approximately 41% of genomic DNA of Drosophila  virilis is comprised of three related repeat sequences of 7 bp. The bottom two sequences differ from the top  sequence at one base pair shown in box.

 

  

Page 823

CLINICAL CORRELATION 19.3  Huntington's Disease and Trinucleotide Repeat Expansions Huntington's disease is an autosomal dominant neurodegenerative disorder characterized  by increasing behavioral disturbance, involuntary movements, cognitive impairment, and  dementia. It can be inherited from either parent. Disease onset often does not occur until  age 40 and death results 10–15 years later from aspiration, trauma, or pneumonia. The  defective gene on chromosome 4 responsible for the disease is dominant over the normal  gene, suggesting the defect causes the gene's protein to gain a deleterious function. This  gene encodes a large protein called "huntingtin" that contains 3144 amino acids found in  many tissues but whose function is unknown. Near the beginning of the gene is a run of  CAGs that encodes a polyglutamine tract in huntingtin. The length of this polyglutamine  tract is 11–34 in normal individuals and 37–121 in Huntington's disease patients. The  larger the number of repeats, the sooner the onset of the disease. Furthermore, the child  of a parent with an abnormally large number of repeats will often have an even larger  number of repeats, resulting in a "genetic anticipation" of the disease. Neither the cause of  the trinucleotide repeat expansions nor the abnormal function of huntingtin with an  expanded polyglutamine is known. However, at least seven other neurological disorders  are caused by trinucleotide repeat expansions in other genes, including X­linked spinal  and bulbar muscular atrophy, fragile X syndrome, and myotonic dystrophy. The reason  for this neuronal toxicity is currently the subject of intense research. These diseases can  be diagnosed molecularly by tests based on the polymerase chain reaction. La Spada, A. R., Paulson, H. L., and Fischbeck, K. H. Trinucleotide repeat expansion in  neurological disease. Ann. Neurol. 36:814, 1994.   many are scattered throughout the DNA. Some moderate repeats are several thousand base pairs in length; other repeats come in a unit size of only a hundred base  pairs. Sometimes the sequence is highly conserved from one repeat to another; in other cases, different repeat units of the same basic sequence will have undergone  considerable divergence. Two examples from the human genome will be described. In mammalian cells the 18S, 5.8S, and 28S rRNAs are transcribed as a single precursor transcript that is subsequently processed to yield the mature rRNAs. In  humans the length of this precursor is 13,400 nucleotides, about one­half of which is comprised of the three mature rRNA sequences. Several posttranscriptional  cleavage steps remove the extra sequences from the ends and the middle of the precursor RNA, releasing the mature rRNA species. DNA that contains the rRNA  genes is a moderately repetitive sequence of about 43,000 bp of which 30,000 bp are nontranscribed spacer DNA. Clusters of this entire DNA unit occur on five  chromosomes. In total, there are about 280 repeats of this unit, which comprise about 0.3% of the total genome (Figure 19.22). The 5S rRNA genes are repeated  about 2000 times but in different clusters. The need for so many rRNA genes is because the rRNAs are structural RNAs. Each transcript from the gene yields only  one copy of each rRNA molecule. On the other hand, each mRNA molecule derived from a ribosomal protein gene can be translated repeatedly to give many protein  molecules. In contrast to tandemly repetitive rRNA genes clustered at a few chromosomal sites, most moderately repetitive sequences in the mammalian genome do not code for  a stable gene product and are interspersed with nonrepetitive sequences that occur only once or a few times in the genome. The average size of these interspersed  repetitive sequences is about 300 bp. Almost one­half of these sequences are members of a general family of moderately repetitive sequences called the Alu family  because they can be cleaved by the restriction enzyme AluI. There are about 300,000 Alu sequences scattered throughout the human haploid genome (on the high side  of being moderately repetitive). Individual members are related in sequence but are frequently not identical. Their average homology with a consensus sequence is  about 87%. Additional repeat symmetry occurs within an Alu sequence. The sequence appears to have arisen by tandem duplication of a 130­bp sequence with a 31­bp insertion  in one of the two adjacent repeats. Some members of the Alu family resemble bacterial transposons in that they are flanked by short direct repeats. This does not  prove that an Alu repeat can be duplicated and transposed to another site like true transposons, but it suggests that such events may occur. The biological function of Alu sequences is unknown. One suggestion is that they serve as multiple origins for the DNA replication during S phase, but more sequences  occur than seem necessary for this function. Alu sequences appear in the introns of some genes and are transcribed as part of large precursor RNAs in which the Alu  sequences are removed during RNA splicing. Other Alu sequences are transcribed into small RNA molecules whose function is unknown. All mammalian genomes  appear to have a counterpart to the human interspersed Alu sequence family although the size of the repeat and its distribution can vary considerably between species.

Figure 19.22  Repetitive sequence in human DNA for rRNA.  In human cells a single transcription unit of  13,400 nucleotides is processed to yield the 18S,  5.8S, and 28S rRNAs. About 280 copies of the  corresponding rRNA genes are clustered on five  chromosomes. Each repeat contains a nontranscribed  spacer region of about 30,000 bp.

 

  

Page 824

19.10—  Genes for Globin Proteins Recombinant DNA Technology Has Been Used to Clone Genes for Many Eukaryotic Proteins Many mammalian structural genes that have been cloned by recombinant DNA techniques specify proteins that either occur in large quantity in a specific cell type, such  as the globin subunits in the red blood cell, or after induction of a specific cell type, for example, growth hormone or prolactin in the pituitary. As a result, more is  known about the regulation of these genes than of other genes whose protein products occur at lower levels in many different cell types. Increasingly, however,  information is being gained about mammalian genes for "rare" proteins with low abundances in the cell. We will discuss organization, structure, and regulation of the  related members of two gene families—the genes for the globin subunits and the growth hormone­like proteins. The first step in characterizing a eukaryotic gene is usually to use recombinant DNA techniques to clone a complementary DNA (cDNA) copy of that gene's  corresponding mRNA. In fact, this is the reason that the most extensively studied mammalian genes code for the major proteins of specific cells; a large fraction of the  total mRNA isolated from these cells codes for protein of interest. Hemoglobin is comprised of two a ­globin subunits (141 amino acids) and two b ­globin subunits  (146 amino acids). Almost all of the mRNA isolated from immature red cells (reticulocytes) codes for these two subunits of hemoglobin. There are several experimental variations of the procedure for synthesizing double­stranded cDNA copies of isolated mRNA in vitro. As discussed in Chapter 18,  many different plasmid and viral DNA vectors are available for cloning the (passenger) cDNA molecules. Figure 19.23 shows one protocol for constructing and  cloning cDNAs prepared from mRNA of reticulocytes. A synthetic oligonucleotide composed of 12–18 residues of deoxythymidine is hybridized to the 3 ­polyadenylate tail of the mRNA and serves as a primer for  reverse transcriptase, an enzyme that copies an RNA sequence into a DNA strand in the presence of the four deoxynucleoside triphosphates. The resulting RNA– DNA heteroduplex is treated with NaOH, which degrades the RNA strand and leaves the DNA strand intact. The 3  end of the remaining DNA strand can then fold  back and serve as a primer for initiating synthesis of a second DNA strand at random locations by reverse transcriptase, the same enzyme used to synthesize the first  strand. The hairpin loop is then nicked by S1 nuclease, an enzyme that cleaves single­stranded DNA but has little activity against double­stranded DNA. The ends of  the resulting double­stranded cDNAs are ligated to small synthetic "linker" oligonucleotides that contain the recognition site for the restriction enzyme HindIII.  Digestion of the resulting DNA with HindIII generates DNA fragments that contain HindIII­specific ends. These fragments can be ligated into the HindIII site of a  plasmid, and when the resulting circular "recombinant" DNA species are incubated with E. coli in the presence of cations such as calcium or rubidium, a few molecules  will be taken up by the bacteria. The incorporated recombinant DNAs will be replicated and maintained in the progeny of the original transformed bacterial cell. The collection of cloned cDNAs synthesized from the total mRNA in a given tissue or cell type is called a cDNA library, for example, a liver cDNA library or a  reticulocyte cDNA library. Since most of the mRNAs of a reticulocyte code for either a ­ or b ­globin, it is relatively easy to identify these globin cDNAs in a  reticulocyte cDNA library using procedures discussed in Chapter 16. Once identified, the nucleotide sequences of the cDNAs can be determined to confirm that they  do code for the known amino acid sequences of the a ­ and b ­globins. In cases in which the amino acid sequence of the protein is not known, other procedures  (sometimes immunological) are used to confirm the identification of the desired cDNA clone.

 

  

Page 825

Figure 19.23  Cloning of globin cDNA.  Step 1: Total RNA is extracted from red blood cells.  Step 2: The total RNA is passed through an oligode­oxythymidylate cellulose  column, which separates polyadenylated mRNA (see Chapter 15) from rRNA  and tRNA. Polyadenylated mRNA of red blood cells contains predominantly  hemoglobin mRNA.  Step 3: The mRNA is reverse­transcribed into first­strand cDNA using reverse  transcriptase, the viral enzyme that synthesizes DNA from RNA templates  (see Chapter 15).  Step 4: The mRNA is hydrolyzed with alkali whereas the DNA is unaffected.  Step 5: The single­stranded cDNA is converted into double­stranded DNA  by reverse transcriptase.  Step 6: The resulting double helix contains a single­stranded hairpin loop that  is removed by S1 nuclease, an enzyme that hydrolyzes single­stranded DNA.  Step 7: The cDNA is now a double helix with A­T base pairs at one end.  To generate cohesive ends for the ligation of this cDNA into a plasmid, a  chemically synthesized decanucleotide is attached to both ends using DNA  ligase from bacteriophage T4. This decanucleotide contains the sequence  recognized by HindIII restriction nuclease.  Step 8a: Treatment with HindIII produces a cDNA molecule with HindIII  cohesive ends.  Step 8b: The plasmid pUC9, which contains an ampicillin­resistance gene,  is cleaved with HindIII and exposed to bacterial alkaline phosphatase, an  enzyme that removes the phosphates from the cleaved 5 ­terminal ends of  the plasmids at the HindIII site. This prevents the cleaved plasmid from  recircularizing without the insertion of the cDNA.  Step 9: The linear plasmid and the cDNA molecules are mixed with T4 DNA  ligase, and circular, dimeric, "recombinant" DNA molecules are formed.  Step 10: This ligation mixture is used to transform E. coli.  Step 11: Individual E. coli cells that take up the plasmid are selected by  their ability to grow on ampicillin. The globin cDNA is confirmed by  determining  the nucleotide sequence of the small DNA fragment released  from the plasmid DNA by HindIII; if the observed nucleotide sequences  corresponded to those expected based on the known amino acid sequence  of a­ and b­globin, then the cDNA is identified.

Comparison of the a ­ and b ­globin cDNA sequences with the corresponding globin genes, which have also been cloned using recombinant DNA techniques, reveals  that all members of both sets of genes contain two introns at approximately the same positions relative to the coding sequences (Figure 19.24). The a  (and a ­like)  genes have an intron of 95 bp between codons 31 and 32 and a second intron of 125 bp between codons 99 and 100. The b  (and b ­like) genes have introns of 125– 150 bp and 800–900 bp located between codons 30

 

  

Page 826

Figure 19.24  Structures of human globin genes.  Structures for the human a­like and b­like  globin genes are drawn to approximate scale.  Red rectangles and open rectangles represent  exons and introns, respectively. Gray rectangles  indicate the (5 ) upstream and (3 ) downstream  nontranslated regions in the DNA. The a­like  globin genes contain introns of approximately  95 and 125 bp, located between codons 31 and  32, and 99 and 100, respectively. The b­like globin  genes contain introns of approximately125–150  and 800–900 bp, located between codons 30  and 31, and 104 and 105, respectively.

and 31 and codons 104 and 105, respectively. Introns separate the coding sequences of different functional domains of a few proteins, including the globins. The  coding region between the two globin introns specifies the region of the protein that interacts with the heme group. The final coding region (after the second intron)  encodes the region of the protein that serves as the interface with the opposite subunit, that is, the a ­globin­ b ­globin interaction. This separation of the coding  sequences for functional domains of a protein by introns is not a general phenomenon, however. The positioning of introns in other genes seems to bear little  relationship to the final three­dimensional structure of the encoded protein. Different a ­like and b ­like globin subunits are synthesized at different developmental stages. These developmentally distinct subunits have slightly different amino acid  sequences and oxygen affinities but are closely related. In humans there are two a ­like chains—that is,  , which is expressed in the embryo during the first 8 weeks,  and a  itself, which replaces   in the fetus and continues through adulthood. There are four b ­like chains. Epsilon ( ) and g are expressed in the embryo, g in the fetus,  and   plus b  in the adult. Each of the different globin chains is coded by at least one gene in the haploid genome. The a ­like genes are clustered on the short arm of human chromosome 16,  and the b ­like genes are clustered on the short arm of chromosome 11. The gene organization within these two clusters is shown in Figure 19.25. The genes within  both clusters are positioned relative to one another in the order of both their transcriptional direction and their developmental expression; that is, 5 –embryonic–fetal­ adult–3 . The a ­gene cluster spans about 28 kb and includes three functional genes and two pseudogenes. The functional genes are the embryonic   gene and two a  genes,  a 1 and a 2, that code for identical a ­globin proteins but have different 3  untranslated regions. The two pseudogenes,   and  , occur between the   and a 1 genes.  They have sequences very similar to the functional genes, but various mutations prevent them from coding for an active globin subunit. Pseudogenes are common in  eukaryotic genomes. They do not seem to be deleterious and probably arose via a duplication of a segment of DNA followed by mutations. The b ­gene cluster encompasses about 60 kb and has five active genes and one pseudogene. Of the five functional genes, two are for the g subunit and specify  proteins that differ only at position 136, which is a glycine in the G variant and an alanine in the A variant. Only a single haploid gene exists for the  ­,  ­, and b ­globin  subunits. Alu repetitive sequences and other moderately repetitive sequences are scattered between some genes of the a ­ and b ­gene clusters.

 

  

Page 827

Figure 19.25  Gene organization forr a ­like and b ­like genes of human hemoglobin.  (a) Linkage of human a­like globin genes on chromosome 16 and locations of some known deletions within  a­like gene cluster. The positions of adult (a1, a2) and embryonic ( ) a­like globin genes and two  pseudogenes ( ,  1) are shown. Pseudogenes have mutations that prevent the formation of  functional proteins from them. For each functional gene the black and white boxes represent exons and  introns, respectively. Horizontal arrow indicates the direction of transcription of each gene. The locations  of DNA deletions associated with the leftward and rightward types of a­thalassemia 2 are indicated  above the linkage map by the rectangles labeled a­thal 2 L and a­thal 2 R. Red areas at the ends of these  rectangles indicate the deletion end points have not been mapped precisely. Locations of deletions  associated with two cases of a­thalassemia 1 (a­thal 1 Thai and a­thal 1 Greek) are shown below the  linkage map. The light green areas and dashed lines indicate uncertainties in the left and right endpoints,  respectively, of each deletion.  (b) Linkage of the human b­like globin genes on chromosome 11 and locations of deletions within the  b­like gene cluster. The positions of the embryonic ( ), fetal (G , A ), and adult (d, b) b­like globin  g

g

genes and one b­like pseudogene (yb1) are shown. For each functional gene the black and white boxes  represent the exons and introns, respectively. The locations of various known deletions within the gene  cluster are shown below the map. Open rectangles represent areas known to be deleted; Red areas and  dashed lines indicate that the endpoints of the deletion have not been determined. For db­thalassemia  and hereditary persistence of fetal hemoglobin (HPFH), the type of fetal globin chain produced  (G  and/or A ) is indicated in the name of each syndrome (e.g., in (G ­A ­db­thalassemia, the G ­ and  g

g

g

g

g

A g­globin chains are produced).  Redrawn from Maniatis, T., Fritsch, E. F. Lauer, J., and Lawn, R. M. Annu. Rev. Genet. 14:145, 1980.  Copyright © 1980 by Annual Reviews, Inc.; and from Karlsson, S., and Nienhuis, A. W.  Annu. Rev. Biochem. 54:1071, 1985. Copyright © 1985 by Annual Reviews, Inc.

Other mammalian species often have a different number of globin­like genes within the two clusters. For example, rabbits have only four b ­like genes, goats have  seven, and mice have as many as nine. Some of these additional genes are pseudogenes. Many patients have been identified who have abnormalities in hemoglobin structure or expression. In many cases the precise molecular defect responsible for these  abnormalities is known. The two that have been the most extensively studied are sickle cell anemia and a family of diseases collectively called thalassemias.

 

  

Page 828

CLINICAL CORRELATION 19.4  Prenatal Diagnosis of Sickle Cell Anemia Sickle cell anemia can be diagnosed from fetal DNA obtained by amniocentesis. This  genetic disease is caused by a single base pair change that converts a glutamate to a  valine in the sixth position of b ­globin. In the normal b ­globin gene, the sequence that  specifies amino acids 5, 6, and 7 (Pro­Glu­Glu) is CCT­GAG­GAG. In a heterozygous  carrier of sickle cell anemia, this sequence is CCT­GTG­GAG. An A in the middle of the  sixth codon has been changed to a T. The restriction enzyme MstII recognizes and  cleaves the sequence CCT­GAG­G, which is present at this position in normal DNA but  not the mutated DNA. Therefore digestion of fetal DNA with MstII followed by the  Southern blot technique (see p. 774) using b ­globin cDNA as the radioactive probe  reveals whether this restriction site is present in one or both allelic copies of the gene. If it  is absent in both copies, the fetus will be homozygous for the sickle trait; if it is missing in  only one copy, the fetus will be heterozygous for the trait. The difference in restriction  enzyme patterns observed between individuals is often called a restriction fragment length  polymorphism (RFLP). Polymerase chain reaction methods can be used to amplify the  desired chromosomal DNA region and greatly speed up the RFLP analysis. Other methods are necessary if the disease mutation does not cause a change in a  restriction site or is not linked to an RFLP. For example, the DNA carrying the mutation  can be amplified by the polymerase chain reaction, and the alleles can be detected by  hybridization with allele­specific oligonucleotides (ASOs). Two ASOs differing at usually  one nucleotide are made so that one ASO matches the normal allele perfectly while the  other ASO matches the abnormal allele. Hybridization conditions are used in which only  the ASO matching perfectly remains bound to the DNA.   Sickle Cell Anemia Is Due to a Single Base Pair Change A single base pair change within the coding region for the b ­globin subunit is responsible for sickle cell anemia. This occurs in the second position of the codon for  position 6 of the b  chain. In the mRNA the codon, GAG, which specifies glutamate in normal b  chains, is converted to GUG, which specifies valine. The resultant  hemoglobin, called hemoglobin S (HbS), has altered surface charge properties (because the negative charge of glutamate has been replaced by valine's nonpolar  group), which is responsible for clinical symptoms. This mutation occurs mainly in peoples of equatorial African descent and is the classic example of a mutation that  confers an adaptive advantage as well as a genetically inheritable disease. Individuals heterozygous for HbS are resistant to infection by the parasites that cause  malaria but do not acquire the symptoms of sickle cell disease exhibited by individuals homozygous for HbS. The life cycle of the malaria­causing parasites includes  an obligatory stage that occurs inside erythrocytes and they do not survive in erythrocytes containing HbS. Carriers of the mutation can be detected by restriction  enzyme digestion of a sample of the potential carrier's DNA followed by Southern hybridization technique with the b ­globin cDNA as described in Clin. Corr. 19.4. Thalassemias Are Caused by Mutations in Genes for the a  or b  Subunits of Globin Thalassemias are a family of related genetic diseases that occur in people who frequently originate from the Mediterranean areas and Asia. If there is a reduced  synthesis or a total lack of synthesis of a ­globin mRNA, the disease is classified as a ­thalassemia; if the b ­globin mRNA level is affected, it is called b ­ thalassemia. Thalassemias can be due to the deletion of one or more globin­like genes in either of the globin gene clusters or be caused by a defect in the  transcription or processing of a globin gene's mRNA. Since each chromosome 16 contains two adjacent a ­globin genes, a normal diploid individual has four copies of this gene. a ­Thalassemic patients may be missing one  to four a ­globin genes. The condition in which one a ­globin gene is missing is referred to as a ­thal 1; when two a ­globin genes are gone, the condition is a ­thal 2.  In both cases the individuals can experience mild to moderate anemia but may have no additional symptoms. When three a ­globin genes are missing, many more b ­ globin molecules are synthesized than a ­globin molecules, resulting in the formation of a globin tetramer of four b ­globins, which causes HbH disease and  accompanying anemia. When all four a ­globin genes are absent, the disease hydrops fetalis occurs, which is fatal at or before birth. Some chromosomal deletions  that have been mapped in the a ­globin gene cluster are shown in Figure 19.25.

b ­Thalassemias also exhibit different degrees of severity and can be caused by a variety of defects or deletions. In one case the b ­globin gene is present but has  undergone a mutation in the codon 17, which generates a termination codon. In another case the b ­globin gene is transcribed in the nucleus but no b ­globin mRNA  occurs in the cytoplasm. Thus a defect has occurred in the processing and/or transport of the primary transcript of the gene. Other b ­thalassemias are caused by deletions within the b ­globin gene cluster on chromosome 11 (Figure 19.25). In some cases these deletions remove the DNA  between two adjacent genes, resulting in a new fusion gene. For example, in the normal person the linked  ­globin and b ­globin genes differ in only about 7% of their  positions. In Hb Lepore a deletion has placed the front portion of the  ­globin gene in register with the back portion of the b ­globin gene. From this fusion gene a new  b ­like globin is produced in which the N­terminal sequence of  ­globin is joined to the C­terminal sequence of b ­globin. Several variants of Hb Lepore are known,  and in each case the globin

 

  

Page 829

CLINICAL CORRELATION 19.5  Prenatal Diagnosis of Thalassemia If a fetus is suspected of being thalassemic because of its genetic background,  recombinant DNA techniques can be used to determine if one or more globin genes are  missing from its genome. Fetal DNA can easily be obtained (in relatively small quantities)  from amniotic fluid cells aspirated early during the second trimester of pregnancy. Regions  of interest are amplified from the fetal DNA by polymerase chain reactions and digested  with restriction enzymes that divide the globin genes among restriction fragments of  several hundred to 2000 base pairs. These fragments are separated by electrophoresis  through an agarose gel and hybridized with radioactive cDNA for a ­ and/or b ­globin  using the Southern blot technique (see p. 774). If one or more globin genes are missing,  the corresponding restriction fragment will not be detected or its hybridization to the  radioactive cDNA probe will be reduced (in the case when only one of two diploid genes  is absent). Benz, E. J. The hemoglobinopathies. In: W. N. Kelly (Ed.), Textbook of Internal  Medicine. Philadelphia: Lippincott, 1989, pp. 1423–1432.   product is a composite of the   and b  sequence, but the actual fusion junction is different. Another fusion b ­like globin is Hb Kenya. This deletion results in a gene product that contains the N­terminal sequence of the g­globin gene and the C­terminal  sequence of the b ­globin gene. Still another series of deletions has been found in which both the  ­ and b ­globin genes are removed, causing HPFH (hereditary  persistence of fetal hemoglobin). Frequently, there are no clinical symptoms of this condition because fetal hemoglobin ( 2 2) continues to be synthesized after the  time at which g­globin gene expression is normally turned off (see Clin. Corr. 19.5.) 19.11—  Genes for Human Growth Hormone­Like Proteins Human growth hormone (hGH, also called somatotropin) is a polypeptide of 191 amino acids. A larger precursor is synthesized in the somatotrophs of the anterior  pituitary, and the mature form is secreted into the circulatory system. Growth hormone induces liver (and perhaps other) cells to produce other hormones called  somatomedins, which are insulin­like growth factors that stimulate proliferation of mesodermal tissues such as bone, cartilage, and muscle. Infants with a deficiency in  growth hormone become dwarfs, whereas those who produce too much become giants. A closely related protein of 191 amino acids, having 85% homology with growth hormone, is human chorionic somatomammotropin (hCS, also called placental  lactogen) synthesized in the placenta. The complete role of this hormone in normal fetal–maternal physiology is still unclear, but it participates in placental growth and  contributes to mammary gland preparation for lactation during pregnancy. The hormones hGH and hCS are examples of two very similar proteins that serve different biological functions and are synthesized in different tissues. It is to be  expected that their genes also are closely related but expressed in a tissue­specific fashion. The genes for hGH and hCS are very similar and occur in the same region  of chromosome 17 (Figure 19.26). Five related genes comprise the human growth hormone gene family. They occur over a distance of about 55 kb and share a common structure of five exons and four  introns, with the exon–intron boundaries always in the same locations. Alu repetitive sequences occur between some of the genes, as in the globin gene clusters. The  order of the genes is 5  hGH­N y hCS­L, hCS­A, hGH­V, hCS­B 3¢. The first gene in this cluster, hGH­N, is expressed in the anterior

Figure 19.26  Organization of human growth hormone (hGH) gene family.  The five structural genes of this family occur as a linear array over about 55 kb on the  long arm of chromosome 17. Two genes (hGH­N and hGH­V) code for growth hormone,  two genes code for the closely related human chorionic somatomammotropin  (hCS­A and hCS­B), and one gene appears to be a pseudogene ( hCS­L). Only  hGH­N is expressed in the pituitary; other genes are expressed in the placenta. The  order of the genes in the array (red boxes) is 5  hGH­N,  hCS­L, hCS­A, hGH­V, hCS­B 3 ,  and all are transcribed in the same direction. Each gene has the same basic structure  of five exons and four introns (not shown).  Redrawn from Chen, E. Y., Liao, Y. C., Smith, D. H., Barrera­Saldana, H. A. et al.  Genomics 2:479, 1989.

 

  

Page 830

pituitary, whereas the others are expressed in the placenta. The hGH and hCS genes have different sequences about 100 bp beyond their polyadenylation sites. The  hGH­N gene codes for normal growth hormone of 22 kDa. Alternative splicing of intron 3 of this gene occurs in about 10% of the primary transcripts, giving rise to a  20­kDa version of growth hormone whose significance is not known. (See Chapter 16.) The hGH­V gene codes for a variant growth hormone that can be expressed  in transgenic animals (see Section 19.14), but whose function in the placenta is unknown. The hCS­A and hCS­B genes specify the same mature hormone but are  expressed at different levels in the placenta. The  hCS­L pseudogene has a single base substitution at an exon­intron splice site that appears to prevent normal  maturation of its primary transcript into mRNA. Expression of hGH and hCS genes is under the regulation of other hormones. Thyroxine and cortisol stimulate increased transcription of these genes. In cultured rat  pituitary tumor cells these hormones act in a synergistic fashion to induce growth hormone mRNA synthesis. Pituitary cells that have only about two molecules of  growth hormone mRNA per cell can be stimulated to a level of 1000 growth hormone mRNA molecules per cell, a 500­fold increase comparable in magnitude to the  induction of many bacterial operons. Only some of the details by which thyroxin and cortisol stimulate this increased transcription are known. Their regulatory effect at the molecular level is clearly more  complicated than is the control of bacterial operon transcription. Two promoter sites lie just upstream of hGH­N and a specific transcription factor, GHF­1 (also  called Pit­1), contributes to this gene's pituitary­specific expression. GHF­1 belongs to a family of homeodomain transcription factors found in organisms as diverse  as yeast and fruitflies. The regulatory hormones are transported into the nucleus and in association either with their receptors or with a binding protein, such as GHF­1,  affect transcription initiation at hGH­N. Alternatively, these other hormones may interact with additional factors in the cell that in turn regulate the level of transcription.  The DNA regulatory site influenced by glucocorticoid hormones is known to be upstream of the site at which transcription of hGS­N begins. An example of the many  transcription initiation protein factors that can interact with the DNA in the vicinity of eukaryotic genes is shown in Figure 16.18. Deletions can occur within the growth hormone gene family. Deletions of hGH­N in both copies of chromosome 17 have been detected in some cases of severe  growth hormone deficiency. These individuals are very short and do not have detectable serum growth hormone. Some such children initially respond very well to  treatment with recombinant human growth hormone synthesized in the bacterium E. coli (see p. 834 and Figure 19.29) but they often develop antibodies against the  growth hormone. Deletions also have been detected in which hCS­A, hGH­V, and hCS­B are lost from both chromosome 17 copies. Despite the fact that maternal  sera of these individuals lack these hormones, fetal development usually proceeds normally, suggesting they either are unnecessary or can be compensated for by other  hormones or factors. 19.12—  Mitochondrial Genes About 0.3% of the DNA of human cells occurs in the mitochondria. Human mitochondrial DNA (mtDNA) is a double­stranded circular molecule of 16,569 bp whose  sequence has been completely determined. As many as 100 molecules of mtDNA can occur in a metabolically active cell. Each mtDNA codes for 2 rRNAs, 22  tRNAs, and 13 proteins, most of which are subunits of multi­subunit complexes in the mitochondrial inner membrane that catalyze oxidative phosphorylation (Figure  19.27). For example, Complex I (NADH dehydrogenase), the first of three proton­pumping complexes involved in oxidative phosphorylation, is comprised of 26  proteins. Seven of these proteins are encoded by the

 

  

Page 831

Figure 19.27  Human mitochondrial DNA.  The 16,569­bp human mtDNA molecule  codes for two ribosomal RNAs (12 S and 16 S rRNA), some of the subunits  for NADH dehydrogenase (ND), cytochrome oxidase (CO), ATP synthase  (ATPase), and cytochrome b (cyt b), and 22 tRNAs (dark gray regions).  Most genes occur on the outer DNA strand but genes for ND6 and a few  tRNAs are on the inner strand.

CLINICAL CORRELATION 19.6  Leber's Hereditary Optic Neuropathy (LHON) Leber's hereditary optic neuropathy, first described in 1871, is a maternally inherited  genetic disease that usually strikes young adults and results in complete or partial  blindness from optic nerve degeneration. Other neurological disorders such as cardiac  dysrhythmia can also be associated with the disease. The cause of this defect in many  patients has been traced to a single base pair mutation in the mitochondrial DNA that  changes an arginine to a histidine at amino acid 340 in NADH dehydrogenase subunit 4  of Complex I in the inner mitochondrial membrane. Although it is not clear why this  mutation leads to blindness, the eyes require a high level of mitochondrial activity and  perhaps become sensitive over time to a small decrease in ATP synthesis by oxidative  phosphorylation. Singh, G., Lott, M. T., and Wallace, D. C. A mitochondrial DNA mutation as a cause of  Leber's Hereditary Optic Neuropathy. N. Eng. J. Med. 320:1300, 1989.   mtDNA. Mitochondrial DNA also contains genes for three cytochrome oxidase subunits, two ATP synthase subunits, and cytochrome b. In contrast to the nucleus,  where much of the chromosomal DNA seems to have no genetic function, virtually every base pair in mtDNA is essential. Regions between the protein­coding genes  usually encode tRNAs and sometimes the last nucleotide of one gene will be the first nucleotide of the adjacent gene. Polyadenylation at the 3  ends of some of the  mitochondrial mRNAs adds the last two A residues of the termination codon, UAA, to create the end of the reading frame. Even more remarkable, the genetic code of mammalian mtDNA is not identical to the genetic code of nuclear or prokaryotic DNA. UGA codes for tryptophan instead  of for termination, AUA codes for methionine rather than isoleucine, and AGA and AGG serve as stop codons instead of specifying arginine. It is not clear why  mitochondria have their own altered genetic system. Perhaps mtDNA is an evolutionary vestige of an early symbiotic relationship between a bacterium and the  progenitor of eukaryotic cells. What is clear is that cells makes a large investment to express the 13 mitochondria­encoded proteins. To produce those proteins a large  group of nucleus­encoded ribosomal proteins and associated translation factors must be imported into the mitochondrion and assembled, as well as all of the enzymes  and binding proteins required for mtDNA replication and transcription. More than 100 nucleus­encoded proteins are probably necessary to maintain the mtDNA and  express its gene products. Since mitochondria are in the cytoplasm, mtDNA molecules are maternally inherited. mtDNA sequences can be used as markers for maternal lineages. In addition,  mutations in mtDNA can lead to genetic diseases that are inherited only from the mother. For example, a single base pair change in mtDNA has been found to be  responsible for Leber's hereditary optic neuropathy (see Clin. Corr. 19.6). Similar mtDNA mutations may be the cause of two other maternally inherited genetic  diseases, myoclonic epilepsy and infantile bilateral striatal necrosis.

 

  

Page 832

19.13—  Bacterial Expression of Foreign Genes Recombinant DNA techniques are now frequently used to construct bacteria that are ''factories" for making large quantities of specific human proteins useful in the  diagnosis or treatment of disease. The two examples to be illustrated here are the construction of bacteria that synthesize human insulin and human growth hormone. Many factors must be considered in designing recombinant plasmids that contain a eukaryotic gene to be expressed in bacteria. First, the cloned eukaryotic gene  cannot have any introns since the bacteria do not have the RNA­splicing enzymes that correctly remove introns from the initial transcript. Thus the actual eukaryotic  chromosomal gene is usually not used for these experiments; instead, the cDNA or a synthetic equivalent of the coding sequence, or a combination of both, is placed in  the bacterial plasmid. Another consideration is that different nucleotide sequences comprise the binding sites for RNA polymerase and ribosomes in bacteria and eukaryotes. Therefore, to  achieve expression of the desired protein, it is necessary to insert the eukaryotic coding sequence directly behind a set of bacterial regulatory elements. This has the  advantage that the foreign gene is now under the regulation of the bacterial control elements, but its disadvantage is that considerable recombinant DNA manipulation is  required to make the appropriate plasmid. Still other factors to be considered are that the foreign gene product must not be degraded by bacterial proteases or require  modification before it is active (e.g., specific glycosylation events that the bacteria cannot perform) and must not be toxic to the bacteria. Even when the bacteria do  synthesize the desired product, it must be isolated from the 1000 or more endogenous bacterial proteins. Recombinant Bacteria Can Synthesize Human Insulin Insulin is produced by the b  cells of the pancreatic islets of Langerhans. It is initially synthesized as preproinsulin, a precursor polypeptide that possesses an N­ terminal signal peptide and an internal C peptide of 33 amino acids that are removed during the subsequent maturation and secretion of insulin (see p. 40). The A  peptide (21 amino acids) and B peptide (30 amino acids) of mature insulin are both derived from this initial precursor and are held together by two disulfide bridges.  Bacteria do not have the processing enzymes that convert the precursor form to mature insulin. Therefore the initial strategy for bacterial synthesis of human insulin  involved the production of the A and B chains by separate bacteria followed by purification of the individual chains and subsequent formation of the proper disulfide  linkages. The first step was to use synthetic organic chemistry methods to prepare a series of single­stranded oligonucleotides (11–18 nucleotides) that were both  complementary and overlapping with each other. When these oligonucleotides were mixed together in the presence of DNA ligase under proper conditions, they  formed a double­stranded fragment of DNA with termini equivalent to those formed by specific restriction enzymes (Figure 19.28). The sequences of the  oligonucleotides were carefully chosen so that one of the two strands contained a methionine codon followed by the coding sequence of the A chain of insulin and a  termination codon. A second set of overlapping complementary oligonucleotides were prepared and ligated together to form another double­stranded DNA fragment  that contained a methionine codon followed by 30 codons specifying the B chain of insulin and a termination codon. These two double­stranded fragments were then individually cloned at a restriction site in the b ­galactosidase gene of the lactose operon in a plasmid. These two  recombinant plasmids were introduced into bacteria. The bacteria could now produce a fusion protein of b ­galactosidase and the A chain (or B chain) whose  expression was under control of the lactose operon. In the absence

 

  

Page 833

Figure 19.28  Bacterial expression of the A and B chains of human insulin.  Step 1: A series of overlapping, complementary oligonucleotides (11 for the A chain and  18 for the B chain) were synthesized and ligated together. One strand of the resulting  small DNA fragments contained a methionine codon followed by coding sequence for A  chain and B chain, respectively.  Step 2: The small DNA fragments were ligated into a restriction site near the end of the  b­galactosidase gene of the lactose operon in a plasmid.  Step 3: Recombinant plasmids were introduced into E. coli and the b­galactosidase gene  was induced with IPTG, an inducer of the lactose operon. A fusion protein was produced  that contained most of the b­galactosidase sequence at the N terminus and the A chain  (or B chain) at the C terminus.  Step 4: Bacterial cell lysates containing the fusion protein were treated with cyanogen  bromide, which cleaves peptide bonds following methionine residues. Step 5: A and  B chains were purified away from all other cyanogen bromide peptides using  biochemical and immunological separation techniques. The –SH groups on the  cysteines were activated and reacted to form intra­ and interchain disulfide  bridges found in mature human insulin.  Redrawn from Crea, R., Krazewski, A., Hirose, T., and Itakura, K. Proc. Natl. Acad.  Sci. USA 75:5765, 1980.

of lactose in the bacterial medium, the lactose operon is repressed and only very small amounts of the fusion protein are synthesized. Using induction with IPTG and  some additional genetic tricks, the bacteria can be forced to synthesize as much as 20% of their protein as the fusion protein. The A peptide (or B peptide) can be  released from this fusion protein by treatment with cyanogen bromide, which cleaves on the carboxyl side of methionine residues. Since neither the A nor B peptide  contains a methionine, they will be liberated intact and can subsequently be purified to homogeneity. The final steps involve chemically activating the free –SH groups  on the cysteines and mixing the activated A and B chains together in a way that the proper disulfide linkages form to generate molecules of mature human insulin.

 

  

Page 834

Recombinant Bacteria Can Synthesize Human Growth Hormone The strategy for generating a recombinant DNA plasmid from which bacteria can synthesize human growth hormone is somewhat different than for insulin synthesis.  First, human growth hormone is 191 amino acids long so the synthetic construction of the corresponding DNA coding sequence is more difficult (although certainly not  impossible) than in the insulin case. On the other hand, growth hormone is a single polypeptide so it is not necessary to deal with the production of two chains and their  subsequent dimerization to form a protein with biological activity. Because of these considerations, the coding sequence was initially cloned into a bacterial expression  plasmid using part of a cloned growth hormone cDNA and several synthetic oligonucleotides (Figure 19.29). The overlapping oligonucleotides were prepared so that,  when ligated together, they would form a small double­stranded DNA containing the codons for the first 24 amino acids of mature human growth hormone. One end  of this DNA

Figure 19.29  Expression of human growth hormone in E. coli.  Step 1: Several overlapping, complementary, oligonucleotides were synthesized  and ligated together. One strand of the resulting small DNA fragment contains  the coding sequence for the first 24 amino acids of mature human growth  hormone (after removal of the N­terminal signal peptide).  Step 2: A recombinant plasmid with a full length human growth hormone (hGH)  cDNA, which is not expressed, is cleaved with restriction enzymes that release a  fragment containing the complete growth hormone coding sequence after codon 24.  Step 3: The synthetic fragment and the partial cDNA­containing fragment are  ligated together to yield a new fragment containing the complete coding  sequence of mature hGH.  Step 4: The new fragment is ligated into a restriction site just downstream from  the lactose promoter–operator region cloned in a plasmid.  Step 5: The resulting recombinant DNA plasmid is introduced into bacteria in  which synthesis of hGH can be induced with IPTG, an inducer of the lactose operon.

 

  

Page 835

fragment was designed so that the fragment could be ligated in front of a restriction fragment of growth hormone cDNA that provided the rest of the coding sequence,  including the termination codon. The other end of the synthetic fragment was chosen so that the composite coding sequence could easily be inserted into a site  immediately downstream of the promoter–operator–ribosome binding site of the lactose operon cloned in a plasmid. After the introduction into bacteria, the bacteria  were induced with IPTG to transcribe this foreign coding region and the greatly overproduced human growth hormone subsequently was purified away from the  bacterial proteins. 19.14—  Introduction of Rat Growth Hormone Gene into Mice The previous section described the use of bacteria to produce large quantities of human proteins for treatment of disease. It is possible to microinject molecules of  purified RNA or DNA directly into eukaryotic cells. This provides a very powerful approach for identifying conditions under which specific genes are expressed in  eukaryotic cells. One of the most dramatic illustrations of this approach was the microinjection of a chromosomal DNA fragment containing the structural gene for rat  growth hormone into the pronuclei of fertilized mouse eggs. The eggs were then reimplanted into the reproductive tracts of foster mouse mothers. Some of the mice  that developed from this procedure were transgenic; one or more copies of the microinjected growth hormone gene integrated into a host mouse chromosome at an  early stage of embryo development. These foreign genes were transmitted through the germline and became a permanent feature in the host chromosomes of the  progeny (Figure 19.30).

Figure 19.30  Schematic illustration of the introduction of rat growth hormone  gene into mice.  Copies of a recombinant plasmid DNA containing rat growth hormone  gene were microinjected into fertilized mouse eggs that were reimplanted  into foster mothers. Some of the resulting progeny contained the foreign  gene integrated into their own genome and greatly overexpressed growth  hormone, growing much larger than their normal­sized littermates.  Redrawn from Palmiter, R. D., Brinster, R. L., Hammer, R. E. et al.  Nature 300:611, 1982.

 

  

Page 836

Analysis of these transgenic mice revealed that in some cases several tandem copies of the rat growth hormone gene had integrated into a mouse chromosome; in other  cases only one gene copy was present. In all cases at least some transcription occurred from the integrated gene(s), and in a few cases a dramatic overproduction of  rat growth hormone resulted. In these latter cases, as much as 800 times more growth hormone was present in the transgenic mice than in normal mice, resulting in  animals more than three times the size and weight of their unaffected littermates. These results present many potential experimental possibilities for the future and raise a number of issues. One implied possibility is the use of similar growth hormone  gene insertions to stimulate rapid growth of commercially valuable animals. This could result in a shorter production time and increased efficiency of food utilization.  Another long­term possibility is the use of this approach to correct certain human genetic diseases or mimic the diseases in experimental animals so that they can be  studied more carefully. One obvious human disease that is a candidate for this "gene therapy" approach is thalassemia. For example, an individual with two to three  missing a ­globin genes might benefit tremendously from receiving bone marrow transplants of his/her own cells that have been established in culture and microinjected  with additional copies of the normal a ­globin gene. This approach to gene therapy is being investigated. Insertion of normal genes into human somatic cells of a  defective tissue or organ does not result in transmission of these genes to the progeny. This lessens the ethical considerations for experiments that do not alter germline  characteristics. Bibliography Prokaryotic Gene Expression Cohen, S. N., and Shapiro, J. A. Transposable genetic elements. Sci. Am. 242:40, 1980. Miller, J. H. The lac gene: its role in lac operon control and its use as a genetic system. In: J. H. Miller and W. S. Resnikoff (Eds.), The Operon. Cold Spring Harbor,  NY: Cold Spring Harbor Laboratory Press, 1978, p. 31. Platt, T. Regulation of gene expression in the tryptophan operon of Escherichia coli. In: J. H. Miller and W. S. Resnikoff (Eds.), The Operon. Cold Spring Harbor,  NY: Cold Spring Harbor Laboratory Press, 1978, p. 263. Simon, M., Zieg, J., Silverman, M., Mandel, G., and Doolittle, R. Phase variation: evolution of a controlling element. Science 209:1370, 1980. Eukaryotic Gene Expression Ahn, A. H., and Kunkel, L. M. The structural and functional diversity of dystrophin. Nature Genetics 3:283, 1993. Chen, E. Y., Liao, Y.­C., Smith, D. H., Barrera­Saldaña, H. A., Gelinas, R. E., and Seeburg, P. H. The human growth hormone locus: nucleotide sequence, biology  and evolution. Genomics 4:479, 1989. Enver, T., and Greaves, D. R. Globin gene switching: a paradigm or what? Curr. Opin. Biotech. 2:787, 1991. Johns, D. R. Mitochondrial DNA and disease. N. Engl. J. Med. 333:638, 1995. Karlsson, S., and Nienhuis, A. W. Developmental regulation of human globin genes. Annu. Rev. Biochem. 54:1071, 1985. Maniatis, T., Fritsch, E. F., Laurer, J., and Lawn, R. M. The molecular genetics of human hemoglobins. Annu. Rev. Genet. 14:145, 1980. Mitchell, P. J., and Tjian, R. Transcription regulation in mammalian cells by sequence­specific DNA binding proteins. Science 245:371, 1989. Palmiter, R. D., Brinster, R. L., Hammer, R. E., et al. Dramatic growth of mice that develop from eggs microinjected with metallothionein­growth hormone fusion  genes. Nature 300:611, 1982. Singh, G., Lott, M. T., and Wallace, D. C. A mitochondrial DNA mutation as a cause of Leber's hereditary optic neuropathy. N. Engl. J. Med. 320:1300, 1989. Struhl, K. Chromatin structure and RNA polymerase II connection: implications for transcription. Cell 84:179, 1996. The Huntington's Disease Collaborative Research Group. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease  chromosomes. Cell 72:971, 1993. Questions J. Baggott and C. N. Angstadt 1. Full expression of the lac operon requires: A. lactose and cAMP. B. allolactose and cAMP. C. lactose alone. D. allolactose alone. E. absence or inactivation of the lac corepressor. 2. In an operon: A. each gene of the operon is regulated independently to achieve levels of expression required by the cell. B. control may be exerted via induction or via repression. C. operator and promoter may be trans to the genes they regulate.

 

  

Page 837

D. the structural genes are either not expressed at all or are fully expressed. E. control of gene expression consists exclusively of induction and repression. 3. The E. coli lacZYA region will be upregulated if: A. there is a defect in binding of the inducer to the product of the lacI gene. B. glucose and lactose are both present in the growth medium, but there is a defect in the cell's ability to bind the CAP protein. C. glucose and lactose are both readily available in the growth medium. D. the operator has mutated so it can no longer bind repressor. E. the lac corepressor is not present. 4. All of the following describe an operon EXCEPT: A. control mechanism for eukaryotic genes. B. includes structural genes. C. expected to code for polycistronic mRNA. D. contains control sequences such as an operator. E. can have multiple promoters. Refer to the following for Questions 5–9: A. repression B. corepression C. attenuation D. stringent response E. RNA splicing 5. Associated with guanosine tetraphosphate and guanosine pentaphosphate. 6. Not found in prokaryotes. 7. Involves rho­independent transcription termination. 8. Involves a leader peptide containing several occurrences of the same amino acid. 9. The only regulatory mechanism for the his operon. 10. Ribosomal operons: A. all contain genes for proteins of just one ribosomal subunit. B. all contain genes for proteins of both ribosomal subunits. C. all contain genes for only ribosomal proteins. D. can have their expression regulated at the level of translation. E. are widely separated in the E. coli chromosome. 11. All of the following phrases describe transposons EXCEPT: A. a means for the permanent incorporation of antibiotic resistance into the bacterial chromosome. B. contain short inverted terminal repeat sequences. C. code for an enzyme that synthesizes guanosine tetraphosphate and guanosine pentaphosphate, which inhibit further transposition. D. include at least one gene that codes for a transposase. E. contain varying numbers of genes, from two to several. 12. Introns: A. are of approximately uniform size. B. are skipped over during translation. C. are found in all eukaryotic genes. D. function to separate functional domains of proteins. E. are smaller and shorter in unicellular lower eukaryotes than in higher, more complicated eukaryotes. 13. Repetitive DNA: A. is common in bacterial and mammalian systems. B. is uniformly distributed throughout the genome. C. includes DNA that codes for rRNA. D. consists mostly of DNA that codes for enzymes catalyzing major metabolic processes. E. is resistant to the action of restriction endonucleases. 14. The b ­gene cluster contains: A. one haploid   gene. B. one haploid b  gene. C. one haploid g gene. D. two haploid   genes. E. two haploid   genes. 15. The number of a  genes in the haploid a ­gene cluster is A. one. B. two. C. three. D. four. E. five. 16. In designing a recombinant DNA for the purpose of synthesizing an active eukaryotic polypeptide in bacteria all of the following should be true EXCEPT: A. the eukaryotic gene may contain its usual complement of introns. B. the foreign polypeptide should be resistant to degradation by bacterial proteases. C. glycosylation of the polypeptide should be unnecessary. D. the foreign polypeptide should be nontoxic to the bacteria. E. bacterial controlling elements are necessary. Answers 1. B A: The true inducer is allolactose, not lactose (p. 806). C: Lactose is converted in the cell to allolactose. D: In addition, cAMP must bind to the CAP protein, and  the cAMP­CAP complex serves as a positive control of transcription (p. 807). E: The lac operon does not involve corepression. 2. B Induction and repression are among the mechanisms used to control operons. A: In an operon the structural genes are under coordinate control. C: The operator  and promoter are elements of the same strand of DNA as the operon they control; they are not diffusible. D: Typically, regulation of operators is somewhat leaky;  some gene product is produced even in the repressed state. E: Another mechanism for regulation of an operon is attenuation (p. 810).

 

  

Page 838

3. D If the operator is unable to bind repressor, the rate of transcription is greater than the basal level (p. 804). A: The product of the lacI gene is the repressor  protein. When this protein binds an inducer, it changes its conformation, no longer binds to the operator site of DNA, and transcription occurs at an increased rate.  Failure to bind an inducer prevents this sequence. B and C: In the presence of glucose catabolite repression occurs. Glucose lowers the intracellular level of cAMP.  The catabolite activator protein (CAP) then cannot complex with cAMP, so there is no CAP–cAMP complex to activate transcription. The same would occur if the  cell had lost its capacity to synthesize cAMP (p. 808). E: The lac operon does not involve corepression. 4. A Operons are prokaryotic mechanisms. B–D: An operon is the complete regulatory unit of a set of clustered genes, including the structural genes (which are  transcribed together to form a polycistronic mRNA), regulatory genes, and control elements, such as the operator (p. 801). E: An operon may have more than one  promoter, as does the tryptophan operon of E. coli (p. 810). 5. D The exact functions of these species are not yet known, but their production is very rapid after the onset of amino acid starvation (p. 815). 6. E Splicing is a eukaryotic phenomenon (p. 820). 7. C The hairpin loop that forms between regions 3 and 4 (Figure 19.11) is followed by an oligo­U region (Figure 19.10). This constellation compromises the signal  for rho­independent termination of transcription. (See pp. 811–812.) 8. C Synthesis of the leader peptide depends strongly on availability of this amino acid, since it must be incorporated several times. When it is insufficiently available,  the ribosome stalls, in region 1 (Figure 19.11), allowing the 2–3 hairpin to form. This in turn prevents formation of the 3–4 hairpin, which would signal termination of  transcription. 9. C In this operon the stalled ribosome acts rather like a positive regulator protein, that is, the cAMP–CAP complex (p. 813). 10. D Excess ribosomal protein binds to its own mRNA, preventing initiation of further synthesis of that protein (p. 814). A, B, C, and E: The genes for one half of the  ribosomal proteins are in two major clusters. There is no pattern to the distribution of genes for the proteins of the two ribosomal subunits, and they are intermixed with  genes for other proteins involved in protein synthesis. 11. C These guanosine phosphates are synthesized by the product of the relA gene; they inhibit initiation of transcription of the rRNA and tRNA genes, shutting off  protein synthesis in general. This is the stringent response (p. 815). 12. E A: Introns are of various sizes. B: They are excised during splicing, not skipped over during translation. C: Although they are common, some genes do not have  them, for example, the histone and interferon genes (p. 821). D: Sometimes they occur between functional domains of proteins, but not always. 13. C This makes sense, since many copies of these structural elements are needed (p. 823). A and B: Highly repetitive and moderately repetitive DNA are found only  in eukaryotes. Highly repetitive sequences tend to be clustered, as are some moderately repetitive sequences (p. 822). D: Most repetitive DNA does not code for a  stable gene product (p. 822). E: The Alu family of moderately repetitive DNA is named for the restriction endonuclease that cleaves them (p. 823). 14. B This means that there are only two b  genes per diploid cell. As a result, in b ­thalassemia, one defective b ­globin gene gives rise to a minor form of the disease,  while two defective genes cause the major form. (See p. 828.) 15. B As a result, a ­thalassemia is more complicated than b ­thalassemia because there are four a ­globin genes per diploid cell, and anywhere between zero and four  of them can be defective. (See p. 828.) 16. A A and C: The bacterial system has no mechanism for posttranscriptional modification of mRNA or for posttranslational (or cotranslational) modification of  protein. E: Bacterial systems need bacterial promoters, and so on (p. 832).

 

  

Page 839

Chapter 20—  Biochemistry of Hormones I: Polypeptide Hormones Gerald Litwack and Thomas J. Schmidt

20.1 Overview

840

20.2 Hormones and the Hormonal Cascade System

841

Cascade System Amplifies a Specific Signal

841

Polypeptide Hormones of the Anterior Pituitary

844

20.3 Major Polypeptide Hormones and Their Actions

846

20.4 Genes and Formation of Polypeptide Hormones

849

Proopiomelanocortin Is a Precursor Polypeptide for Eight Hormones

849

Many Polypeptide Hormones Are Encoded Together in a Single Gene

849

Multiple Copies of a Hormone Can Be Encoded on a Single Gene

852

20.5 Synthesis of Amino Acid­Derived Hormones

853

Epinephrine Is Synthesized from Phenylalanine/Tyrosine

853

Synthesis of Thyroid Hormone Requires Incorporation of Iodine into a  Tyrosine of Thyroglobulin

854

20.6 Inactivation and Degradation of Hormones

857

20.7 Cell Regulation and Hormone Secretion

859

G­Proteins Serve as Cellular Transducers of Hormone Signals

859

Cyclic AMP Activates Protein Kinase A Pathway

862

Inositol Triphosphate Formation Leads to Release of Calcium from  Intracellular Stores

862

Diacylglycerol Activates Protein Kinase C Pathway

865

20.8 Cyclic Hormonal Cascade Systems

866

Melatonin and Serotonin Synthesis Are Controlled by Light and Dark  Cycles

866

Ovarian Cycle Is Controlled by Gonadotropin­Releasing Hormone

867

Absence of Fertilization

868

Fertilization

870

20.9 Hormone–Receptor Interactions

871

Scatchard Analysis Permits Determination of the Number of Receptor­ Binding Sites and Association Constant for Ligand

872

Some Hormone–Receptor Interactions Involve Multiple Hormone  Subunits

872

20.10 Structure of Receptors: b ­Adrenergic Receptor

875

20.11 Internalization of Receptors

876

Clathrin Forms a Lattice Structure to Direct Internalization of Hormone– Receptor Complexes from the Plasma Membrane

877

20.12 Intracellular Action: Protein Kinases Insulin Receptor: Transduction through Tyrosine Kinase

879

Activity of Vasopressin: Protein Kinase A

880

Gonadotropin­Releasing Hormone (GnRH): Protein Kinase C

883

Activity of Atrial Natriuretic Factor (ANF): Protein Kinase G

885

20.13 Oncogenes and Receptor Functions

888

Bibliography

890

Questions and Answers

890

Clinical Correlations

   

  

878

 

20.1 Testing Activity of the Anterior Pituitary

844

20.2 Hypopituitarism

846

20.3 Lithium Treatment of Manic–Depressive Illness: The  Phosphatidylinositol Cycle

863

Page 840

20.1—  Overview Cells are regulated by many hormones, growth factors, neurotransmitters, and certain toxins through interactions of these diverse ligands with their cognate receptors  located at the cell surface. This collection of receptors is the major mechanism through which peptide hormones and amino acid­derived hormones exert their  effects at the cellular level. Another important mechanism involves permeation of the cell membrane by steroid hormones that subsequently interact with their  intracellular cognate receptors (Chapter 21). These two sites, the plasma membrane and the intracellular milieu, represent the principal locations of the initial interaction  between ligands and cellular receptors and are diagrammed in Figure 20.1. Polypeptide hormones and several amino acid­derived hormones bind to cognate receptors  in the plasma membrane. One exception is thyroid hormone, which binds to a receptor that resides in the nucleus much like certain steroid hormone receptors. The hormonal cascade system is applicable to many, but not all, hormones. It begins with signals in the central nervous system (CNS), followed by hormone  secretion by the hypothalamus, pituitary, and end target organ. In this chapter major polypeptide hormones are summarized and the synthesis of specific hormones is  described. Synthesis of the amino acid­derived hormones, epinephrine and triiodo­L­thyronine, is also outlined. Examples of hormone inactivation and degradation are  presented. The remainder of this chapter focuses on receptors, signal transduction, and second messenger pathways. Receptor internalization is described and  examples of cyclic hormonal cascade systems are introduced. Finally, a discussion of oncogenes and receptor function is presented. In terms of receptor mechanisms, aspects of hormone–receptor interactions are presented with a brief mathematical analysis. Signal transduction is considered,  especially in reference to GTP­binding proteins. Second messenger systems discussed include cAMP and the protein kinase A pathway, inositol triphosphate– diacylglycerol and the Ca2+–protein kinase C pathway, and cGMP and

Figure 20.1  Diagram showing the different locations of classes of receptors expressed by a target cell.

 

  

Page 841

the protein kinase G pathway. These pathways are discussed in the context of representative hormone action. Newly identified components of these signal transduction  pathways are defined in terms of the kinase system(s) involved. In addition, the insulin receptor and its tyrosine kinase and second messenger pathways are  considered. 20.2—  Hormones and the Hormonal Cascade System The definition of a hormone has been expanded over the last several decades. Hormones secreted by endocrine glands were originally considered to represent all of  the physiologically relevant hormones. Today, the term hormone refers to any substance in an organism that carries a signal to generate some sort of alteration at the  cellular level. Thus endocrine hormones represent a class of hormones that arise in one tissue, or "gland," and travel a considerable distance through the circulation to  reach a target cell expressing cognate receptors. Paracrine hormones arise from a cell and travel a relatively small distance to interact with their cognate receptors on  another neighboring cell. Autocrine hormones are produced by the same cell that functions as the target for that hormone (neighboring cells may also be targets).  Thus we can classify hormones based on their radii of action. Often, endocrine hormones that travel long distances to their target cells may be more stable than  autocrine hormones that exert their effects over very short distances. Cascade System Amplifies a Specific Signal For many hormonal systems in higher animals, the signal pathway originates with the brain and culminates with the ultimate target cell. Figure 20.2 outlines the sequence  of events in this cascade. A stimulus may originate in the external environment or within the organism in this cascade. This signal may be transmitted as an electrical  pulse (action potential) or as a chemical signal or both. In many cases, but not all, such signals are forwarded to the limbic system and subsequently to the  hypothalamus, the pituitary, and the target gland that secretes the final hormone. This hormone then affects various target cells to a degree that is frequently  proportional to the number of cognate receptors

Figure 20.2  Hormonal cascade of signals from CNS to ultimate  hormone.  The target "gland" refers to the last hormone­producing  tissue in the cascade, which is stimulated by an  appropriate anterior pituitary hormone. Examples would  be thyroid gland, adrenal cortex, ovary, and testis.  Ultimate hormone feeds back negatively on sites  producing intermediate hormones in the cascade.  Amounts [nanogram (ng), microgram (mg), and  milligram (mg)] represent approximate quantities  of hormone released.  Redrawn from Norman, A. W., and Litwack, G.  Hormones. New York: Academic Press, 1987, p. 38.

 

  

Page 842

expressed by that cell. This may be a true cascade in the sense that increasing amounts of hormones are generated at successive levels (hypothalamus, pituitary, and  target gland) and also because the half­lives of these blood­borne hormones tend to become longer in progression from the hypothalamic hormone to the ultimate  hormone. In the case of environmental stress, for example, there is a single stressor (change in temperature, noise, trauma, etc.). This stress results in a signal to the  hippocampal structure in the limbic system that signals the hypothalamus to release a hypothalamic releasing hormone, corticotropin­releasing hormone (CRH), which  is usually secreted in nanogram amounts and may have a t 1/2 in the bloodstream of several minutes. This hormone travels down a closed portal system to gain access to  the anterior pituitary, where it binds to its cognate receptor in the cell membrane of corticotropic cells and initiates a set of metabolic changes resulting in the release  of adrenocorticotropic hormone (ACTH) as well as b ­lipotropin. This hormone, which is released in microgram amounts and has a longer t 1/2 than CRH, circulates in  the bloodstream until it binds to its cognate receptors expressed in the membranes of cells located in the inner layer of the cortex of the adrenal gland (target gland).  Here it affects metabolic changes leading to the synthesis and release in 24 h of the ultimate hormone, cortisol, in multimilligram amounts and this active glucocorticoid  hormone has a substantial t 1/2 in blood. Cortisol is taken up by a wide variety of cells that express varying amounts of the intracellular glucocorticoid receptor. The  ultimate hormone, in this case cortisol, feeds back negatively on cells of the anterior pituitary, hypothalamus, and perhaps higher levels to shut down the overall  pathway in a process that is also mediated by the glucocorticoid receptor. At the target cell level these cortisol­receptor complexes mediate specific transcriptional  responses and the individual hormonal effects summate to produce the systemic effects of the hormone. The cascade is represented in this example by a single  environmental stimulus generating a series of hormones in progressively larger amounts and with increasing stabilities, and by the ultimate hormone that affects most of  the cells in the body. Many other systems operate similarly, there being different specific releasing hormones, anterior pituitary tropic hormones, and ultimate  hormones involved in the process. Clearly, the final number of target cells affected may be large or small depending on the distribution of receptors for each ultimate  hormone. A related system involves the posterior pituitary hormones, oxytocin and vasopressin (antidiuretic hormone), which are stored in the posterior pituitary gland but are  synthesized in neuronal cell bodies located in the hypothalamus. This system is represented in Figure 20.3; elements of Figure 20.2 appear in the central vertical  pathway. The posterior pituitary system branches to the right from the hypothalamus. Oxytocin and vasopressin are synthesized in separate cell bodies of hypothalamic  neurons. More cell bodies dedicated to synthesis of vasopressin are located in the supraoptic nucleus and more cell bodies dedicated to synthesis of oxytocin are  located in the paraventricular nucleus. Their release from the posterior pituitary gland along with neurophysin, a stabilizing protein, occurs separately via specific  stimuli impinging on each of these types of neuronal cells. There are highly specific signals dictating the release of polypeptide hormones along the cascade of this system. Thus there are a variety of aminergic neurons  (secreting amine­containing substances like dopamine and serotonin) which connect to neurons involved in the synthesis and release of the releasing hormones of the  hypothalamus. Releasing hormones are summarized in Table 20.1. These aminergic neurons fire depending on various types of internal or external signals and their  activities account for pulsatile release patterns of certain hormones, such as the gonadotropin­releasing hormone (GnRH), and the rhythmic cyclic release of  other hormones like cortisol. Another prominent feature of the hormonal cascade (Figure 20.3) is the negative feedback system operating when sufficiently high levels of the ultimate hormone  have been secreted into the circulation. Generally, there are three feedback loops—the long feedback, the short feedback, and the ultra­

 

  

Page 843

Figure 20.3  Many hormonal systems involve the hypothalamus.  Cascade of hormonal signals starting with an external or internal environmental signal.  This is transmitted first to the CNS and may involve components of the limbic system,  such as the hippocampus and amygdala. These structures innervate the hypothalamus  in a specific region, which responds with secretion of a specific releasing hormone,  usually in nanogram amounts. Releasing hormones are transported down a closed portal  system connecting the hypothalamus and anterior pituitary, bind to cell membrane  receptors and cause the secretion of specific anterior pituitary hormones, usually in  microgram amounts. These access the general circulation through fenestrated local  capillaries and bind to specific target gland receptors. The interactions trigger release of  an ultimate hormone in microgram to milligram daily amounts, which generate the  hormonal response by binding to receptors in several target tissues. In effect, this overall  system is an amplifying cascade. Releasing hormones are secreted in nanogram amounts and  they have short half­lives on the order of a few minutes. Anterior pituitary hormones are  produced often in microgram amounts and have longer half­lives than releasing hormones.  Ultimate hormones can be produced in daily milligram amounts with much longer half­lives.  Thus the products of mass × half­life constitute an amplifying cascade mechanism. With  respect to differences in mass of hormones produced from hypothalamus to target gland,  the range is nanograms to milligrams, or as much as one million­fold. When the ultimate  hormone has receptors in nearly every cell type, it is possible to affect the body chemistry  of virtually every cell by a single environmental signal. Consequently, the organism is in  intimate association with the external environment, a fact that we tend to underemphasize.  Solid arrows indicate a secretory process. Long arrows studded with open or closed circles  indicate negative feedback pathways (ultra­short, short, and long feedback loops).  Redrawn from Norman, A. W., and Litwack, G. Hormones.  New York: Academic Press, 1987, p. 102.

 

  

Page 844 TABLE 20.1 Hypothalamic Releasing Hormonesa

Releasing Hormone

Number of  Amino Acids in  Structure

Anterior Pituitary Hormone  Released or Inhibited

Thyrotropin­releasing hormone  (TRH)

  3

Thyrotropin (TSH); can also  release prolactin (PRL)  experimentally

Gonadotropin­releasing hormone  (GnRH)

10

Luteinizing and follicle­ stimulating hormones (LH and  FSH) from the same cell type;  leukotriene C4 (LTC4) can also  release LH and FSH by a different  mechanism

Gonadotropin release­inhibiting  factor (GnRIF)

12.2 kDa  LH and FSH release inhibited molecular weight

Corticotropin­releasing hormone  (CRH)

41

ACTH, b ­lipotropin (b ­LPH),  and some b ­endorphin

Arginine vasopressin (AVP)

  9

Stimulates CRH action in ACTH  release

Angiotensin II (AII)

  8

Stimulates CRH action in ACTH  release; releases ACTH weakly

Somatocrinin (GRH)

44

Growth hormone (GH) release

14

GH release inhibited

Somatostatin (GIH) Hypothalamic gastrin­releasing  peptide

 

Inhibits release of GH and PRL

Prolactin­releasing factor (PRF)

 

Releases prolactin (PRL)

Prolactin release­inhibiting factor  (PIF)

 

Evidence that a new peptide may  inhibit PRL release; dopamine  also inhibits PRL release and was  thought to be PIF for some time;  dopamine may be a secondary  PIF: oxytocin may inhibit PRL  release

a

 Melanocyte­stimulating hormone (MSH) is a major product of the pars intermedia  (Figure 20.5) in the rat and is under the control of aminergic neurons. Humans may also  secrete a ­MSH from pars intermedia­like cells although this structure is anatomically  indistinct in the human.

  CLINICAL CORRELATION 20.1  Testing Activity of the Anterior Pituitary Releasing hormones and chemical analogs, particularly of the smaller peptides, are now  routinely synthesized. The gonadotropin­releasing hormone, a decapeptide, is available  for use in assessing the function of the anterior pituitary. This is of importance when a  disease situation may involve either the hypothalamus, the anterior pituitary, or the end  organ. Infertility is an example of such a situation. What needs to be assessed is which  organ is at fault in the hormonal cascade. Initially, the end organ, in this case the gonads,  must be considered. This can be accomplished by injecting the anterior pituitary hormone  LH or FSH. If sex hormone secretion is elicited, then the ultimate gland would appear to  be functioning properly. Next, the anterior pituitary would need to be analyzed. This can  be done by i.v. administration of synthetic GnRH; by this route GnRH can gain access to  the gonadotropic cells of the anterior pituitary and elicit secretion of LH and FSH.  Routinely, LH levels are measured in the blood as a function of time after the injection.  These levels are measured by radioimmunoassay (RIA) in which radioactive LH or hCG  is displaced from binding to an LH­binding protein by LH in the serum sample. The extent  of the competition is proportional to the amount of LH in the serum. In this way a  progress of response is measured that will be within normal limits or clearly deficient. If  the response is deficient, the anterior pituitary cells are not functioning normally and are  the cause of the syndrome. On the other hand, normal pituitary response to GnRH would  indicate that the hypothalamus was nonfunctional. Such a finding would prompt  examination of the hypothalamus for conditions leading to insufficient  availability/production of releasing hormones. Obviously, the knowledge of hormone  structure and the ability to synthesize specific hormones permit the diagnosis of these  disease states. Marshall, J. C., and Barkan, A. L. Disorders of the hypothalamus and anterior pituitary.  In: W. N. Kelley (Ed.), Internal Medicine. New York: Lippincott, 1989, p. 2159; and  Conn, P. M. The molecular basis of gonadotropin­releasing hormone action. Endocr.  Rev. 7:3, 1986.   short feedback loops. In the long feedback loop, the final hormone binds a cognate receptor in/on cells of the anterior pituitary, hypothalamus, and CNS to prevent  further elaboration of hormones from those cells that are involved in the cascade. The short feedback loop is accounted for by the pituitary hormone that feeds back  negatively on the hypothalamus operating through a cognate receptor. In ultra­short feedback loops the hypothalamic releasing factor feeds back at the level of the  hypothalamus to inhibit further secretion of this releasing factor. These mechanisms provide tight controls on the operation of the cascade, responding to stimulating  signals as well as negative feedback, and render this system highly responsive to the hormonal milieu. Clinical Correlation 20.1 describes approaches for testing the  responsiveness of the anterior pituitary gland. Polypeptide Hormones of the Anterior Pituitary The polypeptide hormones of the anterior pituitary are shown in Figure 20.4 together with their controlling hormones from the hypothalamus. The major

 

  

Page 845

Figure 20.4  Overview of anterior pituitary hormones with hypothalamic releasing hormones and their actions.

 

  

Page 846

CLINICAL CORRELATION 20.2  Hypopituitarism The hypothalamus is connected to the anterior pituitary by a delicate stalk that contains  the portal system through which releasing hormones, secreted from the hypothalamus,  gain access to the anterior pituitary cells. In the cell membranes of these cells are specific  receptors for releasing hormones. In most cases, different cells express different releasing  hormone receptors. The connection between the hypothalamus and anterior pituitary can  be disrupted by trauma or tumors. Trauma can occur in automobile accidents or other  local damaging events that may result in severing of the stalk and preventing the releasing  hormones from reaching their target anterior pituitary cells. When this happens, the  anterior pituitary cells no longer have their signaling mechanism for the release of anterior  pituitary hormones. In the case of tumors of the pituitary gland, all of the anterior pituitary  hormones may not be shut off to the same degree or the secretion of some may disappear  sooner than others. In any case, if hypopituitarism occurs this condition may result in a  life­threatening situation in which the clinician must determine the extent of loss of pituitary  hormones, especially ACTH. Posterior pituitary hormones—oxytocin and vasopressin— may also be lost, precipitating a problem of excessive urination (vasopressin deficiency)  that must be addressed. The usual therapy involves administration of the end organ  hormones, such as thyroid hormone, cortisol, sex hormones, and progestin; with female  patients it is also necessary to maintain the ovarian cycle. These hormones can easily be  administered in oral form. Growth hormone deficiency is not a problem in the adult but  would be an important problem in a growing child. The patient must learn to anticipate  needed increases of cortisol in the face of stressful situations. Fortunately, these patients  are usually maintained in reasonably good condition. Marshall, J. C., and Barkan, A. L. Disorders of the hypothalamus and anterior pituitary.  In: W. N. Kelley (Ed.), Internal Medicine. New York: Lippincott, 1989, p. 2159; and  Robinson, A. G. Disorders of the posterior pituitary. In: W. N. Kelley (Ed.), Internal  Medicine, New York: Lippincott, 1989, p. 2172.   hormones of the anterior pituitary are growth hormone (GH), thyrotropin or thyroid­stimulating hormone (TSH), adrenocorticotropic hormone (ACTH), b ­lipotropin  (b ­LTH), b ­endorphin (from pars intermedia­like cells), a ­MSH (from pars intermedia­like cells), b ­MSH (from pars intermedia­like cells), corticotropin­like  intermediary peptide (CLIP; from pars intermedia­like cells), prolactin (PRL), follicle­stimulating hormone (FSH), and luteinizing hormone (LH). Of these, all are  single polypeptide chains, except TSH, FSH, and LH, which are dimers that share a similar or identical subunit, the a  subunit. Since the intermediate lobe in humans is  rudimentary, the circulating levels of free a ­ and b ­MSH are relatively low. It is of interest, particularly in the human, that MSH receptors recognize and are activated  by ACTH, since the first 13 amino acids of ACTH contain the a ­MSH sequence. For this reason, ACTH may be an important contributing factor to skin pigmentation  and may exceed the importance of MSH, especially in conditions where the circulating level of ACTH is high. The clinical consequences of hypopituitarism are  presented in Clin. Corr. 20.2. 20.3—  Major Polypeptide Hormones and Their Actions Since cellular communication is so specific, it is not surprising that there are a large number of hormones in the body and new hormones continue to be discovered.  Limitations of space permit a summary of only a few of the well­characterized hormones. Table 20.2 presents some major polypeptide hormones and their actions. By  inspection of Table 20.2 it becomes evident that many hormones cause the release of other substances, some of which may themselves be hormones. This is  particularly the case for hormonal systems that are included in cascades like that presented in Figures 20.2 and 20.3. Other activities of receptor­hormone complexes  located in cell membranes are to increase the flux of ions into cells, particularly calcium ions, and to activate or suppress activities of enzymes in contact with the  receptor or a transducing protein with which the receptor interacts. Examples of these kinds of activities are discussed later in this chapter. In the functioning of most  membrane–receptor complexes,

 

  

Page 847 TABLE 20.2 Important Polypeptide Hormones in the Body and Their Actionsa Source

Hormone

Action

Hypothalamus

Thyrotropin­releasing hormone (TRH)

Acts on thyrotrope to release TSH

 

Gonadotropin­releasing hormone (GnRH)

Acts on gonadotrope to release LH and FSH from the  same cell

 

Growth hormone­releasing hormone or somatocrinin  (GRH)

Acts on somatotrope to release GH

 

Growth hormone release inhibiting hormone or  somatostatin (GIH)

Acts on somatotrope to prevent release of GH

 

Corticotropin­releasing hormone (CRH) Vasopressin is a helper hormone to CRH in releasing  ACTH; angiotensin II also stimulates CRH action in  releasing ACTH

Acts on corticotrope to release ACTH and b­li­  potropin

 

Prolactin­releasing factor (PRF) (not well established)

Acts on lactotrope to release PRL

 

Prolactin release inhibiting factor (PIF) (not well  established; may be a peptide hormone under control  of dopamine or may be dopamine itself)

Acts on lactotrope to inhibit release of PRL

Anterior pituitary

Thyrotropin (TSH)

Acts on thyroid follicle cells to bring about release of  T4 (T3)

 

Luteinizing hormone (LH) (human chorionic  gonadotropin, hCG, is a similar hormone from the  placenta)

Acts on Leydig cells of testis to increase testosterone  synthesis and release; acts on corpus luteum of ovary  to increase progesterone production and release

 

Follicle­stimulating hormone (FSH)

Acts on Sertoli cells of seminiferous tubule to increase  proteins in sperm and other proteins; acts on ovarian  follicles to stimulate maturation of ovum and  production of estradiol

 

Growth hormone (GH)

Acts on a variety of cells to produce IGFs (or  somatomedins), cell growth, and bone sulfation

 

Adrenocorticotropic hormone (ACTH)

Acts on cells in the adrenal gland to increase cortisol  production and secretion

 

b ­Endorphin

Acts on cells and neurons to produce analgesic and  other effects

 

Prolactin (PRL)

Acts on mammary gland to cause differentiation of  secretory cells (with other hormones) and to stimulate  synthesis of components of milk

 

Melanocyte­stimulating hormone (MSH)

Acts on skin cells to cause the dispersion of melanin  (skin darkening)

Ultimate gland hormones

Insulin­like growth factors (IGF)

Respond to GH and produce growth effects by  stimulating cell mitosis

 

Thyroid hormone (T4/T3) (amino acid­derived  hormone)

Responds to TSH and stimulates oxidation in many  cells

 

Opioid peptides

May derive as breakdown products of g ­lipotropin or b  ­endorphin or from specific gene products; can  respond to CRH or dopamine and may produce  analgesia and other effects

 

Inhibin

Responds to FSH in ovary and in Sertoli cell; regulates  secretion of FSH from anterior pituitary. Second form of  inhibin (activin) may stimulate FSH secretion

 

Corticotropin­like intermediary peptide (CLIP)

Derives from intermediate pituitary by degradation of  ACTH; contains b ­cell tropin activity, which  stimulates insulin release from b cells in presence of  glucose

  (continued)

 

  

Page 848

TABLE 20.2 (Continued) Source

Hormone

Action

Peptide hormones responding to  other signals than anterior pituitary  hormones

Arginine vasopressin (AVP; antidiuretic hormone,  ADH)

Responds to increase in osmoreceptor, which senses  extracellular [Na+]; increases water reabsorption from  distal kidney tubule

 

Oxytocin

Responds to suckling reflex and estradiol; causes milk  ''let down" or ejection in lactating female, involved in  uterine contractions of labor; luteolytic factor  produced by corpus luteum; decreases steroid  synthesis in testis

b Cells of pancreas respond to  glucose and other blood  constituents to release insulin

Insulin

Increases tissue utilization of glucose

a Cells of pancrease respond to low  levels of glucose and falling serum  calcium

Glucagon

Decreases tissue utilization of glucose to elevate blood  glucose

Derived from circulating blood  protein by actions of renin and  converting enzyme

Angiotensin II and III (AII and AIII)

Renin initially responds to decreased blood volume or  decreased [Na+] in the macula densa of the kidney.  AII/AIII stimulate outer layer of adrenal cells to  synthesize and release aldosterone

Released from heart atria in response  Atrial natriuretic factor (ANF) or atriopeptin to hypovolemia; regulated by other  hormones

Acts on outer adrenal cells to decrease aldosterone  release; has other effects also

Generates from plasma, gut, or other  Bradykinin tissues

Modulates extensive vasodilation resulting in  hypotension

Hypothalamus and intestinal  mucosa

Neurotensin

Effects on gut; may have neurotransmitter actions

Hypothalamus, CNS, and intestine

Substance P

Pain transmitter, increases smooth muscle contractions  of the GI tract

Nerves and endocrine cells of gut;  hypothermic hormone

Bombesin

Increases gastric acid secretion

 

Cholecystokinin (CCK)

Stimulates gallbladder contraction and bile flow;  increases secretion of pancreatic enzymes

Stomach antrum

Gastrin

Increases secretion of gastric acid and pepsin

Duodenum at pH values below 4.5

Secretin

Stimulates pancreatic acinar cells to release bicarbonate  and water to elevate duodenal pH

Hypothalamus and GI tract

Vasointestinal peptide (VIP)

Acts as a neurotransmitter in peripheral autonomic  nervous system; relaxes smooth muscles of circulation;  increases secretion of water and electrolytes from  pancreas and gut

Kidney

Erythropoietin

Acts on bone marrow for terminal differentiation and  initiates hemoglobin synthesis

Ovarian corpus luteum

Relaxin

Inhibits myometrial contractions; its secretion  increases during gestation

 

Human placental lactogen (hPL)

Acts like PRL and GH because of large amount of hPL  produced

Salivary gland

Epidermal growth factor

Stimulates proliferations of cells derived from ectoderm  and mesoderm together with serum; inhibits gastric  secretion

Thymus

Thymopoietin (a­thymosin)

Stimulates phagocytes; stimulates differentiation of  precursors into immune competent T cells

  (table continued on next page)

 

  

Page 849

TABLE 20.2 (Continued) Source

Hormone

Action

Parafollicular C cells of thyroid  gland

Calcitonin (CT)

Lowers serum calcium

Parathyroid glands

Parathyroid hormone (PTH)

Stimulates bone resorption; stimulates phosphate  excretion by kidney; raises serum calcium levels

Endothelial cells of blood vessels Endothelin

Vasoconstriction

Source: Part of this table is reproduced from Norman, A. W., and Litwack, G. Hormones. Orlando, FL: Academic Press, 1987. a This is only a partial list of polypeptide hormones in humans. TSH, thyroid­stimulating hormone or thyrotropin; LH, luteinizing hormone, 

FSH, follicle­stimulating hormone; GH, growth hormone; ACTH, adenocorticotropic hormone; PRL, prolactin; T4, thyroid hormone (also T);  IGF, insulin­like growth factor. For the releasing hormones and for some hormones in other categories, the abbreviation may contain "H" at the  end when the hormone has been well characterized, and "F" in place of H to refer to "Factor'' when the hormone has not been well  characterized. Names of hormones may contain "tropic" or "trophic" endings, tropic is mainly used here. Tropic refers to a hormone generating  a change, whereas trophic refers to growth promotion. Both terms can refer to the same hormone at different stages of development. Many of  these hormones have effects in addition to those listed here.

  stimulation of enzymes or flux of ions is followed by a chain of events, which may be described as intracellular cascades, during which a high degree of amplification is  obtained. 20.4—  Genes and Formation of Polypeptide Hormones Genes for polypeptide hormones contain the information for the hormone and the control elements upstream of the transcriptionally active sequence. In some cases,  more than one hormone is encoded in a gene. One example is proopiomelanocortin, a hormone precursor that encodes the following hormones: ACTH, b ­ lipotropin, and other hormones like g­lipotropin, g­MSH, a ­MSH, CLIP, b ­endorphin, and potentially b ­MSH and enkephalins. In the case of the posterior pituitary  hormones, oxytocin and vasopressin, information for these hormones are each encoded on a separate gene together with information for each respective neurophysin, a protein that binds to the completed hormone and stabilizes it. Proopiomelanocortin Is a Precursor Polypeptide for Eight Hormones Proopiomelanocortin, as schematized in Figure 20.5, can generate at least eight hormones from a single gene product. All products are not expressed simultaneously  in a single cell type, but occur in separate cells based on their content of specific proteases required to cleave the propeptide, specific metabolic controls, and the  presence of different positive regulators. Thus, while proopiomelanocortin is expressed in both the corticotropic cell of the anterior pituitary and the pars intermedia  cell, the stimuli and products are different as summarized in Table 20.3. The pars intermedia is a discrete anatomical structure located between the anterior and  posterior pituitary in the rat (Figure 20.6). In the human, however, the pars intermedia is not a discrete anatomical structure, although the cell type may be present in  the equivalent location. Many Polypeptide Hormones Are Encoded Together in a Single Gene An example of another gene and gene products encoding more than one peptide are the genes for vasopressin and oxytocin and their accompanying neurophysin  proteins, products that are released from the posterior pituitary upon specific stimulation. In much the same manner that ACTH and b ­lipotropin

 

  

Page 850

Figure 20.5  Proopiomelanocortin is a polypeptide product encoded by a single gene.  The dark vertical bars represent proteolytic cleavage sites for specific enzymes. The cleavage sites  are Arg­Lys, Lys­Arg, or Lys­Lys. Some specificity also may be conferred by neighboring amino acid  residues. In the corticotropic cell of the anterior pituitary, enzymes are present that cleave at sites 3 and 5,  releasing the major products, ACTH and b­lipotropin, into the general circulation. In the pars intermedia,  especially in vertebrates below humans, these products are further cleaved at major sites 4, 6, and 7 to  release a­MSH, CLIP, g­lipotropin, and b­endorphin into the general circulation. Some b­lipotropin arising  in the corticotroph may be further degraded to form b­endorphin. These two cell types appear to be under  separate controls. The corticotropic cell of the anterior pituitary is under the positive control of the CRH and its  auxiliary helpers, arginine vasopressin (AVP) and angiotensin II. AVP by itself does not release ACTH but  enhances the action of CRH in this process. The products of the intermediary pituitary, a­MSH, CLIP  (corticotropin­like intermediary peptide), g­lipotropin, and b­endorphin, are under the positive control of  norepinephrine, rather than CRH, for release. Obviously there must exist different proteases in these different  cell types in order to generate a specific array of hormonal products. b­Endorphin also contains a  pentapeptide, enkephalin, which potentially could be released at some point (hydrolysis at 8). TABLE 20.3 Summary of Stimuli and Products of  Proopiomelanocortina Cell type

Corticotroph

Pars intermedia

Stimulus

CRH (+)  (Cortisol (­))

Dopamine (­)  Norepinephrine (+)

Auxiliary  stimulus

AVP, AII

 

Major  products

ACTH, b ­lipotropin (b ­ endorphin)

aMSH, CLIP, g­lipotropin, b  ­endorphin

a

 CRH, corticotropin­releasing hormone; AVP, arginine vasopressin;  AII, angiotensin II; ACTH, adrenocorticotropin; a­MSH, a melanocyte­ stimulating hormone; CLIP, corticotropin­like intermediary peptide. Note: Although there are pars intermedia cells in the human pituitary  gland, they do not represent a distinct lobe.

   

  

Page 851

Figure 20.6  The hypothalamus with nuclei in various locations in which the  hypothalamic releasing hormones are synthesized.  Shown is the major vascular network consisting of a primary plexus where  releasing hormones enter its circulation through fenestrations and the  secondary plexus in the anterior pituitary where the releasing hormones are  transported out of the circulation, again through fenestrations in the vessels,  to the region of the anterior pituitary target cells. Also shown are the  resultant effects of the actions of the hypothalamic releasing hormones causing  the secretion into the general circulation of the anterior pituitary hormones.  Adapted from Norman, A. W., and Litwack, G. Hormones. New York: Academic  Press, 1987, p. 104.

(b ­LPH) are split out of the proopiomelanocortin precursor peptide, so are the products vasopressin, neurophysin II, and a glycoprotein of as yet unknown function  split out of the vasopressin precursor. A similar situation exists for oxytocin and neurophysin I (Figure 20.7). Vasopressin and neurophysin II are released by the activity of baroreceptors and osmoreceptors, which sense a fall in blood pressure or a rise in extracellular  sodium ion concentration, respectively. Generally, oxytocin and

Figure 20.7  Prepro­vasopressin and prepro­oxytocin.  Proteolytic maturation proceeds from top to bottom for each  precursor. The organization of the gene translation products is  similar in either case except that a glycopeptide is included on the  proprotein of vasopressin in the C­terminal region. Orange bars of the  neurophysin represent conserved amino acid regions; gray bars  represent variable C and N termini.  Redrawn with permission from Richter, D. VP and OT are expressed as  polyproteins. Trends Biochem. 8:278, 1983.

 

  

Page 852

Figure 20.8  Model of enkephalin precursor.  The distribution of Met­enkephalin sequences (M1–M 6) and  Leu­enkephalin (L) sequences within the protein precursor of bovine  adrenal medulla. CHO, potential carbohydrate attachment sites.  Redrawn from Comb, M., Seeburg, P. H., Adelman, J., Eiden, L., and  Herbert, E. Nature 295:663, 1982.

neurophysin I are released from the posterior pituitary by the suckling response in lactating females or by other stimuli mediated by a specific cholinergic mechanism.  Oxytocin–neurophysin I release can be triggered by injection of estradiol. Release of vasopressin–neurophysin II can be stimulated by administration of nicotine. The  two separate and specific releasing agents, estradiol and nicotine, prove that oxytocin and vasopressin, together with their respective neurophysins, are synthesized and  released from different cell types. Although oxytocin is well known for its milk­releasing action in the lactating female, in the male it seems to have a separate role  associated with an increase in testosterone synthesis in the testes. Other polypeptide hormones are being discovered that are co­encoded together by a single gene. An example is the discovery of the gene encoding GnRH, a  decapeptide that appears to reside to the left of a gene for the GnRH­associated peptide (GAP), which, like dopamine, may be capable of inhibiting prolactin release.  Thus both hormones—GnRH and the prolactin release inhibiting factor GAP—appear to be co­secreted by the same hypothalamic cells. Multiple Copies of a Hormone Can Be Encoded on a Single Gene An example of multiple copies of a single hormone encoded on a single gene is the gene product for enkephalins located in the chromaffin cell of the adrenal medulla.  Enkephalins are pentapeptides with opioid activity; methionine­enkephalin (Met­ENK) and leucine­enkephalin (Leu­ENK) have the structures:

A model of enkephalin precursor in adrenal medulla is presented in Figure 20.8, which encodes several Met­ENK (M) molecules and a molecule of Leu­

Figure 20.9  Nucleic acid sequence for rat proCRH genes.  Representation of the rat proCRH gene. Exons are shown as blocks and the intron by a  double red line. The TATA and CAAT sequence, putative cap site, translation initiation  ATG, translation terminator TGA, and poly(A) addition signals (AATAAA) are indicated.  The location of the CRH peptide is indicated by CRH.  Redrawn from Thompson, R. D., Seasholz, A. F., and Herbert, E. Mol. Endocrinol.  1:363, 1987.

 

  

Page 853

ENK (L). Again, the processing sites to release enkephalin molecules from the protein precursor involve Lys–Arg, Arg–Arg, and Lys–Lys bonds. Many genes for hormones are constructed to encode only one hormone and this may be the general situation. An example of a single hormone gene is shown in Figure  20.9. In this case the information for the hormone CRH is contained in the second exon and the information in the first exon is not expressed. Having cDNAs for use  as probes that contain the information for expression of CRH allows for the localization of the hormone in tissues. Previously it was thought that the hormone should be  restricted to the hypothalamus, the anterior pituitary, and the stalk, which contains the closed vascular transporting system (Figure 20.6). However, RNA extracts from  different tissues probed with this DNA reveal the location of CRH mRNA in testis, brain stem, and adrenal gland in addition to pituitary and hypothalamus. The  presence of the hormone in extrahypothalamic–pituitary axis tissues and its functions there are subjects of active investigation. 20.5—  Synthesis of Amino Acid­Derived Hormones Many hormones and neurotransmitters are derived from amino acids, principally from tyrosine and phenylalanine. Glutamate, aspartate, and other compounds are  important neurotransmitter substances as well. Although there may be some confusion about which compounds are neurotransmitters and which are hormones, it is  clear that epinephrine from the adrenal medulla is a hormone, whereas norepinephrine is a neurotransmitter. This section considers epinephrine and thyroxine or  triiodothyronine. The other biogenic amines, such as dopamine, which are considered to be neurotransmitters, are discussed in Chapter 22. Epinephrine Is Synthesized from Phenylalanine/Tyrosine The synthesis of epinephrine occurs in the adrenal medulla. A number of steroid hormones, including aldosterone, cortisol, and dehydroepiandrosterone (sulfate), are  produced in the adrenal cortex and are discussed in Chapter 21. The biochemical reactions leading to the formation of epinephrine from tyrosine or phenylalanine are  presented in Figure 20.10. Epinephrine is a principal hormone secreted from the adrenal medulla chromaffin cell along with some norepinephrine, enkephalins, and  some of the enzyme dopamine­b ­hydroxylase. Secretion of epinephrine is signaled by the neural response to stress, which is transmitted to the adrenal medulla by  way of a preganglionic acetylcholinergic neuron (Figure 20.11). Release of acetylcholine by the neuron increases the availability of intracellular calcium ion, which  stimulates exocytosis and release of the material stored in the chromaffin granules (Figure 20.11b). This overall system of epinephrine synthesis, storage, and release  from the adrenal medulla is regulated by neuronal controls and also by glucocorticoid hormones synthesized in and secreted from the adrenal cortex in response to  stress. Since the products of the adrenal cortex are transported through the adrenal medulla on their way out to the general circulation, cortisol becomes elevated in the  medulla and induces phenylethanolamine N­methyltransferase (PNMT), a key enzyme catalyzing the conversion of norepinephrine to epinephrine. Thus, in  biochemical terms, the stress response at the level of the adrenal cortex ensures the production of epinephrine from the adrenal medulla (Figure 20.12). Presumably,  epinephrine once secreted into the bloodstream not only affects a  receptors of hepatocytes to ultimately increase blood glucose levels as indicated, but also interacts  with a  receptors on vascular smooth muscle cells and on pericytes to cause cellular contraction and increase blood pressure.

 

  

Page 854

Figure 20.10  Biochemical steps in synthesis of epinephrine by  chromaffin cell of adrenal medulla.

Synthesis of Thyroid Hormone Requires Incorporation of Iodine into a Tyrosine of Thyroglobulin An outline of the biosynthesis and secretion of thyroid hormone, tetraiodo­L­thyronine (T4), also called thyroxine, and its active cellular counterpart, triiodo­L­ thyronine (T3) (structures presented in Figure 20.13) is presented in Figure 20.14. The thyroid gland is differentiated to concentrate iodide from the blood and  through the series of reactions shown in Figures 20.13 and 20.14, monoiodotyrosine (MIT), diiodotyrosine (DIT), T4, and T3 are produced within thyroglobulin  (TG). Thus the iodinated amino acids and thyronines are stored in the thyroid follicle as part of thyroglobulin. Recent work indicates that there

 

  

Page 855

Figure 20.11  Relationship of adrenal medulla chromaffin cells to preganglionic  neuron innervation and the structural elements involved in the synthesis of  epinephrine and the discharge of catecholamines in response to acetylcholine.  (a) Functional relationship between cortex and medulla for control of synthesis  of adrenal catecholamines. Glucocorticoids that stimulate enzymes catalyzing the  conversion of norepinephrine to epinephrine reach the chromaffin cells from  capillaries shown in (b).  (b) Discharge of catecholamines from storage granules in chromaffin cells after   nerve fiber stimulation, resulting in the release of acetylcholine. Calcium enters  the cells as a result, causing the fusion of granular membranes with the plasma  membrane and exocytosis of the contents.  Reprinted with permission from Krieger, D. T., and Hughes, J. C. (Eds.).  Neuroendocrinology. Sunderland, MA: Sinauer Associates, 1980.

are hot spots (regions for very active iodination) in the thyroglobulin sequence for the incorporation of iodine. Apparently, the sequences around iodotyrosyls occur in  three consensus groups: Glu/Asp­Tyr, associated with the synthesis of thyroxine or iodotyrosines; Ser/Thr­Tyr­Ser, associated with the synthesis of

Figure 20.12  Biosynthesis, packaging, and release of epinephrine in the adrenal medulla  chromaffin cell.  PNMT, phenylethanolamine N­methyltransferase; EP, epinephrine; NEP,  norepinephrine. Neurosecretory granules contain epinephrine, dopamine  b­hydroxylase, ATP, Met­ or Leu­enkephalin, as well as larger enkephalin­containing  peptides or norepinephrine in place of epinephrine. Epinephrine and norepinephrine  are stored in different cells. Enkephalins could also be contained in separate cells,  although that is not completely clear.  Adapted from Norman, A. W., and Litwack, G. Hormones.  New York: Academic Press, 1987, p. 464.

 

  

Page 856

Figure 20.13  Synthesis and structures of thyroid hormones T4, T3, and reverse T3.  Step 1, oxidation of iodide:  Step 2, iodination of tyrosine residues;  Step 3, coupling of DIT to DIT;  Step 4, coupling of DIT to MIT  (coupling may be intramolecular or intermolecular).

iodothyronine and iodotyrosine; and Glu­X­Tyr, associated with the remaining iodotyrosyls in the sequence. As depicted in Figure 20.14, secretion of T3 and T4 into  the bloodstream requires endocytosis of the thyroglobulin stored in the follicle and subsequent proteolysis within the epithelial cell. The released DIT and MIT are then  deiodinated and the released iodide ions are recycled and reutilized for hormone synthesis.

 

  

Page 857

Figure 20.14  Cellular mechanisms for T3 and T4 release into the bloodstream.  Iodide trapping at basal membrane of thyroid epithelium concentrates iodide  approximately 30­fold. Secretion of T3 and T4 into bloodstream requires  endocytosis of thyroglobulin and subsequent proteolysis. DIT and MIT are  deiodinated and the released iodide ions are reutilized for hormone synthesis.

20.6—  Inactivation and Degradation of Hormones Most polypeptide hormones are degraded to amino acids by hydrolysis, which presumably occurs in the lysosome. Partial hydrolysis by proteinases is a principal  pathway for degradation. Certain hormones, however, contain modified amino acids; for example, among the hypothalamic releasing hormones, the N­terminal amino  acid can be cycloglutamic acid (or pyroglutamic acid) (Table 20.4) and a C­terminal amino acid amide. Some of the releasing hormones TABLE 20.4 Hypothalamic Releasing Hormones Containing an N­Terminal Pyroglutamate, a a C­Terminal  Amino Acid Amide, or Both Hormone

Sequenceb

Thyrotropin­releasing hormone  pGlu­H­Pro­NH 2 (TRH) Gonadotropin­releasing  hormone (GnRH)

pGlu­HWSYGLRP­Gly­NH2

Corticotropin­releasing  hormone (CRH)

SQEPPISLDLTFHLLREVLEMTKADQLAQQAHSNRKLLDI­Ala­NH2

Somatocrinin (GRH)

YADAIFTNSYRKVLGQLSARKLLQDIMSRQQGESNQERG­ARAR­ Leu­ NH2

a

 The pyroglutamate structure is

  b Single­letter abbreviations used for amino acids: Ala, A; Arg, R; Asn, N; Asp, D; Cys, C; Glu, E, Gln, Q; 

Gly, G; His, H; Ile, I; Leu, L; Lys, K; Met, M; Phe, F; Pro, P; Ser, S; Thr, T; Trp, W; Tyr, Y; Val, V.

   

  

Page 858

that have either or both of these amino acid derivatives are listed in Table 20.4. Apparently, breakage of the cyclic glutamate ring or cleavage of the C­terminal amide  can lead to inactivation of many of these hormones and such enzymic activities have been reported in blood. This activity probably accounts, in part, for the short half­ life of many of these hormones. TABLE 20.5 Examples of Hormones Containing a Cystine  Disulfide Bridge Structure Hormone

Sequencea

Somatostatin

  Oxytocin

  Arginine vasopressin

  a

 Letters refer to single­letter amino acid abbreviations (see  Table 20.4)

  Some hormones contain a ring structure joined by a cystine disulfide bond. A few examples are given in Table 20.5. Peptide hormones, such as those shown in Table  20.5, may be degraded initially by the random action of cystine aminopeptidase and glutathione transhydrogenase as shown in Figure 20.15. Alternatively, as  has been suggested in the case of oxytocin, the peptide may be broken down through partial proteolysis to shorter peptides, some of which may have hormonal actions  on their own. Maturation of prohormones in many cases involves proteolysis, which may be considered as a degradation process in the sense that the prohormone is  degraded to active forms (e.g., Figure 20.5), although degradation is usually thought of as the reduction of active peptides to inactive ones.

Figure 20.15  Degradation of posterior pituitary hormones.  Oxytocin transhydrogenase is similar to degrading enzymes for insulin; presumably,  these enzymes also degrade vasopressin.  Redrawn from Norman, A. W., and Litwack, G. Hormones. New York:  Academic Press, 1987, p. 167.

 

  

Page 859

20.7—  Cell Regulation and Hormone Secretion Hormonal secretion is under specific control. In the cascade system displayed in Figures 20.2 and 20.3, hormones must emanate from one source, cause hormonal  release from the next cell type in line, and so on, down the cascade system. The correct responses must follow from a specific stimulus. The precision of these signals is  defined by the hormone and the receptor as well as by the activities of the CNS, which precedes the first hormonal response in many cases. Certain generalizations can  be made. Polypeptide hormones generally bind to their cognate receptors located in cell membranes. The receptor recognizes structural features of the hormone to  generate a high degree of specificity and affinity. The affinity constants for these interactions are in the range of 109–1011 M–1, representing tight binding. This interaction  usually activates or complexes with a transducing protein in the membrane, such as a G­protein (GTP­binding protein), or other transducer and causes an activation of  some enzymatic function on the cytoplasmic side of the membrane. In some cases receptors undergo internalization to the cell interior; these receptors may or may  not (e.g., the insulin receptor) be coupled to transducing proteins in the cell membrane. A discussion of internalization of receptors is presented in Section 20.11. The  ''activated" receptor complex could physically open a membrane ion channel or have other profound impacts on membrane structure and function. For example,  binding of the hormone to the receptor may cause conformational changes in the receptor molecule, enabling it to associate with transducer in which further  conformational changes may occur to permit interaction with an enzyme on the cytoplasmic side of the plasma membrane. This interaction may cause conformational  changes in an enzyme so that its catalytic site becomes active. G­Proteins Serve as Cellular Transducers of Hormone Signals Most transducers of receptors in the plasma membrane are GTP­binding proteins and are referred to as G­proteins. G­Proteins consist of three types of subunits— a , b , and g. The a  subunit is the guanine nucleotide­binding component and is thought to interact with the receptor indirectly through the b  and g subunits and then  directly with an enzyme, such as adenylate cyclase, resulting in enzyme activation. Actually there are two forms of the a  subunit, designated  s for a stimulatory subunit  and  i for an inhibitory subunit. Two types of receptors, and thus hormones, control the adenylate cyclase reaction: hormone–receptors that lead to a stimulation of the  adenylate cyclase and those that lead to an inhibition of the cyclase. This is depicted in Figure 20.16 with an indication of the role of  s and  i and some of the  hormones that interact with the stimulatory and inhibitory receptors.

Figure 20.16  Components that constitute a  hormone­sensitive adenylate cyclase system  and the subunit composition.  Adenylate cyclase is responsible for conversion  of ATP to cAMP. The occupancy of R  by  s

stimulatory hormones stimulates adenylate  cyclase via formation of an active  dissociated Ga  subunit. The occupancy  s

of Ri by inhibitory hormones results in the  formation of an "active" Ga  complex  i

and concomitant reduction in cyclase  activity. The fate of b and g subunits in  these dissociation reactions is not yet  known. Rs, stimulatory hormone receptor;  Ri, inhibitory hormone receptor.

 

  

Page 860

The sequence of events that occurs when hormone and receptor interact is presented in Figure 20.17 and is as follows: receptor binds hormone in the membrane (Step  1); which produces a conformational change in receptor to expose a site for G­protein (b , g subunit) attachment (Step 2); G­protein can be either stimulatory, Gs, or  inhibitory, Gi, referring to the ultimate effects on the activity of adenylate cyclase; the receptor interacts with b , g subunit of G­protein, enabling the a  subunit to  exchange GTP for bound GDP (Step 3); dissociation of GDP causes separation of G­protein a  subunit from b , g subunit and the a ­binding site for interaction with  adenylate cyclase appears on the surface of the G­protein a  subunit (Step 4); a  subunit binds to adenylate cyclase and activates the catalytic center, so that ATP is  converted to cAMP (Step 5); GTP is hydrolyzed to GDP by the GTPase activity of the a  subunit, returning it to its original conformation and allowing its interaction  with b , g subunit once again (Step 6); GDP associates with the a  subunit and the system is returned to the unstimulated state awaiting another cycle of activity. It is  important to note that there is also evidence suggesting that the b , g complexes may play important roles in regulating certain effectors including adenylate cyclase. In the case where an inhibitory G­protein is coupled to the receptor, the events are similar but inhibition of adenylate cyclase activity may arise by direct interaction of  the inhibitory a  subunit with adenylate cyclase or, alternatively, the inhibitory a  subunit may interact directly with the stimulatory a  subunit on the other side and  prevent the stimulation of adenylate cyclase activity indirectly. Immunochemical evidence suggests multiple Gi subtypes and molecular cloning of complementary DNAs  encoding putative a  subunits has also provided evidence for multiple  i subtypes. Purification and biochemical characterization of G­proteins (Gs as well as Gi) have revealed somewhat unanticipated diversity in this subfamily. Polymerase chain  reaction­based cloning has now brought the number of distinct genes encoding mammalian a  subunits to at least 15. With regard to a  subunits, further diversity is  achieved by alternative splicing of the  s (four forms) gene. There also appears to be diversity among the mammalian b  and g subunits. At least four distinct b  subunit  cDNAs and probably as many g subunits have been described. What is not clear is how these complexes combine to form distinct b , g complexes. Some data suggest  that different b , g complexes may have distinct properties with respect to a  subunit and receptor interactions, but additional research will be required to fully describe  these unique interactions. Table 20.6 lists some activities transduced by G­protein subfamilies. TABLE 20.6 Activities Transduced by G­Protein Subfamilies

a  Subunit

Expression

Effector

 

Gs

Ubiquitous

 

 Adenylate cyclase, Ca2+ channel

Golf

Olfactory

 

 Adenylate cyclase

G+1 (transducin)

Rod photoreceptors

 

 cGMP­phosphodiesterase

G+2 (transducin)

Cone photoreceptors

 

 cGMP­phosphodiesterase

Gi1

Neural > other tissues 

Gi2

Ubiquitous

Gi3

Other tissues > neural

Go

Neural, endocrine

Gq

Ubiquitous 

G11

Ubiquitous

G14

Liver, lung, kidney

G15/16

Blood cells

 Adenylate cyclase

   Ca2+ channel

 

 Phospholipase C

 

Source: Adapted from Spiegel, A. M., Shenker, A., and Weinstein, L. S. Endocr. Rev. 13:536, 1992.

   

  

Page 861

Figure 20.17  Activation of adenylate cyclase by binding  of a hormone to its receptor.  The cell membrane is depicted, which contains on its outer  surface a receptor protein for a hormone. On the inside  surface of the membrane is adenylate cyclase protein and  the transducer protein G. In the resting state GDP is  bound to the a subunit of the G­protein. When a hormone  binds to the receptor, a conformational change occurs  (Step 1). The activated receptor binds to the G­protein  (Step 2), which activates the latter so that it releases GDP  and binds GTP  (Step 3), causing the a and the complex of b and  g­subunits to dissociate  (Step 4). Free G  subunit binds to the adenylate cyclase  a

and activates it so that it catalyzes the synthesis  of cAMP from ATP  (Step 5); this step may involve a conformational change in Ga.  In some cases the b,g complex may play an important role  in regulation of certain effectors including adenylate  cyclase. When GTP is hydrolyzed to GDP, a reaction most  likely catalyzed by Ga itself, Ga is no longer able to  activate adenylate cyclase  (Step 6), and Ga and Gbg reassociate. The hormone  dissociates from the receptor and the system returns to its  resting state.  Redrawn from Darnell, J., Lodish, H., and Baltimore, D.  Molecular Cell Biology. New York: Scientific American  Books, 1986, p. 682.

 

  

Page 862

Cyclic AMP Activates Protein Kinase A Pathway The generation of cAMP in the cell usually activates protein kinase A, referred to as the protein kinase A pathway. The overall pathway is presented in Figure  20.18. Four cAMP molecules are used in the reaction to complex two regulatory subunits (R) and liberating two protein kinase catalytic subunits (C). The liberated  catalytic subunits are able to phosphorylate proteins to produce a cellular effect. In many cases the cellular effect leads to the release of preformed hormones. For  example, ACTH binds to membrane receptors, elevates intracellular cAMP levels, and releases cortisol from the zona fasciculata cells of the adrenal gland by this  general mechanism. Part of the mechanism of release of thyroid hormones from the thyroid gland involves the cAMP pathway as outlined in Figure 20.19. TSH has  been shown to stimulate numerous key steps in this secretory process, including iodide uptake and endocytosis of thyroglobulin (Figure 20.14). The protein kinase A  pathway is also responsible for the release of testosterone by testicular Leydig cells as presented in Figure 20.20. There are many other examples of hormonal actions  mediated by cAMP and the protein kinase A pathway. Inositol Triphosphate Formation Leads to Release of Calcium from Intracellular Stores Uptake of calcium from the cell exterior through calcium channels may be affected directly by hormone­receptor interaction at the cell membrane. In some cases,  ligand­receptor interaction is thought to open calcium channels directly in the cell membrane (Chapter 5, Section 5.5). Another system to increase intracellular Ca2+  concentration derives from hormone­receptor activation of phospholipase C activity transduced by a G­protein (Figure 20.21). A hormone operating through this system binds to a specific cell membrane receptor, which interacts with a G­protein in a mechanism similar to that of the protein  kinase A pathway and transduces the signal, resulting in stimulation of phospholipase C. This enzyme catalyzes the hydrolysis of phosphatidylinositol­4,5­ bisphosphate (PIP2) to form two second messengers, diacylglycerol (DAG) and inositol 1,4,5­triphosphate (IP3). Inositol 1,4,5­triphosphate diffuses to the cytosol and binds to an IP3 receptor on the membrane of a particulate calcium store, either separate from or

Figure 20.18  Activation of protein kinase A.  Hormone–receptor mediated stimulation of  adenylate cyclase and subsequent activation of  protein kinase A. C, catalytic subunit; R,  regulatory subunit.

 

  

Page 863

Figure 20.19  Overview of secretion controls of thyroid  hormone.

CLINICAL CORRELATION 20.3  Lithium Treatment of Manic–Depressive Illness: The Phosphatidylinositol  Cycle Lithium has been used for years in the treatment of manic depression. Our newer  knowledge suggests that lithium therapy involves the phosphatidylinositol (PI) pathway.  This pathway generates the second messengers inositol 1,4,5­triphosphate (IP3) and  diacylglycerol following the hormone/neurotransmitter–membrane receptor interaction and  involves the G­protein complex and activation of phospholipase C. IP3 and its many  phosphorylated derivatives are ultimately dephosphorylated in a stepwise fashion to  generate free inositol. Inositol is then used for the synthesis of phosphatidylinositol  monophosphate. The phosphatase that dephosphorylates IP to inositol is inhibited by Li+.  In addition, Li+ may also interfere directly with G­protein function. The result of Li+  inhibition is that the PI cycle is greatly slowed even in the face of continued  hormonal/neurotransmitter stimulation and the cell becomes less sensitive to these stimuli.  Manic–depressive illness may occur through the overactivity of certain CNS cells,  perhaps as a result of abnormally high levels of hormones or neurotransmitters whose  actions are to stimulate the PI cycle. The chemotherapeutic effect of the Li+ could be to  decrease the cellular responsiveness to elevated levels of agents that might promote high  levels of PI cycle and precipitate manic­depressive illness. Avissar, S., and Schreiber, G. Muscarinic receptor subclassification and G­proteins:  significance for lithium action in affective disorders and for the treatment of the  extrapyramidal side effects of neuroleptics. Biol. Psychiatry 26:113, 1989; Hallcher, L.  M., and Sherman, W. R The effects of lithium ion and other agents on the activity of  myoinositol 1­phosphatase from bovine brain. J. Biol. Chem. 255:896, 1980; and  Pollack, S. J., Atack, J. R., Knowles, M. R., McAllister, G., Ragan, C. I., Baker, R.,  Fletcher, S. R., Iversen, L. L., and Broughton, H. B. Mechanism of inositol  monophosphatase, the putative target of lithium therapy. Proc. Natl. Acad. Sci. USA  91:5766, 1994.   part of the endoplasmic reticulum. IP3 binding results in the release of calcium ions contributing to the large increase in cytosolic Ca2+ levels. Calcium ions may be  important to the process of exocytosis by taking part in the fusion of secretory granules to the internal cell membrane, in microtubular aggregation or in the function of  contractile proteins, which may be part of the structure of the exocytotic mechanism, or all of these. The IP3 is metabolized by stepwise removal of phosphate groups (Figure 20.21) to form inositol. This combines with phosphatidic acid (PA) to form  phosphatidylinositol (PI) in the cell membrane. PI is phosphorylated twice by a kinase to form PIP2, which is ready to undergo another round of hydrolysis and  formation of second messengers (DAG and IP3) upon hormonal stimulation. If the receptor is still occupied by hormone, several rounds of the cycle could occur before  the hormone–receptor complex dissociates or some other feature of the cycle becomes limiting. It is interesting that the conversion of inositol phosphate to inositol is  inhibited by lithium ion (Li+) (Figure 20.21). This could be the metabolic basis for the beneficial effects of Li+ in manic­depressive illness (see Clin. Corr. 20.3).  Finally, it is important to note that not all of the generated IP3 is dephosphorylated during hormonal stimulation. Some of the IP3 is phosphorylated via IP3 kinase to  yield inositol 1,3,4,5­tetraphosphate (IP4), which may mediate some of the slower or more prolonged hormonal responses or facilitate replenishment of intracellular  Ca2+ stores from the extracellular fluid, or both.

 

  

Page 864

Figure 20.20  Overview of the secretion controls and some general actions of  the gonadotropes and testosterone release in males.  In some, but not all, androgen target cells, testosterone is reduced to  the more potent androgen, 5a­ di­hydrotestosterone (5a­DHT).

 

  

Page 865

Figure 20.21  Overview of hormonal signaling through the phosphatidylinositol system generating the second messengers, inositol  1,4,5­trisphosphate (IP ) and diacylglycerol (DAG).  3

The action of IP3 is to increase cytosol Ca2+ levels by a receptor­mediated event in the cellular  calcium store. Steps in pathway: (1) binding of hormone to cell membrane receptor; (2)  production of IP3 from PIP2; (3) binding of IP3 to receptor on calcium storage site; (4) release  of free calcium to the cytosol; (5) release of DAG and subsequent binding to protein kinase  C; (6) phosphorylation of protein substrates by protein kinase C activated by DAG and Ca2+;   and (7) phosphorylation of IP  to yield IP . DAG, diacylglycerol; PA, phosphatidic acid; IP, inositol  3

4

phosphate; IP2, inositol bisphosphate; IP3, inositol 1,4,5­triphosphate; IP4, inositol 1,3,4,5­  tetrakisphosphate; PIP, phosphatidylinositol phosphate; PIP , phosphatidylinositol  2

4,5­bisphosphate; K, kinase; E, esterase.

Diacylglycerol Activates Protein Kinase C Pathway At the same time that the IP3 produced by hydrolysis of PIP2 is increasing the concentration of Ca2+ in the cytosol, the other cleavage product, DAG, mediates  different effects. Importantly, DAG activates a crucial serine/threonine protein kinase called protein kinase C because it is Ca2+ dependent (details of protein kinase  C discussed on p. 883). The initial rise in cytosolic Ca2+ induced by IP3 is believed to somehow alter kinase C so that it translocates from the cytosol to the  cytoplasmic face of the plasma membrane. Once translocated, it is activated by a combination of Ca2+, DAG, and the negatively charged membrane phospholipid,  phosphatidylserine. Once activated, protein kinase C then phosphorylates specific proteins in the cytosol or, in some cases, in the plasma membrane. These  phosphorylated proteins perform specific functions that they could not mediate in their nonphosphorylated states. For example, a phosphorylated protein could  potentially migrate to the nucleus and stimulate mitosis and

 

  

Page 866

growth. It is also possible that a phosphorylated protein could play a role in the secretion of preformed hormones. 20.8—  Cyclic Hormonal Cascade Systems Hormonal cascade systems can be generated by external signals as well as by internal signals. Examples of this are the diurnal variations in levels of cortisol secreted  from the adrenal gland probably initiated by serotonin and vasopressin, the day and night variations in the secretion of melatonin from the pineal gland and the internal  regulation of the ovarian cycle. Some of these biorhythms operate on a cyclic basis, often dictated by daylight and darkness, and are referred to as chronotropic  control of hormone secretion. Melatonin and Serotonin Synthesis Are Controlled by Light and Dark Cycles The release of melatonin from the pineal gland, presented in overview in Figure 20.22, is an example of a biorhythm. Here, as in other such systems, the internal signal  is provided by a neurotransmitter, in this case norepinephrine produced by an adrenergic neuron. In this system, control is exerted by light entering the eyes and is  transmitted to the pineal gland by way of the CNS. The adrenergic neuron innervating the pinealocyte is inhibited by light transmitted through the eyes. Norepinephrine  released as a neurotransmitter in the dark stimulates cAMP formation through a b  receptor in the pinealocyte cell membrane, which leads to the enhanced synthesis of  N­acetyltransferase. The increased activity of this enzyme causes the conversion of serotonin to N­acetylserotonin, and hydroxyindole­O­methyltransferase  (HIOMT) then catalyzes the conversion of N­acetylserotonin to melatonin, which is secreted in the dark hours but not during light hours. Melatonin is circulated to  cells containing receptors that generate effects on reproductive and other functions. For example, melatonin has been shown to exert an antigonadotropic effect,  although the physiological significance of this effect is unclear.

Figure 20.22  Biosynthesis of melatonin in pinealocytes.  HIOMT, hydroxyindole­O­methyltransferase.  Redrawn from Norman, A. W., and Litwack, G. Hormones. New York: Academic  Press, 1987, p. 710.

 

  

Page 867

Ovarian Cycle Is Controlled by Gonadotropin­Releasing Hormone An example of a pulsatile release mechanism is regulation of the periodic release of GnRH. A periodic control regulates the release of this substance at definitive  periods (of about 1 h in higher animals) and is controlled by aminergic neurons, which may be adrenergic (norepinephrine secreting) in nature. The initiation of this  function occurs at puberty and is important in both the male and female. While the male system functions continually, the female system is periodic and known as the  ovarian cycle. This system is presented in Figure 20.23. In the

Figure 20.23  Ovarian cycle in terms of generation of hypothalamic  hormone, pituitary gonadotropic hormones, and  sex hormones.  To begin the cycle at puberty, several centers in the  CNS coordinate with the hypothalamus so that  hypothalamic GnRH can be released in a pulsatile  fashion. This causes the release of the gonadotropic  hormones, LH and FSH, which in turn affect the  ovarian follicle, ovulation, and the corpus luteum.  The hormone inhibin selectively inhibits FSH  secretion. Products of the follicle and corpus luteum,  respectively, are b­estradiol and progesterone.  GnRH, gonadotropin­releasing hormone; FSH,  follicle­stimulating hormone; LH, luteinizing  hormone.

 

  

Page 868

male, the cycling center in the CNS does not develop because its development is blocked by androgens before birth. In the female, a complicated set of signals needs to be organized in the CNS before the initial secretion of GnRH occurs at puberty. The higher centers (CNS  organizer) must harmonize with the tonic and cycling centers and these interact with each other to prime the hypothalamus. The pulsatile system, which innervates the  arcuate nucleus of the hypothalamus, must also function for GnRH to be released, and this system apparently must be functional throughout life for these cycles to be  maintained. Release of GnRH from the axon terminals of the cells that synthesize this hormone is followed by entry of the hormone into the primary plexus of the closed  portal system connecting the hypothalamus and the anterior pituitary (Figure 20.23). The blood–brain barrier preventing peptide transport is overcome in this process  by allowing GnRH to enter the vascular system through fenestrations, or openings in the blood vessels, that permit such transport. The GnRH is then carried down the  portal system and leaves the secondary plexus through fenestrations, again, in the region of the target cells (gonadotropes) of the anterior pituitary. The hormone  binds to its cognate membrane receptor and the signal, mediated by the phosphatidylinositol metabolic system, causes the release of both FSH and LH from the same  cell. The FSH binds to its cognate membrane receptor on the ovarian follicle and, operating through the protein kinase A pathway via cAMP elevation, stimulates  synthesis and secretion of 17b ­estradiol, the female sex hormone, and maturation of the follicle and ovum. Other proteins, such as inhibin, are also synthesized. Inhibin  is a negative feedback regulator of FSH production in the gonadotrope. When the follicle reaches full maturation and the ovum also is matured, LH binds to its cognate  receptor and plays a role in ovulation together with other factors, such as prostaglandin F2a. The residual follicle remaining after ovulation becomes the functional  corpus luteum under primary control of LH (Figure 20.23). The LH binds to its cognate receptor in the corpus luteum cell membrane and, through stimulation of the  protein kinase A pathway, stimulates synthesis of progesterone, the progestational hormone. Estradiol and progesterone bind to intracellular receptors (Chapter 21)  in the uterine endometrium and cause major changes resulting in the thickening of the wall and vascularization in preparation for implantation of the fertilized egg.  Estradiol, which is synthesized in large amount prior to production of progesterone, induces the progesterone receptor as one of its inducible phenotypes. This  induction of progesterone receptors primes the uterus for subsequent stimulation by progesterone secreted by the corpus luteum. Absence of Fertilization If fertilization of the ovum does not occur, the corpus luteum involutes as a consequence of diminished LH supply. Progesterone levels fall sharply in the blood with  the regression of the corpus luteum. Estradiol levels also fall due to the cessation of its production by the corpus luteum. Thus the stimuli for a thickened and  vascularized uterine endometrial wall are lost. Menstruation occurs through a process of programmed cell death of the uterine endometrial cells until the endometrium  reaches its unstimulated state. Ultimately, the fall in blood steroid levels releases the negative feedback inhibition on the gonadotropes and hypothalamus and the cycle  starts again with release of FSH and LH by the gonadotropes in response to GnRH. The course of the ovarian cycle is shown in Figure 20.24 with respect to the relative blood levels of hormones released from the hypothalamus, anterior pituitary,  ovarian follicle, and corpus luteum. In addition, changes in the maturation of the follicle and ovum as well as the uterine endometrium are shown. Aspects of the steroid  hormones, estradiol and progesterone, are discussed in Chapter 21. The cycle first begins at puberty when GnRH is released, corresponding

 

  

Page 869

Figure 20.24  The ovarian cycle.  In the upper diagram, relative blood levels of GnRH, LH, FSH, progesterone, estrogen, and  PGF2a are shown. In the lower diagram, events in the ovarian follicle, corpus luteum, and  uterine endometrium are diagrammed. GnRH, gonadotropin­releasing hormone;  LH, luteinizing hormone; FSH, follicle­stimulating hormone; PGF2 ,  a

prostaglandin F2a; E2, estradiol; E2R, intracellular estrogen receptor; PR,  intracellular progesterone receptor.

to day 1 in Figure 20.24. GnRH is released in a pulsatile fashion, causing the gonadotrope to release FSH and LH; there is a rise in the blood levels of these  gonadotropic hormones in subsequent days. Under the stimulation of FSH the follicle begins to mature (lower section of Figure 20.24) and estradiol (E2) is produced.  In response to estradiol the uterine endometrium begins to thicken (there would have been no prior menstruation in the very first cycle). Eventually, under the continued  action of FSH, the follicle matures with the maturing ovum, and extraordinarily high levels of estradiol are produced (around day 13 of the cycle). These levels of  estradiol, instead of causing feedback inhibition, now generate, through feedback stimulation, a huge release of LH and to a lesser extent FSH from the  gonadotrope. The FSH responds to a smaller extent due to the ovarian production of the hormone inhibin under the influence of FSH. Inhibin is a specific negative  feedback inhibitor of FSH, but not of LH, and probably suppresses the synthesis of the b  subunit of FSH. The high midcycle peak of LH is referred to as the ''LH  spike." Ovulation then occurs at about day 14 (midcycle) through the effects of high LH concentration together with other factors, such as PGF2a. Both LH and PGF2a  act on cell membrane receptors. After ovulation, the function of the follicle declines as reflected by the fall in blood estrogen levels. The spent follicle now differentiates  into the functional corpus luteum driven by the still high levels of blood LH (Figure 20.23, top).

 

  

Page 870

Under the influence of prior high levels of estradiol (estrogen) and the high levels of progesterone produced by the now functional corpus luteum, the uterine  endometrial wall reaches its greatest development in preparation for implantation of the fertilized egg, should fertilization occur. Note that the previous availability of  estradiol in combination with the estrogen receptor (E2R) produces elevated levels of progesterone receptor (PR) within the cells of the uterine wall. The blood levels  of estrogen fall with the loss of function of the follicle but some estrogen is produced by the corpus luteum in addition to the much greater levels of progesterone. In  the absence of fertilization the corpus luteum continues to function for about 2 weeks, then involutes because of the loss of high levels of LH. The production of  oxytocin by the corpus luteum itself and the production or availability of PGF2a cause inhibition of progesterone synthesis and enhances luteolysis by a process of  programmed cell death (Chapter 21). With the death of the corpus luteum there is a profound decline in blood levels of estradiol and progesterone so that the  thickened endometrial wall can no longer be maintained and menstruation occurs, followed by the start of another cycle with a new developing follicle. Fertilization The situation changes if fertilization occurs as shown in Figure 20.25. The corpus luteum, which would have ceased function by 28 days, remains viable due to the  production of chorionic gonadotropin, which resembles and acts like LH, from the trophoblast. Eventually, the production of human chorionic gonadotropin  (hCG) is taken over by the placenta, which continues to produce the hormone at very high levels throughout most of the gestational period. Nevertheless, the corpus  luteum, referred to as the "corpus luteum of pregnancy," eventually dies and, by about 12 weeks of pregnancy, the placenta has taken over the production of  progesterone, which is secreted at high levels throughout pregnancy. Although both progesterone and estrogen are secreted in progressively greater quantities  throughout pregnancy, from the seventh month onward estrogen secretion continues to increase while progesterone secretion remains constant or may even decrease  slightly (Figure 20.25). The increased production of a progesterone­binding protein may also serve to lower the effective concentration of free progesterone in the  myometrium. Thus the estrogen/progesterone ratio increases toward the end of pregnancy and may

Figure 20.25  Effect of fertilization on ovarian cycle in terms of progesterone  and secretion of human chorionic gonadotropin (hCG).

 

  

Page 871

be partly responsible for the increased uterine contractions. Oxytocin secreted by the posterior pituitary also contributes to these uterine contractions. The fetal  membranes also release prostaglandins (PGF2a) at the time of parturition and they also increase the intensity of uterine contractions. Finally, the fetal adrenal glands  secrete cortisol, which not only stimulates fetal lung maturation by inducing surfactant but may also stimulate uterine contractions. As mentioned before, the system in the male is similar, but less complex in that cycling is not involved, and it progresses much as outlined in Figure 20.25. This is only  one example of biorhythmic and pulsatile systems. 20.9—  Hormone–Receptor Interactions Receptors are proteins and differ by their specificity for ligands and by their location in the cell (see Figure 20.1). The interaction of ligand with receptor essentially  resembles a semienzymatic reaction:

The hormone–receptor complex usually undergoes conformational changes resulting from interaction with the hormonal ligand. These changes allow for a subsequent  interaction with a transducing protein (G­protein) in the membrane or for activation to a new state in which active domains become available on the surface of the  receptor. The mathematical treatment of the interaction of hormone and receptor is a function of the concentrations of the reactants, hormone [H] and receptor [R], in  the formation of the hormone–receptor complex [RH], and the rates of formation and reversal of the reaction:

The reaction can be studied under conditions, such as low temperature, that will further reduce reactions involving the hormone–receptor complex. The equilibrium can  thus be expressed in terms of the association constant, Ka, which is equal to the inverse of the dissociation constant, Kd:

The concentrations are equilibrium concentrations that can be restated in terms of the forward and reverse velocity constants, k +1 being the on­rate and k –1 being the  off­rate (on refers to hormone association with the receptor and off refers to hormone dissociation). Experimentally, equilibrium under given conditions is determined  by a progress curve of binding that reaches saturation. A saturating amount of hormone is determined using variable amounts of free hormone and measuring the  amount bound with some convenient assay. The half­maximal value of a plot of receptor­bound hormone (ordinate) versus total free­hormone concentration (abscissa)  approximates the dissociation constant, which will have a specific hormone concentration in molarity as its value. Hormone bound to receptor is corrected for  nonspecific binding of the hormone to the membrane or other nonreceptor intracellular proteins. This can be measured conveniently if the hormone is radiolabelled, by  measuring receptor binding using labeled hormone ("hot" or "uncompeted") and receptor binding using labeled hormone after the addition of an excess (100–1000  times) of unlabeled hormone ("hot" + "cold" or competed). The excess of unlabeled hormone will displace the high­affinity hormone­binding sites but not the low­ affinity nonspecific binding sites. Thus when the ''competed" curve is subtracted from the "uncompeted" curve, as seen in Figure 20.26, an intermediate curve will  represent specific binding of labeled hormone to receptor. This is of critical

Figure 20.26  Typical plot showing specific hormone  binding.

 

  

Page 872

importance when receptor is measured in a system containing other proteins. As an approximation, 20 times the Kd value of hormone is usually enough to saturate the  receptor. Scatchard Analysis Permits Determination of the Number of Receptor­Binding Sites and Association Constant for Ligand Most measurements of Kd are made using Scatchard analysis, which is a manipulation of the equilibrium equation. The equation can be developed by a number of  routes but can be envisioned from mass action analysis of the equation presented above. At equilibrium the total possible number of binding sites (Bmax) equals the  unbound plus the bound sites, so that Bmax = R + RH, and the unbound sites (R) will be equal to R = Bmax – RH. To consider the sites left unbound in the reaction the  equilibrium equation becomes

Thus

The Scatchard plot of bound/free = [RH]/[H] on the ordinate versus bound on the abscissa yields a straight line, as shown in Figure 20.27. When the line is  extrapolated to the abscissa, the intercept gives the value of Bmax (the total number of specific receptor­binding sites). The slope of the negative straight line is –Ka or – 1/Kd. These analyses are sufficient for most systems but become more complex when there are two components in the Scatchard plot. In this case the straight line usually  curves as it approaches the abscissa and a second phase is observed somewhat asymptotic to the abscissa while still retaining a negative slope (Figure 20.28a). In  order to obtain the true value of Kd for the steeper, higher­affinity sites, the low­affinity curve must be subtracted from the first set, which also corrects the extrapolated  value of Bmax. From these analyses information is obtained on Kd, the number of classes of binding sites (usually one or two), and the maximal number of high­affinity  receptor sites (receptor number) in the system (see Figure 20.28b). These curvilinear Scatchard plots can result not only from the existence of more than one distinct  binding component but also as a consequence of what is referred to as negative cooperativity. This term refers to the fact that in some systems the affinity of the  receptor for its ligand is gradually decreased as more and more ligand binds. From application to a wide variety of systems it appears that Kd values for many hormone  receptors range from 10–9 to 10–11 M, indicating very tight binding. These interactions are generally marked by a high degree of specificity so that both parameters  describe interactions of a high order, indicating the uniqueness of receptors and the selectivity of signal reception.

Figure 20.27  Typical plot of Scatchard analysis of  specific binding of ligand to receptor.

Some Hormone–Receptor Interactions Involve Multiple Hormone Subunits Interaction of hormone and receptor can be exemplified by the anterior pituitary hormones, thyrotropin (TSH), luteinizing hormone (LH), and follicle­stimulating  hormone (FSH). These hormones each contain two subunits, an a  and a b  subunit. The a  subunit for all three hormones is nearly identical and the a  subunit of any of  the three can substitute for the other two. Consequently, the a  subunit performs some function in common to all three hormones in their interaction with receptor but is  obviously not responsible for the specificity

 

  

Page 873

Figure 20.28  Scatchard analysis of curves representing two components.  (a) Scatchard curve showing two components.  (b) Scatchard plot with correction of high­affinity component  by subtraction of nonspecific binding attributable to the  low­affinity component. Curve 1: total binding. Curve 2:  Linear extrapolation of high­affinity component that includes  contribution from low­affinity component. Curve 3:  Specific binding of high­affinity component after removal  of nonspecific component.  Redrawn from Chamness, G. C., and McGuire, W.  L. Steroids 26:538, 1975.

required for each cognate receptor. The hormones cannot replace each other in binding to their specific receptor. Thus the specificity of receptor recognition is  imparted by the b  subunit, whose structure is unique for the three hormones. On the basis of topological studies with monoclonal antibodies, a picture of the interaction of LH with its receptor has been suggested as shown in Figure 20.29. In this  model, the receptor recognizes both subunits of the hormonal ligand, but the b  subunit is specifically recognized by the receptor to lead to a response. With the TSH– receptor complex there may be more than one second messenger generated. In addition to the stimulation of adenylate cyclase and the increased intracellular level of  cAMP, the phosphatidylinositol pathway (Figure 20.21) is also turned on. The preferred model is one in which there is a single receptor whose interaction with  hormone activates both the adenylate cyclase and the phospholipid second messenger systems, as shown in Figure 20.30. Thus a variety of reactions could follow the  hormone–receptor interaction through the subsequent stimulation of cAMP levels (protein kinase A pathway) and stimulation of phosphatidylinositol turnover (protein  kinase C pathway).

 

  

Page 874

Figure 20.29  Interaction of the a  and b  subunits  of LH with the LH receptor of rat Leydig cells.  The interaction was determined by  topological analysis with monoclonal  antibodies directed against epitopes on  the a and b subunits of the  hormone. Both a and b subunits  participate in LH receptor binding.  Adapted from Alonoso­Whipple, C.,  Couet, M. L., Doss, R., Koziarz, J.,  Ogunro, E. A., and Crowley, W. E. Jr.  Endocrinology 123:1854, 1988.

Figure 20.30  Model of TSH receptor, which is composed  of glycoprotein and ganglioside component.  After the TSH b subunit interacts  with receptor, the hormone changes its  conformation and the a subunit is brought  into the bilayer, where it interacts  with other membrane components. The  b subunit of TSH may  carry primary determinants recognized  by the glycoprotein receptor component.  It is suggested that the TSH signal to  adenylate cyclase is via the  ganglioside; the glycoprotein component  appears more directly linked to  phospholipid signal system. PI,  phosphatidylinositol; G ,  s

G­protein linked to activation of    adenylate cyclase; Gq G­protein linked to PI cycle.  Adapted with modifications from L. D. Kohn,  et al. Biochemical Actions of  Hormones, 12. G. Litwack (Ed.).  Academic Press, 1985, p. 466.

 

  

Page 875

20.10—  Structure of Receptors: b ­Adrenergic Receptor Structures of receptors are conveniently discussed in terms of functional domains. Consequently, for membrane receptors there will be functional ligand­binding  domains and the transmembrane domains, which for many membrane receptors involve protein kinase activities. In addition, specific immuno­logical domains  contain primary epitopes of antigenic regions. Several membrane receptors have been cloned and studied with regard to structure and function, including the b   receptors ( 1 and  2), which recognize catecholamines, principally norepinephrine, and stimulate adenylate cyclase. The  1 and  2 receptors are subtypes that differ in  affinities for norepinephrine and for synthetic antagonists. Thus  1­adrenergic receptor binds norepinephrine with a higher affinity than epinephrine, whereas the order of  affinities is reversed for the  2­adrenergic receptor. The drug isoproterenol has a greater affinity for both receptors than the two hormones. In Figure 20.31 the amino  acid sequence is shown (with single letter abbreviations for amino acids; see Table 20.4 for list) for the  2­adrenergic receptor. A polypeptide stretch extending from a helix I extends to the extracellular space. There are seven membrane­spanning domains and these appear also in the  1 receptor, where there is extensive homology  with the  2 receptor. Cytosolic peptide regions extend to form loops between I and II, III and IV, and V and VI and an extended chain from VII.

Figure 20.31  Proposed model for insertion of the b 2­adrenergic receptor (AR) in the cell membrane.  The model is based on hydropathicity analysis of the human  2­AR. Standard one­letter codes for amino acid  residues are used. Hydrophobic domains are represented as transmembrane helices. Pink circles with black letters  indicate residues in the human sequence that differ from those in hamster. Also noted are the potential sites of  N­linked glycosylation.  Redrawn from Kobilka, B. K., Dixon, R. A., Frielle, T., Dohlman, H. G., et al. Proc. Natl. Acad. Sci.  USA 84:46, 1987.

 

  

Page 876

Figure 20.32  Proposed arrangement of b ­adrenergic  receptor helices in the membrane.  Lower portion of the figure is a view from  above the plane of the plasma membrane.  It is proposed that helices IV, VI, and VII  are arranged in the membrane in such a  way as to delineate a ligand­binding  pocket, with helix VII centrally located.  Adapted from Frielle, T.,  Daniel, K. W., Caron, M. G., and  Lefkowitz, R. J. Proc. Natl. Acad.  Sci. USA 85:9494, 1988.

The long extended chain from VII may contain phosphorylation sites (serine and threonine residues) of the receptor, which are important in terms of the receptor  regulatory process involving receptor desensitization. Phosphorylation of these residues within the cytoplasmic tail of the receptor results in the binding of an inhibitory  protein, called b  arrestin, which blocks the receptor's ability to activate Gs. Cell exterior peptide loops extend from II to III, IV to V, and VI to VII, but mutational  analysis suggests that the external loops do not take part in ligand binding. It appears that ligand binding may occur in a pocket arranged by the location of the  membrane­spanning cylinders I–VII, which for the  1 receptor appear to form a ligand pocket, as shown from a top view in Figure 20.32. Recently reported work  suggests that transmembrane domain VI may play a role in the stimulation of adenylate cyclase activity. By substitution of a specific cysteine residue in this  transmembrane domain, a mutant was generated that displays normal ligand­binding properties but a decreased ability to stimulate the cyclase. 20.11—  Internalization of Receptors Up to now we have described receptor systems that transduce signals through other membrane proteins, such as G­proteins, which move about in the fluid

Figure 20.33  Diagrammatic summary of the morphological pathway of endocytosis in cells.  The morphological elements of the pathway of endocytosis are not drawn to scale.  The ligands shown as examples are EGF, transferrin, and  ­macroglobulin. EGF is  2

an example of a receptor system in which both ligand and receptor are delivered to  lysosomes; transferrin is shown as an example of a system in which both the ligand  and receptor recycle to the surface;  ­macroglobulin is shown as an example of a  2

system in which the ligand is delivered to lysosomes but the receptor recycles  efficiently back to the cell surface via the Golgi apparatus.  Adapted from Pastan, I., and Willingham, M. C. (Eds.). Endocytosis.  New York: Plenum Press, 1985, p. 3.

 

  

Page 877

cell membrane. However, many types of cell membrane hormone–receptor complexes are internalized, that is, moved from the cell membrane to the cell interior by a  process called endocytosis. This would represent the opposite of exocytosis in which components within the cell are moved to the cell exterior. The process of  endocytosis as presented in Figure 20.33 involves the polypeptide–receptor complex bound in coated pits, which are indentations in the plasma membrane that  invaginate into the cytosol and pinch off from the membrane to form coated vesicles. The vesicles shed their coats, fuse with each other, and form vesicles called  receptosomes. The receptors and ligands on the inside of these receptosomes can have different fates. Receptors can be recycled to the cell surface following fusion  with the Golgi apparatus. Alternatively, the vesicles can fuse with lysosomes for degradation of both the receptor and hormone. In addition, some hormone–receptor  complexes are dissociated in the lysosome and only the hormone is degraded, while the receptor is returned intact to the membrane. In some systems, the receptor  may also be concentrated in coated pits in the absence of exogenous ligand and cycle in and out of the cell in a constitutive, nonligand­dependent manner. Clathrin Forms a Lattice Structure to Direct Internalization of Hormone–Receptor Complexes from the Plasma Membrane The major protein component of the coated vesicle is clathrin, a nonglycosylated protein of mol wt 180,000 whose amino acid sequence is highly conserved. The  coated vesicle contains 70% clathrin, 5% polypeptides of about 35 kDa, and 25% polypeptides of 50–100 kDa. Aspects of the structure of a coated vesicle are  shown in Figure 20.34. Coated vesicles have a lattice­like surface

Figure 20.34  Structure and assembly of a coated vesicle.  (a) A typical coated vesicle contains a membrane vesicle about 40 nm in diameter  surrounded by a fibrous network of 12 pentagons and 8 hexagons. The fibrous coat is  constructed of 36 clathrin triskelions. One clathrin triskelion is centered on each of the  36 vertices of the coat. Coated vesicles having other sizes and shapes are believed to  be constructed similarly: each vesicle contains 12 pentagons but a variable number  of hexagons.  (b) Detail of a clathrin triskelion. Each of three clathrin heavy chains is bent into a  proximal arm and a distal arm. A clathrin light chain is attached to each heavy chain,  most likely near the center.  (c) An intermediate in the assembly of a coated vesicle, containing 10 of the final 36  triskelions, illustrates the packing of the clathrin triskelions. Each of the 54 edges of a  coated vesicle is constructed of two proximal and two distal arms intertwined. The  36 triskelions contain 36 × 3 = 108 proximal and 108 distal arms, and the coated vesicle  has precisely 54 edges.  See Crowther, R. A., and Pearse, B. M. F. J. Cell Biol. 91:790, 1981.  Redrawn from Nathk, I. S., Heuser, J., Lupas, A., Stock, J., Turck, C. W., and  Brodsky, E. M. Cell 68:899, 1992.  Redrawn from Darnell, J., Lodish, H., and Baltimore, D. Molecular Cell  Biology. New York: Scientific American Books, 1986, p. 647.

 

  

Page 878

structure comprised of hexagons and pentagons. Three clathrin molecules generate each polyhedral vertex and two clathrin molecules contribute to each edge. The  smallest such structure would contain 12 pentagons with 4–8 hexagons and 84 or 108 clathrin molecules. A 200­nm diameter coated vesicle contains about 1000  clathrin molecules. Clathrin can form flexible lattice structures that can act as scaffolds for vesicular budding. Completion of the budding process results in the mature  vesicle being able to enter the cycle. The events following endocytosis are not always clear with respect to a specific membrane receptor system. This process can be a means to introduce the intact  receptor or ligand to the cell interior in cases where the nucleus is thought to contain a receptor or ligand­binding site. Consider, for example, growth factors that are  known to bind to a cell membrane receptor but trigger events leading to mitosis. It is possible that signal transmission occurs by the alteration of a specific cytosolic  protein, perhaps by membrane growth factor receptor­associated protein kinase activity, resulting in the nuclear translocation of the covalently modified cytosolic  protein. In the case of internalization, delivery of an intact ligand (or portion of the ligand) could interact with a nuclear receptor. Such mechanisms are speculative.  Nevertheless, these ideas could constitute a rationale for the participation of endocytosis in signal transmission to intracellular components. Endocytosis renders a cell less responsive to hormone. Removal of the receptor to the interior, or cycling of membrane components, alters responsiveness or  metabolism (e.g., glucose receptors can be shuttled between the cell interior and the cell membrane under the control of hormones in certain cells). In another type of  downregulation, a hormone–receptor complex translocated to the nucleus can repress its own receptor mRNA levels by interacting with a specific DNA sequence.  More about this form of receptor downregulation is mentioned in Chapter 21. 20.12—  Intracellular Action: Protein Kinases Many amino acid­derived hormones or polypeptides bind to cell membrane receptors (except for thyroid hormone) and transmit their signal by (1) elevation of cAMP  and transmission through the protein kinase A pathway; (2) triggering of the hydrolysis of phosphatidylinositol 4,5­bisphosphate and stimulation of the protein  kinase C and IP3–Ca2+ pathways; or (3) stimulation of intracellular levels of cGMP and activation of the protein kinase G pathway. There are also other less  prevalent systems for signal transfer, which, for example, affect molecules in the membrane like phosphatidylcholine. As previously discussed in the case of TSH– receptor signaling, it may be possible that two of these pathways are activated. The cAMP system operating through protein kinase A activation has been described. Specific proteins are expected to be phosphorylated by this kinase compared to  other protein kinases, such as protein kinase C. Both protein kinase A and C phosphorylate proteins on serine or threonine residues. An additional protein kinase  system involves phosphorylation of tyrosine, which occurs in cytoplasmic domains of some membrane receptors especially growth factor receptors. This system is  important for the insulin receptor, IGF receptor, and certain oncogenes discussed below. The cellular location of these protein kinases is presented in Figure 20.35. The catalytic domain in the protein kinases is similar in amino acid sequence, suggesting that they have all evolved from a common primordial kinase. The three  tyrosine­specific kinases shown in Figure 20.35 are transmembrane receptor proteins that, when activated by the binding of specific extracellular ligands,  phosphorylate proteins (including themselves) on tyrosine residues inside the cell. Both chains of the insulin receptor are encoded by a single

 

  

Page 879

Figure 20.35  Protein kinases showing the size and location of their catalytic domain.  In each case the catalytic domain (red region) is about 250 amino acid residues long. The regulatory  subunits normally associated with A­kinase and with phosphorylase kinase are not shown. EGF, epidermal  growth factor; NGF, nerve growth factor; VEGF, vascular endothelial growth factor.  Redrawn from Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. Molecular Biology  of the Cell, 3rd ed. New York: Garland Publishing, 1994, p. 760.

gene, which produces a precursor protein that is cleaved into the two disulfide­linked chains. The extracellular domain of the PDGF receptor is thought to be folded  into five immunoglobulin (Ig)­like domains, suggesting that this protein belongs to the Ig superfamily. Proteins that are regulated by phosphorylation–dephosphorylation can have multiple phosphorylation sites and may be phosphorylated by more than one class of  protein kinase. Insulin Receptor:  Transduction through Tyrosine Kinase From Figure 20.35 it is seen that the a  subunits of the insulin receptor are located outside the cell membrane and apparently serve as the insulin­binding site. The  insulin–receptor complex undergoes an activation sequence probably involving conformational changes and phosphorylation (autophosphorylation) of tyrosine  residues located in the cytoplasmic portion of the receptor b  subunits). This results in activation of the tyrosine kinase activity located in the b  subunit, which is now  able to phosphorylate cytosolic proteins that may carry the insulin signal to the interior of the cell. The net results of these phosphorylation events include a series of  short­term metabolic effects, such as increased uptake of glucose, as well as longer­term effects of insulin on cellular differentiation and growth. Although, as already  indicated, the insulin receptor itself is a tyrosine kinase that is activated upon hormone binding, the

 

  

Page 880

Figure 20.36  Hypothetical model depicting two separate biochemical pathways to  explain paradoxical effects of insulin on protein phosphorylation.  Insulin simultaneously produces increases in the serine/threonine phosphorylation  of some proteins and decreases in others. This paradoxical effect may result from  activation of both kinases and phosphatases. Model explains (1) the generation of a  soluble second messenger that directly or indirectly activates serine/threonine  phosphatase and (2) the stimulation of a cascade of protein kinases, resulting in  phosphorylation of cellular proteins.  Redrawn from Saltiel, A. R. The paradoxical regulation of protein phosphorylation  in insulin action. FASEB J. 8:1034, 1994.

subsequent changes in phosphorylation occur predominantly on serine and threonine residues, as indicated in Figure 20.36. As also shown, insulin can simultaneously  stimulate the phosphorylation of some proteins and the dephosphorylation of other proteins. Either of these biochemical events can lead to activation or inhibition of  specific enzymes involved in mediating the effects of insulin. These opposite effects (phosphorylation and dephosphorylation) mediated by insulin suggest that perhaps  separate signal transduction pathways may originate from the insulin receptor to produce these pleiotropic actions. A hypothetical scheme for this bifurcation of signals  in insulin's action is presented in Figure 20.37. The substrates of the insulin–receptor tyrosine kinase are an important current research effort since phosphorylated  proteins could produce the long­term effects of insulin. On the other hand, there is evidence that an insulin second messenger may be developed at the cell membrane  to account for the short­term metabolic effects of insulin. The substance released as a result of insulin–insulin receptor interaction may be a glycoinositol derivative that,  when released from the membrane into the cytosol, could be a stimulator of phosphoprotein phosphatase. This activity would dephosphorylate a variety of enzymes,  either activating or inhibiting them, and produce effects already known to be associated with the action of insulin. In addition, this second messenger, or the direct  phosphorylating activity of the receptor tyrosine kinase, might explain the movement of glucose receptors (transporters) from the cell interior to the surface to account  for enhanced cellular glucose utilization in cells that utilize this mechanism to control glucose uptake. These possibilities are reviewed in Figure 20.37. Activation of the  enzymes indicated in this figure leads to increased metabolism of glucose while inhibition of the enzymes indicated leads to decreased breakdown of glucose or fatty  acid stores. Activity of Vasopressin:  Protein Kinase A An example of the activation of the protein kinase A pathway by a hormone is the activity of arginine vasopressin (AVP) on the distal kidney cell. Here the action of  vasopressin (VP), also called the antidiuretic hormone (Table 20.5),

 

  

Page 881

Figure 20.37  Hypothetical scheme for signal transduction in insulin action.  The insulin receptor undergoes tyrosine autophosphorylation and subsequent kinase  activation upon hormone binding. The receptor phosphorylates intracellular substrates  including IRS­1 and Shc proteins, which associate with SH2­containing proteins like  p85, SYP, or Grb2 upon phosphorylation. Formation of the IRS­1–p85 complex  activates PI 3­kinase; the IRS­1–SYP complex activates SYP, leading to MEK activation.  Formation of the Shc–Grb2 complex mediates the stimulation of P21Ras GTP binding,  leading to a cascade of phosphorylations. These phosphorylations probably occur  sequentially and involve raf proto­oncogene, MEK, MAP kinase, and S6 kinase II.  The receptor is probably separately coupled to activation of a specific phospholipase  C that catalyzes the hydrolysis of the glycosyl­PI molecules in the plasma membrane.  A product of this reaction, inositol phosphate glycan (IPG), may act as a second  messenger, especially with regard to activation of serine/threonine phosphatases  and the subsequent regulation of lipid and glucose metabolism. Abbreviations:  IRS­1, insulin receptor substrate­1; SH, src homology; MAP kinase,  mitogen­activated protein kinase; MEK, MAP kinase kinase; GPI,  glycosylphosphatidylinositol; PLC, phospholipase; SOS, son of sevenless.  Redrawn from Saltiel, A. R. The paradoxical regulation of protein  phosphorylation in insulin action. FASEB J. 8:1034, 1994.

is to cause increased water reabsorption from the urine in the distal kidney. A mechanism for this system is shown in Figure 20.38. Neurons synthesizing AVP  (vasopressinergic neurons) are signaled to release AVP from their nerve endings by interneuronal firing from a baroreceptor responding to a fall in blood pressure or  from an osmoreceptor (probably an interneuron), which responds to an increase in extracellular salt concentration. The high extracellular salt concentration apparently  causes shrinkage of the osmoreceptor cell and generates an electrical signal transmitted down the axon of the osmoreceptor to the cell body of the VP neuron  generating an action potential. This signal is then transmitted down the long axon from the VP cell body to its nerve ending where, by depolarization, the VP– neurophysin II complex is released in to the

 

  

Page 882

Figure 20.38  Secretion and action of arginine vasopressin in the distal kidney.  The release of arginine vasopressin (AVP or VP) from the posterior pituitary begins with a signal from the osmoreceptor, or  baroreceptor (not shown), in the upper right­hand corner of figure. The signal can be an increase in the extracellular concentration  of sodium chloride, which causes the osmoreceptor neuron to shrink and send an electrical message down its axon, which interfaces  with the vasopressinergic cell body. This signal is transmitted down the long axon of the vasopressinergic neuron and depolarizes the  nerve endings causing the release, by exocytosis, of the VP–neurophysin complex stored there. They enter the local circulation  through fenestrations in the vessels and perfuse the general circulation. Soon after release, neurophysin dissociates from VP and VP  binds to its cognate receptor in the cell membrane of the kidney distal tubule cell (other VP receptors are located on the corticotrope of  the anterior pituitary and on the hepatocytes and their mechanisms in these other cells are different from the one for the kidney tubule  cell). NPII, neurophysin II; VP, vasopressin; R, receptor; AC, adenylate cyclase; MF, myofibril; GP, glycogen phosphorylase; PK ,  i

inactive protein kinase; PKa , active protein kinase; R­Ca, regulatory subunit–cyclic AMP complex; TJ, tight junction; PD,  phosphodiesterase. Vasopressin–neurophysin complex dissociates at some point and free VP binds to its cell membrane receptor in  the plasma membrane surface. Through a G­protein adenylate cyclase is stimulated on the cytoplasmic side of the cell membrane,  generating increased levels of cAMP from ATP. Cyclic AMP­dependent protein kinases are stimulated and phosphorylate various  proteins (perhaps including microtubular subunits) which, through aggregation, insert as water channels (aquaporins) in the luminal  plasma membrane, thus increasing the reabsorption of water by free diffusion.  Redrawn in part from Dousa, T. P., and Valtin, H. Cellular actions of vasopressin in the mammalian  kidney. Kidney Int. 10:45, 1975.

extracellular space. The complex enters local capillaries through fenestrations and progresses to the general circulation. The complex dissociates and free VP is able to  bind to its cognate membrane receptors in the distal kidney, anterior pituitary, hepatocyte, and perhaps other cell types. After binding to the kidney receptor, VP  causes stimulation of adenylate cyclase through the stimulatory G­protein and activates protein kinase A. The protein kinase phosphorylates

 

  

Page 883 TABLE 20.7 Examples of Hormones that Operate Through the Protein Kinase A Pathway Hormone

Location of Action

CRH

Corticotrope of anterior pituitary

TSH (also phospholipid metabolism?)

Thyroid follicle

LH

Leydig cell of testis

 

Mature follicle at ovulation and corpus  luteum

FSH

Sertoli cell of seminiferous tubule Ovarian follicle

ACTH

Inner layers of cells of adrenal cortex

Opioid peptides

Some in CNS function on inhibitory pathway   through  Gi

AVP

Kidney distal tubular cell (the AVP hepatocyte receptor  causes phospholipid turn­over and calcium ion uptake;  the AVP receptor in anterior pituitary causes  phospholipid turnover)

PGI2 (prostacyclin)

Blood platelet membrane

Norepinephrine/epinephrine

b ­Receptor

  microtubular subunits that aggregate to form specific water channels, referred to as aquaporins, which are inserted into the luminal membrane for admission of larger  volumes of water than would occur by free diffusion. Water is transported across the kidney cell to the basolateral side and to the general circulation, causing a dilution  of the original high salt concentration (signal) and an increase in blood pressure. These aquaporins, which are a family of integral membrane proteins that function as  selective water channels, consist of six transmembrane a  helical domains. Although aquaporin monomers function as water channels or pores, their stability and proper  functioning may require a tetrameric assembly. Specific mutations in the amino acid sequences of the intracellular and extracellular loops of these proteins result in  nonfunctional aquaporins and the development of diabetes insipidus, which is characterized by increased thirst and production of a large volume of urine. Some hormones that operate through the protein kinase A pathway are listed in Table 20.7. Gonadotropin­Releasing Hormone (GnRH):  Protein Kinase C Table 20.8 presents examples of polypeptide hormones that stimulate the phosphatidylinositol pathway. An example of a system operating through stimulation of the  phosphatidylinositol pathway and subsequent activation of the protein kinase C system is GnRH action, shown in Figure 20.39. Probably under aminergic  interneuronal controls, a signal is generated to stimulate the cell body of the GnRH­ergic neuron where GnRH is synthesized. The signal is transmitted down the long  axon to the nerve ending where the hormone is stored. The hormone is released from the nerve ending by exocytosis resulting from depolarization caused by signal  transmission. The GnRH enters the primary plexus of the closed portal system connecting the hypothalamus and anterior pituitary through fenestrations. Then GnRH  exits the closed portal system through fenestrations in the secondary plexus and binds to cognate receptors in the cell membrane of the gonadotrope (see enlarged view  in Figure 20.39). The signal from the hormone–receptor complex is transduced (through a G­protein) and phospholipase C is activated. This enzyme catalyzes the  hydrolysis of PIP2 to form DAG and IP3. Diacylglycerol activates protein kinase C, which phosphoryl­

 

  

Page 884 TABLE 20.8 Examples of Polypeptide Hormones that Stimulate the  Phosphatidylinositol Pathway Hormone

Location of Action

TRH

Thyrotrope of the anterior pituitary releasing TSH

GnRH

Gonadotrope of the anterior pituitary releasing LH and  FSH

AVP

Corticotrope of the anterior pituitary; assists CRH in  releasing ACTH; hepatocyte: causes increase in cellular  Ca2+

TSH

Thyroid follicle: releasing thyroid hormones causes  increase in phosphatidylinositol cycle as well as increase  in protein kinase A pathway

Angiotensin II/III

Zona glomerulosa cell of adrenal cortex: releases  aldosterone

Epinephrine (thrombin)

Platelet: releasing ADP/serotonin; hepatocyte via a  receptor: releasing intracellular Ca2+

  ates specific proteins, some of which may participate in the resulting secretory process to transport LH and FSH to the cell exterior. The product IP3, which binds to a  receptor on the membrane of the calcium storage particle, probably located near the cell membrane, stimulates the release of calcium ion. Elevated cytosolic Ca2+  causes increased stimulation of protein kinase C and participates in the exocytosis of LH and FSH from the cell.

Figure 20.39  Overview of regulation of secretion of LH and FSH.  A general mode of action of GnRH to release the gonadotropes from the gonadotropic cell of the anterior  pituitary is presented. GnRH, gonadotropin­releasing hormone; FSH, follicle­stimulating hormone; LH,  luteinizing hormone; DAG, diacylglycerol.

 

  

Page 885

Figure 20.40  Common structure of protein kinase C subspecies.  Modified from U. Kikkawa, A. Kishimoto, and Y. Nishizuka, Annu. Rev. Biochem.  58:31, 1989.

Much recent work has focused on protein kinase C. It has been shown to have a number of subspecies; such heterogeneity may indicate that there are multiple  functions for this critical enzyme (Figure 20.40). The enzyme consists of two domains, a regulatory and a catalytic domain, which can be separated by proteolysis at a  specific site. The free catalytic domain, formerly called protein kinase M, can phosphorylate proteins free of the regulatory components. The free catalytic subunit,  however, may be degraded. More needs to be learned about the dynamics of this system and the translocation of the enzyme from one compartment to another. The  regulatory domain contains two Zn2+ fingers usually considered to be hallmarks of DNA­binding proteins (see Chapter 3). This DNA­binding activity has not yet been  demonstrated for protein kinase C and metal fingers may participate in other types of interactions. The ATP­binding site in the catalytic domain contains the G box,  GXGXXG, which is a consensus sequence for ATP binding with a downstream lysine residue. Activity of Atrial Natriuretic Factor (ANF):  Protein Kinase G The third system is the protein kinase G system, which is stimulated by the elevation of cytosolic cGMP (Figure 20.41). Cyclic GMP is synthesized by guanylate  cyclase from GTP. Like adenylate cyclase, guanylate cyclase is linked to a specific biological signal through a membrane receptor. The guanylate cyclase extracellular  domain may serve the role of the hormone receptor. This is directly coupled to the cytosolic domain through one membrane­spanning domain (Figure 20.42), which  may be applicable to the atrial natriuretic factor (ANF; also referred to as atriopeptin) receptor–guanylate cyclase system. Thus the hormone­binding site,  transmembrane domain, and guanylate cyclase activities are all served by a single polypeptide chain.

Figure 20.41  Structure of cGMP.

This hormone is a family of peptides, as shown in Figure 20.43; a sequence of human ANF is shown at the bottom. The functional domains of the ANF receptor are  illustrated in Figure 20.44. Atrial natriuretic factor is released from atrial cells of the heart under control of several hormones. Data from atrial cell culture suggest that  ANF secretion is stimulated by activators of protein kinase C and decreased by activators of protein kinase A. These opposing actions may be mediated by the  actions of a ­ and b ­adrenergic receptors, respectively. An overview of the secretion of ANF and its general effects is shown in Figure 20.45. ANF is released by a  number of signals, such as blood volume expansion, elevated blood pressure directly induced by vasoconstrictors, high salt intake,

 

  

Page 886

Figure 20.42  Model for the regulation of guanylate cyclase  activity after peptide hormone binding.  The enzyme exists in a highly phosphorylated  state under normal conditions. Binding of  hormone markedly enhances enzyme activity,  followed by a rapid dephosphorylation of  guanylate cyclase and a return of activity to  basal state despite continued presence of  hormonal peptide.  Redrawn from Schultz, S., Chinkers, M., and  Garbers, D. L. FASEB J. 3:2026, 1989.

Figure 20.43  Atrial natriuretic peptides.  These active peptides relax vascular smooth muscle and produce vasodilation and  natriuresis as well as other effects discussed in the text.  Adapted from Carlin, M., and Genest, J. The heart and the atrial  natriuretic factor. Endocr. Rev. 6:107, 1985.

 

  

Page 887

Figure 20.44  Functional domains of ANF­R1 receptor.  Hypothetical model shows the sequence of an ANF­binding domain, a  membrane­spanning domain(s), a proteolysis­sensitive region, a guanylate  cyclase catalytic domain, glucosylation sites (CHO), and amino  (H N) and carboxyl terminals (COOH) of receptor.  2

Redrawn from Liu, B., Meloche, S., McNicoll, N., Lord, C., and DeLéan, A.  Biochemistry 28:5599, 1989.

and increased heart pumping rate. ANF is secreted as a dimer that is inactive for receptor interaction and is converted in plasma to a monomer capable of interacting  with receptor. The actions of ANF (Figure 20.45) are to increase the glomerular filtration rate without increasing renal blood flow, leading to increased urine volume  and excretion of sodium ion. Renin secretion is also reduced and aldosterone secretion by the adrenal cortex is lowered. This action reduces aldosterone­mediated  sodium reabsorption. ANF inhibits the vasoconstriction produced by angiotensin II and relaxes the constriction of the renal vessels, other vascular beds, and large  arteries. ANF operates through its mem­

Figure 20.45  Schematic diagram of atrial natriuretic factor–atriopeptin hormonal system.  Prohormone is stored in  granules located in perinuclear atrial cardiocytes. An elevated vascular  volume results in cleavage and release of atriopeptin, which acts on the  kidney (glomeruli and papilla) to increase the glomerular filtration rate  (GFR), to increase renal blood flow (RBF), to increase urine volume (UV)  and sodium excretion (U ), and to decrease plasma renin  Na

activity. Natriuresis and diuresis are also enhanced by the suppression  of aldosterone secretion by the adrenal cortex and the release from the  posterior pituitary of arginine vasopressin. Vasodilatation of blood  vessels also results in a lowering of blood pressure (BP). Diminution of  vascular volume provides a negative feedback signal that suppresses  circulating levels of atriopeptin.  Redrawn from Needleman, P., and Greenwald, J. E. Atriopeptin:  a cardiac hormone intimately involved in fluid, electrolyte, and blood  pressure homeostasis. N. Engl. J.  Med. 314:828, 1986.

 

  

Page 888

brane receptor, which appears to be the extracellular domain of guanylate cyclase. The cGMP produced activates protein kinase G, which further phosphorylates  cellular proteins to express many of the actions of this pathway. More needs to be learned about protein kinase G. Using analogs of ANF it has been shown that the  majority of receptors expressed in the kidney are biologically silent, since they fail to elicit a physiological response. This new class of receptors may serve as specific  peripheral storage–clearance binding sites and as such act as a hormonal buffer system to modulate plasma levels of ANF. 20.13—  Oncogenes and Receptor Functions Oncogenes are genes that are expressed by cancerous transformed cells. A cancer cell may express few or many oncogenes that dictate the aberrant uncontrolled  behavior of the cell. There are three mechanisms by which oncogenes allow a cell to escape dependence on exogenous growth factors; these are presented in Figure  20.46. Some oncogenes are genes for parts of receptors, most often related to growth factor hormone receptors, which can function in the absence of the hormonal  ligand. Thus an oncogene may represent a truncated gene where the ligand­binding domain is missing. This would result in production of the receptor protein, insertion  into the cell membrane, and continuous constitutive function in the absence or presence of ligand (Figure 20.46b,c). In this situation the second messengers would be  produced constitutively at a high rate, instead of being regulated by ligand, and the result would be uncontrolled growth of the cell. Some oncogenes may have tyrosine  protein kinase activity and therefore function like tyrosine kinase normally related to certain cell membrane receptors. Other oncogenes relate to thyroid and steroid  hormone receptors (see Chapter 21) while still others are DNA­binding proteins, some of which may be transactivating factors or related to such factors. Oncogene­ encoded proteins that bind to DNA may be identical with or related to transactivating factors. The oncogene Jun, for example, is a component of activator protein 1  (AP1), a transactivating factor that regulates transcription. Table 20.9 reviews some of the oncogenes, or cancer­causing genes, together with the functions of their  proto­oncogenes (normal proliferation gene).

Figure 20.46  Mechanisms by which oncogenes  can allow a cell to escape dependence on  exogenous growth factors.  (a) By autocrine mechanism, where the cytosolic  oncogene indirectly stimulates expression of  growth factor gene and oversecretion of growth  factors, which then over­stimulates receptors  on same cell;  (b) by receptor alteration so that receptor is  ''permanently turned on" without a requirement  for growth factor binding; and  (c) by transducer alteration, where the intermediate  between the receptor and its resultant activity,  that is, the GTP­stimulatory protein, is  permanently activated, uncoupling the normal  requirement of ligand–receptor binding.  Redrawn from Weinberg, R. A. The action of  oncogenes in the cytoplasm and nucleus.  Science 230:770, 1985.

 

  

Page 889 TABLE 20.9 Known Oncogenes, Their Products and Functionsa

 

 

Name of  Oncogene

  Retrovirus

Oncogenic Protein Virus­Induced Tumor

src

Chicken sarcoma Chicken sarcoma

yes

Chicken sarcoma

fgr

Cat sarcoma

abl

Mouse leukemia Human leukemia

fps

Chicken sarcoma

fes

Cat sarcoma

ros

Cellular Location

Proto­oncogene Function

Plasma membrane

Tyrosine­specific protein kinase

 

Plasma membrane (?)

 

 

(?)

 

Plasma membrane

Tyrosine­specific protein kinase

 

Cytoplasm (plasma membrane?)

 

Sarcoma

Cytoplasm (cytoskeleton?)

Tyrosine­specific protein kinase

Chicken sarcoma

 

(?)

 

erb­B

Chicken  leukemia

Erythroleukemia, fibrosarcoma

Plasma and cytoplasmic  membranes

EGF receptor's cytoplasmic tyrosine­specific  protein kinase domain

fms

Cat sarcoma

Sarcoma

Plasma and cytoplasmic  membranes

Tyrosine­specific protein kinase; macrophage  colony­stimulating factor receptor

mil

Chicken  carcinoma

 

Cytoplasm

(?)

raf

Mouse sarcoma

Sarcoma

Cytoplasm

Protein kinase (serine/threonine) activated by  Ras

mos

Mouse sarcoma

Mouse leukemia

Cytoplasm

(?)

sis

Monkey  sarcoma

Monkey sarcoma

Secreted

PDGF­like growth factor, b­chain

Ha­ras

Rat sarcoma

Human carcinoma, rat carcinoma

Plasma membrane

GTP­binding protein

Ki­ras

Rat sarcoma

Human carcinoma, leukemia, and  sarcoma

Plasma membrane

GTP­binding protein

N­ras



Human leukemia and carcinoma

Plasma membrane

 

myc

Chicken  leukemia

Sarcoma, myelocytoma, and  carcinoma

Nucleus

DNA­binding related to cell proliferation;  transcriptional control

myb

Chicken  leukemia

Human leukemia

Nucleus

(?)

B­lym



Chicken lymphoma, human lymphoma

Nucleus (?)

(?)

ski

Chicken sarcoma

 

Nucleus (?)

(?)

rel

Turkey leukemia Reticuloendotheliosis

(?)

(?)

erb­A

Chicken  leukemia

 

(?)

Thyroid hormone receptor (c­erb­A a1);  related to steroid hormone receptors, retinoic  acid receptor, and vitamin D3 receptor

ets

Chicken  leukemia

 

(?)

DNA binding

elk  (ets­like)

 

 

 

DNA­binding protein

jun

 

Osteosarcoma

 

Products associate to form AP1 gene  transcription factor

fos

 

Fibrosarcoma

 

Products associate to form AP1 gene  transcription factor

Source: Adapted from Hunter, T. The proteins of oncogenes. Sci. Am. 251:70, 1984. a

 The second column gives the source from which each viral oncogene was first isolated and the cancer induced by the oncogene. Some names, such as fps and fes,  may be equivalent genes in birds and mammals. The third column lists human and animal tumors caused by agents other than viruses in which the ras oncogene or  an inappropriately expressed proto­oncogene has been identified.

   

  

Page 890

Bibliography Alberts, B., Bray, D., Lewis, J., Raff, R., Roberts, K., and Watson, J. D. Molecular Biology of the Cell, 3rd ed. New York: Garland Publishing, 1994. Cuatrecasas, P. Hormone receptors, membrane phospholipids, and protein kinases. The Harvey Lectures Series 80:89, 1986. DeGroot, L. J., (Ed.). Endocrinology. Philadelphia: Saunders, 1995. Hunter, T. The proteins of oncogenes. Sci. Am. 251:70, 1984. Krieger, D. T., and Hughes, J. C. (Eds.). Neuroendocrinology. Sunderland, MA: Sinauer Associates, 1980. Litwack, G. (Ed.). Biochemical Actions of Hormones, Vols. 1–14. New York: Academic Press, 1973–1987. Litwack, G. (Ed. in Chief) Vitamins and Hormones, Vol. 50. Orlando: Academic Press, 1995. Norman, A. W., and Litwack, G. Hormones. Orlando: Academic Press, 1987. Richter, D. Molecular events in expression of vasopressin and oxytocin and their cognate receptors. Am. J. Physiol. 255:F207, 1988. Ryan, R. J., Charlesworth, M. C., McCormick, D. J., Milius, R. P., and Keutmann, H. T. FASEB J. 2:2661, 1988. Saltiel, A. R. The paradoxical regulation of protein phosphorylation in insulin action. FASEB J. 8:1034, 1994. Spiegel, A. M., Shenker, A., and Weinstein, L. S. Receptor–effector coupling by G proteins: implication for normal and abnormal signal­transduction. Endocr. Rev.  13:536, 1992. Struthers, A. D. (Ed.) Atrial Natriuretic Factor. Boston: Blackwell Scientific Publications, 1990. Weinberg, R. A. The action of oncogenes in the cytoplasm and nucleus. Science 230:770, 1985. Questions J. Baggott and C. N. Angstadt 1. In a cascade of hormones (e.g., hypothalamus, pituitary, and target tissue), at each successive level: A. the quantity of hormone released and its half­life can be expected to increase. B. the quantity of hormone released increases, but its half­life does not change. C. the quantity of hormone released and its half­life are approximately constant. D. the quantity of hormone released decreases, but its half­life does not change. E. the quantity of hormone released and its half­life can both be expected to decrease. 2. All of the following have an identical (or very similar) a  sub­unit EXCEPT: A. growth hormone. B. thyroid­stimulating hormone. C. luteinizing hormone. D. follicle­stimulating hormone. 3. If a single gene contains information for the synthesis of more than one hormone molecule: A. all the hormones are produced by any tissue that expresses the gene. B. all of the hormone molecules are identical. C. cleavage sites in the gene product are typically pairs of basic amino acids. D. all of the peptides of the gene product have well­defined biological activity. E. the hormones all have similar function. 4. In the sequence of events associated with signal transduction, which one is out of place? Receptor binds hormone. A. Conformational change occurs in receptor. B. Receptor interacts with G­protein. C. a  Subunit of G­protein hydrolyzes GTP. D. a  Subunit of G­protein dissociates from b  and g subunits. E. a  Subunit of G­protein binds to adenylate cyclase. 5. The direct effect of cAMP in the protein kinase A pathway is to: A. activate adenylate cyclase. B. dissociate regulatory subunits from protein kinase. C. phosphorylate certain cellular proteins. D. phosphorylate protein kinase A. E. release hormones from a target tissue. 6. Activation of phospholipase C initiates a sequence of events including all of the following EXCEPT: A. release of inositol 4,5­bisphosphate from a phospholipid. B. increase in intracellular Ca2+ concentration. C. release of diacylglycerol (DAG) from a phospholipid. D. activation of protein kinase C. E. phosphorylation of certain cytoplasmic proteins. 7. In the ovarian cycle: A. GnRH enters the vascular system via transport by a specific membrane carrier. B. the corpus luteum dies only if fertilization does not occur. C. inhibin works by inhibiting the synthesis of the a  subunit of FSH. D. FSH activates a protein kinase A pathway. E. LH is taken up by the corpus luteum and binds to cytoplasmic receptors. 8. The Scatchard plot, shown in the accompanying figure, could be used to determine kinetic parameters of an enzyme. Which letter in the graph corresponds to total  binding sites in a Scatchard plot or Vmax in an enzyme kinetic plot?

9. With the anterior pituitary hormones, TSH, LH, and FSH: A. the a  subunits are all different. B. the b  subunits are specifically recognized by the receptor. C. the b  subunit alone can bind to the receptor. D. hormonal activity is expressed through activation of protein kinase B. E. intracellular receptors bind these hormones.

 

  

Page 891

10. In the interaction of a hormone with its receptor, all of the following are true EXCEPT: A. more than one polypeptide chain of the hormone may be necessary. B. more than one second messenger may be generated. C. an array of transmembrane helices may form the binding site for the hormone. D. receptors have a greater affinity for hormones than for synthetic agonists or antagonists. E. hormones released from their receptor after endocytosis could interact with a nuclear receptor. In the following questions, match the numbered hormone with the lettered kinase it stimulates. A. protein kinase A B. tyrosine kinase C. protein kinase C D. protein kinase G 11. Atrial natriuretic factor. 12. Gonadotropin­releasing hormone. 13. Insulin. 14. Vasopressin. Answers 1. A Each successive step typically releases a larger amount of a longer lived hormone (p. 842). 2. A All of these are anterior pituitary hormones, but only the last three, the glycoprotein hormones, have an a  subunit that is similar or identical from hormone to  hormone (p. 846). 3. C One or more trypsin­like proteases catalyze the reaction (Figure 20.5). A: The POMC gene product is cleaved differently in different parts of the anterior  pituitary (p. 849). B: Multiple copies of a single hormone may occur (p. 852), but not necessarily (Figure 20.5, p. 850). D: Some fragments have no known function.  E: ACTH and b ­endorphin, for example, hardly have similar functions (p. 847; Table 20.2). 4. C Hydrolysis of GTP returns the a  subunit to its original conformation and allows it to associate with the b  and g subunits (p. 861). 5. B cAMP binding causes a conformational change in the regulatory subunits, resulting in the release of active protein kinase A (p. 862). 6. A Inositol 1,4,5­triphosphate (IP3) is released from the phospholipid, phosphatidylinositol 4,5­bisphosphate (PIP2) (p. 862). 7. D A: GnRH enters the vascular system through fenestrations (p. 868). B. The corpus luteum is replaced by the placenta if fertilization occurs (p. 870). C: The  glycoprotein hormones share a common a  subunit. Specific control of them would not involve a subunit they share. E: LH interacts with receptors on the cell  membrane. 8. D A is free ligand concentration (analogous to substrate concentration), B is bound ligand concentration (analogous to  ), C is the equilibrium constant (analogous to  Km ), and D is the extrapolated maximum number of binding sites (analogous to Vmax) (p. 872). 9. B A: The a  subunits are identical or nearly so (p. 872). B and C: Although specificity is conferred by the b  subunits, which differ among the three hormones, binding  to the receptor requires both subunits (p. 873). D: It is protein kinase A, and perhaps also protein kinase C in the case of TSH (p. 873). E: These large glycoprotein  hormones do not penetrate the cell membrane; they bind to receptors on the cell surface (p. 874). See Figure 20.30. 10. D b  Receptors bind isoproterenol more tightly than their hormones (p. 875). A and B: These are true of the glycoprotein hormones (p. 873). C: This appears to  be true for the  1 receptor (Figure 20.31). E: This is possible, but entirely speculative; there are currently no known examples. 11. D See p. 885. 12. C See p. 883. 13. B See p. 879. 14. A See p. 880.

 

  

Page 893

Chapter 21—  Biochemistry of Hormones II: Steroid Hormones Gerald Litwack and Thomas J. Schmidt

21.1 Overview

894

21.2 Structures of Steroid Hormones

894

21.3 Biosynthesis of Steroid Hormones

896

Steroid Hormones Are Synthesized from Cholesterol

896

21.4 Metabolic Inactivation of Steroid Hormones

901

21.5 Cell–Cell Communication and Control of Synthesis and Release of  Steroid Hormones

901

Steroid Hormone Synthesis Is Controlled by Specific Hormones

901

Aldosterone

901

Estradiol

905

Vitamin D3

907

21.6 Transport of Steroid Hormones in Blood

908

Steroid Hormones Are Bound to Specific Proteins or Albumin in Blood

908

21.7 Steroid Hormone Receptors

909

Steroid Hormones Bind to Specific Intracellular Protein Receptors

909

Some Steroid Receptors Are Part of the cErbA Family of Proto­ oncogenes

913

21.8 Receptor Activation: Upregulation and Downregulation

914

Steroid Receptors Can Be Upregulated or Downregulated Depending on  Exposure to the Hormone

915

21.9 A Specific Example of Steroid Hormone Action at Cell Level:  Programmed Death

915

Bibliography

916

Questions and Answers Clinical Correlations

   

  

917  

21.1 Oral Contraception

907

21.2 Apparent Mineralocorticoid Excess Syndrome

911

21.3 Programmed Cell Death in the Ovarian Cycle

916

Page 894

21.1—  Overview Steroid hormones in the human include cortisol as the major glucocorticoid or anti­stress hormone, aldosterone as an important regulator of Na+ uptake, and the sex  and progestational hormones. Sex hormones are 17b ­estradiol in females and testosterone in males. Progesterone is the major progestational hormone. Testosterone  is reduced in some target tissues to dihydrotestosterone, a higher affinity ligand for the androgen receptor. Vitamin D3 is converted to the steroid hormone, dihydroxy  vitamin D3. Genes in the steroid receptor supergene family include retinoic acid receptors and thyroid hormone receptor, although the ligands for these additional  receptors are not derivatives of cholesterol. Retinoic acid and thyroid hormone, however, have six­membered ring structures that could be considered to resemble the  A ring of a steroid.

Figure 21.1  The steroid nucleus.

Steroidal structure will be reviewed with the synthesis and inactivation of steroid hormones. Regulation of synthesis of steroid hormones is reviewed with respect to the  renin–angiotensin system for aldosterone, the gonadotropes, especially follicle­stimulating hormone for 17b ­estradiol, and the vitamin D3 mechanism. Steroid hormone  transport is reviewed with respect to the transporting proteins in blood. A general model for steroid hormone action at the cellular level is presented with information on  receptor activation and regulation of receptor levels. Specific examples of steroid hormone action for programmed cell death and for stress are presented. Finally, the  roles of steroid hormone receptors as transcriptional transactivators and repressors are reviewed. 21.2—  Structures of Steroid Hormones Steroid hormones are derived in specific tissues in the body and are divided into two classes: the sex and progestational hormones, and the adrenal hormones.  They are synthesized from cholesterol and all of these hormones pass through the required intermediate,  5­pregnenolone. The structure of steroid hormones is related  to the cyclopentanoperhydrophenanthrene nucleus. The numbering of the cyclopentanoperhydrophenanthrene ring system and the lettering of the rings is presented  in Figure 21.1. The ring system of the steroid hormones is stable and not catabolized by mammalian cells. Conversion of active hormones to less active or inactive  forms involves alteration of ring substituents rather than the ring structure itself. The parental precursor of the steroid is cholesterol, shown in Figure 21.2. The  biosynthesis of cholesterol is given on p. 410. The major steroid hormones of humans and their actions are shown in Table 21.1. Many of these hormones are similar in gross structure, although the specific receptor  for each hormone is able to distinguish the cognate ligand. In the cases of cortisol and aldosterone, however, there is overlap in the ability of each specific receptor to  bind both ligands. Thus the availability and concen­

Figure 21.2  Structure of cholesterol.

 

  

Page 895 TABLE 21.1 Major Steroid Hormones of Humans Hormone

Structure

Secretion Signala

Secretion from

Functions

Corpus luteum

LH

Maintains (with estradiol) the uterine  endometrium for implantation; differentiation  factor for mammary glands

Ovarian follicle; corpus  luteum; (Sertoli cell)

FSH

Female: regulates gondotropin secretion in  ovarian cycle (see Chapter 20); maintains  (with progesterone) uterine endometrium;  differentiation of mammary gland. Male:  negative feedback inhibitor of Leydig cell  synthesis of testosterone

Leydig cells of testis;  (adrenal gland); ovary

LH

Male: after conversion to  dihydrotestosterone, production of sperm  proteins in Sertoli cells; secondary sex  characteristics (in some tissues testosterone  is active hormone)

Reticularis cells

ACTH

Various protective effects; weak androgen;  can be converted to estrogen; no receptor yet  found; inhibitor of G6­PDH: regulates NAD+  coenzymes

Fasciculata cells

ACTH

Stress adaptation through various cellular  phenotypic expressions; slight elevation of  liver glycogen; killing effect on certain T cells  in high doses; elevates blood pressure;  sodium uptake in luminal epithelia

Glomerulosa cells of  adrenal cortex

Angiotensin II/III

Causes sodium ion uptake via conductance  channel; occurs in high levels during stress;  raises blood pressure; fluid volume increased

Vitamin D arises in skin  cells after irradiation and  then successive  hydroxylations occur in  liver and kidney to yield  active form of hormone

PTH (stimulates  Facilitates Ca2+ and phosphate absorption by  kidney proximal tubule  intestinal epithelial cells; induces intracellular  hydroxylation system) calcium­binding protein

Progesterone

 

17b­Estradiol

 

Testosterone

 

Dehydroepian­  drosterone

 

Cortisol

 

Aldosterone

 

1,25­Dihydroxy­  vitamin D3

  a

 LH, luteinizing hormone; FSH, follicle­stimulating hormone; ACTH, adrenocorticotropic hormone; PTH, parathyroid hormone.

   

  

Page 896

trations of each receptor and the relative amounts of each hormone in a given cell become paramount considerations. The steroid hormones listed in Table 21.1 can be  described as classes based on the carbon number in their structures. Thus a C­27 steroid is 1,25(OH)2D3; C­21 steroids are progesterone, cortisol, and  aldosterone; C­19 steroids are testosterone and dehydroepiandrosterone; and a C­18 steroid is 17 b ­estradiol. Classes, such as sex hormones, can be  distinguished easily by the carbon number, C­19 being androgens, C­18 being estrogens, and C­21 being progestational or adrenal steroids. Aside from the number of  carbon atoms in a class structure, certain substituents in the ring system are characteristic. For example, glucocorticoids and mineralocorticoids (typically aldosterone)  possess a C­11 OH or oxygen function. In rare exceptions, certain synthetic compounds can elicit a response without a C­11 OH group but they require a new  functional group in proximity within the A­B ring system. Estrogens do not have a C­19 methyl group and the A ring is contracted by the content of three double  bonds. Many receptors recognize the ligand A ring primarily, the estrogen receptor can distinguish the A ring of estradiol stretched out of the plane of the B­C­D rings  compared to other steroids in which the A ring is coplanar with the B­C­D rings. These relationships are shown in Figure 21.3. 21.3—  Biosynthesis of Steroid Hormones Steroid Hormones Are Synthesized from Cholesterol Hormonal regulation of steroid hormone biosynthesis is generally believed to be mediated by an elevation of intracellular cAMP and Ca2+, although generation of  inositol triphosphate may also be involved, as shown in Figure 21.4. The stimulatory response of cAMP is mediated via acute (occurring within seconds to minutes)  and chronic (requiring hours) effects on steroid synthesis. The acute effect is to mobilize and deliver cholesterol, the precursor for all steroid hormones, to the  mitochondrial inner membrane, where it is metabolized to pregnenolone by the cytochrome P450 cholesterol side chain cleavage enzyme (see Chapter 22 for  discussion of P450 enzymes). In contrast, the chronic effects of cAMP are mediated via increased transcription of the genes that encode the steroidogenic enzymes  and are thus responsible for maintaining optimal long­term steroid production. Data demonstrate that a protein is induced and that this newly synthesized regulatory  protein actually facilitates the translocation of cholesterol from outer to inner mitochondrial membrane where the P450 enzyme is located. This 30­kDa phosphoprotein  is designated as the steroidogenic acute regulatory (StAR) protein. In humans, StAR mRNA has been shown to be specifically expressed in testis and ovary,  known sites of steroidogenesis. Patients with lipoid congenital adrenal hyperplasia (LCAH), an inherited disease in which both adrenal and gonadal steroidogenesis is  significantly impaired and lipoidal deposits occur in these tissues, express truncated and non­functional StAR proteins. These biochemical and genetic data strongly  suggest that StAR protein is the hormone­induced protein factor that mediates acute regulation of steroid hormone biosynthesis. Pathways for conversion of cholesterol to the adrenal cortical steroid hormones are presented in Figure 21.5. Cholesterol is the major precursor and undergoes side  chain cleavage to form  5­pregnenolone releasing a C6 aldehyde, isocaproaldehyde. D 5­Pregnenolone is mandatory in the synthesis of all steroid hormones. As  shown in Figure 21.5, pregnenolone can be converted directly to progesterone, which requires two cytoplasmic enzymes, 3 b ­ol dehydrogenase and D 4,5­ isomerase. The dehydrogenase converts the 3­OH group of pregnenolone to a 3­keto group and the isomerase moves the double bond from the B ring to the A ring  to produce progesterone. In the corpus luteum the bulk

 

  

Page 897

Figure 21.3  "Ball­and­stick" representations of the structures of some  steroid hormones determined by X­ray crystallographic methods.  Details of each structure are labeled. In aldosterone the acetal grouping   

is 

 where R1, R2,

R3 refer to different substituents.  Reprinted with permission from Glusker, J. P. In G. Litwack (Ed.),  Biochemical Actions of Hormones, Vol. 6. New York:  Academic Press, 1979, pp. 121–204.

 

  

Page 898

Figure 21.4  Overview of hormonal stimulation of steroid hormone biosynthesis.  Nature of the hormone (top of figure) depends on the cell type and receptor (ACTH  for cortisol synthesis; FSH for estradiol synthesis; LH for testosterone  synthesis, etc., as given in Table 20.1). It binds to cell membrane receptor and  activates adenylate cyclase mediated by a stimulatory G­protein. Receptor,  activated by hormone, may directly stimulate a calcium channel or indirectly  stimulate it by activating the phosphatidylinositol cycle (PI cycle) as shown in  Figure 20.25. If the PI cycle is concurrently stimulated, IP3 could  augment cytosol Ca2+ levels from the intracellular calcium store. The  increase in cAMP activates protein kinase A (Figure 21.21) whose phosphorylations  cause increased hydrolysis of cholesteryl esters from the droplet to free  cholesterol and increase cholesterol transport into the mitochondrion. The  combination of elevated Ca2+ levels and protein phosphorylation, as  well as induction of the StAR protein, result in increased side chain cleavage  and steroid biosynthesis. These combined reactions overcome the rate­limiting  steps in steroid biosynthesis and more steroid is produced, which is secreted  into the extracellular space and circulated to the target tissues in the  bloodstream.

of steroid synthesis stops at this point. Progesterone is further converted to aldosterone or cortisol. Conversion of pregnenolone to aldosterone, which occurs in the  adrenal zona glomerulosa cells, requires endoplasmic reticulum 21­hydroxylase, and mitochondrial 11b ­hydroxylase and 18­hydroxylase. To form cortisol, primarily  in adrenal zona fasciculata cells, endoplasmic reticulum 17­hydroxylase and 21­hydroxylase are required together with mitochondrial 11 b ­hydroxylase. The  endoplasmic reticulum (ER) hydroxylases are all cytochrome P450­linked enzymes (see Chapter 22).  5­Pregnenolone is converted to dehydroepiandrosterone in  the adrenal zona reticularis cells by the action of 17a ­hydroxylase of the endoplasmic reticulum to form 17a ­hydroxypregnenolone and then by the action of a  carbon side chain­cleavage system to form dehydroepiandrosterone. Cholesterol is also converted to the sex hormones by way of  5­pregnenolone (Figure 21.6). Progesterone can be formed as described above and further converted  to testosterone by the action of the endoplasmic reticulum enzymes and 17­dehydrogenase. Testosterone, so formed, is a major secretory product in the Leydig cells  of the testis and undergoes conversion to dihydrotestosterone in some androgen target cells before binding to the androgen receptor. This conversion requires the  activity of 5a ­reductase located in the ER and nuclear fractions. Pregnenolone can enter an alternative pathway to form dehydroepiandrosterone as described above.  This compound can be converted to 17b ­estradiol via the aromatase enzyme system and the action of 17­reductase. Also, estradiol can be formed from testosterone  by the action of the aromatase system. The hydroxylases of endoplasmic reticulum involved in steroid hormone synthesis are cytochrome P450 enzymes (Chapter 22). Molecular oxygen (O2) is a substrate  with one oxygen atom incorporated into the steroidal substrate (as an OH) and the second atom incorporated into a water molecule. Electrons

 

  

Page 899

Figure 21.5  Conversion of cholesterol to adrenal cortical hormones.

 

  

Page 900

Figure 21.6  Conversion of cholesterol to sex hormones.  Mt, mitochondrial, cyto, cytoplasmic; and ER, endoplasmic reticulum.

 

  

Page 901

are generated from NADH or NADPH through a flavoprotein to ferredoxin or similar nonheme protein. Various agents can induce the levels of cytochrome P450. Note that there is movement of intermediates in and out of the mitochondrial compartment during the steroid synthetic process. 21.4—  Metabolic Inactivation of Steroid Hormones A feature of the steroid ring system is its great stability. For the most part, inactivation of steroid hormones involves reduction. Testosterone is initially reduced to a  more active form by the enzyme 5a ­reductase to form dihydrotestosterone, the preferred ligand for the androgen receptor. However, further reduction similar to the  other steroid hormones results in inactivation. The inactivation reactions predominate in liver and generally render the steroids more water soluble, as marked by  subsequent conjugation with glucuronides or sulfates (see Chapter 22) that are excreted in the urine. Table 21.2 summarizes reactions leading to inactivation and  excretory forms of the steroid hormones. 21.5—  Cell–Cell Communication and Control of Synthesis and Release of Steroid Hormones Secretion of steroid hormones from cells where they are synthesized is elicited by other hormones. Many, but not all, such systems are described in Chapter 20,  Figures 20.2 and 20.3. The hormones that directly stimulate the biosynthesis and secretion of the steroid hormones are summarized in Table 21.3. The signals for  stimulation of biosynthesis and secretion of steroid hormones are polypeptide hormones operating through cognate cell membrane receptors. In some systems where  both cAMP and the phosphatidylinositol (PI) cycle are involved, it is not clear whether one second messenger predominates. In many such systems, for example,  aldosterone synthesis and secretion, probably several components (i.e., acetylcholine muscarinic receptor, atriopeptin receptor, and their second messengers) are  involved in addition to the signal listed in Table 21.3. Steroid Hormone Synthesis Is Controlled by Specific Hormones The general mechanism for hormonal stimulation of steroid hormone synthesis is presented in Figure 21.4. Figure 21.7 (p. 903) presents the system for stimulation of  cortisol biosynthesis and release. The role of Ca2+ in steroid synthesis and/or secretion is unclear. Rate­limiting steps in the biosynthetic process involve the availability  of cholesterol from cholesteryl esters in the droplet, the transport of cholesterol to the inner mitochondrial membrane (StAR protein), and the upregulation of the  otherwise rate­limiting side chain cleavage reaction. Aldosterone Figure 21.8 (p. 904) shows the overall reactions leading to the secretion of aldosterone in the adrenal zona glomerulosa cell. This set of regulatory controls on  aldosterone synthesis and secretion is complicated. The main driving force is angiotensin II generated from the signaling to the renin–angiotensin system shown in  Figure 21.9 (p. 905). Essentially, the signal is generated under conditions when blood [Na+] and blood pressure (blood volume) are required to be increased. The N­ terminal decapeptide of circulating a 2­globulin (angiotensinogen) is cleaved by renin, a protease. This decapeptide is the hormonally inactive precursor, angiotensin  I. It is converted to the octapeptide hormone, angiotensin II, by the action of converting enzyme. Angiotensin II is converted to the heptapeptide, angiotensin III, by an  aminopeptidase. Both angiotensins

 

  

Page 902 TABLE 21.2 Excretion Pathways for Steroid Hormones

Steroid Class Progestins

Starting Steroid Progesterone

A:B Ring  Junction

Inactivation Steps 1. Reduction of C­20  2. Reduction of 4­ene­3­one

Principal  Conjugate  Presenta

Steroid Structure Representations of Excreted  Product

(cis)

G

  Estrogens

Estradiol

1. Oxidation of 17b­OH 2. Hydroxylation at C­2 with  subsequent methylation 3. Further hydroxylation or ketone  formation at a variety of positions (e.g.,  C­6, C­7, C­14, C­15, C­16, C­18)

 

G

 

Androgens

Testosterone

1. Reduction of 4­ene­3­one  2. Oxidation of C­17 hydroxyl

(cis and trans)

Glucocorticoids

Cortisol

1. Reduction of 4­ene­3­one  2. Reduction of 20­oxo group  3. Side chain cleavage

(trans)

Mineralocorticoids

Aldosterone

1. Reduction of 4­ene­3­one

(trans)

G, S

  G

  G

  Vitamin D metabolites

1,25(OH)2D3

1. Side chain cleavage between C­23  and C­24

?

 

  Source: From Norman, A. W., and Litwack, G. Hormones. Orlando, FL: Academic Press, 1987. a

 G, Glucuronide; S, sulfate.

  II and III can bind to the angiotensin receptor (Figure 21.8), which activates the phosphatidylinositol cycle to generate IP3 and DAG. IP3 stimulates release of calcium  ions from the intracellular calcium storage vesicles. In addition, the activity of the Ca2+ channel is stimulated by the angiotensin–receptor complex. K+ ions are also  required to stimulate the Ca2+ channel and these events lead to a greatly increased level of cytoplasmic Ca2+. The enhanced cytoplasmic Ca2+

 

  

Page 903

Figure 21.7  Action of ACTH on adrenal fasciculata cells to enhance production and secretion of cortisol.  AC, adenylate cyclase; cAMP, cyclic AMP; PKA, protein kinase A; SCC, side chain cleavage system of enzymes. StAR  (steroidogenic acute regulatory)  protein is a cholesterol transporter functioning between the outer and inner mitochondrial membranes. TABLE 21.3 Hormones that Directly Stimulate Synthesis and Release of Steroid Hormones Steroid Hormone

Steroid­Producing Cell or  Structure

Signala 

Second Messenger

Signal System

Cortisol

Adrenal zona fasciculata

ACTH

cAMP, PI cycle, Ca2+

Hypothalamic–pituitary cascade

Aldosterone

Adrenal zona glomerulosa

Angiotensin II/III

PI cycle, Ca2+

Renin–angiotensin system

Testosterone

Leydig cell

LH

cAMP

Hypothalamic–pituitary cascade

17b ­Estradiol

Ovarian follicle

FSH

cAMP

Hypothalamic–pituitary–ovarian  cycle

Progesterone

Corpus luteum

LH

cAMP

Hypothalamic–pituitary–ovarian  cycle

1,25 (OH)2 Vitamin D3

Kidney

PTH

cAMP

Sunlight, parathyroid glands, plasma  Ca2+ level

a ACTH, adrenocorticotropic hormone; LH, luteinizing hormone; FSH, follicle­stimulating hormone; PI, phosphatidylinositol; PTH, parathyroid 

hormone.

   

  

Page 904

Figure 21.8  Reactions leading to the secretion of aldosterone in the adrenal zona  glomerulosa cell.  cGMP, cyclic GMP; ANF, atrial natriuretic factor; see Figure 21.7 for additional  abbreviations.

has a role in aldosterone secretion and together with diacylglycerol stimulates protein kinase C. Acetylcholine released through the neuronal stress signals has similar  effects mediated by the muscarinic acetylcholine receptor to further reinforce Ca2+ uptake by the cell and stimulation of protein kinase C. Enhanced protein kinase C  activity leads to protein phosphorylations that stimulate the rate­limiting steps of aldosterone synthesis leading to elevated levels of aldosterone, which are then secreted  into the extracellular space and finally into the blood. Once in the blood aldosterone enters the distal kidney cell, binds to its receptor, which initially may be  cytoplasmic, and ultimately stimulates expression of proteins that increase the transport of Na+ from the glomerular filtrate to the blood (see p. 1043).

 

  

Page 905

Figure 21.9  Renin–angiotensin system.  Amino acid abbreviations are found on p. 27.  NEP, norepinephrine.

Signals opposite to those that activate the formation of angiotensin generate atrial natriuretic factor (ANF) or atriopeptin from the heart atria (Figure 21.8; see also  Figure 20.45). ANF binds to a specific zona glomerulosa cell membrane receptor and activates guanylate cyclase, which is part of the same receptor polypeptide so  that the cytosolic level of cGMP increases. Cyclic GMP antagonizes the synthesis and secretion of aldosterone as well as the formation of cAMP by adenylate cyclase.  Involvement of ACTH in aldosterone synthesis and release may involve adenylate cyclase but may be of secondary importance. Aldosterone should be regarded as a stress hormone since its presence in elevated levels in blood occurs as a result of stressful situations. In contrast, cortisol, also  released in stress has an additional biorhythmic release (possibly under control of serotonin and vasopressin), which accounts for a substantial reabsorption of Na+  probably through glucocorticoid stimulation of the Na+–H+ antiport in luminal epithelial cells in addition to the many other activities of cortisol (e.g., anti­inflammatory  action, control of T­cell growth factors, synthesis of glycogen, and effects on carbohydrate metabolism).

Figure 21.10  Formation and secretion of 17  b­estradiol and progesterone.

Estradiol Control of formation and secretion of 17 b ­estradiol, the female sex hormone, is shown in Figure 21.10. During development, control centers for the steady­state and  cycling levels arise in the CNS. Their functions are required to initiate the ovarian cycle at puberty. These centers must harmonize with the firing of other neurons, such  as those producing a clock­like mechanism via release of catecholamines or other amines to generate the pulsatile release of gonadotropin­releasing hormone (GnRH),  probably at hourly intervals. Details of these reactions are presented on page 867, Chapter 20. The FSH circulates and binds to, its cognate receptor on the cell  membrane of the ovarian follicle cell and

 

  

Page 906

through its second messengers, primarily cAMP and the activation of cAMP­dependent protein kinase, there is stimulation of the synthesis and secretion of the female  sex hormone, 17b ­estradiol. At normal stimulated levels of 17b ­estradiol, there is a negative feedback on the gonadotrope (anterior pituitary), suppressing further  secretion of FSH. Near ovarian midcycle, however, there is a superstimulated level of 17b ­estradiol produced that has a positive rather than a negative feedback  effect on the gonadotrope. This causes very high levels of LH to be released, referred to as the LH spike, and elevated levels of FSH. The level of FSH released is  substantially lower than LH because the follicle produces inhibin, a polypeptide hormone that specifically inhibits FSH release without affecting LH release. The  elevation of LH in the LH spike participates in the process of ovulation. After ovulation, the remnant of the follicle is differentiated into the functional corpus luteum,  which now synthesizes progesterone (and also some estradiol), under the influence of elevated LH levels. Progesterone, however, is a feedback inhibitor of LH  synthesis and release (operating through a progesterone receptor in the gonadotropic cell) and eventually the corpus luteum dies, owing to a fall in the level of  available LH and the production of oxytocin, a luteolytic agent, by the corpus luteum. Prostaglandin F2a may also be involved. With the death of the corpus luteum,  the blood levels of progesterone and estradiol fall, causing menstruation as well as a decline in the negative feedback effects of these steroids on the anterior pituitary  and hypothalamus, and the cycle begins again. Clinical Correlation 21.1 describes how oral contraceptives interrupt this sequence. The situation is similar in males with respect to regulation of gonadotropin secretion, but LH acts principally on the Leydig cell for the stimulated production of  testosterone, and FSH acts on the Sertoli cells to stimulate production of inhibin and sperm proteins. Production of testosterone is subject to the negative feedback  effect of 17b ­estradiol synthesized in the Sertoli cell. The 17b ­estradiol so produced operates through a nuclear estrogen receptor in the Leydig cell to produce  inhibition of testosterone synthesis at the transcriptional level. In all cases of steroid hormone production, the synthetic system resembles that shown in Figure 21.4.

Figure 21.11  The vitamin D endocrine system.  Pi, inorganic phosphate.  Adapted from Norman, A. W. and Litwack, G.  Hormones. Orlando, FL: Academic  Press, 1987, f. 379.

 

  

Page 907

CLINICAL CORRELATION 21.1  Oral Contraception Oral contraceptives usually contain an estrogen and a progestin. Taken orally, the levels  of these steroids increase in blood to a level where secretion of FSH and LH is  repressed. Consequently, the gonadotropic hormone levels in blood fall and there is  insufficient FSH to drive development of the ovarian follicle. As a result, the follicle does  not mature and ovulation cannot occur. In addition, any corpora lutea cannot survive  because of low LH levels. In sum, the ovarian cycle ceases. The uterine endometrium  thickens and remains in this state, however, because of elevated levels of estrogen and  progestin. Pills without the steroids (placebos) are usually inserted in the regimen at about  the 28th day and, as a result, blood levels of steroids fall dramatically and menstruation  occurs. When oral contraceptive steroids are resumed, the blood levels of estrogen and  progestin increase again and the uterine endometrium thickens. This sequence creates a  false ''cycling" because of the occurrence of menstruation at the expected time in the  cycle. The ovarian cycle and ovulation are suppressed by the oral contraceptive based on  the negative feedback effects of estrogen and progestin on the secretion of the anterior  pituitary gonadotropes. It is also possible to provide contraception by implanting in the  skin silicone tubes containing progestins. The steroid is slowly released, providing  contraception for up to 3–5 years. Zatuchni, G. I. Female contraception. In: K. L. Becker (Ed.), Principles and Practice  of Endocrinology and Metabolism. New York: Lippincott, 1990, p. 861; and Shoupe,  D., and Mishell, D. R. Norplant: subdermal implant system for long term contraception.  Am. J. Obstet. Gynecol. 160:1286, 1988.   Vitamin D3 Activation of vitamin D to dihydroxy vitamin D3 produces a hormone that has the general features of a steroid hormone. The active form of vitamin D stimulates  intestinal absorption of dietary calcium and phosphorus, the mineralization of bone matrix, bone resorption, and reabsorption of calcium and phosphate in the renal  tubule. The vitamin D endocrine system is diagrammed in Figure 21.11.7­Dehydrocholesterol is activated in the skin by sunlight to form vitamin D3  (cholecalciferol). This form is hydroxylated first in the liver to 25­hydroxy vitamin D3 (25­hydroxycholecalciferol) and subsequently in the kidney to form the  1a ,25­vitamin D3 (1,25(OH)2D3)(1a ,25­dihydroxycholecalciferol). The hormone can bind to nuclear 1,25(OH)2D3 receptors in intestine, bone, and kidney and  then transcriptionally activate genes encoding calcium­binding proteins whose actions may lead to the absorption and reabsorption of Ca2+ (as well as phosphorus).  The subcellular mode of action is presented in Figure 21.12. In this scheme the active form of vitamin D3 enters the intestinal cell from the blood side and migrates to  the nucleus. Once inside it binds to the high­affinity vitamin D3 receptor, which probably undergoes an activation event, and associates with a vitamin D3­responsive  element to activate genes responsive to the hormone. Messenger RNA is produced and translated in the cytoplasm; these RNAs encode calcium­binding proteins,  Ca2+­ATPase, other ATPases, membrane components, and facilitators of vesicle formation. Increased levels of calcium­binding proteins may cause increased uptake  of Ca2+ from the intestine or may simply buffer the cytoplasm against high free Ca2+ levels. With each of the steroid­producing systems discussed, feedback controls are operative whereby sufficient amounts of the circulating steroid hormone inhibit the further  production and release of intermediate hormones in the pathway at the levels of the pituitary and hypothalamus, as viewed in Figure 20.3. In the case of the vitamin D  systems, the controls are different since the steroid production is not stimulated by the cascade process applicable to estra­

Figure 21.12  Schematic model to describe the action of 1,25(OH)2D3 in the intestine in  stimulating intestinal calcium transport.  Redrawn from Nemere, I., and Norman, A. W. Biochim.  Biophys. Acta 694:307, 1982.

 

  

Page 908

diol. When the circulating levels of the active form of vitamin D (1, 25(OH)2D3) are high, hydroxylations at the 24 and 25 positions are favored and the inactive 24,25 (OH)2­vitamin D3 compound is generated. 21.6—  Transport of Steroid Hormones in Blood Steroid Hormones Are Bound to Specific Proteins or Albumin in Blood There are four major proteins in the circulation that account for much of the steroid hormones bound in the blood. They assist in maintaining a level of these hormones  in the circulation and protect the hormone from metabolism and inactivation. The binding proteins of importance are corticosteroid­binding globulin protein, sex  hormone­binding protein, androgen­binding protein, and albumin. Corticosteroid­binding globulin (CBG) or transcortin is about 52 kDa, is 3–4 mg% in human plasma, and binds about 80% of the total 17­hydroxysteroids in the  blood. In the case of cortisol, which is the principal antistress corticosteroid in humans, about 75% is bound by CBG, 22% is bound in a loose manner to albumin, and  8% is in free form. The unbound cortisol is the form that can permeate cells and bind to intracellular receptors to produce biological effects. The CBG has a high  affinity for cortisol with a binding constant (Ka) of 2.4 × 107 M–1. Critical structural determinants for steroid binding to CBG are the  4­3­ketone and 20­ketone  structures. Aldosterone binds weakly to CBG but is also bound by albumin and other plasma proteins. Normally, 60% of aldosterone is bound to albumin and 10% is  bound to CBG. In human serum, albumin is 1000­fold the concentration of CBG and binds cortisol with an affinity of 103 M–1, much lower than the affinity of CBG for  cortisol. Thus cortisol will always fill CBG­binding sites first. During stress, when secretion of cortisol is very high, CBG sites will be filled but there will be sufficient  albumin to accommodate excess cortisol. Sex hormone binding globulin (SHBG) (40 kDa) binds androgens with an affinity constant of about 109 M–1, which is much tighter than albumin binding of  androgens. One to three percent of testosterone is unbound in the circulation and 10% is bound to SHBG, with the remainder bound to albumin. The level of SHBG is  probably important in controlling the balance between circulating androgens and estrogens along with the actual amounts of these hormones produced in given  situations. About 97–99% of bound testosterone is bound reversibly to SHBG but much less estrogen is bound to this protein in the female. As mentioned above, only  the unbound steroid hormone can permeate cells and bind to intracellular receptors, thus expressing its activity. The level of SHBG before puberty is about the same in  males and females, but, at puberty, when the functioning of the sex hormones becomes important, there is a small decrease in the level of circulating SHBG in females  and a larger decrease in males, ensuring a relatively greater amount of the unbound, biologically active sex hormones—testosterone and 17b ­estradiol. In adults, males  have about one­ half as much circulating SHBG as females, so that the unbound testosterone in males is about 20 times greater than in females. In addition, the total  (bound plus unbound) concentration of testosterone is about 40 times greater in males. Testosterone itself lowers SHBG levels in blood, whereas 17 b ­estradiol raises  SHBG levels in blood. These effects have important ramifications in pregnancy and in other conditions. Androgen binding protein (ABP) is produced by Sertoli cells in response to testosterone and FSH, both of which stimulate protein synthesis in these cells.  Androgen­binding protein is doubtless not of great importance in the entire blood circulation but is important because it maintains a ready supply of testosterone for the  production of protein constituents of spermatozoa. Its

 

  

Page 909

role may be to maintain a high local concentration of testosterone in the vicinity of the developing germ cells within the tubules. From a variety of studies it is clear that these, as well as other transport proteins, protect the circulating pool of steroid hormones. They supply free steroids that can  enter cellular targets after dissociation from the bound forms as more free hormone is utilized, thus serving the needs of target cells by a mass action effect. 21.7—  Steroid Hormone Receptors Steroid Hormones Bind to Specific Intracellular Protein Receptors The general model for steroid hormone action presented in Figure 21.13 takes into account the differences among steroid receptors in terms of their location within the  cell. In contrast to polypeptide hormone receptors that are generally located on/in the cell surface, steroid hormone receptors, as well as other related receptors for  nonsteroids (i.e. thyroid hormone, retinoic acid, vitamin D3), are located in the cell interior. Among the steroid receptors there appear to be some differences as to the  subcellular location of the non­DNA­binding forms of these receptors. The glucocorticoid receptor and possibly the aldosterone receptor appear to reside in the  cytoplasm, whereas the other receptors, for which suitable data have been collected, may be located within the nucleus, presumably in association with DNA, although  not necessarily at productive acceptor sites on the DNA. Figure 21.13, Step 1, shows a bound and a free form of a steroid hormone(s). The free form may enter the  cell by a process of diffusion. In the case of glucocorticoids, like cortisol, the steroid would bind

Figure 21.13  Model of steroid hormone action.  Step 1—Dissociation of free hormone (biologically active) from circulating transport protein;  Step 2—diffusion of free ligand into cytosol or nucleus;  Step 3—binding of ligand to unactivated cytoplasmic or nuclear receptor;  Step 4—activation of cytosolic or nuclear hormone–receptor complex to activated,  DNA­binding form;  Step 5—translocation of activated cytosolic hormone–receptor complex into nucleus;  Step 6—binding of activated hormone–receptor complexes to specific response elements  within the DNA;  Step 7—synthesis of new proteins encoded by hormone­responsive genes; and  Step 8—alteration in phenotype or metabolic activity of target cell mediated by  specifically induced proteins.

 

  

Page 910 9

–1

7

to an unactivated receptor with an open ligand binding site (Step 3). The binding constant for this reaction is on the order of 10  M , compared to about 10  M–1 for  the binding to CBG (see above). The non­DNA­binding form also referred to as the unactivated or nontransformed receptor is about 300 kDa, because other proteins  may be associated in the complex. Many investigators believe that a dimer of the 90­kDa protein, which is a heat shock protein that is induced when cells are stressed  (heat shock proteins), is associated with the receptor in this form and occludes its DNA­binding domain, accounting for its non­DNA­binding activity. Associated  with this dimer of hsp90 is another heat shock protein designated as hsp56, which interestingly also functions as an immunophilin and, as such, can bind to a number of  potent immunosuppressive drugs. The dimer of the 90­kDa heat shock protein is depicted by the pair of red ovals attached to the cytoplasmic receptor that block the  DNA­binding domain pictured as a pair of "fingers" in the subsequently activated form. Activation or transformation to the DNA­binding form is accomplished by  release of the 90­kDa heat shock proteins (Step 4). It is not clear what actually drives the activation step(s). Clearly, the binding of the steroidal ligand is important but  other factors may be involved. A low molecular weight component has been proposed to be part of the cross­linking between the nonhomologous proteins and the  receptor in the DNA­binding complex. In the case of glucocorticoid receptor, only the non­DNA­binding form has a high affinity for binding steroidal ligand. Following  activation and exposure of the DNA­binding domain, the receptor translocates to the nucleus (Step 5), binds to DNA, and "searches" the DNA for a high­affinity  acceptor site. At this site the bound receptor complex, frequently a homodimer, acts as a transactivation factor, which together with other transactivators allows for the  starting of RNA polymerase and stimulation of transcription. In some cases the binding of the receptor may lead to repression of transcription and this effect is less well  understood. New mRNAs are translocated to the cytoplasm and assembled into translation complexes for the synthesis of proteins (Step 7) that alter metabolism and  functioning of the target cell (Step 8). When the unoccupied (nonliganded) steroid hormone receptor is located in the nucleus, as may be the case with the estradiol, progesterone, androgen, and vitamin D3  receptors (see Figure 21.12), the steroid must travel through the cytoplasm and cross the perinuclear membrane. It is not clear whether this transport through the  cytoplasm (aqueous environment) requires a transport protein for the hydrophobic steroid molecules. Once inside the nucleus the steroid can bind to the high­affinity,  unoccupied receptor, presumably already on DNA, and cause it to be "activated" to a form bound to the acceptor site. The ligand might promote a conformation that  decreases the off­rate of the receptor from its acceptor, if it is located on or near its acceptor site, or might cause the receptor to initiate searching if the unoccupied  receptor associates with DNA at a locus remote from the acceptor site. Consequently, the mechanism underlying activation of nuclear receptors is less well understood  as compared to activation of cytoplasmic receptors. After binding of activated receptor complexes to DNA acceptor sites, enhancement or repression of transcription  occurs. Consensus DNA sequences defining specific hormone response elements (HREs) for the binding of various activated steroid hormone–receptor complexes are  summarized in Table 21.4. Receptors for glucocorticoids, mineralocorticoids, progesterone, and androgen all bind to the same HRE on the DNA. Thus, in a given cell  type, the extent and type of receptor expressed will determine the hormone sensitivity. For example, sex hormone receptors are expressed in only a few cell types and  the progesterone receptor is likewise restricted to certain cells, whereas the glucocorticoid receptor is expressed in a large number of cell types. In cases where  aldosterone and cortisol receptors are coexpressed, only one form may predominate depending on the cell type. Some tissues, such as the kidney and colon, are  known targets for aldosterone

 

  

Page 911

CLINICAL CORRELATION 21.2  Apparent Mineralocorticoid Excess Syndrome Some patients (usually children) exhibit symptoms, including hypertension, hypokalemia,  and suppression of the reninangiotensin–aldosterone system, that would be expected if  they were hypersecreting aldosterone. Since bioassays of plasma and urine sometimes fail  to identify any excess of mineralocorticoids, these patients are said to suffer from the  apparent mineralocorticoid excess (AME) syndrome. This syndrome results as a  consequence of the failure of cortisol inactivation by the 11b ­hydroxy­steroid  dehydrogenase enzyme. Inactivity of this key enzyme gives cortisol direct access to the  renal mineralocorticoid receptor. Since cortisol circulates at much higher concentrations  than aldosterone, this glucocorticoid saturates these mineralocorticoid receptors and  functions as an agonist, causing sodium retention and suppression of the renin– angiotensin–aldosterone axis. Although this AME syndrome can result from a congenital  defect in the distal nephron 11b ­hydroxysteroid dehydrogenase isoform, which renders  the enzyme incapable of converting cortisol to cortisone (binds poorly to  mineralocorticoid receptors), it can also be acquired by ingesting excessive amounts of  licorice. The major component of licorice is glycyrrhizic acid and its hydrolytic product,  glycyrrhetinic acid (GE). This active ingredient (GE) acts as a potent inhibitor of 11b ­ hydroxysteroid dehydrogenase. By blocking activity of this inactivating enzyme, GE  facilitates the binding of cortisol to renal mineralocorticoid receptors and hence induces  AME syndrome. Edwards, C. R. W. Primary mineralocorticoid excess syndromes. In: L. J. DeGroot  (Ed.), Endocrinology. Philadelphia: Saunders, pp. 1775–1803, 1995; and Shackleton,  C. H. L., and Stewart, P. M. The hypertension of apparent mineralocorticoid excess  syndrome. In: E. G. Biglieri and J. C. Melby (Eds.), Endocrine Hypertension. New  York: Raven Press, 1990, pp. 155–173.   TABLE 21.4 Steroid Hormone Receptor Responsive DNA Elements:  Consensus Acceptor Site Element

DNA Sequencea

POSITIVE Glucocorticoid responsive  element (GRE) Mineralocorticoid responsive  element (MRE) 5 ­GGTACAnnnTGTTCT­3 Progesterone responsive element  (PRE) Androgen responsive element  (ARE) Estrogen responsive element (ERE)

  5 AGGTCAnnnTCACT­3

NEGATIVE Glucocorticoid responsive element

5 ­ATYACNnnnTGATCW­3

Source: Data are summarized from work of Beato, M. Cell 56:355, 1989. a n, any nucleotide; Y, a purine; W, a pyrimidine.

  and express relatively high levels of mineralocorticoid receptors as well as glucocorticoid receptors. These mineralocorticoid target tissues express the enzyme 11 b ­ hydroxysteroid dehydrogenase (see Clin. Corr. 21.2). This enzyme converts cortisol and corticosterone, both of which can bind to the mineralocorticoid receptor  with high affinity, to their 11­keto analogs, which bind poorly to the mineralocorticoid receptor. This inactivation of corticosterone and cortisol, which circulate at much  higher concentrations than aldosterone, facilitates the binding of aldosterone to the mineralocorticoid receptors in these classical target tissues. In tissues that express  mineralocorticoid receptors but are not considered target tissues, this enzyme may not be expressed, and in these situations the mineralocorticoid receptors may simply  function as pseudo­glucocorticoid receptors and mediate the effects of low circulating levels of cortisol (predominant glucocorticoid in humans). Thus the  mineralocorticoid and glucocorticoid receptors may regulate the expression of an overlapping gene network in various target tissues. As also indicated in Table 21.4,  the activated estrogen–receptor complex recognizes a distinct or unique response element. All of the response elements listed at the top of Table 21.4 function as  positive elements, since binding of the indicated steroid receptors results in an increase in the rate of transcription of the associated gene. Glucocorticoid hormones also repress transcription of specific genes. For example, glucocorticoids are known to repress transcription of the proopio­melanocortin  gene (POMC) (see p. 849), which contains the ACTH sequences. Glucocorticoid­mediated repression of POMC gene expression thus plays a key role in the  negative feedback loop regulating the rate of secretion of ACTH and ultimately cortisol. Negative glucocorticoid response elements (nGREs) mediate this repression of  the POMC gene as well as other important genes. A general model of positive as well as negative transcriptional effects mediated by steroid receptors is shown in  Figure 21.14: In (a) binding of a steroid receptor (R) homodimer to its response element allows it to interact synergistically with a positive transcription factor (TF) and  hence induce gene transcription; in (b) binding of a receptor dimer to its response element displaces a positive transcription factor (TF) but has no or weak  transactivation potential because no synergizing factor is nearby; and in (c) the DNA­AP1 (positive factor) may interact in a protein–protein fashion in such a way that  the transactivating functions of both proteins are inhibited and gene transcription is repressed.

 

  

Page 912

Figure 21.14  Positive and negative transcriptional effects  of steroid receptors.  Redrawn from Renkowitz, R. Ann. N.Y.  Acad. Sci. 684:1, 1993.

Some members of this receptor supergene family can mediate gene silencing. Silencer elements, in analogy to enhancer elements, function independently of their  position and orientation. The silencer for a particular gene consists of modules that independently repress gene activity. In the absence of their specific ligands, the  thyroid hormone receptor (T3R) and retinoic acid receptor (RAR) appear to bind to specific silencer elements and repress gene transcription. This silencing  activity may occur via destabilization of the transcription initiation complex or via direct or indirect effects on the carboxy­terminal domain of RNA polymerase II. After  binding of their respective ligands, these two receptors lose this silencing activity and are converted into transactivators of gene transcription. As indicated in Figure 21.14, dimerization of receptor monomers is a prerequisite for efficient DNA binding and transcriptional activation by most steroid receptors.  Strong interactions between these monomers are mediated by the ligand­binding domains of several steroid receptors. The dimerization domain of the ligand­binding  domain has been proposed to form a helical structure containing a succession of hydrophobic sequences that would generate a leucine zipper­like structure or a helix– turn–zipper motif (see p. 110), which are known to be necessary for the dimerization of other transcription factors. Although the majority of receptors in this  superfamily form homodimers, heterodimers have also been detected. More specifically, a distinct class of retinoic acid receptors, classified as retinoid X receptors  (RXRs), regulate gene expression via heterodimerization with the other distinct form of the retinoic acid receptor (RAR), the thyroid hormone receptor, and other  members of this receptor superfamily. A model for the stabilization of the transcriptional preinitiation complex by an RXR/RAR heterodimer is presented in Figure  21.15. Thus the changes produced in different cells by the activation of steroid hormone receptors may be different in different cells that contain the relevant receptor in  suitable concentration. The whole process is triggered by the entry of the steroidal ligand in amounts that supersede the dissociation constant of the receptor. The  different phenotypic changes in different cell types in response to a specific hormone then summate to give the systemic or organismic response to the hormone.

Figure 21.15  Model for stabilization of preinitiation complex by an RXR/RAR heterodimer.  TF, transcription factor; LBD, ligand­binding domain; DBD, DNA­binding  domain; AF1, activation function located in amino­terminal region of receptor, which  may provide contact with cell­specific proteins; AF2, activation function located  within ligand­binding domain, which interacts directly with transcriptional machinery.

 

  

Page 913

Figure 21.16  Model of a typical steroid hormone receptor.  The results are derived from studies on cDNA in various laboratories,  especially those of R. Evans and K. Yamamoto.

Some Steroid Receptors Are Part of the cErbA Family of Proto­oncogenes The glucocorticoid receptor is conveniently divided into three major functional domains (Figure 21.16). Starting at the C terminus, the steroid­binding domain is  indicated and has 30–60% homology with the ligand­binding domains of other receptors in the steroid receptor family. The more alike two steroids that bind  different receptors are, the greater the extent of homology to be anticipated in this domain. The steroid­binding domain contains a sequence that may be involved in the  binding of molybdate and a dimer of the 90­kDa heat shock protein whose function would theoretically result in the assembly of the high molecular weight unactivated– nontransformed steroid–receptor complex. To the left of that domain is a region that modifies transcription. In the center of the molecule is the DNA­binding domain.  Among the steroid receptors there is 60–95% homology in this domain. Two zinc fingers (see p. 108) interact with DNA. The structure of the zinc finger DNA­binding  motif is shown in Figure 21.17. The N­terminal domain contains the principal antigenic domains and a site that modulates transcriptional activation. The amino acid  sequences in this site are highly variable among the steroid receptors. These features are common to all steroid receptors. The family of steroid receptors is  diagrammed in Figure 21.17. The ancestor to which these receptor genes are related is v­erbA or c­erbA (see p. 889). v­ErbA is an oncogene that binds to DNA but  has no ligand­binding domain. In some cases the DNA­binding domains are homologous enough that more than one receptor will bind to a common responsive  element (consensus sequence on DNA) as shown in Table 21.4. In addition to those genes pictured in Figure 21.18, the aryl hydrocarbon receptor (Ah) may also  be a member of this family. The Ah receptor binds carcinogens with increasing affinity paralleling increasing carcinogenic potency and translocates the carcinogen to the  cellular nucleus unless the receptor is already located in the nucleus. The N­terminal portions of the receptors usually contain major antigenic sites and may also contain  a site that is active in modulating binding of the receptor to DNA.

Figure 21.17  Structure of the zinc finger located within the glucocorticoid  receptor DNA­binding domain as determined by X­ray crystallography.  Yellow circles indicate amino acid residues (located in GR monomer) that  interact with base pairs. Blue circles are those making phosphate backbone  contacts. Green circles are those participating in dimerization.  Redrawn from Luisi, B. F., Schwabe, J. W. R., and Freedman, L. P. In:  G. Litwack (Ed.), Vitamins and Hormones, Vol. 49. San Diego,  Academic Press, 1994, pp. 1–47.

 

  

Page 914

Figure 21.18  Steroid receptor gene superfamily.  T3, triiodothyronine; RA, retinoic acid; D3,  dihydroxy vitamin D ; E2, estradiol;  3

CORT, cortisol; ANDR, androgen; PROG,  progesterone; ALDO, aldosterone. Figure shows  roughly the relative sizes of the genes for these  receptors. Information derived from the  laboratories of R. Evans, K. Yamamoto, P. Chambon,  and others. In some cases there is high homology  in DNA­binding domains and lower homology in  ligand­binding domains.

Thyroid hormone and retinoic acid receptors are also members of the same superfamily of receptors although their ligands are not steroids. They do contain six­ membered rings as shown in Figure 21.19. For some steroid receptors the A ring is the prominent site of recognition by the receptor, presenting the likelihood that the  A ring inserts into the binding pocket of the receptor. In some cases, derivatives of the structures with a six­membered ring bind to the estradiol and glucocorticoid  receptors. Thus the ring structures of thyroid hormone and retinoic acid have structural similarities not unlike many of the steroidal ligands involved in binding. The receptors in this large gene family may act as transcriptional activators that together with other transcriptional regulators bring about gene activation. 21.8—  Receptor Activation:  Upregulation and Downregulation Little is known about activation of steroid receptors. Activation converts a non­DNA­binding form (unactivated–nontransformed) of the receptor to a form (activated– transformed) that is able to bind nonspecific DNA or specific DNA (hormone­responsive element). The likelihood that certain receptors are cytoplasmic  (glucocorticoid receptor and possibly the mineralocorticoid receptor) while others seem to be nuclear (progesterone, estradiol, vitamin D3, and androgen receptors)  may have a bearing on the significance of the activation phenomena. Most information is available for cytoplasmic receptors. The current view is that the non­DNA­ binding form is a heteromeric trimer consisting of one molecule of receptor and a dimer of 90­kDa heat shock protein, as shown in Figure 21.20. The DNA­binding  site of the receptor is blocked by the heteromeric proteins or by some other factor or by a combination of both. Upon activation–transformation a stepwise  disaggregation of this complex could occur, leading to the activated receptor having its DNA­binding site fully exposed. The reaction may be initiated by the binding of  steroid to the ligand­binding site that produces a conformational change in the receptor protein.

Figure 21.19  Structures of retinoic acid (vitamin  A acid) and 3,5,3 ­triiodothyronine.

Although the conditions required to induce activation in vitro are well known, the primary signal within the cell is not. Many believe that the binding of ligand alone is  not sufficient to cause the activation process. Clearly, elevated temperature is a requirement for this conformational change, since incubation of target cells with  appropriate steroids at low temperatures fails to result in in vivo activation and subsequent translocation. Once the liberated receptor is free in the cytoplasm it crosses  the perinuclear membrane, perhaps through a nucleopore, to enter the nucleus. It binds nonspecifically and specifically to

 

  

Page 915

chromatin, probably as a dimer, presumably in search of the specific response element (Table 21.4). Thus these receptors are transacting factors and may act in  concert with other transacting factors to provide the appropriate structure to initiate transcription. Most steroid receptors have in their DNA­binding domains an  SV40­like sequence (i.e., Pro­Lys­Lys­Lys­Arg­Lys­Val) known to code for nuclear translocation. Steroid receptors have variants of this sequence; some  degeneracy is permitted but probably a specific lysine residue cannot be altered. This signal may provide recognition for the nucleopore.

Figure 21.20  Hypothetical minimal model of a  non­DNA­binding form of a steroid receptor.  This form of the receptor cannot bind to DNA  because the DNA­binding site is blocked by  the 90­kDa hsp proteins or by some other  constituent. Mass of this complex is  approximately 300 kDa.

Steroid Receptors Can Be Upregulated or Downregulated Depending on Exposure to the Hormone In general, many membrane or intracellular receptors are downregulated when the cell has been exposed to a certain amount of the hormonal ligand. In some cases,  the downregulation is called "desensitization." Downregulation can take many forms. For membrane receptors the mechanism may be internalization by endocytosis  of the receptors after exposure to hormone (see p.876). Internalization reduces the number of receptors on the cell surface and renders the cell less responsive to  hormone; that is, desensitizes the cell. In the case of intracellular steroid receptors, downregulation generally takes the form of reducing the level of receptor mRNA,  which decreases the concentration of receptor molecules. The receptor gene may have a specific responsive element on its promoter whose action results in an  inhibition of transcription of receptor mRNA or the receptor may stimulate transcription of a gene that codes for a protein that degrades the mRNA of the receptor.  Sequences are now being recognized on receptor gene promoters that may bind activated steroid–receptor complexes and result in inhibition of transcription (Table  21.4). Downregulation of receptors by their own ligands plays an important physiological role because it prevents overstimulation of target cells when circulating  hormone levels are elevated. Although downregulation of steroid receptor levels by their cognate hormones appears to be the most frequently detected form of autoregulation, it is not common to  all target cells. In fact, glucocorticoid­mediated upregulation has been reported in a number of responsive cells. Since all of these cells are growth inhibited by these  hormones, it was initially suggested that hormone­mediated upregulation may be required for subsequent growth inhibition. However, the fact that glucocorticoid­ mediated upregulation also occurs in human lymphoid cells, which express glucocorticoid receptors but are not growth inhibited by these steroids, demonstrates that  this positive autoregulation is neither the result nor cause of hormone­mediated growth arrest. 21.9—  Specific Example of Steroid Hormone Action at Cell Level: Programmed Death Programmed cell death or apoptosis is a suicide process by which cells die according to a program that may be beneficial for the organism. It can result from the rise  or fall in the level of a specific hormone(s). Uterine endometrial cells at the beginning of menstruation are an example where programmed cell death is initiated by the  fall in levels of progesterone and estradiol in the blood (see Clin. Corr. 21.3). Another case is apoptosis of thymus cells during development when the adrenal cortex  becomes functional and begins to synthesize and secrete relatively large amounts of cortisol. A newborn has a large thymus but when cortisol is synthesized and  released the thymus cortical cells begin to die until a resistant core of cells is reached and the gland achieves its adult size. Thus programmed cell death is a mechanism  used in development for the maturation of certain organs as well as in cyclic systems where cells

 

  

Page 916

CLINICAL CORRELATION 21.3  Programmed Cell Death in the Ovarian Cycle During the ovarian cycle, the ovarian follicle expels the mature ovum at day 14 and the  remaining cells of the follicle are differentiated into a functional corpus luteum. The  corpus luteum produces some estradiol to partially replace that provided earlier by the  maturing follicle. However, its principal product is progesterone. Estradiol and  progesterone are the main stimulators of uterine endometrial wall thickening in preparation  for implantation. One of the proteins induced by estradiol action in the endometrium is the  progesterone receptor. Thus the uterine endometrial cells become exquisitely sensitive to  estradiol as well as progesterone. The corpus luteum supplies the latter, but in the  absence of fertilization and development of an embryo, the corpus luteum lives only for a  short while and then atrophies because of lack of LH or chorionic gonadotropin, a  hormone produced by the early embryo. The production of oxytocin and PGF2a in the  ovary may bring about the destruction of the corpus luteum (luteolysis). Blood levels of  estradiol and progesterone fall dramatically after luteolysis and the stimulators of uterine  endometrial cells disappear, causing degeneration of this thickened, vascularized layer of  tissue and precipitating menstruation. These cells die by programmed cell death  (apoptosis) due to the withdrawal of steroids. The hallmark of programmed cell death is  internucleosomal cleavage of DNA. Thus programmed cell death appears to play specific  roles in development and in tissue cycling either due to a specific hormonal stimulus or to  withdrawal of hormone(s) as described here. Erickson, G. F., and Schreiber, J. R. Morphology and physiology of the ovary. In: K. L.  Becker (Ed.), Principles and Practice of Endocrinology and Metabolism. New York:  Lippincott, 1990, p. 776; Rebar, R. W., Kenigsberg, D., and Hogden, G. D. The normal  menstrual cycle and the control of ovulation. In: K. Becker (Ed.), Principles and  Practice of Endocrinology and Metabolism. New York: Lippincott, 1990, p. 788; and  Hamburger, L., Hahlin, M., Hillensjo, T., Johanson, C., and Sjogren, A. Luteotropic and  luteolytic factors regulating human corpus luteum function. Ann. N.Y. Acad. Sci. 541:485,  1988.   proliferate and then regress until another cycle is initiated to begin the proliferation all over again, as is the case with the ovarian cycle. Glucocorticoid­induced apoptosis in thymocytes is mediated by the intracellular glucocorticoid receptor. There are two phases to this complex process: inhibition of  cell proliferation (cytostatic phase) followed by a cytolytic phase characterized by internucleosomal DNA cleavage and ultimate cell death (cytolytic phase). These two  phases are not necessarily linked, since some cells are growth inhibited, but not lysed, by glucocorticoid hormones. The precise mechanism by which glucocorticoid– receptor complexes induce cell death is not fully understood. Exposure to hormone may result in a conformational change in chromatin with the unmasking of  internucleosomal linker DNA regions, which are substrates for a nuclease. Treatment of thymocytes with glucocorticoids results in the activation of a constitutive,  endogenous Ca2+/Mg2+­dependent endonuclease, while similar treatment of human leukemic T cells results in the activation of Ca2+/Mg2+­independent nuclease.  Recent studies have demonstrated that the Ca2+/Mg2+­dependent nuclease that is activated by glucocorticoids in rat thymocytes is homologous with a cyclophilin.  These proteins are high­affinity binding proteins for the immunosuppressive drug, cyclosporin A, and have Ca2+/Mg2+­dependent nuclease activity. The mechanism(s)  by which glucocorticoid hormones induce lysis of thymocytes versus leukemic T cells appears to differ in several other respects. Treatment of sensitive T cells with  these hormones results in upregulation of glucocorticoid receptor mRNA levels, while identical treatment of thymocytes appears to result in down­regulation of mRNA  levels. Also, the mRNA levels for an important growth factor, c­myc, are repressed in glucocorticoid­treated T cells but induced in thymocytes. Thus the cytostatic  and cytolytic phases of apoptosis may be mediated by slightly different pathways in these two different cell types. Bibliography Argentin, S., Sun, Y. L., Lihrmann, I., Schmidt, T. J., Drouin, J., and Nemer, M. Distal cis­acting promoter sequences mediate glucocorticoid stimulation of cardiac  ANF gene transcription. J. Biol. Chem. 266:23315, 1991. Baulieu, E.­E. Steroid hormone antagonists at the receptor level: a role for heat­shock protein MW 90,000 (hsp 90). J. Cell Biochem. 35:161, 1987. Beato, M. Gene regulation by steroid hormones. Cell 56:335, 1989.

 

  

Page 917

Carson­Jurica, M. A., Schrader, W. T., and O'Malley, B. W. Steroid receptor family: structure and functions. Endocr. Rev. 11:201, 1990. Chrousos, G. P., Loriaux, D. L., and Lipsett, M. B. (Eds.). Steroid Hormone Resistance. New York: Plenum Press, 1986. Drouin, J., Sun, Y. L. Tramblay, S., Schmidt, T. J., deLean A., and Nemer, M. Homodimer formation is rate­limiting for high affinity DNA binding by glucocorticoid  receptor. Mol. Endocrinol, 6:1299, 1992. Evans, R. M. The steroid and thyroid hormone receptor superfamily. Science 240:889, 1988. Giguere, V., Hollenberg, S. M., Rosenfeld, M. G., and Evans, R. M. Functional domains of the human glucocorticoid receptor. Cell 46:645, 1986. Green, S., Kumar, V., Theulaz, I., Wahli, W., and Chambon, P. The N­terminal DNA­binding ''zinc­finger" of the estrogen and glucocorticoid receptors determines  target gene specificity. EMBO J. 7:3037, 1988. Gustafsson, J. A., Carlstedt­Duke, J., Poellinger, L., et al. Biochemistry, molecular biology, and physiology of the glucocorticoid receptor. Endocr. Rev. 8:185, 1987. Huft, R. W., and Pauerstein, C. J. Human Reproduction: Physiology and Pathophysiology. New York: Wiley, 1979. Litwack, G. (Ed.). Biochemical Actions of Hormones, Vols. 1–14. New York: Academic Press, 1973–1987. Litwack, G. (Ed.). Steroids. Vitamins and Hormones, Vol. 49. San Diego: Academic Press, 1994. Litwack, G. (Ed. in Chief). Vitamins and Hormones, Vol. 51. San Diego: Academic Press, 1995. Mester, J., and Baulieu, E.­E. Nuclear receptor superfamily. In: L. J. DeGroot, (Ed.) Endocrinology, 3rd ed., Philadelphia: Saunders, 1995, pp. 93–118. Norman, A. W., and Litwack, G. Hormones. Orlando, FL: Academic Press, 1987. O'Malley, B. W., Tsai, S. Y., Bagchi, M., Weigel, N. L., Schrader, W. T., and Tsai, M.­J. Molecular mechanism of action of a steroid hormone receptor. Recent  Prog. Horm. Res. 47:1, 1991. Renkawitz, R. Repression mechanisms of v­erbA and other members of the steroid receptor superfamily. Ann. N.Y. Acad. Sci. 684:1, 1993. Rusconi, S., and Yamamoto, K. R. Functional dissection of the hormone and DNA binding activities of the glucocorticoid receptor. EMBO J. 6:1309, 1987. Schmidt, T. J., and Meyer, A. S. Autoregulation of corticosteroid receptors. How, when, where and why? Receptor 4:229, 1994. Schwabe, J. W. R., and Rhodes, D. Beyond zinc fingers: steroid hormone receptors have a novel structural motif for DNA recognition. Trends Biochem. Sci. 16:291,  1991. Wahli, W., and Martinez, E. Superfamily of steroid nuclear receptors—positive and negative regulators of gene expression. FASEB J. 5:2243, 1991. Questions J. Baggott and C. N. Angstadt 1. The C­21 steroid hormones include: A. aldosterone. B. dehydroepiandrosterone. C. estradiol. D. testosterone. E. vitamin D3. 2. Side chain cleavage enzyme complex activity may be stimulated by all of the following EXCEPT: A. cAMP. B. Ca2+ released via stimulation of the IP3 pathway. C. Ca2+ entering the cell through a channel. D. 5 ­AMP. E. induction of the StAR protein. 3. 

5

­Pregnenolone is a precursor of all of the following EXCEPT:

A. aldosterone. B. cortisol. C. 17b ­estradiol. D. progesterone. E. vitamin D3. 4. Major steps in the inactivation and excretion of ALL classes of steroid hormones (except vitamin D3) include: A. conjugation to glucuronic acid. B. conjugation to sulfuric acid. C. hydroxylation. D. oxidation. E. side chain cleavage. 5. All of the following may be involved in the action of steroid hormone receptors EXCEPT: A. binding of the hormone to an intracellular receptor. B. activation of a G­protein. C. association with a heat shock protein (hsp90) with a cytoplasmic receptor. D. binding to a receptor in the nucleus. E. translocation of a cytoplasmic hormone–receptor complex into the nucleus. 6. Retinoic acid and its derivatives: A. may activate genes by preventing the binding of receptor proteins to silencer elements. B. bind to homodimeric proteins, which in turn bind to DNA. C. bind to DNA via leucine zipper motifs. D. are vitamin derivatives and hence have no effect on regulation of gene expression. E. may substitute for thyroid hormones in binding to the thyroid hormone receptor. 7. Reactions in the pathway of synthesis of active vitamin D involve all of the following organs EXCEPT: A. skin. B. kidney. C. liver. D. intestine. Refer to the following for Questions 8–11: A. corticosteroid binding globulin B. serum albumin C. sex hormone­binding globulin D. androgen­binding protein E. transferrin 8. Major aldosterone carrier in blood. 9. Produced by the Sertoli cells. 10. Binds about 20% of the cortisol in the plasma. 11. At puberty decreases more in males than in females. 12. Receptors for steroid hormones are found in: A. cell membranes. B. cytoplasm. C. ribosomes. D. mitochondria. E. Golgi apparatus.

 

  

Page 918

13. Which of the following involve(s) a response element of DNA that differs from all of the other listed hormones? A. estrogen B. glucocorticoid C. mineralocorticoid D. progesterone 14. All of the following receptors may belong to the steroid receptor gene superfamily EXCEPT: A. aryl hydrocarbon receptor. B. erbA protein. C. retinoic acid receptor. D. thyroid hormone receptor. E. a ­tocopherol receptor. Refer to the following for Questions 15 and 16: A. programmed cell death B. stress response C. downregulation of steroid receptors D. upregulation of steroid receptors E. silencing 15. Mechanism for the maturation of certain organs. 16. Receptor mRNA is reduced. Answers 1. A B and D: These are C­19 androgens. C: Estradiol is a C­18 estrogen. E: Vitamin D3 is a C­27 compound (pp. 899–900). 2. D See Figure 21.4, p. 898. 3. E See Figure 21.5 (p. 899) and Figure 21.6 (p. 900) for the synthesis of A–D. The synthesis of vitamin D3 is summarized in Figure 21.11 (pp. 906–907). 4. A Oxidation (including hydroxylation) and reduction are common in steroid hormone degradation. Glucocorticoids undergo side chain cleavage. Conjugation to  sulfate is important in the excretion of androgens. But conjugation to glucuronide is significant for all steroid hormones except vitamin D3 (Table 21.2, p. 902). 5. B G­proteins are generally associated with signal transduction for receptors on the membrane surface. See Figure 21.13 for the roles of the other choices. 6. A The retinoic acid receptor (RAR) binds to specific silencer elements in the absence of the ligand, retinoic acid. When bound, transcription of the gene is  repressed. In addition, there are retinoid X receptors (RXR), which also affect gene expression, via heterodimerization with RAR (p. 912, Figure 21.14b). 7. D Intestine is a target organ of the active hormone, but is not involved in synthesis. See Figure 21.11, p. 906. A: Light­induced cleavage of 7­dehydrocholesterol  occurs in the skin. B: Hydroxylation of 25­(OH)D3 occurs in the kidney. C: Hydroxylation of D3 occurs in the liver. 8. B See p. 908. 9. D See p. 908. 10. B Cortisol­binding globulin carries most of the cortisol. Serum albumin, however, nonspecifically binds a large number of hydrophobic substances, including  cortisol (p. 908). 11. C As a result, there is more unbound testosterone circulating in the blood of males (p. 908). 12. B B: In addition, the nucleus contains steroid hormone receptors. See Figure 21.13, p. 909. A: Membrane receptors are generally associated with  nonhydrophobic hormones, such as epinephrine and peptide hormones (Figure 21.4, p. 898). 13. A The positive glucocorticoid response element is the same as the mineralocorticoid response element and the progesterone response element. The estrogen  response element differs (Table 21.4, p. 911). 14. E Note that c­erbA is a protooncogene. See p. 913, Figure 21.17. 15. A Thymus cortical cells are killed by cortisol (p. 915). 16. C This contrasts with cell membrane receptors, which can be internalized to render the cell less responsive (p. 915).

 

  

Page 919

Chapter 22—  Molecular Cell Biology Thomas E. Smith

22.1 Overview

920

22.2 Nervous Tissue: Metabolism and Function

920

ATP and Transmembrane Electrical Potential in Neurons

921

Neuron–Neuron Interaction Occurs through Synapses

923

Synthesis, Storage, and Release of Neurotransmitters

924

Termination of Signals at Synaptic Junctions

927

Acetylcholine

928

Catecholamines

929

5­Hydroxytryptamine (Serotonin)

930

4­Aminobutyrate (g­Aminobutyrate)

931

Neuropeptides Are Derived from Precursor Proteins

931

22.3 The Eye: Metabolism and Vision

932

The Cornea Derives ATP from Aerobic Metabolism

933

Lens Consists Mostly of Water and Protein

933

The Retina Derives ATP from Anaerobic Glycolysis

935

Visual Transduction Involves Photochemical, Biochemical, and Electrical  Events

936

Photoreceptor Cells Are Rods and Cones

937

Color Vision Originates in the Cones

944

Other Physical and Chemical Differences between Rods and Cones

945

22.4 Muscle Contraction

946

Skeletal Muscle Contraction Follows an Electrical to Chemical to  Mechanical Path

946

Myosin Forms the Thick Filament of Muscle

949

Actin, Tropomyosin, and Troponin Are Thin Filament Proteins

951

Muscle Contraction Requires Ca2+ Interaction

954

Energy for Muscle Contraction Is Supplied by ATP Hydrolysis

954

Model for Skeletal Muscle Contraction

957

Calcium Regulates Smooth Muscle Contraction

959

22.5 Mechanism of Blood Coagulation Clot Formation Is a Membrane­Mediated Process

960

Reactions of the Intrinsic Pathway

961

Reactions of the Extrinsic Pathway

963

Thrombin Converts Fibrinogen to Fibrin

964

Major Roles of Thrombin

966

Formation of a Platelet Plug

967

Properties of Some of the Proteins Involved in Coagulation

968

Role of Vitamin K in Protein Carboxylase Reactions

970

Control of the Synthesis of Gla­Proteins

971

Dual Role of Thrombin in Promoting Coagulation and Clot Dissolution

972

The Allosteric Role of Thrombin in Controlling Coagulation

974

Inhibitors of the Plasma Serine Proteinases

975

Fibrinolysis Requires Plasminogen and Tissue Plasminogen Activator (t­ PA) to Produce Plasmin

975

Bibliography

976

Questions and Answers

977

Clinical Correlations

   

  

960

 

22.1 Lambert–Eaton Myasthenic Syndrome

927

22.2 Myasthenia Gravis: A Neuromuscular Disorder

929

22.3 Macula Degeneration: Other Causes of Vision Loss

936

22.4 Niemann–Pick Disease and Retinitis Pigmentosa

938

22.5 Retinitis Pigmentosa Resulting from a De Novo Mutation in the  Gene Coding for Peripherin

940

22.6 Abnormalities in Color Perception

946

22.7 Troponin Subunits As Markers for Myocardial Infarction

954

22.8 Voltage­Gated Ion Channelopathies

956

22.9 Intrinsic Pathway Defects: Prekallikrein Deficiency

963

22.10 Classic Hemophilia

969

22.11 Thrombosis and Defects of the Protein C Pathway

971

Page 920

22.1—  Overview Animals sense their environment through the responses of certain organs to stimuli: touch, pain, heat, cold, intensity (light or noise), color, shape, position, pitch, quality,  acid, sweet, bitter, salt, alkaline, fragrance, and so on. Externally, these generally reflect responses of the skin, eye, ear, tongue, and nose to stimuli. Some of these  signals are localized to the point at which they occur; others—sound and sight—are projected in space, that is, the environment outside and distant to the animal. Discrimination of these signals occurs at the point of reception, but acknowledgment of what they are occurs as a result of secondary stimulation of the nervous system  and transmission of the signals to the brain. In many instances, a physical response is indicated, which results in muscular activity, either voluntary or involuntary.  Common to these events is electrical activity associated with signal transmission along neurons and chemical activity associated with signal transmission across synaptic  junctions. In all cases, stimuli received from the environment in the form of pressure (skin, feeling), light (eye, sight), noise (ear, hearing), taste (tongue), or smell (nose)  are converted (transduced) into electrical impulses and to some other form of energy in order to effect the desired terminal response dictated by the brain. A  biochemical component is associated with each of these events. General biochemical mechanisms of signal transduction and amplification will be discussed as they relate to biochemical events involved in nerve transmission, vision,  and muscular contraction. Finally, a specialized case of biochemical signal amplification will be discussed, namely, blood coagulation. This process is initiated on  membrane surfaces as a result of the exposure of specific proteins that act as receptors and form nucleation sites for formation of multienzyme complexes. These  multienzyme complexes lead to the amplification of blood coagulation through a cascade mechanism. 22.2—  Nervous Tissue:  Metabolism and Function Knowledge of the chemical composition of the brain began with the work of J. L. W. Thudichum in 1884 and the publication of his monogram, "A Treatise on the  Chemical Composition of the Brain, Based Throughout on Original Research" (cited in West and Todd, Textbook of Biochemistry, MacMillan, 1957). Thudichum's  research was supplemented with the work of others during those earlier years. There have been almost explosive advances during more recent years, through the use  of molecular biological techniques, not only in our knowledge of the composition of the brain but also of molecular mechanisms involved in many brain/neuronal  functions. About 2.4% of an individual's body weight is nervous tissue, of which approximately 83% is the brain. The nervous system provides the communications network  between the senses, the environment, and all parts of the body. The brain is the command center. This system is always functioning and requires a large amount of  energy to keep it operational. Under normal conditions, the brain derives its energy from glucose metabolism. Ketone bodies can cross the blood–brain barrier  and be metabolized by brain tissue. Their metabolism becomes more prominent during starvation, but even then they cannot replace the need for glucose. The human  brain uses approximately 103 g of glucose per day. For a 1.4­kg brain, this corresponds to a rate of utilization of approximately 0.3  mol min–1 g–1 of tissue. This rate  of glucose utilization represents a capacity for ATP production through the tricarboxylic acid (TCA) cycle alone of approximately 6.8  mol min–1 g–1 of tissue. Of  course, the TCA cycle is not 100% efficient for ATP production, nor is all of the glucose metabolized through it. Most of the ATP used by the brain and other nervous  tissue is

 

  

Page 921

generated aerobically through the TCA cycle, which functions at near maximum capacity. Glycolysis functions at approximately 20% capacity. Much of the energy  used by the brain is to maintain ionic gradients across the plasma membranes, to effect various storage and transport processes, and for the synthesis of  neurotransmitters and other cellular components. Two features of brain composition are worth noting. It contains specialized and complex lipids, but they appear to function to maintain membrane integrity (see  Chapter 5) rather than to have metabolic roles. There is generally a rapid turnover rate of brain proteins relative to other body proteins in spite of the fact that the  cells do not divide after they have differentiated. Cells of the nervous system responsible for collecting and transmitting messages are the neurons. They are very highly specialized (Figure 22.1). Each neuron consists  of a cell body, dendrites that are short antenna­like protrusions that receive signals from other cells, and an axon that extends from the cell body and transmits signals  to other cells. The central nervous system (CNS) is a highly integrated system where individual neurons can receive signals from a variety of different sources, including  both inhibitory and excitatory stimuli.

Figure 22.1  A motor nerve cell and investing membranes.

Cells other than neurons exist in the CNS. In the brain, there are about 10 times more glial cells than there are neurons. Glial cells occupy spaces between neurons and  provide some electrical insulation. Glial cells are generally not electrically active, and they are capable of division. There are basically five types of glial cells: Schwann  cells, oligodendrocytes, microglia, ependymal cells, and astrocytes. Each type of glial cell has a specialized function, but only astrocytes appear to be directly  associated with biochemical functions related to neuronal activity. One is metabolic (see discussion below on GABA) and the other anatomical. Astrocytes send out processes at the external surfaces of the CNS. These processes are linked to form anatomical complexes that provide sealed barriers and isolate  the CNS from the external environment. Astrocytes also send out similar processes to the circulatory system, inducing the endothelial cells of the capillaries to become  sealed by forming tight junctions that prevent the passive entry into the brain of water­soluble molecules. These tight junctions form what is commonly known as the  blood–brain barrier. Water­soluble compounds enter the brain only if there are specific membrane transport systems for them. The normal individual has between 1011 and 1013 neurons, and communication between them is by electrical and chemical signals. Electrical signals transmit nerve  impulses down the axon and chemicals transmit signals across the gap between cells. Some of the biochemical events that give the cell its electrical properties and are  involved in the propagation of an impulse will be discussed. ATP and Transmembrane Electrical Potential in Neurons Adenosine triphosphate generated from the metabolism of glucose is used to help maintain an equilibrium electrical potential across the membrane of the neuron of  approximately –70 mV, with the inside being more negative than the outside. This potential is maintained by the action of the Na+, K+ ion pump (see pp. 206–207),  the energy for which is derived from the hydrolysis of ATP to give ADP and inorganic phosphate. This system pumps Na+ out of the cell by an antiport mechanism,  whereas K+ is moved into the cell. The channels through which Na+ enters the cell are voltage gated; that is, the proteins of the channel undergo a charge­dependent  conformation change and open when the electrical potential across the membrane decreases (specifically, becomes less negative) by a value greater than some  threshold value. When the membrane becomes depolarized, Na+, whose concentration is higher outside the cell than inside, flows into the cell and K+, whose  concentration is greater inside the cell, flows out of the cell, both going down their respective concentration gradients. The channels are open in a particular  geographical

 

  

Page 922

Figure 22.2  Schematic of Na+ channels opening and closing during nerve impulse transmission.  Redrawn from Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J.  Molecular Biology of the Cell, 2d ed. New York: Garland Publishing, 1989, p. 1071.

region of the cell for fractions of a millisecond (Figure 22.2). The localized depolarization (voltage change) causes a conformation change in the neighboring proteins  that make up the voltage­gated ion channels. These channels open momentarily to allow more ions in and, thus, by affecting adjacent channel proteins, allow the  process to continue down the axon. There is a finite recovery

 

  

Page 923

time. During this time, the proteins that form the channels cannot repeat the process of opening. Thus charge propagation proceeds in one direction. It is the  progressive depolarization and repolarization along the length of the axon that allow electrical impulses to be propagated undiminished in amplitude. Electrical impulse  transmission is a continuous process in nervous tissue, and it is the ATP generated primarily from the metabolism of glucose that keeps the system operational. A current area of active research in biochemistry involves the use of gene cloning and engineering techniques to isolate ion channel proteins and to determine their  structures and elucidate their mechanisms of action. A considerable amount of information has been obtained in recent years on how mutations in voltage­gated ion  channels may affect muscle function. Considerably less is known, however, about the relationship between structural disorders of ion channels in neurons and clinical  disorders. TABLE 22.1 Some of the Neurotransmitters Found  in Nervous Tissue EXCITATORY Acetylcholine Aspartate Dopamine Histamine Norepinephrine Epinephrine ATP Glutamate 5­Hydroxytryptamine INHIBITORY 4­Aminobutyrate Glycine Taurine

  Neuron–Neuron Interaction Occurs through Synapses There are generally two mechanisms for neuron–neuron interaction: through electrical synapses or through chemical synapses. Electrical synapses permit the  more rapid transfer of signals from cell to cell. Chemical synapses allow for various levels of versatility in cell–cell communication. T. R. Elliot, in a paper published in  1904, was one of the first scientists to clearly express the idea that signaling between nerves could be chemical. Needless to say, considerably more information is now  known about this mode of neuron–neuron communication. Chemical synapses are of two types: those that bind directly to an ion channel and cause it to open or to  close, and those that bind to a receptor that releases a second messenger that reacts with the ion channel to cause it to open or to close. Primary emphasis here is on  chemical synapses. Chemical neurotransmitters fit the following criteria: they are found in the presynaptic axon terminal; enzymes necessary for their syntheses are present in the  presynaptic neuron; stimulation under physiological conditions results in their release; mechanisms exist (within the synaptic junction) for rapid termination of their  action; and their direct application to the postsynaptic terminal mimics the action of nervous stimulation. A sixth criterion, as a corollary of the five criteria listed above,  is that drugs that modify the metabolism of the neurotransmitter should have predictable physiological effects in vivo, assuming that the drug is transported to the site  where the neurotransmitter acts. Chemical neurotransmitters may be excitatory or inhibitory. Excitatory neurotransmitters include acetylcholine and the catecholamines. Inhibitory  neurotransmitters include g­aminobutyric acid (also referred to as GABA or 4­aminobutyric acid), glycine, and taurine (Table 22.1). The two major inhibitory neurotransmitters in the central nervous system are glycine and GABA. Glycine acts predominantly in the spinal cord and the brain stem, and  g­aminobutyric acid (GABA) acts predominantly in all other parts of the brain. Strychnine (Figure 22.3), a highly poisonous alkaloid obtained from Nux vomica and  related plants of the genus Strychnos, binds to glycine receptors of the CNS. It has been used in very small amounts as a CNS stimulant. Can you propose how it  works? The GABA receptor also reacts with a variety of pharmacologically significant agents such as benzodiazepines (Figure 22.4) and barbiturates. As with  strychnine and glycine, there is little structural similarity between GABA and benzodiazepines.

Figure 22.3  Structures of  glycine and strychnine.

The genes for the acetylcholine receptor, which also binds nicotinic acid, the glycine receptor, and the GABA receptor have been cloned and their amino acid  sequences inferred. There is a relatively high degree of homology in their primary amino acid sequences.

Figure 22.4  Structures of GABA and  diazepam.

A model of one­half of the GABA receptor is shown in Figure 22.5. This receptor has an 

 

  

2

 composition. The polypeptides are synthesized with "signal

2

Page 924

peptides" that direct their transport to the membrane. The a  subunit has 456 amino acid residues and the b  subunit has 474. The signal peptides are cleaved, leaving a   and b  subunits of 429 and 449 amino acid residues, respectively. Interestingly, the pharmaceutical agents bind to the a  subunit, whereas GABA, the natural inhibitory  neurotransmitter, binds to the b  subunit. The protrusion of an extended length of the amino­terminal end of each polypeptide to the extracellular side of the membrane  suggests that the residues to which the channel regulators bind are at the N terminal. A smaller C­terminal segment is also on the extracellular side of the membrane.  The four subunits of the receptor form a channel through which small negative ions (Cl–) can flow, depending on what is bound to the receptor end of the molecule. All neurotransmitters are made and stored in presynaptic neurons. They are released after stimulation of the neuron, traverse the synapse, and bind to a specific  receptor on the postsynaptic junction to elicit a response in the next cell. If the neurotransmitter is an excitatory one, it causes depolarization of the membrane as  described above. If it is an inhibitory neurotransmitter, it binds to a channel­linked receptor and causes a conformation change that opens the pores and permits small  negatively charged ions, specifically Cl–, to enter. The net effect of this is to increase the chloride conductance of the postsynaptic membrane, making it more difficult  for it to become depolarized—that is, effectively causing hyperpolarization. Synthesis, Storage, and Release of Neurotransmitters Nonpeptide neurotransmitters may be synthesized in almost any part of the neuron, in the cytoplasm near the nucleus, or in the axon. Most nonpeptide  neurotransmitters are amino acids, derivatives of amino acids, or other intermediary metabolites. Synthesis and degradation of many of them have been discussed  elsewhere, but some aspects of their metabolism relative to nerve transmission will be discussed later in this chapter. Neurotransmitters travel rapidly across the synaptic junction (which is about 20 nm across), bind to receptors on the postsynaptic side, induce

Figure 22.5  Schematic model of one­half of the GABA receptor  embedded in the cell membrane.  The complete receptor has an  2 2 structure and forms an  ion channel. The site labeled P is a serine residue that may  be phosp­ horylated by a cAMP­dependent protein kinase.  Redrawn from Schofield, P. R., Darlison, M. G.,  Fujita, N. et al. Nature 328:221, 1987.

 

  

Page 925

Figure 22.6  Schematic drawing of the relative arrangement of proteins of the  synaptic vesicle (SV).  Rab proteins are attached by isoprenyl groups and cysteine string  proteins by palmitoyl chains to SVs. The N and C termini of proteins  are marked by N and C, respectively. Phosphorylation  sites are indicated by P.  Redrawn from Sudhof, T. C. Nature 375:645, 1995.

conformational changes in receptors and/or that membrane, and start the process of electrical impulse propagation in the postsynaptic neuron. Storage and release of  neurotransmitters are intricate processes, but many details of the mechanism of these processes have begun to unfold. It has been shown by conventional techniques  that some neurons contain more than one chemical type of neurotransmitter. The significance of this observation is unclear. Release of neurotransmitter is a quantal  event; that is, a nerve impulse reaching the presynaptic terminal results in the release of transmitters from a fixed number of synaptic vesicles. Release of  neurotransmitters involves attachment of the synaptic vesicle to the membrane and exocytosis of their content into the synaptic cleft. Storage of neurotransmitters occurs in large or small vesicles in the presynaptic terminal. Small vesicles are the predominant type and exist in two pools: free and  attached to cytoskeletal proteins, mainly actin. Small vesicles contain only "classical" small molecule type transmitters, whereas large vesicles may contain "classical"  small molecule neurotransmitters and neuropeptides. Some may also contain enzymes for synthesis of norepinephrine from dopamine. A schematic diagram of a small  synaptic vesicle is shown in Figure 22.6. The genes for many of the proteins attached to the synaptic vesicle have been cloned and significant amounts of information  about their functions are known. Table 22.2 contains a list of some of those proteins. Some of their properties are briefly described. Figure 22.7 shows schematically  how some of them may be arranged on the synaptic vesicle and how they may interact with the plasma membrane of the presynaptic neuron. 1. Synapsin exists as a family of proteins encoded by two genes. The proteins differ primarily in the C­terminal end (Figure 22.8). They constitute about 9% of the  total protein of the synaptic vesicle membrane. All can be phosphorylated near their N termini by either cAMP­dependent protein kinase and/or calcium– calmodulin (CaM) kinase I, which is considered to be the physiologically important one relative to nerve transmission. Synapsins Ia and Ib can also be  phosphorylated by CaM kinase II near their C termini, a region that is missing in synapsin IIa and IIb. Synapsin has a major role in determining whether the synaptic vesicles are in the free pool and available for binding to the presynaptic membrane. Nerve stimulation  leads to the entry of Ca2+ into the presynaptic vesicle (see Clin. Corr. 22.1). CaM kinase I (II also) is activated and phosphorylates synapsin. This either prevents  binding of synaptic vesicles to the cytoskeletal proteins or TABLE 22.2 List of Synaptic Vesicle Proteins Synapsin

Ia

 

Ib

 

IIa

 

IIb Synaptophysin Synaptotagmin (p65) Syntaxin (p35) Synaptobrevin/VAMP Rab3 and rabphilin SV­2 Vacuolar proton pump

   

  

Page 926

Figure 22.7  Schematic diagram showing how some of the synaptic vesicle proteins  may interact with plasma membrane proteins to effect exocytosis.  Redrawn from Bennett, M. K., and Scheller, R. H.  Proc. Natl. Acad. Sci. USA, 90:2559, 1993.

releases them from those binding sites. The result is an increase in the free pool of synaptic vesicles. It has also been observed that calcium–calmodulin itself can bind  synapsin and competitively block its interaction with actin. Calcium–calmodulin therefore regulates the number of free synaptic vesicles in the two pools by two  mechanisms. 2. Synaptophysin is an integral membrane protein of synaptic vesicles that is structurally similar to gap junction proteins. It may be involved in the formation of a  channel from the synaptic vesicle through the presynaptic membrane to permit the passage of neurotransmitters into the synaptic cleft. 3. Synaptotagmin is also an integral membrane protein of synaptic vesicles that interacts in a Ca2+­dependent manner with specific proteins localized

Figure 22.8  Structural arrangement of the synapsin family of proteins.  Redrawn from Chilcote, T. J., Siow, Y. L., Schaeffer, E., et al.  J. Neurochem. 63:1568, 1994.

 

  

Page 927

CLINICAL CORRELATION 22.1  Lambert–Eaton Myasthenic Syndrome Lambert–Eaton myasthenic syndrome (LEMS) is an autoimmune disease in which the  body raises antibodies against voltage­gated calcium channels (VGCC) located on  presynaptic nerve termini. Upon depolarization of presynaptic neurons, calcium channels  at presynaptic nerve termini open, permitting the influx of calcium ions. This increase in  calcium ion concentration initiates events of the synapsin cycle and leads to release of  neurotransmitters into synaptic junctions. When autoantibodies against VGCC react with  neurons at neuromuscular junctions, calcium ions cannot enter and the amount of  acetylcholine released into synaptic junctions is diminished. Since action potentials to  muscles may not be induced, the effect mimics that of classic myasthenia gravis. LEMS has been observed in conjunction with other conditions such as small cell lung  cancer. Some patients have shown a neurological disorder manifesting itself as subacute  cerebellar degeneration (SCD). Plasma exchange (removal of antibodies) and  immunosuppressive treatments have been effective for LEMS, but the latter treatment is  less effective on SCD. Diagnostic assays for LEMS depend on the detection of antibodies in patients' sera  against VGCC. There are at least four subtypes of VGCC: T, L, N, and P. It has been  found that the P subtype may be the one responsible for initiating neurotransmitter release  at the neuromuscular junction in mammals. A peptide toxin produced by a cone snail  (Conus magnus) binds to P­type VGCC in cerebella extracts. This small peptide has  been labeled with 125I, bound to VGCC in cerebella extracts, and the radiolabeled  complex was precipitated by sera of patients who have been clinically and  electrophysiologically defined as LEMS positive. This assay may prove useful not only in  detecting LEMS but also in providing a means of finding out more about the antigenicity  of the area(s) on the VGCCs to which antibodies are raised. Goldstein, J. M., Waxman, S. G., Vollmer, T. L., et al. Subacute cerebellar degeneration  and Lambert–Eaton myasthenic syndrome associated with antibodies to voltage­gated  calcium channels: differential effect of immunosuppressive therapy on central and  peripheral defects. J. Neurol. Neurosurg. Psychiatry 57:1138, 1994; and Motomura,  M., Johnston, I., Lang, B., et al. An improved diagnostic assay for Lambert–Eaton  myasthenic syndrome. J. Neurol. Neurosurg. Psycbiatry 58:85, 1995.   on the presynaptic plasma membrane. It is probably involved in the process of docking of synaptic vesicles to the membrane. 4. Syntaxin is an integral membrane protein of the plasma membrane of the presynaptic neuron. Syntaxin binds synaptotagmin and mediates its interaction with Ca2+  channels at the site of release of the neurotransmitters. It also appears to have a role in exocytosis. 5. Synaptobrevin/VAMP (or vesicle­associated membrane protein) exists as a family of two small proteins of 18 and 17 kDa. They are anchored in the cytoplasmic  side of the synaptic vesicle membrane through a single C­terminal domain and appear to be involved in vesicle transport and/or exocytosis. VAMPs appear to be  involved in the release of synaptic vesicles from the plasma membrane of the presynaptic neuron. Tetanus and botulinum toxins bind VAMPs, causing slow and  irreversible inhibition of transmitter release. 6. Rab3 is one among a large rab family of GTP­binding proteins. Rab3 is specific for synaptic vesicles and is involved in the docking and fusion process of  exocytosis. Rab3 is anchored to the membrane through a polyprenyl side chain near its C­terminal end. Elimination by genetic engineering of the polyprenyl side  chain binding site did not alter its function in vitro, but it is not clear whether this is also true in vivo. 7. SV­2 is a large glycoprotein with 12 transmembrane domains. No function has yet been assigned to it. 8. Vacuolar proton pump is an ATPase found in the vesicle membrane that is responsible for the transport of neurotransmitters into the synaptic vesicle. Termination of Signals at Synaptic Junctions Neurotransmitter action may be terminated by metabolism, reuptake, and/or diffusion into other cell types. Neurotransmitters responsible for fast responses are  generally inactivated by one or both of the first two mechanisms. The following sections will outline some biochemical pathways involved in the synthesis and the  degradation of representative fast­acting neurotransmitters—

 

  

Page 928

specifically, acetylcholine, catecholamines, 5­hydroxytryptamine, and 4­aminobutyrate (GABA). Acetylcholine Reactions involving acetylcholine at the synapse are summarized in Figure 22.9. Acetylcholine is synthesized by the condensation of choline and acetyl CoA in a  reaction catalyzed by choline acetyltransferase found in the cytosol of the neuron. The reaction is

Choline is derived mainly from the diet; however, some may come from reabsorption from the synaptic junction or from other metabolic sources (see p. 460). The  major source of acetyl CoA is the decarboxylation of pyruvate by the pyruvate dehydrogenase complex in mitochondria. Since choline acetyltransferase is present  in the cytosol, acetyl CoA must get into the cytosol for the reaction to occur. The same mechanism discussed previously (see p. 371) for getting acetyl CoA across the  inner mitochondrial membrane (as citrate) operates in presynaptic neurons. Acetylcholine is released and reacts with the nicotinic–acetylcholine receptor located in the postsynaptic membrane (see Clin. Corr. 22.2). The action of  acetylcholine at the postsynaptic membrane is terminated by the action of the enzyme acetylcholinesterase, which hydrolyzes the acetylcholine to acetate and choline:

Choline is taken up by the presynaptic membrane and reutilized for synthesis of more acetylcholine. Acetate probably gets reabsorbed into the blood and is  metabolized by tissues other than nervous tissue. An X­ray crystallographic structure of acetylcholinesterase is shown in Figure 22.10. Its mechanism of action is similar to that of serine proteases (see p. 97). It too  has a catalytic triad, but the amino acids in that triad, from N to C

Figure 22.9  Summary of the reactions of acetylcholine at the synapse.  AcCoA, acetyl coenzyme A.

 

  

Page 929

CLINICAL CORRELATION 22.2  Myasthenia Gravis: A Neuromuscular Disorder Myasthenia gravis is an acquired autoimmune disease characterized by muscle weakness  due to decreased neuromuscular signal transmission. The neurotransmitter involved is  acetylcholine. The sera of more than 90% of patients with myasthenia gravis have  antibodies to the nicotinic–acetylcholine receptor (AChR) located on the postsynaptic  membrane of the neuromuscular junction. Antibodies against the AChR interact with it  and inhibit its function, either its ability to bind acetylcholine or its ability to undergo  conformation changes necessary to effect ion transport. Evidence in support of  myasthenia gravis as an autoimmune disease affecting the AChR is the finding that the  number of AChRs is reduced in patients with the disease, and experimental models of  myasthenia gravis have been generated by either immunizing animals with the AChR or by  injecting them with antibodies against it. It is not known what events trigger the onset of the disease. There are a number of  environmental antigens that have epitopes resembling those on the AChR. A rat  monoclonal antibody of the IgM type prepared against AChRs reacts with two proteins  obtained from the intestinal bacterium Escherichia coli. Both of the proteins are  membrane proteins of 38 and 55 kDa, the smaller of which is located in the outer  membrane. This does not suggest that exposure to E. coli proteins is likely to trigger the  disease. The sera of both normal individuals and myasthenia gravis patients have  antibodies against a large number of E. coli proteins. Some environmental antigens from  other sources also react with antibodies against AChRs. The thymus gland, which is involved in antibody production, is also implicated in this  disease. Antibodies have been found in thymus glands of myasthenia gravis patients that  react with AChRs and with environmental antigens. The relationship between  environmental antigens, thymus antibodies against AChRs, and onset of myasthenia gravis  is unclear. Myasthenia gravis patients may receive one or a combination of several therapies.  Pyridostigmine bromide, a reversible inhibitor of acetylcholine esterase (AChE) that does  not cross the blood–brain barrier, has been used. The inhibition of AChE within the  synapse by drugs of this type increases the half­time for acetylcholine hydrolysis. This  leads to an increase in the concentration of acetylcholine, stimulation of more AChR, and  increased signal transmission. Other treatments include use of immunosuppressant drugs,  steroids, and surgical removal of the thymus gland to decrease the rate of production of  antibodies. Future treatment may include the use of anti­idiotype antibodies to the AChR  antibodies, and/or the use of small nonantigenic peptides that compete with AChR  epitopes for binding to the AChR antibodies. Stefansson, K., Dieperink, M. E., Richman, D. P., Gomez, C. M., and Marton, L. S. N.  Engl. J. Med. 312:221, 1985; Drachman, D. B. (Ed.). Myasthenia gravis: biology and  treatment. Ann. N.Y. Acad. Sci. 505:1, 1987; and Steinman, L., and Mantegazza, R.  FASEB J. 4:2726, 1990.   termini, are in reverse order to those of the serine proteases, and glutamate instead of aspartate is involved. Catecholamines The catecholamine neurotransmitters are dopamine (3,4­dihydroxyphenylethylamine), norepinephrine, and epinephrine (Figure 22.11). Their biosynthesis has  been discussed (see p. 466). The action of catecholamine neurotransmitters is terminated by reuptake into the presynaptic neuron by specific transporter proteins. Cocaine, for example, binds to  the dopamine transporter and blocks its reuptake. Dopamine remains within the synapse for a prolonged period of time and continues to stimulate the receptors of  the postsynaptic neuron. Once inside the neuron,

Figure 22.10  Space­filling stereo view of acetylcholinesterase looking down into the active site.  Aromatic residues are in green, Ser­200 is red, Glu­199 is cyan, and other residues  are gray.  Reproduced with permission from Sussman, J. L., Harel,  M., Frolow, F., et al. Science 253:872, 1991. Copyright 1991 American  Association for the Advancement of Science. Photograph generously supplied by Dr.  J. L. Sussman.

 

  

Page 930

these neurotransmitters may be either repackaged into synaptic vesicles or metabolized. The two enzymes primarily involved in their metabolism are catechol­O­ methyltransferase and monoamine oxidase. The metabolic reactions are shown in Figure 22.12. Catechol­O­methyltransferase catalyzes the transfer of a methyl  group from S­adenosylmethionine to one of the phenolic OH groups. Monoamine oxidase catalyzes the oxidative deamination of these amines to aldehydes and  ammonium ions. Monoamine oxidase can use them as substrates whether or not they have been altered by the methyltransferase. The end product of dopamine  metabolism is homovanillic acid, and that of epinephrine and norepinephrine is 3­methoxy­4­hydroxymandelic acid.

Figure 22.11  Catecholamine  neurotransmitters.

5­Hydroxytryptamine (Serotonin) Serotonin, 5­hydroxytryptamine, is derived from tryptophan (see p. 476). Like dopamine, the action of serotonin is terminated by its reuptake into the presynaptic  neuron by a specific transporter. Some types of depression are associated with low brain levels of serotonin. The action of some antidepressants such as Paxil  (paroxetine hydrochloride), Prozac (fluoxetine hydrochloride), and Zoloft (sertraline hydrochloride) is linked to their ability to inhibit

Figure 22.12  Pathways of catecholamine degradation.  COMT, catechol­O­methyltransferase (requires S­adenosylmethionine);  MAO, monoamine oxidase; Ox, oxidation; Red, reduction. The end  product of epinephrine and norepinephrine metabolism is  3­methoxy­4­hydroxymandelic acid.

 

  

Page 931

serotonin reuptake. Once inside the presynaptic neuron, serotonin may be either repackaged in synaptic vesicles or metabolized. The primary route for its  degradation is oxidative deamination to the corresponding acetaldehyde catalyzed by the enzyme monoamine oxidase (Figure 22.13). The aldehyde is further  oxidized to 5­hydroxyindole­3­acetate by an aldehyde dehydrogenase.

Figure 22.13  Degradation of  5­hydroxytryptamine (serotonin).

4­Aminobutyrate (g ­Aminobutyrate)

g ­Aminobutyrate (GABA), an inhibitory neurotransmitter, is synthesized and degraded through a series of reactions commonly known as the GABA shunt. In brain  tissue, it appears that GABA and glutamate, an excitatory neurotransmitter, may share some common routes of metabolism in astrocytes (Figure 22.14). Both are  taken up by astrocytes and converted to glutamine, which is then transported back into presynaptic neurons. In excitatory neurons, glutamine is converted to glutamate  and repackaged in synaptic vesicles. In inhibitory neurons, glutamine is converted to glutamate and then to GABA, which is repackaged in synaptic vesicles. It has been suggested that brain levels of GABA in some epileptic patients may be low. Valproic acid (2­propylpentanoic acid) apparently increases brain levels of  GABA. The mechanism by which it does so is not clear. Valproic acid is metabolized primarily in the liver by glucuronidation and urinary excretion of the glucuronides,  or by mitochondrial b ­oxidation and microsomal oxidation. Neuropeptides Are Derived from Precursor Proteins Peptide neurotransmitters are generally synthesized as larger proteins and are cleaved by proteolysis to produce the neuropeptide molecules. Their synthesis  requires the same biochemical machinery as does any protein synthesis and takes place in the cell body, not the axon. They travel down the axon to the presynaptic  region by one of two generic mechanisms: fast axonal transport at a rate of about 400 millimeters per day and slow axonal transport at a rate of 1–5 millimeters per  day. Since axons may vary in length from 1 millimeter to 1 meter, theoretically the total transit time could vary from 150 milliseconds to 200 days. It is highly unlikely  that the latter transit time occurs under normal

Figure 22.14  Involvement of the astrocytes in the metabolism of GABA and glutamate.

 

  

Page 932 TABLE 22.3 Peptides Found in Brain Tissuea Peptide

Structure

b­endorphin

Y G G F M T S E K S Q T P L V T

 

L F K N A I I K N A Y K K G E

Met­enkephalin

Y G G F M

Leu­enkephalin

Y G G F L

Somatostatin

  Luteinizing hormone­  releasing hormone

p­E H W S Y G L R P G­NH2

Thyrotropin­releasing  hormone

p­E H P­NH2

Substance P

R P K P E E F F G L M­NH2

Neurotensin

p­E L Y E N K P R R P Y I L

Angiotensin I

D R V Y I H P F H L

Angiotensin II

D R V Y I H P F

Vasoactive

H S D A V F T D N Y T R L R

intestinal peptide

K E M A V K K Y L N S I L N­NH2

a Peptides with p preceding the structure indicate that the N terminal is 

pyroglutamate. Those with NH2 at the end indicate that the C terminal  is an amide.

  physiological conditions, and the upper limit is probably hours rather than days. Recent experiments suggest that the faster transit times prevail. Neuropeptides mediate sensory and emotional responses such as those associated with hunger, thirst, sex, pleasure, and pain. Included in this category are  enkephalins, endorphins, and substance P. Substance P is an excitatory neurotransmitter that has a role in pain transmission, whereas endorphins have roles in  eliminating the sensation of pain. Some of the peptides found in brain tissue are shown in Table 22.3. Note that Met­enkephalin is derived from the N­terminal region  of b ­endorphin. The N­terminal or both the N­ and C­terminal amino acids of many of the neuropeptide transmitters are modified. For a further discussion of these  peptides, see Chapter 20. 22.3—  The Eye:  Metabolism and Vision The eye, our window to the outside world, allows us to view the beauties of nature, the beauties of life, and, vide this textbook, the beauties of biochemistry. What are  the features of this organ that permit this view? A view through any window, through any camera lens, is clearest when unobstructed. The eye has evolved in such a  way that a similar objective has been achieved. It is composed of live tissues that require continuous nourishment for survival. Energy and metabolites for growth and  maintenance are derived from nutrients by conventional biochemical mechanisms, but the structures responsible for these processes are arranged and distributed such  that they do not interfere with the visual process. Also, the brain has devised an enormously efficient filtering system that makes invisible objects within the eye that may  appear to lead to visual distortion. In addition, different tissues use specific metabolic pathways to accommodate their unique needs. A schematic diagram of a cross  section of the eye is shown in Figure 22.15. Light entering the eye passes progressively through the cornea; the anterior chamber, which consists of the aqueous humor; the lens; the vitreous body, which  consists of the vitreous humor; and finally focuses on the retina, which contains the visual sensing apparatus. The exterior of the cornea is bathed by

 

  

Page 933

Figure 22.15  Schematic of a horizontal section of the left eye.

tears, while the interior is bathed by the aqueous humor, an iso­osmotic fluid containing salts, albumin, globulin, glucose, and other constituents. The aqueous humor  brings nutrients to the cornea and to the lens, and it removes end products of metabolism from them. The vitreous humor is a collagenous or gelatinous mass that helps  maintain the shape of the eye while allowing it to remain somewhat pliable. The Cornea Derives ATP from Aerobic Metabolism The eye is an extension of the nervous system, and like other tissues of the central nervous system, the major metabolic fuel is glucose. The cornea, which is not a  homogeneous tissue, obtains a relatively large percentage of its ATP from aerobic metabolism. About 30% of glucose used by the cornea is metabolized by glycolysis  and about 65% by the hexose monophosphate pathway. On a relative weight basis, the cornea has the highest activity of the hexose monophosphate pathway of  any other mammalian tissue. It also has a high activity of glutathione reductase, an activity that requires NADPH, a product of the hexose monophosphate pathway.  Corneal epithelium is permeable to atmospheric oxygen, that is necessary for various oxidative reactions. The reactions of oxygen can result in the formation of various  active oxygen species that are harmful to the tissues, perhaps in some cases by oxidizing protein sulfhydryl groups to disulfides. Reduced glutathione (GSH) is used  to reduce those disulfide bonds back to their original native states while GSH itself is converted to oxidized glutathione (GSSG). Furthermore, oxidized glutathione  (GSSG) may also be formed by auto­oxidation. Glutathione reductase uses NADPH to reduce GSSG to 2GSH.

The activities of the hexose monophosphate pathway and the glutathione reductase maintain this tissue in an appropriately reduced state by effectively neutralizing the  active oxygen species. Lens Consists Mostly of Water and Protein The lens is bathed on one side by the aqueous humor and supported on the other side by the vitreous humor. The lens has no blood supply, but it is metabolically  active. It gets nutrients from the aqueous humor and eliminates waste into the aqueous humor. The lens is mostly water and proteins. The majority of the proteins are  the a ­, b ­, and g ­crystallins. There are also albuminoids, enzymes, and membrane proteins that are synthesized in an epi­

 

  

Page 934 TABLE 22.4 Eye Lens Crystallins and Their Relationships with Other Proteins Crystallin

a

  b

Distribution

Small heat shock proteins (aB)

 

[Schistosoma mansoni antigen]

All vertebrates

[Myxococcus xanthus protein S]

g Taxon­specific enzyme  crystallins

[Related] or Identical

All vertebrates

(embryonic g not in birds)

[Physarum polycephalum spherulin 3a]

 

 

Most birds, reptiles

Argininosuccinate lyase ( 2)

Crocodiles, some birds

Lactate dehydrogenase B

Guinea pig, camel, llama

NADPH: quinone oxidoreductase

Elephant shrew

Aldehyde dehydrogenase I

Source: Wistow, G. TIBS 18:301, 1993.

  thelial layer around the edge of the lens. Some other types of proteins that are found in lens, including the lens of species other than vertebrates, are shown in Table  22.4. This shows that lens proteins may have different genetic origins and functions in other tissues. The most important physical requirement of these proteins is that  they maintain a clear crystalline state. The center area of the lens, the core, consists of the lens cells that were present at birth. The lens grows from the periphery  (Figure 22.16). The human lens increases in weight and thickness with age and becomes less elastic. This is accompanied by a loss of near vision (Table 22.5); a  condition referred to as presbyopia. On average the lens may increase threefold in size and approximately 1 1/2­fold in thickness from birth to about age 80. Lens proteins must be maintained in a native unaggregated state. They are sensitive to various insults such as changes in oxidation–reduction state, osmolarity,  excessively increased concentrations of metabolites, and physical insults such as UV irradiation. Reactions that help maintain structural integrity of the lens are the Na+,  K+–ATPase for osmotic balance, glutathione reductase for redox state balance, and protein synthesis for growth and maintenance. Energy for these processes comes  from the metabolism of glucose. About 85% of the glucose metabolized by the lens is by glycolysis, 10% by the hexose monophosphate pathway, and 3% by the  tricarboxylic acid cycle, presumably by the cells located at the periphery. Cataract is the only known disease of the lens. Cataracts are opacities of lenses brought about by a loss of osmolarity and a change in solubility of some

Figure 22.16  Schematic representation of a meridional section of a mammalian lens.

 

  

Page 935

of the proteins, resulting in regions of high light scatter. Cataracts affect about 1 million people per year in the United States, and there are no known cures or  preventative measures. The remedy is lens replacement, a very common operation in the United States. There are basically two types of cataracts: senile cataracts  and diabetic cataracts. Both are the result of changes in the solubility and aggregation state of the lens crystallins. In senile cataracts, changes in the architectural  arrangement of the lens crystallins are age­related and due to such changes as breakdown of the protein molecules starting at the C­terminal ends, deamidation, and  racemization of aspartyl residues. Diabetic cataracts result from loss in osmolarity of the lens due to the activity of aldose reductase and polyol (aldose)  dehydrogenase of the polyol metabolic pathway. When the glucose concentration in the lens is high, aldose reductase reduces some of it to sorbitol (Figure 22.17),  which may be converted to fructose by polyol dehydrogenase. In human lens, the ratio of activities of these two enzymes favors sorbitol accumulation, especially since  sorbitol is not used otherwise, and it diffuses out of the lens rather slowly. Accumulation of sorbitol in the lens increases osmolarity of the lens, affects the structural  organization of the crystalline proteins within the lens, and enhances the rate of protein aggregation and denaturation. The areas where this occurs will have increased  light scattering properties—which is the definition of cataracts. Normally, sorbitol formation is not a problem because the Km  of aldose reductase for glucose is about  200 mM and very little sorbitol would be formed. In diabetics, where the circulating concentration of glucose is high, activity of this enzyme can be significant. TABLE 22.5 Changes in Focal Distance with  Age Age

Focal Distance (in.)

10

2.8

20

4.4

35

9.8

45

26.2

70

240.0

Source: Adapted from Koretz, J. F., and  Handelman, G. H. Sci. Am., 92, July 1988.

  The Retina Derives ATP from Anaerobic Glycolysis The retina, like the lens, depends heavily on anaerobic glycolysis for ATP production. Unlike the lens, the retina is a vascular tissue, but there are essentially no blood  vessels in the area where visual acuity is greatest, the fovea centralis (see Clin. Corr. 22.3). Mitochondria are present in the retina, including in the rods and in the  cones. There are no mitochondria in the outer segments of the rods and cones where the visual pigments are located. NADH produced during glycolysis can be used to reduce pyruvate to lactate. The lactate dehydrogenase of the retina can use either NADH or NADPH, the

Figure 22.17  Metabolic interrelationships of lens metabolism.

 

  

Page 936

CLINICAL CORRELATION 22.3  Macula Degeneration: Other Causes of Vision Loss Many diseases of the eye affect vision, not all of which have clear, direct biochemical  origins. The most serious eye diseases are those that result in blindness. Glaucoma is the  most common and there is a direct causal relationship with diabetes, the biochemistry of  which is fairly well known. Glaucoma can be treated and blindness does not have to be a  result. Macula degeneration leads to blindness and there is no cure. The macula is a circular area  of the retina, the center of which is the fovea centralis, the area containing the greater  concentration of cones and the one of greatest visual acuity. Macula degeneration may be  among the leading causes of blindness in people over the age of 50. Macula degeneration  is of two types: dry and wet. The dry form develops gradually over time, whereas the wet  form develops rapidly and can lead to blindness within days. Macula degeneration occurs  when blood vessels rupture under the macula, leading to a loss of the nutrient supply and  a rapid loss of vision. Experimental procedures are in progress to surgically remove scar  tissue that develops and to transplant tissue from the rear of the eye to restore  nourishment to the photoreceptor cells. Rupture of blood vessels that obscure macula details and result in rapid onset of blindness  may be temporary in some cases. Six cases of sudden visual loss associated with sexual  activity have been reported that are not associated with a sexually transmitted disease.  Vision was lost in one eye apparently during, but most often reported a few days after  engaging in, ''highly stimulatory" sexual activity. Blindness was due to rupture of blood  vessels in the macula area. When patients did see an ophthalmologist, most were reluctant  to discuss what they were doing when sight loss was first observed. Four of the patients  recovered with restoration of vision upon reabsorption of blood. In one case, where  blood was trapped between the vitreous gel and the retinal surface directly in front of the  fovea, the hemorrhage cleared only slightly during the next month, but visual acuity did not  improve. The patient did not return for a follow­up examination, but there was no  indication that the condition was permanent. Since most of the persons affected by this  phenomenon were over the age of 39, it may be a worry more to professors than to  students. It also may give a new meaning to the phrase "love is blind." Friberg, T. R., Braunstein, R. A., and Bressler, N. M. Arch. Ophthalmol. 113:738,  1995.   latter being formed from the hexose monophosphate pathway. It is not clear whether lactate dehydrogenase of the retina plays any substantial role in mediating the  regulation of glucose metabolism through either of these pathways by its selective use of NADH or NADPH. Visual Transduction Involves Photochemical, Biochemical, and Electrical Events Figure 22.18 shows an electron micrograph and schematic of the retinal membrane. Light entering the eye through the lens passes the optic nerve fibers, the ganglion  neurons, the bipolar neurons, and the nuclei of the rods and cones before it reaches the outer segment of the rods and cones where the signal transduction process  begins. The pigmented epithelial layer of the eye, the choroid, lies behind the retina, absorbs the excess light, and prevents reflections back into the rods and cones  where it may cause distortion or blurring of the image (see Clin. Corr. 22.4). The eye may be compared with a video camera. The camera collects images, converts them into electrical pulses, records them on magnetic tape, and allows their  visualization by decoding the taped information. The eye focuses on an image by projecting that image onto the retina. A series of events begins, the first of which is  photochemical, followed by biochemical events that amplify the signal, and finally electrical impulses are sent to the brain where the image is reconstructed in "the  mind's eye." During this process, the initial event has been transformed from a physical event to a chemical reaction, through a series of biochemical reactions, to an  electrical event, to a conscious acknowledgment of the presence of an object in the environment outside the body. When photons of light enter the eye and are absorbed by photoreceptors in the outer segments of rods or cones, they cause isomerization of the visual pigment,  retinal, from the 11­cis form to the all­trans form. This isomerization causes a conformation change in the protein moiety of the complex and affects the resting  membrane potential of the cell, resulting in an electrical signal being

 

  

Page 937

Figure 22.18  Electron micrograph and schematic representation of cells of the human retina.  Tips of rods and cones are buried in the pigmented epithelium of the outermost  layer. Rods and cones form synaptic junctions with many bipolar neurons,  which in turn form synapses with cells in the ganglion layer that send axons  through the optic nerve to the brain. The synapse of a rod or cone with many  cells is important for the integration of information. HC, horizontal cells;  AC, amacrine cell; MC, Müller cell; BL, basal lamina.  Reprinted with permission from Kessel, R. G., and Kardon, R. H., Tissues  and Organs: A Text­Atlas of Scanning Electron Microscopy.  New York: W. H. Freeman, 1979, p. 87.

transmitted by way of the optic nerve to the brain. These processes will be discussed later in more detail. Photoreceptor Cells Are Rods and Cones The photoreceptor cells of the eye are the rods and the cones (Figure 22.18). Each type has flattened disks that contain a photoreceptor pigment. This pigment is  rhodopsin in the rod cells, and red, green, or blue pigment in the cone cells. Rhodopsin is a transmembrane protein to which is bound a prosthetic group, 11­cis­ retinal. Rhodopsin minus its prosthetic group is opsin. The three proteins that form the red, green, and blue pigments of cone cells are different from each other and  from opsin. Rhodopsin, an approximately 40­kDa protein, contains seven transmembrane a ­helices. The 11­cis­retinal is attached through a protonated Schiff base to the  ­ amino group of lysine­296 on the seventh helix. Lysine­296 lies about midway between the two faces of the membrane (Figure 22.19a). A 9­Å resolution three­ dimensional (3­D) model for rhodopsin, obtained by cryomicroscopy, shows that most of the helices are perpendicular to the surface of the membrane

 

  

Page 938

CLINICAL CORRELATION 22.4  Niemann–Pick Disease and Retinitis Pigmentosa There are central nervous system disorders associated with the Niemann–Pick group of  diseases that can become evident by ocular changes. Some of these are observed as  abnormal macula with gray discoloration and granular pigmentation or granule opacities  about the fovea. Acute type I Niemann–Pick disease, lipidosis with sphingomyelinase deficiency and  primary sphingomyelin storage, may show a cherry red spot in the retina in as many as  50% of patients. Macula halo syndrome applies to the crystalloid opacities seen in some  patients with subacute type I disease. They form a halo approximately one­half the disk  diameter at their outer edge and are scattered throughout the various layers of the retina.  They do not interfere with vision. In an 11­year­old girl who had type II disease, more extensive ocular involvement was  observed. There was sphingomyelin storage in the keratocytes of the cornea, the lens, the  retinal ganglion cells, the pigmented epithelium, the corneal tract, and the fibrous  astrocytes of the optic nerve. Retinitis pigmentosa is a secondary effect of the abnormal biochemistry associated with  Niemann–Pick disease. Spence, M. W., and Callahan, J. W. In: C. R. Schriver, A. L. Beaudet, W. Sly, and D.  Volle (Eds.), The Metabolic Basis of Inherited Disease, New York: McGraw­Hill,  1989, pp. 1656–1676.   (Figure 22.19b). Some, however, are distorted from this perpendicular arrangement. It is not known whether the orientation of those distorted helices is associated  with binding of 11­cis­retinal since this low­resolution structure will not permit tracing of the carbon backbone structure of rhodopsin. See also Clin. Corr. 22.5. Reactions involved in the formation of 11­cis­retinal from b ­carotene and rhodopsin from opsin and 11­cis­retinal are shown in Figure 22.20. The 11­cis­retinal is  derived from vitamin A and/or b ­carotene of the diet. These are

Figure 22.19  Rhodopsin.  (a) A model of the structure of vertebrate rhodopsin.  (b) A 9­Å resolution 3­D model for rhodopsin obtained by cryomicroscopy.  (a) Redrawn from Stryer, L. Annu. Rev. Neurosci. 9:87, 1986 (based  on Dratz and Hargrave, 1983).  (b) Reproduced with permission from Unger, V. M. and Schertler,  G. F. X. Biophys. J. J. 68:1776, 1995. Photograph generously supplied by  Dr. G. F. X. Schertler.

 

  

Page 939

Figure 22.20  Formation of 11­cis­retinal and rhodopsin from b­carotene.

 

  

Page 940

CLINICAL CORRELATION 22.5  Retinitis Pigmentosa Resulting from a De Novo Mutation in the Gene Coding for  Peripherin A group of heterogeneous diseases of variable clinical and genetic origins have been  placed under the category of retinitis pigmentosa (RP). Several of these have origins in  abnormal lipid metabolism. Approximately 1.5 million people throughout the world are  affected by this disease. It is a slowly progressive condition associated with loss of night  and peripheral vision. It can be inherited through an autosomal dominant, recessive, or X­ linked mode. RP has been associated with mutations in the protein moiety of rhodopsin  and in a related protein, peripherin/RDS, both of which are integral membrane proteins.  Peripherin is a 344 amino acid residue protein located in the rim region of the disk  membrane. Structural models of these two proteins are shown in the figure below. Filled  circles and other notations in the figure mark residues or regions that have been  correlated with RP or other retinal degenerations. A case has been described where a de novo mutation in exon 1 of the gene coding for  peripherin resulted in the onset of RP. Using molecular biological techniques, Lam et al.  (1995) found the specific change in peripherin to be a C­to­T transition in the first  nucleotide of codon 46. This resulted in changing an arginine to a stop codon (R46X).  The pedigree of this family is shown in the figure on next page. Neither parent had the  mutation and genetic typing analysis (20 different short tandem repeat polymorphisms)  showed that the probability that the proband's parents are not his actual biological  parents is less than 1 in 10 billion. This establishes with near certainty that the mutation is  de novo.

Schematic representation of structural models for rhodopsin  (top) and peripherin/RDS (bottom).  The location of mutations in amino acid  residues that  segregate with RP or other retinal degenerations are shown as solid red circles.

  (Table continued on next page)

 

  

Page 941

Clin. Corr. 22.5 (continued)

Pedigree of family.  Males are squares, females are circles.  Solid squareindicates the proband.  A slash through a symbol indicates deceased.  From Lam et al. (1995).

This R46X mutation has been observed in another unrelated patient. These observations  demonstrate the importance of the use of DNA analysis to establish the genetic basis for  RP, especially considering that RP symptoms have been associated with a variety of  other diseases, such as those related to abnormal lipid metabolism. Shastry, B. S. Am. J. Med. Genet. 52:467, 1994; and Lam, B. L., Vandenburgh, K.,  Sheffield, V. C., and Stone, E. M. Am. J Ophthalmol. 119:65, 1995.   transported to specific sites in the body while attached to specific carrier proteins. Cleavage of b ­carotene yields two molecules of all­trans­retinol. There is an  enzyme in the pigmented epithelial cell layer of the retina that catalyzes the isomerization of all­trans­retinol to 11­cis­retinol. Oxidation of the 11­cis­retinol to 11­cis­ retinal and its binding to opsin occur in the rod outer segment. The absorption spectra of 11­cis­retinal and the four visual pigments are shown in Figure 22.21. There is a shift in the wavelength of maximum absorption of 11­cis­ retinal upon binding to opsin and the protein components of the other visual pigments. Absorption bands for the pigments are coincident with their light sensitivity.

Figure 22.21  Absorption spectra of  11­cis­retinal and the four visual pigments.  Absorbance is relative and was obtained for  pigments as difference spectra from  reconstituted recombinant apoproteins. The  spectrum for 11­cis­retinal (11­cR) is in the  absence of protein. B, blue pigment; Rh,  rhodopsin; G, green; R, red.  Adapted from Nathans, J. Cell, 78:357, 1994.

The magnitude of change in the electrical potential of photoreceptor cells following exposure to a light pulse is different in magnitude from that of neurons during  depolarization. The resting potential of rod cell membrane is approximately –30 mV instead of the –70 mV observed with neurons. Excitation of rod cells causes  hyperpolarization of the membrane, from about –30 mV to about –35 mV (Figure 22.22). It takes hundreds of milliseconds for the potential to reach its maximum  state of hyperpolarization. A number of biochemical events take place during this time interval and before the potential returns to its resting state. The initial events, absorption of photons of light and the subsequent isomerization of 11­cis­retinal, are rapid, requiring only picoseconds. Following this, a series of  changes take place in rhodopsin, leading to various short­lived conformational states (Figure 22.23), each of which has specific absorption characteristics. Finally,  rhodopsin dissociates, giving opsin and all­trans­retinal. At 37°C, activated rhodopsin has decayed in slightly more than 1 millisecond through several intermediates to metarhodopsin II. Metarhodopsin II has a half­life of  approximately 1 minute. It is the active rhodopsin species, R*, that is involved in the biochemical reactions of interest. Metarhodopsin II will have begun to form  within hundredths of microseconds of the initial event. All of the first series of reactions shown in Figure 22.23 take place in the disk of the

 

  

Page 942

rod outer segment. Upon dissociation of metarhodopsin into opsin and all­trans­retinal, the all­trans­retinal is enzymatically converted to all­trans­retinol by all­ trans­retinol dehydrogenase that is located in the rod outer segment. All­trans­retinol is transported (or diffuses) into the pigmented epithelium where a specific  isomerase converts it to 11­cis­retinol. The 11­cis­retinol is then transported (or diffuses) back into the rod outer segment and is reoxidized to 11­cis­retinal. Since the  all­trans­retinol dehydrogenase appears to have only about 6% as much activity with 11­cis­retinal, it appears that another enzyme may be responsible for its  oxidation. Once the aldehyde is formed, it can recombine with opsin to form rhodopsin. Rhodopsin is now in a state to begin the cycle again. The same events take  place in the cones with the three proteins of the red, green, and blue pigments.

Figure 22.22  Changes in the potential of a  rod cell membrane after a light pulse.  Redrawn from Darnell, J., Lodish, H., and  Baltimore, D. Molecular Cell  Biology. New York: Scientific American  Books, 1986, p. 763.

There are three interconnecting "mini" biochemical cycles involved in the conversion of light energy to nerve impulses (Figure 22.24). These cycles describe the  reactions of rhodopsin, transducin, and phosphodiesterase, respectively. The net result of their operation is to cause a hyperpolarization of the plasma membrane of  the rod (or cone) cells, that is, from –30 mV to approximately –35 mV. It is important to understand first what the biochemical mechanism is for maintaining the plasma  membrane at –30 mV.

Figure 22.23  Light activation of rhodopsin.

Rod cells of a fully dark­adapted human can detect a flash of light that emits as few as 50 photons. The rod is a specialized type of neuron in that the signal generated  does not depend on an all­or­none event. The signal may be graded in intensity, reflected by the extent that the millivolt potential of the plasma membrane changes  from its steady­state value of –30 mV. This steady­state potential is maintained at a more positive value because Na+ channels of the photoreceptor cells are  ligand gated and are maintained in a partially opened state. The ligand responsible for keeping some of the Na+ channels open is cyclic GMP (cGMP). cGMP binds  to them in a concentration­dependent, kinetically dynamic manner. Biochemical events that affect the concentration of cGMP within rod and cone cells also affect the  number of Na+ channels that are open and, hence, the membrane potential (Figure 22.24). Active rhodopsin (R*, namely, metarhodopsin II) forms a complex with transducin. Transducin is a classical type of G­protein and functions in a manner very similar  to that described on page 859 in relation to the action of some hormones. In the R*–transducin complex (R*–Ta,b ,g complex), transducin undergoes a conformation  change that facilitates an exchange of its bound GDP with GTP. When this occurs, the a  subunit (Ta) of the trimeric molecule dissociates from its b , g subunits. Ta  interacts with and activates phosphodiesterase (PDE), which hydrolyzes cGMP to 5¢­GMP, resulting in a decreased concentration of cGMP and a decrease in the  number of channels held open. The membrane potential becomes more negative, that is, hyperpolarized. The diagram of Figure 22.24 shows in cartoon form two such channels embedded in the plasma membrane, one of which has cGMP bound to it and is open. The  other does not have cGMP bound to it and it is closed. By this mechanism, the concentration of Na+ in the cell is directly linked to the concentration of cGMP and,  thus, also to the membrane potential. PDE in rod cells is a heterotetrameric protein consisting of one each a  and b  catalytic subunits and two g regulatory subunits. Ta–GTP forms a complex with the g  subunits of PDE, resulting in their dissociation from the catalytic subunits, freeing the catalytically active a ,b ­dimeric PDE subunit complex. Ta has GTPase activity.  Hydrolysis of bound GTP to GDP and inorganic phosphate (Pi) results in dissociation of Ta from the regulatory g subunits of PDE, permitting them to reassociate with  the catalytic subunits and to inhibit the PDE activity. The same reactions occur in cone cells, but the catalytic subunit of cone cell PDE is composed of two a  catalytic  subunits instead of a ,b  subunits as are present in rod cells. cGMP concentration is regulated by intracellular Ca2+ concentration. Calcium enters rod cells in the dark through sodium channels, increasing its concen­

 

  

Page 943 2+

tration to the 500­nM range. At these concentrations, activity of guanylate cyclase is low. When sodium channels are closed, Ca  entry is inhibited, but efflux  mediated by the sodium/calcium–potassium exchanger is unchanged (top complex of the plasma membrane in Figure 22.24). This results in a decrease in the  intracellular Ca2+ concentration, which in turn leads to activation of guanylate cyclase and increased production of cGMP from GTP. Both the resynthesis of cGMP and the hydrolysis of Ta–GTP play important roles in stopping the reactions of the visual cycle. The inactivation of activated rhodopsin,  R*, is also very important. Activated rhodopsin, R*, is phosphorylated by rhodopsin kinase in the presence of ATP (Figure 22.24). The R*–Pi has high binding affinity for the cytosolic protein,  arrestin. The arrestin–R*–Pi complex is no longer capable of interacting with transducin. The kinetics of arrestin binding to the activated­phosphorylated rhodopsin is  sufficiently rapid in vivo to stop the cascade of reactions. Rhodopsin is regenerated through another series of reactions and the cycle can be initiated again by photons of light. Figure 22.23 shows that the series of reactions  leading to the regeneration of rhodopsin includes the dissociation of all­trans­retinal from metarhodopsin. The regeneration of 11­cis­retinal from all­trans­retinal  occurs by reactions previously described and occurs before it is used again to form rhodopsin. Major proteins involved in the visual cycle are listed in Table 22.6.

Figure 22.24  Cascade of biochemical reactions involved in the visual cycle.  Redrawn from Farber, D. B. Invest. Ophthalmol. Vis. Sci. 36:263, 1995.

 

  

Page 944 TABLE 22.6 Major Proteins Involved in the Phototransduction Cascade Protein Rhodopsin

Relation to  Membrane Intrinsic

Molecular  Mass (kDa)

Concentration in  Cytoplasm (m M)

  39



Transducin (a + b + g)

Peripheral or  soluble

  80

500

Phosphodiesterase

Peripheral

200

150

Rhodopsin kinase

Soluble

  65

    5

Arrestin

Soluble

  48

500

Guanylate cyclase

Attached to  cytoskeleton

?

?

cGMP­activated  channel

Intrinsic

  66

?

  Color Vision Originates in the Cones Even though there are photographic artists, such as the late Ansel Adams, who make the world look beautiful in black and white, the intervention of colors in the  spectrum of life's pictures brings another degree of beauty to the wonders of nature and the beauty of life . . . even the ability to make a distinction between tissues  from histological staining. The ability of humans to distinguish colors resides within a relatively small portion of the visual system, the cones. The number of cones within  the human eye are few compared with the number of rods. Some animals like dogs have even fewer cones, and other animals, like birds, have many more. The general mechanism by which light stimulates cone cells is exactly the same as it is for rod cells. There are three types of cone cells, defined by the visual  pigments they contain, which are either blue, green, or red. Normally, only one type of visual pigment occurs in a single cell. The blue pigment has optimum absorbance  at 420 nm, the green pigment at 535 nm, and the red pigment at 565 nm (Figure 22.21). Each of these pigments has 11­cis­retinal as the prosthetic group, and, when  activated by light, the 11­cis­retinal isomerizes to all­trans­retinal in exactly the same manner as it does in the rod cells. Colors other than those of the visual pigments  are distinguished by graded stimulation of the different cones and comparative analysis by the brain. Color vision is trichromatic. The characteristic of color discrimination by cone cells is an inherent property of the proteins of the visual pigments to which the 11­cis­retinal is attached. The 11­cis­ retinal is attached to each of the proteins through a protonated Schiff base. The conjugated double­bond system of 11­cis­retinal influences the absorption spectrum of  the pigment (Figure 22.21). When 11­cis­retinal is bound to different visual proteins, amino acid residues in the local areas around the protonated base and the  conjugated  ­bond system influence the energy level and give different absorption spectra with absorption maxima that are different for the different color pigments. Genes for the color pigments have been cloned and their amino acid sequences inferred from the gene sequences. A structural comparison of the sequences of the  visual pigments is shown in Figure 22.25. Open circles represent amino acids that are the same, and closed circles represent amino acids that are different. A string of  closed circles at either end may represent an extension of the chain of one protein relative to the other. The red and green pigments show the greatest degree of  homology, about 96% identity, whereas the degree of homology between different pairs of the others is between 40% and 45%. Genes encoding the visual pigments have been mapped to specific chromosomes (see Clin. Corr. 22.6). The rhodopsin gene resides on the third chromo­

 

  

Page 945

Figure 22.25  Comparisons of the amino acid sequences of the human visual pigments.  Each red dot indicates an amino acid difference.  Adapted from Nathans, J. Annu. Rev. Neurosci. 10:163, 1987.

some, the gene encoding the blue pigment resides on the seventh chromosome, and the two genes for the red and green pigments reside on the X chromosome.  Abnormal color vision results from mutations in one or more of these genes (see Clin. Corr. 22.6). In spite of their great similarity, the red and green pigments are  distinctly different proteins. Individuals have been identified with inherited variations that affect one but not both pigments simultaneously. In addition, there may be  more than one gene for the green pigment, but it appears that only one is expressed. The person who developed the atomic theory of chemistry, John Dalton (1766–1844), was color blind. He thought his color blindness was due to the vitreous humor  being tinted blue, selectively absorbing longer wavelengths of light. He instructed that after his death his eyes be examined to determine whether his theory was correct.  An autopsy revealed that the vitreous humor was "perfectly pellucid," normal. Using DNA analysis on his preserved eyes obtained from the British Museum, it has now  been demonstrated that Dalton was missing the blue pigment. Thus, instead of having trichromatic vision, he was dichromatic with a vision type referred to as  deuteranopia. The type of color blindness of one who is missing the green pigment is protanopia. Other Physical and Chemical Differences between Rods and Cones The sensitivity and the response time of the rods are different from that of the cones. Absorption of a single photon by photoreceptors in rod cells generates a current  of approximately 1–3 picoamperes (1–3 × 10–12 pA), whereas the same event in the cones generates a current of approximately 10 femtoamperes (10 × 10–15 fA),  about 1/100th of the rod response. The response time of cone cells, however, is about four times faster than that of rod cells. Thus the cones

 

  

Page 946

CLINICAL CORRELATION 22.6  Abnormalities in Color Perception The chromosomal arrangement of genes for vision precludes inheritance of a single  defective gene from one parent that would render recipients sightless. Genes that code for  visual pigments occur on chromosomes that exist in pairs except in males where there is a  single X chromosome containing the genes for red and green pigments. In females, there  is a pair of X chromosomes and, therefore, color vision abnormalities in females are rare,  affecting only about 0.5% of the population. By contrast, about 8% of males have  abnormal color vision that affects red or green perception and, on rare occasions, both.  For the sake of simplicity, the proteins coded for by the different genes will be referred to  as pigments in spite of the fact that they become visual pigments only when they form  complexes with 11­cis­retinal. The gene that codes for the protein moiety of rhodopsin, the rod pigment, is located on  the third chromosome. Genes that code for the three pigment proteins of cone cells are  located on two different chromosomes. The gene for the blue pigment is on the seventh  chromosome. The genes for the red and green pigments are tightly linked and are on the  X chromosome, which normally contains one gene for the green pigment and from one to  three genes for the green pigment. In a given set of cones, only one of these gene types is  expressed, either the gene for the red pigment or one of the genes for the green pigment. Genetic mutations may cause structural abnormalities in the proteins that influence the  binding of retinal or the environment in which retinal resides. In addition, the gene for the  protein of a specific pigment may not be expressed. If 11­cis­retinal does not bind or one  of the proteins is not expressed, the individual will have dichromatic color vision and be  color blind for the color of the missing pigment. If the mutation changes the environment  around the 11­cis­retinal, shifting the absorption spectrum of the pigment, the individual  will have abnormal trichromatic color vision; that is, the degree of stimulation of one or  more of the three cone pigments will be abnormal. This will result in a different integration  of the signal and hence a different interpretation of color. Vollrath, D., Nathans, J., and Davis, R. W. Science 240:1669, 1988; and Nathans, J.  Cell 78:357, 1994.   are better suited for discerning rapidly changing events and the rods are better suited for low­light visual sensitivity. 22.4—  Muscle Contraction On the basis of an extensive evaluation of electron micrographs of skeletal muscle tissue, the sliding filament model for muscle contraction was proposed. This  simple but eloquent model has weathered the test of time. Genes for many of the proteins found in muscle tissue have been cloned, and the amino acid sequences of  the proteins they encode inferred from their cDNA sequences. Three­dimensional structures of some of these proteins have also been published. Although the detailed  picture of muscle contraction has not been completed, a clearer understanding of the process is emerging. In this section, some biochemical aspects of the mechanism  of muscle contraction will be discussed. Primary emphasis will be on skeletal muscle rather than cardiac and smooth muscles. Skeletal Muscle Contraction Follows an Electrical to Chemical to Mechanical Path The signal for skeletal muscle contraction begins with an electrical impulse from a nerve. This is followed by a chemical change occurring within the unit cell of the  muscle, and is followed by contraction, a mechanical process. Thus the signal transduction process goes from electrical to chemical to mechanical. Figure 22.26 is a schematic diagram showing the structural organization of skeletal muscle. Muscle consists of bundles of fibers (diagram c). Each bundle is called a  fasciculus (diagram b). The fibers are made up of myofibrils (diagram d), and each myofibril is a continuous series of muscle cells or units called sarcomeres. The  muscle cell is multinucleated and is no longer capable of division. Most muscle cells survive for the life of the animal, but they can be replaced when lost or lengthened  by fusion of myoblast cells.

 

  

Page 947

Figure 22.26  Structural organization of skeletal muscle.  Redrawn from Bloom, W. D., and Fawcett, D. W. Textbook of  Histology, 10th ed. Philadelphia: Saunders, 1975.

A muscle cell is shown diagrammatically in Figure 22.27. Note that the myofibrils are surrounded by a membranous structure called the sarcoplasmic reticulum. At  discrete intervals along the fasciculi and connected to the terminal cisterna of the sarcoplasmic reticulum are transverse tubules. The transverse tubules are connected  to the external plasma membrane that surrounds the

 

  

Page 948

Figure 22.27  Schematic representation of a bundle of six myofibrils.  The lumen of the transverse tubules connects with the  extracellular medium and enters the fibers at the Z disk.  Reprinted with permission from Darnell, J., Lodish, H., and  Baltimore, D. Molecular Cell Biology. New York:  Scientific American Books, 1986, p. 827.

entire structure. The nuclei and the mitochondria lie just inside the plasma membrane. The single contractile unit, the sarcomere, extends from Z line to Z line (Figures 22.26d and 22.27). Bands seen in the sarcomere are due to the arrangement of  specific proteins (Figure 22.26e). Two types of fibers are apparent: long thick ones with protrusions on both ends lie near the center of the sarcomere, and long thin  ones are attached to the Z line. The I band (isotropic) extends for a short distance on both sides of the Z line. This region contains only thin filaments that are  attached to a protein band within the Z line. The H band is in the center of the sarcomere. There are no thin filaments within this region. In the middle of the H band,  there is a somewhat diffuse band due to the presence of other proteins that assist in cross­linking the fibers of the heavy filaments (Figure 22.26, pattern h). The A  band (anisotropic) is located between the inner edges of the I bands. When the muscle contracts, the H and I bands shorten, but the distance between the Z line and  the near edge of the H band remains constant. The distance between the innermost edges of the I bands on both ends of the sarcomere also remains constant. This  occurs because the length of the thin filaments and the thick filaments does not change during contraction. Contraction therefore results when these filaments ''slide"  past each other. TABLE 22.7 Molecular Weights of Skeletal  Muscle Contractile Proteins Myosin

500,000

Heavy chain

200,000

Light chain

20,000

Actin monomer (G­actin)

42,000

Tropomyosin

70,000

Troponin

76,000

Tn­C subunit

18,000

Tn­I subunit

23,000

Tn­T subunit

37,000

a­Actinin

200,000

C­protein

150,000

b­Actinin

60,000

M­protein

100,000

  The contractile elements, sarcomeres, consist of many different proteins, eight of which are listed in Table 22.7. The two most abundant proteins in the sarcomere are  myosin and actin. About 60–70% of the muscle protein is myosin and about 20–25% is actin. The thick filament is mostly myosin and the thin filament is mostly actin.  Three other proteins listed in Table 22.7 are associated with thin filaments, and two are associated with thick filaments.

 

  

Page 949

Myosin Forms the Thick Filament of Muscle The schematic drawing of the myosin molecule in Figure 22.28a is a representation of the electron micrographs in Figure 22.28b. Myosin, a long molecule with two  globular heads on one end, is composed of two heavy chains of about 230 kDa each. Bound to each heavy chain in the vicinity of the head group is a dissimilar pair of  light chains, each of which is approximately 20 kDa. The

Figure 22.28  Myosin.  (a) Electron micrographs of the myosin molecule.  (b) Schematic drawing of a myosin molecule. Diagram shows the two  heavy chains and the two light chains of myosin. Also shown are the  approximate positions of cleavage by trypsin and papain.  Reprinted with permission from Alberts, B., Bray, D., Lewis, J., Raff, M.,  Roberts, K., and Watson, J. Molecular Biology of the Cell,  2nd ed. New York: Garland Publishing, 1983.

 

  

Page 950

light chains are "calmodulin­like" proteins that bind calcium. One from each myosin can be removed easily without affecting in vitro function. The carboxyl end of myosin is located in the tail section. The tail section of the two heavy chains are coiled around each other in an a ­helical arrangement (Figure  22.28a). Trypsin cleaves the tail section at about one­third of its length from the head to produce heavy meromyosin (the head group and a short tail) and light  meromyosin (the remainder of the tail section). Only light meromyosin has the ability to aggregate under physiological conditions, suggesting that aggregation is one of  its roles in heavy chain formation. The head section can be separated from the remainder of the tail section by treatment with papain. The myosin head group resulting  from this cleavage is referred to as subfragment 1 or S­1. Action of these proteases also demonstrates that the molecule has at least two hinge points in the vicinity of  the head–tail junction (Figure 22.28a). cDNAs for myosin from many different species and from different types of muscle have been cloned and amino acid sequences for these myosin molecules inferred.  Myosin has evolved very slowly, and there is a very high degree of homology among them, particularly within the head, or globular, region. There is somewhat less  sequence homology within the tail region, but functional homology exists to an extraordinarily high degree regardless of length, which ranges from about 86 to about  150 nm for different species. The myosin head group contains nearly one­half of the total number of amino acid residues of the entire molecule in mammals, and it  varies in the number of residues from only about 839 to about 850. Myosin forms a symmetrical tail­to­tail aggregate around the M line of the H zone in the sarcomere. Its tail sections are aligned in a parallel manner on both sides  of the M line with the head groups pointing towards the Z line. Each thick filament contains about 400 molecules of myosin. The C­protein (Table 22.7) is involved in  their assembly. The M­protein is also involved, presumably to hold the tail sections together as well as to anchor them to the M line of the H zone. The globular head section of myosin contains the ATPase activity that provides energy for contraction and the actin binding site. The S­1 fragment also contains the  binding sites for the essential light chain and the regulatory light chain. A space­filling model of the three­dimensional structure of the myosin S­1 fragment is  shown in Figure 22.29. The actin binding region is located at the lower right­hand corner and the cleft, visible in that region of the molecule, points toward the active  site region where ATP binds. The 25­, 50­, and 20­kDa domains of the heavy chain are colored green, red, and blue, respectively. The essential light chain (ELC) and  the regulatory light chain (RLC) are shown in yellow and magenta, respectively. The active (ATP binding) site is also an open cleft about 13 Å deep and 13 Å wide. It is separated from the actin binding site by approximately 35 Å.

Figure 22.29  Space­filling model of the amino acid residues in myosin  S­1 fragment.  The 25­, 50­, and 20­kDa domains of the heavy chain are  green, red, and blue, respectively. The essential and regulatory  light chains are yellow and magenta, respectively.  Reprinted with permission from Rayment, I., Rypniewski,  W. R., Schmidt­Bäse, K., Smith, R., et al. Science 261:50, 1993.  Copyright 1993 American Association for the Advancement  of Science. Photograph generously supplied by Dr. I. Rayment.

 

  

Page 951

Myosin binding to actin shows stereo specificity. The ELC and RLC are associated with a single long helix that connects the head region with the tail section. There is  room for flexibility, which requires only a low energy expenditure, between the ELC and the connecting single helix. The conformation of myosin that has ATP bound  to it has an affinity for actin that is 1/10,000 that of the conformation of myosin that does not have ATP bound to it! Thus the process of chemical energy transduction  to mechanical work depends on the primary event of protein conformation changes that occur upon binding of ATP, its hydrolysis, and product dissociation. Actin, Tropomyosin, and Troponin Are Thin Filament Proteins Actin is a major protein of the thin filament and makes up about 20–25% of muscle protein. It is synthesized as a 42­kDa globular protein. It has a better than 90%  conserved amino acid sequence among a variety of species. This is shown in Table 22.8 for skeletal muscle, smooth muscle, and cardiac muscle actin in three different  species of animals. Differences are observed at most in about seven different positions. In fact, the primary amino acid sequences of more than 30 different actin  isotypes, with the longest containing 375 amino acid residues, reveal that a maximum of only 32 residues in any of them had been substituted. A significant number of  them occurred at the N terminal, which may be predicted considering that all actin molecules are posttranscriptionally modified at the N terminal. The N­terminal  methionine is acetylated and removed, and the next amino acid is acetylated. The process may end at this stage or it may be repeated one or two additional times. In  all cases, the N­terminal amino acid will be acetylated. As first synthesized, actin is called G­actin for globular actin. The structure in Figure 22.30 shows that it is not strictly globular. Actin has two distinct domains of  approximately equal size that, historically, have been designated as large (left) and small (right) domains. Each of these domains consists of two subdomains. Both the  N­terminal and C­terminal amino acid residues are located within subdomain 1 of the small domain. The molecule has polarity, and when it aggregates to form F­actin,  or fibrous actin, it does so with a specific directionality. This is important for the "stick and pull" processes involved in sarcomere shortening during muscular  contraction. G­actin contains a specific binding site, located between the two major domains, for ATP and a divalent metal ion. Mg2+ ion is most likely the physiologically important  cation, but Ca2+ also binds tightly and competes with Mg2+ for the same tight binding site. It is the G­actin–ATP–Mg2+ complex that aggregates to form the F­actin  polymer (see Figure 22.34). Aggregation can occur from either direction, but kinetic data indicate that the preferred direction of aggregation is TABLE 22.8 Summary of the Amino Acid Differences Between Chicken Gizzard Smooth  Muscle Actin, Skeletal Muscle Actin, and Bovine Cardiac Actin Residue Number

  Actin Type

1

2

3

17

89

298

357

Val

Thr

Met

Thr

Leu

Ser

Leu

Ser

Asp

Glu

Asp

Cardiac muscleb

 

Asp

Glu

Smooth musclec

Absent

a

Skeletal muscle

 

Glu

 

  Cys

Ser

Source: Adapted from Vandekerckhove, J., and Weber K. FEBS Lett. 102:219, 1979. a From rabbit, bovine, and chicken skeletal muscle. b From bovine heart. c From chicken gizzard.

   

  

Page 952

Figure 22.30  Secondary structural elements of G­actin crystal structure.  ADP and the metal ion are shown in the cleft between the two large domains.  Redrawn with permission from Lorenz, M., Popp, D., and Holmes, K. C. J. Mol. Biol.  234:826, 1993. By permission of the publisher, Academic Press Limited, London.

by extension from the large end of the molecule where the rate is diffusion controlled. ATP hydrolysis occurs by orders of magnitude faster in the aggregated actin than  it does in the monomer. G­actin–ADP–Mg2+ also aggregates to form F­actin but at a slower rate. Orientation of G­actin molecules in F­actin is such that subdomains  1 and 2 are to the outside where myosin binding sites are located. F­actin may be viewed as either (1) a single­start, left­handed helix with rotation of the monomers  through an approximate 166° with a rise of 27.5 Å or (2) a two­start, right­handed helix with a half pitch of 350–380 Å. There are a number of proteins in the cytosol that bind actin. b ­Actinin binds to F­actin and plays a major role in limiting the length of the thin filament. a ­Actinin, a  homodimeric protein with a subunit molecular weight of 90–110 kDa, binds adjacent actin monomers of F­actin at positions 86–117 and 350–375 and strengthens the  fiber. It also helps to anchor the actin filament to the Z line of the sarcomere. There are two other major proteins associated with the thin filament, tropomyosin and  troponin. Tropomyosin is a rod­shaped protein consisting of two dissimilar subunits, each of about 35 kDa. It forms aggregates in a head­to­tail configuration. This polymerized  protein interacts in a flexible manner with the thin filament throughout its entire length. It fits within the groove of the helical assembly of the actin monomers of F­actin.  Each of the single tropomyosin molecules interacts with about seven monomers of actin. The site on actin with which tropomyosin interacts is between subdomains 1  and 3. Tropomyosin helps to stabilize the thin filament and to transmit signals for conformation change to other components of the thin filament upon Ca2+ binding.  Bound to each individual tropomyosin molecule is one molecule of troponin.

 

  

Page 953

Figure 22.31  Best fit model for the 4 Ca2+ ∙ Tn­C ∙ Tn­I complex.  A model for the complex of 4 Ca2+ ∙ troponin C ∙  troponin I based on neutron scattering studies with deuterium labeling and  contrast variation (Olah, C. A., and Trewhella, J., Biochemistry 33:12800,  1994). (Right) A view showing the spiral path of troponin I (green  crosses) winding around the 4 Ca2+ ∙ troponin  C that is represented by an a­carbon backbone trace (red  ribbon) with the C, E, and G helices labeled. (Left) The same view with 4  Ca2+ ∙ troponin C represented as a CPK model.  Photograph generously supplied by Dr. J. Trewhella. The publisher  recognizes that the U. S. Government retains a nonexclusive, royalty­free  license to publish or reproduce the published form of this contribution or  to allow others to do so, for U. S. Government purposes.

Troponin has three dissimilar subunits designated Tn­C, Tn­I, and Tn­T with molecular weights of about 18 kDa, 21 kDa, and 37 kDa, respectively. The Tn­T  subunit binds to tropomyosin. The Tn­I subunit is involved in the inhibition of the binding of actin to myosin in the absence of Ca2+. The Tn­C subunit, a calmodulin­like  protein, binds Ca2+ and induces a conformation change that alters the conformation of Tn­I and tropomyosin, resulting in exposure of the actin–myosin binding sites. A three­dimensional structure of Tn­C shows it to be a dumbbell­shaped molecule with much similarity to calmodulin. A structural model of the calcium saturated Tn­ C–Tn­I complex is shown in Figure 22.31. The Tn­I subunit fits around the central region of Tn­C in a helical coil conformation and forms caps over it at each end.  The cap regions of Tn­I are in close contact with Tn­C when Tn­C is fully saturated with calcium ions. Tn­C has four divalent metal ion binding sites. Two are in the  N­terminal region, are high affinity (Kdissoc of about 10–7 M), and are presumed to be always occupied since this is about the concentration of calcium ions in resting  cells. Under these conditions, Tn­I has a conformation that permits its interaction with binding sites on actin, inhibiting myosin binding and preventing contraction. Upon  excitation, the calcium ion concentration increases to about 10–5 M, high enough to effect calcium binding to sites within the N­terminal region of Tn­C. Tn­I now binds  preferentially to Tn­C in a capped structural conformation as shown in Figure 22.31. Myosin binding sites on actin are now exposed. The relatively loose interaction of  tropomyosin with actin gives it the flexibility to alter its conformation as a function of calcium ion concentration and to assist in blockage of the myosin binding sites on  actin. (See Clin. Corr. 22.7 for additional information about troponin.) Figure 22.26i shows schematically a cross section of the sarcomere and the relative arrangement of the thin and thick filaments. There are six thin filaments surrounding  each thick filament. The arrangement of myosin head groups around the thick filaments and the flexibility of those head groups make it possible for each thick filament  to interact with multiple thin filaments. When cross­bridges are formed between the thick and thin filaments, they do so in patterns consistent with that shown in the  electron micrograph of Figure 22.32. This figure shows a two­dimensional view of the myosin of the thick filament interacting with the actin of the thin filaments lying on  either side of it. Similar interactions of myosin occur with the actin of the other four thin filaments that surround it.

 

  

Page 954

CLINICAL CORRELATION 22.7  Troponin Subunits as Markers for Myocardial Infarction Troponin has three subunits (Tn­T, Tn­I, and Tn­C) each of which is expressed by more  than one gene. Two genes code for skeletal muscle Tn­I, one in fast­ and one in slow­ skeletal muscle; and one gene codes for cardiac muscle Tn­I. The genes that code for  Tn­T have the same distribution pattern. They differ in that the slow­skeletal muscle gene  for Tn­I is also expressed in fetal heart tissue. The gene for the cardiac form of Tn­I  appears to be specific for heart tissue. Tn­C is encoded by two genes, but neither gene  appears to be expressed only in cardiac tissue. The cardiac form of Tn­I in humans is about 31 amino acids longer than the skeletal  muscle form, which makes it easy to differentiate from others. Serum levels of Tn­I  increase within four hours of an acute myocardial infarction and remain high for about  seven days in about 68% of patients tested. Almost 25% of one group of patients tested  also showed a slight increase in the cardiac­form of Tn­I after acute skeletal muscle  injury. This would be a good but not a very sensitive test for myocardial infarction. Two isoforms of cardiac Tn­T, Tn­T1, and Tn­T2, are present in adult human cardiac  tissue. Two additional isoforms are also present in fetal heart tissue. Speculation is that the  isoforms are the result of alternative splicing of mRNA. Serum levels of Tn­T2 increase  within four hours of acute myocardial infarction and remain high for up to 14 days. The  appearance of Tn­T2 in serum is 100% sensitive and 95% specific for detection of  myocardial infarction. In the United States, the Food and Drug Administration has given  approval for marketing of the first Tn­T assay for acute myocardial infarction. Myocardial  infarcts are either undiagnosed or misdiagnosed in hospital patients admitted for other  causes, or in 5 million or more people who go to doctors for episodes of chest pain. It is  believed that this test will be sufficiently specific to diagnose myocardial incidents and to  help direct doctors to proper treatment of these individuals. Anderson, P. A. W., Malouf, N. N., Oakeley, A. E., Pagani, E. D., and Allen, P. D.  Circ. Res. 69:1226, 1991; and Ottlinger, M. E., and Sacks, D. B. Clin. Lab. News, 33,  1994.   Muscle Contraction Requires Ca2+ Interaction Contraction of skeletal muscle is initiated by transmission of nerve impulses across the neuromuscular junction mediated by release into the synaptic cleft of the  neurotransmitter acetylcholine. The acetylcholine receptors are associated with the plasma membrane and are ligand gated. Binding of acetylcholine causes them  to open and to permit Ca2+/Na+ to enter the sarcomere. The electron micrograph and accompanying diagrams of Figure 22.33 provide a picture of the anatomical  relationship between the presynaptic nerve and the sarcomere. There are transverse tubules along the membrane in the vicinity of the Z lines that are connected to the  terminal cisternae of the sarcoplasmic reticulum. Nerve impulses result in a depolarization of the plasma membrane and the transverse tubules, and an influx of Ca2+ into  the sarcomere. As indicated above, Ca2+ concentration increases about 100­fold, permitting it to bind to the low­affinity sites of Tn­C and to initiate the contraction  process. (See Clin. Corr. 22.8.) Energy for Muscle Contraction Is Supplied by ATP Hydrolysis ATP is an absolute requirement for muscular contraction. ATP hydrolysis by the myosin–ATPase to give the myosin–ADP complex and inorganic phos­

Figure 22.32  Electron micrograph of actin–myosin cross­bridges in a striated insect  flight muscle.  Reproduced with permission from Darnell, J., Lodish, H., and Baltimore,  D. Molecular Cell Biology. New York: Scientific American Books, 1986.

 

  

Page 955

Figure 22.33  Neuromuscular junction.  (a) Electron micrograph of a neuromuscular junction.  (b) Schematic diagram of the neuromuscular junction shown in (a).  Reproduced with permission from Alberts, B., Bray, D., Lewis, J., Raff, M.,  Roberts, K., and Watson, J. Molecular Biology of the Cell. New York: Garland Publishing, 1983.

phate leads to a myosin conformation that has an increased binding affinity for actin. Additional ATP is required for the dissociation of the myosin–actin complex. The concentration of ATP in the sarcomere remains fairly constant even during strenuous muscle activity, because of increased metabolic activity and of the action of  two enzymes: creatine phosphokinase and adenylate kinase. Creatine phosphokinase catalyzes the transfer of phosphate from phosphocre­

 

  

Page 956

CLINICAL CORRELATION 22.8  Voltage­Gated Ion Channelopathies Action potentials in nerve and muscle are propagated by the operation of voltage­gated ion  channels. Generally, there are three recognized types of voltage­gated cation channels: Na+,  Ca2+, and K+. Each of these has been cloned, primary sequence inferred from the DNA  sequence, and a model constructed of how each may be assembled in the membrane. Each is a  heterogeneous protein

Transmembrane organization of ion channel subunits.  Glycosylation and phosphorylation sites are marked. From Catterall (1995).

   

  

Page 957

consisting of various numbers of a  and b  subunits. A linear model of the arrangement of each of  these is shown in the figure above. In actual fact, they are arranged in a more­or­less circular  manner with a channel formed through the middle of the a  subunits. Roles of the b  subunits are  still being elucidated, but they appear to help stabilize and/or regulate activity of a  subunits. Toxins are being used to study subunit function. Tetrodotoxin and saxitoxin block Na+ channel  pores of the a  subunit. Scorpion toxins also bind to the a  subunit and appear to affect activation  and inactivation gating. Experiments of this type suggest that the a  subunit is involved in both  conductance and gating. Even though the Na+ channel was first cloned from nerve tissue, the electroplax of the eel, more  is known about how mutations affect its function in muscle. Voltage­gated channels from nerve  and muscle tissue show high homology in many of the transmembrane domains but are less  conserved in the intracellular connecting loops. A common effect of mutations in Na+ channels is  muscle weakness or paralysis. Some inherited sodium voltagegated ion channelopathies are listed  below. Each of these is reported to result from a single amino acid change in the a  subunit. The  inheritance pattern generally is dominant. Disorder

Unique Clinical Feature

Hyperkalemic periodic paralysis

Induced by rest after exercise, or the intake of K+

Paramyotonia congenita

Cold­induced myotonia

Sodium channel myotonia

Constant myotonia

It has been surmised (by Hoffman, 1995; see Catterall, 1995) that if the membrane potential is  slightly more positive (i.e., changes from –70 to –60 mV), the myofiber can reach the threshold  more easily and the muscle becomes hyperexcitable. If the membrane potential becomes even  more positive (i.e., up to –40 mV) the fiber cannot fire an action potential. This inability to  generate an action potential is synonymous with paralysis. The fundamental biochemical defect in  each case is a mutation in the channel protein. Catterall, W. A. Annu. Rev. Biochem. 64:493, 1995; and Hoffmann, E. P. Annu. Rev. Med.  46:431, 1995.   atine to ADP in an energetically favored manner:

If the metabolic process is insufficient to keep up with the energy demand, the creatine phosphokinase system serves as a "buffer" to maintain cellular levels of ATP.  The second enzyme is adenylate kinase that catalyzes the reaction

ATP depletion brings about rather rigid consequences to muscle cells. When the ATP supply of the muscle is exhausted and the intracellular Ca2+ concentration is no  longer controlled, myosin will exist exclusively bound to actin, a condition called rigor mortis. The function of ATP binding in muscular contraction is to promote  dissociation of the actin–myosin complex, not to promote its association. Model for Skeletal Muscle Contraction A model of the actin–myosin complex is shown in Figure 22.34. The myosin head undergoes conformation changes upon binding of ATP, hydrolysis of ATP, and  release of products. ATP binding leads to closure of the active site cleft and opening of the cleft in the region of the actin binding site. Hydrolysis of ATP and release of  inorganic phosphate result in closure of the cleft in the actin binding region. The conformation change that occurs is evident by the movement of two cysteine­containing  helices. The distance between the two cysteine residues (697 and 707) changes from about 19 A to about 2 A. If further conformation change is prevented by cross­ linking these two cysteines, ADP is trapped within its binding site. A stereo view of myosin showing the reactive cysteine pocket is shown in Figure 22.34b. The sequence of events leading to muscle contraction from its resting state, following Ca2+ entry into the cell, probably begins with the hydrolysis of bound ATP.  Myosin–ATP complex has a very low affinity for actin. Thus, even with exposed actin binding sites, any interaction between myosin and actin would be weak. The first  significant interaction between myosin and actin probably

 

  

Page 958

Figure 22.34  Model of actin–myosin interaction.  (a) Myosin is shown as a ribbon structure and actin as space­filling. Each G­actin  monomer is represented by different colors.  (b) Stereo view of myosin showing the pocket that contains the mobile ''reactive"  cysteine residues.  Reproduced with permission from Rayment, I., and Holden, H. M. TIBS 19:129, 1994.  Photograph generously supplied by Dr. I. Rayment.

 

  

Page 959

occurs upon release of inorganic phosphate. Release of ADP leads to tight binding (approximately a 10,000­fold increase) and another conformation change that  results in opening of the reactive cysteine pocket. The conformation change results in a movement of the upper portion of the myosin head in the direction of the arrows  in Figure 22.34a and movement of the thin filament in a direction away from the Z line, the power stroke. The thick filament is anchored in the center of the sarcomere  and the myosin head groups are polarized in opposite directions on each side of the M line. Each thick filament contains hundreds of S­1 or myosin head units  surrounded by six actin­containing thin filaments. Individual myosin units function in an asynchronous manner—possibly like changes in the position of hands on a rope  in the game of tug­of­war. Thus when some myosin head groups bind with high affinity, others have low affinity. Calcium Regulates Smooth Muscle Contraction Calcium ions play an important role in smooth muscle contraction also, but there are some important differences in the mechanism by which it acts. A mechanism for  calcium regulation of smooth muscle contraction is shown in Figure 22.35. Key elements of this mechanism are as follows. (1) A phosphorylated form of myosin light  chain stimulates Mg­ATPase, which supplies energy for the contractile process. (2) Myosin light chain is phosphorylated by a myosin light chain kinase (MLCK).  (3) MLCK is activated by a Ca2+–calmodulin (CaM) complex. (4) Formation of the Ca2+–CM complex is dependent on the concentration of intracellular Ca2+.  Release of Ca2+ from its intracellular stores or an increase in its flux across the plasma membrane is important for control. (5) Contraction is stopped by the action of a  myosin phosphatase or the transport of Ca2+ out of the cell. It is apparent that, in smooth muscle, many more biochemical steps are involved in the regulation of  contraction, steps that can be regulated in a progressive manner by hormones and other agents. These serve the function of smooth muscles well, namely, giving them  the ability to develop various degrees of tension and to retain it for prolonged periods of time.

Figure 22.35  Schematic representation of the mechanism of regulation of smooth muscle contraction.  Heavy arrows show the pathway for tension development and  light arrows show the pathway for release of tension. The Mg2+­ATPase  activity is highest in the actin–myosin­P complex. CaM, calmodulin; MLCK,  myosin light chain kinase.  Adapted from Kramm, K. E., and Stull, J. T. Annu. Rev. Pharmacol. Toxicol. 25:593, 1985.

 

  

Page 960

22.5—  Mechanism of Blood Coagulation The circulation of blood is essential for life, and the integrity of the process must be maintained. Some aspects of the importance of blood circulation in the maintenance  of pH, in the transport of oxygen and nutrients to cells, and in the transport of carbon dioxide and waste products from cells are well known. This section deals  primarily with a description of the system responsible for clot formation and dissolution. Blood circulation occurs in a very specialized type of closed system in which the volume of circulating fluid is maintained fairly constant. This system is also one in  which the transfer of solutes across its boundaries is a necessary function. Like any system of pipes and tubes, leaks can occur and must be repaired. The process of  blood clotting primarily addresses the question of stopping the leaks. Secondarily, small clots may form due to disease and other abnormalities that are independent of  total rupture of vesicles. Discussion of the function of the process must therefore extend beyond the primary one of leak prevention to include clot dissolution. The purpose of this section is to give a general picture of the mechanism of blood clotting from a biochemical viewpoint. To this end, this section will focus on the  relationship between blood clot formation, blood clot dissolution, and the enzymes and other proteins involved—their activation, regulation, inhibition, and synthesis.  Blood clotting is not a process of signal transduction in the same sense as are the other topics of this chapter. Instead, it is a dynamic process of signal amplification and  modulation. Some of the primary questions to be addressed are: (1) What initiates the clotting process? (2) What substances, reactions, and mechanisms are  responsible for forming the clot? (3) What factors and mechanisms are involved in inhibiting the clotting process once it is initiated? (4) How is the clot dissolved? It is important for the body to maintain hemostasis, that is, no bleeding. Thus the process of blood clotting is designed to stop as rapidly as possible the loss of blood  following vascular injury. When such an injury occurs, three major events take place: (1) aggregation of a protein, fibrin, into an insoluble network, or clot, to cover  the ruptured area to prevent the loss of blood; (2) clumping of blood platelets at the site of injury in an effort to form a physical plug to stop the leak; and (3)  vasoconstriction in an effort to reduce the blood flow through the area. Equally important is regulation of the process to prevent excessive clot formation. The processes mentioned above are emergency mechanisms for stopping the loss of blood. The process is not complete, however, until the ruptured vessel itself is  repaired and the clot dissolved. Many of the proteins involved in blood coagulation contain epidermal growth factor (EGF)­like domains. Whether these EGF­like  domains act directly to facilitate the regrowth of blood vesicles is not clear. Some of the major proteins (players) involved in this process (silent drama) are listed in Table 22.9, not necessarily in order of appearance. All are important and, as  time goes on, others are sure to be added. In fact, protein Z that occurs to a larger extent in children could be added but its role and function are not clear. Clot Formation Is a Membrane­Mediated Process Clot formation initially follows two separate pathways: intrinsic or contact factor pathway and extrinsic or tissue factor pathway (see Figures 22.36 and 22.38).  These pathways merge with the formation of factor Xa, the proteinase component of the multienzyme complex that catalyzes the formation of thrombin from  prothrombin. From this point on, there is a single pathway for clot formation. Historically, the term intrinsic pathway came from the observa­

 

  

Page 961 TABLE 22.9 Some of the Factors Involved in Blood Coagulation, Control, and Clot Dissolution Factor

Name

Pathway

Characteristic

Concentrationa

I Fibrinogen

Both

 

9.1

II Prothrombin

Both

Contains N­terminal Gla residues

1.4

III Tissue factor

Extrinsic

Transmembrane protein

       —

IV Calcium

Both

 

Both

Protein cofactor

0.03b

Extrinsic

Endopeptidase with Gla residues

0.010c

Intrinsic

Protein cofactor

0.0003b

Intrinsic

Endopeptidase with Gla residues

0.089

Both

Endopeptidase with Gla residues

0.136

Intrinsic

Endopeptidase

0.031

Intrinsic

Endopeptidase

0.375

Both

Transpeptidase

0.031b

V Proaccelerin VII Proconvertin VIII Antihemophilic IX Christmas factor X Stuart factor XI Thromboplastin  antecedent XII Hageman factor XIII Proglutamidase

 

 

Protein C

(Both)

Endopeptidase with Gla residues

0.065

 

Protein S

(Both)

Cofactor with Gla residues

0.30

 

Prekallikrein

Intrinsic

Zymogen/activator factor­XII

0.581

 

HMWKd

Intrinsic

Receptor protein

0.636

 

Antithrombin III

Both

Thrombin inhibitor

3.0

 

Plasminogen

 

Zymogen/clot dissolution

2.4

 

Heparin Co­II

Both

Thrombin inhibitor

1.364 0.952

 

a2­Antiplasmin

 

Plasmin inhibitor

 

Protein C inhibitor

 

Protein C inhibitor

0.070

 

a2­Macroglobulin

 

Proteinase inhibitor

2.9

 

LACIe

 

Extrinsic pathway inhibitor

0.003

a Concentrations are approximate and shown in micromolar. b These values approximate solution concentrations since some are complexed with other proteins in platelets. c This factor probably circulates as both VII and VIIa. d HMWK is high molecular weight kininogen. e LACI is lipoprotein­associated coagulation factor.

  tion that blood clotting would occur spontaneously when blood was placed in clean glass test tubes, leading to the idea that all components for the clotting process  were intrinsic to the circulating blood. Glass contains anionic surfaces that formed the nucleation points that initiate the process. In mammals, anionic surfaces are  exposed upon rupture of the endothelial lining of the blood vessels and are the binding sites for specific factors that initiate clotting in the intrinsic pathway. Similarly,  the term extrinsic came from the observation that there was another factor extrinsic to circulating blood that facilitates blood clotting. This factor was identified as  factor III, tissue factor (see Figure 22.39a). Whether intrinsic or extrinsic, the process of blood coagulation is initiated on the membrane and is continued on the  membrane surface at the site of injury. Reactions of the Intrinsic Pathway Reactions of the intrinsic pathway are shown in Figure 22.36. Upon injury to the endothelial lining of blood vessels and exposure of external membrane surfaces, the  proteinase zymogen factor XII binds directly to anionic surfaces and undergoes a conformation change that increases its catalytic activity 104­to 105­fold.  Prekallikrein and factor XI, also zymogens, circulate in blood as separate complexes with high molecular weight kininogen (HMWK): either a factor XI– HMWK complex or a prekallikrein–HMWK complex. In Figure 22.37 is a schematic diagram showing the functional regions of HMWK. The binding site on  HMWK for prekallikrein consists of approximately 31 amino acid residues. Factor XI binds to approximately 58 amino acid residues that include the

 

  

Page 962

Figure 22.36  Intrinsic pathway of blood coagulation.  HMWK, high molecular weight  kininogen. Activated factors are designated  with an "a."  Adapted from Kalafatis, M.,  Swords, N. A., Rand M. D., and Mann, K. G.  Biochim. Biophys. Acta 1227:113, 1994.

Figure 22.37  Schematic diagram of the functional regions of human high molecular weight kininogen (HMWK).  Bradykinin is derived from near the middle of HMWK by proteolysis. The resulting two  chains are held together by disulfide bonds (horizontal arrows).  Redrawn from Tait. J. F., and Fujikawa, K. J. Biol. Chem. 261:15396, 1986.

 

  

Page 963

31 to which prekallikrein binds. Factor XI and prekallikrein are attached to anionic sites of exposed membrane surfaces through their interactions with HMWK. This  brings those zymogens to the site of injury and in direct proximity to factor XII. The membrane­bound "activated" form of factor XII activates prekallikrein, a 619  amino acid protein, by cleavage at Arg371–Ile372, to yield kallikrein. Kallikrein contains two chains covalently linked by a single disulfide bond. Kallikrein, whose C­ terminal domain (248 amino acid residues) contains the catalytic site, further activates factor XII to give XIIa. Factor XI, which is membrane bound through its  noncovalent attachment to HMWK, is activated by XIIa through proteolytic cleavage to XIa. Factor XIa activates factor IX to IXa. Factor IXa in the presence of  factor VIIIa, a protein cofactor, forms the intrinsic factor ten'ase (intrinsic Xase) that can now activate factor X to Xa. Factor Xa is the catalytic moiety of the  proteinase complex responsible for the activation of prothrombin to thrombin (see Clin. Corr. 22.9). This is essentially a four­step cascade started by the "contact"  activation of factor XII and the autocatalytic action between factor XII and kallikrein to give XIIa (step 1). Factor XIIa activates factor XI (step 2); factor XIa  activates factor IX (step 3); and factor IXa, in the presence of VIIIa, activates factor X (step 4). If each enzyme molecule activated also catalyzed the formation of  100 others before it is inactivated, the amplification factor would be 1 × 106. Reactions of the Extrinsic Pathway A diagram of the extrinsic pathway is shown in Figure 22.38. The membrane receptor that initiates this process is factor III or tissue factor. Tissue factor (Figure  22.39a) is a transmembrane protein of 263 amino acids. Residues 243–263 are located on the cytosolic side of the membrane. Residues 220–242 are hydrophobic  residues and represent the transmembrane sequence. Residues 1–219 are on the outside of the membrane, are exposed after injury, and form the receptor for  formation of the initial complex of the extrinsic pathway. This domain is glycosylated and contains four cysteine residues. A stereo representation of a section of it  highlighting some of the amino acid residues involved in factor VII binding is shown in Figure 22.39b.

Figure 22.38  Extrinsic pathway of blood coagulation.

Tissue factor (factor III or TF) and factor VII are unique to the extrinsic pathway and are essentially all of its major components. Factor VII is a g ­ carboxyglutamyl or Gla­containing protein that binds to tissue factor only CLINICAL CORRELATION 22.9  Intrinsic Pathway Defects: Prekallikrein Deficiency Components of the intrinsic pathway include factor XII (Hageman factor), factor XI,  prekallikrein (Fletcher factor), and high molecular weight kininogen. Clinical disorders  have been associated with defects in each of these components. Inherited disorders in  each appear to be autosomal recessive. Each appears to be associated with an increase  in activated partial thromboplastin time (APTT). The only one of these components  directly associated with a clinical bleeding disorder is factor XI deficiency. In some cases where there is a prekallikrein (Fletcher factor) deficiency, autocorrection  after prolongation of the preincubation phase of the APTT test occurs. This phenomenon  is explained by the ability of factor XII to be activated by an autocatalytic mechanism.  The reaction is very slow in prekallikrein deficiency since the rapid reciprocal  autoactivation between factor XII and prekallikrein cannot take place. Prekallikrein  deficiency may be due to a decrease in the amount of the protein synthesized, to a genetic  alteration in the protein itself that interferes with its ability to be activated, or its ability to  activate factor XII. A lack of knowledge of the structure of the gene for prekallikrein  precludes definitive explanations of the mechanisms operational in patients with  prekallikrein deficiency. Specific deficiencies of the intrinsic pathway, however, can be  localized to a specific factor if the appropriate number of tests are performed. These may  include a direct measurement of the amount of each of the factors present in the patient's  plasma in addition to APTT test performed with and without prolonged preincubation  time. Use of these direct measurements helped diagnose a prekallikrein deficiency in a 9­ year­old girl who had a prolonged APTT. The functional level of prekallikrein in this  patient was less than 1/50th of the minimum normal value. Immunological test (ELISA)  showed an antigen level of 20–25%, suggesting that she was synthesizing a dysfunctional  molecule. Coleman, R. W., Rao, A.K., and Rubin, R. N. Am. J. Hematol. 48:273, 1995.    

  

Page 964 2+

2+

in the presence of Ca . The resulting TF–VII–Ca  complex is the catalytically active species. It catalyzes the formation of factor Xa from X. The zymogen form of factor VII is initially activated through protein–protein interaction as a result of its binding to tissue factor. Additional factor VII is activated  by Xa of the complex through proteolytic cleavage. Unlike other proteinases of the blood coagulation scheme, factor VIIa has a long half­life in circulating blood.  Once dissociated from tissue factor, VIIa is not catalytically active, and its presence in blood would be harmless. Formation of the initial complex with TF could  involve some of the already preformed factor VIIa, making it difficult to state with absolute certainty whether the zymogen form of VII in complex with tissue factor is  totally responsible for the initial activation of factor X. A 3­D ribbon structural representation of factor VIIa is shown in Figure 22.40. The region for tissue factor  interaction, Ca2+ binding, and the substrate binding pocket are highlighted. Thrombin Converts Fibrinogen to Fibrin The final phase in the formation of the fibrin clot (Figure 22.41) begins with action of the complex, factor Xa–Va, on prothrombin. A stereo view of factor

Figure 22.39  Tissue factor.  (a) Amino acid sequence of human tissue factor derived from its  cDNA sequence.  (b) A stereo representation of the carbon chain of the extracellular  domain of tissue factor. Residues important for binding of factor  VII are shown in yellow. Clusters of aromatic and charged  residues are shown in light blue.  (a) Redrawn from Spicer, E. K., Horton, R., Bloem, L., et al.  Proc. Natl. Acad. Sci. USA 84:5148, 1987.  (b) Reproduced with permission from Muller, Y. A., Ultsch,  M. H., Kelley, R. F., and deVos, A. M. Biochemistry 33:10864, 1994.  Copyright 1994 American Chemical Society. Photograph generously  supplied by Dr. A. de Vos.

 

  

Page 965

Figure 22.40  Ribbon structural representation of the protease domain  of factor VIIa.  The dark ribbon labeled  "TF inhibitory peptide" represents a section involved in  binding to tissue factor. The catalytic triad is shown in  the substrate binding pocket as H, S, and D for  His193, Ser344, and Asp338, respectively. The arrow  labeled 

 lies in the putative extended substrate  binding region.  Redrawn with permission from Sabharwal, A. K., Birktoft,  J. J., Gorka, J., et al. J. Biol. Chem.  270:1553, 1995.

Figure 22.41  Clot forming pathway.  Adapted from Kalafatis, M., Swords, N. A., Rand,  M. D., and Mann, K. G. Biochim. Biophys.  Acta 1227:113, 1994.

Xa is shown in Figure 22.42. Factor Xa is formed by both the extrinsic and the intrinsic pathways by cleavage of factor X at positions 145 and 151 with elimination of  a six amino acid peptide. Although the enzyme primarily responsible for activation of factor V is thrombin, factor Xa also catalyzes formation of Va. Thus the  prothrombinase complex, Xa–Va, appears early in the process. Thrombin, which circulates in plasma as prothrombin, catalyzes the conversion of fibrinogen to fibrin. Prothrombin, a 72­kDa protein (Figure 22.43), contains ten g­ carboxyglutamate (Gla) residues in its N­terminal region. Binding of calcium ions to these residues facilitates binding of prothrombin to membrane surfaces and to the  Xa–Va complex at the site of injury. The prothrombinase complex (Xa–Va) activates prothrombin by making two proteolytic cleavages on the carboxyl side of  arginine residues, first at position 320 and then at position 284. The active thrombin molecule (a ­thrombin) consists of two chains, one of 6 kDa and the other of 31  kDa, that are covalently linked by a disulfide

Figure 22.42  Stereo view of the CN­backbone structure of factor Xa.  The EGF­like domain is in bold.  Redrawn from Padmanabhan, K., Padmanabhan, K. P., Tulinsky, A., et al.  J. Mol. Biol 232:947, 1993.

 

  

Page 966

Figure 22.43  Schematic diagram of prothrombin activation.

bond. A stereo view of the active a ­thrombin molecule is shown in Figure 22.44. Regions involved in some of its functions are highlighted. The substrate for thrombin  is fibrinogen. Fibrinogen is a large molecule of approximately 340 kDa consisting of two tripeptide units with a ,b ,g structure (Figure 22.45). The subunits are "tied" together at their  N­terminal regions by a group of disulfide bonds. Fibrinogen has three globular domains, one on each end and one in the middle where the chains are joined. The  globular domains are separated by rod­like domains. A short segment of the free N­terminal regions projects out from the central globular domain. The N­terminal  region of the   and the   subunits, through charge–charge repulsion, prevent aggregation of fibrinogen. Thrombin cleaves these N­terminal peptides and allows the  resulting fibrin molecules to aggregate and to form the "soft" clot. The soft clot is stabilized and strengthened by the action of factor XIIIa, transglutamidase. This  enzyme catalyzes the formation of an isopeptide linkage by replacing the  ­amide group of glutamine residues of one chain with the  ­amino group of lysine residues  of another chain (Figure 22.46) with the release of ammonia. This cross­linking of fibrin completes the steps involved in the formation of the hard clot. Major Roles of Thrombin

a ­Thrombin activates the protein cofactors V and VIII and it is also involved in platelet aggregation. Factor V is a 330,000 molecular weight protein. Activation of  factor V by thrombin occurs through proteolytic cleavage at Arg709 and Arg1545. Factor Va is a heterodimer consisting of an N­terminal domain of 105

Figure 22.44  Stereo view of the active site cleft of human a ­thrombin.  Dark blue, basic amino acids, red, acid; light blue, neutral. The active site goes from left to right. Figure  courtesy of Dr. M. T. Stubbs II, Max­Planck Institut für Biochemie, Martinsreid, Germany.

 

  

Page 967

kDa and a C­terminal domain of 74 kDa. These two subunits are noncovalently held together by a calcium ion (Figure 22.47). Factor VIII circulates in plasma attached to another protein, von Willebrand's factor (vWF). Factor VIII is a 285­kDa protein that is activated by thrombin  cleavage at Arg372, Arg740, Arg1648, and Arg1689. The latter cleavage releases VIIIa from vWF. Factor VIIIa is a heterotrimer (Figure 23.47) composed of N­terminal  peptides of 40 kDa (A2) and 50 kDa (A1), and a C­terminal peptide of 74 kDa (A3). Factor VIIIa also contains a Ca2+ bridge between the N­ and C­terminal  domains. Classic hemophilia results from a deficiency in factor VIII (see Clin. Corr. 22.10). Thrombin also activates factor XIII, transglutamidase (Figure 22.48). Protransglutamidase exists in both plasma and platelets. The structural form of the platelet  enzyme is  2, whereas that of the plasma form is  2 2. Thrombin cleaves the a  subunit of both the platelet and the plasma forms of transglutaminase. Cleavage of the a subunit of the plasma form of the enzyme leads to dissociation of the b  subunit, which is not catalytically active. The platelet form of the enzyme is released at the site of  fibrin aggregation and is activated just by cleavage of the a  subunit.

Figure 22.45  Diagrammatic representation  of the fibrinogen molecule and its  conversion to the soft clot of fibrin.

Formation of a Platelet Plug The clumping of platelets at the site of injury is mediated by the presence of thrombin. There is a thrombin receptor, a member of the seven­transmembrane­domain  family of receptors, on the outside of endothelial cells. This receptor is exposed upon injury and is activated by a ­thrombin. Aggregation of platelets is facilitated by  their initial binding to this activated receptor. In addition to the formation of a physical plug, platelets undergo a morphological change and release other chemicals that  elicit other actions (Figure 22.49): ADP, serotonin, some types of phospholipids, and proteins that aid in coagulation and tissue repair. A glycoprotein, von  Willebrand's factor (vWF) is released, concentrates in the area of the injury, and also forms a link between the exposed receptor and the platelets. von Willebrand's  factor also serves as a carrier for factor VIII. Activation and release of factor VIII from vWF have been discussed.

Figure 22.46  Reactions catalyzed by transglutamidase.

 

  

Page 968

Figure 22.47  Organizational structure of cofactor proteins, factors VIII and V.  Positions for thrombin cleavage are shown. A's and C's represent structural domains.  Redrawn from Kalafatis, M., Swords, N. A., Rand, M. D., and Mann, K. G.  Biochim. Biophys. Acta 1227:113, 1994.

Platelet aggregation becomes autocatalytic with the release of ADP and thromboxane A2. Platelet factor IV, heparin binding protein, prevents heparin–antithrombin  III complexes from inhibiting serine proteinase coagulation factors, and it attracts cells with anti­inflammatory activity to the site of injury. About 20% of factor V exists  in platelets as does one form of factor XIII, the transglutamidase. Intact vascular endothelium does not normally initiate platelet aggregation since receptors and other elements are not exposed and activators such as ADP are rapidly  degraded or are not in blood in sufficient concentration to be effective. The endothelium also secretes prostacyclin(PGI2), a potent inhibitor of platelet aggregation.

Figure 22.48  Activation of transglutamidase  by thrombin.

Properties of Some of the Proteins Involved in Coagulation Calcium ions have at least two important functions in blood coagulation. They form complexes with factors that contain g­carboxyglutamyl (Gla) residues and

Figure 22.49  Action of platelets in blood coagulation.

 

  

Page 969

CLINICAL CORRELATION 22.10  Classic Hemophilia Hemophilia is an inherited disorder characterized by a permanent tendency for  hemorrhages, spontaneous or traumatic, due to a defective blood clotting system. Classic  hemophilia, hemophilia A, is an X­linked recessive disorder characterized by a deficiency  of factor VIII. About 1 in 10,000 males is born with a deficiency of factor VIII. Of the  approximate 25,000 hemophiliacs in the United States, more than 80% are of the A type.  Hemophilia B is due to a dysfunction in factor IX. Some hemophilia A patients may have a normal prothrombin time if the concentration of  tissue factor is high. One possible explanation for this is that factor V in human plasma is  much lower in concentration than factor X. Activation of an amount of factor X to Xa in  excess of that required to bind all of factor Va would initiate blood clotting by the  extrinsic pathway and give a normal prothrombin time. The intrinsic pathway would not  function normally due to the deficiency in factor VIII. Without the two pathways  operating in concert, the overall process of blood clotting would be impaired. Both factor  Xa and thrombin activate factor V and are involved in a number of other reactions. If the  overall process is not accelerated at its onset by intervention of the intrinsic pathway, due  to kinetics of the interaction of thrombin and factor Xa with the normally low  concentration of factor V, the clotting disorder is expressed. The blood levels of factor  VII in severe hemophilia A patients are less than 5% of normal. These patients have  generally been treated by blood transfusion with its associated dangers: the possibility of  contraction of hepatitis or HIV, and the 6% possibility of patients making autoantibodies.  Treatment of hemophiliacs has been made much safer as a result of cloning and  expression of the gene for factor VIII. The pure protein can be administered to patients  with none of the associated dangers mentioned above. Nemerson, Y. Blood 71:1, 1988.   induce conformational and electronic states that facilitate their interaction with membrane ''receptors" for initiation and localization of their reactions. Calcium ions also  bind at sites other than Gla residues, producing protein conformational changes that enhance catalytic activity. Evidence for this second role for calcium ions comes  from the observation that activation of at least one of the enzymes leads to both the cleavage and elimination of the N­terminal region containing the Gla residues, but  calcium ions are still required for its effective participation in blood coagulation. A schematic representation of the structural arrangement of five of the Gla­containing proteins listed in Table 22.9 is shown in Figure 22.50. Gla­containing residues  are located in the N­terminal region of the molecules followed by a structural component that resembles epidermal growth factor. The position of proteolytic cleavage  by activation proteinases is generally at an amino acid residue located between cysteine residues that form a disulfide bond. Activation may or may not result in loss of  a small peptide. Prothrombin is the only one whose activation is by cleavage outside the bridging disulfide bond and results in elimination of the Gla peptide.  Factor VII is activated by cleavage of a single Arg152–Ile153 bond. Factor IX is activated by cleavages at Arg145 and Arg180 with the release of an approximately 11­ kDa peptide. Factor X consists of two chains connected by a disulfide bridge. It is activated by cleavage of its heavy chain at Arg194–Ile195. The Gla residues are  located in the light chain. Protein C also consists of a heavy and a light chain connected by a disulfide bond. Cleavage of an Arg–Ile bond at position 169 results in its  activation.

Figure 22.50  Gla­containing proteins.  (a) General structure of the g­carboxyglutamyl­containing proteins.  (b) Structural organization of the zymogens and their  cleavage sites for activation.

 

  

Page 970

Role of Vitamin K in Protein Carboxylase Reactions Modification of prothrombin, protein C, protein S, and factors VII, IX, and X to form Gla residues occurs during synthesis by a carboxylase located on the  luminal side of the rough endoplasmic reticulum. Vitamin K (phytonadione, the "koagulation" vitamin) is an essential cofactor for this carboxylase. During the reaction,  the dihydroquinone or reduced form of vitamin K (Figure 22.51), vit K(H2), is oxidized to the epoxide form, vit K(O), using molecular oxygen. A plausible mechanism  involves the addition of molecular oxygen to the C­1 position of dihydro­vitamin K and its subsequent rearrangement to an alkoxide with a pKa of ~20. This  intermediate serves as a strong base and abstracts a

Figure 22.51  The vitamin K cycle as it functions in protein glutamyl carboxylation reaction.  X­(SH)  and X­S  represent the reduced and oxidized forms, respectively, of a  2

2

thioredoxin. The NADH­dependent and the dithiol­dependent vitamin K reductases are different  enzymes. The dithiol­dependent K and KO reductases are inhibited by dicumarol (I) and warfarin (II).  *Possible alkoxide intermediate (III).  Redrawn and modified from Vermeer, C. Biochem. J. 266:625, 1990.

 

  

Page 971

proton from the g­methylene carbon of glutamate, yielding a carbanion that can add to CO2 by a nucleophilic mechanism (Figure 22.51). The vitamin K epoxide  formed is converted back to the dihydroquinone by enzymes that require dithiols like thioredoxin as cofactors. Analogs of vitamin K inhibit dithiol­requiring vitamin  K reductases and result in conversion of all available vitamin K to the epoxide form that is not functional in this reaction. The overall carboxylation reaction is

The structure of two analogs, dicumarol and warfarin, that interfere with the action of vitamin K are shown in Figure 22.51. In animals treated with these compounds,  prothrombin, protein C, protein S, and factors VII, IX, and X are not posttranslationally modified, are deficient in Ca2+ binding, and cannot participate in blood  coagulation. Dicumarol and warfarin have no effect on blood coagulation in the test tube. Control of the Synthesis of Gla­Proteins Gla­peptides that are released from prothrombin upon activation are removed from circulation by the liver. These N­terminal Gla­containing peptides stimulate the de  novo synthesis of Gla­requiring proteins of the blood coagulation scheme (Figure 22.52). The proteins are synthesized even in the absence of vitamin K CLINICAL CORRELATION 22.11  Thrombosis and Defects of the Protein C Pathway Four major proteins are involved in the action of protein C in regulating blood  coagulation: protein C itself; protein S, a cofactor for protein C action; factor Va; and  factor VIIIa. The latter two are substrates for catalytic action of the protein C–protein S  complex. Mutations, generally inherited, in any of them can result in venous thrombosis  with various degrees of severity. De novo mutations have also been identified in patients showing type I protein C  deficiency. One was the result of a missense mutation, a transition of T to C, resulting in  the change of a codon for amino acid residue 270 from TCG to CCG. This gave Pro  instead of Ser at that position, resulting in a conformational change that affected activity.  The gene for protein C is on chromosome 2 and has 9 exons and 8 introns. In another  patient, a de novo mutation located at the exon VI–intron f junction was detected. A 5­ bp deletion (underlined below) occurred, resulting in a "read through" of sections of the  intron.

Normal sequence: 

Mutated sequence:  The normally translated sequence is in bold type. The degree of severity of thrombotic  events depends on the extent to which the gene inherited from the other parent is normal  and the extent to which it is expressed. Resistance to the action of activated protein C as a result of single point mutations in its  substrates, factor Va and factor VIIIa, can occur. This prevents or retards their  inactivation through the proteolytic action of protein C. The most commonly identified  cause of inherited resistance to the action of activated protein C is single point mutations  in the gene for factor V. A third cause of protein C­related thrombosis is a defect in protein S. Fewer specific  details are available that permit a definition of the mechanism of the interaction between  protein C and protein S, and likewise of the mutations that affect its function. It is quite  clear, however, that protein S deficiency leads to thrombotic events. Venous thrombosis  occurs in almost one­half of patients at some stage of their lives if they have deficiencies in  functional amounts of protein S. Gandrille, S., Jude, B., Alhenc­gelas, M., et al. Blood 84:2566, 1994; Zoller, B.,  Berntsdotter, A., Garcia de Frutos, P., et al. Blood 85:3518, 1995; and Reistma, P. H.,  Bernardi, F., Doig, R. G., et al. Thromb. Haemost. 73:876, 1995.    

  

Page 972

Figure 22.52  Role of Gla peptides in the regulation of de novo synthesis of coagulation factors.

or in the presence of antagonists of vitamin K. They are not secreted into the circulation, however. When vitamin K is restored, or is added in high enough  concentrations to overcome the effects of antagonists, the preformed proteins are carboxylated and secreted into the circulation. Activation of blood coagulation is a one­way process. The use of the activation peptides released from prothrombin to signal the liver to synthesize more of these  proteins is an efficient mechanism for maintaining their concentrations in blood at effective levels. Monitoring of patients on long­term therapy with vitamin K  antagonists is necessary to assure that posttranslational modification to produce the Gla­containing proteins is not shut down completely. Dual Role of Thrombin in Promoting Coagulation and Clot Dissolution The process of blood coagulation is self­controlling. One protein involved is protein C. Protein C, a Gla­containing protein, is activated in a membrane­

 

  

Page 973

bound complex of thrombin, thrombomodulin, and calcium. Thrombomodulin is an integral glycoprotein of the endothelial cell membrane that contains 560 amino acid  residues. Thrombomodulin shows amino acid sequence homology with the low­density lipoprotein receptor but very little with tissue factor. There is, however, a  great deal of similarity in functional domains between tissue factor and thrombomodulin, each of which functions as a receptor and activator for a proteinase.  Thrombomodulin carries out this function for thrombin for activation of the proteinase, protein C. Binding of thrombin to thrombomodulin reduces its catalytic  specificity for fibrinogen and enhances its specificity for protein C. Protein C inhibits coagulation by

Figure 22.53  Primary structure of recombinant protein C.  Redrawn with permission from Christiansen, W. T., Geng, J. P.,  and Castillino, F. J. Biochemistry 34:8082, 1995. Copyright 1995 American Chemical Society.

 

  

Page 974

inactivating factors Va and VIIIa. Another Gla­containing protein, protein S (a 75­kDa protein), is a cofactor for protein C. Deficiency in protein S and/or protein C,  leads to thrombotic diseases (see Clin. Corr. 22.11). A schematic representation of protein C showing some of its reactive regions is depicted in Figure 22.53. The Allosteric Role of Thrombin in Controlling Coagulation Important reactions of thrombin relative to its dual role in the processes of promoting and stopping coagulation are summarized in Figure 22.54. Thrombin exists in two  conformational forms: one is stabilized by Na+ and has high

Figure 22.54  Allosteric reactions  of thrombin and its actions on  fibrinogen and protein C.

Figure 22.55  Proposed mechanism of inhibition of the extrinsic pathway.  LACI is lipoprotein­associated coagulation factor whose structure is shown in (b).  Kunitz domain 1 inhibits factor VIIa and Kunitz domain 2 inhibits factor Xa. Arrows  indicate the presumed location of the active­site inhibitor region for each domain.  Redrawn with permission from Broze, G. J., Girard, T. J., and Novotny, W. F.  Biochemistry 29:7539, 1990. Copyright 1990 American Chemical Society.

 

  

Page 975

specificity for catalyzing the conversion of fibrinogen to fibrin; the other conformational form predominates in the absence of sodium, has low specificity for fibrinogen  conversion, but high specificity for thrombomodulin binding and activity on protein C. These forms are referred to as "fast" and "slow," respectively. This dynamic  "feedback'' mechanism is important for stopping the clotting process at its point of origin. Many thrombotic diseases are associated with mutations in protein C that  affect its activation by thrombin. Inhibitors of the Plasma Serine Proteinases Proteinase inhibitors in blood interact with enzymes of the blood coagulation system. Most of these fit into the serpin family of inhibitors. The term serpin was coined  by Carrell and Travis and stands for serine proteinase inhibitor. There is a tertiary structural similarity between them with a common core domain of about 350 amino  acids. Antithrombin III is one of the major serpins and inhibits most of the serine proteinases of coagulation. Inhibition of the proteinases is a kinetic process that  can begin almost as soon as coagulation itself begins. Initially, formation of inhibitor complexes is slow because the concentrations of the enzymes with which the  inhibitors interact are low. As activation of the enzymes proceeds, inhibition increases and becomes more prominent. These reactions, and destruction of protein  cofactors, eventually stop the coagulation process completely. In general, proteinase–inhibitor complexes do not dissociate readily and are removed intact from  blood by the liver. Inhibition of the extrinsic pathway, that is, the TF–VIIa–Ca2+–Xa complex, is unique and involves specific interaction with a lipoprotein­associated coagulation  inhibitor (LACI), formerly known as anticonvertin. LACI is a 32­kDa protein that contains three tandem domains (Figure 22.55, p. 974). Each domain is a  functionally homologous protease inhibitor that resembles other individual protease inhibitors such as the bovine pancreas trypsin inhibitor. LACI inhibits the  extrinsic pathway by interacting specifically with the TF–VIIa–Ca2+–Xa complex. Domain 1 binds to factor Xa and domain 2 binds to factor VIIa of the complex.  Binding of LACI to VIIa does not occur unless Xa is present. The uniqueness of this reaction is that LACI is a multi­enzyme inhibitor in which each of its separate  domains inhibits the action of one of the enzymes of the multi­enzyme complex of the extrinsic pathway.

Figure 22.56  Reactions involved in clot dissolution.

Fibrinolysis Requires Plasminogen and Tissue Plasminogen Activator (t­PA) to Produce Plasmin Reactions of fibrinolysis are shown in Figure 22.56. Lysis of the fibrin clot occurs through action of the enzyme plasmin, which is formed from plasminogen through  the action of tissue plasminogen activator (t­PA or TPA). Plasminogen has high affinity for fibrin clots and forms complexes with fibrin throughout various regions of  the fibrin network. t­PA also binds to fibrin clots and activates plasminogen to plasmin by specific bond cleavage. The clot is then solubilized by the action of plasmin. t­PA is a 72­kDa protein with several functional domains. It has a growth factor domain near its N terminus, two adjacent Kringle domains that interact with fibrin,  and a domain with protease activity that is close to its C terminus. Kringle domains are conserved sequences that fold into large loops stabilized by disulfide bonds.  These domains are important structural features for protein–protein interactions that occur with several blood coagulation factors. t­PA is activated by cleavage  between an Arg–Ile bond, resulting in a molecule with a heavy and a light chain. The serine protease activity is located within the light chain.

 

  

Page 976

Activity of t­PA is regulated by protein inhibitors. Four immunologically distinct types of inhibitors have been identified, two of which are of greater physiological  significance because they react rapidly with t­PA and are specific for it. They are plasminogen activator­inhibitor type 1 (PAI­1) and plasminogen activator­ inhibitor type 2 (PAI­2). The human PAI­2 is a 415 amino acid protein. Starting and stopping blood coagulation follow essentially the same type process, binding and proteolysis. Both are one­way processes and the only mechanism for  replenishing the proteins once they are used is by resynthesis. Bibliography General Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. Molecular Biology of the Cell, 2nd ed. New York: Garland Publishing, 1989. Nerve Bennett, M. K., and Scheller, R. H. The molecular machinery for secretion is conserved from yeast to neurons. Proc. Natl. Acad. Sci. USA 90:2559, 1993 Fried, G. Synaptic vesicles and release of transmitters: new insights at the molecular level. Acta Physiol. Scand. 154:1, 1995. Greengard, P., Valtorta, F., Czernik, A. J., and Benfenati, F. Synaptic vesicle phosphoproteins and regulation of synaptic function. Science 259:780, 1993. Grenningloh, G., Rienitz, A., Schmitt, D., et al. The strychnine­binding subunit of the glycine receptor shows homology with nicotinic acetylcholine receptors. Nature  328:215, 1987. Goodman, S. R., Zimmer, W. E., Clark, M. B., Zegon, I. S., Barker, J. E., and Bloom, M. L. Brain spectrin: of mice and men. Brain Res. Bull. 36:593, 1995. Pleribone, V. A., Shupllakov, O., Brodin, L., Hilfiker­Rothenfluh, S., Czernik, A. J., and Greengard, P. Distinct pools of synaptic vesicles in neurotransmitter release.  Nature 375:493, 1995. Sudhof, T. The synaptic vesicle cycle: a cascade of protein–protein interactions. Nature 375:645, 1995. Taylor, P. The cholinesterases. J. Biol. Chem. 266:4025, 1991. Vision Abrahamson, E. W., and Ostroy, S. E. (Eds.). Molecular Processes in Vision, Benchmark Papers in Biochemistry/3. Stroudsburg, PA: Hutchinson Ross Publishing,  1981. Farber, D. B. From mice to men: the cyclic GMP phosphodiesterase gene in vision and disease. Invest. Ophthalmol. Vis. Sci. 36:263, 1995. Nathans, J. Molecular biology of visual pigments. Annu. Rev. Neurosci. 10:163, 1987. Nathans, J., Davenport, C. M., Maumenee, I. H., et al. Molecular genetics of human blue cone monochromacy. Science 245:831, 1989. Palczewski, K. Is vertebrate phototransduction solved? New insights into the molecular mechanism of phototransduction. Invest. Ophthalmol. Vis. Sci. 35:3577,  1994. Stryer, L. Visual excitation and recovery. J. Biol. Chem. 266:10711, 1991. Zigler, J. S. Jr., and Goosey, J. Aging of protein molecules: lens crystallins as a model system. Trends Biochem. Sci. 7:133, 1981. Muscle Anderson, P. A. W., Malouf, N. N., Oakley, A. E., Pagani, E. D., and Allen, P. D. Troponin T isoform expression in humans: a comparison among normal and failing  adult heart, fetal heart, and adult and fetal skeletal muscle. Circ. Res. 69:1226, 1991. Carlier, M.­F. Actin: protein structure and filament dynamics. J. Biol. Chem. 266:1, 1991. da Silva, A. C. R., and Reinach, F. C. Calcium binding induces conformational changes in muscle regulatory proteins. Trends Biochem. Sci. 16:53, 1991. dos Remedios, C. G., and Moens, P. D. J. Actin and the actomyosin interface: a review. Biochim. Biophys. Acta 1228:99, 1995. Ebashi, S. Excitation–contraction coupling and the mechanism of muscle contraction. Annu. Rev. Physiol. 53:1, 1991. Gerisch, G., Noegel, A. A., and Schleicher, M. Genetic alteration of proteins in actin­based motility systems. Annu. Rev. Psychol. 53:607, 1991. Hirose, K., Franzini­Armstrong, C., Goldman, Y. E., and Murray, J. M. Structural changes in muscle crossbridges accompanying force generation. J. Cell Biol.  127:763, 1994. Huxley, H. E. The mechanism of muscular contraction. Science 164:1356, 1969. Lorenz, M., Popp, D., and Holmes, K. C. Refinement of the F­actin model against X­ray fiber diffraction data by the use of a directed mutation algorithm. J. Mol.  Biol. 234:826, 1993. Blood Antalis, T. M., Clark, M. A., Barnes, T., et al. Cloning and expressing of a cDNA coding for a human monocyte­derived plasminogen activator inhibitor. Proc. Natl.  Acad. Sci. USA 85:985, 1988. Colombatti, A., and Bonaldo, P. The superfamily of proteins with von Willebrand factor type A­like domains: one theme common to components of extracellular  matrix, hemostasis, cellular adhesion, and defense mechanisms. Blood 77:2305, 1991. Cooper, D. N. The molecular genetics of familial venous thrombosis. Blood Rev. 5:55, 1991. Dowd, P., Hershline, R., Ham, S. W., and Naganathan, S. Vitamin K and energy transduction: a base strength amplification mechanism. Science 269:1684, 1995. Hessing, M. The interaction between complement component C4b­binding protein and the vitamin K­dependent protein S forms a link between blood coagulation and  the complement system. Biochem. J. 277:581, 1991. Kalafatis, M., Sworde, N. A., Rand, M. D., and Mann, K. G. Membrane­dependent reactions in blood coagulation: role of the vitamin K­dependent enzyme  complexes. Biochim. Biophys. Acta 1227:113, 1994. Kuliopulus, A., Hubbard, B. R., Lam, Z., Koski, I. J., Furie, B., Furie, B. C., and Walsh, C. T. Dioxygen transfer during vitamin K­dependent carboxylase catalysis.  Biochemistry 31:7722, 1992. McClure, D. B., Walls, J. D., and Grinnell, B. W. Post­translational processing events in the secretion pathway of human protein C, a complex vitamin K­dependent  antithrombotic factor. J. Biol. Chem. 267:19710, 1992. Ny, T., Elgh, F., and Lund, B. The structure of the human tissue­type plasminogen activator gene: correlation of intron and exon structures to functional and structural  domains. Proc. Natl. Acad. Sci. USA 81:5355, 1984. Palston, P. A., and Gettings, P. G. W. A database of recombinant wild­type and mutant serpins. Thromb. Haemost. 72:166, 1994. Reitsma, P. H., Bernardi, F., Doig, R. G., et al. Protein C deficiency: a database of mutations, 1995 update. Thromb. Haemost. 73:876, 1995. Vermeer, C. g­Carboxyglutamate­containing proteins and the vitamin K­dependent carboxylase. Biochem. J. 266:625, 1990. Zeheb, R., and Gelehrter, T. D. Cloning and sequencing of cDNA for the rat plasminogen activator inhibitor­1. Gene 73:459, 1988.

 

  

Page 977

Questions C. N. Angstadt and J. Baggott 1. In the propagation of a nerve impulse by an electrical signal: A. the electrical potential across the membrane maintained by the ATP­driven Na+,K+ ion pump becomes more negative. B. local depolarization of the membrane causes protein conformational changes that allow Na+ and K+ to move down their concentration gradients. C. charge propagation is bidirectional along the axon. D. "voltage­gated" ion channels have a finite recovery time so the amplitude of the impulse changes as it moves along the axon. E. astrocytes are the antenna­like protrusions that receive signals from other cells. 2. All of the following are characteristics of nonpeptide neurotransmitters EXCEPT: A. they transmit the signal across the synapse between cells. B. they must be made in the cell body and then travel down the axon to the presynaptic terminal. C. electrical stimulation increasing Ca2+ in the presynaptic terminal fosters their release from storage vesicles. D. binding to receptors on the postsynaptic terminal induces a conformational change in proteins of that membrane. E. their actions are terminated by specific mechanisms within the synaptic junction. Refer to the following for Questions 3–5. A. acetylcholine B. 4­aminobutyrate (GABA) C. catecholamines D. 5­hydroxytryptamine (serotonin) 3. Binding to its receptor opens a channel for Cl–, causing hyperpolarization of the cell. 4. Termination of the signal typically involves the actions of both methyltransferase and monoamine oxidase, as well as reuptake into the presynaptic neuron. 5. Action is terminated by an esterase. 6. Which of the following is a correct statement about biochemical events occurring in the eye is (are) true? A. Glucose in the lens is metabolized by the TCA cycle in order to provide ATP for the Na+, K+–ATPase. B. Controlling the blood glucose level might reduce the incidence of diabetic cataracts by allowing the production of sorbitol. C. The high rate of the hexose monophosphate pathway in the cornea is necessary to provide NADPH as a substrate for glutathione reductase. D. The retina contains mitochondria so it depends on the TCA cycle for its production of ATP. E. Cataracts are the result of increasing blood flow in the lens leading to disaggregation of lens proteins. 7. Which of the following statements about rhodopsin is true? A. Rhodopsin is the primary photoreceptor of both rods and cones. B. The prosthetic group of rhodopsin is all­trans­retinol derived from cleavage of b ­carotene. C. Conversion of rhodopsin to activated rhodopsin, R*, by a light pulse requires depolarization of the cell. D. Rhodopsin is located in the cytosol of the cell. E. Absorption of a photon of light by rhodopsin causes an isomerization of 11­cis­retinal to all­trans­retinal. 8. All of the following statements about the transduction of the light signal on rhodopsin are true EXCEPT: A. cGMP is involved in the transmission of the signal between the disk membrane and the plasma membrane. B. it involves the G­protein, transducin. C. cGMP concentration is increased in the presence of an activated rhodopsin–transducin–GTP complex. D. the signal is turned off, in part, by the GTPase activity of the a  subunit of transducin. E. both guanylate cyclase and phosphodiesterase are regulated by calcium concentration. 9. The cones of the retina: A. are responsible for color vision. B. are much more numerous than the rods. C. have red, blue, and green light­sensitive pigments that differ because of small differences in the retinal prosthetic group. D. do not use transducin in signal transduction. E. are better suited for discerning rapidly changing visual events because a single photon of light generates a stronger current than it does in the rods. 10. When a muscle contracts, the: A. transverse tubules shorten, drawing the myofibrils and sarcoplasmic reticulum closer together. B. thin filaments and the thick filaments of the sarcomere shorten. C. light chains dissociate from the heavy chains of myosin. D. H bands and I bands of the sarcomere shorten because the thin filaments and thick filaments slide past each other. E. cross­linking of proteins in the heavy filaments increases. 11. All of the following statements about actin and myosin are true EXCEPT: A. the globular head section of myosin has domains for binding ATP and actin. B. actin is the major protein of the thick filament. C. the binding of ATP to the actin–myosin complex promotes dissociation of actin and myosin. D. F­actin, formed by aggregation of G­actin–ATP–Mg2+ complex, is stabilized when tropomyosin is bound to it. E. binding of calcium to the calmodulin­like subunit of troponin induces conformational changes that permit myosin to bind to actin. 12. ATP concentration is maintained relatively constant during muscle contraction by: A. increasing the metabolic activity. B. the action of adenylate kinase. C. the action of creatine phosphokinase. D. all of the above. E. none of the above.

 

  

Page 978

13. The nerve impulse that initiates muscular contraction: A. begins with the binding of acetylcholine to receptors in the sarcoplasmic reticulum. B. causes both the plasma membrane and the transverse tubules to undergo hyperpolarization. C. causes opening of calcium channels, which leads to an increase in calcium concentration within the sarcomere. D. prevents Na+ from entering the sarcomere. E. prevents Ca2+ from binding to troponin C. 14. Platelet aggregation: A. is initiated at the site of an injury by conversion of fibrinogen to fibrin. B. is inhibited in uninjured blood vessels by the secretion of prostacyclin by intact vascular endothelium. C. causes morphological changes and a release of the vasodilator, serotonin. D. is inhibited by the release of ADP and thromboxane A2. E. is inhibited by von Willebrand factor (vWF). 15. In the formation of a blood clot: A. proteolysis of g­carboxyglutamate residues from fibrinogen to form fibrin is required. B. the clot is stabilized by the cross­linking of fibrin molecules by the action of factor XIII, transglutamidase. C. antagonists of vitamin K inhibit the formation of g­carboxyglutamate residues in various proteins, thus facilitating the clotting process. D. tissue factor, factor III, must be inactivated for the clotting process to begin. E. the role of calcium is primarily to bind fibrin molecules together to form the clot. 16. Factor Xa, necessary for conversion of prothrombin to thrombin, is formed by the action of the TF–VII–Ca2+ complex on factor X: A. only in the extrinsic pathway for blood clotting. B. only in the intrinsic pathway for blood clotting. C. as part of both the extrinsic and intrinsic pathways. D. only if the normal blood clotting cascade is inhibited. 17. Lysis of a fibrin clot: A. is in equilibrium with formation of the clot. B. begins when plasmin binds to the clot. C. requires the hydrolysis of plasminogen into heavy and light chains. D. is regulated by the action of protein inhibitors on plasminogen. E. requires the conversion of plasminogen to plasmin by t­PA (tissue plasminogen activator). Answers 1. B This is the mechanism for impulse propagation. A: The potential becomes less negative. C: It is unidirectional. D: "Voltagegated" channels do have a finite  recovery time so the amplitude remains constant. E: This describes dendrites. Astrocytes are glial cells that are involved in processes isolating the CNS from the  external environment (pp. 921–923). 2. B This is true for neuropeptides, but many nonpeptide neurotransmitters are synthesized in the presynaptic terminal (pp. 923–924). A: This is a difference between  electrical and chemical signals (p. 921). C: What is the role of synapsin I in this process (p. 925)? E: Make sure you know the three types of processes involved (p.  927). 3. B GABA is an inhibitory neurotransmitter. All the others are excitatory ones that cause depolarization of the cells (p. 931). 4. C Methylation by catecholamine­O­methyltransferase is an important part of the metabolism of the catecholamines. A: Acetylcholinesterase terminates the action of  this (p. 928). B: GABA is converted into an intermediate of the TCA cycle (p. 931). D: Monoamine oxidase is the primary enzyme responsible for terminating  serotonin's action (p. 930). 5. A The enzyme is acetylcholinesterase (p. 928). 6. C Make sure you understand the role of glutathione in protecting against harmful by­products from atmospheric oxygen (p. 933). A: Most of the ATP (85%) in the  lens is generated by glycolysis (p. 934). B: Controlling glucose reduces sorbitol formation (p. 935). D: Its metabolism is similar to that of other eye tissues directly  involved in the visual process. Thus its major source of energy is from glycolysis (p. 935). E: Lens has no blood supply. In diabetic cataracts there is increased  aggregation of lens proteins because of increasing sorbitol (p. 935). 7. E This causes the conformational change of the protein that affects the resting membrane potential and initiates the rest of the events. A: Cones have the same  prosthetic group but different proteins, so rhodopsin is in rods only (p. 937). B: This is the precursor of the prosthetic group 11­cis­retinal (p. 938). C: Isomerization  of the prosthetic group leads to hyperpolarization (p. 939 Figure 22.20). D: Rhodopsin is a transmembrane protein (p. 937). 8. C The transducin complex activates the phosphodiesterase, thus lowering [cGMP] (p. 942). A: This is an example of a second messenger type chemical synapse. B  and D: Transducin meets the criteria for a typical G­protein. E: The enzymes are regulated in opposite directions by Ca2+, thus controlling [cGMP] (p. 943). 9. A Rods are responsible for low light vision. C: All three pigments have 11­cis­retinal; the proteins differ and are responsible for the slightly different spectra (pp.  937 and 944). D: The biochemical events are believed to be the same in rods and cones (p. 944). E: Cones are better suited for rapid events because their response  rate is about four times faster than rods, even though their sensitivity to light is much less (p. 945). 10. D This occurs because of association–dissociation of actin and myosin (pp. 948 and 957). A: Depolarization in the transverse tubules may be involved in  transmission of the signal but not directly in the contractile process (p. 954). B: The filaments do not change in length, but slide past each other (p. 948). C: This is not  physiological. E: Cross­linking occurs in the H band of the sarcomere but does not change during the contractile process (p. 953). 11. B A: See Figure 22.28. C: Note that the role of ATP in contraction is to favor dissociation, not formation, of the actin–myosin

 

  

Page 979

complex (p. 957). D. and E: Tropomyosin, troponin, and actin are the three major proteins of the filament. Their actions are closely interconnected (pp. 951–953). 12. D Make sure you know the reactions catalyzed by these two enzymes (pp. 954–957). 13. C A: Acetylcholine receptors are on the plasma membrane. B: The impulse results in depolarization of both of these structures. D: Both Ca2+ and Na2+ enter the  sarcomere when the channels open. E: Binding of Ca+2 to Tn­C initiates contraction (p. 954). 14. B The "ying–yang" nature of PGI2 and TXA2 help to control platelet aggregation until there is a need for it. A: Initiation is by contact with an activated receptor at  the site of injury. Clot formation requires activation of various enzymes (pp. 960–961). C: Serotonin is a vasoconstrictor. Vasodilation would be contraindicated in this  situation (p. 967). D: TXA2 facilitates aggregation. E: vWF forms a link between the receptor and platelets, promoting aggregation (p. 967). 15. B The cross­linking occurs between a glutamine and a lysine (Figure 22.46). A and E: g­Carboxyglutamate residues are on various enzymes; they bind calcium and  facilitate the interaction of these proteins with membranes that form the sites for initiation of reaction (pp. 968–969). C: Vitamin K is an activator for the g­ carboxylation reaction, which is a necessary posttranslational modification of some of the enzymes involved in clot formation (p. 970). D: TF, factor III, is the primary  receptor for initiation of the clotting process (p. 963). 16. A Tissue factor and factor VII are unique to the extrinsic pathway. B,C: The membrane interaction with the intrinsic pathway is with high­molecular­weight  kininogen and pre­kallikrein (p. 961). 17. E The clot is solubilized by plasmin. A: Both formation and lysis of clots are unidirectional. B: Both plasminogen and t­PA bind to the clot. C and D: Both of these  refer to t­PA (pp. 975–976).

 

  

Page 981

Chapter 23—  Biotransformations:  The Cytochromes P450 Richard T. Okita and Bettie Sue Siler Masters

23.1 Overview

982

23.2 Cytochromes P450: Nomenclature and Overall Reaction

982

23.3 Cytochromes P450: Multiple Forms

984

Multiplicity of Genes Produces Many Forms of Cytochrome P450

984

Substrate Specificity

985

Induction of Cytochromes P450

985

Polymorphisms

986

23.4 Inhibitors of Cytochromes P450

986

23.5 Cytochrome P450 Electron Transport Systems

987

NADPH–Cytochrome P450 Reductase Is the Flavoprotein Donor in the  Endoplasmic Reticulum

988

NADPH–Adrenodoxin Reductase Is the Flavoprotein Donor in  Mitochondria

989

23.6 Physiological Functions of Cytochromes P450

989

Cytochromes P450 Participate in Synthesis of Steroid Hormones and  Oxygenation of Eicosanoids

990

Cytochromes P450 Oxidize Exogenous Lipophilic Substrates

992

23.7 Other Hemoprotein­ and Flavoprotein­Mediated Oxygenations: The  Nitric Oxide Synthases

995

Three Distinct Nitric Oxide Synthase Gene Products Display Diverse  Physiological Functions

995

Structural Aspects of Nitric Oxide Synthases

996

Bibliography 997

 

Questions and Answers Clinical Correlations

   

  

998  

23.1 Consequences of Induction of Drug­Metabolizing Enzymes

986

23.2 Genetic Polymorphisms of Drug­Metabolizing Enzymes

987

23.3 Deficiency of Cytochrome P450 Steroid 21­Hydroxylase  (CYP21A2)

992

23.4 Steroid Hormone Production during Pregnancy

993

23.5 Clinical Aspects of Nitric Oxide Production

996

Page 982

23.1—  Overview The term cytochrome P450 refers to a family of heme proteins present in all mammalian cell types, except mature red blood cells and skeletal muscle cells, which  catalyze oxidation of a wide variety of structurally diverse compounds. Cytochrome P450 also occurs in prokaryotes. Substrates for this enzyme system include  endogenously synthesized compounds, such as steroids and fatty acids (including prostaglandins and leukotrienes), and exogenous compounds, such as drugs, food  additives, or industrial by­products that enter the body through food sources, injection, inhalation from the air, or absorption through the skin. The cytochrome P450  system has far­reaching effects in medicine. It is involved in (1) inactivation or activation of therapeutic agents; (2) conversion of chemicals to highly reactive molecules,  which may produce unwanted cellular damage, cell death, or mutations; (3) production of steroid hormones; and (4) metabolism of fatty acids and their derivatives.  Other heme­binding, cysteine thiolate­containing proteins also exist, including thromboxane, prostacyclin, and allene oxide synthases, as well as the nitric oxide  synthases. This chapter will address the cytochromes P450 in detail and will introduce the isoforms of nitric oxide synthase. Clinical implications of these oxygenation  systems will be presented. 23.2—  Cytochromes P450:  Nomenclature and Overall Reaction Designation of a particular protein as a cytochrome P450 originated from its spectral properties before its catalytic function was known. This group of proteins has a  unique absorbance spectrum that is obtained by adding a reducing agent, such as sodium dithionite, to a suspension of endoplasmic reticulum vesicles, frequently  referred to as microsomes, followed by bubbling of carbon monoxide gas into the solution. Carbon monoxide is bound to the reduced heme protein and produces an  absorbance spectrum with a peak at approximately 450 nm (Figure 23.1); thus the name P450 for a pigment with an absorbance at 450 nm. Specific forms of  cytochrome P450 differ in their maximum absorbance wavelengths, with a range between 446 and 452 nm. The many forms of cytochrome P450 are classified,  according to their sequence similarities, into various gene subfamilies; this system of nomenclature is being adopted almost universally. Individual cytochrome P450  forms are given an Arabic number to designate a specific family, followed by a capital letter to identify its subfamily, followed by another Arabic number designating the  individual P450 form, for example, 1A2 or 2D6. The term CYP, which represents the first two letters of cytochrome and the first letter in P450, is used as a preface to  designate a gene or protein as a cytochrome P450 form. Thus cytochromes P450 1A2 and P450 2D6 are designated CYP1A2 and CYP2D6 in this nomenclature  system. Members of the same family share at least 40% amino acid sequence homology and members of the same subfamily share at least 55% sequence homology.  Table 23.1 lists several human cytochrome P450 forms. In certain families several subfamilies have been identified such as in CYP2 (CYP2A and CYP2B) and CYP4  (CYP4A and CYP4B), whereas in others only a single gene has been reported (CYP17, CYP19, and CYP21).

Figure 23.1  Absorbance spectrum of the carbon  monoxide­bound cytochrome P450.  The reduced form of this heme  protein binds carbon monoxide to  produce a maximum absorbance at  approximately 450 nm. Hence this  cytochrome was designated P450.

The general reaction catalyzed by cytochrome P450 is written as follows:

where the substrate (S) may be a steroid, fatty acid, drug, or other chemical that has an alkane, alkene, aromatic ring, or heterocyclic ring substituent that can serve as  a site for oxygenation. The reaction is referred to as a monooxygenation and the enzyme as a monooxygenase because only one of the two oxygen atoms is  incorporated into the substrate. In mammalian cells, cytochromes P450

 

  

Page 983 TABLE 23.1 Human Cytochrome P450 Forms Cytochrome P450 Subfamilies CYP1

CYP2

CYP3

CYP4

CYP11

CYP17

CYP19

CYP21

1A1

2A6

3A3

4A9

11A1

 

 

 

 

 

 

 

 

Individual Forms

1A2

21A2

2A7

3A4

4A11

11B1

 

2B6

3A5

4B1

11B2

 

2C8

3A7

4F2

 

 

 

 

 

2C9

 

4F3

 

 

 

 

 

2C10

 

 

 

 

 

 

 

2C18

 

 

 

 

 

 

 

2C19

 

 

 

 

 

 

 

2D6

 

 

 

 

 

 

 

2E1

 

 

 

 

 

 

  serve as terminal electron acceptors in electron transport systems, which are present either in the endoplasmic reticulum or inner mitochondrial membrane. The  cytochrome P450 proteins contain a single iron protoporphyrin IX prosthetic group (see p. 1009), which binds oxygen, and the resulting heme protein contains  binding sites for the substrate. Heme iron of all known cytochromes P450 is bound to the four pyrrole nitrogen atoms of the porphyrin ring and two axial ligands, one  of which is a sulfhydryl group from a cysteine residue located toward the carboxyl end of the molecule (Figure 23.2). Heme iron may exist in two different spin states:  (1) a hexa­coordinated low­spin

Figure 23.2  Binding of protoporphyrin IX prosthetic group of cytochromes P450.  The cysteine thiolate ligand (Cys 357) liganded to the heme iron is shown in the top of  the figure and the space­filling model shows the camphor in the active site of the    cytochrome P450c a m. Generated by Dr. John Salerno from Dr. Tom Poulos' P450c a m structure using Biosym's Insight program run on a Silicon Graphics Indigo Extreme platform.

 

  

Page 984

iron or (2) a penta­coordinated high­spin state. Low­ and high­spin states are descriptions of the electronic shells within the iron atom. When a cytochrome P450  molecule binds a substrate, there is a perturbation of the structure surrounding heme iron such that a more positive reduction potential (–170 mV) results than in the  absence of substrate (–270 mV). This accelerates the rate at which cytochrome P450 may be reduced by electrons donated from NADPH through the flavoprotein  enzyme NADPH–cytochrome P450 reductase (Figure 23.3). In order for hydroxylation (monooxygenation) to occur, heme iron must be reduced from the ferric  (Fe3+) to its ferrous (Fe2+) state so that oxygen may bind to the heme iron. A total of two electrons is required for the mono­oxygenation reaction. Electrons are  transferred to the cytochrome P450 molecule individually, the first to allow oxygen binding and the second to cleave the oxygen molecule to generate the active oxygen  species for insertion into the reaction site of the substrate. 23.3—  Cytochromes P450: Multiple Forms Since the mid­1950s it has been known that one atom of molecular O2 is inserted into a substrate being metabolized. This process of monooxygenation is also  performed by other specialized proteins such as flavoprotein monooxygenases (hydroxylases). None of the other proteins classified as oxygenases, however, displays  the versatility of the members of the cytochrome P450 family. In the past decade, information on the sequence and structure of cytochromes P450 has led to a further  understanding of their evolution and regulation. Multiplicity of Genes Produces Many Forms of Cytochrome P450 Many cytochrome P450 forms have emerged due to gene duplication events occurring in the last 5–50 million years. The different forms of cytochrome P450 among  various animal species have likely arisen from the selective pressure of environmental influences, such as dietary habits or exposure to environmental agents. It is logical  that the primordial genes gave rise to those cytochromes P450 that metabolized endogenous substrates. Examination of the phylogenetic tree, generated by comparing  amino acid sequences and assuming a constant evolutionary change rate, leads to the conclusion that the earliest cytochromes P450 evolved to metabolize cholesterol  and fatty acids. Therefore they may have played a role in the maintenance of membrane integrity in early eukaryotes.

Figure 23.3  Sequence of reactions at cytochrome P450.  Diagram demonstrates the binding of substrate, transfer of the first and  second electrons from NADPH–cytochrome P450 reductase, and binding  of molecular oxygen.

 

  

Page 985

Substrate Specificity By the mid­1990s, nucleotide sequences for over 300 cytochrome P450 genes, coding for different proteins catalyzing the oxygenation of a variety of endogenous and  exogenous substrates, had been characterized. There remain other members of this gene superfamily for which sequences have not yet been determined. One of the  ways of characterizing these enzymes is the determination of substrate specificity. While this has been possible with many of the members of this family, the similarity  of molecular weights and other molecular properties has made purification of individual cytochromes P450 from the same organ or even the same subcellular organelle  very difficult, if not impossible. One way of determining the substrate specificity of a cytochrome P450 has been to express the cDNA for the particular protein via an  expression vector in an appropriate cellular expression system in which that specific cytochrome P450 form is not expressed constitutively. This has been achieved in  bacterial, insect, yeast, and mammalian cell systems and permits the unequivocal determination of substrate specificity uncomplicated by impurities of protein  purification. The assumption is that knowing the nucleotide sequences of the expressed genes leaves little doubt as to the source of enzyme activity expressed in those  cells. Induction of Cytochromes P450 Induction of various cytochromes P450 by both endogenous and exogenous compounds has been known since the mid­1960s. The mechanisms of induction of  cytochromes P450 have been demonstrated to be at either the transcriptional or posttranscriptional level and it is not possible to predict the mode of induction  based on the inducing compound. For example, a single cytochrome P450 can be induced by different mechanisms. In one case, induction occurs at the transcriptional  level and, in the other, it involves posttranscriptional events, that is, stabilization of mRNA. An example of the complexity of the induction process occurs with rat  CYP2E1 as a result of treatment with small organic molecules, such as ethanol, acetone, or pyrazole, or during fasting or diabetic conditions. Administration of these  small organic compounds produces larger amounts of the CYP2E1 protein without affecting the levels of mRNA. While the mechanism is not completely understood,  pyrazole may stabilize this specific cytochrome P450 from proteolytic degradation. However, in diabetic rats the sixfold induction of CYP2E1 protein is accompanied  by a tenfold increase in mRNA in the absence of an increase in gene transcription, suggesting stabilization of the mRNA. The role of specific cytosolic receptor proteins has been indicated in the case of some of the known inducing agents. One of the most extensively studied is the  interaction of 2,3,7,8­tetrachlorodibenzo­p­dioxin (TCDD) with its cytosolic receptor, called the aryl hydrocarbon (or Ah) receptor, which functions in the  induction of CYP1A1 and CYP1A2 forms. Polycyclic aromatic hydrocarbons serve as ligands which bind to the Ah receptor, producing a complex that is  translocated to the nucleus and is involved in binding to the upstream regulatory regions (specific response elements) of cytochrome P450 genes. A second protein  called the Ah receptor nuclear translocator or Arnt protein was found to interact with the ligand bound Ah receptor. The Arnt protein was essential for enabling  this ligand–Ah receptor complex to recognize and bind to its specific DNA response element. Utilizing cytochrome P450 gene transfection and expression vector  technology, it has been possible to express those portions of the cytochrome P450 genomic DNA representing the RNA polymerase II promoter region and the  upstream DNA sequences in conjunction with another gene coding for an enzyme that is not expressed constitutively in eukaryotes. In an assay of the prokaryotic  enzyme activity, for example, chloramphenicol acetyltransferase (CAT) in the expression system, it is possible to determine which specific nucleotide sequences of  DNA are involved in

 

  

Page 986

CLINICAL CORRELATION 23.1  Consequences of Induction of Drug­Metabolizing Enzymes Induction of the cytochrome P450 system may result in altered efficacy of therapeutic  drugs, as the accelerated rate of hydroxylation will increase the inactivation and/or  enhance the excretion rate of drugs. Induction of this protein system may also produce  unexpected and unwanted side effects of therapeutic agents due to increased formation of  toxic metabolites that may cause cell injury if produced in large enough concentrations.  The induction of different cytochrome P450 forms by a drug may stimulate the  metabolism of itself or other drugs that are substrates for the cytochrome P450 system.  Clinical problems may develop as a consequence of cytochrome P450 induction. The increase in clearance of oral contraceptives by rifampicin, an antituberculosis drug  and CYP3A4 inducer, has been shown to decrease the effectiveness of the contraceptive  agent and increase the incidence of pregnancy in women who are prescribed both drugs.  Fatalities have been reported in patients who are simultaneously treated with  phenobarbital, a long­acting sedative and potent cytochrome P450 inducer, and warfarin,  an anticoagulant, which is prescribed to patients with clotting disorders. Higher doses of  warfarin are required in these patients to maintain the same effective concentration of the  drug to delay coagulation because warfarin is a substrate for the cytochrome P450  induced by phenobarbital. Consequently, the drug is metabolized and cleared at a faster  rate, which reduces its therapeutic efficacy. Clinical problems are created when  phenobarbital is removed from the treatment regimen with no corresponding decrease in  warfarin levels. With time, cytochrome P450 levels decrease to the noninduced state but  the high concentrations of warfarin, proper under conditions of accelerated metabolism  and clearance, are in excess and produce unwanted hemorrhaging. Induction of CYP2E1 by chronic alcohol use has led to a warning for consumers of  acetaminophen, a common over­the­counter analgesic agent, because this cytochrome  P450 will metabolize acetaminophen to a toxic metabolite that may lead to liver cell  damage. These represent classic examples of cytochrome P450–drug interactions that  can lead to unwanted and unexpected clinical problems.   regulating these genes. These nucleotide sequences are referred to as xenobiotic regulatory elements or XREs. Another much studied inducer of cytochrome P450 genes is phenobarbital, which increases the transcription rate of certain cytochrome P450 forms. A receptor that  binds phenobarbital has not been described, but a specific DNA response element that is essential for phenobarbital­mediated induction has been identified in the  upstream regulatory region of CYP2B2 and CYP3A1 genes. Although the mechanism by which phenobarbital increases transcription is unknown, the intracellular  messenger, adenosine 2 ,3 ­cyclic monophosphate (cAMP), is a negative modifier, suppressing phenobarbital­mediated cytochrome P450 gene expression. An  increase in cAMP levels in rat hepatocytes was found to prevent the phenobarbital­directed induction of CYP2B2 and CYP3A1 by activating protein kinase A  activity. Some clinical consequences of induction of drug­metabolizing enzymes are presented in Clin. Corr. 23.1. Polymorphisms In addition to exposure to different inducing agents, individuals may differ in their rates of metabolism of a particular drug because of differences in the cytochrome  P450 genes they possess. Different forms of a cytochrome P450 gene may exist in a given population, which will alter the functional activity of the complement of  cytochromes P450. These genetic polymorphisms may be present in a small percentage of the population and cause an individual to be unable to metabolize a drug  at a sufficient rate, thereby producing significantly elevated drug levels. These ''poor metabolizers" may be at risk for a dose­dependent toxicity if the unmetabolized  form of the drug is pharmacologically active. Examples of genetic polymorphisms in drug metabolism are described in Clin. Corr. 23.2. 23.4—  Inhibitors of Cytochromes P450 Due to the many forms of cytochrome P450, it is of interest to examine the metabolic roles of these various enzymes in the organs in which they function. Several  inhibitors have been utilized to demonstrate that cytochrome(s) P450 may be involved in a metabolic pathway, for example, the metabolism of steroids in the adrenal  or specific reproductive organs. As has been discussed, the detection of cytochrome P450 in most tissues can be ascertained by the reduced­carbon monoxide  difference spectrum. Carbon monoxide (CO) binds to the heme iron, in lieu of oxygen, with a much higher binding affinity and thereby is a potent inhibitor of its  function. The identity of a cytochrome P450 in the catalysis of a putative substrate in a metabolic pathway rested on the reversal of CO inhibition by light at 450 nm,  corresponding to the reduced­CO absorption maximum. This was first demonstrated for steroids as substrates for adrenal mitochondrial cytochromes P450 and later  for drugs metabolized by liver microsomal cytochromes P450. However, this is a nonspecific inhibition characteristic of most cytochromes P450 and does not  differentiate among the various forms. More specific inhibitors are needed that can determine the role of a specific cytochrome P450 in a particular metabolic pathway. Although monospecific polyclonal  and monoclonal antibodies have been developed to a number of cytochromes P450, it is not always possible to determine that a single form is responsible because of  inhibition of a given reaction. The strong structural homology among the various forms may allow cross­reactivity among cytochromes P450. This is particularly true of  members of the same gene family that exhibit immune cross­reactivity. Recently, efforts have been directed to develop mechanism­based inhibitors, so­called suicide substrates, which bear strong resemblance to the sub­

 

  

Page 987

CLINICAL CORRELATION 23.2  Genetic Polymorphisms of Drug­Metabolizing Enzymes Genetic polymorphisms of cytochromes P450 result in the expression of cytochromes  P450 that are nonfunctional or exhibit lower enzymatic activities. This may result in  unwanted side effects because of the inability to eliminate the active form of the drug,  causing elevated concentrations in the body. It may also result in the absence of a  therapeutic effect because the active form of a drug is not formed. The discovery of an individual who suffered exaggerated hypotensive effects when  administered the antihypertensive drug, debrisoquine, led to the characterization of  individuals who metabolized substrates catalyzed by the CYP2D6 form inefficiently.  Approximately 5–10% of the Caucasian population, 2% of the Asian, and 1% of the  Arabic populations were deficient for the catalytically active CYP2D6 form. In addition  to debrisoquine, other drugs that are metabolized by CYP2D6 are sparteine,  amitriptyline, dextromethorphan, and codeine. In the case of codeine, CYP2D6 catalyzes  the O­demethylation of codeine to morphine. Approximately 10% of the dose of codeine  is metabolized to morphine in individuals who have a normal CYP2D6 and this  metabolism is responsible for the analgesic effects of this drug. Individuals who lack the  normal gene for CYP2D6 are unable to catalyze this reaction and are unable to achieve  the analgesic effects associated with codeine. Another genetic polymorphism was demonstrated in individuals who were poor  metabolizers of the drug mephenytoin. This drug is used in the treatment of epilepsy. Poor  metabolizers of this drug suffer greater sedative effects at normal dosages. The 4­ hydroxylation of the S­enantiomer of mephenytoin is carried out by CYP2C19.  Approximately 14–22% of the Asian population are reported to be poor metabolizers of  the S­isomer of mephenytoin whereas only 3–6% of the Caucasian population are  affected. These genetic polymorphisms may explain some of the interindividual or  interracial differences in the way individuals respond to therapeutic drugs. Eichelbaum, M., and Gross, A. S. The genetic polymorphism of debrisoquine/sparteine  metabolism—clinical aspects. In: W. Kalow (Ed.), Pharmacogenetics of Drug  Metabolism. New York: Pergamon Press, 1992, Chap. 21, p. 625; and Meyer, U. A.,  Skoda, R. D., Zanger, U. M., Heim, M., and Broly, F. The genetic polymorphism of  debrisoquine/sparteine metabolism—molecular mechanism. In: W. Kalow (Ed.),  Pharmocogenetics of Drug Metabolism. New York: Pergamon Press, 1992, Chap.  20, p. 609.   strate(s) of the specific cytochrome P450, but which during catalytic turnover form an irreversible inhibition product with the enzyme prosthetic group or protein.  Because of their structural resemblance to the substrate(s), these inhibitors become highly specific for that particular form of cytochrome P450. These inhibitors contain  functional groups that are metabolized to intermediates that result in their covalent binding to the enzymes, thereby accounting for their irreversibility. This represents a  possible tactical approach to drug design. 23.5—  Cytochrome P450 Electron Transport Systems Although cytochrome P450­catalyzed reactions require two electrons to accomplish the tasks of heme iron reduction, oxygen binding, and oxygen cleavage, a basic  mechanistic problem is the direct and simultaneous transfer of electrons from NADPH to the cytochrome P450. Pyridine nucleotides are two electron donors (see p.  250), but cytochrome P450, with its single heme prosthetic group, may only accept one electron at a time. Thus a protein that serves to transfer electrons from  NADPH to the cytochrome P450 molecule must have the capacity to accept two electrons but serve as a one­electron donor. This problem is solved by the presence  of a NADPH­dependent flavoprotein reductase, which accepts two electrons from NADPH simultaneously but transfers the electrons individually either to an  intermediate iron–sulfur protein (mitochondria) or directly to cytochrome P450 (endoplasmic reticulum). The active redox group of the flavin moiety is the  isoalloxazine ring (see p. 251). The isoalloxazine nucleus is uniquely suited to perform this chemical task since it can exist in oxidized and one­ and two­electron  reduced states (Figure 23.4). The transfer of electrons from NADPH to cytochrome P450 is accomplished by two distinct electron transport systems that reside  almost exclusively in either mitochondria or endoplasmic reticulum.

 

  

Page 988

Figure 23.4  Isoalloxazine ring of FMN or FAD in its oxidized, semiquinone (1e– reduced), or fully reduced (2e– reduced) states.

NADPH–Cytochrome P450 Reductase Is the Flavoprotein Donor in the Endoplasmic Reticulum In the endoplasmic reticulum, NADPH donates electrons to a flavoprotein called NADPH–cytochrome P450 reductase. The rat enzyme has a mass of 76,962 Da and  contains both flavin adenine dinucleotide (FAD) and flavin mononucleotide (FMN) as prosthetic groups. Until the recent characterization of nitric oxide synthases, it  was the only mammalian flavoprotein known to contain both FAD and FMN. A significant number of residues at the amino end of the molecule are hydrophobic, and  this segment of the protein is embedded in the endoplasmic reticulum (Figure 23.5). FAD serves as the entry point for electrons from NADPH, and FMN serves as  the exit point, transferring electrons individually to cytochrome P450. Because the flavin molecule may exist as one­ or two­electron­reduced forms and two flavin  molecules are bound per reductase molecule, the enzyme may receive electrons from NADPH and store them between the two flavin molecules before transferring  them individually to the cytochrome P450. In certain reactions catalyzed by the microsomal cytochrome P450, the transfer of the second electron may not be directly from NADPH–cytochrome P450  reductase but may occur from cytochrome b5, a small heme protein of molecular mass 15,330 Da. Cytochrome b5, is reduced either by NADPH–cytochrome P450  reductase or another microsome­bound flavoprotein, NADH–cytochrome b5 reductase. It is not known why reactions catalyzed by specific cytochromes P450  apparently require cytochrome b5 for optimal enzymatic activity. In addition, NADH–cytochrome b5 reductase and cytochrome b5 constitute the electron transfer  system for NADH to the iron–sulfur protein, fatty acid desaturase, which catalyzes the formation of double bonds in fatty acids (see p. 372).

Figure 23.5  Components of the endoplasmic reticulum (microsomal) cytochrome P450 system.  NADPH–cytochrome P450 reductase is bound by its hydrophobic tail to the membrane,  whereas cytochrome P450 is deeply embedded in the membrane. Also shown is  cytochrome b 5, which may participate in selected cytochrome P450­mediated reactions.

 

  

Page 989

NADPH–Adrenodoxin Reductase Is the Flavoprotein Donor in Mitochondria In mitochondria, a flavoprotein reductase also acts as the electron acceptor from NADPH. This protein is referred to as NADPH–adrenodoxin reductase because  its characteristics were described for the flavoprotein first isolated from the adrenal gland. This protein contains only FAD and the bovine NADPH–adrenodoxin  reductase has a mass of 50,709 Da. Adrenodoxin reductase is only weakly associated with its membrane milieu, unlike NADPH–cytochrome P450 reductase of  endoplasmic reticulum. Adrenodoxin reductase cannot directly transfer either the first or second electron to heme iron of cytochrome P450 (Figure 23.6). A small  molecular weight protein, called adrenodoxin (12,500 Da), serves as an intermediate between the adrenodoxin reductase and mitochondrial cytochrome P450.  The adrenodoxin molecule is also weakly associated with the inner mitochondrial membrane through interaction with the membrane­bound cytochrome P450.  Adrenodoxin contains two iron–sulfur clusters, which serve as redox centers for this molecule and function as an electron shuttle between the adrenodoxin reductase  and the mitochondrial cytochromes P450. One adrenodoxin molecule receives an electron from its mitochondrial flavoprotein reductase and interacts with a second  adrenodoxin, which then transfers its electron to the cytochrome P450 (Figure 23.6). Components of the mitochondrial cytochrome P450 system are synthesized in  the cytosol as larger molecular weight precursors, transported into mitochondria, and processed by proteases into smaller molecular weight, mature proteins. 23.6—  Physiological Functions of Cytochromes P450 Cytochromes P450 metabolize a variety of lipophilic compounds of endogenous or exogenous origin. These enzymes may catalyze simple hydroxylations of the carbon  atom of a methyl group, insertion of a hydroxyl group into a methylene carbon of an alkane, hydroxylation of an aromatic ring to form a phenol, or addition of an  oxygen atom across a double bond to form an epoxide. In dealkylation reactions, the oxygen is inserted into the carbon–hydrogen bond, but the resulting product is  unstable and rearranges to the primary alcohol, amine, or sulfhydryl compound. Oxidation of nitrogen, sulfur, and phosphorus atoms and dehalogenation reactions are  also catalyzed by cytochromes P450. Reactions catalyzed by cytochrome P450 forms are shown in Figure 23.7.

Figure 23.6  Components of mitochondrial cytochrome P450 system.  Cytochrome P450 is an integral protein of the inner mitochondrial membrane.  NADPH–adrenodoxin reductase and adrenodoxin (ADR) are peripheral  proteins and are `not embedded in the membrane.

 

  

Page 990

Figure 23.7  Reaction types catalyzed by cytochromes P450.

Cytochromes P450 Participate in Synthesis of Steroid Hormones and Oxygenation of Eicosanoids The importance of cytochrome P450­catalyzed reactions is illustrated by the synthesis of steroid hormones from cholesterol in the adrenal cortex and sex organs.  Mitochondrial and endoplasmic reticulum cytochrome P450 systems are required to metabolize cholesterol stepwise into aldosterone and cortisol in adrenal cortex,  testosterone in testes, and estradiol in ovaries. Cytochromes P450 are responsible for several steps in the adrenal synthesis of aldosterone, the mineralocorticoid responsible for regulating salt and water balance,  and cortisol, the glucocorticoid that governs protein, carbohydrate, and lipid metabolism. In addition, adrenal cytochromes P450 catalyze the synthesis of small  quantities of the androgen, androstenedione, a precursor of both estrogens and testosterone (see p. 900). Production of androstenedione regulates secondary sex  characteristics. Figure 23.8 presents a summary of these pathways. In adrenal mitochondria, a cytochrome P450 (CYP11A1) catalyzes the side chain cleavage converting cholesterol to pregnenolone, a committed step in steroid  synthesis. The removal of isocaproic aldehyde results from a cytochrome P450­catalyzed reaction involving sequential hydroxylation at C­22 and C­20 to produce  22­hydroxycholesterol and then 20,22­dihydroxycholesterol (Figure 23.9). An additional P450­catalyzed step is necessary to cleave the bond between C­20 and C­ 22 to produce pregnenolone. This reaction sequence, which requires

 

  

Page 991

Figure 23.8  Steroid hormone synthesis in the adrenal gland.  The reactions catalyzed by cytochromes P450 (CYP) are indicated.

3 NADPH and 3 O2 molecules, results in the breakage of a carbon–carbon bond and is catalyzed by a single cytochrome P450 enzyme, CYP11A1. After  pregnenolone is produced in mitochondria, it is transported into the cytosol where it is oxidized by 3b ­hydroxysteroid dehydrogenase/ 4,5­isomerase to progesterone.  Progesterone is metabolized to 11­deoxycorticosterone(DOC) by an endoplasmic reticulum cytochrome P450 (CYP21), which catalyzes the 21­hydroxylation  reaction. DOC is hydroxylated by an additional mitochondrial

Figure 23.9  Side chain cleavage reaction of cholesterol.  Three sequential reactions are catalyzed by cytochrome P450 to produce pregnenolone and isocaproic aldehyde.

 

  

Page 992

CLINICAL CORRELATION 23.3  Deficiency of Cytochrome P450 Steroid 21­Hydroxylase (CYP21A2) The adrenal cortex is a major site of steroid hormone production during fetal and adult  life. The adrenal gland is metabolically more active in fetal life and may produce 100–200  mg of steroids per day in comparison to the 20–30 mg produced per day in the non­ stressed adult adrenal gland. A number of enzymes are required for the production of  cortisol, and enzyme deficiencies have been reported at all steps of cortisol production.  Diseases associated with insufficient cortisol production are referred to as congenital  adrenal hyperplasias (CAHs). The enzyme deficiency that is most common in CAH is the  cytochrome P450­dependent 21­hydroxylase or CYP21A2. A deficiency in a functional  21­hydroxylase enzyme prevents the metabolism of 17a ­hydroxyprogesterone to 11­ deoxycortisol and subsequently to cortisol. This causes an increase in ACTH secretion,  the pituitary hormone that regulates adrenal cortex production of cortisol. Prolonged  periods of elevated ACTH levels causes adrenal hyperplasia and an increased production  of the androgenic hormones, DHEA and androstenedione. Clinical problems arise  because the additional production of androgenic steroids causes virilization in females,  precocious sex organ development in prepubertal males, or diseases related to salt  imbalance because of decreased levels of aldosterone. Clinical consequences of severe  21­hydroxylase deficiency may be recognizable at birth, particularly in females, because  the excessive buildup of androgenic steroids may cause obvious irregular development of  their genitalia. In male newborns, a deficiency in 21­hydroxylase activity may be  overlooked, because male genitalia will appear normal, but there will be precocious  masculinization and physical development. In late onset CAH, individuals are born  without obvious signs of prenatal exposure to excessive androgen levels, and clinical  symptoms may vary considerably from early development of pubic hair, early fusion of  epiphyseal growth plates causing premature cessation of growth, or male baldness  patterns in females. Donohoue, P. A., Parker, K., and Migeon, C. J. Congenital adrenal hyperplasia. In: C.  S. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (Eds.), The Metabolic and  Molecular Bases of Inherited Disease, 7th ed., Vol. II. New York; McGraw­Hill,  1995, Chap. 94, p. 2929.   cytochrome P450 (CYP11B2), which catalyzes both the 11b ­hydroxylase and 18­hydroxylase activities to form the mineralocorticoid, aldosterone, in the zona  glomerulosa (Chapter 21, p. 899). Synthesis of cortisol proceeds from either pregnenolone or progesterone and involves a cytochrome P450 (CYP17), an endoplasmic reticulum cytochrome P450,  which catalyzes the 17a ­hydroxylation reaction. Hydroxylation of the C­21 of 17a ­hydroxyprogesterone by CYP21 produces 11­deoxycortisol, which is transported  into mitochondria where it is hydroxylated at carbon atom 11 by CYP11B1 to form cortisol. These reactions occur primarily in the zona fasciculata of the adrenal  cortex. The consequences of a genetic polymorphism in CYP21 is presented in Clin. Corr. 23.3. Synthesis of steroids containing 19 carbon atoms from 17a ­hydroxypregnenolone or 17a ­hydroxyprogesterone is the result of the loss of the acetyl group at C­17.  This reaction is catalyzed by CYP17, identified as the same cytochrome P450 that hydroxylates C­17. Thus cleavage of the bond between C­17 and C­20 with loss  of the acetyl group is also catalyzed by a cytochrome P450 molecule. The factors that determine whether this cytochrome P450 performs only a single hydroxylation  step to produce the 17­OH product or proceeds further to cleave the C­17–C­20 bond has not been determined. The products are dehydroepiandrosterone  (DHEA) from 17a ­hydroxypregnenolone or androstenedione from 17a ­hydroxyprogesterone. DHEA in the sex organs may be metabolized by dehydrogenation of  the 3­OH group to androstenedione, a potent androgenic steroid that serves as the immediate precursor of testosterone. Another physiologically important reaction catalyzed by cytochromes P450 is synthesis of estrogens from androgens, collectively called aromatization because an  aromatic ring is introduced into the product. This is a complex reaction not unlike the side chain cleavage of cholesterol in which multiple hydroxylation reactions are  carried out by a single cytochrome P450 enzyme to form the aromatic ring and remove the methyl group at C­19. Figure 23.10 outlines the aromatization reaction of  ring A. Two cytochrome P450­mediated hydroxylation reactions at the methyl carbon atom at position 19 introduce an aldehyde group. It has been proposed that the  final step involves a peroxidative attack at C­19 with loss of the methyl group and elimination of the hydrogen atom to produce the aromatic ring. The reaction steps of  this sequence are catalyzed by the same cytochrome P450 and the enzyme is called aromatase or P450arom. P450arom is a member of the CYP19 subfamily. The  complexity of steroid hormone production and the role of cytochromes P450 are illustrated in Clin. Corr. 234. Other cytochromes P450 metabolize vitamin D3 to produce the 1,25­dihydroxy vitamin D3, which is the active form of this important hormone (see p. 907),  leukotriene B4 to produce 20­hydroxy­leukotriene B4, which is the less active form of this chemotactic agent (see p. 438), and arachidonic acid to produce epoxides,  hydroxy and dihydroxy derivatives of arachidonic acid, which may have important regulatory functions (see p. 433). Cytochromes P450 Oxidize Exogenous Lipophilic Substrates Exogenous substrates are often referred to as xenobiotics, meaning "foreign to life." They include therapeutic drugs, chemicals used in the workplace, industrial by­ products that become environmental contaminants, and food additives. Cytochromes P450 oxidize a variety of xenobiotics, particularly lipophilic compounds. The  addition of a hydroxyl group makes the compound more polar and thus more soluble in the aqueous environment of the cell. Many exogenous compounds are highly  lipophilic and accumulate within cells, interfering with cellular function over a period of time. Examples of xenobiotics that are oxidized by cytochromes P450 are  presented in Tables 23.2 and 23.3 (p. 994). In many cases the action of the cytochromes P450 leads to a compound

 

  

Page 993

Figure 23.10  Sequence of reactions leading to aromatization of androgens to estrogens.  Adapted from Graham­Lorence, S., Amarneh,  B., White, R. E., Peterson, J. A., and Simpson E. R. Protein Sci. 4:1065, 1995.

with reduced pharmacological activity or toxicity, which can readily be excreted in the urine or bile. Modified and unmodified xenobiotics can be altered chemically by  a variety of conjugating enzyme systems forming products that are even less toxic and that can readily be eliminated from the body. A list of enzymes that metabolize  xenobiotics is presented in Table 23.3; many occur primarily in the liver. One xenobiotic that has received considerable attention is benzol[a]pyrene, a common environmental contaminant produced from the burning of CLINICAL CORRELATION 23.4  Steroid Hormone Production during Pregnancy Steroid hormone production increases dramatically during pregnancy and, at term, the  pregnant woman produces 15–20 mg of estradiol, 50–100 mg of estriol, and  approximately 250 mg of progesterone per 24­h period. The amount of estrogens  synthesized during pregnancy far exceeds the amount synthesized by nonpregnant  women. For example, the pregnant woman at the end of gestation produces 1000 times  more estrogen than premenopausal women per day. Production of progesterone and estrogens in pregnant women is decidedly different from  that in the nonpregnant woman. The corpus luteum of the ovary is the major site for  estrogen production in the first few weeks of pregnancy, but at approximately 4 weeks of  gestation, the placenta begins synthesizing and secreting progesterone and estrogens.  After 8 weeks of gestation, the placenta becomes the dominant source for the synthesis of  progesterone. An interesting difference between the steroid hydroxylating systems in the  placenta and the ovary is that the human placenta lacks the cytochrome P450  (CYP11A1) that catalyzes the 17b ­hydroxylation reaction and the cleavage of the 17,20  carbon–carbon bond (see Chapter 21, p. 898, for details of synthesis of steroid  hormones). Thus the placenta cannot, by itself, synthesize estrogens from cholesterol. The  placenta catalyzes the side chain cleavage reaction to form pregnenolone from cholesterol  and oxidizes pregnenolone to progesterone but releases this hormone into the maternal  circulation. How then does the placenta produce estrogens if it cannot synthesize DHEA  or androstenedione from progesterone? This is accomplished in the fetal adrenal gland,  which represents a significant proportion of the total fetal weight compared to its adult  state. The fetal adrenal gland catalyzes the synthesis of DHEA from cholesterol and  releases it into the fetal circulation. A large proportion of the fetal DHEA is metabolized  by the fetal liver to 16a ­hydroxy­DHEA, and this product is aromatized in the placenta  to the estrogen, estriol. This is an elegant demonstration of the cooperativity of the  cytochrome P450­mediated hydroxylating systems in the fetal and maternal organ  systems leading to the progressive formation of estrogens during the gestational  development of the human fetus. Cunningham, F. G., MacDonald, P. C., Gant, N. F., Leveno, K. J., and Gilstrap, L. C.  The placental hormones. In: Williams Obstetrics, 19th ed. East Norwalk, CT: Appleton  & Lange, 1993, Chap. 6, p. 139.    

  

Page 994 TABLE 23.2 Xenobiotics Metabolized by Cytochromes P450 Reaction

Examples

Aliphatic hydroxylation

Valproic acid, pentobarbital

Aromatic hydroxylation

Debrisoquine, acetanilide

Epoxidation

Benzene, benzo[a]pyrene

Dealkylation

Aminopyrine, phenacetin, 6­ methyl­thiopurine

Oxidative deamination

Amphetamine

Nitrogen or sulfur oxidation 2­Acetylaminofluorene,  chlorpromazine Dehalogenation

Halothane

Alcohol oxidation

Ethanol

  coal, from the combustion of plant materials in tobacco, from food barbecued on charcoal, and as an industrial by­product. Benzo[a]pyrene binds to the aryl  hydrocarbon receptor and induces cytochromes P450 in the 1A subfamily, thus increasing its own metabolism. Several sites of the molecule may be hydroxylated by  different forms of cytochrome P450. Benzo[a]pyrene is metabolized to a carcinogen in animals and a mutagen in bacteria, prompting considerable work in identifying  the enzymes involved in this process. The product found to represent the ultimate carcinogen is benzo[a]pyrene­7,8­dihydrodiol­9,10­epoxide, the formation of  which is illustrated in Figure 23.11. The initial step involves a cytochrome P450­catalyzed epoxidation at the 7,8 position, hydrolysis by epoxide hydrolase to the  vicinal hydroxylated compound, benzo[a]pyrene­7,8­dihydrodiol, and then another epoxidation reaction to form benzo[a]pyrene­7,8­dihydrodiol­9,10­epoxide. The  parent compound, benzo[a]pyrene, is a weak carcinogen and, like most carcinogens that have been characterized, requires metabolic activation to its more potent  carcinogenic form. In a number of cases, the cytochrome P450 system is responsible for generation of the ultimate carcinogen Formation of toxic compounds by the cytochrome P450  system, however, does not mean that cell damage or cancer will occur, because many other factors will determine whether or not the toxic metabolite will cause cell  injury. These include the involvement of detoxification enzyme systems, the status of the immune system, nutritional state, genetic predisposition, and environmental  factors. One may ask why the body should possess an enzyme system that would create highly toxic compounds? As indi­ TABLE 23.3 Xenobiotic­Metabolizing Enzymes Type of Reaction

Enzyme

Representative  Substrate

Oxidation

Cytochrome P450

Toluene

 

Alcohol dehydrogenase

Ethyl alcohol

 

Flavin­containing monooxygenase Dimethylaniline

Reduction

Ketone reductase

Metyrapone

Hydration

Epoxide hydrolase

Benzo[a]pyrene­ 7,8­ epoxide

Hydrolysis

Esterase

Procaine

Conjugation

UDP­glucuronyltransferase

Acetaminophen

 

Sulfotransferase

b­Naphthol

 

N­acetyltransferase

Sulfanilamide

 

Methyltransferase

Thiouracil

 

Glutathionetransferase

Acetaminophen

   

  

Page 995

Figure 23.11  Metabolism of benzo[a]pyrene by cytochrome P450 and  epoxide hydrolase to form benzo[a]pyrene­7,8­dihydrodiol­9,10­epoxide.

cated, the purpose of the cytochrome P450 system is to add or expose functional groups making the molecule more polar and/or more susceptible to attack by  additional detoxification enzyme systems. In addition, many of these compounds resemble hormones that are our natural communication signals and would interfere  with cell–cell or organ–organ communication. Thus the cytochrome P450 system plays a significant role in the health and disease of humans. Different cytochromes P450 are responsible for generation of essential  steroid hormones, the regulation of blood levels of therapeutic agents, the removal of unwanted chemicals that would accumulate because of their lipophilicity, and the  generation of potentially toxic metabolites that may cause acute cell injury or damage to genetic material and lead to production of tumors. 23.7—  Other Hemoprotein­ and Flavoprotein­Mediated Oxygenations: The Nitric Oxide Synthases Three Distinct Nitric Oxide Synthase Gene Products Display Diverse Physiological Functions Release of nitric oxide from therapeutic drugs has been used as a treatment for angina pectoris since 1867, when Sir Thomas Lauder Brunton reported the use of  nitroglycerin and amyl nitrate in his patients. However, it was not known until the 1980s that nitric oxide, or NO∙, was the active agent in the dilation of blood vessels.  The demonstration that this free radical diatomic gas was the primary endogenous vasodilator released by the vascular endothelium led to the search for an enzymatic  source of NO∙. The source of NO∙ is the guanidino group of the naturally occurring amino acid, L­arginine. The reaction catalyzed by the enzymes responsible for the  conversion of L­arginine to L­citrulline and NO∙ is shown below:

Nitric oxide synthases have been examined in whole animals, tissues, and cells for functional properties and recently three genes have been identified for

 

  

Page 996

CLINICAL CORRELATION 23.5  Clinical Aspects of Nitric Oxide Production Although the role of NO∙ in the tumoricidal and bactericidal functions of macrophages is  essential in these cells, the overproduction of NO∙ (from the inducible isoform of nitric  oxide synthase, iNOS or NOS­II) has been implicated in septic/cytokine­induced  circulatory shock in humans through the activation of guanylate cyclase. This mechanism is  responsible for profound hypotension in patients after abdominal surgery or abdominal  trauma complicated by bacterial infections that produce endotoxins, as well as in patients  with neoplasias treated by IL­2 chemotherapy. Hypotension in these patients is often  refractory to treatment with conventional vasoconstrictor drugs. Therapeutic interventions  using NOS inhibitors are being examined in gastrointestinal inflammatory diseases, such as  pancreatitis and ulcerative colitis, and in arthritis. Administration of NOS inhibitors (e.g.,  specific to iNOS) might be a treatment of choice in such patients. The endothelial isoform of nitric oxide synthase, eNOS or NOS­III, is thought to play a  critical role in maintaining a basic vasotonus in hemodynamic regulation such that an  imbalance in the production of NO∙ could result in hypertension, thrombosis, or  atherosclerosis. Direct application of NO∙ gas may also be beneficial in the treatment of  pulmonary hypertension. In addition, recent experiments with mice in which the gene for  the neuronal isoform of nitric oxide synthase, nNOS or NOS­I, has been deleted have  resulted in animals with distended stomachs due to constriction of the pyloric sphincter.  This work has unexpectedly produced a model for the clinical disease, infantile  hypertrophic pyloric stenosis. It has also been shown that these nNOS­deficient mice are  resistant to brain damage as a result of ischemic injury usually resulting in vascular strokes.  While the direct connection to human disease has not yet been made, in this instance, it  presents a paradigm that can now be examined in clinical and pathological settings. The development of potent, specific inhibitors of the isoforms of nitric oxide synthase is an  active area of research being pursued collaboratively by investigators in academia and the  pharmaceutical industry.   the isoforms responsible for the activities in various tissues. Accordingly, the respective enzymes have been designated as neuronal (NOS­I), macrophage or  induced (NOS­II), or endothelial (NOS­III). Any tissue or cell may contain more than one isoform of nitric oxide synthase, thus contributing to the production of  NO∙ under various physiological circumstances. Studies of the macrophage type of nitric oxide synthase led to the conclusion that, upon treatment of animals with  cytokines or lipopolysaccharide, the increase in production of NO∙ was due to this isoform, since it is quantitatively the major source of NO∙. Subsequently, L­arginine  was shown to be the precursor of NO∙ in both endothelial and neuronal cells. Production of NO∙ is necessary for maintenance of vascular tone, platelet aggregation,  neural transmission, and bacterial and/ or tumor cytotoxicity (see Clin. Corr. 23.5). As further evidence of the importance of heme enzymes, signaling events require binding to the heme prosthetic group of guanylate cyclase of NO∙ produced in  neuronal and endothelial cells for activation of signaling events. The formation of cGMP leads to the subsequent downregulation of intracellular Ca2+ concentrations  and to a cellular response appropriate to the specific cell involved. For example, the production of cGMP in vascular smooth muscle cells resulting from NO∙  production leads to the lowering of Ca2+ concentrations, resulting in vasodilatation due to smooth muscle relaxation. Structural Aspects of Nitric Oxide Synthases Although the written reaction does not reveal the overall stoichiometry, it is representative of a monooxygenation reaction and the mechanism is similar to that catalyzed  by cytochromes P450. The oxygen atoms incorporated into both L­citrulline and NO∙ are derived from atmospheric oxygen. It was originally assumed that  oxygenation was occurring through mediation of tetrahydrobiopterin (BH4), a required cofactor for the overall reaction, analogous to the phenylalanine hydroxylase  reaction (see p. 464). The discovery that heme (iron protoporphyrin IX) is a functional prosthetic group associated with all three isoforms of nitric oxide synthase has  directed subsequent studies to include interactions between the flavoprotein and hemoprotein domains of these enzymes. These complex proteins must now be  understood from the standpoint of the roles of the flavins, heme, and BH4, under the control of Ca2+/calmodulin in the case of the neuronal (NOS­I) and endothelial  (NOS­II) isoforms. Figure 23.12 shows the overall structural organization of the neuronal NOS isoform. In addition to the various protein modules or domains of  NOS­I which are involved in electron transfer, substrate binding, oxygen activation, and calcium binding, a four­amino acid motif (glycine–leucine–glycine– phenylalanine, GLGF) has been identified in the amino terminal region of NOS­I. Although the function of this amino acid motif in NOS­I has not been established,  studies on other proteins containing this motif indicate that it may serve to target proteins to specific sites in the cell. The flow of electrons is assumed to occur in an analogous fashion to that of cytochrome P450­mediated electron systems. The electron donor is NADPH, which  donates two electrons to the enzyme­bound entry FAD, which, in turn, reduces the exit FMN. It is the latter flavin that reduces the heme iron prosthetic group to Fe2+  to which oxygen can now bind for the oxygenation of the substrate, L­arginine. The overall reaction is inhibited by carbon monoxide and enzyme activity is totally  dependent on bound calmodulin, which requires high concentrations of Ca2+ for the neuronal and endothelial isoforms. Calmodulin is involved in the control of  electron flow between the flavin prosthetic groups and between the exit flavin, FMN, and the heme prosthetic group in the oxygenase module. While the precise  residues constituting the binding site of BH4 have not been identified, its location has been narrowed to the

 

  

Page 997

Figure 23.12  Modular structure of neuronal nitric oxide synthase showing  approximate locations of prosthetic groups and cofactors.  Adapted from Masters, B.S.S., McMillan, K., Sheta, E. A., Nishimura, J. S. et al.  FASEB J.,10:552, 1996.

oxygenase module in the vicinity of the heme­binding site. The analogy between the systems synthesizing nitric oxide and the cytochrome P450­mediated systems is  remarkable, but the differences are significant and the oxygenase module probably represents an example of convergent evolution with the cytochromes P450. The  three­dimensional structures of mammalian representatives of either the cytochromes P450 or the nitric oxide synthases are yet to be determined. Bibliography Drug Metabolism and Cytochrome P450 Eichelbaum, M., and Gross, A. S. The genetic polymorphism of debrisoquine/sparteine metabolism—clinical aspects. In: W. Kalow (Ed.), Pharmacogenetics of  Drug Metabolism. New York: Pergamon Press, 1992, Chap. 21, p. 625. Gibson, G. G., and Skett, P. Introduction to Drug Metabolism, 2nd ed. London: Blackie Academic & Professional, 1994. Guengerich, F. J. Reactions and significance of cytochrome P­450 enzymes. J. Biol. Chem. 266:10019, 1991. Meyer, U. A., Skoda, R. C., Zanger, U. M., Heim, M., and Broly, F. The genetic polymorphism of debrisoquine/sparteine metabolism—molecular mechanism. In: W.  Kalow (Ed.), Pharmacogenetics of Drug Metabolism. New York: Pergamon Press, 1992, Chap. 20, p. 609. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W.,  Gunsalus, I. C., and Nebert, D. W. The P450 superfamily: update on new sequences, gene mapping, accession numbers and nomenclature. Pharmacogenetics 6:1,  1996. Parkinson, A. Biotransformation of xenobiotics In: C. D. Klassen (Ed.), Casarett and Doull's Toxicology. The Basic Science of Poisons. New York: McGraw­ Hill, 1996, Chap. 6, p. 113. Wilkinson, G. R., Guengerich, F. P., and Branch, R. A. Genetic polymorphism of S­mephenytoin hydroxylation. In: W. Kalow (Ed.), Pharmacogenetics of Drug  Metabolism. New York: Pergamon Press, 1992, Chap. 23, p. 657. Electron Transport and Cytochrome P450 Estabrook, R. W., Cooper, D. Y., and Rosenthal, O. The light reversible carbon monoxide inhibition of the steroid C21­hydroxylase system of the adrenal cortex.  Biochem. Zeit. 338:741, 1963. Peterson, J. A., and Prough, R. A. Cytochrome P­450 reductase and cytochrome b5 in cytochrome P­450 catalysis. In: P. R. Ortiz de Montellano (Ed.),  Cytochrome P­450 Structure, Mechanism, and Biochemistry. New York: Plenum Press, 1986, Chap. 4, p. 89. Induction of Cytochrome P450 Denison, M. S., and Whitlock, J. P. Xenobiotic­inducible transcription of cytochrome P450 genes. J. Biol. Chem. 270:18175, 1995. Sidhu, J. S., and Omiecinski, C. J. cAMP­associated inhibition of phenobarbital­inducible cytochrome P450 gene expression in primary rat hepatocyte cultures. J.  Biol. Chem. 270:12762, 1995. Endogenous Substrates and Cytochrome P450 Cunningham, F. G., MacDonald, P. C., Gant, N. F., Leveno, K. J., and Gilstrap, L. C. The placental hormones. In: Williams Obstetrics, 19th ed. East Norwalk, CT:  Appleton & Lange, 1993, Chap. 6, p. 139. Donohoue, P. A., Parker, K., and Migeon, C. J. Congenital adrenal hyperplasia. In: C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (Eds.), The Metabolic  and Molecular Bases of Inherited Disease, 7th ed., Vol. II. New York: McGraw­Hill, 1995, Chap. 94, p. 2929. Graham­Lorence, S., Amarneh, B., White, R. E., Peterson, J. A., and Simpson, E. R. A three­dimensional model of aromatase cytochrome P450. Protein Sci.  4:1065, 1995. Masters, B. S. S., Muerhoff, A. S., and Okita, R. T. Enzymology of extrahepatic cytochromes P450. In: F. P. Guengerich (Ed.), Mammalian Cytochromes P450.  Boca Raton, FL: CRC Press, 1987, Chap. 3, p. 107. Waterman, M. R., John, M. E., and Simpson, E. R. Regulation of synthesis and activity of cytochrome P­450 enzymes in physiological pathways. In: P. R. Ortiz de  Montellano (Ed.), Cytochrome P­450. Structure, Mechanism, and Biochemistry. New York: Plenum Press, 1986, Chap. 10, p. 315. Biochemistry and Physiology of Nitric Oxide Formation Bredt, D. S., and Snyder, S. H. Nitric oxide: a physiologic messenger molecule. Annu. Rev. Biochem. 63:175, 1994. Garthwaite, J., and Boulton, C. L. Nitric oxide signalling in the central nervous system. Annu. Rev. Physiol. 57:683, 1995. Griffith, O. W., and Stuehr, D. J. Nitric oxide synthases—properties and catalytic mechanism. Annu. Rev. Physiol. 57:707, 1995. Ignarro, L. J., Buga, G. M., Wood, K. S., Byrns, R. E., and Chaudhuri, G. Endothelium­derived relaxing factor produced and released from artery and vein is nitric  oxide. Proc. Natl. Acad. Sci. USA 84:9265, 1987. Khan, M. T., and Furchgott, R. F. Additional evidence that endothelium­derived relaxing factor is nitric oxide. In: M. J. Rand and C. Raper (Eds.), Pharmacology.  Amsterdam: Elsevier, 1987, p. 341. Marletta, M. A. Approaches toward selective inhibition of nitric oxide synthase. J. Med. Chem. 37:1899, 1994.

 

  

Page 998

Masters, B. S. S. Nitric oxide synthases: Why so complex? Annu. Rev. Nutr. 14:131, 1994. Masters, B. S. S., McMillan, K., Sheta, E. A., Nishimura, J. S., Roman, L. J., and Martasek, P. Neuronal nitric oxide synthase, a modular enzyme formed by  convergent evolution: structure studies of a cysteine thiolate­liganded heme protein that hydroxylates L­arginine to produce NO as a cellular signal. FASEB J. 10:552,  1996. Palmer, R. M. J., Ferrige, A. G., and Moncada, S. Nitric oxide release acccounts for the biological activity of endothelium­derived relaxing factor. Nature 327:524,  1987. Questions C. N. Angstadt and J. Baggott 1. All of the following are correct about a molecule designated as a cytochrome P450 EXCEPT: A. it contains a heme as a prosthetic group. B. it catalyzes the hydroxylation of a hydrophobic substrate. C. it may accept electrons from a substance such as NADPH. D. it undergoes a change in the heme iron upon binding a substrate. E. it comes from the same gene family as all other molecules designated as cytochromes P450. 2. Known roles for cytochromes P450 include all of the following EXCEPT: A. synthesis of steroid hormones. B. conversion of some chemicals to mutagens. C. hydroxylation of an amino acid. D. inactivation of some hydrophobic drugs. E. metabolism of fatty acid derivatives. 3. The induction of cytochromes P450: A. occurs only by endogenous compounds. B. occurs only at the transcriptional level. C. necessarily results from increased transcription of the appropriate mRNA. D. necessitates the formation of an inducer–receptor protein complex. E. may occur by posttranscriptional processes. 4. Flavoproteins are usually intermediates in the transfer of electrons from NADPH to cytochrome P450 because: A. NADPH cannot enter the membrane. B. flavoproteins can accept two electrons from NADPH and donate them one at a time to cytochrome P450. C. they have a more negative reduction potential than NADPH and so accept electrons more readily. D. as redox proteins, they can directly reduce cytochromes P450 while the nonprotein NADPH cannot. E. they contain iron–sulfur centers. 5. NADPH–cytochrome P450 reductase: A. uses both FAD and FMN as prosthetic groups. B. is found in mitochondria. C. requires an iron–sulfur center for activity. D. always passes its electrons to cytochrome b5. E. can use NADH as readily as NADPH. 6. The system necessary for the formation of double bonds in fatty acids: A. is the cytochrome P450 electron transport system in the endoplasmic reticulum. B. is the cytochrome P450 electron transport system in the mitochondria. C. contains NADH–cytochrome b5 reductase. D. uses NADPH–adrenodoxin reductase to reduce cytochrome b5. E. uses both FAD and FMN as prosthetic groups. 7. NADPH–adrenodoxin reductase: A. is located in the endoplasmic reticulum B. passes its electrons to a protein with iron–sulfur centers. C. has a stretch of hydrophobic amino acid residues at the N­terminal end. D. reacts directly with cytochrome P450. E. reacts directly with cytochrome b5. 8. Cytochrome P450 systems are able to oxidize: A. —CH2— groups. B. benzene rings. C. nitrogen atoms in an organic compound. D. sulfur atoms in an organic compound. E. all of the above. 9. In the conversion of cholesterol to steroid hormones in the adrenal gland: A. all of the cytochrome P450 oxidations occur in the endoplasmic reticulum. B. all of the cytochrome P450 oxidations occur in the mitochondria. C. side chain cleavage of cholesterol to pregnenolone is one of the cytochrome P450 systems that uses adrenodoxin reductase. D. cytochrome P450 is necessary for the formation of aldosterone and cortisol but not for the formation of the androgens and estrogens. E. aromatization of the first ring of the steroid does not use cytochrome P450 because it involves removal of a methyl group, not a hydroxylation. 10. Many xenobiotics (exogenous substrates) are oxidized by cytochromes P450 in order to: A. make them carcinogenic. B. increase their solubility in an aqueous environment. C. enhance their deposition in adipose tissue. D. increase their pharmacological activity. E. all of the above. 11. Benzo[a]pyrene, a xenobiotic produced by combustion of a variety of substances: A. induces the synthesis of cytochrome P450. B. undergoes epoxidation by a cytochrome P450. C. is converted to a potent carcinogen in animals by cytochrome P450. D. would be rendered more water­soluble after the action of cytochrome P450. E. all of the above.

 

  

Page 999

12. Phenobarbital is a potent inducer of cytochrome P450. Warfarin, an anticoagulant, is a substrate for cytochrome P450 with the result that the drug is metabolized  and cleared from the body more rapidly than normal. If phenobarbital is added to the therapeutic regimen of a patient, with no change in the dosage of warfarin, the  expected consequence would be: A. no change in the clinical results. B. an increased possibility of clot formation. C. an increased possibility of hemorrhaging. 13. Nitric oxide: A. is formed spontaneously by a reduction of NO2. B. is synthesized only in macrophages. C. is synthesized from arginine. D. acts as a potent vasoconstrictor. E. has three isoforms. 14. Nitric oxide synthase: A. catalyzes a dioxygenase reaction. B. is similar mechanistically to phenylalanine hydroxylase since it requires tetrahydrobiopterin. C. accepts electrons from NADH. D. uses a flow of electrons from NADPH to FAD to FMN to heme iron. E. is inhibited by Ca2+. Answers 1. E Several gene families are known. The number after CYP designates the family. B: The types of substrates are hydrophobic. It is classified as a monooxygenase.  C: See Figure 23.3. D: The change from hexa to penta coordinated gives the compound a more positive reduction potential (pp. 982–984). 2. C Cytochromes P450 are not the only hydroxylases and other types are active with amino acids (pp. 982, 989–995). 3. E There may be a stabilization of mRNA (as seen in diabetic rats) or decrease in the degradation of the protein, which may be a mechanism for pyrazole (p. 985).  A: One of the roles of cytochromes P450 is in the metabolism of exogenous substances. B and C: Transcriptional modification is only one of the mechanisms of  induction (see E). D: This has been shown with induction by some compounds, but with others, like phenobarbital, this is not so (p. 985). 4. B Heme can accept only one electron at a time while NADPH always donates two at a time. A: NADPH passes only electrons; it does not have to enter the  membrane. C: If this were true, the flow of electrons would not occur in the way it does. D: Protein–protein binding is not known to play a role here. E: Iron–sulfur  centers play a role in some, but not all, systems. Flavoproteins are not the only system with iron–sulfur centers (p. 987). 5. A This enzyme is one of two mammalian proteins known to do so. B: This is in the endoplasmic reticulum. C: Some reductases do so but not this one. D: Only  certain reactions catalyzed by the enzyme do. E: There are NADH­dependent reductases but they are different enzymes (p. 988). 6. C This enzyme reduces desaturase. A: Desaturase does not react with cytochrome P450. B: Desaturase is in the endoplasmic reticulum. D: This is a mitochondrial  system. E: This is not one of the two enzymes that use both flavins (p. 988). 7. B The iron–sulfur protein is adrenodoxin, which passes the electron to cytochrome P450. A and C: This is a mitochondrial enzyme. D and E: See B (p. 989). 8. E See Figure 23.7, p. 989. 9. C This is a mitochondrial process (Figure 23.8, p. 990). A and B: Hormone synthesis involves a series of reactions that move back and forth between mitochondria  and endoplasmic reticulum (p. 991). D and E: Removal of side chains frequently begins with oxidation reactions (p. 992). 10. B The types of xenobiotics oxidized by cytochrome P450 are usually highly lipophilic but must be excreted in the aqueous urine or bile. A: This may happen but is  certainly not the purpose. C: They do that prior to oxidation. D: Oxidation tends to reduce pharmacological activity (p. 992). 11. E A: It is not uncommon for xenobiotics to induce synthesis of something that will enhance their own metabolism. B and C: Epoxidation is the first step in the  conversion of this compound to one that is carcinogenic—again, a common occurrence (p. 994). D: Benzo[a]pyrene, with its four fused benzene rings, is highly  hydrophobic; introducing oxygens increases water solubility (p. 993). 12. B If warfarin is metabolized and cleared more rapidly by cytochrome P450, its therapeutic efficiency is decreased. Therefore, at the same dosage, it will be less  effective as an anticoagulant. Think what would happen if the warfarin dosage were adjusted for a proper response, and then phenobarbital were withdrawn without  adjusting the warfarin dose (see Clin. Corr. 23.1). 13. C The other product is citrulline. A and E: Three isoforms of nitric oxide synthase have been identified. B: One of the isoforms of NO synthase has been found in  macrophages but neuronal and endothelial isoforms also exist. D: Nitric oxide is a vasodilator, which is the basis for the use of nitroglycerin in angina pectoris (p. 995). 14. D This is the second mammalian enzyme known to use both FAD and FMN. A: The reaction is a monooxygenation. B: BH4 is required but the action of the  enzyme is similar to a cytochrome P450­mediated system. C: The donor is NADPH. E: The system requires Ca2+–calmodulin, at least the neuronal and endothelial  isoforms (p. 996).

 

  

Page 1001

Chapter 24—  Iron and Heme Metabolism William M. Awad, Jr.

24.1 Iron Metabolism: Overview

1002

24.2 Iron­Containing Proteins

1003

Transferrin Transports Iron in Serum

1003

Lactoferrin Binds Iron in Milk

1003

Ferritin Is a Protein Involved in Storage of Iron

1004

Other Nonheme Iron­Containing Proteins Are Involved in Enzymatic  Processes

1004

24.3 Intestinal Absorption of Iron

1005

24.4 Molecular Regulation of Iron Utilization

1006

24.5 Iron Distribution and Kinetics

1007

24.6 Heme Biosynthesis

1009

Enzymes in Heme Biosynthesis Occur in Both Mitochondria and  Cytosol

1011

Aminolevulinic Acid Synthase

1011

ALA Dehydratase

1012

Porphobilinogen Deaminase

1013

Uroporphyrinogen Decarboxylase

1014

Coproporphyrinogen Oxidase

1015

Protoporphyrinogen Oxidase

1016

Ferrochelatase

1016

ALA Synthase Catalyzes Rate­Limiting Step of Heme Biosynthesis

1017

24.7 Heme Catabolism Bilirubin Is Conjugated to Form Bilirubin Diglucuronide in Liver

1018

Intravascular Hemolysis Requires Scavenging of Iron

1020

Bibliography

1021

Questions and Answers

1022

Clinical Correlations

   

  

1017

 

24.1 Iron Overload and Infection

1003

24.2 Duodenal Iron Absorption

1005

24.3 Mutant Iron­Responsive Element

1007

24.4 Ceruloplasmin Deficiency

1008

24.5 Iron­Deficiency Anemia

1009

24.6 Hemochromatosis: Molecular Genetics and the Issue of Iron­ Fortified Diets

1011

24.7 Acute Intermittent Porphyria

1013

24.8 Neonatal Isoimmune Hemolysis

1020

24.9 Bilirubin UDP­Glucuronosyltransferase Deficiency

1020

24.10 Elevation of Serum Conjugated Bilirubin

1021

Page 1002

24.1—  Iron Metabolism: Overview Iron is closely involved in the metabolism of oxygen, permitting the transportation and participation of oxygen in a variety of biochemical processes. The common  oxidation states are either ferrous (Fe2+) or ferric (Fe3+); higher oxidation levels occur as short­lived intermediates in certain redox processes. Iron has an affinity  for electronegative atoms such as oxygen, nitrogen, and sulfur, which provide the electrons that form the bonds with iron. These can be of very high affinity when  favorably oriented on macromolecules. In forming complexes, no bonding electrons are derived from iron. There is an added complexity to the structure of iron: the  nonbonding electrons in the outer shell of the metal (the incompletely filled 3d orbitals) can exist in two states. Where bonding interactions with iron are weak, the outer  nonbonding electrons will avoid pairing and distribute throughout the 3d orbitals. Where bonding electrons interact strongly with iron, however, there will be pairing of  the outer nonbonding electrons, favoring lower­energy 3d orbitals. These two different distributions for each oxidation state of iron can be determined by electron spin  resonance measurements. Dispersion of 3d electrons to all orbitals leads to the high­spin state, whereas restriction of 3d electrons to lower energy orbitals, because of  electron pairing, leads to a low­spin state. Some iron–protein complexes reveal changes in spin state without changes in oxidation during chemical events (e.g., binding  and release of oxygen by hemoglobin). At neutral and alkaline pH ranges, the redox potential for iron in aqueous solutions favors the Fe3+ state; at acid pH values, the equilibrium favors the Fe2+ state. In the  Fe3+ state iron slowly forms large polynuclear complexes with hydroxide ion, water, and other anions that may be present. These complexes can become so large as to  exceed their solubility products, leading to their aggregation and precipitation with pathological consequences. Iron can bind to and influence the structure and function of various macromolecules, with deleterious results to the organism. To protect against such reactions, several  iron­binding proteins function specifically to store and transport iron. These proteins have both a very high affinity for the metal and, in the normal physiological state,  also have incompletely filled iron­binding sites. The interaction of iron with its ligands has been well characterized in some proteins (e.g., hemoglobin and myoglobin),  whereas for others (e.g., transferrin) it is presently in the process of being defined. The major area of ignorance in the biochemistry of iron lies in the in vivo transfer  processes of iron from one macromolecule to another. Several proposed mechanisms may explain the process of iron transfer. Two are supported by excellent model  studies but have varying degrees of relevance to the physiological state. The proposed processes are the following. First, the redox change of iron has been an  attractive mechanism because it is supported by selective in vitro studies and because in some cases macromolecules have a very selective affinity for Fe3+, binding  Fe2+ poorly. Thus reduction of iron would permit ferrous ions to dissociate, and reoxidation would allow the iron to redistribute to appropriate macromolecules. Redox  mechanisms have only been defined in a very few settings, some of which will be described below. An alternative hypothesis involves chelation of ferric ions by specific  small molecules with high affinities for iron; this mechanism has been supported also by selective in vitro studies. The chelation mechanism suffers from the lack of a  demonstrably specific in vivo chelator. Because the redox potential strongly favors ferric ion at almost all tissue sites and because Fe3+ binds so strongly to liganding  groups, the probability is that there are cooperating mechanisms regulating the intermolecular transfer of iron.

 

  

Page 1003

CLINICAL CORRELATION 24.1  Iron Overload and Infection If an individual is overloaded with iron by any of several causes, the serum transferrin  value can be close to saturation, making small amounts of free serum iron available.  Microorganisms that are usually nonpathogenic, because they are iron dependent and  cannot compete against partially saturated transferrin in the normal individual, can now  become pathogenic under these circumstances. For example, Vibrio vulnificus, a marine  halophile, is found in a small percentage of oysters and commercial shellfish. Individuals  who are iron overloaded can develop a rapidly progressive infection, with death ensuing  within 24 h after ingestion of the offending meal, whereas normal individuals consuming  the same food are entirely free of symptoms. Muench, K. H. Hemochromatosis and infection: alcohol and iron, oysters and sepsis.  Am. J. Med. 87:3, 1989.   24.2—  Iron­Containing Proteins Iron binds to proteins either by incorporation into a protoporphyrin IX ring (see below) or by interaction with other protein ligands. Ferrous­ and ferric­ protoporphyrin IX complexes are designated heme and hematin, respectively. Heme­containing proteins include those that transport (e.g., hemoglobin) and store  (e.g., myoglobin) oxygen, and certain enzymes that contain heme as part of their prosthetic groups (e.g., catalase, peroxidases, tryptophan pyrrolase, prostaglandin  synthase, guanylate cyclase, NO synthase, and the microsomal and mitochondrial cytochromes.). Discussions on structure–function relationships of heme proteins are  presented in Chapters 6 and 25. Nonheme proteins include transferrin, ferritin, a variety of redox enzymes that contain iron at the active site, and iron–sulfur proteins. A significant body of information  has been acquired that relates to the structure–function relationships of some of these molecules. Transferrin Transports Iron in Serum The protein in serum involved in the transport of iron is transferrin, a b 1­glycoprotein synthesized in the liver, consisting of a single polypeptide chain of 78,000 Da  with two noncooperative iron­binding sites. The protein is a product of gene duplication derived from a putative ancestral gene coding for a protein binding only one  atom of iron. Several metals bind to transferrin; the highest affinity is for Fe3+; Fe2+ ion is not bound. The binding of each Fe3+ ion is absolutely dependent on the  coordinate binding of an anion, which in the physiological state is carbonate as indicated below:

Estimates of the association constants for the binding of Fe3+ to transferrins from different species range from 1019 to 1031 M–1, indicating for practical purposes that  wherever there is excess transferrin free ferric ions will not be found. In the normal physiological state, approximately one­ninth of all transferrin molecules are saturated  with iron at both sites; four­ninths of transferrin molecules have iron at either site; and four­ninths of circulating transferrin are free of iron. Unsaturated transferrin  protects against infections (see Clin. Corr. 24.1). The two iron­binding sites show differences in sequences and in affinities for other metals. Transferrin binds to  specific cell surface receptors that mediate the internalization of the protein. The transferrin receptor is a transmembrane protein consisting of two subunits of 90,000 Da each, joined by a disulfide bond. Each subunit contains one  transmembrane segment and about 670 residues that are extracellular and bind a transferrin molecule, favoring the diferric form. Internalization of the receptor­ transferrin complex is dependent on receptor phosphorylation by a Ca2+–calmodulin–protein kinase C complex. Release of the iron atoms occurs within the acidic  milieu of the lysosome after which the receptor–apotransferrin complex returns to the cell surface where the apotransferrin is released to be reutilized in the plasma. Lactoferrin Binds Iron in Milk Milk contains iron that is bound almost exclusively to a glycoprotein, lactoferrin, closely homologous to transferrin, with two sites binding the metal. The iron content  of the protein varies, but it is never saturated. Studies on the function of lactoferrin have been directed toward its antimicrobial effect, protecting the newborn from  gastrointestinal infections. Microorganisms require iron for

 

  

Page 1004

replication and function. Presence of incompletely saturated lactoferrin results in the rapid binding of any free iron, leading to the inhibition of microbial growth by  preventing a sufficient amount of iron from entering these microorganisms. Other microbes, such as Escherichia coli, which release competitive iron chelators, are able  to proliferate despite the presence of lactoferrin, since the chelators transfer the iron specifically to the microorganism. Lactoferrin is present in granulocytes being  released during bacterial infections. It is also present in mucous secretions. Besides its bacteriostatic function it is believed to facilitate iron transport and storage in milk.  Lactoferrin has been found in urine of premature infants fed human milk. Ferritin Is a Protein Involved in Storage of Iron Ferritin is the major protein involved in the storage of iron. The protein consists of an outer polypeptide shell 130 Å in diameter with a central ferric­hydroxide­ phosphate core 60 Å across. The apoprotein, apoferritin, consists of 24 subunits of a varying mixture of H subunits (178 amino acids) and L subunits (171 amino  acids) that provide various isoprotein forms. H subunits predominate in nucleated blood cells and heart, L subunits in liver and spleen. Synthesis of the subunits is  regulated mainly by the concentration of free intracellular iron. The bulk of iron storage occurs in hepatocytes, reticuloendothelial cells, and skeletal muscle. The ratio of  iron to polypeptide is not constant, since the protein has the ability to gain and release iron according to physiological needs. With a capacity of 4500 iron atoms, the  molecule contains usually less than 3000. Channels from the surface permit the accumulation and release of iron. When iron is in excess, the storage capacity of newly  synthesized apoferritin may be exceeded. This leads to iron deposition adjacent to ferritin spheres. Histologically, such amorphous iron deposition is called  hemosiderin. The H chains of ferritin oxidize ferrous ions to the ferric state. Ferritins derived from different tissues of the same species differ in electrophoretic mobility  in a fashion analogous to the differences noted with isoenzymes. In some tissues ferritin spheres form lattice­like arrays, which are identifiable by electron microscopy.  Plasma ferritin (low in iron, rich in L subunits) has a half­life of 50 h and is cleared by reticuloendothelial cells and hepatocytes, and its concentration, although very  low, correlates closely to the size of the body iron stores.

Figure 24.1  Structure of ferredoxins.  Dark red circles represent iron atoms; light red  circles represent the inorganic sulfur  atoms; and small gray circles represent the  cysteinyl sulfur atoms derived from the  polypeptide chain. Variation in type IV  ferredoxins can occur where one of the  cysteinyl residues can be substituted by a  solvent oxygen atom of an OH group.

Other Nonheme Iron­Containing Proteins Are Involved in Enzymatic Processes Many iron­containing proteins are involved in enzymatic processes, most of which are related to oxidation mechanisms. The structural features of the ligands binding  the iron are not well known, except for a few components involved in mitochondrial electron transport. These latter proteins, termed ferredoxins, are characterized by  iron being bonded, with one exception, only to sulfur atoms. Four major types of such proteins are known (see Figure 24.1). The smallest, type I (e.g., nebredoxin),  found only in microorganisms, consists of a small polypeptide chain with a mass of about 6000 and contains one iron atom bound to four cysteine residues. Type II  consists of ferredoxins found in both plants and animal tissues where two iron atoms are found, each liganding to two separate cysteine residues and sharing two sulfide  anions. The most complicated of the iron–sulfur proteins are the bacterial ferredoxins, type III, which contain four atoms of iron, each of which is linked to single  separate cysteine residues but also shares three sulfide anions with neighboring iron molecules to form a cube­like structure. In some anaerobic bacteria, a family of  ferredoxins may contain two type III iron–sulfur groups per macromolecule. Type IV ferredoxins contain structures with three atoms of iron, each linked to two  separate cysteine residues and each sharing two sulfide anions, forming a

 

  

Page 1005

CLINICAL CORRELATION 24.2  Duodenal Iron Absorption Mucin in the duodenal lumen helps to solubilize ferric ions with presentation of the metal  to an integrin, a transmembrane protein consisting of a heterodimer of 230 kDa. The  cytosolic surface of the integrin interacts with a 56­kDa protein known as mobilferrin. The  integrin transfers the iron from the luminal to cytoplasmic surface of the cell, where it is  bound by mobilferrin. Mobilferrin acts as a cytosolic shuttle, transferring iron either to  cytosolic ferritin or to the opposite pole of the duodenal cell where the iron is transported  by an as yet undefined mechanism to capillaries to be picked up by transferrin. Conrad, M. D., and Umbreit, J. N. Iron absorption—the mucin–mobilferrin–integrin  pathway. A competitive pathway for metal absorption. Am. J. Hematol. 42:67, 1993.   planar ring. In one example of this ferredoxin type, an exception of iron atoms being liganded only to sulfur atoms was found where the sulfur of a cysteinyl residue was  substituted by a solvent oxygen atom. The redox potential afforded by these different ferredoxins varies widely and is in part dependent on the environment of the  surrounding polypeptide chain that envelops these iron–sulfur groups. In nebredoxin the iron undergoes ferric–ferrous conversion during electron transport. With the  plant and animal ferredoxins (type II iron–sulfur proteins) both irons are in the Fe3+ form in the oxidized state; upon reduction only one iron goes to Fe2+. In the  bacterial ferredoxin (type III iron–sulfur protein) the oxidized state can be either 2 Fe3+ ∙ 2 Fe2+ or 3 Fe3+ ∙ Fe2+, with corresponding reduced forms of Fe3+ ∙ 3 Fe2+ or  2 Fe3+ ∙ 2 Fe2+. 24.3—  Intestinal Absorption of Iron The high affinity of iron for both specific and nonspecific macromolecules leads to the absence of significant formation of free iron salts, and thus this metal is not lost  via usual excretory routes. Rather, excretion of iron occurs only through the normal sloughing of tissues that are not reutilized (e.g., epidermis and gastrointestinal  mucosal cells). In the healthy adult male the loss is about 1 mg day–1. In premenopausal women, the normal physiological events of menses and parturition substantially  augment iron loss. A wide variation of such loss exists, depending on the amounts of menstrual flow and the multiplicity of births. In the extremes of the latter settings, a  premenopausal woman may require an amount of iron that is four to five times that needed in an adult male for prolonged periods of time. The postmenopausal woman  who is not iron­deficient has an iron requirement similar to that of the adult male. Children and patients with blood loss naturally have increased iron requirements. Cooking of food facilitates the breakdown of ligands attached to iron, increasing the availability of the metal in the gut. The low pH of stomach contents permits the  reduction of Fe3+ to Fe2+, facilitating dissociation from ligands. The latter requires the presence of an accompanying reductant, which is usually achieved by adding  ascorbate to the diet. The absence of a normally functioning stomach reduces substantially the amount of iron that is absorbed. Some iron­containing compounds bind  the metal so tightly that it is not available for assimilation. Contrary to popular belief, spinach is a poor source of iron because of an earlier erroneous record of the iron  content and because some of the iron is bound to phytate (inositol hexaphosphate), which is resistant to the chemical actions of the gastrointestinal tract. Specific  protein cofactors derived from the stomach or pancreas have been suggested as being facilitators of iron absorption in the small intestine. The major site of absorption of iron is in the small intestine, with the largest amount being absorbed in the duodenum and a gradient of lesser absorption occurring in  the more distal portions of the small intestine. The metal enters the mucosal cell either as the free ion or as heme; in the latter case the metal is split off from the  porphyrin ring in the mucosal cytoplasm. The large amount of bicarbonate secreted by the pancreas neutralizes the acidic material delivered by the stomach and thus  favors the oxidation of Fe2+ to Fe3+. The major barrier to the absorption of iron is not at the luminal surface of the duodenal mucosal cell. Whatever the requirements of  the host are, in the face of an adequate delivery of iron to the lumen, a substantial amount of iron will enter the mucosal cell. Regulation of iron transfer occurs between  the mucosal cell and the capillary bed (see Figure 24.2 and Clin. Corr. 24.2). In the normal state, certain processes define the amount of iron that will be transferred.  Where there is iron deficiency, the amount of transfer increases; where there is iron overload in the host, the amount transferred is curtailed substantially. One  mechanism that has been demonstrated to regulate this transfer of iron across the mucosal–capillary

 

  

Page 1006

Figure 24.2  Intestinal mucosal regulation of iron absorption.  The flux of iron in the duodenal mucosal cell is indicated. A fraction of the iron  that is potentially acceptable is transferred from the intestinal lumen  into the epithelial cell. A large portion of ingested iron is not absorbed,  in part because it is not presented in a readily acceptable form. Some  iron is retained within the cell, bound by apoferritin to form ferritin. This  iron is sloughed into the intestinal lumen with the normal turnover of  the cell. A portion of the iron within the mucosal cell is absorbed and  transferred to the capillary bed to be incorporated into transferrin.  During cell division, which occurs at the bases of the intestinal  crypts, iron is incorporated for cellular requirements. These fluxes  change dramatically in iron­depleted or iron­excess states.

interface is the synthesis of apoferritin by the mucosal cell. In situations in which little iron is required by the host, a large amount of apoferritin is synthesized to trap the  iron within the mucosal cell and prevent transfer to the capillary bed. As the cells turn over (within a week), their contents are extruded into the intestinal lumen without  absorption occurring. In situations in which there is iron deficiency, virtually no apoferritin is synthesized so as not to compete against the transfer of iron to the deficient  host. There are other as yet undefined positive mechanisms that increase the rate of iron absorption in the iron­deficient state. Iron transferred to the capillaries is  trapped exclusively by transferrin. 24.4—  Molecular Regulation of Iron Utilization Cytosol contains at least two proteins that respond to changes in iron concentration. They act as effector molecules controlling the translation of mRNAs, which are  important in iron metabolism. These iron regulatory proteins (IRPs) bind to specific stem–loop structures on certain mRNAs. IRP­1 is the best defined of these  proteins. It contains an Fe4S4 cubane group when the cellular concentration of iron is high. This prosthetic group activates IRP­1 so that it possesses an aconitase  activity. However, since neither citrate nor isocitrate is present in significant amounts in the cytosol, the activity is only a potential

Figure 24.3  Iron­responsive protein­1.  Dark blue circles represent iron atoms and open circles inorganic sulfur atoms.

 

  

Page 1007

Figure 24.4  Structure of transferrin receptor mRNA.

CLINICAL CORRELATION 24.3  Mutant Iron­Responsive Element Single mutations have been described of two adjacent bases in the loop segment of the  iron­responsive element of ferritin light chain mRNA with an increased amount of  apoferritin being synthesized but without an increase in total body iron. This mutation  leads to a 28­fold lower affinity for IRP­1 in one case and perhaps an even lower affinity  in the other. The reason why these patients have cataracts is unknown. The gene for MP­ 19, an abundant protein in the lens, which is very close to the light chain gene on  chrosomome 19, might possibly be affected by the regulatory process on the mRNA.  However, it is more probable that a greatly increased synthesis of ferritin in the lens leads  to an increased amount of iron­catalyzed reactions with well­described oxidative  lenticular damage. Girelli, D., Corrocher, R., Bisceglia, L., et al. Molecular basis for the recently described  hereditary hyperferritinemia–cataract syndrome: a mutation in the iron­responsive  elements of ferritin L­subunit gene (the ''Verona mutation"). Blood 86:4050, 1995; and  Beaumont, C., Leneuve, P., Devaux, I., Scoazec, J. Y., et al. Mutation in the iron  responsive element of the L ferritin mRNA in a family with dominant hyperferritinaemia  and cataract. Nature Genet. 11:444, 1995.   one. At low iron concentrations, the cubane structure collapses, dissociating from the protein and leaving an apoenzyme without catalytic activity. However, it can now  bind to specific mRNA stem–loop structures, known as iron­responsive elements (IREs) (Figure 24.3). Five mRNAs are known to contain IREs: those for the  light and heavy chains of ferritin, the erythrocytic form of amino­levulinic acid synthase, the mitochondrial form of aconitase, and transferrin receptor. (Mitochondrial  aconitase, the physiologically active isozyme, has no IRP function.) The first four mRNAs have single IREs in the 5  flanking region, which bind a single IRP. In  contrast, the transferrin receptor has five tandem IREs that bind IRPs in the 3  flanking region. The binding of the 5  and 3  flanking IREs leads to different translational  effects. In the iron­deprived state, binding to the 3  IRE of transferrin receptor (Figure 24.4) leads to stabilization of the mRNA with reduced turnover and, therefore,  an increased number of receptor­specific RNA molecules, thereby leading to the increased synthesis of receptor protein. The single 5  stem–loop of ferritin mRNA  (Figure 24.5) is homologous to the 3  stem–loops of the transferrin receptor mRNA. However, in the former case, binding of the IRP leads to a decreased rate of  translation of the mRNA and, thereby, to a decreased concentration of ferritin molecules. Note that the molecular events that are controlled are different in the  syntheses of transferrin receptor and apoferritin (see Clin. Corr. 24.3). In summary, low iron concentrations lead to activation of an IRP that binds to the mRNAs for transferrin receptor and ferritin. In the former case, more receptor is  synthesized, while in the latter case less apoferritin is synthesized. The net effect is utilization of iron by proliferating cells. In contrast, high iron concentrations lead to  loss of binding by the IRPs to IREs, with a shift of iron from uptake by proliferating cells to storage in the liver. IRP­1 is regulated by its change from active to inactive states in mRNA­binding properties as noted above. IRP­2, a second regulatory protein, also responds to  varying concentrations of iron, but in this case, the protein is regulated by increased synthesis at low iron concentrations and increased degradation by a proteasome at  high iron concentrations. In addition to the effects of changed iron concentration, increased production of NO (see p. 995) also acts to regulate IRPs. 24.5—  Iron Distribution and Kinetics A normal 70­kg male has 3–4 g of iron, of which only 0.1% (3.5 mg) is in the plasma. Approximately 2.5 g are in hemoglobin. Table 24.1 lists the distribution

Figure 24.5  Structure of apoferritin H­subunit mRNA.

 

  

Page 1008

CLINICAL CORRELATION 24.4  Ceruloplasmin Deficiency A deficiency of ceruloplasmin, a copper­containing protein, but not its absence, is  associated with Wilson's disease in which there is progressive hepatic failure and  degeneration of the basal ganglia, associated with a characteristic copper deposition in the  cornea (Kayser–Fleischer rings). Because there was no evidence for significant  impairment of mobilization of iron in Wilson's disease, it was originally thought that the  ferroxidase activity of ceruloplasmin was not physiologically important. However, a  recently discovered very rare genetic defect in ceruloplasmin biosynthesis, where the  protein was virtually absent in serum, leads to a marked elevation of liver­iron content and  serum ferritin levels. These patients develop diabetes, retinal degeneration, and central  nervous system findings. The diabetes and central nervous system findings are associated  with increased iron in the pancreas and brain, respectively. Thus, in contrast to earlier  considerations, it appears that ceruloplasmin has a significant role in iron metabolism. Harris, E. D. The iron–copper connection: the link to ceruloplasmin grows stronger.  Nutr. Rev. 53:226, 1995.   of iron in humans. Normally about 33% of the sites on transferrin contain iron. Iron picked up from the intestine is delivered primarily to the marrow for incorporation  into the hemoglobin of red blood cells. The mobilization of iron from the mucosa and from storage sites involves in part the reduction of iron to the ferrous state and its  reoxidation to the ferric form. The reduction mechanisms have not been well described. On the other hand, conversion of the Fe2+ back to Fe3+ state is regulated by  serum enzymes called ferroxidases as indicated below:

TABLE 24.1 Approximate Iron  Distribution: 70–kg Man g

 

%

Hemoglobin

2.5

  68

Myoglobin

0.15

    4

Transferrin

0.003

    0.1

Ferritin, tissue

1.0

  27

Ferritin, serum

0.0001

    0.004

Enzymes

0.02

    0.6

Total

3.7

100

  Ferroxidase I is also known as ceruloplasmin (see Clin. Corr. 24.4). Another serum protein, ferroxidase II, appears to be the major serum component that  oxidizes ferrous ions. In any disease process in which iron loss exceeds iron repletion, a sequence of physiological responses occurs. The initial events are without  symptoms to the subject and involve depletion of iron stores without compromise of any physiological function. This depletion will be manifested by a reduction or  absence of iron stores in the liver and in the bone marrow and also by a decrease in the content of the very small amount of ferritin that is normally present in plasma.  Serum ferritin levels reflect slow release from storage sites during the normal cellular turnover that occurs in the liver; measurements are made by radioimmunoassays.  Serum ferritin is mostly apoferritin in form, containing very little iron. During this early phase, the level and percentage saturation of serum transferrin are not distinctly  abnormal. As the iron deficiency progresses, the level of hemoglobin begins to fall and morphological changes appear in the red blood cells. Concurrently, the serum  iron falls with a rise in the level of total serum transferrin, the latter reflecting a physiological adaptation in an attempt to absorb more iron from the gastrointestinal tract.  At this state of iron depletion a very sensitive index is the percentage saturation of serum transferrin with iron (normal range, 21–50%). At this point the patient usually  comes to medical attention, and the diagnosis of iron deficiency is made. In countries in which iron deficiency is severe without available corrective medical measures, a  third and severe stage of iron deficiency can occur, where a depletion of iron­containing enzymes leads to very pronounced metabolic effects (see Clin. Corr. 24.5). Iron overload can occur in patients so that the iron content of the body can be elevated to values as high as 100 g. This may happen for a variety of reasons. Some  patients have a recessive heritable disorder associated with a marked inappropriate increase in iron absorption. In such cases the serum transferrin can be almost  completely saturated with iron. This state, which is known as idiopathic hemochromatosis, is more commonly seen in men because women with the abnormal gene  are protected somewhat by menstrual and childbearing events. The accumulation of iron in the liver, pancreas, and heart can lead to cirrhosis and liver tumors, diabetes  mellitus, and cardiac failure, respectively. Treatment for these patients is periodic withdrawals of large amounts of blood, where the iron is contained in the hemoglobin.  Another group of patients has severe anemias, among the most common of which are the thalassemias, a group of hereditary hemolytic anemias. In these cases the  subjects require transfusions throughout their lives, leading to the accumulation of large amounts of iron derived from the transfused blood. Clearly bleeding would be  an inappropriate measure in these cases; rather, the patients are treated by the administration of iron chelators, such as desferrioxamine, which leads to the excretion of  large amounts of complexed iron in the urine. Rarely, a third group of patients will acquire excess iron because they ingest large amounts of both iron and ethanol, the  latter promoting iron absorption. In these cases excess stored iron can be removed by bleeding (see Clin. Corr. 24.6).

 

  

Page 1009

CLINICAL CORRELATION 24.5  Iron­Deficiency Anemia Microscopic examination of a blood smear in patients with iron­deficiency anemia usually  reveals the characteristic findings of microcytic (small in size) and hypochromic  (underpigmented) red blood cells. These changes in the red cell result from decreased  rates of globin synthesis when heme is not available. A bone marrow aspiration will reveal  no storage iron to be present and serum ferritin values are virtually zero. The serum  transferrin value (expressed as the total iron­binding capacity) will be elevated (upper  limits of normal: 410  g dL–1) with a serum iron saturation of less than 16%. Common  causes for iron deficiency include excessive menstrual flow, multiple births, and  gastrointestinal bleeding that may be occult. The common causes of gastrointestinal  bleeding include medications that can cause ulcers or erosion of the gastric mucosa  (especially aspirin or cortisone­like drugs), hiatal hernia, peptic ulcer disease, gastritis  associated with chronic alcoholism, and gastrointestinal tumor. The management of such  patients must include both a careful examination for the cause and source of bleeding and  supplementation with iron. The latter is usually provided in the form of oral ferrous sulfate  tablets; occasionally, intravenous iron therapy may be required. Where the iron deficiency  is severe, transfusion with packed red blood cells may also be indicated. Finch, C. A., and Huebers, H. Perspectives in iron metabolism. N. Engl. J. Med.  306:1520, 1982.   24.6—  Heme Biosynthesis Heme is produced in virtually all mammalian tissues. Its synthesis is most pronounced in the bone marrow and liver because of the requirements for incorporation into  hemoglobin and the cytochromes, respectively. As depicted in Figure 24.6, heme is largely a planar molecule. It consists of one ferrous ion and a tetrapyrrole ring,  protoporphyrin IX. The diameter of the iron atom is a little too large to be accommodated within the plane of the porphyrin ring, and thus the metal puckers out to  one side as it coordinates with the apical nitrogen atoms of the four pyrrole groups. Heme is one of the most stable compounds, reflecting its strong resonance features. Figure 24.7 depicts the pathway for heme biosynthesis. The following are the important aspects to be noted. First, the initial and last three enzymatic steps are  catalyzed by enzymes that are in the mitochondrion, whereas the intermediate steps take place in the cytoplasm. This is important in considering the regulation by heme  of the first biosynthetic step; this aspect is discussed below. Second, the organic portion of heme is derived totally from eight residues each of glycine and succinyl  CoA. Third, the reactions occurring on the side groups attached to the tetrapyrrole ring involve the colorless intermediates known as porphyrinogens. The latter  compounds, though exhibiting reso­

Figure 24.6  Structure of heme.

 

  

Page 1010

Figure 24.7  Pathway for heme biosynthesis.  Numbers indicate enzymes involved in each step as follows: 1, ALA  synthase; 2, ALA dehydratase; 3, porphobilinogen deaminase; 4, uroporphyrinogen III cosynthase; 5,  uroporphyrinogen decarboxylase; 6, coproporphyrinogen III oxidase; 7, protoporphyrinogen IX oxidase; 8,  ferrochelatase. Pyrrole ligands are indicated as follows: P, propionic (b­carboxyethyl); A,  acetic (carboxymethyl); M, methyl; V, = vinyl.

nance features within each pyrrole ring, do not demonstrate resonance between the pyrrole groups. As a consequence, the porphyrinogens are unstable and can  readily be oxidized, especially in the presence of light, by nonenzymatic means to their stable porphyrin products. In the latter cases resonance between pyrrole  groups is established by oxidation of the four methylene bridges. Figure 24.8 depicts the enzymatic conversion of protoporphyrinogen to protoporphyrin

 

  

Page 1011

CLINICAL CORRELATION 24.6  Hemochromatosis: Molecular Genetics and the Issue of Iron­Fortified Diets The hemochromatosis gene is heterozygous in about 9% of the population. The disease is  expressed primarily in the homozygous state; about 0.25% of all individuals are at risk.  Normal individuals have a major histocompatibility complex class­1 gene (HLA­H) with  unknown function that encodes for the a ­chain, containing three immunoglobulin­like  domains. The normal gene product has a structure that cannot present an antigen. Most  individuals with hemochromatosis are homozygous for a Cys282­Tyr mutation which  prevents the normal conformation of an immunoglobulin domain. A controversy has developed as to whether food should be fortified with iron because of  the prevalence of iron­deficiency anemia, especially among premenopausal women. It  was suggested that dietary iron deficiency would be reduced if at least 50 mg of iron was  incorporated per pound of enriched flour. Others suggested that toxicity from excess iron  absorption through iron fortification was too great. Sweden has mandated iron  fortification for 45 years and about 42% of the average daily intake of iron is derived  from these sources. However, 5% of males had elevation of serum iron values, with 2%  having iron stores consonant with the distribution found in early stages of  hemochromatosis, pointing out the danger of iron­fortified diets. In countries where iron  deficiency is widespread, however, fortification may still be the most appropriate  measure. McLaren, C. E., Gorddeuk, V. R., Looker, A. C., et al. Prevalence of heterozygotes for  hemochromatosis in the white population of the United States. Blood 86:2021, 1995;  Feder, J. N., Gnirki, A., Thomas, W., et al. A novel MHC class 1­like gene is mutated in  patients with hereditary haemochromatosis. Nature Genetics 13:399, 1996; Olsson, K.  S., Heedman, P. A., and Staugard, F. Preclinical hemochromatosis in a population on a  high­iron­fortified diet. J. Am. Med. Assoc. 239:1999, 1978; Olsson, K. S., Marsell, R.,  Ritter, B., Olander, B., et al. Iron deficiency and iron overload in Swedish male  adolescents. J. Intern. Med. 237:187, 1995.   by this oxidation mechanism. This is the only known porphyrinogen oxidation that is enzyme regulated in humans; all other porphyrinogen   porphyrin conversions are  nonenzymatic and catalyzed by light rather than catalyzed by specific enzymes. Fourth, once the tetrapyrrole ring is formed, the order of the R groups as one goes  clockwise around the tetrapyrrole ring defines which of the four possible types of uro­ or coproporphyrinogens are being synthesized. These latter compounds have  two different substituents, one each for every pyrrole group. Going clockwise around the ring, the substituents can be arranged as ABABABAB (where A is one  substituent and B the other), forming a type I porphyrinogen, or the arrangement can be ABABABBA, forming a type III porphyrinogen. In principle, two other  arrangements can occur to form porphyrinogens II and IV, and these can be synthesized chemically; however, they do not occur naturally. In protoporphyrinogen and  protoporphyrin there are three types of substituents, and the classification becomes more complicated; type IX is the only form that is synthesized naturally. Derangements of porphyrin metabolism are known clinically as the porphyrias. This family of diseases is of great interest because it has revealed that the regulation of  heme biosynthesis is complicated. The clinical presentations of the different porphyrias provide a fascinating exposition of biochemical regulatory abnormalities and  their relationship to pathophysiological processes. Table 24.2 lists the details of the different porphyrias (see Clin. Corr. 24.7). Enzymes in Heme Biosynthesis Occur in Both Mitochondria and Cytosol Aminolevulinic Acid Synthase Aminolevulinic acid (ALA) synthase controls the rate­limiting step of heme synthesis in all tissues studied. Synthesis of the enzyme is not directed by mitochondrial  DNA but occurs rather in the cytosol, being directed by mRNA derived from the nucleus. The enzyme is incorporated into the matrix of the mitochondrion. Succinyl  CoA is one of the substrates and is found only in the mitochondrion. This protein has been purified to homogeneity from rat liver mitochondria. The cytosolic protein is  a dimer of a 71,000­Da subunit, containing a basic N­terminal signaling sequence that directs the enzyme into the mitochondrion. An ATP­dependent 70,000­Da  cytostolic component, known as a chaperone protein, maintains ALA synthase in the unfolded extended state, the only form that can pass through the mitochondrial  membrane. Thereafter, the N­

 

  

Page 1012

Figure 24.8  Action of protoporphyrinogen IX oxidase, an example of the  conversion of a porphyrinogen to a porphyrin.

terminal signaling sequence is cleaved by a metal­dependent protease in the mitochondrial matrix, yielding an ALA synthase with subunits of 65,000 Da each. Within  the matrix another oligomeric chaperone protein, of 14 subunits of 60,000 Da each, catalyzes the correct folding of the protein in a second ATP­dependent process  (Figure 24.9, p. 1014). The ALA synthase has a short biological half­life (~60 min). Both the synthesis and activity of the enzyme are subject to regulation by a variety  of substances; 50% inhibition of activity occurs in the presence of 5 mM of hemin, and virtually complete inhibition is noted at a 20­mM concentration. The enzymatic  reaction involves the condensation of a glycine residue with a residue of succinyl CoA. The reaction has an absolute requirement for pyridoxal phosphate. Two  isoenzymes exist for ALA synthase; only the erythrocytic form contains an IRE. ALA Dehydratase Aminolevulinic acid dehydratase (280 Da) (or porphobilinogen synthase) is a cytosol component consisting of eight subunits, of which only four interact with the  substrate. This protein also interacts with the substrate to form a Schiff base, but in this case the  ­amino group of a lysine residue binds to the ketonic carbon of the  substrate molecule (Figure 24.10, p. 1015). Two molecules of TABLE 24.2 Derangements in Porphyrin Metabolism Disease State

  Genetics

  Tissue

Enzyme

    Activity

Organ Pathology

Acute intermittent porphyria Dominant

Liver

1. ALA synthase

Increase

Nervous system

 

 

 

2. Porphobilinogen deaminase

Decrease

 

 

 

 

3. 

Decrease

 

Hereditary coproporphyria

Dominant

Liver

1. ALA synthase

Increase

Nervous system; skin

 

 

 

2. Coproporphyrinogen  oxidase

Decrease

 

Variegate porphyria

Dominant

Liver

1. ALA synthase

Increase

Nervous system; skin

 

 

 

2. Protoporphyrinogen  oxidase

Decrease

 

Porphyria cutanea tarda

Dominant

Liver

1. Uroporphyrinogen  decarboxylase

Decrease

Skin, induced by liver disease

Hereditary protoporphyria

Dominant

Marrow

1. Ferrochelatase

Decrease

Gallstones, liver disease, skin

Erythropoietic porphyria

Recessive

Marrow

1. Uroporphyrinogen III  cosynthase

Decrease

Skin and appendages;  reticuloengothelial system

Lead poisoning

None

All tissues

1. ALA dehydrase

Decrease

 

 

 

Nervous system; blood; others

2. Ferrochelatase

Decrease

   

  

4­5a­Reductase

Page 1013

CLINICAL CORRELATION 24.7  Acute Intermittent Porphyria A 40­year­old woman appears in the emergency room in an agitated state, weeping and  complaining of severe abdominal pain. She has been constipated for several days and has  noted marked weakness in the arms and legs and that "things do not appear to be quite  right." Physical examination reveals a slightly rapid heart rate (100/min) and moderate  hypertension (blood pressure of 160/110 mmHg). There have been earlier episodes of  severe abdominal pain; operations undertaken on two occasions revealed no  abnormalities. The usual laboratory tests are normal. The neurological complaints are not  localized to an anatomical focus. The decision is made that the present symptoms are  largely psychiatric in origin and have a functional rather than an organic basis. The patient  is sedated with 60 mg of phenobarbital; a consultant psychiatrist agrees by telephone to  see the patient in about 4 h. The staff notices a marked deterioration; generalized  weakness rapidly appears, progressing to a compromise of respiratory function. This  ominous development leads to immediate incorporation of a ventilatory assistance  regimen, with transfer to intensive care for physiological monitoring. Her condition  deteriorates and she dies 48 h later. A urine sample of the patient is reported later to have  a markedly elevated level of porphobilinogen. This patient had acute intermittent  porphyria, a disease of incompletely understood derangement of heme biosynthesis.  There is a dominant pattern of inheritance associated with an overproduction of the  porphyrin precursors, ALA and porphobilinogen. Three enzyme abnormalities are noted  in the cases that have been studied carefully. These include (1) a marked increase in ALA  synthase, (2) a reduction by one­half of activity of porphobilinogen deaminase, and (3) a  reduction of one­half of the activity of steroid  4­5a ­reductase. The change in content of  the second enzyme is consonant with a dominant expression. The change in content of the  third enzyme is acquired and not apparently a heritable expression of the disease. It is  believed that a decrease in porphobilinogen deaminase leads to a minor decrease in  content of heme in liver. The lower concentration of heme leads to a failure both to  repress the synthesis and to inhibit the activity of ALA synthase. Almost never manifested  before puberty, the disease is thought to appear only with the induction of  4­5b ­ reductase at adolescence. Without a sufficient amount of  4­5a ­reductase, the observed  increase in the 5b  steroids is due to a shunting of  4 steroids into the 5b ­reductase  pathway. The importance of abnormalities of this last metabolic pathway in the  pathogenesis of porphyria is controversial. Pathophysiologically, the disease poses a great  riddle: the derangement of porphyrin metabolism is confined to the liver, which  anatomically appears normal, whereas the pathological findings are restricted to the  nervous system. In the present case, involvement of (1) the brain led to the agitated and  confused state and the respiratory collapse, (2) the autonomic system led to the  hypertension, increased heart rate, constipation, and abdominal pain, and (3) the  peripheral nervous system and spinal cord led to the weakness and sensory disturbances.  Experimentally, no known metabolic intermediate of heme biosynthesis can cause the  pathology noted in acute intermittent porphyria. There should have been a greater  suspicion of the possibility of porphyria early in the patient's presentation. The analysis of  porphobilinogen in the urine is a relatively simple test. The treatment would have been  glucose infusion, the exclusion of any drugs that could cause elevation of ALA synthase  (e.g., barbiturates), and, if her disease failed to respond satisfactorily despite these  measures, the administration of intravenous hematin to inhibit the synthesis and activity of  ALA synthase. Acute hepatic porphyria is of historic political interest. The disease has  been diagnosed in two descendants of King George III, suggesting that the latter's  deranged personality preceding and during the American Revolution could possibly be  ascribed to porphyria. Meyer, U. A., Strand, L. J., Doss, M., et al. Intermittent acute porphyria: demonstration  of a genetic defect in porphobilinogen metabolism. N. Engl. J. Med. 286:1277, 1972;  and Stein, J. A., and Tschudy, D. D. Acute intermittent porphyria: a clinical and  biochemical study of 46 patients. Medicine (Baltimore) 49:1, 1970.   ALA condense asymmetrically to form porphobilinogen. The ALA dehydratase is a sulfhydryl enzyme and is very sensitive to inhibition by heavy metals. A  characteristic finding of lead poisoning is the elevation of ALA in the absence of an elevation of porphobilinogen. Porphobilinogen Deaminase Synthesis of the porphyrin ring is a complicated process. A sulfhydryl group on porphobilinogen deaminase forms a thioether bond with a porphobilinogen residue  through a deamination reaction. Thereafter, five additional porphobilinogen residues are deaminated successively to form a linear hexapyrrole adduct with the enzyme.  The adduct is cleaved hydrolytically to form both an enzyme–dipyrromethane complex and the linear tetrapyrrole, hydroxymethylbilane. The enzyme–dipyrromethane  complex is then ready for another cycle of addition of four porphobilinogen residues to generate another tetrapyrrole. Thus dipyrromethane is the covalently attached  novel cofactor for the enzyme. Porphobilinogen deaminase has no ring­closing function; hydroxymethylbilane closes in an enzyme­independent step to form  uroporphyrinogen I if no additional factors are present. However, the deaminase is closely associated with a second protein,

 

  

Page 1014

Figure 24.9  Synthesis of  ­aminolevulinic acid synthase.

uroporphyrinogen III cosynthase, which directs the synthesis of the III isomer. The formation of the latter involves a spiro intermediate generated from  hydroxymethylbilane; this allows inversion of one of the pyrrole groups (Figure 24.11, p. 1016). In the absence of the cosynthase, uroporphyrinogen I is synthesized  slowly; in its presence, the III isomer is synthesized rapidly. A rare recessively inherited disease, erythropoietic porphyria, associated with marked cutaneous light  sensitization, is due to an abnormality of red blood cell cosynthase. Here, large amounts of the type I isomers of uroporphyrinogen and coproporphyrinogen are  synthesized in the bone marrow. Two isoenzymes exist for porphobilinogen deaminase due to alternative splicing of exon 1 or exon 2 to the rest of the mRNA. Uroporphyrinogen Decarboxylase This enzyme acts on the side chains of the uroporphyrinogens to form the coproporphyrinogens. The protein catalyzes the conversion of both I and III isomers of  uroporphyrinogen to the respective coproporphyrinogen isomers. Uroporphyrinogen decarboxylase is inhibited by iron salts. Clinically, the most common cause of  porphyrin derangement is associated with patients who have a single gene abnormality for this enzyme, leading to 50% depression of the enzyme's activity. This  disease, which shows cutaneous manifestations primarily with sensitivity to light, is known as porphyria cutanea tarda. The condition

 

  

Page 1015

Figure 24.10  Synthesis of porphobilinogen.

is not expressed unless patients either take drugs that cause an increase in porphyrin synthesis or drink large amounts of alcohol, leading to the accumulation of iron,  which then acts to inhibit further the activity of uroporphyrinogen decarboxylase. Coproporphyrinogen Oxidase This mitochondrial enzyme is specific for the type III isomer of coproporphyrinogen, not acting on the type I isomer. Coproporphyrinogen III enters the mitochondrion  and is converted to protoporphyrinogen IX. The mechanism of action is not understood. A dominant disease associated with a deficiency of this

 

  

Page 1016

Figure 24.11  Synthesis of uroporphyrinogens I and III.  Enzyme in blue is uroporphyrinogen I synthase.

enzyme leads to a form of hereditary hepatic porphyria, known as hereditary coproporphyria. Protoporphyrinogen Oxidase This mitochondrial enzyme generates a product, protoporphyrin IX, which, in contrast to the other heme precursors, is very water­insoluble. Excess amounts of  protoporphyrin IX that are not converted to heme are excreted by the biliary system into the intestinal tract. A dominant disease, variegate porphyria, is due to a  deficiency of protoporphyrinogen oxidase. Ferrochelatase Ferrochelatase inserts ferrous iron into protoporphyrin IX in the final step of the synthesis of heme. The protein is sensitive to the effects of heavy metals (especially  lead) and, of course, to iron deprivation. In these latter instances, zinc instead of iron is incorporated to form a zinc–protoporphyrin IX complex. In contrast to heme,  the zinc–protoporphyrin IX complex is brilliantly fluorescent and easily detectable in small amounts. The enzyme contains an Fe2S2 group and has been proposed as an  IRP­3 that controls translation of the erythrocytic ALA synthase mRNA.

 

  

Page 1017

ALA Synthase Catalyzes Rate­Limiting Step of Heme Biosynthesis ALA synthase controls the rate­limiting step of heme synthesis in all tissues. Succinyl CoA and glycine are substrates for a variety of reactions. The modulation of the  activity of ALA synthase determines the quantity of the substrates that will be shunted into heme biosynthesis. Heme (and also hematin) acts both as a repressor of the  synthesis of ALA synthase and as an inhibitor of its activity. Since heme resembles neither the substrates nor the product of the enzyme's action, it is probable that the  latter inhibition occurs at an allosteric site. Almost 100 different drugs and metabolites can cause induction of ALA synthase; for example, a 40­fold increase is noted in  the rat after treatment with 3,5­dicarbethoxy­1,4­dihydrocollidine. The effect of pharmacological agents has led to the important clinical feature where some patients  with certain kinds of porphyria have had exacerbations of their condition following the inappropriate administration of certain drugs (e.g., barbiturates). ALA  dehydratase is also inhibited by heme; but this is of little physiological consequence, since the activity of ALA dehydrase is about 80­fold greater than that of ALA  synthase, and thus heme­inhibitory effects are reflected first in the activity of ALA synthase. Glucose or a proximal metabolite serves to inhibit heme biosynthesis in a mechanism that is not yet defined. This is of clinical relevance, since some patients manifest  their porphyric state for the first time when placed on a very low caloric (and therefore glucose) intake. Other regulators of porphyrin metabolism include certain  steroids. Steroid hormones (e.g., oral contraceptive pills) with a double bond in ring A between C­4 and C­5 atoms can be reduced by two different reductases. The  product of 5a ­reduction has little effect on heme biosynthesis; however, the product of 5a ­reduction serves as a stimulus for the synthesis of ALA synthase. 24.7—  Heme Catabolism Catabolism of heme­containing proteins presents two requirements to the mammalian host: (1) development of a means of processing the hydrophobic products of  porphyrin ring cleavage and (2) retention and mobilization of the contained iron so that it may be reutilized. Red blood cells have a life span of approximately 120 days.  Senescent cells are recognized by their membrane changes and removed and engulfed by the reticuloendothelial system at extravascular sites. The globin chains  denature, releasing heme into the cytoplasm. The globin is degraded to its constituent amino acids, which are reutilized for general metabolic needs. Figure 24.12 depicts the events of heme catabolism. Heme is degraded primarily by a microsomal enzyme system in reticuloendothelial cells that requires molecular  oxygen and NADPH. Heme oxygenase is substrate inducible and catalyzes the cleavage of the a ­methene bridge, which joins the two pyrrole residues containing  the vinyl substituents. The a ­methene carbon is converted quantitatively to carbon monoxide. The only endogenous source of carbon monoxide in humans is the a ­ methene carbon. A fraction of the carbon monoxide is released via the respiratory tract. Thus the measurement of carbon monoxide in an exhaled breath provides an  index to the quantity of heme that is degraded in an individual. The oxygen present in the carbon monoxide and in the newly derivatized lactam rings are generated  entirely from molecular oxygen. The stoichiometry of the reaction requires 3 mol of oxygen for each ring cleavage. Heme oxygenase will only use heme as a substrate,  with the iron possibly participating in the cleavage mechanism. Thus free protoporphyrin IX is not a substrate. The linear tetrapyrrole biliverdin IX is the product  formed by the action of heme oxygenase. Biliverdin IX is reduced by biliverdin reductase to bilirubin IX.

Figure 24.12  Formation of bilirubin from heme.  Greek letters indicate the labeling of the  methene carbon atoms in heme.

 

  

Page 1018

Bilirubin Is Conjugated to Form Bilirubin Diglucuronide in Liver Bilirubin is derived not only from senescent red cells but also from the turnover of other heme­containing proteins, such as the cytochromes. Studies with labeled  glycine as a precursor have revealed that an early­labeled bilirubin, with a peak within 1–3 h, appears a very short time after a pulsed administration of the labeled  precursor. A larger amount of bilirubin appears much later at about 120 days, reflecting the turnover of heme in red blood cells. Early­labeled bilirubin can be divided  into two parts: an early–early part, which reflects the turnover of heme proteins in the liver, and a late–early part, which consists of both the turnover of heme­ containing hepatic proteins and the turnover of bone marrow heme, which is either poorly incorporated or easily released from red blood cells. The latter is a  measurement of ineffective erythropoiesis and can be very pronounced in disease states such as pernicious anemia (see Chapter 28) and the thalassemias. Bilirubin is poorly soluble in aqueous solutions at physiological pH values. When transported in plasma, it is bound to serum albumin with an association constant  greater than 106 M–1. Albumin contains one such high­affinity site and another with a lesser affinity. At the normal albumin concentration of 4 g dL–1, about 70 mg of  bilirubin per deciliter of plasma can be bound on the two sites. However, bilirubin toxicity (kernicterus), which is manifested by the transfer of bilirubin to membrane  lipids, commonly occurs at concentrations greater than 25 mg dL–1. This suggests that the weak affinity of the second site does not allow it to serve effectively in the  transport of bilirubin. Bilirubin on serum albumin is rapidly cleared by the liver, where there is a free bidirectional flux of the tetrapyrrole across the sinusoidal– hepatocyte interface. Once in the hepatocyte, bilirubin is bound to several cytostolic proteins, of which only one has been well characterized. The latter component,  ligandin, is a small basic component making up to 6% of the total cytosolic protein of rat liver. Ligandin has been purified to homogeneity from rat liver and  characterized as having two subunits with molecular masses of 22 kDa and 27 kDa. Each subunit contains glutathione S­epoxidetransferase activity, a function  important in detoxification mechanisms of aryl groups. The stoichiometry of binding is one bilirubin molecule per complete ligandin molecule. The functional role of  ligandin and other hepatic bilirubin­binding proteins remains to be defined. Once in the hepatocyte the propionyl side chains of bilirubin are conjugated to form a diglucuronide (Figure 24.13). The reaction utilizes uridine diphosphoglucuronate  derived from the oxidation of uridine diphosphoglucose. The former serves as a glucuronate donor to bilirubin. In normal bile, the diglucuronide is the major form of  excreted bilirubin, with only small amounts of the monoglucuronide or other glycosidic adducts present. Bilirubin diglucuronide is much more water­soluble than free  bilirubin, and thus the transferase facilitates excretion of the bilirubin into bile. Bilirubin diglucuronide is poorly absorbed by the intestinal mucosa. The glucuronide  residues are released in the terminal ileum and large intestine by bacterial hydrolases; the released free bilirubin is reduced to the colorless linear tetrapyrroles known as  urobilinogens. Urobilinogens can be oxidized to colored products known as urobilins, which are excreted in the feces. A small fraction of urobilinogen can be  reabsorbed by the terminal ileum and large intestine to be removed by hepatic cells and resecreted in bile. When urobilinogen is reabsorbed in large amounts in certain  disease states, the kidney serves as a major excretory site. In the normal state, plasma bilirubin concentrations are 0.3–1 mg dL–1, and this is almost all in the unconjugated state. In the clinical setting, conjugated bilirubin is  expressed as direct bilirubin because it can be coupled readily with diazonium salts to yield azo dyes; this is the direct van den Bergh reaction. Unconjugated  bilirubin is bound noncovalently to albumin and will not react until it is released by the addition of an organic solvent such as

 

  

Page 1019

Figure 24.13  Biosynthesis of bilirubin diglucuronide.

ethanol. The reaction with diazonium salts yielding the azo dye after the addition of ethanol is the indirect van den Bergh reaction, and this measures the indirect  bilirubin or the unconjugated bilirubin. Unconjugated bilirubin binds so tightly to serum albumin and lipid that it does not diffuse freely in plasma and therefore does not  lead to an elevation of bilirubin in the urine. Unconjugated bilirubin has a high affinity for membrane lipids, which leads to the impairment of cell membrane function,  especially in the nervous system. In contrast, conjugated bilirubin is relatively water­soluble, and elevations of this bilirubin form lead to high urinary concentrations with  the characteristic deep yellow­brown color. The deposition of conjugated and unconjugated bilirubin in skin and the sclera gives the yellow to yellow­green color seen  in patients with jaundice. A third form of plasma bilirubin occurs only with hepatocellular disease in which a fraction of the bilirubin binds so tightly that it is not released from serum albumin by  the usual techniques and is linked covalently to the protein. In some cases up to 90% of total bilirubin can be in this covalently bound form. The normal liver has a very large capacity to conjugate and mobilize the bilirubin that is delivered. As a consequence, hyperbilirubinemia due to excess heme  destruction, as in hemolytic diseases, rarely leads to bilirubin levels that exceed 5 mg dL–1, except in situations in which functional derangement of the liver is present  (see Clin. Corr. 24.8). Thus marked elevation of unconjugated bilirubin reflects primarily a variety of hepatic diseases, including those that are heritable and those that  are acquired (see Clin. Corr. 24.9). Elevations of conjugated bilirubin level in plasma are attributable to liver and/or biliary tract disease. In simple uncomplicated biliary tract obstruction, the major  component of the elevated serum bilirubin is the diglucuronide form, which is released by the liver into the vascular compartment. Biliary tract disease may be  extrahepatic or intrahepatic, the latter involving the canaliculi and biliary ductules (see Clin. Corr. 24.10).

 

  

Page 1020

CLINICAL CORRELATION 24.8  Neonatal Isoimmune Hemolysis Rh­negative women pregnant with Rh­positive fetuses will develop antibodies to Rh  factors. These antibodies will cross the placenta to hemolyze fetal red blood cells. Usually  this is not of clinical relevance until about the third Rh­positive pregnancy, in which the  mother has had antigenic challenges with earlier babies. Antenatal studies will reveal rising  maternal levels of IgG antibodies against Rh­positive red blood cells, indicating that the  fetus is Rh­positive. Before birth, placental transfer of fetal bilirubin occurs with excretion  through the maternal liver. Because hepatic enzymes of bilirubin metabolism are poorly  expressed in the newborn, infants may not be able to excrete the large amounts of  bilirubin that can be generated from red cell breakdown. At birth these infants usually  appear normal; however, the unconjugated bilirubin in the umbilical cord blood is elevated  up to 4 mg dL–1; due to the hemolysis initiated by maternal antibodies. During the next 2  days the serum bilirubin rises, reflecting continuing isoimmune hemolysis, leading to  jaundice, hepatosplenomegaly, ascites, and edema. If untreated, signs of central nervous  system damage can occur, with the appearance of lethargy, hypotonia, spasticity, and  respiratory difficulty, constituting the syndrome of kernicterus. Treatment involves  exchange transfusion with whole blood, which is serologically compatible with both the  infant's blood and maternal serum. The latter requirement is necessary to prevent  hemolysis of the transfused cells. Additional treatment includes external phototherapy,  which facilitates the breakdown of bilirubin. The entire problem can be prevented by  treating Rh­negative mothers with anti­Rh globulin. These antibodies recognize the fetal  red cells, block the Rh antigens, and cause them to be destroyed without stimulating an  immune response in the mothers. Mauer, H. M., Shumway, C. N., Draper, D. A., and Hossaini, A. A. Controlled trial  comparing agar, intermittent phototherapy, and continuous phototherapy for reducing  neonatal hyperbilirubinemia. J. Pediatr. 82:73, 1973; and Bowman, J. J. Management of  Rh­isoimmunization. Obstet. Gynecol. 52:1, 1978.   Intravascular Hemolysis Requires Scavenging of Iron In certain diseases destruction of red blood cells occurs in the intravascular compartment rather than in the extravascular reticuloendothelial cells. In the former case the  appearance of free hemoglobin and heme in the plasma potentially could lead to the excretion of these substances through the kidney with a substantial loss of iron. To  prevent this occurrence, specific plasma proteins are involved in scavenging mechanisms. Transferrin binds free iron and thus permits its reutilization. Free hemoglobin,  after oxygenation in the pulmonary capillaries, dissociates into a ,b  dimers, which are bound to a family of circulating CLINICAL CORRELATION 24.9  Bilirubin UDP­Glucuronosyltransferase Deficiency Bilirubin UDP­glucuronosyltransferase has two isoenzyme forms, derived from alternative  mRNA splicing between variable forms of exon 1 and common exons 2, 3, 4, and 5. The  latter exons define the part of the protein that binds the UDP­glucuronate, whereas the  various exons 1 have defined specificities for either bilirubin or other acceptors, such as  phenol. Two exons have bilirubin specificity leading to two forms of bilirubin UDP­ glucuronosyltransferase forms. Two major families of diseases are seen with deficiencies  of the enzyme. Crigler–Najjar syndrome is seen in infants and is associated with  extraordinarily high serum unconjugated bilirubin due to an autosomal recessive  inheritance of mutations on both alleles in exons 2, 3, 4, or 5. Gilbert's syndrome is also  associated with a deficiency of the enzyme's activity, but only to about 25% of normal.  The patients appear jaundiced but without other clinical symptoms. The major  complication is an exhaustive search by the physician looking for some serious liver  disease and failing to recognize the benign condition. Two different findings that may be  restricted to different populations account for the condition. In Japan a dominant pattern  of inheritance is noted with a mutation on only one allele. The 75% reduction of activity is  ascribed to the fact that the enzyme exists as an oligomer, where mutant and normal  monomers might associate to form heterooligomers. The explanation is that not only is the  mutant monomer inactive, but it forces conformational effects on the normal subunit,  reducing its activity substantially. In contrast, in the Western world the condition is due  largely to a homozygous expansion of the bases in the promoter region with less efficient  transcription of the gene. Aono, S., Adachi, Y., Uyama, S., et al. Analysis of genes for bilirubin UDP­ glucuronosyltransferase in Gilbert's syndrome. Lancet 345:958, 1995; and Bosma, P. J.,  Chowdhury, J. R., Bakker, C., et al. The genetic basis of the reduced expression of  bilirubin UDP­glucuronosyltransferase 1 in Gilberts syndrome. N. Engl. J. Med.  333:1171, 1995.    

  

Page 1021

CLINICAL CORRELATION 24.10  Elevation of Serum Conjugated Bilirubin Elevations of serum conjugated bilirubin are attributable to liver and/or biliary tract  disease. In simple uncomplicated biliary tract obstruction, the major component of the  elevated serum bilirubin is the diglucuronide form, which is released by the liver into the  vascular compartment. Biliary tract disease may be extrahepatic or intrahepatic, the latter  involving the canaliculi and biliary ductules. Dubin–Johnson syndrome is an autosomal  recessive disease involving a defect in the biliary secretory mechanisms in liver. Excretion  through the biliary tract of a variety of (but not all) organic anions is affected. Retention of  melanin­like pigment in the liver in this disorder leads to a characteristic gray­black color  of this organ. A second heritable disorder associated with elevated levels of serum  conjugated bilirubin is Rotor's syndrome. In this poorly defined disease no hepatic  pigmentation occurs. Kitamura, T., Alroy, J., Gatmaitan, Z., et al. Defective biliary excretion of epinephrine  metabolites in mutant (TR­) rats: relation to the pathogenesis of black liver in the Dubin– Johnson syndrome and Corriedale sheep with an analogous excretory defect.  Hepatology 15:1154, 1992.   plasma proteins, the haptoglobins, having a high affinity for the oxyhemoglobin dimer. Since deoxyhemoglobin does not dissociate into dimers in physiological settings,  it is not bound by haptoglobin. The stoichiometry of binding is two a ,b ­oxyhemoglobin dimers per haptoglobin molecule. Interesting studies have been made with  rabbit antihuman­hemoglobin antibodies on the haptoglobin–hemoglobin interaction. Human haptoglobin interacts with a variety of hemoglobins from different species.  The binding of human haptoglobin with human hemoglobin is not affected by the binding of rabbit antihuman­hemoglobin antibody. These studies suggest that  haptoglobin binds to sites on hemoglobin that are highly conserved in evolution and therefore are not sufficiently antigenic to generate antibodies. The most likely site  for the molecular interaction of hemoglobin and haptoglobin is the interface of the a  and b  globins of the tetramer that dissociates to yield a ,b  dimers. Sequence  determinations have indicated that these contact regions are highly conserved in evolution. The haptoglobins are a 2­globulins. Synthesized in the liver, they consist of two pairs of polypeptide chains (a  being the lighter and b  the heavier). The genes for the a   and a  chains are linked so that a single mRNA is synthesized, generating a single polypeptide chain that is cleaved to form the two different chains. The b  chains are  glycopeptides of 39 kDa and are invariant in structure; a  chains are of several kinds. The haptoglobin peptide chains are joined by disulfide bonds between the a  and  b  chains and between the two a  chains. Interaction of haptoglobin with hemoglobin forms a complex that is too large to be filtered through the renal glomerulus. Free hemoglobin (appearing in renal tubules  and in urine) will occur during intravascular hemolysis only when the binding capacity of circulating haptoglobin has been exceeded. Haptoglobin delivers hemoglobin to  the reticuloendothelial cells. The heme in free hemoglobin is relatively resistant to the action of heme oxygenase, whereas the heme residues in an a ,b  dimer of  hemoglobin bound to haptoglobin are very susceptible. The measurement of serum haptoglobin is used clinically as an indication of the degree of intravascular hemolysis. Patients who have significant intravascular hemolysis  will have little or no levels of haptoglobin because of the removal of haptoglobin–hemoglobin complexes by the reticuloendothelial system. Haptoglobin levels can also  be low in severe extravascular hemolysis, in which the large load of hemoglobin in the reticuloendothelial system leads to the transfer of free hemoglobin into plasma. Free heme and hematin appearing in plasma are bound by a b ­globulin, hemopexin (57 kDa). One heme residue binds per hemopexin molecule. Hemopexin transfers  heme to liver, where further metabolism by heme oxygenase occurs. Normal plasma hemopexin contains very little bound heme, whereas in intravascular hemolysis, the  hemopexin is almost completely saturated by heme and is cleared with a half­life of about 7 h. In the latter instance, excess heme binds to albumin, with newly  synthesized hemopexin serving as a mediator for the transfer of the heme from albumin to the liver. Hemopexin also binds free protoporphyrin. Bibliography Battersby, A. R. The Bakerian Lecture, 1984. Biosynthesis of the pigments of life. Proc. R. Soc. Lond. B Biol. Sci. 225:1, 1985. Bothwell, T. H., Charlton, R. W., and Motulsky, A. G. Hemochromatosis. In: C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (Eds.), The Metabolic and  Molecular Bases of Inherited Disease, 7th ed., Vol. II. New York: McGraw­Hill, 1995, p. 2237. Braig, K., Otwinowski, Z., Hegde, R., Boisvert, D. C., Joachimiak, A., Horwich, A. L., and Sigler, P. B. The crystal structure of the bacterial chaperonin GroEL at  2.8 Å. Nature 371:578, 1994. Casey, J. L., Hentze, M. W., Koeller, D. H., Caughman, S., Rouault, T. A., Klausner, R. D., and Harford, J. B. Iron­responsive elements: regulatory RNA  sequences that control mRNA levels and translation. Science 240:924, 1988. Chowdhury, J. R., Wolkoff, A. W., Chowdhury, N. R., and Arias, I. M. Hereditary jaundice and disorders of bilirubin metabolism. In: C. R. Scriver, A. L. Beaudet,  W. S. Sly, and D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, 7th ed., Vol. II. New York: McGraw­Hill, 1995, p. 2161.

 

  

Page 1022

Fenton, W. A., Kashi, Y., Furtak, K., and Horwich, A. L. Residues in chapteronin GroEL required for polypeptide binding and release. Nature 371:614, 1994. Ferreira, G. C. Ferrochelatase binds the iron­responsive element present in the erythroid 5­aminolevulinate synthase mRNA. Biochem. Biophys. Res. Commun.  214:875, 1995. Fleet, J. C. A new role for lactoferrin: DNA binding and transcription activation. Nutr. Rev. 53:226, 1995. Guo, B., Phillips, J.D., Yu, Y., and Leibold, E.A. Iron regulates the intracellular degradation of iron regulatory protein 2 by the proteasome. J. Biol. Chem.  270:21645, 1995. Hartl, F. U. Molecular chaperones in cellular protein folding. Nature 381:571, 1996. Huebers, H. A., and Finch, C. A. Transferrin: physiologic behavior and clinical implications. Blood 64:763, 1984. Kappas, A., Sassa, S., Galbraith, R. A., and Nordmann, Y. The porphyrias. In: C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (Eds.), The Metabolic and  Molecular Bases of Inherited Disease, 7th ed., Vol. II. New York: McGraw­Hill, 1995, p. 2103. Lustbader, J. W., Arcoleo, J. P., Birken, S., and Greer, J. Hemoglobin­binding site on haptoglobin probed by selective proteolysis. J. Biol. Chem. 258:1227, 1983. Maeda, N., Yang, F., Barnett, D. R., Bowman, B. H., and Smithies, O. Duplication within the haptoglobin Hp2 gene. Nature 309:131, 1984. Mascotti, D. P., Rup, D., and Thach, R. E. Regulation of iron metabolism: translational effects mediated by iron, heme, and cytokines. Annu. Rev. Nutr. 15:239,  1995. May, W. S., Sahyoun, N., Jacobs, S., Wolf, M., and Cuatracasas, P. Mechanism of phorbol­diester­induced regulation of surface transferrin receptor involves the  action of activated protein kinase C and an intact cytoskeleton. J. Biol. Chem. 260:9419, 1985. Melefors, O., and Hentze, M. W. Iron regulatory factor—the conductor of cellular iron regulation. Blood Rev. 7:251, 1993. Osterman, J., Horwich, A. L., Neupert, W., and Hartl, F.­U. Protein folding in mitochondria requires complex formation with hsp60 and ATP hydrolysis. Nature  341:125, 1989. Pantopoulos, K., Gray, N. K., and Hentze, M. W. Differential regulation of two related RNA­binding proteins, iron regulatory protein (IRP) and IRPB. RNA 1:155,  1995. Pantopoulos, K., and Hentze, M. W. Nitric oxide signalling to iron­regulatory protein: direct control of ferritin mRNA translation and transferrin receptor mRNA  stability in transfected fibroblasts. Proc. Natl. Acad. Sci. USA 92:1267, 1995. Weiss, J. S., Gautam, A., Lauff, J. J., et al. The clinical importance of a protein­bound fraction of serum bilirubin in patients with hyperbilirubinemia. N. Engl. J. Med.  309:147, 1983. Yamashiro, D. J., Tycko, B., Fluss, S. R., and Maxfield, F. R. Segregation of transferrin to a mildly acidic (pH 6.5) para­golgi compartment in the recycling pathway.  Cell 37:789, 1984. Questions C. N. Angstadt and J. Baggott Refer to the following for Questions 1–3: A. ferritin B. ferredoxin C. hemosiderin D. lactoferrin E. transferrin 1. A type of protein in which iron is specifically bound to sulfur. 2. Exhibits an antimicrobial effect in the intestinal tract of newborns because of its ability to bind iron. 3. Delivers iron to tissues by binding to specific cell surface receptors. 4. In the intestinal absorption of iron: A. the presence of a reductant like ascorbate enhances the availability of iron. B. regulation of uptake occurs between the lumen and mucosal cells. C. the amount of apoferritin synthesized in the mucosal cell is directly related to the need for iron by the host. D. iron bound tightly to a ligand, such as phytate, is more readily absorbed than free iron. E. low pH in the stomach inhibits absorption by favoring Fe3+. 5. Which of the following statements about iron distribution is correct? A. Iron overload cannot occur because very efficient excretory mechanisms are available. B. Cells cannot regulate their uptake of iron with changing iron content. C. Transferrin decreases in iron deficiency to facilitate storage of iron. D. Iron homeostasis is maintained in part by iron regulatory proteins binding to iron­responsive elements in mRNA. E. In the early stages of iron depletion, serum ferritin levels rise rapidly as iron is released from storage forms. 6. The biosynthesis of heme requires all of the following EXCEPT: A. propionic acid. B. succinyl CoA. C. glycine. D. ferrous ion. 7. Uroporphyrin III: A. is an intermediate in the biosynthesis of heme. B. does not contain a tetrapyrrole ring. C. differs from coproporphyrin III in the substituents around the ring. D. is formed from uroporphyrinogen III by an oxidase. E. formation is the primary control step in heme synthesis. 8. Aminolevulinic acid synthase: A. requires NAD for activity. B. is allosterically activated by heme. C. synthesis is inhibited by steroids. D. is synthesized in mitochondria. E. synthesis can be induced by a variety of drugs. 9. Lead poisoning would be expected to result in an elevated level of: A. aminolevulinic acid. B. porphobilinogen. C. protoporphyrin I. D. heme. E. bilirubin.

 

  

Page 1023

10. Ferrochelatase: A. is an iron­chelating compound. B. releases iron from heme in the degradation of hemoglobin. C. binds iron to sulfide ions and cysteine residues. D. is inhibited by heavy metals. E. is involved in the cytoplasmic portion of heme synthesis. 11. Heme oxygenase: A. can oxidize the methene bridge between any two pyrrole rings of heme. B. requires molecular oxygen. C. produces bilirubin. D. produces carbon dioxide. E. can use either heme or protoporphyrin IX as substrate. 12. The substance deposited in skin and sclera in jaundice is: A. biliverdin. B. only unconjugated bilirubin. C. only direct bilirubin. D. both bilirubin and bilirubin diglucuronide. E. hematin. 13. Hepatic disease leads to major elevation of the blood level of: A. heme. B. biliverdin. C. bilirubin. D. bilirubin diglucuronide. E. direct bilirubin. 14. Biliary obstruction leads to major elevation of the blood level of: A. only direct bilirubin. B. only indirect bilirubin. C. both direct and indirect bilirubin. D. heme but not bilirubin. E. biliverdin but not bilirubin. 15. Acute intermittent porphyria is accompanied by an increased urinary level of: A. biliverdin. B. direct bilirubin. C. heme. D. indirect bilirubin. E. porphobilinogen. 16. Haptoglobin binds: A. a globin monomer. B. an oxyhemoglobin molecule. C. a ,b ­oxyhemoglobin dimers. D. a deoxyhemoglobin molecule. E. a ,b ­deoxyhemoglobin dimers. 17. Haptoglobin: A. helps prevent loss of iron following intravascular red blood cell destruction. B. levels in serum are elevated in severe intravascular hemolysis. C. inhibits the action of heme oxygenase. D. binds heme and hematin as well as hemoglobin. E. is a b ­globulin. Answers 1. B Animal ferredoxins, also known as nonheme iron­containing proteins, have two irons bound to two cysteine residues and sharing two sulfide ions (p. 1004). 2. D As long as lactoferrin is not saturated, its avid binding of iron diminishes the amount available for growth of microorganisms (p. 1003). 3. E Internalization of the receptor–transferrin complex is mediated by a Ca2+–calmodulin–protein kinase C complex. Internalization is followed by release of the iron  and recycling of the apotransferrin to the plasma (p. 1003). Ferritin and hemosiderin (p. 1004) are storage forms of iron. 4. A A and D: Ascorbate facilitates reduction to the ferrous state and, therefore, dissociation from ligands and absorption. B: Substantial iron enters the mucosal cell  regardless of need, but the amount transferred to the capillary beds is controlled. C: Iron bound to apoferritin is trapped in mucosal cells and not transferred to the  host. E: Oxidation to Fe3+ is favored by higher pH (p. 1005). 5. D B and D: In the presence of low iron this mechanism leads to increased synthesis of transferrin receptor and decreased synthesis of apoferritin. A: The high affinity  of many macromolecules for iron prevents efficient excretion. C: Transferrin increases in iron deficiency to improve absorption. E: Serum ferritin is normally small and  decreases (pp. 1007–1008). 6. A B and C: The organic portion of heme comes totally from glycine and succinyl CoA; the propionic acid side chain comes from the succinate. D: The final step of  heme synthesis is the insertion of the ferrous ion (p. 1010, Figure 24.7). 7. C A, B, and D: The tetrapyrrole porphyrins (except for protoporphyrin IX) are not intermediates but end products formed from the porphyrinogens  nonenzymatically. E: Synthesis of aminolevulinic acid is the rate­limiting step (p. 1010, Figure 24.7). 8. E The enzyme is induced in response to need (many drug detoxifications are cytochrome P450­dependent). A: The mechanism involves a Schiff base with glycine.  B: Heme both allosterically inhibits and suppresses synthesis of the enzyme. C: One reduction product of catabolic steroids stimulates synthesis. D: The gene for this  enzyme is on nuclear DNA (pp. 1009–1012). 9. A A–D: Lead inhibits ALA dehydratase so it inhibits synthesis of porphobilinogen and subsequent compounds. Heme certainly would not be elevated, because lead  also inhibits ferrochelatase. E: Bilirubin is a breakdown product of heme, not an intermediate in synthesis (p. 1013). 10. D This enzyme, in the mitochondria, catalyzes the last step of heme synthesis, the insertion of Fe2+, and is sensitive to the effects of heavy metals (p. 1016). 11. B Oxygenases usually use O2. A: The enzyme is specific for the methene between the two rings containing the vinyl groups (a ­methene bridge). C and D: The  products are biliverdin and CO; the measurement of CO in the breath is an index of heme degradation. E: Iron is necessary for activity (p. 1017). 12. D Both conjugated (direct) and unconjugated (indirect) bilirubin are deposited (p. 1019).

 

  

Page 1024

13. C Since the liver is responsible for conjugating bilirubin, hepatic disease leads to the elevation of unconjugated (indirect) bilirubin in blood. A and B: Catabolism of  heme to bilirubin occurs in reticuloendothelial cells. D and E: These are the same and require conjugation by the liver (p. 1019). 14. A Conjugated (direct) bilirubin is excreted in the bile. B and C: As long as the liver is functioning, bilirubin (indirect) will be conjugated. D and E: These occur in the  reticuloendothelial cells so bilirubin will be formed (pp. 1020–1021). 15. E The disease is characterized by increased ALA synthase and decreased porphobilinogen deaminase activities. A, B, and E: These all represent heme catabolism.  D: Heme synthesis is reduced (p. 1013). 16. C Haptoglobin binds dimers, two per haptoglobin molecule, specifically the oxyhemoglobin dimers since deoxyhemoglobin does not dissociate to dimers  physiologically (p. 1021). 17. A Haptoglobin is part of the scavenging mechanism to prevent urinary loss of heme and hemoglobin from intravascular degradation of red blood cells. B: Since the  scavenged complex is taken up by the reticuloendothelial system, the haptoglobin levels in serum are low. C: Heme residues in the dimers bound to haptoglobin are  more susceptible than free heme to oxidation by heme oxygenase. D and E: Heme and hematin are bound by a b ­globulin, while haptoglobin is an b ­globulin (p.  1021).

 

  

Page 1025

Chapter 25—  Gas Transport and pH Regulation James Baggott

25.1 Introduction to Gas Transport

1026

25.2 Need for a Carrier of Oxygen in Blood

1026

Respiratory System Anatomy Affects Blood Gas Concentration

1027

A Physiological Oxygen Carrier Must Have Unusual Properties

1027

The Steep Part of the Curve Lies in the Physiological Range

1028

25.3 Hemoglobin and Allosterism: Effect of 2,3­Bisphosphoglycerate

1029

25.4 Other Hemoglobins

1030

25.5 Physical Factors That Affect Oxygen Binding

1031

High Temperature Weakens Hemoglobin's Oxygen Affinity

1031

Low pH Weakens Hemoglobin's Oxygen Affinity

1031

25.6 Carbon Dioxide Transport

1031

Blood CO2 Is Present in Three Major Forms

1031

Bicarbonate Formation

1032

Carbaminohemoglobin Formation

1032

Two Processes Regulate [H+] Derived from CO2 Transport

1033

Buffering

1033

Isohydric Mechanism

1034

HCO3– Distribution between Plasma and Erythrocytes

1035

25.7 Interrelationships among Hemoglobin, Oxygen, Carbon Dioxide,  Hydrogen Ion, and 2,3­Biphosphoglycerate

1036

25.8 Introduction to pH Regulation

1036

25.9 Buffer Systems of Plasma, Interstitial Fluid, and Cells

1036

25.10 The Carbon Dioxide–Bicarbonate Buffer System

1038

The Chemistry of the System

1038

The Equilibrium Expression Involves an Anhydride Instead of an  Acid

1038

The Carbon Dioxide–Bicarbonate Buffer System is an Open System

1039

Graphical Representation: The pH–Bicarbonate Diagram

1040

25.11 Acid–Base Balance and Its Maintenance The Kidney Plays a Critical Role in Acid–Base Balance

1043

Urine Formation Occurs Primarily in the Nephron

1043

The Three Fates of Excreted H+

1043

Total Acidity of the Urine

1045

25.12 Compensatory Mechanisms

1046

Principles of Compensation

1046

The Three States of Compensation Defined

1046

Specific Compensatory Processes

1047

Respiratory Acidosis

1047

Respiratory Alkalosis

1047

Metabolic Acidosis

1048

Metabolic Alkalosis

1048

25.13 Alternative Measures of Acid–Base Imbalance

1049

25.14 The Significance of Na+ and Cl– in Acid–Base Imbalance

1050

Bibliography

1052

Questions and Answers

1052

Clinical Correlations

   

  

1041

 

25.1 Diaspirin Hemoglobin

1026

25.2 Cyanosis

1028

25.3 Chemically Modified Hemoglobins: Methemoglobin and  Sulfhemoglobin

1030

25.4 Hemoglobins with Abnormal Oxygen Affinity

1032

25.5 The Case of the Variable Constant

1039

25.6 The Role of Bone in Acid–Base Homeostasis

1042

25.7 Acute Respiratory Alkalosis

1047

25.8 Chronic Respiratory Acidosis

1048

25.9 Salicylate Poisoning

1049

25.10 Evaluation of Clinical Acid–Base Data

1051

25.11 Metabolic Alkalosis

1052

Page 1026

CLINICAL CORRELATION 25.1  Diaspirin Hemoglobin Shock is a condition of inadequate tissue perfusion due, for example, to loss of blood.  Hemorrhagic shock is a major cause of death following trauma. Rapid blood transfusion  can be life­saving, but cross­matching must be done before transfusing blood, and  transfusion is associated with a significant risk of disease. In addition, blood (or blood of  the correct type) may be in short supply under certain circumstances. Hence there is  considerable interest in developing a safe, effective blood substitute. Hemoglobin in plasma has a very short lifetime. It rapidly dissociates into   dimers,  which bind to the plasma protein, haptoglobin, and are removed from circulation.  Hemoglobin can be specifically cross­linked with bis(3,5­dibromosalicyl) fumarate at the  Lys 99 of the a  chains; the product is called diaspirin cross­linked hemoglobin (DCLHb).  DCLHb has a longer lifetime in plasma than hemoglobin, and its lifetime can be extended  still further by polymerizing the DCLHb. DCLHb has performed well as a blood  replacement in experimental animals, and the possibility of using it in humans is being  pursued.   25.1—  Introduction to Gas Transport Large organisms, especially terrestrial ones, require a relatively tough, impermeable outer covering to help shield them from dust, twigs, nonisotonic fluids like rain and  seawater, and other elements in the environment that might be harmful to living cells. One of the consequences of being large and having an impermeable covering is  that individual cells of the organism cannot exchange gases directly with the atmosphere. Instead there must exist a specialized exchange surface, such as a lung or a  gill, and a system to circulate the gases (and other materials, such as nutrients and waste products) in a manner that will meet the needs of every living cell in the body. The existence of a system for the transport of gases from the atmosphere to cells deep within the body is not merely necessary, it has definite advantages. Oxygen is a  good oxidizing agent, and at its partial pressure in the atmosphere, about 160 mmHg or 21.3 kPa, it would oxidize and inactivate many components of the cells, such  as essential sulfhydryl groups of enzymes. By the time O2 gets through the transport system of the body its partial pressure is reduced to a much less damaging 20  mmHg (2.67 kPa) or less. In contrast, CO2 is relatively concentrated in the body and becomes diluted in transit to the atmosphere. In the tissues, where it is produced,  its partial pressure is 46 mmHg (6.13 kPa) or more. In the lungs it is 40 mmHg (5.33 kPa), and in the atmosphere only 0.2 mmHg (0.03 kPa), less abundant than the  rare gas, argon. Its relatively high concentration in the body permits it to be used as one component of a physiologically important buffering system, a system that is  particularly useful because, upon demand, the concentration of CO2 in the extracellular fluid can be varied over a rather wide range. This is discussed in more detail  later in the chapter. Oxygen and CO2 are carried between the lungs and the other tissues by the blood. In the blood some of each gas is present in simple physical solution, but mostly each  is involved in some sort of interaction with hemoglobin, the major protein of the red blood cell. There is a reciprocal relation between hemoglobin's affinity for O2 and  CO2, so that the relatively high level of O2 in the lungs aids the release of CO2, which is to be expired, and the high CO2 level in other tissues aids the release of O2 for  their use. Thus a description of the physiological transport of O2 and CO2 is the story of the interaction of these two compounds with hemoglobin. 25.2—  Need for a Carrier of Oxygen in Blood An O2 carrier is needed in blood because O2 is not soluble enough in blood plasma to meet the body's needs. At 38°C, 1 L of plasma dissolves only 2.3 mL of O2.  Whole blood, because of its hemoglobin, has a much greater oxygen capacity (see Clin. Corr. 25.1). One liter of blood normally contains about 150 g of hemoglobin  (contained within the erythrocytes), and each gram of hemoglobin can combine with 1.34 mL of O2. Thus the hemoglobin in 1 L of blood can carry 200 mL of O2, 87  times as much as plasma alone would carry. Without an O2 carrier, the blood would have to circulate 87 times as fast to provide the same amount of O2. As it is, the  blood makes a complete circuit of the body in 60 s under resting conditions, and in the aorta it flows at the rate of about 18.6 m s–1. An 87­fold faster flow would  require a fabulous high­pressure pump, would produce tremendously turbulent flow and high shear forces in the plasma, would result in uncontrollable bleeding from  wounds, and would not even allow the blood enough time in the lungs to take up O2. The availability of a carrier not only permits us to avoid these impracticalities, but  also gives us a way of controlling oxygen delivery, since the O2 affinity of the carrier is responsive to changing physiological conditions.

 

  

Page 1027

Respiratory System Anatomy Affects Blood Gas Concentration The respiratory system includes the trachea, in the neck, which bifurcates in the thorax into right and left bronchi, as shown schematically in Figure 25.1. The bronchi  continue to bifurcate into smaller and smaller passages, ending with tiny bronchioles, which open into microscopic gas­filled sacs called alveoli. It is in the alveoli that  gas exchange takes place with the alveolar capillary blood.

Figure 25.1  Diagram showing the respiratory tract.

As we inhale and exhale, the alveoli do not appreciably change in size. Rather, it is the airways that change in length and diameter as the air is pumped into and out of  the lungs. Gas exchange between the airways and the alveoli then proceeds simply by diffusion. These anatomical and physiological facts have two important  consequences. In the first place, since the alveoli are at the ends of long tubes that constitute a large dead space, and the gases in the alveoli are not completely  replaced by fresh air with each breath, the gas composition of the alveolar air differs from that of the atmosphere, as shown in Table 25.1. Oxygen concentration is  lower in the alveoli because it is removed by the blood. Carbon dioxide concentration is higher because it is added. Since we do not usually breathe air that is  saturated with water vapor at 38°C, water vapor is generally added in the airways. The concentration of nitrogen is lower in the alveoli, not because it is taken up by  the body, but simply because it is diluted by the CO2 and water vapor. A second consequence of the existence of alveoli of essentially constant size is that the blood that flows through the pulmonary capillaries during expiration, as well as  the blood that flows through during inspiration, can exchange gases. This would not be possible if the alveoli collapsed during expiration and contained no gases, in  which case the composition of the blood gases would fluctuate widely, depending on whether the blood passed through the lungs during an inspiratory or expiratory  phase of the breathing cycle. A Physiological Oxygen Carrier Must Have Unusual Properties We have seen that an O2 carrier is necessary. Clearly this carrier would have to be able to bind oxygen at an O2 tension of about 100 mmHg (13.3 kPa), the partial  pressure of oxygen in the alveoli. The earner must also be able to release O2 to the extrapulmonary tissues. The O2 tension in the capillary bed of an active muscle is  about 20 mmHg (2.67 kPa). In resting muscle it is higher, but during extreme activity it is lower. These O2 tensions represent the usual limits within which an oxygen  carrier must work. An efficient carrier would be nearly fully saturated in the lungs but should be able to give up most of this to a working muscle. Let us first see whether a carrier that binds O2 in a simple equilibrium represented by

TABLE 25.1 Partial Pressures of Important Gases Given in Millimeters  of Hg (kPa)

  Gas

In the Atmosphere

In the Alveoli of the Lungs

mmHg

kPa

mmHg

kPa

O2

159

21.2

100

13.3

N2

601

80.1

573

76.4

CO2

       0.2

       0.027

  40

    5.33

H2O

   0

0

  47

    6.27

Total

760

   

  

     101

760

    101

Page 1028

Figure 25.2  Oxygen saturation curves for two hypothetical  oxygen carriers and for hemoglobin.  Curve A: Hypothetical  carrier with hyperbolic saturation curve (a  simple carrier), 90% saturated in the lungs  and 66% saturated at the partial pressure  found in interstitial fluid. Curve  B: Hypothetical carrier with  hyperbolic saturation curve (another simple  carrier), 56% saturated in the lungs and  20% saturated at the partial pressure found  in interstitial fluid. Dashed curve:  Hemoglobin in whole blood.

CLINICAL CORRELATION 25.2  Cyanosis Cyanosis is a condition in which a patient's skin or mucous membrane appears gray or (in  severe cases) purple­magenta. It is due to an abnormally high concentration of  deoxyhemoglobin below the surface, which is responsible for the observed color. The  familiar blue of superficial veins is due to their deoxyhemoglobin content and is a normal  manifestation of this color effect. Cyanosis is most commonly caused by diseases of the cardiac or pulmonary systems,  resulting in inadequate oxygenation of the blood. It can also be caused by certain  hemoglobin abnormalities. Severely anemic individuals cannot become cyanotic; they do  not have enough hemoglobin in their blood for the characteristic color of its deoxy form to  be apparent. Albert, R. K. Approach to the patient with cyanosis and/or hypoxemia. In: W. N. Kelley  (Ed.), Textbook of Internal Medicine. Philadelphia: Lippincott, 1989, pp. 2041–2044.   would be satisfactory. For this type of carrier the dissociation constant would be given by the simple expression

and the saturation curve would be a rectangular hyperbola. This model would be valid even for a carrier with several oxygen­binding sites per molecule, which we  know is the case for hemoglobin, as long as each site were independent and not influenced by the presence or absence of O2 at adjacent sites. If such a carrier had a dissociation constant that permitted 90% saturation in the lungs, then, as shown in Figure 25.2, curve A, at a partial pressure of 20 mmHg (2.67  kPa) it would still be 66% saturated and would have delivered only 24% of its O2 load. This would not be very efficient. What about some other simple carrier, one that bound O2 less tightly and therefore released most of it at low partial pressure, so that the carrier was, say, only 20%  saturated at 20 mmHg (2.67 kPa)? Again, as shown in Figure 25.2, curve B, it would be relatively inefficient; in the lungs this carrier could fill only 56% of its maximum  O2 capacity and would deliver only 36% of what it could carry. It appears then that the mere fivefold change in O2 tension between the lungs and the unloading site is  not compatible with efficient operation of a simple carrier. Simple carriers are not sensitive enough to respond massively to a signal as small as a fivefold change. Figure 25.2 also shows the oxygen­binding curve of hemoglobin in normal blood. The curve is sigmoid, not hyperbolic, and it cannot be described by a simple  equilibrium expression. Hemoglobin, however, is a very good physiological O2 carrier. It is 98% saturated in the lungs and only about 33% saturated in the working  muscle. Under these conditions it delivers about 65% of the O2 it can carry. It can be seen in Figure 25.2 that hemoglobin is 50% saturated with O2, at a partial pressure of 27 mmHg (3.60 kPa). The partial pressure corresponding to 50%  saturation is called the P50. The term P50 is the most common way of expressing hemoglobin's O2 affinity. By analogy with Km  for enzymes, a relatively high P50  corresponds to a relatively low O2 affinity. The Steep Part of the Curve Lies in the Physiological Range Note that the steep part of hemoglobin's saturation curve lies in the range of O2 tensions that prevail in the extrapulmonary tissues. This means that relatively small  decreases in oxygen tension in these tissues will result in large increases in O2 delivery, this effect becoming more pronounced as the partial pressure of O2 diminishes  within the physiological range. Furthermore, small shifts of the curve to the left or right will also strongly influence O2 delivery. In Sections 25.3, 25.5, and 25.6 we see  how physiological signals effect such shifts and result in enhanced delivery under conditions of increased O2 demand. Small decreases of O2 tension in the lungs,  however, such as occur at moderately high altitudes, do not seriously compromise hemoglobin's ability to bind oxygen. This will be true as long as the alveolar partial  pressure of O2 remains in a range that corresponds to the relatively flat region of hemoglobin's O2 dissociation curve (see Clin. Corr. 25.2). Finally, we can see from Figure 25.2 that the binding of oxygen by hemoglobin is cooperative. At very low O2 tension the hemoglobin curve tends to follow the  hyperbolic curve, which represents relatively weak O2 binding, but at higher tensions it actually rises above the hyperbolic curve that represents tight binding. Thus  hemoglobin binds O2 weakly at low oxygen tension and tightly at high tension. The binding of the first O2 to each hemoglobin molecule enhances the binding of  subsequent O2 molecules.

 

  

Page 1029

Hemoglobin's ability to bind O2 cooperatively is reflected in its Hill coefficient, which has a value of about 2.7. (The Hill equation is derived and interpreted on p.  119.) Since the maximum value of the Hill coefficient for a system at equilibrium is equal to the number of cooperating binding sites, a value of 2.7 means that  hemoglobin, with its four oxygen­binding sites, is more cooperative than would be possible for a system with only two cooperating binding sites, but it is not as  cooperative as it could be.

Figure 25.3  2,3­Bisphosphoglycerate  (BPG).

25.3—  Hemoglobin and Allosterism:  Effect of 2,3­Bisphosphoglycerate Hemoglobin's binding of O2 was the original example of a homotropic effect (cooperativity and allosterism are discussed in Chapter 4), but hemoglobin also exhibits a  heterotropic effect of great physiological significance. This involves its interaction with 2,3­bisphosphoglycerate (BPG) (Figure 25.3), which is closely related to the  glycolytic intermediate, 1,3­bisphosphoglycerate, from which it is biosynthesized. It had been known for many years that hemoglobin in the red cell bound oxygen less tightly than purified hemoglobin could (Figure 25.4). It had also been known that  the red cell contained high levels of BPG, nearly equimolar with hemoglobin. Finally, the appropriate experiment was done to demonstrate the relationships between  these two facts. It was shown that the addition of BPG to purified hemoglobin produced a shift to the right of its oxygen­binding curve, bringing it into congruence with  the curve observed for whole blood. Other organic polyphosphates, such as ATP and inositol pentaphosphate, also have this effect. Inositol pentaphosphate is the  physiological effector in birds, where it replaces BPG, and ATP plays a similar role in some fish.

Figure 25.4  Oxygen dissociation curves for myoglobin,  for hemoglobin that has been stripped of  CO  and organic phosphates, and for whole  2

red blood cells.  Data from Brenna, O., Luzzana,  M., Pace, M., et al.  Adv. Exp. Biol. Med. 28:19, 1972.  Adapted from McGilvery, R. W.  Biochemistry: A Functional  Approach, 2nd ed. Philadelphia:  Saunders, 1979, p. 236.

Monod's model of allosterism explains heterotropic interaction. Applying this model to hemoglobin, in the deoxy conformation (the T state) a cavity large enough to  admit BPG exists between the b  chains of hemoglobin. This cavity is lined with positively charged groups and firmly binds one molecule of the negatively charged  BPG. In the oxy conformation (the R state) this cavity is smaller, and it no longer accommodates BPG as easily. The result is that the binding of BPG to  oxyhemoglobin is much weaker. Since BPG binds preferentially to the T state, the presence of BPG shifts the R–T equilibrium in favor of the T state; the  deoxyhemoglobin conformation is thus stabilized over the oxyhemoglobin conformation (Figure 25.5). For oxygen to overcome this and bind to hemoglobin, a higher  concentration of oxygen is required. Oxygen tension in the lungs is sufficiently high under most conditions to saturate hemoglobin almost completely, even when BPG  levels are high. The physiological effect of BPG can, therefore, be expected to be upon release of oxygen to the extrapulmonary tissues, where O2 tensions are low.

Figure 25.5  Schematic representation  of equilibria among BPG, O2,  and the T and R states of  hemoglobin.

The significance of a high BPG concentration is that the efficiency of O2 delivery is increased. Concentrations of BPG in the red cell rise in conditions associated with  tissue hypoxia, such as various anemias, cardiopulmonary insufficiency, and high altitude. These high levels of BPG enhance the formation of deoxyhemoglobin at low  partial pressures of oxygen; hemoglobin then delivers more of its O2 to the tissues. This effect can result in a substantial increase in the amount of O2 delivered because  the venous blood returning to the heart of a normal individual is (at rest) at least 60% saturated with O2. Much of this O2 can dissociate in the peripheral tissues if the  BPG concentration rises. The BPG mechanism works very well as a compensation for tissue hypoxia as long as the partial pressure of oxygen in the lungs remains high enough that oxygen  binding in the lungs is not compromised. Since, however, BPG shifts the oxygen­binding curve to the right, the mechanism will not compensate for tissue hypoxia when  the partial pressure of O2 in the lungs falls too low. Then

 

  

Page 1030

CLINICAL CORRELATION 25.3  Chemically Modified Hemoglobins: Methemoglobin and Sulfhemoglobin Methemoglobin is a form of hemoglobin in which the iron is oxidized from the iron (II)  state to the iron (III) state. A tendency for methemoglobin to be present in excess of its  normal level of about 1% may be due to a hereditary defect of the globin chain or to  exposure to oxidizing drugs or chemicals. Sulfhemoglobin is a species that forms when a  sulfur atom is incorporated into the porphyrin ring of hemoglobin. Exposure to certain  drugs or to soluble sulfides produces it. Sulfhemoglobin is green. Hemoglobin subunits  containing these modified hemes do not bind oxygen, but they change the oxygen­binding  characteristics of the normal subunits in hybrid hemoglobin molecules containing some  normal subunits and one or more modified subunits. The accompanying figure shows the  oxygen­binding curve of normal HbA, 15% methemoglobin and 12% sulfhemoglobin.  The presence of methemoglobin shifts the curve to the left, impairing the delivery of the  decreased amount of bound oxygen. In contrast, the sulfhemoglobin curve is shifted to the  right, a BPG­like effect. As a result, oxygen delivery is enhanced, partially compensating  for the inability of the sulfur­modified hemes to bind oxygen.

Oxygenation curves of unmodified hemoglobin  A (squares) of a 15% oxidized hemolysate  (circles) and of a hemolysate containing  12% sulfhemoglobin (triangles) in 0.1 M  phosphate, pH 7.35, at 20°C.  Data from Park, C. M., and Nagel, R. L., N.  Engl. J. Med. 310:1579, 1984.

  the increased efficiency of O2 unloading to the tissues is counterbalanced by a decrease in the efficiency of loading in the lungs. This may be a factor in determining the  maximum altitude at which people choose to establish permanent dwellings, which is about 18,000 ft (~5500 m). There is evidence that a better adaptation to  extremely low ambient partial pressures of O2 would be a shift of the curve to the left. 25.4—  Other Hemoglobins Although hemoglobin A is the major form of hemoglobin in adults and in children over seven months of age, accounting for about 90% of their total hemoglobin, it is  not the only normal hemoglobin species. Normal adults also have 2–3% of hemoglobin A2, which is composed of two a  chains like those in hemoglobin A and two    chains. It is represented as  2 2. The   chains differ in amino acid sequence from the b  chains and are under independent genetic control. Hemoglobin A2 does not  appear to be important in normal individuals. Several species of modified hemoglobin A also occur normally. These are designated A1a1, A1a2, A1b, and A1c. They are adducts of hemoglobin with various sugars,  such as glucose, glucose 6­phosphate, and fructose 1,6­bisphosphate. The quantitatively most significant is hemoglobin A1c, formed by covalent binding of a glucose  residue to the N terminal of the b  chain at a rate that depends on the concentration of glucose. As a result, hemoglobin A1c forms more rapidly in uncontrolled diabetics  and can comprise up to 12% of their total hemoglobin. Hemoglobin A1c or total glycosylated hemoglobin levels are a useful measure of how well diabetes has been  controlled during the days and weeks before the measurement is taken; measurement of blood glucose only indicates how well diabetes is under control when the  blood sample is taken. Chemical modification of hemoglobin A can also occur from interaction with drugs or environmental pollutants (see Clin. Corr. 25.3). Fetal hemoglobin, hemoglobin F, is the major hemoglobin in newborn infants. It contains two g chains in place of the b  chains and is represented as  2 2. Shortly  before birth g­chain synthesis diminishes and b ­chain synthesis is initiated, and by the age of seven months well over 90% of the infant's hemoglobin is hemoglobin A. Hemoglobin F is adapted to the environment of the fetus, who gets oxygen from maternal blood, a source that is far poorer than the atmosphere. To compete with the  maternal hemoglobin for O2, fetal hemoglobin must bind O2 more tightly; its oxygen­binding curve is thus shifted to the left relative to hemoglobin A. This is  accomplished through a difference in the influence of BPG upon the maternal and fetal hemoglobins. In hemoglobin F two of the groups that line the BPG­binding  cavity have neutral side chains instead of the positively charged ones that occur in hemoglobin A. Consequently, hemoglobin F binds BPG less tightly and thus binds  oxygen more tightly than hemoglobin A does. Also, about 15–20% of the hemoglobin F is acetylated at the N terminals; this is referred to as hemoglobin F1.  Hemoglobin F1 does not bind BPG, and its affinity for oxygen is not affected at all by BPG. The postnatal change from hemoglobin F to hemoglobin A, combined with  a rise in red cell BPG that peaks three months after birth, results in a gradual shift to the right of the infant's oxygen­binding curve (Figure 25.6). The result is greater  delivery of oxygen to the tissues at this age than at birth, in spite of a 30% decrease in the infant's total hemoglobin concentration. In many inherited anomalies of hemoglobin synthesis there is formation of a structurally abnormal hemoglobin; these are called hemoglobinopathies. They may involve  the substitution of one amino acid in one type of polypeptide chain for some other amino acid or they may involve absence of one or more amino acid residues of a  polypeptide chain. In some cases the change is clinically insignificant, but in others it causes serious disease (see Clin. Corr. 25.4).

 

  

Page 1031

25.5—  Physical Factors That Affect Oxygen Binding High Temperature Weakens Hemoglobin's Oxygen Affinity Temperature has a significant effect on O2 binding by hemoglobin (Figure 25.7). At below­normal temperatures the binding is tighter, resulting in a leftward shift of the  curve; at higher temperatures the binding becomes weaker, and the curve is shifted to the right. The effect of elevated temperature is like that of high levels of BPG, in  that both enhance unloading of oxygen. The temperature effect is physiologically useful, as it makes additional O2 available to support the high metabolic rate found in  fever or in exercising muscle with its elevated temperature. The relative insensitivity to temperature of O2 binding at high partial pressure of oxygen minimizes  compromise of O2 uptake in the lungs under these conditions.

Figure 25.6  Oxygen dissociation curves after birth.  Adapted from Oski, F. A., and  Delivoria­Papadopoulos, M.  J. Pediatr. 77:941, 1970.

The tighter binding of O2 that occurs in hypothermic conditions is not important in hypothermia induced for surgical purposes. Decreased O2 utilization by the body and  increased solubility of O2 in plasma at lower temperatures, as well as the increased solubility of CO2, which acidifies the blood, compensate for hemoglobin's  diminished ability to release O2.

Figure 25.7  Oxygen dissociation curve for whole blood at  various temperatures.  From Lambertson, C. J. In: P. Bard  (Ed.), Medical Physiology, 11th ed. St.  Louis, MO: Mosby, 1961, p. 596.

Low pH Weakens Hemoglobin's Oxygen Affinity Hydrogen ion concentration influences hemoglobin's O2 binding. As shown in Figure 25.8, low pH shifts the curve to the right, enhancing O2 delivery, whereas high pH  shifts the curve to the left. It is customary to express O2 binding by hemoglobin as a function of plasma pH because it is this value, not the pH within the erythrocyte,  that is usually measured. Erythrocyte cell sap pH is lower than the plasma pH, but these two fluids are in equilibrium, and changes in one reflect changes in the other. The influence of pH upon O2 binding is physiologically significant, since a decrease in pH is often associated with increased oxygen demand. Increased metabolic rate  increases production of carbon dioxide and, as in muscular exercise and hypoxic tissue, lactic acid. These acids produced by metabolism help release oxygen to  support that metabolism. The increase in acidity of hemoglobin as it binds O2 is known as the Bohr effect; an equivalent statement is that the Bohr effect is the increase in basicity of  hemoglobin as it releases oxygen. The effect may be expressed by the equation

This equation gives the same information as Figure 25.8—that increases in hydrogen ion concentration favor formation of free oxygen from oxyhemoglobin, and  conversely, that oxygenation of hemoglobin lowers the pH of the solution.

Figure 25.8  Oxygen dissociation curve for whole blood at  various values of plasma pH.  Adapted from Lambertson, C. J. In: P. Bard  (Ed.), Medical Physiology, 11th ed. St.  Louis, MO: Mosby, 1961, p. 596.

25.6—  Carbon Dioxide Transport The carbon dioxide we produce is excreted by the lungs, to which it is transported by the blood. Carbon dioxide transport is closely tied to hemoglobin and to the  problem of maintaining a constant pH in the blood, a problem that will be discussed subsequently. Blood CO2 Is Present in Three Major Forms Carbon dioxide is present in the blood in three major forms, as dissolved CO2, as HCO3– (formed by ionization of H2CO3 produced when CO2 reacts with H2O), and as  carbaminohemoglobin (formed when CO2 reacts with amino groups of protein). Each of these is present both in arterial blood and in venous blood

 

  

Page 1032

(see the top three lines of Table 25.2). Net transport to the lungs for excretion is represented by the concentration difference between arterial and venous blood,  shown in the last column. Note that for each form of carbon dioxide the arterial–venous difference is only a small fraction of the total amount present; venous blood  contains only about 10% more total carbon dioxide (total CO2 is the sum of HCO3–, dissolved CO2, and carbaminohemoglobin) than arterial blood. After carbon dioxide enters the bloodstream for transport, it generates hydrogen ions. Most come from formation of bicarbonate ion, which occurs in the following  manner. Bicarbonate Formation Carbon dioxide enters the blood and diffuses into erythrocytes, whose membranes, like most biological membranes, are freely permeable to dissolved CO2. Within the  erythrocytes most of the carbon dioxide is acted on by the intracellular enzyme, carbonic anhydrase, which catalyzes the reaction

This reaction proceeds in the absence of a catalyst, as is well known to all who drink carbonated beverages. Without the catalyst, however, it is too slow to meet the  body's needs, taking over 100 s to reach equilibrium. Recall that at rest the blood makes a complete circuit of the body in 60 s. Carbonic anhydrase is a very active  enzyme, having a turnover number of the order of 106, and inside the erythrocytes the reaction reaches equilibrium within 1 s, less than the time spent by the blood in  the capillary bed. The enzyme contains zinc and accounts in part for our dietary requirement for this metal. The ionization of carbonic acid, 

, is a rapid, spontaneous reaction. It produces equivalent amounts of H+ and HCO3–. Since, as shown in the last 

column of line 2 in Table 25.2, 1.69 meq of bicarbonate was added to each liter of blood by this process, 1.69 meq of H+ must also have been generated per liter of blood. Addition of this  much acid, over 10–3 equiv of H+, to 1 L of water would give a final pH below 3. Since the pH of venous plasma averages 7.37, most of the H+ generated during HCO3– production must be  consumed by buffer action and/or other processes. This is discussed below.

Because of the compartmentalization of carbonic anhydrase, essentially all conversion of CO2 to H2CO3, and ultimately to HCO3–, occurs inside the erythrocyte.  Negligible amounts of CO2 react nonenzymatically in the plasma. Thus virtually all of the increase in HCO3– in venous as compared to arterial blood is generated in  erythrocytes. Most of this diffuses into the plasma, so that venous plasma HCO3– is higher than the arterial, but the erythrocyte was the site of its formation. Carbaminohemoglobin Formation It has been observed that in the presence of carbonic anhydrase inhibitors, such as acetazolamide or cyanide, blood will still take up a certain amount of carbon  dioxide rapidly. This is due to the reaction of carbon dioxide with amino groups of proteins within erythrocytes to form carbamino groups (Figure 25.9). Hemoglobin  is quantitatively the most important protein involved in this reaction. Deoxyhemoglobin forms carbamino hemoglobin more readily than oxyhemoglobin. Oxygenation  causes release of CO2 in carbaminohemoglobin. Carbaminohemoglobin formation occurs only with uncharged aliphatic amino groups, not with the charged form, R–NH3+. The pH within erythrocytes is normally  about 7.2, somewhat more acidic than the plasma. Since protein amino groups have pK values well to the alkaline side of 7.2, they will be mostly in the charged  (undissociated acid) form. Removal of some of the un­ CLINICAL CORRELATION 25.4  Hemoglobins with Abnormal Oxygen Affinity Some abnormal hemoglobins have an altered affinity for oxygen. If oxygen affinity is  increased (P50 decreased), oxygen delivery to the tissues will be diminished unless some  sort of compensation occurs. Typically, the body responds by producing more  erythrocytes (polycythemia) and more hemoglobin. Hb Rainier is an abnormal hemoglobin  in which the P50 is 12.9 mmHg, far below the normal value of 27 mmHg. In the accompanying figure the oxygen content in volume percent (mL of O2 per 100 mL  of blood) is plotted versus partial pressure of oxygen, both for normal blood (curve a)  and for the blood of a patient with Hb Rainier (curve b). Obviously, the patient's blood  carries more oxygen; this is because it contains 19.5 g of Hb per 100 mL instead of the  usual 15 g per 100 mL. Since the partial pressure of oxygen in mixed venous blood is about 40 mmHg, the  volume of oxygen the blood of each individual can deliver may be obtained from the  graph by subtracting the oxygen content of the blood at 40 mmHg from its oxygen  content at 100 mmHg. As shown in the figure, the blood of the patient with Hb Rainier  delivers nearly as much oxygen as normal blood does, although Hb Rainier delivers a  significantly smaller fraction of the total amount it carries. Evidently, polycythemia is an  effective compensation for this condition, at least in the resting state.

Oxygen content plotted against  partial pressure of oxygen.

  (continued)

 

  

Page 1033

(Table continued from previous page) Curve a shows the oxygen dissociation  curve of normal blood with a hemoglobin  of 15 g dL–1,P  27 mmHg, n 2.8, at pH 7.4, 37°C.  50

Curve b shows that of blood from a patient  with Hb Rainier, having a hemoglobin of 19.5 g dL–1,  P50 12.9 mmHg, n 1.2, at the same  pH and temperature. (1 mmHg   133.3 Pa.)  On the right is shown the oxygen delivery.  The compensatory polycythemia  and hyperbolic curve of Hb Rainier  result in practically normal arterial and venous  oxygen tensions. Arrow indicates normal  mixed venous oxygen tension.  From Bellingham, A. J. Br. Med.  Bull. 32:234, 1976.

  charged form via carbamino group formation shifts the equilibrium, generating more uncharged amino groups and an equivalent amount of H+, as shown in Figure  25.10. Carbamination, like HCO3– formation, generates H+. The N­terminal a ­amino groups of proteins have pK values in the range of 7.6–8.4. The N terminals of hemoglobin's polypeptide chains are the principal sites of  carbamination. If they are blocked chemically by reaction with cyanate, carbamino formation does not occur. The N­terminal amino groups of the b ­globin chains are part of the binding site for BPG. Since they cannot bind BPG and also form carbamino groups, a competition  arises. Carbon dioxide diminishes the effect for BPG and, conversely, BPG diminishes the ability of hemoglobin to form carbaminohemoglobin. Ignorance of the latter  interaction led to a major overestimation of the role of carbaminohemoglobin in carbon dioxide transport. Prior to the discovery of the BPG effect, careful  measurements were made of the capacity of purified hemoglobin (no BPG present) to form carbaminohemoglobin. The results were assumed to be applicable to  hemoglobin in the erythrocyte, leading to the erroneous conclusion that carbaminohemoglobin accounted for 25–30% or more of CO2 transport. It now appears that  13–15% of CO2 transport is via carbaminohemoglobin. Table 25.3 summarizes the contribution of each major form of blood carbon dioxide to overall CO2 transport.

Figure 25.9  Carbamino formation from a  free amino group and carbon dioxide.

Two Processes Regulate [H+] Derived from CO2 Transport Buffering Hemoglobin, besides carrying O2 and CO2 in the covalently bound form of a carbamino group, also plays the major role in handling the H+ produced in CO2 transport.  It does this by buffering and by the isohydric mechanism (discussed below). Hemoglobin's buffering power resides in its ionizable groups with pK values close to the  intraerythrocyte pH. These include the four N­terminal amino groups and the imidazole side chains of the histidine residues. There are 38 histidines per hemoglobin  tetramer; these provide most of hemoglobin's buffering ability.

Figure 25.10  Dissociation of an  ammonium ion to yield a free  amino group and H+.

In whole blood, buffering takes up about 60% of the acid generated in normal carbon dioxide transport. Although hemoglobin is by far the most important  nonbicarbonate buffer in blood, the organic phosphates in the eryth­ TABLE 25.2 Properties of Blood of Humans at Resta Arterial

  Serum

  ­1

Hb carbamino groups (meq L  of  blood)

 

Venous

Cells

Blood

       1.13

         1.13

Serum

 

Cells

Blood

      1.42

   1.42

Serum

 

Cells

Blood

+0.29

  +0.29

HCO3­ (meq L­1 of blood)

   13.83

      5.73

       19.56

     14.84

      6.41

  21.25

+1.01

+0.68

  +1.69

Dissolved CO2 (meq L­1 of blood)

     0.71

      0.48

         1.19

       0.82

      0.56

   1.38

+0.11

+0.08

  +0.19

Total CO2 (meq L­1 of blood)

   14.54

      7.34

       21.88

     15.66

      8.39

   24.05

+1.12

+1.05

  +2.17

Free O2 (mmol L­1 of blood)

 

 

         0.10

 

 

    0.04

 

 

  ­0.06

Bound O2 (mmol L­1 of blood)

 

 

         8.60

 

 

    6.01

 

 

  ­2.59

Total O2 (mmol L­1 of blood)

 

 

         8.70

 

 

    6.05

 

 

  ­2.65

 

 

      88.0

 

 

     37.2

 

 

­50.8

 

 

      41.0

 

 

     47.5

 

 

  +6.5

 (mmHg)  (mmHg)

      7.40

      7.19

Volume (cc L­1 of blood)

551.7

448.3

H2O (cc L­1 of blood)

517.5

Cl­ (meq L­1 of blood)

    57.71

pH

       7.37

      7.17

1000

548.9

451.1

322.8

     840.0

514.7

    24.30

       82.01

     56.84

 

Source: From Baggott, J. Trends Biochem. Sci 3:N207, 1978, with permission of the publisher. a

 Hemoglobin, 9 mM; serum protein, 39.8 g L­1 of blood; respiratory quotient, 0.82.

   

  

A­V Difference

­0.03

­0.02

1000

­2.8

+2.8

    0.0

325.6

    840.0

­2.8

+2.8

    0.0

    25.17

      82.01

­0.88

+0.88

    0.0

 

 

Page 1034 TABLE 25.3 Major Forms of Carbon Dioxide  Transport Species

Transport (%)

HCO3

78

CO2 (dissolved)

  9

Carbaminohemoglobin

13



  TABLE 25.4 Processes occurring at the N  Terminals of the a  Chains and b  Chains of  Hemoglobin N Terminals

  Process Carbamino formation BPG binding –

H  binding in the Bohr  effect

a  Chains

b  Chains

Yes

Yes

No

Yes

Yes

No

  TABLE 25.5 Control of the Excess H+ Generated  During Normal Carbon Dioxide Transport Buffering

 

By hemoglobin

50%

By other buffers

10%

Isohydric mechanism  (hemoglobin)

40%

  rocytes, the plasma proteins, and so on also make a significant contribution. Buffering by these compounds accounts for about 10% of the H+, leaving about 50% of  acid control specifically attributable to buffering by hemoglobin. These buffer systems minimize the change in pH that occurs when acid or base is added but do not  altogether prevent that change. A small difference in pH between arterial and venous blood is therefore observed. Isohydric Mechanism The remainder of the H+ arising from carbon dioxide is taken up by hemoglobin, but not by buffering. Recall that when hemoglobin becomes oxygenated it becomes a  stronger acid and releases H+ (the Bohr effect). In the capillaries, where O2 is released, the opposite occurs:

Simultaneously, CO2 enters the capillaries and is hydrated:

Addition of these two equations gives

revealing that to some extent this system can take up H+ arising from CO2, and can do so without a change in H+ concentration (i.e., with no change in pH).  Hemoglobin's ability to do this, through the operation of the Bohr effect, is referred to as the isohydric carriage of CO2. As already pointed out, there is a small A–V  difference in plasma pH. This is because the isohydric mechanism cannot handle all the acid generated during normal CO2 transport; if it could, no such difference  would occur. Figure 25.11 is a schematic representation of

Figure 25.11  Schematic representation of oxygen transport and the isohydric  carriage of CO  by hemoglobin.  2

In the lungs (left) O2 from the atmosphere reacts with deoxyhemoglobin, forming  oxyhemoglobin and H+. The H+ combines with the HCO O and CO .  3– to form H2

2

The CO2 is exhaled. Oxyhemoglobin is carried to extrapulmonary tissues (right),  where it dissociates in response to low 

. The O2 is used by metabolic processes,   

and CO2 is produced. CO2 combines with H2O to give HCO3

– and H+. H+ can then react

with deoxyhemoglobin to give HHb, which returns to the lungs, and the cycle repeats.

 

  

Page 1035

O2 transport and the isohydric mechanism, showing what happens in the lungs and in the other tissues. Estimates of the importance of the isohydric mechanism in handling normal respiratory acid production have changed upward and downward over the years. The  older, erroneous estimates arose out of a lack of knowledge of the multiple interactions in which hemoglobin participates. The earliest experiments, titrations of purified  oxyhemoglobin and purified deoxyhemoglobin, revealed that oxygenation of hemoglobin resulted in release of an average of 0.7 H+ for every O2 bound. This figure still  appears in textbooks, and much is made of it. Authors point out that with a Bohr effect of this magnitude the isohydric mechanism alone could handle all of the acid  produced by the metabolic oxidation of fat (RQ of fat is 0.7), and buffering would be unnecessary. Unfortunately, the experimental basis for this interpretation is  physiologically unrealistic; the titrations were done in the total absence of carbon dioxide, which we now know binds to some of the Bohr groups, forming carbamino  groups and diminishing the effect. When later experiments were carried out in the presence of physiological amounts of carbon dioxide, there was a drastic diminution  of the Bohr effect, so much so that at pH 7.45 the isohydric mechanism was able to handle only the amount of acid arising from carbamino group formation. This  work, however, was done prior to our appreciation of the competition between BPG and CO2 for the same region of the hemoglobin molecule (see Table 25.4).  Finally, in 1971, careful titrations of whole blood under presumably physiological conditions were carried out, yielding a value of 0.31 H+ released per O2 bound. This  value is the basis of the present assertion that the isohydric mechanism accounts for about 40% of the H+ generated during normal carbon dioxide transport. The  quantitative contributions of various mechanisms to the handling of H+ arising during carbon dioxide transport are summarized in Table 25.5. The major role of  hemoglobin in handling this acid is obvious. HCO3– Distribution between Plasma and Erythrocytes We have seen that essentially all of HCO3– formation is intracellular, catalyzed by carbonic anhydrase, and that the vast bulk of the H+ generated by CO2 is handled  within the erythrocyte. These two observations bear upon the final distribution of HCO3– between plasma and the erythrocyte. Intracellular formation of HCO3– increases its intracellular concentration. Since HCO3– and Cl– exchange freely across the erythrocyte membrane, HCO3– will diffuse  out of the erythrocyte, increasing the plasma HCO3– concentration. Electrical neutrality must be maintained across the membrane as this happens. Maintenance of  neutrality can be accomplished in principle either by having a positively charged ion accompany HCO3– out of the cell or by having some other negatively charged ion  enter the cell in exchange for the HCO3–. Since the distribution of the major cations, Na+ and K+ , is under strict control, it is the latter mechanism that is seen, and the  ion that is exchanged for HCO3– is Cl–. Thus as HCO3– is formed in red cells during their passage through the capillary bed, it moves out into the plasma and Cl– comes  in to replace it. The increase in intracellular Cl– is shown in the last line of Table 25.2. In the lungs, all events that occur in the peripheral capillary beds are reversed;  HCO3– enters the erythrocytes to be converted to CO2 for exhalation, and Cl– returns to the plasma. The exchange of Cl– and HCO3– between the plasma and the  erythrocyte is called the chloride shift (Figure 25.12).

Figure 25.12  Schematic representation of the  chloride shift.  (a) In the capillaries of the extrapu­  lmonary tissues, CO  produced by  2

tissue metabolism is converted to  HCO3– in the erythrocytes. This  HCO – exits the erythrocytes in  3

exchange for Cl–.  (b) In the capillaries of the lungs,  HCO – enters the erythrocytes in  3

exchange for Cl–. Within the  erythrocytes HCO3– is converted  to CO2. CO2 subsequently diffuses  out of the erythrocytes and is exhaled.

The intraerythrocytic buffering of H+ from carbon dioxide causes these cells to swell, giving venous blood a slightly (0.6%) higher hematocrit than arterial blood.  (Hematocrit is the volume percent of red cells in the blood.) This occurs because the charge on the hemoglobin molecule becomes more positive with every H+ that  binds to it. Each bound positive charge requires an accompanying negative charge to maintain neutrality. Thus as a result of buffering there is a net accumulation of  HCO3– or Cl– inside the erythrocyte.

 

  

Page 1036

An increase in the osmotic pressure of the intracellular fluid results from this increase in concentration of particles. As a consequence, water enters the cells, causing  them to swell slightly. Typically, an arterial hematocrit might be 44.8 and a venous hematocrit 45.1, as shown in Table 25.2 by the line labeled ''volume (cc L–1 of  blood)."

Figure 25.13  Interaction of H+, BPG, CO2, and O2 with  hemoglobin.  This is a schematic, intended  to denote the direction of the equilibrium,  not the stoichiometry of the reaction.

25.7—  Interrelationships among Hemoglobin, Oxygen, Carbon Dioxide, Hydrogen Ion, and 2,3­Bisphosphoglycerate By now it should be clear that multiple interrelationships of physiological significance exist among the ligands of hemoglobin. These interrelationships are summarized  schematically in Figure 25.13. This equation shows that changes in the concentration of H+, BPG, or CO2 have similar effects on O2 binding. The equation will help you  remember the effect of changes in any one of these variables upon hemoglobin's O2 affinity. BPG levels in the erythrocytes are controlled by product inhibition of its synthesis and by pH. Hypoxia results in increased levels of deoxyhemoglobin on a time­ averaged basis. Since deoxyhemoglobin binds BPG more tightly, in hypoxia there is less free BPG to inhibit its own synthesis, and so BPG levels will rise due to  increased synthesis. The effect of pH is that high pH increases BPG synthesis and low pH decreases BPG synthesis; this reflects the influence of pH on the activity of  BPG mutase, the enzyme that catalyzes BPG formation. Since changes in BPG levels take many hours to become complete, this means that the immediate effect of a  decrease in blood pH is to enhance oxygen delivery by the Bohr effect. If the acidosis is sustained (most causes of chronic metabolic acidosis are not associated with a  need for enhanced oxygen delivery), diminished BPG synthesis leads to a decrease in intracellular BPG concentration, and hemoglobin's oxygen affinity returns toward  normal (Figure 25.14). This system can respond appropriately to acute conditions, such as vigorous exercise, but when faced with a prolonged abnormality of pH, it  readjusts to restore normal (and presumably optimal) oxygen delivery.

Figure 25.14  In chronic acidosis, BPG  concentration decreases, returning  hemoglobin's oxygen affinity toward normal.  This schematic diagram illustrates the  rapid decrease in hemoglobin's oxygen  affinity due to decreased pH. Lowering pH  immediately lowers the activity of BPG  mutase. In consequence, the concentration  of BPG gradually diminishes as normal  degradation proceeds. As BPG concentration  diminishes, hemoglobin's oxygen affinity  rises.

25.8—  Introduction to pH Regulation We have noted the large amount of H+ generated by carbon dioxide transport, and we considered the ways in which the blood pH is controlled. This is important  because changes in blood pH will affect intracellular pH, which in turn may profoundly alter metabolism. Protein conformation is affected by pH, as is enzyme activity.  In addition, the equilibria of important reactions that consume or generate hydrogen ions, such as any of the oxidation–reduction reactions involving pyridine  nucleotides, are shifted by changes in pH. Normal arterial plasma pH is 7.40 ± 0.05; the pH range compatible with life is about 6.8–7.8. Intracellular pH varies with cell type; that of the erythrocyte is nearly  7.2, but that of most other cells is lower, about 7.0. Values as low as 6.0 have been reported for skeletal muscle. It is fortunate for both diagnosis and treatment of diseases that the acid–base status of intracellular fluid influences and is influenced by the acid–base status of the  blood. Blood is readily available for analysis, and when alteration of body pH becomes necessary, intravenous administration of acidifying or alkalinizing agents is  efficacious. 25.9—  Buffer Systems of Plasma, Interstitial Fluid, and Cells Each body water compartment is defined spatially by one or more differentially permeable membranes. Each contains characteristic kinds and concentrations

 

  

Page 1037

Figure 25.15  Chief chemical constituents of the three fluid compartments.  Height of left half of each column indicates total concentration of  cations; that of right half, concentration of anions. Both are expressed  in milliequivalents per liter (meq L–1) of water. Note that chloride and  sodium values in cell fluid are questioned. It is probable that, at least in  muscle, the intracellular phase contains some sodium but no chloride.  Adapted from Gregersen, M. I. In: P. Bard (Ed.),  Medical Physiology, 11th ed. St. Louis, MO: Mosby, 1961, p. 307.

of solutes, some of which are buffers in the physiological range of pH. Although the solutes in each type of cell are different, most cells are similar enough to be  considered together for purposes of acid–base balance. Thus there are, from this point of view, three major body water components: plasma, within the circulatory  system; interstitial fluid, the fluid that bathes the cells; and intracellular fluid. The compositions of these fluids are given in Figure 25.15. In plasma the major cation is Na+; small amounts of K+, Ca2+, and Mg2+ are also present. The two  dominant anions are HCO3– and Cl–; smaller amounts of protein, phosphate, and SO42– are also present, along with a mixture of organic anions (amino acids, etc.), each of which would  be insignificant if taken separately. The sum of the anions equals, of course, the sum of the cations. It is apparent at a glance that the composition of interstitial fluid is very similar. The  major difference is that interstitial fluid contains much less protein than plasma contains (capillary endothelium is not normally permeable to plasma proteins) and, correspondingly, a lower  cation concentration. Plasma and interstitial fluid together comprise the extracellular fluid, and low molecular weight components equilibrate fairly rapidly between them. For example, H+  equilibrates between the plasma and interstitial fluid within about 1/2 h. The composition of intracellular fluid is strikingly different. The major cation is K+, while organic phosphates (ATP,  BPG, glycolytic intermediates, etc.) and protein are the major anions. TABLE 25.6 Acid Dissociation Constants of Major  Physiological Buffers Buffer System

  pK

HCO3–/CO2

6.1

Phosphate

 

HPO4 /H2PO4

6.7–7.2

Organic phosphate esters

6.5–7.6

2–



Protein

 

Histidine side chains

5.6–7.0

N­terminal amino groups

7.6–8.4

  Because of these differences among the fluid compartments, each fluid makes a different contribution to buffering. The major buffer of extracellular fluid, for example, is  the HCO3–/CO2 system. Since its pK is 6.1 (Table 25.6 lists the major physiological buffers and their pK values), extracellular fluid at a pH of 7.4 is not very effective  in resisting changes in pH arising from changes in   changes. We have already seen the importance of buffering by hemoglobin and organic phosphates within  erythrocytes. On the other hand, for reasons that will be explained

 

  

Page 1038

in Section 25.10, the bicarbonate buffer system is quite effective in controlling pH changes from causes other than changes in  . Extracellular and intracellular fluids  share almost equally in buffering strong organic or inorganic acids (see Table 25.7). Plasma is therefore an excellent indicator of the whole body's capacity to handle  additional loads of these acids. TABLE 25.7 Buffering of Metabolic Acids Tissue

Buffering (%)

Extracellular fluids

42

Red cells

  6

Tissue cells

52

  Since acid–base imbalance arising from metabolic production of organic acids is common and potentially life­threatening, and since plasma is such a good indicator of  the whole body's capacity to handle further metabolic acid loads, plasma composition is of major clinical concern. It is hydrogen ion concentration that must be kept  within acceptable limits, but measuring pH alone is like walking on thin ice while observing merely whether or not you are still on the surface. Knowledge of [HCO3–]  tells you how close the ice is to the breaking point and how deep the water is underneath. Because of the importance of the bicarbonate buffer system and its interaction with the other buffers of blood and other tissues, we will consider blood as a buffer in  some detail. We will begin with a brief consideration of a model buffer. Every buffer consists of a weak acid, HA, and its conjugate base, A–. Examples of conjugate base/weak acid pairs are acetate–/acetic acid, NH3/NH4+, and HPO42– /H2PO4–. Note that the weak acid may be neutral, positively charged, or negatively charged, and that its conjugate base must (since a H+ has been lost) have one less  positive charge (or one more negative charge) than the weak acid. The degree of ionization of a weak acid depends on the concentration of free hydrogen ions. This  may be expressed in the form of the Henderson–Hasselbalch equation (derived on p. 9) as follows:

This is a mathematical rearrangement of the fundamental equilibrium equation. It states that there is a direct relationship between pH and the ratio [conjugate base]/ [acid]. It is important to realize that this ratio, not the absolute concentration of any particular species, is the factor that is related to pH. Use of this equation will help  you to understand the operation of and to predict the effects of various alterations upon acid–base balance in the body. Blood plasma is a mixed buffer system; in the plasma the major buffers are HCO3–/CO2, HPO42–/H2PO4–, and protein/Hprotein. The pH is the same throughout the  plasma, so each of these buffer pairs distributes independently according to its own Henderson–Hasselbalch equation, shown in Figure 25.16. Because each pair has a  different pK, the [conjugate base]/[acid] ratio is also different for each. Note, though, that if the ratio is known for any given buffer pair, information about the others  can be calculated (assuming the pK values are known). 25.10—  The Carbon Dioxide–Bicarbonate Buffer System As we have seen, the major buffer of plasma and interstitial fluid is the bicarbonate buffer system. The bicarbonate system has two peculiar properties that make its  operation unlike that of typical buffers. We will examine this important buffer in some detail, since a firm understanding of it is the key to a grasp of acid–base balance.

Figure 25.16  Some of the Henderson–Hasselbalch  equations that are obeyed  simultaneously in plasma.

The Chemistry of the System The Equilibrium Expression Involves an Anhydride Instead of an Acid In the first place, the component that we consider to be the acid in this buffer system is CO2, which is an acid anhydride, not an acid. It reacts with water to

 

  

Page 1039

CLINICAL CORRELATION 25.5  The Case of the Variable Constant In clinical laboratories plasma pH and   are commonly measured with suitable  electrodes, and plasma [HCO3–] is then calculated from the Henderson–Hasselbalch  equation using pK = 6.1. Although this procedure is generally satisfactory, there have  been several reports of severely erroneous results in patients whose acid–base status was  changing rapidly.* Clinicians who are attuned to this phenomenon urge that direct  measurements of all three variables be made in acutely ill patients. The clinical literature discusses this problem in terms of departure of the value of pK from  6.1. Studies of model systems suggest that this interpretation is incorrect; pK does change  with ionic strength, temperature, and so on, and so does a , but not enough to account for  the magnitude of the clinical observations. Astute commentators have speculated that the real basis of the phenomenon is  disequilibrium. The detailed nature of the putative disequilibrium has not yet been  established, but it is probably related to the difference in pH across the erythrocyte  membrane. Normally, the pH of the erythrocyte is about 7.2, and the plasma pH is 7.4. If  the plasma pH changes rapidly in an acute illness, the pH of the erythrocyte will also  change, but the rate of change within the erythrocyte is not known. If the change within  the erythrocyte lags sufficiently behind the change in the plasma, the system would indeed  be in gross disequilibrium, and equilibrium calculations would not apply. *See Hood, I., and Campbell, E. J. M. N Engl. J. Med. 306:864, 1982.   form carbonic acid, which is indeed a typical weak acid:

Carbonic acid rapidly ionizes to give H+ and HCO3–:

If these two equations are added, H2CO3 cancels out, and the sum is

Elimination of H2CO3 from formal consideration is realistic, since not only does it simplify matters, but H2CO3 is, in fact, quantitatively insignificant. Because the  equilibrium of the reaction,

lies far to the left, H2CO3 is present only to the extent of 1/200 of the concentration of dissolved CO2. Since the concentration of H2O is virtually constant, it need not  be included in the equilibrium expression for the reaction, and we may write:

The value of K is 7.95 × 10–7. The concentration of a gas in solution is proportional to its partial pressure. Thus we measure partial pressure of CO2( the millimolar concentration of dissolved CO2.

 multiplied by a conversion factor, a , gives 

a  has a value of 0.03 meq L–1 mmHg–1 (or 0.225 meq L–1 kPa–1) at 37°C. The equilibrium expression thus becomes

and the Henderson–Hasselbalch equation for this buffer system becomes

with [HCO3–] expressed in units of meq L–1 (see Clin. Corr. 25.5). The Carbon Dioxide–Bicarbonate Buffer System Is an Open System We said earlier that the bicarbonate buffer system, with a pK of 6.1, is not effective against carbonic acid in the pH range of 7.8–6.8 but is effective against  noncarbonic acids. The usual rules of chemical equilibrium dictate that a buffer is not very useful in a pH range more than about one unit beyond its pK. Thus we need  to explain how the bicarbonate system can be effective against noncarbonic acids; its failure to buffer carbonic acid is expected. The way it buffers noncarbonic acids  in a pH range far from its pK is the second unusual property of this buffer system. Note that the explanation of this property in the following paragraph involves the  flow of materials in a living system, and so departs from mere equilibrium considerations. Consider first a typical buffer, consisting of a mixture of a weak acid and its conjugate base. When a strong acid is added, most of the added H+ combines with the  conjugate base. As a result, [weak acid] increases and simultaneously [conjugate base] diminishes. The ratio [conjugate base]/[weak acid] changes, and so does the  pH, but much less than if there were no buffer present. Now imagine that the weak acid, as it is generated by reaction of added strong acid with conjugate base, is  somehow removed so that while [conjugate base] diminishes, [weak acid] remains nearly constant. In this case the ratio of [conjugate base]/[weak acid] would change  much less for a given addition of strong acid, and the pH would also change much less. This is exactly what happens with the body's bicarbonate buffer system. As  strong acid is added, [HCO3–]

 

  

Page 1040

diminishes and CO2 is formed. But the excess CO2 is exhaled, so that the ratio of  keeping with the prediction of chemical equilibrium.

 changes strikingly, and the bicarbonate system would be relatively ineffective, in 

Figure 25.17  pH–Bicarbonate diagram  including the 40­mmHg (5.33­kPa)  CO  isobar, and showing the  2

normal values of plasma pH and bicarbonate  ion concentration.

Graphical Representation:  The pH–Bicarbonate Diagram A graphical representation of the Henderson–Hasselbalch equation for the bicarbonate buffer system assists in learning and understanding how this system reflects the  body's acid–base status. A common representation is the pH–bicarbonate diagram, shown in Figure 25.17. [HCO3–] up to 40 meq L–1 is shown on the ordinate;  enough to deal with most situations. Since plasma pH does not exceed 7.8 or (except transiently) fall below 7.0 in living patients, the abscissa is limited to 7.0–7.8. The  normal plasma [HCO3–], 24 meq L–1, and the normal plasma pH, 7.4, are indicated. The third variable, CO2, can be shown on a two­dimensional graph by assigning a  fixed value to 

 is 40 mmHg (5.33­kPa), pH and [HCO3–] must be somewhere on that line.

Figure 25.18  pH–Bicarbonate diagram showing CO2 isobars  from 10 to 100 mmHg.

Similarly, we can plot isobars for various abnormal values of   (Figure 25.18). The range of values given covers those found in patients. Any point on the graph  gives the values of the three variables of the Henderson–Hasselbalch equation for the bicarbonate system at that point. Since only two variables are needed to locate a  point, the third can be read directly from the graph. Let us now see how the bicarbonate buffer system behaves when it is in the presence of other buffers, as it is in whole blood. First, let us acidify the system by  increasing the concentration of the acid­producing component, CO2. For every CO2 that reacts with water to produce a H–, one HCO3– forms. Most of the H+, however,  is buffered by protein and phosphate. As a result, [HCO3–] rises much more than [H+]. Similarly, if acid is removed from this system by decreasing 

 is the only variable that is changed, 

the response of the system is confined to movements along this line.

The slope of the buffering line depends on the concentration of the nonbicarbonate buffers. If they were more concentrated, they would better resist changes in pH. An  increase in   to 80 mmHg (10.7 kPa) would then cause a smaller drop in pH, and since the more concentrated buffers would react with more hydrogen ions  (produced by the ionization of carbonic acid), [HCO3–] would rise higher. Thus the slope of the buffering line would be steeper.

 

  

Page 1041

Hemoglobin is quantitatively the second most important blood buffer, exceeded only by the bicarbonate buffer system. Since hemoglobin concentration in the blood  can fluctuate widely in various disease states, it is the most important physiological determinant of the slope of the blood buffer line. Figure 25.20 shows how this slope  varies with hemoglobin concentration.

Figure 25.19  The buffering line of blood.  This pH–bicarbonate  diagram shows the changes in pH that occur in  whole blood in vitro when   is changed.  Note that the relationship between pH and  – [HCO ] is described by a straight line with  3

a nonzero slope.

Having now seen how the bicarbonate buffer system in blood responds to changes in 

 are represented by points confined to the CO2 isobar.

The effects on blood of changing   or of adding acid or alkali, as we have just described, are realistic qualitative models of what happens in certain disease states.  We next see how these changes occur in the body and how the body compensates for them. 25.11—  Acid–Base Balance and Its Maintenance It should come as no surprise that mechanisms exist whereby the body normally rids itself of excess acid or alkali. The physiological implication is that if a patient is in a  state of continuing acidosis (excess acid or deficiency of alkali in the body) or alkalosis (excess alkali or deficiency of acid in the body), there must be a continuing  cause of the imbalance. In such a situation the body's first task is to somehow compensate so plasma pH does not exceed the limits compatible with life. Assistance  from the physician is sometimes necessary. The body's second task is to eliminate the primary cause of the imbalance, that is, to cure the disease, so that a normal  acid–base status can be reestablished. Again, intervention by the physician may be needed.

Figure 25.20  Slope of the buffering line of blood as it varies with  hemoglobin concentration.  From Davenport, H. W. The ABC of Acid–Base  Chemistry, 6th ed. revised.  Chicago: University of Chicago Press, 1974, p. 55.

 

  

Page 1042

CLINICAL CORRELATION 25.6  The Role of Bone in Acid–Base Homeostasis The average adult skeleton contains 50,000 meq of Ca2+ in the form of salts that are  alkaline relative to the pH of plasma. In chronic acidosis this large reservoir of base is  drawn upon to help control the plasma pH. Thus people with chronic kidney disease and  severely impaired renal acid excretion do not experience a continuous decline in plasma  pH and [HCO3–]. Rather, the pH and [HCO3–] stabilize at some below­normal level. The  resulting change in bone composition is not inconsequential, and clinical and  roentgenologic evidence of rickets or osteomalacia often appear. Bone healing has been  shown in these patients after prolonged administration of alkali in the form of sodium  bicarbonate or citrate sufficient to restore plasma [HCO3–]. Lemann, J. Jr. and Lennon, E. J. Kidney Int. 1:275, 1972.   All individuals, in sickness or in health, produce large amounts of acids every day. The major acid is CO2, the amount depending on the individual's caloric expenditure,  and ranging between 12,500 and nearly 50,000 meq day–1. In an average young adult male, about 22,000 meq of CO2 are produced daily. This acid is volatile and is  normally excreted by the lungs. Inability of the lungs to do this adequately leads to respiratory acidosis or alkalosis. Respiratory acidosis is the result of  hypoventilation of the alveoli, so that CO2 accumulates in the body. Alveolar hypoventilation occurs when the depth or rate of respiration diminishes. Airway  obstruction, neuromuscular disorders, and diseases of the central nervous system are common causes of acute respiratory acidosis. Chronic respiratory acidosis is  seen in patients with chronic obstructive lung disease, such as emphysema. Obviously, since the common element in all these conditions is increased alveolar    would also cause respiratory acidosis.

Figure 25.21  Effect of adding noncarbonic  acid or alkali to whole blood with  at 40 mmHg.

Respiratory alkalosis, on the other hand, arises from decreased alveolar 

 fixed 

 also falls, producing chronic respiratory alkalosis.

Nonvolatile acids are also produced by the body. The diet and physiological state of the individual determine the kinds and amounts of these acids. Oxidation of sulfur­ containing amino acids produces H+ and SO42–, the equivalent of sulfuric acid. Hydrolysis of phosphate esters is equivalent to the formation of phosphoric acid. The contribution of  these processes depends on the amount of acid precursors ingested; on an average American diet, net acid production is about 60 meq day–1.

Metabolism normally produces lactic acid, acetoacetic acid, and b ­hydroxybutyric acid. In some physiological or pathological states these are produced in excess, and  accumulation of the excess causes acidosis. When an ammonium salt of a strong acid, such as ammonium chloride, or when arginine hydrochloride or lysine  hydrochloride is administered, it is converted to urea, and the corresponding strong acid (HCl) is synthesized. Ingestion of salicylates, methyl alcohol, or ethylene glycol  results in production of strong organic acids. Accumulation of any of these nonvolatile acids leads to metabolic acidosis. While it is obvious that excess acid production can cause acidosis, the same net effect can arise from abnormal loss of base, as predicted from the Henderson­ Hasselbalch equation for the bicarbonate buffer system. Renal tubular acidosis is a condition in which this occurs. Abnormal amounts of HCO3– escape from the blood  into the urine, leaving the body acidotic (see Clin. Corr. 25.6). A more common cause of bicarbonate depletion is severe diarrhea. In this chapter it will be assumed  that kidney function is normal. Mammals do not synthesize alkaline compounds from neutral starting materials. Metabolic alkalosis therefore arises from intake of excess alkali or abnormal loss of  acid. A commonly ingested alkali is sodium bicarbonate. A less obvious source of alkali is the salt of any metabolizable organic acid. Sodium lactate is often  administered to combat acidosis; normal metabolism converts it to sodium bicarbonate. The net reaction is as follows:

Most dietary fruits and vegetables have a net alkalinizing effect on the body for this reason. They contain a mixture of organic acids, which are metabolized to CO2 and  H2O, and therefore have no long­term effect on acid–base balance, and salts of organic acids, which give rise to bicarbonate. Abnormal loss of acid, as occurs in  prolonged vomiting or gastric lavage, causes alkalosis. Alkalosis may also be produced by rapid loss of body water, as in diuresis, which may temporarily increase  [HCO3–] in the plasma and extracellular fluid. Table 25.8 summarizes the causes of acid­base imbalances.

 

  

Page 1043

The Kidney Plays a Critical Role in Acid­Base Balance Excess nonvolatile acid and excess bicarbonate are excreted by the kidney. As a result, urine pH varies as a function of the body's need to excrete these materials. For  an individual on a typical American diet, urine pH is about 6, indicating a net acidification as compared to plasma. This is consistent with our knowledge that the typical  diet results in a net production of acid. Urine pH can range from 4.4 to 8.0. A typical daily urine volume is about 1.2 L. At the minimum urine pH of 4.4, [H+] is only 0.04 meq L–1, and it would take 1250 L of urine to excrete 50 meq of acid as  free hydrogen ions. Clearly, most of the acid we excrete must be in a form other than H+. A form that can be excreted in a reasonable concentration, such as H2PO4–  or NH4+, is needed. TABLE 25.8 Causes of Acid­Base Imbalance  Summarized Acidosis Respiratory Alveolar hypoventilation Metabolic H+ overproduction HCO3– overexcretion Alkalosis Respiratory Alveolar hyperventilation Metabolic Alkali ingestion H+ overexcretion

  Urine Formation Occurs Primarily in the Nephron Let us now see how the kidney accomplishes the excretion of acid or base. Figure 25.22 shows the fundamental functioning unit of the kidney, a nephron. Each human  kidney contains at least a million, which first filter the blood and then modify the filtrate into urine.

Figure 25.22  Essential features of a typical  nephron in the human kidney.  Reprinted with  permission from Smith, H. W. The  Physiology of the Kidney. London: Oxford  University Press, 1937, p. 6.

Filtration occurs in the glomerulus, a tuft of capillaries enclosed by an epithelial envelope called the glomerular capsule (formerly Bowman's capsule). Water and low  molecular weight solutes, such as inorganic ions, urea, sugars, and amino acids (but not normally substances with molecular weights above 70,000, such as plasma  proteins), pass from these capillaries into the capsular space. This ultrafiltrate of plasma then passes through the proximal convoluted tubule, where most of the water  and solutes are reabsorbed. The tubule fluid continues through the loop of the nephron (loop of Henle) and through the distal convoluted tubule, where further  reabsorption of some solutes or secretion of others occurs. The tubule fluid then passes into the collecting tubule, where additional concentration can occur if  necessary. The fluid may now be called urine; it contains 1% or less of the water and solutes of the original glomerular filtrate. The kidney regulates acid–base balance by controlling bicarbonate reabsorption and by secreting acid. Both processes depend on formation of H+ and HCO3– from   and H2O within the tubule cells, shown in Figure 25.23a. The H+ formed in this reaction is actively secreted into the tubule fluid in exchange for Na+. Na+ uptake  CO2 by the tubule cell is partly passive, with Na+ flowing down the electrochemical gradient, and partly active, via a Na+, H+­antiport system. At this point Na+ has been  reabsorbed in exchange for H+, and sodium bicarbonate has been generated within the tubule cell. The sodium bicarbonate is then transported out of the cell into the  interstitial fluid, which equilibrates with the plasma. The Three Fates of Excreted H+ The H+ that has been secreted into the tubule fluid can now experience one of three fates. First, it can react with a HCO3–, as shown in Figure 25.23b, to form CO2 and H2O.  The overall net effect of this process is to move sodium bicarbonate from the tubule fluid back into the interstitial fluid. The name given to this is reabsorption of  sodium bicarbonate. As reabsorption of sodium bicarbonate proceeds, the tubule fluid becomes depleted of HCO3–, and the pH drops from its initial value, which was identical to the pH of  the plasma from which it was derived. As HCO3– becomes less available and the pH comes closer to the pK of the HPO42–/H2PO4– buffer system, more and more of  the H+ will be taken up by this buffer. Buffering is the second fate of H+, represented in Figure 25.23c. H2PO4– is not readily reabsorbed by the kidney. It passes out  in the urine, and its loss represents net excretion of H+.

 

  

Page 1044

Figure 25.23  Role of the exchange of tubular cell H+ ions in tubular fluid in renal regulation  of acid–base balance.  (a) Basic ion exchange mechanism  (b) Reabsorption of bicarbonate.  (c) Excretion of titrable acid.  (d) Excretion of ammonia.  Adapted from Pitts, R. E. N. Engl. J. Med. 284:32,  1971, with permission of the publisher.

Although phosphate is normally the most important buffer in the urine, other ions can become significant. For example, in diabetic ketoacidosis, plasma levels of  acetoacetate and b ­hydroxybutyrate are elevated. These pass into the glomerular filtrate and appear in the tubule fluid. Since acetoacetic acid has a pK = 3.6 and b ­ hydroxybutyric acid has a pK = 4.7, as the urine pH approaches its minimum of 4,4, these begin to serve as buffers. The effect of buffering is not only to excrete acid but to regenerate the bicarbonate that was lost when the acid was first neutralized. Let us consider a situation in which  the metabolic defect of a diabetic patient has produced the elements of b ­hydroxybutyric acid. The protons are neutralized by sodium

 

  

Page 1045

bicarbonate, leaving sodium b ­hydroxybutyrate. In the kidney, then, b ­hydroxybutyrate appears in the filtrate, it is converted to b ­hydroxybutyric acid, which is  excreted, and sodium bicarbonate returns to the extracellular fluid. Net acid excretion and bicarbonate regeneration occur no matter what anion in the tubule fluid acts  as the H+ acceptor. The amount of acid excreted as the acid component of a urinary buffer is measured by titrating the urine back to the normal pH of the plasma, 7.4. The amount of base  required is identical to the amount of acid excreted in this form and is called the titratable acidity of the urine. The formation of titratable acidity accounts for about one­third to one­half of our normal daily acid excretion. It is thus an important mechanism for acid excretion and  can put out as much as 250 meq of acid daily. There is, however, a limit to the amount of acid that can be excreted in this manner. Titratable acidity can be increased  only by lowering the pH of the urine or by increasing the concentration of buffer in the urine, and neither of these processes can proceed indefinitely. The urine pH  cannot go below about 4.4; evidently the Na+/H+ exchange mechanism is incapable of pumping H+ out of the tubule cells against more than a 1000­fold concentration  gradient. Buffer excretion is limited not only by the solubility of the buffer, but by limitations to the supply of the buffer ion and of the cations that are necessarily part of  the important buffer systems. If a 600 meq day–1 of acid were excreted as NaH2PO4, the body would be totally depleted of sodium in less than one week. The third fate that H+ can experience in the tubule fluid is neutralization by NH3. Tubule cells produce NH4+ from amino acids, particularly glutamine, as shown in  Figure 25.23d. Elimination of NH4+ in the urine contributes to net acid excretion. NH4+ is normally a major urinary acid. Typically, one­half to two­thirds of our daily acid load is excreted as NH4+. For three reasons it becomes even more important  in acidosis. In the first place, since the pK of NH4+ is 9.3, acid can be excreted in this form without lowering the pH of the urine, whereas formation of titratable acidity  requires a decrease in urine pH. Second, enormous amounts of acid can be excreted in this form. Ammonia is readily available from amino acids, and in prolonged  acidosis the NH4+ excretion system becomes activated. This activation, however, takes several days; it does not begin to adapt until after 2–3 days, and the process is  not complete until 5–6 days after the onset of acidosis. Once complete, though, amounts of acid in excess of 500 meq can be excreted daily as NH4+. The third role of  NH4+ in acidosis is that it spares the body's stores of Na+ and K+. Excretion of titratable acid, such as H2PO4–, and of the anions of strong acids, such as acetoacetate,  requires simultaneous excretion of a cation to maintain electrical neutrality. At the onset of acidosis this is Na+, but as the body's Na+ stores become depleted, K+  excretion rises. If NH4+ were not available, even a moderate acidosis could quickly become fatal. Total Acidity of the Urine Total acid excretion, the total acidity of the urine, is the sum of titratable acidity and NH4+. Strictly speaking, we should subtract from this sum the urinary HCO3–,  but this is seldom done in practice, since in severe metabolic acidosis, where the total acid excretion would be of greatest interest, the urine would be so acidic that  [HCO3–] would be nil. In alkalosis the kidney's role is simply to allow HCO3– to escape. Metabolic alkalosis is therefore seldom long­lasting unless alkali is continuously administered or  HCO3– elimination is somehow prevented. HCO3– elimination may be restricted if the kidney receives a strong signal to conserve Na+ at a time when there is a  deficiency of an easily reabsorbable anion, such as Cl–, to be reabsorbed with it. Some diuretics cause this. The first renal response is to put out K+ in exchange for  Na+ from the tubule fluid. When K+ stores are depleted, H+ is exchanged for Na+. This results in the production of an acidic urine by

 

  

Page 1046 –

an alkalotic patient. If NaCl is administered, alkalosis associated with volume and Cl  depletion may correct itself. 25.12—  Compensatory Mechanisms We have defined four primary types of acid–base imbalances and we have seen their chemical causes. Respiratory acidosis arises from an increased plasma  . In  metabolic acidosis addition of strong organic or inorganic acid (or loss of HCO3–) results in decreased plasma [HCO3–]. Conversely, in metabolic alkalosis loss of acid  from the body or ingestion of alkali raises the plasma [HCO3–]. Recall that in an acute respiratory acid–base imbalance, as long as there is no attempt to compensate,  pH will be abnormal, and [HCO3–] will be somewhere on the buffer line. In an acute metabolic acid–base imbalance, if there is no attempt to compensate, pH will be  abnormal and [HCO3–] will be somewhere on the 40­mmHg (5.33­kPa) isobar. Principles of Compensation When the plasma pH deviates from the normal range, various compensatory mechanisms begin to operate. The general principle of compensation is that, since an  abnormal condition has directly altered one term of the [HCO3–]/[CO2] ratio, plasma pH can be readjusted back toward normal by a compensatory alteration of the  other term. For example, if a diabetic patient becomes acidotic due to excess production of ketone bodies, plasma [HCO3–] will decrease. Compensation would  involve decreasing plasma [CO2] so that the [HCO3–]/[CO2] ratio, and therefore the pH, is readjusted back toward normal. Note that compensation does not involve a  return of [HCO3–] and [CO2] toward normal. Rather, compensation is a secondary alteration in one of these that counteracts the primary alteration in the other. The  result is that the plasma pH is readjusted toward normal. That this is necessarily so is evident from the Henderson–Hasselbalch equation.

If [HCO3–] changes, the only way to restore the original [HCO3–]/[CO2] ratio is to change  direction.

, the original ratio can be restored only by altering [HCO3–] in the same 

The Three States of Compensation Defined Although some compensatory mechanisms begin to operate rapidly and produce their effects rapidly, others are slower and show stages of compensation. First is the  acute stage, before any significant degree of compensation could possibly occur. After the acid–base imbalance has been in effect for a period of time the patient may  become compensated. This means the compensatory mechanisms have come into play in a normal manner, as expected on the basis of experience with other  individuals with an acid–base imbalance of similar type and degree. The ''compensated state" does not necessarily imply that the plasma pH is within the normal range.  Alternatively, the patient may show no sign of compensation and may be in the uncompensated state; this occurs because compensation cannot occur due to some  other abnormality. Finally, there is an intermediate state where compensation is occurring but is not yet as complete as it should be. This is the partially compensated  state. Factors that limit the compensatory processes will be discussed at the end of this section.

 

  

Page 1047

Specific Compensatory Processes Respiratory Acidosis Let us now follow the course of acute onset of each type of acid–base imbalance and of the compensatory process. Each of these will be schematically illustrated in a  pH–bicarbonate diagram. Imagine an individual in normal acid–base balance who goes into acute respiratory acidosis from breathing a gas mixture containing a high  level of CO2. As   will occur. The abnormal condition has fixed this patient on an abnormally high CO2 isobar. If the condition is returned to normal, he/she can  drop back to the 40­mmHg (5.33­kPa) isobar and all will be well, but until that time all compensatory processes are confined to the higher CO2 isobar.  Compensation, of course, consists of renal excretion of H+. Since this is a bicarbonate­producing process, [HCO3–] should rise, even though it is already above  normal. This could have been predicted from the pH–HCO2– diagram with no knowledge of the renal mechanism of compensation. Since it is assumed that the  individual is fixed on the high CO2 isobar by the abnormal condition, the only way the pH can possibly be adjusted toward normal is by sliding up the isobar to point B  in Figure 25.24. This movement is necessarily linked to an increase in [HCO3–]. Thus the correct analysis of this compensation could be made either from an  understanding of the nature of the compensatory mechanism or from an appreciation of the physical chemistry of the bicarbonate buffer system as expressed in the  pH–HCO3– diagram.

Figure 25.24  pH–Bicarbonate diagram  showing compensation for respiratory  acidosis (normal state to point B) and for  respiratory alkalosis (normal state to point D).

Although the path we have described, up the buffer line to point A and then up the isobar to point B, is a real possibility, it is also possible that a respiratory acidosis  would develop gradually, with compensation occurring simultaneously. The points describing this progress would fall on a curved line from the normal state to point B. Respiratory Alkalosis In sudden onset respiratory alkalosis 

), and the plasma pH

CLINICAL CORRELATION 25.7  Acute Respiratory Alkalosis An anesthetized surgical patient with a urethral catheter in place was hyperventilated as an  adjunct to the general anesthesia. Prior to hyperventilation normal values of plasma    was 25 mmHg and the pH was 7.55. Plasma HCO3– was not directly measured, but  interpolation from a pH–bicarbonate diagram (e.g., Figure 25.17) or calculation from the  Henderson–Hasselbalch equation reveals that the plasma [HCO3–] decreased to 21.2  meq L–1. Analysis of the urine showed negligible loss of HCO3– through the kidneys. It  can be concluded that the decrease in [HCO3–] was due to titration of bicarbonate by the  acid components of the body's buffer systems. The point representing the patient's new  steady­state condition clearly must be on the buffering line that represents whole body  buffering. (Since the buffers of the whole body are not identical in type or concentration  to the blood buffers, the buffer line for the whole body will be analogous, but not  identical, to the blood buffer line.) Magarian, G. J. Medicine(Baltimore) 61:219, 1982.    

  

Page 1048

CLINICAL CORRELATION 25.8  Chronic Respiratory Acidosis H.W. was admitted to the hospital with marked dyspnea, cyanosis, and signs of mental  confusion. As his acute problems were relieved by appropriate treatment, his symptoms  disappeared except for a continuing dyspnea. Blood gas analysis performed eight days  later yielded the following data: pH, 7.32;  , 70 mmHg; [HCO3–], 34.9 meq L–1. This  is a typical compensation for this degree of chronic respiratory acidosis. Another patient, C.Q., with chronic obstructive lung disease was found to have arterial  plasma pH, 7.40; [HCO3–], 35.9 meq L–1; and   of 60 mmHg, a plasma pH of 7.4 lies  outside the 95% probability range. Close questioning of the patient revealed that he had  surreptitiously been taking a relative's thiazide diuretic, which superimposed a metabolic  alkalosis upon respiratory acidosis. Rastegar, A., and Thier, S. O. Chest 63:355, 1972.   decreases toward normal. This is described in Figure 25.24 by movement along the isobar from point C to point D. With a gradual onset of respiratory alkalosis, the  bicarbonate buffer system would follow points along the curved line from the normal state to point D. Metabolic Acidosis In metabolic acidosis two mechanisms are usually available for dealing with the excess acid. One is that kidneys increase their H+ excretion, but this is slow and  inadequate to return [HCO3–] and pH to normal. The other, which begins to operate almost instantly, is respiratory compensation. Acidosis stimulates the respiratory  system to hyperventilate, decreasing the   but also a further small decrease in [HCO3–]. This is due to the same factor that causes the buffer line to have a slope:  titration of nonbicarbonate buffers. The inevitability and magnitude of the further decrease in [HCO3–] can be seen clearly in the pH–bicarbonate diagram.

Figure 25.25  pH–Bicarbonate diagram showing compensation  for metabolic acidosis (normal state to point F)  and for metabolic alkalosis (normal state to point H).

Metabolic Alkalosis The principles governing compensation for metabolic alkalosis are like those for metabolic acidosis, but operate in the opposite direction. In metabolic alkalosis the  primary defect is an increase in plasma [HCO3–]; it rises from the normal state to point G in Figure 25.25. The immediate physiological response is hypoventilation,  followed by increased renal excretion of HCO3–. As a result of hypoventilation 

 increases along the line from G to H, and a further small rise in [HCO3–] occurs.

The respiratory response to metabolic acid–base imbalance is rapid, and the bicarbonate buffer system would in most cases be expected to follow points along the  curved line from the normal state to the compensated state. An acute metabolic imbalance will not generally be seen outside the experimental laboratory. Indeed, if a  physician sees a patient whose plasma pH, [HCO3–], and   would be abnormal. How complete can compensation be? Can the body totally compensate (bring the pH back to the normal range) for any imbalance? Generally, the answer is no. The  compensatory organs, the lungs and kidneys, do not exist exclusively to deal with acid–base imbalance. There is a limit to how much one can hyperventilate; it is simply  impossible to move air into and out of the lungs at an indefinitely high rate for an indefinitely long time. Also, one cannot suspend respiration merely to raise  , rises  above 70 mmHg (9.33 kPa) in respiratory acidosis, renal mechanisms for reabsorbing HCO3– fail to keep pace, and further increases in plasma [HCO3–] are only  about what could be expected from titration of nonbicarbonate buffers (see Clin. Corr. 25.8). In respiratory alkalosis renal excretion of excess HCO3– can, with time,  be sufficient to return plasma pH to within the normal range. Individuals who dwell at high altitude are typically

 

  

Page 1049

CLINICAL CORRELATION 25.9  Salicylate Poisoning Salicylates are the most common cause of poisoning in children. A typical pathway of  salicylate intoxication is plotted in the accompanying figure. The first effect of salicylate  overdose is stimulation of the respiratory center, resulting in respiratory alkalosis. Renal  compensation occurs, lowering the

A typical pathway of salicylate intoxication.  Data replotted from Singer, R. B. Medicine  (Baltimore) 33:1, 1954.

plasma [HCO3–]. A second, delayed effect of salicylate may then appear, metabolic  acidosis. Since [HCO3–] had been lowered by the previous compensatory process, the  victim is at a particular disadvantage in dealing with the metabolic acidosis. In addition,  but not shown in the graph, respiratory stimulation sometimes persists after the acidosis  has run its course. Rational management of salicylate intoxication requires knowledge of  the plasma pH and the plasma [HCO3–] or its equivalent throughout the course of the  condition.   in compensated respiratory alkalosis, with their plasma pH within the normal range. For the other types of acid–base imbalance, the exact degree of compensation  expected of a patient with a given clinical picture is well worked out, but a detailed discussion is beyond the scope of this chapter. Suffice it to say that if a patient is  compensating, but not as well as expected, this is taken to mean that the patient cannot compensate appropriately and must therefore have a mixed acid–base  disturbance. 25.13—  Alternative Measures of Acid–Base Imbalance Modern clinical laboratories generally report plasma bicarbonate concentration, and the value is used by physicians just as we have used it here. Some laboratories,  however, report total plasma CO2, that is, the sum of bicarbonate and dissolved CO2, and this is always slightly higher than [HCO3–]. At pH 7.4, for example, the  ratio of [HCO3–] to [CO2] is 20 : 1 (dissolved CO2 is only 1 : 21 of the total CO2); if [HCO3–] is 24 meq L–1, [CO2] is 1.2 meq L–1 and total CO2 is 25.2 meq L–1. At  pH 7.1, HCO3– is still 10 times as concentrated as dissolved CO2. Because the major contributor to total CO2 is HCO3–, total CO2 is often used in the same manner as  bicarbonate to make clinical judgments. Strictly speaking, total CO2 also includes that in carbamino proteins, but current clinical laboratory practice is to ignore this  when making a blood gas and pH report. If it were included in a total CO2 measurement, it would not change the interpretation of the measurement, since the CO2 in  carbamino proteins, like dissolved CO2, represents only a small fraction of the total CO2. The clinical importance of bicarbonate as a gauge of the whole body's ability to buffer further loads of metabolic acid (see Clin. Corr. 25.9) has led to several ways of  expressing what the [HCO3–] would be if there were no respiratory component or respiratory compensation involved in a patient's condition. Base excess is one of  these expressions. It is defined as the amount of acid that would have to be added to blood to titrate it to pH 7.4 at a  , only the metabolic contribution to acid– base imbalance (primary metabolic imbalance and nonrespiratory compensatory processes) would be measured. If a blood sample were acidic under the conditions of  the titration, alkali would have to be added instead of acid, and the base excess would be negative. The concept and the quantitation of base excess are most easily understood from the pH–bicarbonate diagram. In our discussion of the blood buffer line we saw how  increasing the   in blood, where other buffers are present, would result in a rise in [HCO3–] and a virtually identical decrease in the concentration of other buffer  bases. This was because equivalent amounts of the other buffer bases were consumed as they buffered carbonic acid. Since virtually all the carbonic acid formed was  buffered, for every HCO3– formed one conjugate base of some other system was consumed. In this situation the total base in the blood is not measurably changed;  only the distribution of HCO3– and nonbicarbonate buffer conjugate base is changed. Thus, as long as one remains on the blood buffer line, [HCO3–] can change but  total base will not. There will be no positive or negative base excess. If, however, renal activity, diet, or some metabolic process adds or removes HCO3–, then a positive or negative base excess will occur. The patient's status will no  longer be described by a point on the buffer line, and the base excess will be the difference between the observed plasma [HCO3–] and the [HCO3–] on the buffer line  at the same pH (Figure 25.26). To calculate this difference, the position of the buffer line, which can be determined from knowledge of the slope and the point  representing the normal state, must be known. In the

 

  

Page 1050

clinical laboratory it can be estimated from hemoglobin concentration and assuming that it is the major nonbicarbonate buffer.

Figure 25.26  Calculation of base excess for a  point above the blood buffer line, and  calculation of negative base excess for a  point below the blood buffer line.  Base excess is 32 – 28 = 4 meq  L–1. Negative base excess  is 30 – 8 = 22 meq L–1.

The buffer line, then, is the dividing line between positive and negative base excess. Any point above it is in the region of positive base excess, and any point below it is  in the region of negative base excess. This gives rise to situations that may seem peculiar at first. In Figure 25.27 the [HCO3–] at point A is normal, but the patient has a  negative base excess. A positive or negative base excess occurs as a result of compensation for a respiratory acid–base imbalance or directly from a metabolic one.  Respiratory compensation for a metabolic acid–base imbalance, since it involves movement along a line parallel to the buffer line (Figure 25.25), would cause no  further change in the value of the base excess. Clinical Correlation 25.10 involves consideration of base excess.

Figure 25.27  Examples showing the sign of the  base excess at various points.  At points A and C there is a negative base  excess. At point B the base excess is positive.

25.14—  The Significance of Na+ and Cl– in Acid–Base Imbalance An important concept in diagnosing certain acid–base disorders is the anion gap. Most clinical laboratories routinely measure plasma Na+, K+, Cl–, and HCO3–. A  glance at the graph in Figure 25.15 confirms that in the plasma of a normal individual the sum of the concentrations of Na+ and K+ is greater than the sum of the concentrations of Cl– and  HCO3–. This difference is called A, the anion gap; it represents the other plasma anions (Figure 25.15), which are not routinely measured. It is calculated as follows:

The normal value of A is in the range of 12–16 meq L–1. In some clinical laboratories K+ is not measured; then the normal value is 8–12 meq L–1. The gap is changed  only by conditions that change the sum of the cations or the sum of the anions, or by conditions that change both sums by different amounts. Thus administration or  depletion of sodium bicarbonate would not change the anion gap because [Na+] and [HCO3–] would be affected equally. Metabolic acidosis due to HCl or NH4Cl  administration would also leave the anion gap unaffected; here [HCO3–] would decrease, but [Cl–] would increase by an equivalent amount, and the sum of [HCO3–]  plus [Cl–] would be unchanged. In contrast, diabetic ketoacidosis or methanol poisoning involves production of organic acids, which react with HCO3–, decreasing its  concentration. But since the [HCO3–] is replaced by some organic anion, the sum of [HCO3–] plus [Cl–] decreases, and the anion gap increases. The anion gap is most commonly used to establish a differential diagnosis for metabolic acidosis. In a metabolic acidosis with an increased anion gap, H+ must have  arisen in the body with some anion other than chloride. Metabolic acidosis without an increased anion gap must be due either to accumulation of H+ with chloride or to  a decrease in the concentration of sodium bicarbonate. Thus, on the basis of the anion gap, certain diseases can be ruled out, while others would have to be  considered. This information can be especially important in dealing with patients who cannot give good histories due to language barriers, unconsciousness, and so on. Electrolytes of body fluids interact in a multitude of ways. One important way involves the capacity of K+ and H+ to substitute for one another under certain  circumstances. This can occur in cells, where K+ is the major cation. In acidosis intracellular [H+] rises, and it replaces some of the intracellular K+. The displaced K+  appears in plasma and is excreted by the kidneys. This leaves the patient with normal plasma [K+] (normokalemia), but with seriously depleted body K+ stores  (hypokalia). Subsequent excessively rapid correction of the acidosis may then reverse events. As plasma pH rises, K+ flows back into the cells, and plasma [K+] may  decline to the point where muscular weakness sets in and respiratory insufficiency may become life­threatening.

 

  

Page 1051

CLINICAL CORRELATION 25.10  Evaluation of Clinical Acid–Base Data In a 1972 study of total parenteral nutrition of infants, it was found that infants who  received amino acids in the form of a hydrolysate of the protein fibrin maintained normal  acid–base balance. In contrast, infants receiving two different mixtures of synthetic amino  acids, FreAmine and Neoaminosol, became acidotic. Both synthetic mixtures contained  adequate amounts of all the essential amino acids, but neither contained aspartate or  glutamate. The fibrin hydrolysate contained all of the common amino acids. The accompanying figure shows the blood acid–base data from these infants. Note that  the normal values for infants, given by the dashed lines, are not quite the same as normal  values for adults. (A child is not a small adult.) The blood pH data show that the infants  receiving synthetic mixtures were clearly acidotic. The low [HCO3–] of the Neoaminosol  group immediately suggests a metabolic acidosis, and the   values indicate respiratory  acidosis, but a simple respiratory acidosis should be associated with a slightly elevated  [HCO3–]. The absence of this finding in most of the infants indicates that the acidosis must  also have a metabolic component. This is confirmed by the observation that all the infants  receiving FreAmine have a significant negative base excess. The infants with mixed acid–base disturbances did, in fact, have pneumonia or respiratory  distress syndrome. The metabolic acidosis, which all the infants receiving synthetic  mixtures experienced, was due to synthesis of aspartic acid and glutamic acid from a  neutral starting material (presumably glucose). Subsequent incorporation of these acids  into body protein imposed a net acid load on the body. Addition of aspartate and/or  glutamate to the synthetic mixtures was proposed as a solution of the problem.

Blood acid–base data of patients receiving fibrin hydrolysate   and of those receiving synthetic L­amino acid mixtures, FreAmine  .  Values are those observed at the time of the lowest blood base excess.  Dashed lines represent accepted normal values for infants.  Adapted from W. C. Heird, N. Engl. J. Med. 287:943, 1972.

  In kidneys the reciprocal relationship between K+ and H+ results in an association between metabolic alkalosis and hypokalemia. If hypokalemia arises from long­term  insufficiency of dietary potassium or long­term diuretic therapy, intracellular K+ levels diminish, and intracellular [H+] will increase. This leads to increased acid  excretion, acidic urine, and an alkaline arterial plasma pH. We have already seen how in an alkalotic individual a hormonal signal to absorb Na+ can lead to K+ loss  and then to an exacerbation of the metabolic alkalosis (p. 1045). The opposite also occurs, with alkalosis leading to hypokalemia. In this case increased amounts of  Na+ + HCO3– are presented to the distal convoluted tubules, where all K+ secretion normally takes place (all filtered K+ is reabsorbed; K+ loss is due to distal tubular secretion). The  distal tubules take up some Na+ but since HCO3– does not readily follow across that membrane, the increased Na+ uptake is linked to increased K+ secretion. K+ excretion is complicated,  being controlled by a variety of hormones and other

 

  

Page 1052

CLINICAL CORRELATION 25.11  Metabolic Alkalosis Prolonged gastric lavage produces a metabolic alkalosis that is a good experimental  model of the metabolic alkalosis that results from repeated vomiting. The following table  gives plasma and urine acid–base and electrolyte data from a healthy volunteer on a low­ sodium diet who, after a control period, was subjected to gastric lavage for two days.  After a five day recovery period, he was placed on a low­potassium diet and given a  sodium (130 meq day–1) and chloride (121 meq day–1) supplement. During the control  period the data are within normal limits. After gastric lavage that selectively removed HCl  (Na+ , K+ , and H2O lost with the gastric juice were restored), an uncomplicated metabolic  alkalosis developed. Note that the subject excreted an alkaline urine, containing a  substantial amount of HCO3–]. The Na+  excretion increased, depleting the body's Na+   stores. Plasma   was not measured, but plotting the values of pH and [HCO3–] on a  pH­bicarbonate diagram (e.g., Figure 25.18) allows one to interpolate a value of about  47 mmHg. Clearly, respiratory compensation was occurring. Plasma [K+ ] was  decreased. Plasma [Cl–] decreased, but no more than would be expected on the basis of  the changes in [Na+ ], [K+ ], and [HCO3–]. When the subject was placed on a low­potassium diet the alkalosis grew worse, and  plasma [HCO3–] rose. Additional compensatory hypoventilation evidently prevented a  further rise in plasma pH. Note, though, that the urine became acidic, in spite of the  increased severity of the alkalosis. The Na+  was conserved, not in exchange for K+ , but  in exchange for H+ . After several days of Na+  and Cl– administration, however, the  subject was able to restore the depleted Cl–, excrete the excess HCO3–, and repair the  acid–base imbalance with no other treatment. After  Lavage

  Control

Low  KCl

After  NaCl

Plasma pH

       7.4

      7.50

     7.48

      7.41

HCO3–

     29.3

    35.3

   38.1

    26.1

Na+ (meq L–1)

   138

  134

141

  144

K+ (meq L–1)

       4.2

      3.2

     2.9

      3.2

Cl– (meq L–1)

   101

    88

   85

  108

 

 

 

 

       6.12

      7.48

     5.70

      7.19

Urine pH – (meq/day–1)

HCO3

       3

    51

     1

    17

NH4

+  (meq/day–1)

     22

      4

   36

    14

Titratable acidity  (meq/day–1)

     10

      0

   14

      1

Total acidity (meq/day– 1 )

     29

  –49

   49

    –2

Na+ (meq/day–1)

       2

     28

     1

     95

 

Source: Data from Kassirer, J. P., and Schwartz, W. B., Am. J. Med. 40:10, 1966.

  factors. The end result, however, is that metabolic alkalosis and hypokalemia go hand in hand, so that the term "hypokalemic alkalosis" is often used synonymously  with metabolic alkalosis. Clinical Correlation 25.11 discusses a case of experimental metabolic alkalosis in which this occurred. Bibliography Gas Transport Bunn, H. F., and Forget, B. G. Hemoglobin: Molecular, Genetic, and Clinical Aspects. Philadelphia: Saunders, 1986. Bunn, H. F., Gabbay, K. H., and Gallop, P. M. The glycosylation of hemoglobin: relevance to diabetes mellitus. Science 200:21, 1978. Kilmartin, J. V. Interaction of haemoglobin with protons, CO2 and 2,3­diphosphoglycerate. Br. Med. Bull. 32:209, 1976. Perutz, M. F., and Lehmann, H. Molecular pathology of human haemoglobin. Nature 219:902, 1968. Steffes, M. W., and Mauer, S. M. Toward a basic understanding of diabetic complications. N. Engl. J. Med. 325:883, 1991. pH Regulation Davenport, H. W. The ABC of Acid–Base Chemistry, 6th ed. Chicago: University of Chicago Press, 1974. Gabow, P. A., Kaehny, W. D., Fennessey, P. V., et al. Diagnostic importance of an increased serum anion gap. N. Engl. J. Med. 303:854, 1980. Gamble, J. L. Jr., and Bettice, J. A. Acid­base relationships in the different body compartments: the basis for a simplified diagnostic approach. Johns Hopkins Med.  J. 140:213, 1977. Masoro, E.J., and Siegel, P. D. Acid–Base Regulation: Its Physiology, Patho­physiology and the Interpretation of Blood–Gas Analysis, 2nd ed. Philadelphia:  Saunders, 1977. Siggaard­Andersen, O. The Acid–Base Status of the Blood, 4th ed. Baltimore: Williams & Wilkins, 1974. Questions J. Baggott and C. N. Angstadt 1. During a breathing cycle: A. the alveolar gases are completely exchanged for atmospheric gases. B. gas exchange between the alveoli and the capillary blood can occur at all times. C. gas exchange with the capillary blood occurs at the surface of all the airways. D. there is net uptake of nitrogen by the blood. E. atmospheric water vapor is taken up by the lungs.

 

  

Page 1053

2. From an oxygen saturation curve for normal blood we can determine that: A. P50 is in the 

 range found in extrapulmonary tissues.

B. oxygen binding is hyperbolic. C. an oxygen carrier is necessary. D. tighter oxygen binding occurs at lower 

.

E. shifts of the curve to the left or right would have little effect on oxygen delivery. Refer to the following for Questions 3–5. A. hemoglobin a ­chains B. hemoglobin b ­chains C. hemoglobin g­chains D. hemoglobin  ­chains E. hemoglobin  ­chains 3. Found in HbA, HbA2, and HbF. 4. Modified in HbAic. 5. Form the major binding sites for 2,3­bisphosphoglycerate (BPG). 6. At a 

 of 30 mmHg hemoglobin's percent saturation will:

A. increase with increasing temperature. B. increase with decreasing pH. C. increase with increasing  D. increase with increasing 2,3­bisphosphoglycerate concentration. E. none of the above. 7. Significant contributors to the total carbon dioxide of whole blood include all of the following EXCEPT: A. bicarbonate ion. B. dissolved carbon dioxide (CO2). C. carbaminohemoglobin. D. carbonic acid (H2CO3). 8. 2,3­Bisphosphoglycerate (BPG): A. is absent from the normal erythrocyte. B. is a homotropic effector for hemoglobin. C. binds more tightly to HbF than to HbA. D. synthesis increases when hemoglobin's T 

 R equilibrium is shifted in favor of the T state.

E. synthesis decreases when the erythrocyte pH rises. 9. Which of the following buffer systems is far less effective in controlling changes in physiological pH due to CO2 than changes due to metabolic acids, like acetoacetic  acid? A. bicarbonate B. inorganic phosphate C. organic phosphate esters D. intracellular protein E. extracellular protein 10. The slope of the blood buffer line is most sensitive to pathological changes in the blood concentration of: A. plasma bicarbonate. B. plasma phosphate. C. hemoglobin. D. plasma proteins. E. organic phosphates of the erythrocyte. 11. As 

 is increased in a normal individual:

A. the plasma [CO2] remains unchanged. B. plasma bicarbonate increases. C. the slope of the blood buffer line changes. D. the base excess increases. E. the base excess decreases. 12. All of the following produce H+ EXCEPT: A. formation of bicarbonate ion from CO2 and water. B. formation of carbaminohemoglobin from CO2 and hemoglobin. C. binding of oxygen by hemoglobin. D. oxidation of sulfur­containing amino acids. E. metabolism of sodium lactate. 13. A substantial fraction of the urinary titratable acidity of a normal individual consists of: A. H2CO3. B. NH4+. C. acetoacetic acid. D. H2PO4–. E. HCO3–. 14. In a patient with diabetic ketoacidosis of long duration: A. the major urinary acid is H2PO4–. B. hemoglobin's oxygen dissociation curve would be shifted to the right. C. the distribution of hemoglobin species would be the same as in a normal individual. D. 1 mol of bicarbonate is regenerated for every mole of H2PO4– formed in the renal tubule. E. hypoventilation would be expected. 15. The following laboratory data are obtained from a patient:  buffer line. We conclude:

 = 60 mmHg, HCO3– = 27 meq L–1, pH = 7.28. These values define a point on the patient's blood 

A. The patient has an acute condition. B. The condition would lead to production of an alkaline urine. C. Of the blood buffers, the bicarbonate buffer system is the most important in resisting this pH change. D. Increasing the alveolar 

 could restore the plasma to normal.

E. Hyperventilation due to anxiety could cause this. 16. During compensation for a metabolic acid­base imbalance, which of the following would become increasingly abnormal? A. plasma pH B. blood  C. base excess D. total hemoglobin E. none of the above 17. In respiratory alkalosis: A. the acute state is associated with an abnormally low plasma [HCO3–]. B. the mechanism of compensation causes an increase in the plasma HCO3–. C. the plasma pH never returns to the normal range in the fully compensated state. D. in the partially compensated state, there will be a negative base excess equal to the difference between 24 meq L–1 and the actual plasma HCO3–. E. compensation involves changing 

 

  

Page 1054

18. Hypokalemia can be expected to: A. occur if the plasma pH is rapidly raised. B. lead to increased urine acidity. C. be associated with a high plasma [HCO3–]. D. decrease the value of the anion gap slightly. E. all of the above. Answers 1. B A and C: The alveoli, where gas exchange with the blood occurs, are of constant size and exchange gases with the airways by diffusion. D and E: Water vapor  and CO2 are added to the alveolar gases by the lung tissue, diluting the nitrogen (p. 1027). 2. A 

 (p. 1028, Figure 25.2). C: If O2 were soluble enough in plasma, no carrier would be necessary (p. 1026). E: Shifts profoundly affect delivery (p. 1028).

3. A It is the non­ a ­chain that differs among these (p. 1030). 4. B The b ­chains are nonenzymatically glycosylated in HbA1c (p. 1030). 5. B BPG binds between the N terminals of the b ­chains (p. 1029). 6. E All effects are opposite to those proposed in the question. A­C: High temperature, low pH (and therefore high  saturation (p. 1031). D: High BPG has the same effect (p. 1029).

 favor dissociation; that is, decreased 

7. D Carbonic acid is present in very small amounts; the equilibrium strongly favors CO2 and H2O (p. 1033, Table 25.2; see also p. 1039). 8. D A and B: BPG is a normal component of the red cell, where it serves as a heterotropic effector of HbA (p. 1029, Figure 25.4). C: It binds weakly or not at all to  the HbF (p. 1030). D and E: BPG binds to the T state, relieving product inhibition of BPG synthesis; BPG synthesis is inhibited by low pH (p. 1036). 9. A The bicarbonate system is a major extracellular buffer; with a pK of 6.1 it is ineffective toward CO2. The other buffers (phosphates and protein) are, effective (p.  1037, Table 25.6). All of these buffers, however, are effective against noncarbonic acids. The bicarbonate buffer system is included here because the response of the  respiratory system to low pH, exhaling CO2, compensates for the innate ineffectiveness of this system at a pH fairly distant from its pK (p. 1039). 10. C The slope of the blood buffer line is determined by the concentration of the nonbicarbonate buffers. Of these, hemoglobin is quantitatively the most important  and is susceptible to change (i.e., anemias from any cause) (p. 1040). 11. B This is because the resulting H2CO3 is buffered by various nonbicarbonate buffers, producing HCO3– (p. 1040). 12. E Metabolism of sodium lactate produces sodium bicarbonate and is used clinically to control acidosis (p. 1042). A and B are reactions whose products include  H+ (pp. 1032–1033). C is the Bohr effect (p. 1031). D is a major source of acid in the typical American diet (p. 1042). 13. D A: The level of H2CO3 is very low. B: NH4+ is an important urinary acid, but its pK is too high to be titrated at pH 7.4, the endpoint. C: Acetoacetic acid would  appear only in some kinds of severe acidosis. E: HCO3– is physiologically a base; its dissociable H+ has a pK that is far above the physiological range (pp. 1043 and  1045). 14. D See pp. 1043–1045. A: After adaptation to acidosis NH4+ excretion rises enormously, becoming the major urinary acid (p. 1045). B: True only in acidosis of  short duration; decreasing BPG in prolonged acidosis tends to restore the normal position (p. 1036). C: Large amounts of HbA1c would be expected (p. 1030). E:  Hyperventilation, to expel CO2, would be expected (p. 1048). 15. A High  , low pH point on the blood buffer line define an acute respiratory acidosis. Buffering by nonbicarbonate buffer systems and excretion of acid in the  urine would be the physiological responses (p. 1047). 16. B   would decrease during compensation for acidosis or rise during compensation for alkalosis. A: Plasma pH would be restored. C: Base excess would be  unchanged. D: Hemoglobin would participate in buffering, but its total concentration would not be expected to change (pp. 1048–1050, Figures 25.25 and 25.26). 17. A A and B: See p. 1047, Figure 25.24. C: This is the only acid­base abnormality in which compensation is expected to restore the plasma pH to 7.4 (p. 1047). D:  There is a negative base excess equivalent to the difference between the patient's [HCO3–] and the [HCO3–] of the point on the blood buffer line at the same pH, a  point that will be less than 24 meq L–1 (p. 1050, Figure 25.26). E: This would either be a cure or an exacerbation, depending on the direction of the change; it would  not be compensation. 18. E A, B, and C: See p. 1050. D: Decreasing K+ would lower the anion gap by a small amount (p. 1050).

 

  

Page 1055

Chapter 26—  Digestion and Absorption of Basic Nutritional Constituents Ulrich Hopfer

26.1 Overview

1056

Gastrointestinal Organs Have Multiple Functions in Digestion

1057

26.2 Digestion: General Considerations

1059

Pancreas Supplies Enzymes for Intestinal Digestion

1059

Digestive Enzymes Are Secreted as Proenzymes

1059

Regulation of Secretion Occurs through Secretagogues

1060

26.3 Epithelial Transport

1063

Solute Transport May Be Transcellular or Paracellular

1063

NaCl Absorption Has Both Active and Passive Components

1064

NaCl Secretion Depends on Contraluminal Na+,K+–ATPase

1066

Concentration Gradients or Electrical Potentials Drive Transport of  Nutrients

1067

Gastric Parietal Cells Secrete HCl

1069

26.4 Digestion and Absorption of Proteins

1070

Mixture of Peptidases Assures Efficient Protein Digestion

1070

Pepsins Catalyze Gastric Digestion of Protein

1070

Pancreatic Zymogens Are Activated in Small Intestine

1071

Intestinal Peptidases Digest Small Peptides

1072

Free Amino Acids and Dipeptides Are Absorbed by Carrier­Mediated  Transport

1072

Fetus and Neonate Can Absorb Intact Proteins

1073

26.5 Digestion and Absorption of Carbohydrates

1073

Di­ and Polysaccharides Require Hydrolysis

1073

Monosaccharides Are Absorbed by Carrier­Mediated Transport

1076

26.6 Digestion and Absorption of Lipids Lipid Digestion Requires Overcoming the Limited Water Solubility of  Lipids

1077

Lipids Are Digested by Gastric and Pancreatic Lipases

1078

Bile Acid Micelles Solubilize Lipids during Digestion

1078

Most Absorbed Lipids Are Incorporated into Chylomicrons

1082

26.7 Bile Acid Metabolism

1083

Bibliography

1085

Questions and Answers

1085

Clinical Correlations

   

  

1077

 

26.1 Cystic Fibrosis

1067

26.2 Bacterial Toxigenic Diarrheas and Electrolyte Replacement  Therapy

1068

26.3 Neutral Amino Aciduria (Hartnup Disease)

1073

26.4 Disaccharidase Deficiency

1075

26.5 Cholesterol Stones

1081

26.6 A­ b ­Lipoproteinemia

1082

Page 1056

26.1—  Overview Secretion of digestive fluids and digestion of food were some of the earliest biochemical events to be investigated at the beginning of the era of modern science. Major  milestones were the discovery of hydrochloric acid secretion by the stomach and enzymatic hydrolysis of protein and starch by gastric juice and saliva, respectively.  The discovery of gastric HCl production goes back to the American physician William Beaumont (1785–1853). In 1822 he treated a patient with a stomach wound.  The patient recovered from the wound, but retained a gastric fistula (abnormal opening through the skin). Beaumont seized the opportunity to obtain and study gastric  juice at different times during and after meals. Chemical analysis revealed, to the surprise of chemists and biologists, the presence of the inorganic acid HCl. This  discovery established the principle of unique secretions into the gastrointestinal tract, which are elaborated by specialized glands. Soon thereafter, the principle of enzymatic breakdown of food was recognized. In 1836 Theodor Schwann, a German anatomist and physiologist (1810–1882),  noticed that gastric juice degraded albumin in the presence of dilute acid. He recognized that a new principle was involved and coined the word pepsin from the Greek  pepsis, meaning digestion. Today the process of secretion of digestive fluids, digestion of food, and absorption of nutrients and of electrolytes can be described in  considerable detail. The basic nutrients fall into the classes of proteins, carbohydrates, and fats. Many different types of food can satisfy the nutritional needs of humans, even though they  differ in the ratios of proteins to carbohydrates and to fats and in the ratio of digestible to nondigestible materials. Unprocessed plant products are especially rich in  fibrous material that can be neither digested by human enzymes nor easily degraded by intestinal bacteria. The fibers are mostly carbohydrates, such as cellulose  (b ­1,4­glucan) or pectins (mixtures of methyl esters of polygalacturonic acid, polygalactose, and polyarabinose). High­fiber diets enjoy a certain popularity nowadays  because of a postulated preventive effect on development of colonic cancer. 26.1 describes average contributions of different food classes to the diet of North Americans. The intake of individuals may substantially deviate from the average, as  food consumption depends mainly on availability and individual tastes. The ability to utilize a wide variety of food is possible because of the great adaptability and  digestive reserve capacity of the gastrointestinal tract. Knowledge of the nature of proteins and carbohydrates in the diet is important from a clinical point of view. Certain proteins and carbohydrates, although good  nutrients for most humans, cannot be properly digested by some individuals and produce gastrointestinal ailments. Omission of the offending TABLE 26.1 Contribution of Major Food Groups to Daily Nutrient Supplies in the United States

Total Daily  Consumption  (g)

Dairy  Products,  Except Butter    (%)

Protein

100

22

Carbohydrate

381

  7

Fat

155

13

35

Type of  Nutrient

   

  

Meat,  Poultry,  Fish  (%)

Eggs  (%)

Fruits, Nuts,  Vegetables  (%)

Flour,  Cereal  (%)

Sugar  Sweeteners  (%)

Fats,  Oils (%)

42

6

12

18

  0

  0

      0.1

    0.1

19

36

37

  0

3

  4

  1

  0

42

Page 1057

material and change to another diet can eliminate these gastrointestinal problems. Examples of food constituents that can be the cause of gastrointestinal disorders are  gluten, one of the protein fractions of wheat, and lactose, the disaccharide in milk. Gastrointestinal Organs Have Multiple Functions in Digestion The bulk of ingested nutrients consists of large polymers that have to be broken down to monomers before they can be absorbed and made available to all cells of  the body. The complete process from food intake to absorption of nutrients into the blood consists of a complicated sequence of events, which at the minimum  includes (Figure 26.1): 1. Mechanical homogenization of food and mixing of ingested solids with fluids secreted by the glands of the gastrointestinal tract. 2. Secretion of digestive enzymes that hydrolyze macromolecules to oligomers, dimers, or monomers. 3. Secretion of electrolytes, acid, or base to provide an appropriate environment for optimal enzymatic digestion. 4. Secretion of bile acids as detergents to solubilize lipids and facilitate their absorption. 5. Hydrolysis of nutrient oligomers and dimers by enzymes on the intestinal surface. 6. Transport of nutrient molecules and of electrolytes from the intestinal lumen across the epithelial cells into blood or lymph. To accomplish these functions, the gastrointestinal tract contains specialized glands and surface epithelia: Organ

Major Function in Digestion and Absorption

Salivary glands

Elaboration of fluid and digestive enzymes

Stomach

Elaboration of HCl and proteases

Pancreas

Elaboration of NaHCO3 and enzymes for intraluminal  digestion

Liver

Elaboration of bile acids

Gallbladder

Storage and concentration of bile

Small intestine

Terminal digestion of food, absorption of nutrients and  electrolytes

Large intestine

Absorption of electrolytes

  The pancreas and small intestine are essential for digestion and absorption of all basic nutrients. Fortunately, both organs have large reserve capacities. For  example, maldigestion due to pancreatic failure becomes a problem only when the pancreatic secretion rate of digestive enzymes drops below one­tenth of the normal  rate. The secretion of the liver (bile) is important for efficient lipid absorption, which depends on the presence of bile acids. In contrast, gastric digestion of food is  nonessential for adequate nutrition, and loss of this function can be compensated for by the pancreas and the small intestine. Yet normal gastric digestion greatly  increases the smoothness and efficiency of the total digestive process. The stomach aids in the digestion through its reservoir function, its churning ability, and initiation  of protein hydrolysis, which, although small, is important for stimulation of pancreatic and gallbladder output. Peptides and amino acids liberated in the stomach  stimulate the coordinated release of pancreatic juice and bile into the lumen of the small intestine, thereby ensuring efficient digestion of food.

 

  

Page 1058

Figure 26.1  Gastrointestinal organs and their functions.

 

  

Page 1059

26.2—  Digestion:  General Considerations Pancreas Supplies Enzymes for Intestinal Digestion Most of the breakdown of food is catalyzed by soluble enzymes and occurs within the lumen of the stomach or small intestine. The pancreas, not the stomach, is the  major organ that synthesizes and secretes the large amounts of enzymes needed for digestion. Secreted enzymes amount to at least 30 g of protein per day in a healthy  adult. The pancreatic enzymes together with bile are poured into the lumen of the second (descending) part of the duodenum, so that the bulk of the intraluminal  digestion occurs distal to this site in the small intestine. However, pancreatic enzymes cannot completely digest all nutrients to forms that can be absorbed. Even after  exhaustive contact with pancreatic enzymes, a substantial portion of carbohydrates and amino acids are present as dimers and oligomers that depend for final digestion  on enzymes present on the luminal surface or within the chief epithelial cells that line the lumen of the small intestine (enterocytes). The luminal plasma membrane of enterocytes is enlarged by a regular array of projections, termed microvilli, which give it the appearance of a brush and have led to  the name brush border for the luminal pole of enterocytes. This membrane contains many di­ and oligosaccharidases, amino­ and dipeptidases, as well as  esterases (Table 26.2). Many of these enzymes protrude up to 100 Å into the intestinal lumen, attached to the plasma membrane by an anchoring polypeptide that  itself has no role in the hydrolytic activity. The substrates for these enzymes are the oligomers and dimers that result from pancreatic digestion. The surface enzymes are  glycoproteins that are relatively stable against digestion by pancreatic proteases or the effects of detergents. A third site of digestion is the cytoplasm of enterocytes. Intracellular digestion is of some importance for the hydrolysis of di­ and tripeptides, which can be  absorbed across the luminal plasma membrane. Digestive Enzymes Are Secreted as Proenzymes Salivary glands, gastric mucosa, and pancreas contain specialized cells that synthesize and store digestive enzymes until the enzymes are needed during TABLE 26.2 Digestive Enzymes of the Small Intestinal Surface Enzyme (Common Name)

Substrate

Maltase

Maltose

Sucrase/isomaltase

Sucrose/a ­limit dextrin

Glucoamylase

Amylose

Trehalase

Trehalose

b ­Glucosidase

Glucosylceramide

Lactase

Lactose

Endopeptidase 24.11

Protein (cleavage at internal hydrophobic amino  acids)

Aminopeptidase A

Oligopeptide with acidic NH2 terminus

Aminopeptidase N

Oligopeptide with neutral NH2 terminus

Dipeptidyl aminopeptidase IV

Oligopeptide with X­Pro or X­Ala at NH2 terminus

Leucine aminopeptidase

Peptides with neutral amino acid at NH2 terminus

g ­Glutamyltransferase

Glutathione + amino acid

Enteropeptidase (enterokinase)

Trypsinogen

Alkaline phosphatase

Organic phosphates

   

  

Page 1060

Figure 26.2  Exocrine secretion of digestive enzymes.  Redrawn with permission from Jamieson, J. D. Membrane  and secretion. In: G. Weissmann and R. Claiborne (Eds.),  Cell Membranes: Biochemistry, Cell Biology and Pathology.  New York: HP Publishing Co., 1975. Figure by B. Tagawa.

a meal. The enzymes are then released into the lumen of the gastrointestinal tract (Figure 26.2). This secretion is termed exocrine because of its direction toward the  lumen. Proteins destined for secretion are synthesized on the polysomes of the rough endoplasmic reticulum (see p. 739 for synthesis and glycosylation of membrane  and secreted proteins) and transported via the Golgi complex to storage vesicles in the apical cytoplasm. The storage vesicles (zymogen granules) have a diameter of  about 1  m. Most digestive enzymes are produced and stored as inactive proenzymes (zymogens) (see p. 101). The zymogen granules are bounded by a typical  cellular membrane. When an appropriate stimulus for secretion is received by the cell, the granules move closer to the luminal plasma membrane, where their  membranes fuse with the plasma membrane and release the contents into the lumen (exocytosis). Activation of proenzymes occurs only after they are released from  the cells. Regulation of Secretion Occurs through Secretagogues The processes involved in the secretion of enzymes and electrolytes are regulated and coordinated. Elaboration of electrolytes and fluids simultaneously with that of  enzymes is required to flush any discharged digestive enzymes out of the gland into the gastrointestinal lumen. The physiological regulation of secretion occurs through  secretagogues that interact with receptors on the surface of the exocrine cells (Table 26.3). Neurotransmitters, hormones, pharmacological agents, and certain  bacterial toxins can be secretagogues. Different exocrine cells, for example, in different glands, usually possess different sets of receptors. Binding of the secretagogues  to receptors sets off a chain of signaling events that ends with fusion of zymogen granules with the plasma membrane. Two major signaling pathways have been  identified (Figure 26.3): (1) activation of phosphatidylinositol­specific phospholipase C with liberation of inositol 1,4,5­triphosphate and diacylglycerol (see p.  862); in turn, triggering Ca2+ release into the cytosol and activation of protein kinase C, respec­

 

  

Page 1061

Figure 26.3  Cellular regulation of exocrine secretion in the pancreas.  Abbreviations: PI­4,5P2, phosphatidylinositol­4,5­bisphosphate;  DG, diacylglycerol; IP , inositol­1,4,5­triphosphate; PLC,  3

phospolipase C. Adapted from Gardner, J. D. Annu. Rev.  Physiol. 41:63, 1979.  Copyright © 1979 by Annual Reviews, Inc.

tively; and (2) activation of adenylate or guanylate cyclase, resulting in elevated cAMP or cGMP levels, respectively (see p. 859). Secretion can be stimulated  through either pathway. Acetylcholine (Figure 26.4) elicits salivary, gastric, and pancreatic enzyme and electrolyte secretion. It is the major neurotransmitter for stimulating secretion, with  input from the central nervous system in salivary and gastric glands, or via local reflexes in gastric glands and the pancreas. The acetylcholine receptor of exocrine cells  is of the muscarinic type; that is, it can be blocked by atropine (Figure 26.5). Most people have experienced the effect of atropine because it is used by dentists to ''dry  up" the mouth for dental work.

Figure 26.4  Acetylcholine.

Another class of secretagogues are the biogenic amines, consisting of histamine and 5­hydroxytryptamine. Histamine (Figure 26.6) is a potent stimulator of HCl  secretion. It interacts with a gastric­specific histamine receptor, also referred to as the H2 receptor, on the contraluminal plasma membrane of parietal cells. Histamine  is normally secreted by specialized regulatory cells in the stomach wall (enterochromaffin­like cells, ECC). Histamine analogs that are antagonists at the H2 receptor  are used medically to decrease HCl output during treatment for peptic ulcers. 5­Hydroxytryptamine (serotonin) is pres­

Figure 26.5  (a) L(+)­Muscarine and (b) atropine.

Figure 26.6  Histamine. TABLE 26.3 Physiological Secretagogues Organ

Secretion

Secretagogue

Salivary gland

NaCl, amylase

Acetylcholine, (catecholamines?)

Stomach

HCl, pepsinogen

Acetylcholine, histamine, gastrin

Pancreas—acini

NaCl, digestive enzymes Acetylcholine, cholecystokinin (secretin)

Pancreas—duct

NaHCO3, NaCl

Secretin

Small intestine

NaCl

Acetylcholine, serotonin, vasoactive  intestinal peptide (VIP), guanylin

   

  

Page 1062

ent in relatively high amounts in the gastrointestinal tract (Figure 26.7). It stimulates secretion of NaCl by the small intestinal mucosa.

Figure 26.7  5­OH­Tryptamine (serotonin).

A third class of secretagogues consists of peptide­neurotransmitters and ­hormones (Table 26.4). The intestinal nerve cells are rich in peptide­neurotransmitters that  stimulate NaCl secretion. Vasoactive intestinal peptide (VIP) is a particularly potent one in this respect in the intestines and pancreas. Furthermore, the  gastrointestinal tract contains many specialized epithelial cells that produce biologically active amines and peptides. The peptides are localized in granules, usually close  to the contraluminal pole of these cells, and are released into the blood. Hence these cells are classified as epithelial endocrine cells. Of particular importance are the  peptides gastrin, cholecystokinin (pancreozymin), and secretin. In contrast, a recently identified peptide—namely, guanylin—is released into the lumen and stimulates  NaCl secretion by binding to a brush border receptor that activates guanylate cyclase and thus elevates cGMP levels. Gastrin occurs as either a peptide of 34 amino acids (G­34) or one of 17 residues (G­17) from the COOH terminus of G­34. The functional portion of gastrin resides  mainly in the last five amino acids of the COOH terminus. Thus pentagastrin, an artificial pentapeptide containing only the last five amino acids, can be used specifically  to stimulate gastric HCl and pepsin secretion. Gastrin as well as cholecystokinin have an interesting chemical feature, a sulfated tyrosine, which considerably  enhances the potency of both hormones. Cholecystokinin and pancreozymin denote the same peptide. The different names allude to the different functions elicited by the peptide and had been coined before  purification. The peptide stimulates gallbladder contraction (cholecystokinin) as well as secretion of pancreatic enzymes (pancreozymin). It is secreted by epithelial  endocrine cells of the small intestine, particularly in the duodenum, and this secretion is stimulated by luminal amino acids and peptides, usually derived from gastric  proteolysis, by fatty acids, and by an acid pH. Cholecystokinin and gastrin are thought to be related in an evolutionary sense, as both share an identical amino acid  sequence at the COOH terminus.

 

  

Page 1063

Secretin is a polypeptide of 27 amino acids. This peptide is secreted by yet other endocrine cells of the small intestine. Its secretion is stimulated particularly by luminal  pH less than 5. The major biological activity of secretin is stimulation of secretion of pancreatic juice rich in NaHCO3. Pancreatic NaHCO3 is essential for neutralization  of gastric HCl in the duodenum. Secretin also enhances pancreatic enzyme release, acting synergistically with cholecystokinin. 26.3—  Epithelial Transport Solute Transport May Be Transcellular or Paracellular Solute movement across an epithelial cell layer is determined by the properties of epithelial cells, particularly their plasma membranes, and by the intercellular tight  junctional complexes (Figure 26.8). The tight junctions extend in a belt­like manner around the perimeter of each epithelial cell and connect neighboring cells.  Therefore the tight junctions constitute part of the barrier between the two extracellular spaces on either side of the epithelium, that is, the lumen of the gastrointestinal  tract and the intercellular (interstitial) space on the other (blood or serosal) side. The tight junction marks the boundary between the luminal and contraluminal region of  the plasma membrane of epithelial cells. Two potentially parallel pathways for solute transport across epithelial cell layers can be distinguished: through the cells (transcellular) and through the tight  junctions between cells (paracellular) (Figure 26.8). The transcellular route in turn consists mainly of two barriers in series, formed by the luminal and contraluminal  plasma membranes. Because of this combination of different barriers in parallel (cellular and paracellular pathways) and in series (luminal and contraluminal plasma  membranes), biochemical and biophysical information on all three barriers as well as their mutual influence is required for understanding the overall transport properties  of the epithelium. A major function of gastrointestinal epithelial cells is active transport of nutrients, electrolytes, and vitamins. The cellular basis for this vectorial solute movement lies in  the different properties of the luminal and contraluminal regions of the plasma membrane. The small intestinal cells provide a prominent example of the differentiation  and specialization of the two types of membrane. The luminal and contraluminal plasma membranes differ in morphological appearance, enzymatic composition,  chemical composition, and transport functions (Table 26.5). The luminal membrane is in contact with the nutrients in the chyme (the semifluid mass of partially digested  food) and is specialized for terminal digestion of nutrients through its digestive enzymes and for nutrient absorption through transport systems that accomplish  concentrative uptake. Transport systems are present for monosaccharides, amino acids, peptides, and electrolytes. In contrast, the contraluminal plasma membrane,  which is in contact with the intercellular fluid, capillaries, and lymph, has properties similar to the

Figure 26.8  Pathways for transport across epithelia.

 

  

Page 1064 TABLE 26.5 Characteristic Differences Between Luminal and Contraluminal Plasma  Membrane of Small Intestinal Epithelial Cells Parameter

Luminal

Contraluminal

Morphological  appearance

Microvilli in ordered arrangement (brush  border)

Few microvilli

Enzymes

Di­ and oligosaccharidases

Na+,K+–ATPase

 

Aminopeptidase

Adenylyl cyclase

 

Dipeptidases

 

 

g ­Glutamyltransferase

 

 

Alkaline phosphatase

 

 

Guanylate cyclase

 

Transport systems

Na+–monosaccharide cotransport  (SGLT1)

Facilitated monosac­ charide  transport (GLUT2)

 

Facilitated fructose transport (GLUT5)

Facilitated neutral amino acid  transport

 

Na+–neutral amino acid cotransport

 

 

Na+–bile acid cotransport

 

  plasma membrane of most cells. It possesses receptors for hormonal or neuronal regulation of cellular functions, a Na+,K+–ATPase for removal of Na+ from the cell,  and transport systems for the entry of nutrients for consumption by the cell. In addition, the contraluminal plasma membrane contains the transport systems necessary  for exit of the nutrients derived from the lumen so that the digested food can become available to all cells of the body. Some of the transport systems in the  contraluminal plasma membrane may fulfill both the function of catalyzing exit when the intracellular nutrient concentration is high after a meal and that of mediating their  entry when the blood levels are higher than those within the cell. NaCl Absorption Has Both Active and Passive Components Transport of Na+ plays a crucial role not only for epithelial NaCl absorption or secretion, but also in the energization of nutrient uptake. The Na+,K+–ATPase  provides the dominant mechanism for transduction of chemical energy in the form of ATP into osmotic energy of a concentration (chemical) or a combined  concentration and electrical (electrochemical) ion gradient across the plasma membrane. In epithelial cells this enzyme is located exclusively in the contraluminal plasma  membrane (Figure 26.9). The stoichiometry of the Na+,K+–ATPase reaction is 1 mol of ATP coupled to the outward pumping of 3 mol of Na+ and the simultaneous  inward pumping of 2 mol of K+. The Na+,K+–ATPase maintains the high K+ and low Na+ concentrations in the cytosol and is directly or indirectly responsible for an  electrical potential of about –60 mV of the cytoplasm relative to the extracellular solution. The direct contribution comes from the charge movement when 3Na+ ions  are replaced by 2K+; the indirect contribution is by way of the K+ gradient, which becomes the dominant force for establishing the potential by the movement of K+  through K+ channels. Transepithelial NaCl movements are produced by the combined actions of the Na+,K+–ATPase and additional "passive" transport systems in the plasma  membrane, which allow the entry of Na+ or Cl– into the cell. NaCl absorption results from Na+ entry into the cell across the luminal plasma membrane and its extrusion  by the Na+,K+–ATPase across the contraluminal membrane. Epithelial cells of the lower portion of the large intestine possess a luminal Na+ channel (epithelial Na+  channel or ENaC) that allows the uncoupled entry of Na+ down its electrochemical gradient (Figure 26.10). This Na+ flux is electrogenic; that is, it is associated  with an electrical current, and it can be inhibited by

 

  

Page 1065

Figure 26.9  Na+ concentrations and electrical potentials in enterocytes.

Figure 26.10  Model for electrogenic NaCl absorption in the  lower intestine.

the diuretic drug amiloride at micromolar concentrations (Figure 26.11). The presence of this transport system, and hence NaCl absorption, is regulated by  mineralocorticoid hormones of the adrenal cortex.

Figure 26.11  Amiloride.

Epithelial cells of the small intestine possess a transport system in the brush border membrane, which catalyzes an electrically neutral Na+/H+ exchange (Na/H  exchanger or NHE) (Figure 26.12). The exchange is not affected by low concentrations of amiloride and not regulated by mineralocorticoids. The Na+ absorption  secondarily drives Cl– absorption through a specific Cl–/HCO3– exchanger (anion exchanger or AE) in the luminal plasma membrane, as illustrated in Figure 26.12. The necessity for  two types of NaCl absorption may arise from the different functions of upper and lower intestine, which require different regulation. The upper intestine reabsorbs the bulk of NaCl from the  diet and from secretions of the exocrine glands after each meal, while the lower intestine participates in the fine regulation of NaCl retention, depending on the overall electrolyte balance of  the body.

Figure 26.12  Model for electrically neutral NaCl absorption in  the small intestine.

 

  

Page 1066

Figure 26.13  Ionic composition of secretions of the gastrointestinal tract.  Serum included for comparisons. Note the high H+ concentration  in gastric juice (pH + 1) and the high HCO3– concentration in pancreatic  , inorganic and organic sulfate; 

juice. P, organic and inorganic phosphate; SO4

Ca, calcium; Mg, magnesium; bile a., bile acids.  Adapted from Biological Handbooks. Blood and  Other Body Fluids. Federation of American Societies for Experimental  Biology, 1961.

NaCl Secretion Depends on Contraluminal Na+,K+–ATPase The epithelial cells of most regions of the gastrointestinal tract have the potential for electrolyte and fluid secretions. The major secreted ions are Na+ and Cl–. Water  follows passively because of the osmotic forces exerted by any secreted solute. Thus NaCl secretion secondarily results in fluid secretion. The fluid may be either  hypertonic or isotonic, depending on the contact time of the secreted fluid with the epithelium and the tissue permeability to water. The longer the contact and the  greater the water permeability, the closer the secreted fluid gets to osmotic equilibrium, that is, isotonicity. Ionic compositions of gastrointestinal secretions are  presented in Figure 26.13.

Figure 26.14  Model for epithelial NaCl secretion.

The cellular mechanisms for NaCl secretion involve the Na+,K+–ATPase located in the contraluminal plasma membrane of epithelial cells (Figure 26.14). The enzyme  is implicated because cardiac glycosides, inhibitors of this enzyme, abolish salt secretion. However, the involvement of Na+, K+–ATPase does not provide a  straightforward explanation for a NaCl movement from the capillary side to the lumen because the enzyme extrudes Na+ from the cell toward the capillary side. Thus  the active step of Na+ transport across one of the plasma membranes has a direction opposite to that of overall transepithelial NaCl movements. The apparent paradox  is resolved by an electrical coupling of Cl– secretion across the luminal plasma membrane and Na+ movements via the paracellular route, illustrated in Figure 26.14.  The Cl– secretion depends on coupled uptake of 2 Cl– ions with Na+ and K+ via a specific cotransporter in the contraluminal plasma membrane and specific luminal  Cl– channels. The Na+,K+,2 Cl–­cotransporter, which can be identified by specific inhibitors such as the common diuretic furosemide (Figure 26.15), utilizes the  energy of the Na+ gradient to accumulate Cl– within the cytoplasmic compartment above its electrochemical equilibrium concentration. Subsequent opening of luminal  Cl– channels allows efflux of Cl– together with a negative charge (see Clin. Corr. 26.1 and 26.2).

Figure 26.15  Furosemide.

In the pancreas a fluid rich in Na+ and Cl– is secreted by acinar cells. This fluid provides the vehicle for the movement of digestive enzymes from the acini, where they  are released, to the lumen of the duodenum. The fluid is modified in the ducts by the additional secretion of NaHCO3 (Figure 26.16). The HCO3– concentration in the  final pancreatic juice can reach concentrations of up to 120 mM.

The permeability of the tight junction to H2O, Na+ or other ions modifies

 

  

Page 1067

Figure 26.16  Model for epithelial NaHCO3 secretion.  Note that two different mechanisms for H+  secretion exist in the contraluminal plasma  membrane: (1) Na+/H+ exchange and  (2) H+–ATPase.

CLINICAL CORRELATION 26.1  Cystic Fibrosis Cystic fibrosis is an autosomal recessive inherited disease due to a mutation in the cystic  fibrosis transmembrane regulatory (CFTR) protein. This protein contains 1480 amino  acids organized into two membrane­spanning portions, which contain six transmembrane  regions each, two ATP­binding domains, and a regulatory domain that undergoes  phosphorylation by cAMP­dependent protein kinase. Some 400 mutations have been  discovered since the gene was cloned in 1989. The normal form of this protein is a Cl– channel that is found in the luminal plasma  membrane of epithelial cells in many tissues. The channel is normally closed but opens  when phosphorylated by protein kinase A, thus providing regulated Cl– and fluid  secretion. The most common and severe mutation lacks one amino acid ( F508 CFTR),  which prevents the protein from properly maturing and reaching the plasma membrane.  People who inherit this mutant CFTR from both parents lack Cl– and fluid secretion in  tissues that depend on CFTR for this function. Failure to secrete fluid, in turn, can lead to  gross organ impairment due to partial or total blockage of passageways, for example, the  ducts in the pancreas, the lumen of the intestine, or airways. (See Clin. Corr. 26.2 for  activation of the CFTR Cl– channel.)   active transepithelial solute movements. For example, a high permeability is necessary to allow Na+ to equilibrate between extracellular solutions of the intercellular and  luminal compartments during NaCl or NaHCO3 secretion. Different regions of the gastrointestinal tract differ not only with respect to the transport systems that  determine the passive entry (see above for amiloride­sensitive Na+ channel and Na+/H+ exchange), but also with respect to the permeability characteristics of the tight  junction. The distal portion (colon) is much tighter so as to prevent leakage of Na+ from blood to lumen, in accordance with its function of scavenging of NaCl from the  lumen. Concentration Gradients or Electrical Potentials Drive Transport of Nutrients Many solutes are absorbed across the intestinal epithelium against a concentration gradient. Energy for this "active" transport is directly derived from the Na+  concentration gradient or the electrical potential across the luminal plasma membrane, rather than from the chemical energy of a covalent bond change, such as ATP  hydrolysis. Glucose transport provides an example of uphill solute transport that is driven directly by the electrochemical Na+ gradient and only indirectly by ATP  (Figure 26.17). Glucose is absorbed from the intestinal lumen into the blood against a concentration gradient. This vectorial transport is the combined result of several separate  membrane events (Figure 26.18): (1) ATP­dependent Na+ transport out of the cell at the contraluminal pole that establishes an electrochemical Na+ gradient across  the plasma membrane; (2) K+ channels that convert a K+ gradient into a membrane potential; (3) asymmetric insertion of two different transport systems for glucose  into the luminal and contraluminal plasma membranes; and (4) coupling of Na+ and glucose transport across the luminal membrane. The luminal plasma membrane contains a transport system that facilitates a tightly coupled movement of Na+ and D­glucose or structurally similar sugars

 

  

Page 1068

CLINICAL CORRELATION 26.2  Bacterial Toxigenic Diarrheas and Electrolyte Replacement Therapy Voluminous, life­threatening intestinal electrolyte and fluid secretion (diarrhea) occurs in  patients with cholera, an intestinal infection by Vibrio cholerae. Certain strains of E. coli  also cause (traveler's!) diarrhea that can be serious in infants. The secretory state is a  result of enterotoxins produced by the bacteria. The mechanisms of action of some of  these enterotoxins are well understood at the biochemical level. Cholera toxin activates  adenylyl cyclase by causing ADP­ribosylation of the Gas­protein, which stimulates the  cyclase (see p. 859). Elevated cAMP levels in turn activate protein kinase A, which  opens the luminal CFTR Cl– channel and inhibits the Na+/H+ exchanger by protein  phosphorylation. The net result is gross NaCl secretion. Escherichia coli produces a  heat­stable toxin that binds to the receptor for the physiological peptide "guanylin,"  namely, the brush border guanylyl cyclase. When the receptor is occupied on the luminal  side by either guanylin or the heat­stable E. coli toxin, the guanylyl cyclase domain of the  protein on the cytosolic side is activated and cGMP levels rise. Elevated cGMP levels  have the same effect on Cl– secretion as elevated cAMP levels, except that a cGMP­ activated protein kinase is involved in protein phosphorylation. Modern, oral treatment of cholera takes advantage of the presence of Na+­glucose  cotransport in the intestine, which is not regulated by cAMP and remains fully active in  this disease. In this case, the presence of glucose allows uptake of Na+ to replenish body  NaCl. Composition of solution for oral treatment of cholera patients is glucose 110 mM,  Na+ 99 mM, Cl– 74 mM, HCO3– 29 mM, and K+ 4 mM. The major advantages of this form of  therapy are its low cost and ease of administration when compared with intravenous fluid therapy.

Carpenter, C. C. J. In: M. Field, J. S. Fordtran, and S. G. Schultz (Eds.), Secretory  Diarrhea. Bethesda, MD: American Physiological Society, 1980, pp. 67–83.  

Figure 26.17  Model for epithelial glucose absorption.

Figure 26.18  Transepithelial glucose transport as translocation reactions across the plasma  membranes and the tight junction.  SGLT1 (sodium glucose transporter 1) and GLUT2 (glucose transporter 2) are  specific intestinal gene products mediating Na+–glucose cotransport and  facilitated glucose transport, respectively. Numbers in the left column indicate  the minimal turnover of individual reactions to balance the overall reaction.

 

  

Page 1069 +

(Sodium GLucose Transporter or SGLT). The most common intestinal sodium­glucose cotransporter is SGLT1 and it couples the movement of 2 Na  ions with  that of 1 glucose molecule. It mediates glucose and Na+ transport equally well in both directions. However, because of the higher Na+ concentration in the lumen and  the negative potential within the cell, the observed direction is from lumen to cell, even if the cellular glucose concentration is higher than the luminal one. In other  words, downhill Na+ movement normally supports concentrative glucose transport. Concentration ratios of up to 20­fold between intracellular and extracellular  glucose have been observed in vitro under conditions of blocked efflux of cellular glucose. In some situations Na+ uptake via this route is actually more important than  glucose uptake (see Clin. Corr. 26.2). The contraluminal plasma membrane contains a member of the GLUcose Transporter (or GLUT) family, which facilitates glucose exit and entry. The intestine  contains the GLUT2 transporter, which accepts many monosaccharides, including glucose. The direction of net flux is determined by the sugar concentration gradient.  The two glucose transport systems SGLT1 and GLUT2 in the luminal and contraluminal plasma membranes, respectively, share glucose as substrate, but otherwise  differ considerably in terms of amino acid sequence, secondary protein structure, Na+ as cosubstrate, specificity for other sugars, sensitivity to inhibitors, or biological  regulation. Since both SGLT and GLUT are not inherently directional, "active" transepithelial glucose transport can be maintained under steady­state conditions only if  the Na+,K+–ATPase continues to move Na+ out of the cell. Thus the active glucose transport is indirectly dependent on a supply of ATP and an active Na+,K+– ATPase. The advantage of an electrochemical Na+ gradient serving as intermediate is that the Na+,K+–ATPase can energize the transport of many different nutrients. The only  requirement is presence of a transport system catalyzing cotransport of the nutrient with Na+. Gastric Parietal Cells Secrete HCl The parietal (oxyntic) cells of gastric glands are capable of secreting HCl into the gastric lumen. Luminal H+ concentrations of up to 0.14 M (pH 0.8) have been  observed (see Figure 26.13). As the plasma pH = 7.4, the parietal cell transports protons against a concentration gradient of 106.6. The free energy required for HCl  secretion under these conditions is minimally 9.1 kcal mol–1 of HCl (= 38 J mol–1 of HCl), as calculated from

A K+­activated ATPase (K+,H+–ATPase) is intimately involved in the mechanism of active HCl secretion. This enzyme is unique to the parietal cell and is found only  in the luminal region of the plasma membrane. It couples the hydrolysis of ATP to an electrically neutral obligatory exchange of K+ for H+, secreting H+ and taking K+  into the cell. The stoichiometry appears to be 1 mol of transported H+ and K+ for each mole of ATP.

Figure 26.19  Omeprazole, an inhibitor of K+,H+–ATPase.  This drug accumulates in an acidic  compartment (pKa  ~ 4) and is converted  to a reactive sulfenamide, which reacts  with cysteine SH groups.  From Sachs, G. The gastric  H,K­ATPase. In. L. R. Johnson (Ed.),  Physiology of the Gastrointestinal Tract.  New York: Raven Press, 1994, p. 1133.

As the K+,H+–ATPase generates a very acidic solution, protein reagents that are activated by acid can become specific inhibitors of this enzyme. Figure 26.19 shows  an example of such a reagent used to treat peptic ulcers. In the steady state, HCl can be elaborated by K+, H+–ATPase only if the luminal membrane is permeable to  K+ and Cl– and the contraluminal plasma membrane catalyzes an exchange of Cl– for HCO3– (Figure 26.20). The exchange of Cl– for HCO3– is essential to resupply the cell with  Cl– and to prevent accumulation of base within the cell. Thus, under steady­state conditions, secretion of HCl into the gastric lumen is coupled to movement of HCO3– into the plasma.

 

  

Page 1070

26.4—  Digestion and Absorption of Proteins

Figure 26.20  Model for secretion of hydrochloric acid.

Mixture of Peptidases Assures Efficient Protein Digestion The total daily protein load to be digested consists of about 70–100 g of dietary proteins and 35–200 g of endogenous proteins from digestive enzymes and sloughed­ off cells. Digestion and absorption of proteins are very efficient processes in healthy humans, since only about 1–2 g of nitrogen are lost through feces each day, which  is equivalent to 6–12 g of protein. Except for a short period after birth, oligo­ and polypeptides (proteins) are not absorbed intact in appreciable quantities by the intestine. Proteins are broken down by  hydrolases with specificity for the peptide bond, that is, by peptidases. This class of enzymes is divided into endopeptidases (proteases), which attack internal bonds  and liberate large peptide fragments, and exopeptidases, which cleave off one amino acid at a time from either the COOH (carboxypeptidases) or the NH2  terminus (aminopeptidases). Endopeptidases are important for an initial breakdown of long polypeptides into smaller products, which can then be attacked more  efficiently by exopeptidases. The final products are free amino acids and di­ and tripeptides, which are absorbed by epithelial cells (Figure 26.21). The process of protein digestion can be divided into a gastric, a pancreatic, and an intestinal phase, depending on the source of peptidases. Pepsins Catalyze Gastric Digestion of Protein Gastric juice is characterized by the presence of HCl and therefore a low pH less than 2 as well as the presence of proteases of the pepsin family. The acid serves to  kill off microorganisms and also to denature proteins. Denaturation makes proteins more susceptible to hydrolysis by proteases. Pepsins are unique in that they are  acid stable; in fact, they are active at acid but not at neutral pH. The catalytic mechanism that is effective for peptide hydrolysis at the acid pH depends on two  carboxylic groups at the active site of the enzymes. Pepsin A, the major gastric protease, prefers peptide bonds formed by the amino group of aromatic acids (Phe,  Tyr) (Table 26.6). Active pepsin is generated from the proenzyme pepsinogen by the removal of 44 amino acids from the NH2 terminus (pig enzyme). Cleavage between residues 44  and 45 of pepsinogen occurs as either an intramolecular reaction (autoactivation) below pH 5 or by active pepsin (autocatalysis). The liberated peptide from the NH2  terminus remains bound to pepsin and acts as "pepsin

Figure 26.21  Digestion and absorption of proteins.

 

  

Page 1071 TABLE 26.6 Gastric and Pancreatic Peptidases Enzyme

Proenzyme

Activator

Cleavage Point

R

CARBOXYL PROTEASES  Pepsin A

Pepsinogen A

Autoactivation, pepsin

 

Tyr, Phe, Leu

SERINE PROTEASES  Trypsin

Trypsinogen

Enteropeptidase, trypsin

 

Arg, Lys

Chymotrypsin

Chymotrypsinogen

Trypsin

 

Tyr, Trp, Phe, Met, Leu

Elastase

Proelastase

Trypsin

 

Ala, Gly, Ser

ZINC PEPTIDASES  Carboxypeptidase A

Procarboxypeptidase A

Trypsin

 

Val, Leu, Ile, Ala

Carboxypeptidase B

Procarboxypeptidase B

Trypsin

 

Arg, Lys

  inhibitor'' above pH 2. This inhibition is released either by a drop of the pH below 2 or further degradation of the peptide by pepsin. Thus, once favorable conditions  are reached, pepsinogen is converted to pepsin by autoactivation and subsequent autocatalysis at an exponential rate. The major products of pepsin action are large peptide fragments and some free amino acids. The importance of gastric protein digestion does not lie so much in its  contribution to the breakdown of ingested macromolecules, but rather in the generation of peptides and amino acids that act as stimulants for cholecystokinin release  in the duodenum. The gastric peptides therefore are instrumental in the initiation of the pancreatic phase of protein digestion. Pancreatic Zymogens Are Activated in Small Intestine Pancreatic juice is rich in proenzymes of endopeptidases and carboxypeptidases (Figure 26.22), which are activated after they reach the lumen of the small  intestine. Enteropeptidase (old name: enterokinase), a protease produced by duodenal epithelial cells, activates pancreatic trypsinogen to trypsin by scission of a  hexapeptide from the NH2 terminus. Trypsin in turn autocatalytically activates more trypsinogen to trypsin and also acts on the other proenzymes, thus liberating the  endopeptidases chymotrypsin and elastase and the carboxypeptidases A and B. Since trypsin plays a pivotal role among pancreatic enzymes in the activation  process, pancreatic juice normally contains a small­molecular­weight peptide that acts as a trypsin inhibitor and neutralizes any trypsin formed prematurely within the  pancreatic cells or pancreatic ducts. Trypsin, chymotrypsin, and elastase have different substrate specificity, as shown in Table 26.6. They are active only at neutral pH and depend on pancreatic NaHCO3 for neutralization of gastric HCl. Their mechanism of catalysis involves an essential serine residue (see p. 97) and is thus similar to serine esterases, such as acetyl  choline esterase. Reagents that interact with serine and modify it, inactivate serine esterases and peptidases. A prominent example of such a reagent is the highly toxic  diisopropylphosphofluoridate, which was developed originally for chemical warfare (neurotoxic because of inhibition of acetyl choline esterase).

 

  

Page 1072

Figure 26.22  Secretion and activation of pancreatic enzymes.  Abbreviation: CCK, cholecystokinin.  Reproduced with permission from Freeman, H. J., and  Kim, Y. S. Annu. Rev. Med. 29:102, 1978.  Copyright © 1978 by Annual Reviews, Inc.

Polypeptides generated from ingested proteins are degraded within the small intestinal lumen by carboxypeptidases A and B. The pancreatic carboxypeptidases  are Zn2+ metalloenzymes and possess a different type of catalytic mechanism than the carboxyl or serine peptidases. The combined action of pancreatic peptidases  results in the formation of free amino acids and small peptides of 2–8 residues. Peptides account for about 60% of the amino nitrogen at this point. Intestinal Peptidases Digest Small Peptides Since pancreatic juice does not contain appreciable aminopeptidase activity, final digestion of di­ and oligopeptides depends on small intestinal enzymes. The luminal  surface of epithelial cells is particularly rich in endopeptidase and aminopeptidase activity, but also contains dipeptidases (Table 26.2). The end products of the cell  surface digestion are free amino acids and di­ and tripeptides, which are absorbed via specific amino acid or peptide transport systems. Transported di­ and  tripeptides are generally hydrolyzed within the cytoplasmic compartment before they leave the cell. The cytoplasmic dipeptidases explain why practically only free  amino acids are found in the portal blood after a meal. The virtual absence of peptides had previously been taken as evidence that luminal protein digestion had to  proceed all the way to free amino acids before absorption could occur. However, it is now established that a large portion of dietary amino nitrogen is absorbed in the  form of small peptides with subsequent intracellular hydrolysis. However, di­ and tripeptides containing proline and hydroxyproline or unusual amino acids, such as b ­ alanine as carnosine (b ­alanylhistidine) or anserine (b ­alanyl 1­methylhistidine), are absorbed without intracellular hydrolysis because they are not good substrates for  the intestinal cytoplasmic dipeptidases. b ­Alanine is present in chicken meat. Free Amino Acids and Dipeptides Are Absorbed by Carrier­Mediated Transport The small intestine has a high capacity to absorb free amino acids and small peptides. Most L­amino acids can be transported across the epithelium against a  concentration gradient, although the need for concentrative transport in vivo is not obvious, since luminal concentrations are usually higher than the plasma levels of  0.1–0.2 mM. Amino acid and peptide transport in the small intestine has all the characteristics of carrier­mediated transport, such as discrimination between D­ and L­ amino acids and energy and temperature dependence. In addition, genetic defects are known to occur in humans (see Clin. Corr. 26.3).

 

  

Page 1073

CLINICAL CORRELATION 26.3  Neutral Amino Aciduria (Hartnup Disease) Transport functions, like enzymatic functions, are subject to modification by mutations. An  example of a genetic lesion in epithelial amino acid transport is Hartnup disease, named  after the family in which the disease entity resulting from the defect was first recognized.  The disease is characterized by the inability of renal and intestinal epithelial cells to absorb  neutral amino acids from the lumen. In the kidney, in which plasma amino acids reach the  lumen of the proximal tubule through the ultrafiltrate, the inability to reabsorb amino acids  manifests itself as excretion of amino acids in the urine (amino aciduria). The intestinal  defect results in malabsorption of free amino acids from the diet. Therefore the clinical  symptoms of patients with this disease are mainly those due to essential amino acid and  nicotinamide deficiencies. The pellagra­like features (see p. 1121) are explained by a  deficiency of tryptophan, which serves as precursor for nicotinamide. Investigations of  patients with Hartnup disease revealed the existence of intestinal transport systems for di­  or tripeptides, which are different from the ones for free amino acids. The genetic lesion  does not affect transport of peptides, which remains as a pathway for absorption of  protein digestion products. Silk, D. B. A. Disorders of nitrogen absorption. In: J. T. Harries (Ed.), Clinics in  Gastroenterology: Familial Inherited Abnormalities, Vol. 11: London: Saunders,  1982, pp. 47–73.   On the basis of genetics, transport experiments, and expression cloning, at least seven brush border specific transport systems for the uptake of L­amino acids or  small peptides in the luminal membrane can be distinguished: (1) for neutral amino acids with short or polar side chains (Ser, Thr, Ala); (2) for neutral amino acids with  aromatic or hydrophobic side chains (Phe, Tyr, Met, Val, Leu, Ile); (3) for imino acids (Pro, Hyp); (4) for b ­amino acids (b ­Ala, taurine); (5) for basic amino acids  and cystine (Lys, Arg, Cys­Cys); (6) for acidic amino acids (Asp, Glu); and (7) for dipeptides (Pept1) (Gly­sarcosine). The concentration mechanisms for neutral L­amino acids appear to be similar to those discussed for D­glucose (see Figure 26.17). Na+­dependent transport  systems have been identified in the luminal (brush border) membrane and Na+­independent transporters in the contraluminal plasma membrane of small intestinal  epithelial cells. Similarly, as for active glucose transport, the energy for concentrative amino acid transport is derived directly from the electrochemical Na+ gradient and  only indirectly from ATP. Amino acids are not chemically modified during membrane transport, although they may be metabolized within the cytoplasmic compartment.  The brush border transport for the other amino acids is energized in more complicated ways. For example, the acidic amino acid transporter mediates cotransport  of the amino acid with 2 Na+ ions and counter transport with 1 K+ ion. Neutral dipeptides are cotransported across the brush border membrane with a proton and thus are energized through the proton electrochemical gradient across this  membrane. However, because of the Na+/H+ exchange, both gradients tend to be similar and interdependent. The dipeptide transporter also accepts b ­lactam  antibiotics (aminopenicillins) and is important for absorption of orally administered antibiotics of this class. Fetus and Neonate Can Absorb Intact Proteins The fetal and neonatal small intestines can absorb intact proteins. The uptake occurs by endocytosis, that is, the internalization of small vesicles of plasma membrane,  which contain ingested macromolecules. The process is also termed pinocytosis because of the small size of vesicles. The small intestinal pinocytosis of protein is  thought to be important for the transfer of maternal antibodies (g­globulins) to the offspring, particularly in rodents. The pinocytotic uptake of proteins is not important  for nutrition, and its magnitude usually declines after birth. Persistence of low levels of this process beyond the neonatal period may, however, be responsible for  absorption of sufficient quantities of macro­molecules to induce antibody formation. 26.5—  Digestion and Absorption of Carbohydrates Di­ and Polysaccharides Require Hydrolysis Dietary carbohydrates provide a major portion of the daily caloric requirement. They consist of mono­, di­, and polysaccharides (Table 26.7). Monosaccharides  need not be hydrolyzed for absorption. Disaccharides require the small intestinal surface enzymes for hydrolysis into monosaccharides, while polysaccharides depend  on pancreatic amylase for degradation (Figure 26.23). Starch, a major nutrient, is a plant polysaccharide with a molecular mass of more than 100 kDa. It consists of a mixture of linear chains of glucose molecules linked by  a ­1,4­glucosidic bonds (amylose) and of branched chains with branch points made up by a ­1,6 linkages (amylopectin). The ratio of 1,4­ to 1,6­glucosidic bonds is  about 20 : 1. Glycogen is an animal polysaccharide similar in structure to amylopectin. The two compounds differ in terms of the number of branch points, which  occur more frequently in glycogen.

 

  

Page 1074 TABLE 26.7 Dietary Carbohydrates Carbohydrate

Typical Source

 

Structure

Amylopectin

Potatoes, rice, corn, bread

a­Glc(1   4)nGlc with a­Glc(1   6)  branches

 

Amylose

Potatoes, rice, corn, bread

a­Glc(1   4)nGlc

 

Sucrose

Table sugar, desserts

a­Glc(1   2)b­Fru

Trehalose

Young mushrooms

a­Glc(1   1)a­Glc

Lactose

Milk, milk products

b­Gal(1   4)Glc

Fructose

Fruit, honey

Fru

Glucose

Fruit, honey, grape

Glc

Raffinose

Leguminous seeds

a­Gal(1   6)a­Glc (1   2)b­Fru

 

 

 

 

 

 

  Hydrated starch and glycogen are attacked by the endosaccharidase a ­amylase present in saliva and pancreatic juice (Figure 26.24). Hydration of the  polysaccharides occurs during heating and is essential for efficient digestion. Amylase is specific for internal a ­1,4­glucosidic bonds; a ­1,6 bonds are not attacked,  nor are a ­1,4 bonds of glucose units that serve as branch points. The pancreatic isoenzyme is secreted in large excess relative to starch intake and

 

  

Page 1075

Figure 26.23  Digestion and absorption of carbohydrates.

CLINICAL CORRELATION 26.4  Disaccharidase Deficiency Intestinal disaccharidase deficiencies are encountered relatively frequently in humans.  Deficiency can be present in one enzyme or several enzymes for a variety of reasons  (genetic defect, physiological decline with age, or the result of "injuries" to the mucosa).  Of the disaccharidases, lactase is the most common enzyme with an absolute or relative  deficiency, which is experienced as milk intolerance. The consequences of an inability to  hydrolyze lactose in the upper small intestine are inability to absorb lactose and bacterial  fermentation of ingested lactose in the lower small intestine. Bacterial fermentation results  in the production of gas (distension of gut and flatulence) and osmotically active solutes  that draw water into the intestinal lumen (diarrhea). The lactose in yogurt has already been  partially hydrolyzed during the fermentation process of making yogurt. Thus individuals  with lactase deficiency can often tolerate yogurt better than unfermented dairy products.  The enzyme lactase is commercially available to pretreat milk so that the lactose is  hydrolyzed. Buller, H. A., and Grant, R. G. Lactose intolerance. Annu. Rev. Med. 41:141, 1990.   is more important than the salivary enzyme from a digestive point of view. The products of the digestion by a ­amylase are mainly the disaccharide maltose, the  trisaccharide maltotriose, and so­called a ­limit dextrins containing on average eight glucose units with one or more a ­1,6­glucosidic bonds. Final hydrolysis of di­ and oligosaccharides to monosaccharides is carried out by surface enzymes of the small intestinal epithelial cells (Table 26.8). Most of the  surface oligosaccharidases are exoenzymes that cleave off one monosaccharide at a time from the nonreducing end. The capacity of the a ­glucosidases is normally  much greater than that needed for completion of the digestion of starch. Similarly, there is usually excess capacity for sucrose (table sugar) hydrolysis relative to dietary  intake. In contrast, b ­galactosidase (lactase) can be rate­limiting in humans for hydrolysis and utilization of lactose, the major milk carbohydrate (see Clin. Corr.  26.4). Di­, oligo­, and polysaccharides that are not hydrolyzed by a ­amylase and/ or intestinal surface enzymes cannot be absorbed; therefore they reach the lower tract of  the intestine, which from the lower ileum on contains bacteria. Bacteria can utilize many of the remaining carbohydrates because they possess many more types of  saccharidases than humans. Monosaccharides that are released as a result of bacterial enzymes are predominantly metabolized anaerobically by the bacteria  themselves, resulting in degradation products such as short­chain fatty acids, lactate, hydrogen gas (H2), methane (CH4), and carbon

Figure 26.24  Digestion of amylopectin by salivary and pancreatic a­amylase.

 

  

Page 1076 TABLE 26.8 Saccharidases of the Surface Membrane of the Small Intestine Enzyme

Specificity

Natural Substrate

Product

exo­1,4­a­Glucosidase  (glucoamylase)

a­(1   4)Glucose Amylose

Glucose

Oligo­1,6­glucosidase  (isomaltase)

a­(1   6)Glucose Isomaltose, a­ dextrin

Glucose

a­Glucosidase (maltase)

a­(1   4)Glucose Maltose,  maltotriose

Glucose

Sucrose­a­Glucosidase  (sucrase)

a­Glucose

a,a­Trehalase

a­(1   1)Glucose Trehalose

b­Glucosidase

b ­Glucose

Glucosylceramide

Glucose,  ceramide

b­Galactosidase (lactase)

b­Galactose

Lactose

Glucose,  galactose

Sucrose

Glucose, fructose Glucose

  dioxide (CO2). These compounds can cause fluid secretion, increased intestinal motility, and cramps, either because of increased intraluminal osmotic pressure, and  distension of the gut, or a direct irritant effect of the bacterial degradation products on the intestinal mucosa. The well­known problem of flatulence after ingestion of leguminous seeds (beans, peas, and soya) is caused by oligosaccharides, which cannot be hydrolyzed by  human intestinal enzymes. The leguminous seeds contain modified sucrose to which one or more galactose moieties are linked. The glycosidic bonds of galactose are in  the a  configuration, which can only be split by bacterial enzymes. The simplest sugar of this family is raffinose (see Table 26.7). Trehalose, a disaccharide that occurs in young mushrooms, requires a special disaccharidase, trehalase. Monosaccharides Are Absorbed by Carrier­Mediated Transport The major monosaccharides that result from digestion of di­ and polysaccharide are D­glucose, D­galactose, and D­fructose. Absorption of these and other minor  monosaccharides are carrier­mediated processes that exhibit such features as substrate specificity, stereospecificity, saturation kinetics, and inhibition by specific  inhibitors. At least two types of monosaccharide transporters catalyze monosaccharide uptake from the lumen into the cell: (1) a Na+­monosaccharide cotransporter, existing  probably as a tetramer of 75­kDa peptides, has high specificity for D­glucose and D­galactose and catalyzes "active" sugar absorption (SGLT); and (2) a Na+­ independent, facilitated­diffusion type of monosaccharide transport system with specificity for D­fructose (GLUT5). In addition, a Na+­independent  monosaccharide transporter (GLUT2), consisting of 57­kDa peptide(s), which accepts all three monosaccharides, is present in the contraluminal plasma membrane.  GLUT2 is also located in the liver and kidney, and other members of the GLUT family of glucose transporters are found in all cells. All GLUT transporters mediate  uncoupled D­glucose flux down its concentration gradient. GLUT2 of gut, liver, and kidney moves D­glucose out of the cell into the blood under physiological  conditions, while in other tissues GLUT1 (in erythrocytes and brain) or the insulin­sensitive GLUT4 (in fat and muscle tissue) are mainly involved in D­glucose uptake.  Properties of intestinal SGLT1 and of GLUT2 are compared in Table 26.9, and their role in transepithelial glucose absorption is illustrated in Figure 26.18.

 

  

Page 1077 TABLE 26.9 Characteristics of Glucose Transport Systems in the Plasma Membranes of  Enterocytes Characteristic

Luminal

Contraluminal

Designation

SGLT1

GLUT2

Subunit molecular weight  (kDa)

75

57

Effect of Na+

Cotransport with Na+

None

Good substrates

D­Glc, D­Gal, a­methyl­D­Glc D­Glc, D­Gal, D­Man, 2­deoxy­D­ Glc

Inhibitor

Phlorizin (Figure 26.25)

Cytochalasin B (Figure 26.26)

  26.6—  Digestion and Absorption of Lipids Lipid Digestion Requires Overcoming the Limited Water Solubility of Lipids An adult man ingests about 60–150 g of lipid per day. Triacylglycerols constitute more than 90% of the dietary fat. The rest is made up of phospholipids,  cholesterol, cholesterol esters, and free fatty acids. In addition, 1–2 g of cholesterol and 7–22 g of phosphatidylcholine (lecithin) are secreted into the small intestine  lumen as constituents of bile.

Figure 26.25  Phlorizin (phloretin­2 ­glucoside).

Lipids are defined by their good solubility in organic solvents and their sparing or lack of solubility in aqueous solutions. The poor water solubility presents problems for  digestion because the substrates are not easily accessible to the digestive enzymes in the aqueous phase. In addition, even if ingested lipids are hydrolyzed into simple  constituents, the products tend to aggregate to larger complexes that make poor contact with the cell surface and therefore are not easily absorbed. These problems  are overcome by (1) increases in the interfacial area between the aqueous and lipid phase and (2) "solubilization" of lipids with detergents. Thus changes in the  physical state of lipids are intimately connected to chemical changes during digestion and absorption.

Figure 26.26  Cytochalasin B.

At least five different phases can be distinguished (Figure 26.27): (1) hydrolysis of triacylglycerols to free fatty acids and monoacylglycerols; (2) solubilization by  detergents (bile acids) and transport from the intestinal lumen toward the cell surface; (3) uptake of free fatty acids and monoacylglycerols into the cell and resynthesis  to triacylglycerols; (4) packaging of newly synthesized triacylglycerols into special lipid­rich globules, called chylomicrons, and (5) exocytosis of chylomicrons from  cells and release into lymph.

Figure 26.27  Digestion and absorption of lipids.

 

  

Page 1078

Figure 26.28  Changes in physical state during triacylglycerol digestion.  Abbreviations: TG, triacylglycerol; DG, diacylglycerol; MG, monoacylglycerol; FA, fatty acid.

Lipids Are Digested by Gastric and Pancreatic Lipases Digestion of lipids is initiated in the stomach by an acid­stable lipase, most of which is thought to originate from glands at the back of the tongue. However, the rate of  hydrolysis is slow because the ingested triacylglycerols form a separate lipid phase with a limited water–lipid interface. The lipase adsorbs to that interface and converts  triacylglycerols into fatty acids and diacylglycerols (Figure 26.28). The importance of the initial hydrolysis is that some of the water­immiscible triacylglycerols are  converted to products that possess both polar and nonpolar groups. Such surfactive products spontaneously adsorb to water–lipid interfaces and confer a hydrophilic  surface to lipid droplets thereby providing a stable interface with the aqueous environment. At constant volume of the lipid phase, any increase in interfacial area  produces dispersion of the lipid phase into smaller droplets (emulsification) and provides more sites for adsorption of more lipase molecules. The major enzyme for triacylglycerol hydrolysis is the pancreatic lipase (Figure 26.29). This enzyme is specific for esters in the a ­position of glycerol and prefers  long­chain fatty acids with more than ten carbon atoms. Hydrolysis by the pancreatic enzyme also occurs at the water–lipid interface of emulsion droplets. The  products are free fatty acids and b ­monoacylglycerols. The purified form of the enzyme is strongly inhibited by the bile acids that normally are present in the small  intestine during lipid digestion. The problem of inhibition is overcome by colipase, a small protein (12 kDa) that binds to both the water–lipid interface and to lipase,  thereby anchoring and activating the enzyme. It is secreted by the pancreas as procolipase and depends on tryptic removal of a NH2­terminal decapeptide for full  activity.

Figure 26.29  Mechanism of action of lipase.

Pancreatic juice also contains another less specific lipid esterase, which acts on cholesterol esters, monoglycerides, or other lipid esters, such as esters of vitamin A  with carboxylic acids. In contrast to triacylglycerol lipase, this lipid esterase requires bile acids for activity. Phospholipids are hydrolyzed by specific phospholipases. Pancreatic secretions are especially rich in the proenzyme for phospholipase A2 (Figure 26.30). As other  pancreatic proenzymes, this one is also activated by trypsin. Phospholipase A2 requires bile acids for activity. Bile Acid Micelles Solubilize Lipids during Digestion Bile acids are biological detergents that are synthesized by the liver and secreted as conjugates of glycine or taurine with the bile into the duodenum. At physiological  pH values, they are present as anions, which have detergent

 

  

Page 1079

Figure 26.30  Mechanism of action of phospholipase A2.

properties. Therefore the terms bile acids and bile salts are often used interchangeably (Figure 26.31). Bile acids at pH values above the pK (Table 26.10)  reversibly form aggregates at concentrations above 2–5 mM. These aggregates are called micelles, and the minimal concentration necessary for micelle formation is  the critical micellar concentration (Figure 26.32). The bile acids in micelles are in equilibrium with those free in solution. Thus micelles, in contrast to emulsified  lipids, are equilibrium structures with well­defined sizes that are much smaller than emulsion droplets. Micelle sizes typically range between 40 and 600  m depending  on bile acid concentration and the ratio of bile acids to lipids. The arrangements of bile acids in micelles is such that the hydrophobic portions are removed from contact with water, while hydrophilic groups remain exposed to the  water. The hydrophobic region of bile acids is formed by one surface of the fused ring system, while the carboxylate or sulfonate ion and the hydroxyl groups on the  other side of the ring system are hydrophilic. Since the major driving forces for micelle formation are the removal of apolar, hydrophobic groups from and the  interaction of polar groups with water molecules, the distribution of polar and apolar regions places some constraints on the stereochemical arrangements of bile acid  molecules within a micelle. Four bile acid molecules are sufficient to form a very simple micelle as shown in Figure 26.33. Bile salt micelles can solubilize other lipids,  such as phospholipids and fatty acids. These mixed micelles have disk­like shapes, whereby the phospholipids and fatty acids form a bilayer and the bile acids  occupy the edge positions, rendering the edge of the disk hydrophilic (Figure 26.34). Within the mixed phospholipid–bile acid micelles, other water­insoluble lipids,  such as cholesterol, can be accommodated and thereby "solubilized" (for potential problems see Clin. Corr. 26.5).

Figure 26.31  Cholic acid, a bile acid.

Figure 26.32  Solubility properties of bile acids in aqueous  solutions. Abbreviation: CMC, critical  micellar concentration.

Figure 26.33  Diagrammatic representation of a Na+  cholate micelle. Adapted from Small, D. M.  Biochim. Biophys. Acta 176: 178, 1969.

 

  

Page 1080

Figure 26.34  Proposed structure of the intestinal mixed micelle.  The bilayer disk has a band of bile salt at its periphery and  other, more hydrophobic components (fatty acids, monoacylglycerol,  phospholipids, and cholesterol) protected within its interior.  Redrawn based on figure from Carey, M. C. In: A. M.  Arias, H. Popper, D. Schachter, et al. (Eds.), The Liver: Biology and  Pathology, New York: Raven Press, 1982.

 

  

Page 1081

CLINICAL CORRELATION 26.5  Cholesterol Stones Liver secretes phospholipids and cholesterol together with bile acids into the bile.  Because of the limited solubility of cholesterol, its secretion in bile can result in cholesterol  stone formation in the gallbladder. Stone formation is a relatively frequent complication;  up to 20% of North Americans will develop stones during their lifetime. Cholesterol is practically insoluble in aqueous solutions. However, it can be incorporated  into mixed phospholipid–bile acid micelles up to a mole ratio of 1:1 for  cholesterol/phospholipids and thereby ''solubilized" (see accompanying figure). The liver  can produce supersaturated bile with a higher ratio than 1:1 of cholesterol/phospholipid.  This excess cholesterol has a tendency to come out of solution and to crystallize. Such  bile with excess cholesterol is considered lithogenic, that is, stoneforming. Crystal  formation usually occurs in the gallbladder, rather than the hepatic bile ducts, because  contact times between bile and any crystallization nuclei are greater in the gallbladder. In  addition, the gallbladder concentrates bile by absorption of electrolytes and water. The  bile salts chenodeoxycholate and ursodeoxycholate are now available for oral use to  dissolve gallstones. Ingestion of these bile salts reduces cholesterol excretion into the bile  and allows cholesterol in stones to be solubilized. The tendency to secrete bile supersaturated with respect to cholesterol is inherited and  found more frequently in females than in males, often associated with obesity.  Supersaturation also appears to be a function of the size and nature of the bile acid pool  as well as the secretion rate. Schoenfield, L. J., and Lachin, J. M. Chenodiol (chenodeoxycholic acid) for dissolution  of gallstones: The National Cooperative Gallstone Study. A controlled trial of safety and  efficacy. Ann. Intern. Med. 95:257, 1981; and Carey, M. C., and Small, D. M. The  physical chemistry of cholesterol solubility in bile. J. Clin. Invest. 61:998, 1978.

Diagram of the physical states of mixtures of 90% water and 10% lipid. The 10% lipid is  made up of bile acids, lecithin, and cholesterol, and the triangle represents all possible  ratios of the three lipid constituents. Each point within the triangle corresponds to a  particular composition of the three components, which can be read off the graph as  indicated; each point on one of the sides corresponds to a particular composition of just  two components. The left triangle contains the composition of gallbladder bile samples from  patients without stones (red  ). Lithogenic bile  has a composition that falls outside the "one liquid" area in the lower left corner.  Redrawn from Hofmann, A. F., and Small, D. M. Annu. Rev. Med. 18:362, 1967.  Copyright © 1967 by Annual Reviews, Inc.

  During triacylglycerol digestion, free fatty acids and monoacylglycerols are released at the surface of fat emulsion droplets. In contrast to triacylglycerols, which are water­ insoluble, free fatty acids and monoacylglycerols are slightly water­soluble, and molecules at the surface equilibrate with those in solution. The latter in turn become  incorporated into bile acid micelles. Thus the products of triacylglycerol hydrolysis are continuously transferred from emulsion droplets to the micelles (see Figure 26.27). Micelles provide the major vehicle for moving lipids from the intestinal lumen to the cell surface where absorption occurs. Because the fluid layer next to the cell surface is  poorly mixed, the major transport mechanism for solute

 

  

Page 1082

CLINICAL CORRELATION 26.6  A­b ­Lipoproteinemia A­ b ­lipoproteinemia is an autosomal recessive disorder characterized by the absence of  all lipoproteins containing apo­ b ­lipoprotein, that is, chylomicrons, very low density  lipoproteins (VLDLs), and low density lipoproteins (LDLs). Serum cholesterol is  extremely low. This defect is associated with severe malabsorption of triacylglycerol and  lipid­soluble vitamins (especially tocopherol and vitamin E) and accumulation of apo B in  enterocytes and hepatocytes. The defect does not appear to involve the gene for apo B,  but rather one of several proteins involved in processing of apo B in liver and intestinal  mucosa, or in assembly and secretion of triacylglycerol­rich lipoproteins, that is,  chylomicrons and VLDLs from these tissues, respectively. Kane, J. P. Apolipoprotein B: structural and metabolic heterogeneity. Annu. Rev.  Physiol. 45:673, 1983; and Kane, J. P., and Havel, R. J. In: C. R. Scriver, A. L.  Beaudet, W. S. Sly, and D. Valle (Eds.), The Metabolic and Molecular Bases of  Inherited Disease, Vol. 1, 7th ed. New York: McGraw­Hill, 1995, p. 1853.   flux across this "unstirred" fluid layer is diffusion down the concentration gradient. With this type of transport mechanism, the delivery rate of nutrients at the cell surface  is proportional to their concentration difference between luminal bulk phase and cell surface. Obviously, the unstirred fluid layer presents problems for sparingly soluble  or insoluble nutrients, in that reasonable delivery rates cannot be achieved. Bile acid micelles overcome this problem for lipids by increasing their effective concentration  in the unstirred layer. The increase in transport rate is nearly proportional to the increase in effective concentration and can be 1000­fold over that of individually  solubilized fatty acids, in accordance with the different solubility of fatty acids as micelles or as individual molecules. This relationship between flux and effective  concentration holds because the diffusion constant, another parameter that determines the flux, is only slightly smaller for the mixed micelles as compared to lipid  molecules free in solution. Thus efficient lipid absorption depends on the presence of sufficient bile acids to "solubilize" the ingested and hydrolyzed lipids in micelles. In  the absence of bile acids, the absorption of triacylglycerols does not completely stop, although the efficiency is drastically reduced. The residual absorption depends on  the slight water solubility of the free fatty acids and monoacylglycerols. Unabsorbed lipids reach the lower intestine where a small part can be metabolized by bacteria.  The bulk of unabsorbed lipids, however, is excreted with the stool (this is called steatorrhea). Micelles also transport cholesterol and the lipid­soluble vitamins A, D, E, and K through the unstirred fluid layers. Bile acid secretion is absolutely essential for their  absorption. Most Absorbed Lipids Are Incorporated into Chylomicrons Uptake of lipids by the epithelial cells occurs by diffusion through the plasma membrane. Absorption is virtually complete for fatty acids and monoacylglycerols, which  are slightly water­soluble. It is less efficient for water­insoluble lipids. For example, only 30–40% of the dietary cholesterol is absorbed. Within the intestinal cells, the fate of absorbed fatty acids depends on chain length. Fatty acids of medium chain length (6–10 carbon atoms) pass through the cell  into the portal blood without modification. Long­chain fatty acids (>12 carbon atoms) become bound to a cytosolic, specifically intestinal fatty acid­binding protein  (I­FABP) and are transported to the endoplasmic reticulum, where they are resynthesized into triacylglycerols. Glycerol for this process is derived from the absorbed  2­monoacylglycerols and, to a minor degree, from glucose. The resynthesized triacylglycerols form lipid globules to which surface­active phospholipids and special  proteins, termed apolipoproteins, adsorb. The lipid globules migrate within membrane­bounded vesicles through the Golgi to the basolateral plasma membrane. They  are finally released into the intercellular space by fusion of the vesicles with the basolateral plasma membrane. Because the lipid globules can be several micrometers in  diameter and because they leave the intestine via lymph vessels, they are called chylomicrons (chyle = milky lymph that is present in the intestinal lymph vessels,  lacteals, and the thoracic duct after a lipid meal; chyle is derived from the Greek chylos, which means juice). The intestinal apolipoproteins are distinctly different from  those of the liver and are designated A­1 and B. Apolipoprotein B is essential for chylomicron release from enterocytes (see Clin. Corr. 26.6). While dietary medium­chain fatty acids reach the liver directly with the portal blood, the long­chain fatty acids bypass the liver by being released in the form of  chylomicrons into the lymphatics. The intestinal lymph vessels drain into the large body veins via the thoracic duct. Blood from the large veins first reaches the lungs and  then the capillaries of the peripheral tissues, including adipose tissue and muscle, before it comes into contact with the liver. Fat and

 

  

Page 1083

muscle cells in particular take up large amounts of dietary lipids for storage or metabolism. The bypass of the liver may have evolved to protect this organ from a lipid  overload after a meal. The differential handling of medium­ and long­chain fatty acids by intestinal cells can be specifically exploited to provide the liver with high­caloric nutrients in the form  of fatty acids. Short­ and medium­chain fatty acids are not very palatable; however, triacylglycerols synthesized from these fatty acids are quite palatable and can be  used as part of the diet. 26.7—  Bile Acid Metabolism All bile acids are synthesized within the liver from cholesterol but can be modified by bacterial enzymes in the intestinal lumen. Primary bile acids synthesized by the  liver are cholic and chenodeoxycholic (or chenic) acid. The secondary bile acids are derived from the primary bile acids by bacterial dehydroxylation in position 7  of the ring structure, resulting in deoxycholate and lithocholate, respectively (Figure 26.35). Primary and secondary bile acids are reabsorbed by the intestine into the portal blood, taken up by the liver, and then resecreted into bile. Within the liver, primary as  well as secondary bile acids are linked to either glycine or

Figure 26.35  Bile acid metabolism in the rat.  Green and black arrows indicate reactions catalyzed by liver enzymes; red arrows  indicate those of bacterial enzymes within the intestinal lumen. (NH—), glycine  or taurine conjugate of the bile acids.

 

  

Page 1084

taurine via an isopeptide bond. These derivatives are called glyco­ and tauro­conjugates, respectively, and constitute the forms that are secreted into bile. With the  conjugation, the carboxyl group of the unconjugated acid is replaced by an even more polar group. The pK values of the carboxyl group of glycine and of the sulfonyl  group of taurine are lower than that of unconjugated bile acids, so that conjugated bile acids remain ionized over a wider pH range (see Table 26.10). The conjugation  is partially reversed within the intestinal lumen by hydrolysis of the isopeptide bond. The total amount of conjugated and unconjugated bile acids secreted per day by the liver is 16–70 g for an adult. As the total body pool is only 3–4 g, bile acids have  to recirculate 5–14 times each day between the intestinal lumen and the liver. Reabsorption of bile acids is important to conserve the pool. Most of the uptake is  probably by passive diffusion along the entire small intestine. In addition, the lower ileum contains a specialized Na+­bile acid cotransport system for concentrative  reuptake. Thus during a meal, bile acids from the gallbladder and liver are released into the lumen of the upper small intestine, pass with the chyme down the small  intestinal lumen, are reabsorbed by the epithelium of the lower small intestine into the portal blood, and are then extracted from the portal blood by the liver  parenchymal cells. The process of secretion and reuptake is referred to as the enterohepatic circulation (Figure 26.36). Reabsorption of bile acids by the intestine is  quite efficient as only about 0.5 g of bile acids escapes reuptake each day and is secreted with the feces. Serum levels of bile acids normally vary with the rate of  reabsorption and therefore are highest during a meal. Cholate, deoxycholate, chenodeoxycholate, and their conjugates continuously participate in the enterohepatic circulation. In contrast, most of the lithocholic acid that  is produced by bacterial enzymes is sulfated during the next passage through the liver. The sulfate ester of lithocholic acid is not a substrate for the bile acid transport  system in the ileum and therefore is excreted in the feces.

Figure 26.36  Enterohepatic circulation of bile acids.  Redrawn from Clark, M. L., and Harries, J. T. In: I. McColl and G. E. Sladen (Eds.),  Intestinal Absorption in Man. New York: Academic Press, 1975, p. 195.

 

  

Page 1085

Bibliography Cristofaro, E., Mottu, F., and Wuhrmann, J. J. Involvement of the raffinose family of oligosaccharides in flatulence. In: H. L. Sipple and K. W. McNutt (Eds.), Sugars  in Nutrition. New York: Academic Press, 1974, p. 314. Field, M., and Semrad, C. E. Toxigenic diarrheas, congenital diarrheas, and cystic fibrosis. Annu. Rev. Physiol. 55:631–655, 1993. Hediger, M. A., and Rhoads, D. B. Molecular physiology of sodium­glucose cotransporters. Physiol. Rev. 74:993–1026, 1994. Johnson, L. R. (ed.­in­chief). Physiology of the Gastrointestinal Tract, Vols. 1 and 2, 2nd and 3rd eds. New York: Raven Press, 1987, 1994. Mathews, D. M. Protein Absorption. New York: Wiley­Liss, 1991. Pandol, S., and Raybauld, H. E. Integrated response to a meal. The Undergraduate Teaching Project in Gastroenterology and Liver Disease. Unit #29. American  Gastroenterological Assoc. Timonium, MD: Milner­Fenwick, Inc., 1995. Porter, R., and Collins, G. M. Brush Border Membranes. Volume 95, Ciba Foundation Symposium. London: Pitman, 1983. Schultz, S. G. (section ed.). Handbook of Physiology. Section 6: The Gastrointestinal System. Vol. IV. Intestinal Absorption and Secretion (M. Field and R. A.  Frizzell, Eds.). Bethesda, MD: American Physiological Society, 1991 Sleisenger, M. H. Malabsorption and nutritional support. Clin. Gastroenterol. 12:323, 1983. Thomson, A. B. R., Schoeller, C., Keelan, M., Smith, L., and Clandinin, M. T. Lipid absorption: passing through the unstirred layers near the brush border  membrane, and beyond. Can. J. Physiol. Pharmacol. 71:531–555, 1993. Questions J. Baggott and C. N. Angstadt Refer to the following for Questions 1–5: A. liver B. pancreas C. spleen D. stomach E. none of the above 1. Has no role in digestion. 2. Synthesizes an essential emulsifier of lipids. 3. Participates in a nonessential manner in protein digestion. 4. Transports HCO3– from the cytoplasm across the contraluminal plasma membrane. 5. Site of chymotrypsinogen synthesis. 6. Active forms of enzymes that digest food may normally be found in all of the following EXCEPT: A. in soluble form in the lumen of the stomach. B. in the saliva. C. attached to the luminal surface of the plasma membrane of intestinal epithelial cells. D. dissolved in the cytoplasm of intestinal epithelial cells. E. in zymogen granules of pancreatic exocrine cells. 7. Histamine is a physiologically important secretagogue of: A. amylase by the salivary glands. B. HCl by the stomach. C. gastrin by the stomach. D. hydrolytic enzymes by the pancreas. E. NaHCO3 by the pancreas. 8. The contraluminal membranes of small intestinal epithelial cells contain: A. aminopeptidases. B. Na+,K+–ATPase. C. disaccharidases. D. GLUT5. E. Na+–monosaccharide transport (SGLT1). 9. Oral administration of large amounts of tyrosine could be expected to interfere with the intestinal absorption of: A. leucine. B. lysine. C. glycine. D. aspartate. E. none of the above. 10. Which of the following has two carboxyl groups essential for peptidase activity? A. carboxypeptidase B. chymotrypsin C. elastase D. pepsin E. trypsin 11. Starch digestion is more efficient after heating the starch with water because heating: A. hydrates the starch granules, making them more susceptible to pancreatic amylase. B. converts a ­1,4 links to b ­1,4 links, which are more susceptible to attack by mammalian amylases. C. partly hydrolyzes a ­1,6 links. D. converts the linear amylose to branched amylopectin, which resembles glycogen. E. inactivates amylase inhibitors, which are common in the tissues of starchy plants. 12. In the cytoplasm of intestinal cells: A. all di­ and tripeptides are hydrolyzed. B. aminopeptidases are especially active. C. during the neonatal period ingested proteins may be found. D. most disaccharides are hydrolyzed. E. raffinose and related sugars are degraded to yield hydrogen, methane, and carbon dioxide. 13. In the digestion and absorption of triacylglycerols: A. a pancreatic lipase initiates the process. B. an important colipase is activated by tryptic hydrolysis. C. hydrolysis occurs in the interior of the lipid droplets. D. most of the triacylglycerol hydrolysis is carried out by a lipase of gastric origin. E. efficiency is greatly increased if bile acids are absent.

 

  

Page 1086

14. Micelles: A. are the same as emulsion droplets. B. form from bile acids at all bile acid concentrations. C. although they are formed during lipid digestion, do not significantly enhance utilization of dietary lipid. D. always consist of only a single lipid species. E. are essential for the absorption of vitamins A and K. 15. In the metabolism of bile acids: A. the liver synthesizes the primary bile acids, cholic, and deoxycholic acids. B. secondary bile acids are produced by conjugation of primary bile acids to glycine or taurine. C. physiologically active bile acids are formed from primary bile acids by intestinal bacteria. D. daily bile acid secretion by the liver is approximately equal to daily bile acid synthesis. E. conjugation reduces the polarity of bile acids, enhancing their ability to interact with lipids. Answers 1. C The spleen has no role in the digestion of food, though it does participate in other degradation processes. 2. A Bile acids are synthesized in the liver and are stored in the gallbladder (p. 1057). 3. D Loss of the stomach function can be compensated for by the intestinal processes (p. 1057). 4. D This occurs in the parietal (oxyntic) cells during HCl secretion (p. 1069). 5. B 6. E Zymogen granules contain inactive proenzymes or zymogens, which are not activated until after release from the cell (p. 1060). 7. B Stimulation of H2 receptors of the stomach causes HCl secretion (p. 1061). 8. B. Only the contraluminal surface contains the Na+,K+–ATPase. All other activities are associated with the luminal surface (Table 26.5, p. 1064). 9. A Tyrosine shares a transport system with Val, Leu, Met, Phe, and Ile (p. 1073). 10. D The carboxylic acid groups are involved in the mechanism that depends on an acid pH (p. 1070). 11. A a ­Amylase attacks hydrated starch more readily than unhydrated; heating hydrates the starch granules (p. 1074). 12. C They are taken up by pinocytosis (p. 1073). 13. B This colipase is required to overcome bile acid inhibition of pancreatic lipase, the major enzyme of lipid digestion. The colipase is secreted by the pancreas as a  procolipase and must be activated by tryptic cleavage (p. 1078). A: Lipid digestion is initiated in the stomach by acid­stable lipase (p. 1078). 14. E The lipid­soluble vitamins must be dissolved in mixed micelles as a prerequisite for absorption (p. 1082). A: Micelles are of molecular dimensions and are highly  ordered structures; emulsion droplets are much larger and are random (p. 1078, Figure 26.28; p. 1080, Figure 26.34). B: Micelle formation occurs only above the  critical micellar concentration (CMC); below that concentration the components are in simple solution (p. 1079, Figure 26.32). C: See item 13. D: Micelles may  consist of only one component, or they may be mixed (p. 1079.) 15. C The primary bile acids (cholic and chenodeoxycholic acids) are synthesized in the liver. In the intestine they may be dehydroxylated by bacteria to form the  secondary bile acids—deoxycholate and lithocholate. Only a small fraction of the bile acid escapes reuptake; this must be replaced by synthesis. Both are reabsorbed  and recirculated (enterohepatic circulation). Both are conjugated to glycine or taurine, increasing their polarity (p. 1078).

 

  

Page 1087

Chapter 27—  Principles of Nutrition I: Macronutrients Stephen G. Chaney

27.1 Overview

1088

27.2 Energy Metabolism

1088

Energy Content of Food Is Measured in Kilocalories

1088

Energy Expenditure Is Influenced by Four Factors

1088

27.3 Protein Metabolism Dietary Protein Serves Many Roles Including Energy Production

1089

Nitrogen Balance Relates Intake of Nitrogen to Its Excretion

1090

Essential Amino Acids Must Be Present in the Diet

1090

Protein Sparing Is Related to Dietary Content of Carbohydrate and Fat

1091

Normal Adult Protein Requirements Depend on Diet

1091

Protein Requirement Increases during Growth and Recovery from  Illness

1092

27.4 Protein­Energy Malnutrition

1093

27.5 Excess Protein­Energy Intake

1094

Obesity Has Dietary and Genetic Components

1094

Metabolic Consequences of Obesity Have Significant Health Implications

1094

27.6 Carbohydrates

1095

27.7 Fats

1097

27.8 Fiber

1097

27.9 Composition of Macronutrients in the Diet

1098

Composition of the Diet Affects Serum Cholesterol

1098

Effects of Refined Carbohydrate in the Diet Are Not Straightforward

1100

Mixed Vegetable and Animal Proteins Meet Nutritional Protein  Requirements

1101

An Increase in Fiber from Varied Sources Is Desirable

1101

Current Recommendations Are for a "Prudent Diet"

1101

Bibliography

1103

Questions and Answers

1104

Clinical Correlations

   

  

1089

 

27.1 Vegetarian Diets and Protein–Energy Requirements

1091

27.2 Low­Protein Diets and Renal Disease

1092

27.3 Providing Adequate Protein and Calories for the Hospitalized  Patient

1093

27.4 Carbohydrate Loading and Athletic Endurance

1096

27.5 High­Carbohydrate Versus High­Fat Diets for Diabetics

1096

27.6 Polyunsaturated Fatty Acids and Risk Factors for Heart Disease

1099

27.7 Metabolic Adaptation: The Relationship between Carbohydrate  Intake and Serum Triacylglycerols

1100

Page 1088

27.1—  Overview Nutrition is best defined as the utilization of foods by living organisms. Since the process of food utilization is biochemical, the major thrust of the next two chapters is a  discussion of basic nutritional concepts in biochemical terms. Simply understanding basic nutritional concepts is no longer sufficient. Nutrition attracts more than its  share of controversy in our society, and a thorough understanding of nutrition almost demands an understanding of the issues behind these controversies. These  chapters also explore the biochemical basis for some of the most important nutritional controversies. Study of human nutrition can be divided into three areas: undernutrition, overnutrition, and ideal nutrition. Undernutrition is not a primary concern in this country  because nutritional deficiency diseases are now quite rare. Overnutrition is a particularly serious problem in developed countries. Current estimates suggest that  between 15% and 30% of the U.S. population is obese, and obesity is known to have a number of serious health consequences. Finally, there is increasing interest  today in the concept of ideal or optimal nutrition. This is a concept that has meaning only in an affluent society. Only when food supply becomes abundant enough so  that deficiency diseases are a rarity does it become possible to consider long­range effects of nutrients on health. This is probably the most exciting area of nutrition  today. 27.2—  Energy Metabolism Energy Content of Food Is Measured in Kilocalories You should be well acquainted with the energy requirements of the body. Much of the food we eat is converted to ATP and other high­energy compounds, which are  utilized to drive biosynthetic pathways, generate nerve impulses, and power muscle contraction. We generally describe the energy content of foods in terms of  calories. Technically speaking, we are actually referring to kilocalories of heat energy released by combustion of that food in the body. Some nutritionists prefer the  term kilojoule (a measure of mechanical energy), but since the American public is likely to be counting calories rather than joules in the foreseeable future, we will  restrict ourselves to that term. Caloric values of protein, fat, carbohydrate, and alcohol are roughly 4, 9, 4, and 7 kcal g–1, respectively. Given these data and the  composition of the food, it is simple to calculate the caloric content (input) of the foods we eat. Calculating caloric content of foods does not appear to be a major  problem in this country. Millions of Americans are able to do it with ease. The problem lies in balancing caloric input with caloric output. Where do these calories go? Energy Expenditure Is Influenced by Four Factors There are four principal factors that affect individual energy expenditure: surface area (which is related to height and weight), age, sex, and activity level. (1) The effects  of surface area are thought to be simply related to the rate of heat loss by the body—the greater the surface area, the greater the rate of heat loss. While it may seem  surprising, a lean individual actually has a greater surface area, and thus a greater energy requirement, than an obese individual of the same weight. (2) Age may reflect  two factors: growth and lean muscle mass. In infants and children more energy expenditure is required for rapid growth, and this is reflected in a higher basal  metabolic rate (rate of energy utilization in resting state). In adults (even lean adults), muscle tissue is gradually replaced with fat and water during the aging process,  resulting in a 2% decrease

 

  

Page 1089

in basal metabolic rate (BMR) per decade of adult life. (3) As for sex, women tend to have a lower BMR than men due to a smaller percentage of lean muscle mass  and the effects of female hormones on metabolism. (4) The effect of activity levels on energy requirements is obvious. However, most of us overemphasize the  immediate, as opposed to the long­term, effects of exercise. For example, one would need to jog for over an hour to burn up the calories found in one piece of apple  pie. Yet, the effect of a regular exercise program on energy expenditure can be quite beneficial. Regular exercise increases lean muscle mass, which has a higher basal  metabolic rate than adipose tissue, allowing one to burn up calories more rapidly 24 hours a day. A regular exercise program should be designed to increase lean  muscle mass and should be repeated 3–5 days a week but need not be aerobic exercise to have an effect on basal metabolic rate. For an elderly or infirm individual,  even daily walking may, with time, help to increase basal metabolic rate slightly. Hormone levels are important also, since thyroxine, sex hormones, growth hormone, and, to a lesser extent, epinephrine and cortisol increase BMR. The effects of  epinephrine and cortisol probably explain in part why severe stress and major trauma significantly increase energy requirements. Finally, energy intake itself has an  inverse relationship to expenditure in that during periods of starvation or semistarvation BMR can decrease up to 50%. This is of great survival value in cases of  genuine starvation, but not much help to the person who wishes to lose weight on a calorie­restricted diet. 27.3—  Protein Metabolism Dietary Protein Serves Many Roles Including Energy Production Protein carries a certain mystique as a "body­building" food. While it is true that protein is an essential structural component of all cells, protein is equally important for  maintaining the output of essential secretions such as digestive enzymes and peptide or protein hormones. Protein is also needed to synthesize plasma proteins, which  are essential for maintaining osmotic balance, transporting substances through the blood, and maintaining immunity. However, the average adult in this country  consumes far more protein than needed to carry out these essential functions. Excess protein is treated as a source of energy, with the glucogenic amino acids being  converted to glucose and the ketogenic amino acids converted to fatty acids and keto acids. Both kinds of amino acids will eventually be converted to triacylglycerol  in adipose tissue if fat and carbohydrate supplies are already adequate to meet energy requirements. Thus for most of us the only body­building obtained from high­ protein diets is in adipose tissue. It has always been popular to say that the body has no storage depot for protein, and thus adequate dietary protein must be supplied with every meal. However, in  actuality, this is not quite accurate. While there is no separate class of "storage" protein, there is a certain percentage of body protein that undergoes a constant process  of breakdown and resynthesis. In the fasting state the breakdown of this store of body protein is enhanced, and the resulting amino acids are utilized for glucose  production, synthesis of nonprotein nitrogenous compounds, and synthesis of the essential secretory and plasma proteins described above (see also Chapter 14). Even  in the fed state, some of these amino acids are utilized for energy production and as biosynthetic precursors. Thus the turnover of body protein is a normal process— and an essential feature of what is called nitrogen balance.

 

  

Page 1090

Nitrogen Balance Relates Intake of Nitrogen to Its Excretion Nitrogen balance (Figure 27.1) is a comparison between intake of nitrogen (chiefly in the form of protein) and excretion of nitrogen (chiefly in the form of undigested  protein in the feces and urea and ammonia in urine). A normal adult is in nitrogen equilibrium, with losses just balanced by intake. Negative nitrogen balance results  from inadequate dietary intake of protein, since amino acids utilized for energy and biosynthetic reactions are not replaced. It also occurs in injury when there is net  destruction of tissue and in major trauma or illness when the body's adaptive response causes increased catabolism of body protein stores. Positive nitrogen balance is  observed whenever there is a net increase in the body protein stores, such as in growing children, pregnant women, or convalescing adults. Essential Amino Acids Must Be Present in the Diet In addition to the amount of protein in the diet, several other factors must be considered. One is the complement of essential amino acids present in the diet. Essential  amino acids are those amino acids that cannot be synthesized by the body (Chapter 11). If just one of these essential amino acids is missing from the diet, the body  cannot synthesize new protein to replace the protein lost due to normal turnover, and a negative nitrogen balance results (Figure 27.1).

Figure 27.1  Factors affecting nitrogen balance.  Schematic representations  of the metabolic interrelationship involved in determining nitrogen  balance. Each figure represents the nitrogen balance resulting from a  particular set of metabolic conditions. The dominant pathways in each  situation are indicated by heavy red arrows.

 

  

Page 1091

Obviously then, the complement of essential amino acids in any dietary protein will determine how well it can be used by the body. Generally, most animal proteins contain all essential amino acids in about the quantities needed by the human body. Vegetable proteins, on the other hand, often lack  one or more essential amino acids and may, in some cases, be more difficult to digest. Even so, vegetarian diets can provide adequate protein provided enough extra  protein is consumed to provide sufficient quantities of the essential amino acids and/or two or more different proteins are consumed together, which complement each  other in amino acid content. For example, if corn (which is deficient in lysine) is combined with legumes (deficient in methionine but rich in lysine), the efficiency of  utilization for the combination of the two vegetable proteins approaches that of animal protein. The adequacy of vegetarian diets with respect to protein and calories is  discussed more fully in Clin. Corr. 27.1, and the need for high­quality protein in low­protein diets in renal disease is discussed in Clin. Corr. 27.2. Protein Sparing Is Related to Dietary Content of Carbohydrate and Fat Another factor that must be considered in determining protein requirements is dietary intake of fat and carbohydrate. If these components are present in insufficient  quantities, some dietary protein must be used for energy generation and is unavailable for building and replacing tissue. Thus as energy (calorie) content of the diet from  carbohydrate and fat increases, the need for protein decreases. This is referred to as protein sparing. Carbohydrate is somewhat more efficient at protein sparing  than fat—presumably because carbohydrate can be used as an energy source by almost all tissues, whereas fat cannot. Normal Adult Protein Requirements Depend on Diet Assuming adequate calorie intake and 75% efficiency of utilization, which is typical of mixed protein in the average American diet, the recommended CLINICAL CORRELATION 27.1  Vegetarian Diets and Protein–Energy Requirements One of the most important problems of a purely vegetarian diet (as opposed to a lacto­ ovo vegetarian diet) is the difficulty in obtaining sufficient calories and protein. Potential  caloric deficit results from the fact that the caloric densities of fruits and vegetables are  much less than the meats they replace (30–50 cal per 100 g versus 150–300 cal per 100  g). The protein problem is generally threefold: (1) most plant products contain much less  protein (1–2 g of protein per 100 g versus 15–20 g per 100 g); (2) most plant protein is  of low biological value; and (3) some plant proteins are incompletely digested. Actually,  well­designed vegetarian diets usually provide enough calories and protein for the average  adult. In fact, the reduced caloric intake may well be of benefit because strict vegetarians  do tend to be lighter than their nonvegetarian counterparts. However, whereas an adult male may require about 0.8 g of protein and 40 cal kg–1 of  body weight, a young child may require 2–3 times that amount. Similarly, a pregnant  woman needs an additional 10 g of protein and 300 cal day–1 and a lactating woman an  extra 15 g of protein and 500 cal. Thus both young children and pregnant and lactating  women run a risk of protein–energy malnutrition. Children of vegetarian mothers generally  have a lower birth weight than children of mothers consuming a mixed diet. Similarly,  vegetarian children generally have a slower rate of growth through the first 5 years, but  generally catch up by age 10. It is possible to provide sufficient calories and protein even for these high­risk groups  provided the diet is adequately planned. Three principles should be followed to design a  calorie–protein­sufficient vegetarian diet for young children: (1) whenever possible,  include eggs and milk in the diet; they are both excellent sources of calories and high­ quality protein; (2) include liberal amounts of those vegetable foods with high­caloric  density in the diet, including nuts, grains, dried beans, and dried fruits; and (3) include  liberal amounts of high­protein vegetable foods that have complementary amino acid  patterns. It used to be thought that these complementary proteins must be present in the  same meal. Recent animal studies, however, suggest that a meal low in (but not devoid of)  an essential amino acid may be supplemented by adding the limiting amino acid at a  subsequent meal. First International Congress on Vegetarian Nutrition. Proc. Am. J. Clin. Nutr. 48(Suppl.  1):707, 1988; and Saunders, T. A. B. Vegetarian diets and children. Pediatr. Nutr.,  42:955, 1995.    

  

Page 1092

CLINICAL CORRELATION 27.2  Low­Protein Diets and Renal Disease Chronic renal failure is characterized by the buildup of the end products of protein  catabolism, mainly urea. Some degree of dietary protein restriction is usually necessary  because these toxic end products are responsible for many of the symptoms associated  with renal failure. The amount of protein restriction is dependent on the severity of the  disease. It is easy to maintain patients in nitrogen balance for prolonged periods on diets  containing as little as 40 g of protein/day if the diet is calorically sufficient. Diets containing  less than 40 g/day pose problems. Protein turnover continues and a balance must be  found between enough protein to avoid negative nitrogen balance and little enough to  avoid buildup of waste products. The strategy employed in such diets is twofold: (1) provide a minimum of protein,  primarily protein of high BV, and (2) provide the rest of the daily calories as  carbohydrates and fats. The goal is to provide just enough essential amino acids to  maintain positive nitrogen balance. In turn, the body should be able to synthesize the  nonessential amino acids from other nitrogen­containing metabolites. Enough  carbohydrate and fat are provided so that essentially all dietary protein can be spared  from energy metabolism. With this type of diet, it is possible to maintain a patient on 20 g  of protein per day for considerable periods. Because of the difficulty in maintaining  nitrogen equilibrium at such low­protein intakes, the patient's protein status should be  monitored. This can be done by measuring parameters such as serum albumin and  transferrin. Moreover, such diets are extremely monotonous and difficult to follow. A typical 20­g  protein diet is shown below: 1. One egg plus 3/4 cup milk or 1 additional egg or 1 oz of meat. 2. One­half pound of deglutenized (low­protein) wheat bread; all other breads and  cereals must be avoided—this includes almost all baked goods. 3. A limited amount of low­protein, low­potassium fruits and vegetables. 4. Sugars and fats to make up the rest of the needed calories; however, cakes, pies, and  cookies need to be avoided. The palatability of these diets can be improved considerably by starting with a vegan diet  and supplementing it with a mixture of essential amino acids and ketoacid analogs of the  essential amino acids. Recent studies indicate that this technique will help preserve renal  function and allow a somewhat greater variety of foods. Goodship, T. H. J., and Mitch, W. E. Nutritional approaches to preserving renal function.  Adv. Intern. Med. 33:377, 1988; Dwyer, J. Vegetarian diets for treating nephrotic  syndrome. Nutr. Rev. 51:44, 1993; and Barsotti, G., Morrell, E., Cupisti, A., Bertoncini,  P., and Giovannetti, S. A special supplemented ''vegan" diet for nephrotic patients. Am.  J. Nephrol. 11:380, 1991.   protein intake is 0.8 g/kg–1 (body weight) day–. This amounts to about 58 g protein day–1 for a 72­kg (160­lb) man and about 44 g day–1 for a 55­kg (120­lb)  woman. These recommendations would need to be increased on a vegetarian diet if overall efficiency of utilization were less than 75%. Protein Requirement Increases during Growth and Recovery from Illness Because dietary protein is essential for synthesis of new body tissue, as well as for maintenance and repair, the need for protein increases markedly during periods of  rapid growth. Such growth occurs during pregnancy, infancy, childhood, and adolescence. Once growth requirements have been considered, age does not seem to have much effect on protein requirements. If anything, the protein requirement may decrease  slightly with age. However, older people need and generally consume less calories, so high­quality protein should provide a larger percentage of their total calories.  Furthermore, some older people may have special protein requirements due to malabsorption problems. Illness, major trauma, and surgery all cause a major catabolic response. Energy needs are very large, and the body responds by increasing production of glucagon,  glucocorticoids, epinephrine, and certain cytokines. In these situations breakdown of body protein is greatly accelerated and a negative nitrogen balance results unless  protein intake is increased (Figure 27.1). Although this increased protein requirement is of little significance in short­term illness, it can be vitally important in the  recovery of hospitalized patients as discussed in the next section (see also Clin. Corr. 27.3).

 

  

Page 1093

CLINICAL CORRELATION 27.3  Providing Adequate Protein and Calories for the Hospitalized Patient The normal metabolic response to infection, trauma, and surgery is a complex and  carefully balanced catabolic state. As discussed in the text, epinephrine, glucagon,  cortisol, and cytokines are released, greatly accelerating the rates of lipolysis, proteolysis,  and gluconeogenesis. The net result is an increased supply of fatty acids, amino acids, and  glucose to meet the increased energy demands of such major stress. The high serum,  glucose results in elevation of circulating insulin levels, which is more than counter­ balanced by increased levels of epinephrine and other hormones. Skeletal muscle, for  example, uses very little of the serum glucose but continues to rely on free fatty acids and  its own catabolized protein as a primary source of energy. It also continues to export  amino acids, primarily alanine, for use elsewhere in the body, resulting in a very rapid  depletion of body protein stores. A highly catabolic hospitalized patient may require 35–45 kcal kg–1 day–1 and 2–3 g of  protein kg–1 day–1. A patient with severe burns may require even more. A physician has a  number of options available to provide this postoperative patient with sufficient calories  and protein to ensure optimal recovery. When the patient is simply unable to ingest  enough food, it may be adequate to supplement the diet with high­calorie–high­protein  preparations, which are usually mixtures of homogenized cornstarch, egg, milk protein,  and flavorings. When the patient is unable to ingest solid food or unable to digest complex  mixtures of foods adequately, elemental diets are usually administered via a nasogastric  tube. Elemental diets consist of small peptides or purified amino acids, glucose and  dextrins, some fat, vitamins, and electrolytes. These diets are sometimes sufficient to meet  most of the short­term caloric and protein needs of a moderately catabolic patient. When  a patient is severely catabolic or unable to digest and absorb foods normally, parenteral  (intravenous) nutrition is necessary. The least invasive method is to use a peripheral, slow­ flow vein in a manner similar to any other i.v. infusion. The main limitation of this method is  hypertonicity. However, a solution of 5% glucose and 4.25% purified amino acids can be  used safely. This solution will usually provide enough protein to maintain positive nitrogen  balance but will rarely provide enough calories for long­term maintenance of a catabolic  patient. The most aggressive nutritional therapy is total parenteral nutrition. Usually an indwelling  catheter is inserted into a large fast­flow vessel such as the superior vena cava, so that the  very hypertonic infusion fluid can rapidly be diluted. This allows solutions of up to 60%  glucose and 4.25% amino acids to be used, providing sufficient protein and most of the  calories for long­term maintenance. Intravenous lipid infusion is often added to boost  calories and provide essential fatty acids. All of these methods can prevent or minimize  the negative nitrogen balance associated with surgery and trauma. The actual choice of  method depends on the patient's condition. As a general rule it is preferable to use the  least invasive technique. Streat, S. J., and Hill, G. L. Nutritional support in the management of critically ill patients  in surgical intensive care. World J. Surg. 11:194, 1987; and The Veterans Affairs Total  Parenteral Nutrition Cooperative Study Group. Perioperative total parenteral nutrition in  surgical patients. N. Engl. J. Med. 325:25, 1991.   27.4—  Protein–Energy Malnutrition The most common form of malnutrition in the world is protein–energy malnutrition (PEM). In developing countries inadequate intake of protein and energy is all  too common, and it is usually the infants and young children who suffer most. While the symptoms of protein–energy insufficiency vary widely from case to case, it is  common to classify most cases as either marasmus or kwashiorkor. Marasmus is usually defined as inadequate intake of both protein and energy. Kwashiorkor is  defined as inadequate intake of protein with adequate energy intake. Often the diets associated with marasmus and kwashiorkor may be similar, with the kwashiorkor  being precipitated by conditions of increased protein demand such as infection. The marasmic infant will have a thin, wasted appearance and will be small for his/her  age. If PEM continues long enough the child will be permanently stunted in both physical and mental development. In kwashiorkor the child will often have a  deceptively plump appearance due to edema. Other telltale symptoms associated with kwashiorkor are dry, brittle hair, diarrhea, dermatitis of various forms, and  retarded growth. Perhaps the most devastating result of both marasmus and kwashiorkor is reduced ability of the afflicted individuals to fight off infection. They have a  reduced number of T lymphocytes (and thus diminished cell­mediated immune response) as well as defects in the generation of phagocytic cells and production of  immuno­globulins, interferon, and other components of the immune system. Many of

 

  

Page 1094

these individuals die from secondary infections, rather than from the starvation itself. The most common form of PEM seen in the United States occurs in the hospital setting. A typical course of events is as follows: The patient has not been eating well  for several weeks or months prior to entering the hospital due to chronic or debilitating illness. He/she enters the hospital with major trauma, severe infection, or for  major surgery, all of which cause a large negative nitrogen balance. This is often compounded by difficulties in feeding the patient or by the necessity of fasting in  preparation for surgery or diagnostic tests. The net result is PEM as measured by low levels of serum albumin and other serum proteins or by decreased cellular  immunity tests. Recent studies have shown that hospitalized patients with demonstrable PEM have delayed wound healing, decreased resistance to infection, increased  mortality, and increased length of hospitalization. Most major hospitals have programs to monitor the nutritional status of their patients and to intervene where  necessary to maintain a positive nitrogen and energy balance (see Clin. Corr. 27.3). 27.5—  Excess Protein–Energy Intake Much has been said in recent years about the large amount of protein that the average American consumes. Certainly most consume far more than needed to maintain  positive nitrogen balance. An average American currently consumes 99 g of protein, 68% from animal sources. However, most studies show that a healthy adult can  consume that amount of protein with no apparent harm. Concern has been raised about possible effects of high­protein intake on calcium requirements. Some studies  suggest that high­protein intake increases urinary loss of calcium and may accelerate bone demineralization associated with aging. However, this issue is far from  settled. Obesity Has Dietary and Genetic Components Perhaps the more serious nutritional problem is excessive energy consumption. In fact, obesity is the most frequent nutritional disorder in the United States. It would,  however, be unfair to label obesity as simply a problem of excess consumption. Overeating plays an important role in many individuals, as does inadequate exercise,  but there is also a strong genetic component as well. While the biochemical mechanisms for this genetic predisposition are unclear, investigators have recently identified  an obesity gene in mice that appears to regulate obesity through effects on both appetite and deposition of fat. A similar gene exists in humans, but its metabolic  function is still not known (see p. 378). Detailed characterization of this and other genes that predispose to obesity in animals may yield valuable clues to the causes  and treatment of obesity in humans. Metabolic Consequences of Obesity Have Significant Health Implications A discussion of the treatment of obesity is clearly beyond the scope of this chapter, but it is worthwhile to consider some of the metabolic consequences of obesity.  One striking clinical feature of overweight individuals is a marked elevation of serum free fatty acids, cholesterol, and triacylglycerols irrespective of the dietary intake  of fat. Why is this? Obesity is obviously associated with an increased number and/or size of adipose cells. These cells contain fewer Insulin receptors and thus  respond more poorly to insulin, resulting in increased activity of the hormone­sensitive lipase. The increased lipase activity

 

  

Page 1095

along with the increased mass of adipose tissue is probably sufficient to explain the increase in circulating free fatty acids. These excess fatty acids are carried to the  liver and metabolized to acetyl CoA, a precursor for triacylglycerol and cholesterol synthesis. Excess triacylglycerol and cholesterol are released as very low density  lipoprotein particles, leading to higher circulating levels of both triacylglycerol and cholesterol (see Chapters 9 and 10). A second striking finding in obese individuals is higher fasting blood sugar levels and decreased glucose tolerance. Fully 80% of adult­onset diabetics are overweight.  Again the culprit appears to be the decrease in insulin receptors, since many adult­onset diabetics have higher than normal insulin levels. This hyperinsulinemia appears  to stimulate the sympathetic nervous system, leading to sodium and water retention and vasoconstriction, which tend to increase blood pressure. Because of these  metabolic changes, obesity is a primary risk factor in coronary heart disease, hypertension, and diabetes. This is nutritionally significant because all of these metabolic  changes are reversible. Quite often reduction to ideal weight is the single most important aim of nutritional therapy. Furthermore, when the individual is at ideal body  weight, the composition of the diet becomes a less important consideration in maintaining normal serum lipid and glucose levels. Any discussion of weight reduction regimens should include a mention of one other metabolic consequence of obesity. As discussed above, obesity can lead to  increased retention of both sodium and water. As the fat stores are metabolized, they produce water (which is denser than the fat), and the water may largely be  retained. In fact, some individuals may actually observe short­term weight gain on certain diets, even though the diet is working perfectly well in terms of breaking  down their adipose tissue. This metabolic fact of life can be psychologically devastating to dieters, who expect quick results for all their sacrifice. 27.6—  Carbohydrates The chief metabolic role of carbohydrates in the diet is for energy production. Any carbohydrate in excess of that needed for energy is converted to glycogen and  triacylglycerol for long­term storage. The body can adapt to a wide range of carbohydrate levels in the diet. Diets high in carbohydrate result in higher steady­state  levels of glucokinase and some of the enzymes involved in the hexose monophosphate shunt and triacylglycerol synthesis. Diets low in carbohydrate result in higher  steady­state levels of some of the enzymes involved in gluconeogenesis, fatty acid oxidation, and amino acid catabolism. Glycogen stores are also affected by the  carbohydrate content of the diet (see Clin. Corr. 27.4). The most common nutritional problems involving carbohydrates are seen in those individuals with various carbohydrate intolerances. The most common form of  carbohydrate intolerance is diabetes mellitus, caused either by lack of insulin production or lack of insulin receptors. This causes an intolerance to glucose and sugars  that can readily be converted to glucose. Dietary treatment of diabetes is discussed in Clinical Correlation 27.5. Lactase insufficiency is also a common disorder of  carbohydrate metabolism affecting over 30 million people in the United States alone. It is most prevalent among blacks, Asians, and Hispanics. Without the enzyme  lactase, the lactose is not significantly hydrolyzed or absorbed. It remains in the intestine where it acts osmotically to draw water into the gut and serves as a substrate  for conversion to lactic acid, CO2, and H2S by intestinal bacteria. The end result is bloating, flatulence, and diarrhea—all of which can be avoided simply by eliminating  milk and milk products from the diet (see p. 1075).

 

  

Page 1096

CLINICAL CORRELATION 27.4  Carbohydrate Loading and Athletic Endurance The practice of carbohydrate loading dates back to observations made in the early 1960s  that endurance during vigorous exercise was limited primarily by muscle glycogen stores.  Of course, the glycogen stores are not the sole energy source for muscle. Free fatty acids  are present in the blood during vigorous exercise and are utilized by muscle along with the  glycogen stores. Once the glycogen stores have been exhausted, however, muscle cannot  rely entirely on free fatty acids without tiring rapidly. This is probably related to the fact  that muscle becomes partially anaerobic during vigorous exercise. While glycogen stores  are utilized equally well aerobically or anaerobically, fatty acids can only be utilized  aerobically. Under those conditions, fatty acids cannot provide ATP rapidly enough to  serve as the sole energy source. Thus the practice of carbohydrate loading to increase glycogen stores was devised for  track and other endurance athletes. Originally, it was thought that it would be necessary  to trick the body into increasing glycogen stores. The original carbohydrate loading  regimen consisted of a 3–4­day period of heavy exercise while on a low­carbohydrate  diet, followed by 1–2 days of light exercise while on a high­carbohydrate diet. The initial  low­carbohydrate–high­energy demand period caused a depletion of muscle glycogen  stores. Apparently, the subsequent change to a high­carbohydrate diet resulted in a slight  rebound effect, with the production of higher than normal levels of insulin and growth  hormone. Under these conditions glycogen storage was favored and glycogen stores  reached almost twice the normal amounts. This practice did increase endurance  significantly. In one study, test subjects on a high­fat and high­protein diet had less than  1.6 g of glycogen per 100 g of muscle and could perform a standardized workload for  only 60 min. When the same subjects then consumed a high­carbohydrate diet for 3 days,  their glycogen stores increased to 4 g per 100 g of muscle and the same workload could  be performed for up to 4 h. While the technique clearly worked, the athletes often felt lethargic and irritable during the  low­carbohydrate phase of the regimen, and the high­fat diet ran counter to current health  recommendations. Fortunately, recent studies show that regular consumption of a high  complex­carbohydrate–low­fat diet during training increases glycogen stores without the  need for tricking the body with sudden dietary changes. Current recommendations are for  endurance athletes to consume a high­carbohydrate diet (with emphasis on complex  carbohydrates) during training. Then carbohydrate intake is increased further (to 70% of  calories) and exercise tapered off during the 2–3 days just prior to an athletic event. This  procedure increases muscle glycogen stores to levels comparable to the original  carbohydrate loading regimen. Conlee, R. K. Muscle glycogen and exercise endurance: a twenty­year perspective.  Exerc. Sport Sci. Rev. 15:1, 1987; Ivey, J. L., Katz, A. L., Cutler, C. L., Sherman, W.  M., and Cayle, E. F. Muscle glycogen synthesis after exercise: effect of time of  carbohydrate ingestion. J. Appl. Physiol. 64:1480, 1988; and Probart, C. K., Bird, P.  J., and Parker, K. A. Diet and athletic performance. Med. Clin. North Am. 77:757,  1993.   CLINICAL CORRELATION 27.5  High­Carbohydrate Versus High­Fat Diets for Diabetics For years the American Diabetes Association has recommended diets that were low in fat  and high in complex carbohydrates and fiber for diabetics. The logic of such a  recommendation seemed to be inescapable. Diabetics are prone to hyperlipidemia with  attendant risk of heart disease, and low­fat diets appeared likely to reduce risk of  hyperlipidemia and heart disease. In addition, numerous clinical studies had suggested that  the high­fiber content of these diets resulted in improved control of blood sugar. This  recommendation has proved to be controversial. An understanding of the controversies  involved illustrates the difficulties in making dietary recommendations for population  groups rather than individuals. In the first place, it is very difficult to make any major  changes in dietary composition without changing other components of the diet. In fact,  most of the clinical trials of the high­carbohydrate–high­fiber diets have resulted in  significant weight reduction, either by design or because of the lower caloric density of the  diet. Since weight reduction improves diabetic control, it is not entirely clear whether the  improvements seen in the treated group were due to the change in diet composition per  se or because of the weight loss. Second, there is significant individual variation in how  diabetics respond to these diets. Many diabetic patients appear to show poorer control  (as evidenced by higher blood glucose levels, elevated VLDL and/or LDL levels, and  reduced HDL levels) on the high­carbohydrate–high­fiber diets than they do on diets high  in monounsaturated fatty acids. However, diets high in monounsaturated fatty acids tend  to have higher caloric density and are inappropriate for overweight individuals with type 2  (non­insulin dependent) diabetes. Thus a single diet may not be equally appropriate for all  diabetics. Even the "glycemic index" concept (Table 27.2) may also turn out to be difficult  to apply to the diabetic population as a whole, because of individual variation. Thus in  1994 the American Diabetes Association abandoned the concept of a single diabetic diet.  Instead, their recommendations focus on achievement of glucose, lipid, and blood  pressure goals, with weight reduction and dietary recommendations based on individual  preferences and what works best to achieve metabolic control in that individual. Anderson, J. W., Gustafson, N.J., Bryant, C. A., and Tietyen­Clark, J. Dietary fiber and  diabetes: a comprehensive review and practical application. J. Am. Diet Assoc. 87:1189,  1987; Jenkins, D. J. A., Wolener, T. M. S., Jenkins, A. L., and Taylor, R. H. Dietary  fiber, carbohydrate metabolism and diabetes. Mol. Aspects Med. 9:97, 1987; Garg, A.,  Grundy, S. M., and Unger, R. H. Comparison of the effects of high and low  carbohydrate diets on plasma lipoproteins and insulin sensitivity in patients with mild  NIDDM. Diabetes 41;1278, 1992; and American Diabetes Association. Nutritional  recommendations and principles for people with diabetes. Diabetes Care 17:519, 1994.    

  

Page 1097

27.7— Fats Triacylglycerols, or fats, are directly utilized by many tissues of the body as an energy source and, as phospholipids, are an important part of membrane structure.  Excess fat in the diet can only be stored as triacylglycerol. As with carbohydrate, the body adapts to a wide range of fat intakes. However, problems develop at the  extremes (either high or low) of fat consumption. At the low end, essential fatty acid (EFA) deficiencies may become a problem. The fatty acids linoleic, linolenic,  and arachidonic acid cannot be made by the body and thus are essential components of the diet. These EFAs are needed for maintaining the function and integrity of  membrane structure, for fat metabolism and transport, and for synthesis of prostaglandins. The most characteristic symptom of essential fatty acid deficiency is a  scaly dermatitis. EFA deficiency is very rare in the United States, occurring primarily in low­birth­weight infants fed artificial formulas lacking EFA and in hospitalized  patients maintained on total parenteral nutrition for long periods of time. At the high end of the scale, there is concern that excess dietary fat causes elevation of serum  lipids and thus an increased risk of heart disease. Recent studies also suggest that high­fat intakes are associated with increased risk of colon, breast, and prostate  cancer, but it is not yet certain whether the cancer risk is associated with fat intake per se or with the excess calories associated with a high­fat diet. To the extent that  fat intake is associated with cancer risk, animal studies suggest that polyunsaturated fatty acids of the w­6 series may be more tumorigenic than other unsaturated fatty  acids. The reason for this is not known, but it has been suggested that prostaglandins derived from the w­6 fatty acids may stimulate tumor progression. 27.8— Fiber Dietary fiber is defined as those components of food that cannot be broken down by human digestive enzymes. It is incorrect, however, to assume that fiber is  indigestible since some fibers are, in fact, at least partially broken down by intestinal bacteria. Knowledge of the role of fiber in human metabolism has expanded  significantly in the past decade. Our current understanding of the metabolic roles of dietary fiber is based on three important observations: (1) there are several different  types of dietary fiber, (2) they each have different chemical and physical properties, and (3) they each have different effects on human metabolism, which can be  understood, in part, from their unique properties. The major types of fiber and their properties are summarized in Table 27.1. Cellulose and most hemicelluloses increase stool bulk and decrease transit time. These  are the types of fiber that should most properly be associated with the effects of fiber on regularity. They decrease intracolonic pressure and appear to play a beneficial  role with respect to diverticular diseases. By diluting out potential carcinogens and speeding their transit through the colon, they may also play a role in reducing the risk  of colon cancer. Lignins have a slightly different role. In addition to their bulk­enhancing properties, they adsorb organic substances such as cholesterol and appear to  have a cholesterol­lowering effect. Mucilaginous fibers, such as pectin and gums, tend to form viscous gels in the stomach and intestine and slow the rate of gastric  emptying, thus slowing the rate of absorption of many nutrients. The most important clinical role of these fibers is to slow the rate at which carbohydrates are digested  and absorbed. Thus both the rise in blood sugar and the subsequent rise in insulin levels are significantly decreased if these fibers are ingested along with carbohydrate­ containing foods. Water­soluble fibers (pectins, gums, some hemicelluloses, and storage polysaccharides) also help to lower serum cholesterol levels in most people.  Whether this is due to their effect on insulin levels (insulin

 

  

Page 1098 TABLE 27.1 Major Types of Fiber and Their Properties Type of Fiber

Major Source in Diet

Chemical Properties

Physiological Effects

Cellulose

Unrefined cereals

Nondigestible

Increases stool bulk

 

Bran

Water insoluble

Decreases intestinal transit time

 

Whole wheat

Absorbs water

Decreases intracolonic pressure

Hemicellulose

Unrefined cereals

Partially digestible

Increases stool bulk

 

Some fruits and vegetables

Usually water insoluble

Decreases intestinal transit time

 

Whole wheat

Absorbs water

Decreases intracolonic pressure

Lignin

Woody parts of vegetables

Nondigestible

Increases stool bulk

 

 

Water insoluble

Bind cholesterol

 

 

Absorbs organic substances

Bind carcinogens

Pectin

Fruits

Digestible

Decreases rate of gastric emptying

 

 

Water soluble

Decreases rate of sugar uptake

 

 

Mucilaginous

Decreases serum cholesterol

Gums

Dried beans

Digestible

Decreases rate of gastric emptying

 

Oats

Water soluble

Decreases rate of sugar uptake

 

 

Mucilaginous

Decreases serum cholesterol

  stimulates cholesterol synthesis and export) or to other metabolic effects (perhaps caused by end products of partial bacterial digestion) is unknown. Vegetables,  wheat, and most grain fibers are the best sources of the water­insoluble cellulose, hemicellulose, and lignin. Fruits, oats, and legumes are the best source of the water­ soluble fibers. Obviously, a balanced diet should include food sources of both soluble and insoluble fiber. 27.9—  Composition of Macronutrients in the Diet From the foregoing discussion it is apparent that there are relatively few instances of macronutrient deficiencies in the American diet. Thus much of the interest in recent  years has focused on whether there is an ideal diet composition consistent with good health. It would be easy to pass off such discussions as purely academic, yet our  understanding of these issues could well be vital. Heart disease, stroke, and cancer kill many Americans each year, and if some experts are even partially correct, many  of these deaths could be preventable with prudent diet. Composition of the Diet Affects Serum Cholesterol With respect to heart disease, the current discussion centers around two key issues: (1) Can serum cholesterol and triacylglycerol levels be controlled by diet? (2)  Does lowering serum cholesterol and triacylglycerol levels protect against heart disease? The controversies centered around dietary control of cholesterol levels  illustrate perfectly the trap one falls into by trying to look too closely at each individual component of the diet instead of the diet as a whole. For example, there are at  least four dietary components that can be identified as having an effect on serum cholesterol: cholesterol itself, polyunsaturated fatty acids (PUFAs), saturated  fatty acids (SFAs), and fiber. It would seem that the more cholesterol one eats, the higher the serum cholesterol should be. However, cholesterol synthesis is tightly  regulated via a feedback control at the hydroxymethylglutaryl­CoA reductase step, so decreases in dietary cholesterol have relatively little effect on serum cholesterol  levels (see p. 415). One can obtain a more significant reduction in cholesterol and triacylglycerol levels by

 

  

Page 1099

increasing the ratio of PUFA/SFA in the diet. Finally, some plant fibers, especially the water­soluble fibers, appear to decrease cholesterol levels significantly. While the effects of various lipids in the diet can be dramatic, the biochemistry of their action is still uncertain. Saturated fats inhibit receptor­mediated uptake of LDL,  but the mechanism is complex. Palmitic acid (saturated, C16) raises cholesterol levels while stearic acid (saturated, C18) is neutral. Polyunsaturated fatty acids lower  both LDL and HDL cholesterol levels, while oleic acid (monounsaturated, C18) appears to lower LDL without affecting HDL levels. Furthermore, the w­3 and w­6  polyunsaturated fatty acids have slightly different effects on lipid profiles (see Clin. Corr. 27.6). However, these mechanistic complexities do not significantly affect  dietary recommendations. Most foods high in saturated fats contain both palmitic and stearic acid and are atherogenic. The data showing oleic acid lowers LDL levels  mean that olive oil, and possibly peanut oil, may be considered as beneficial as polyunsaturated oils. There is very little disagreement with respect to these data. The question is, what can be done with the information? Much of the disagreement arises from the tendency  to look at each dietary factor in isolation. For example, it is debatable whether it is worthwhile placing a patient on a highly restrictive 300­mg cholesterol diet (1 egg =  213 mg of cholesterol) if his serum cholesterol is lowered by only 5–10%. Likewise, changing the PUFA/SFA ratio from 0.3 (the current value) to 1.0 would either  require a radical change in the diet by elimination of foods containing saturated fat (largely meats and fats) or an addition of large amounts of rather unpalatable  polyunsaturated fats to the diet. For many Americans this would be unrealistic. Fiber is another good example. One could expect, at the most, a 5% decrease in serum  cholesterol by adding any reasonable amount of fiber to the diet. (Very few people would eat the CLINICAL CORRELATION 27.6  Polyunsaturated Fatty Acids and Risk Factors for Heart Disease Recent studies confirming that reduction of elevated serum cholesterol levels can reduce  risk of heart disease have rekindled interest in the effects of diet on serum cholesterol  levels and other risk factors for heart disease. We have known for years that one of the  most important dietary factors regulating serum cholesterol levels is the ratio of  polyunsaturated fats (PUFAs) to saturated fats (SFAs) in the diet. One of the most  interesting recent developments is the discovery that different types of polyunsaturated  fatty acids have different effects on lipid metabolism and on other risk factors for heart  disease. As discussed in Chapter 9, there are two families of polyunsaturated essential  fatty acids—the w­6, or linoleic family, and the w­3, or linolenic family. Recent clinical  studies have shown that the w­6 PUFAs (chief dietary source is linoleic acid from plants  and vegetable oils) primarily decrease serum cholesterol levels, with only modest effects  on triacylglycerol levels. The w­3 PUFAs (chief dietary source is eicosapentaenoic acid  from certain ocean fish and fish oils) cause modest decreases in serum cholesterol levels  and significantly lower triacylglycerol levels. The biochemical mechanism behind these  different effects on serum lipid levels is unknown. The w­3 PUFAs have yet another unique physiological effect that may decrease the risk  of heart disease—they decrease platelet aggregation. The mechanism of this effect is a  little clearer. Arachidonic acid (w­6 family) is known to be a precursor of thromboxane  A2 (TXA2), which is a potent proaggregating agent, and prostaglandin I2 (PGI2), which is  a weak antiaggregating agent (see p. 436). The w­3 PUFAs are thought to act by one of  two mechanisms: (1) Eicosapentaenoic acid (w­3 family) may be converted to  thromboxane A3 (TXA3), which is only weakly proaggregating, and prostaglandin I3  (PGI3), which is strongly antiaggregating. Thus the balance between proaggregation and  antiaggregation would be shifted toward a more antiaggregating condition as the w­3  PUFAs displace w­6 PUFAs as a source of precursors to the thromboxanes and  prostaglandins. (2) The w­3 PUFAs may also act by simply inhibiting the conversion of  arachidonic acid to TXA2. The unique potential of eicosapentaenoic acid and other w­3 PUFAs in reducing the risk  of heart disease is being tested in numerous clinical trials. Although the results may affect  dietary recommendations in the future, it is well to keep in mind that no long­term clinical  studies of the w­3 PUFAs have been carried out. No major health organization has  recommended that we replace w­6 with w­3 PUFAs in the American diet. Holub, B. J. Dietary fish oils containing eicosapentaenoic acid and the prevention of  atherosclerosis and thrombosis. Can. Med. Assoc. J. 139:377, 1988; Simopoulos, A. P.  Omega­3 fatty acids in health and disease and in growth and development. Am. J. Clin.  Nutr. 54:438, 1991; and Gapinski, J. P., Van Ruiswyk, J. V., Heudebert, G. R., and  Schectman, G. S. Preventing restenosis with fish­oils following coronary angioplasty. A  meta­analysis. Arch. Intern. Med. 153:1595, 1993.    

  

Page 1100

ten apples per day needed to lower serum cholesterol by 15%.) Are we to conclude then that any dietary means of controlling cholesterol levels is useless? Only if  each element of the diet is examined in isolation. For example, recent studies have shown that vegetarians, who have lower cholesterol intakes plus higher PUFA/SFA  ratios and higher fiber intakes, may average 25–30% lower cholesterol levels than their nonvegetarian counterparts. Perhaps, more to the point, diet modifications of  the type acceptable to the average American have been shown to cause a 10–15% decrease in cholesterol levels in long­term studies. A 7­year clinical trial sponsored  by the National Institutes of Health has proved conclusively that lowering serum cholesterol levels reduces the risk of heart disease in men. It is important to keep in  mind that serum cholesterol is just one of many risk factors. Effects of Refined Carbohydrate in the Diet Are Not Straightforward Much of the nutritional dispute in the area of carbohydrates centers around the amount of refined carbohydrate in the diet. In the past, simple sugars (primarily  sucrose) have been blamed for almost every ill from tooth decay to heart disease and diabetes. In the case of tooth decay, these assertions were clearly correct. In the  case of heart disease, however, the linkage is more obscure (see Clin. Corr. 27.7). The situation with respect to diabetes is probably even less direct. Whereas  restriction of simple sugars is often desirable in patients who already have diabetes, recent studies show less than expected correlation between the type of  carbohydrate ingested and the subsequent rise in serum glucose levels (Table 27.2). Ice cream, for example, causes a much smaller increase in serum glucose levels  than either potatoes or whole wheat bread. It turns out that other components of food—such as protein, fat, and the soluble fibers—are much more important than the  type of carbohydrate present in determining how rapidly glucose will enter the bloodstream. CLINICAL CORRELATION 27.7  Metabolic Adaptation: The Relationship between Carbohydrate Intake and  Serum Triacylglycerols In evaluating the nutrition literature, it is important to be aware that most clinical trials are  of rather short duration (2–6 weeks), while some metabolic adaptations may take  considerably longer. Thus even apparently well­designed clinical studies may lead to  erroneous conclusions that will be repeated in the popular literature for years to come.  For example, several studies carried out in the 1960s and 1970s tried to assess the  effects of carbohydrate intake on serum triacylglycerol levels. Typically, young collegeage  males were given a diet in which up to 50% of their fat calories were replaced with  sucrose or other simple sugars for a period of 2–3 weeks. In most cases serum  triacylglycerol levels increased markedly (up to 50%). This led to the tentative conclusion  that high intake of simple sugars, particularly sucrose, might increase the risk of heart  disease, a notion that was popularized by nutritional best sellers such as "Sugar Blues"  and "Sweet and Dangerous." Unfortunately, while the original conclusions were promoted  in the lay press, the experiments themselves were questioned. Subsequent studies showed  that if these trials were continued for longer periods of time (3–6 months), the  triacylglycerol levels usually normalized. The nature of this slow metabolic adaptation is  unknown. It should be noted that while the interpretation of the original clinical trials may have been  faulty, the ensuing dietary recommendations may not have been entirely incorrect. Many  of the snack and convenience foods in the American diet that are high in sugar are also  high in fat and in caloric density. Thus removing some of these foods from the diet can aid  in weight control, and being overweight is known to contribute to  hypertriacylglycerolemia. Also, some individuals exhibit carbohydrate­induced  hypertriacylglycerolemia. Triacylglycerol levels in these individuals respond dramatically to  diets that substitute foods containing complex carbohydrates and fiber for these foods  containing primarily simple sugars as a carbohydrate source. MacDonald, I. Effects of dietary carbohydrates on serum lipids. Prog. Biochem.  Pharmacol. 8:216, 1973; and Vrana, A., and Fabry, P. Metabolic effects of high  sucrose or fructose intake. World Rev. Nutr. Diet 42:56, 1983.    

  

Page 1101 TABLE 27.2 Glycemic Indexa of Some Selected Foods Grain and cereal products Bread (white)  Bread (whole wheat) Rice (white)  Sponge cake Breakfast cereals

69 ± 5 72 ± 6 72 ± 9 46 ± 6

51 ± 5  80 ± 6  49 ± 8  67 ± 10

Dairy products Ice cream  Milk (whole) Yogurt

Beans (kidney)  Beans (soy)  Peas (blackeye) Fruits

 

Sweet corn  Frozen peas

Beets  Carrots  Potato (white)  Potato (sweet) Dried legumes

 

All bran  Cornflakes  Oatmeal  Shredded wheat Vegetables

Root vegetables

 

59 ± 11 51 ± 6

Apple (Golden Delicious) Banana  Oranges Sugars

  36 ± 8 34 ± 6 36 ± 4

Fructose Glucose  Honey  Sucrose

    64 ± 16    92 ± 20    70 ± 8    48 ± 6

    29 ± 8    15 ± 5    33 ± 4

    39 ± 3    62 ± 9    40 ± 3

    20 ± 5  100    87 ± 8    59 ± 10

Source: Data from Jenkins, D. A., et al. Glycemic index of foods: a physiological basis for  carbohydrate exchange. Am. J. Clin. Nutr. 34:362, 1981. a Glycemic index is defined as the area under the blood glucose response curve for each food 

expressed as a percentage of the area after taking the same amount of carbohydrate as glucose  (mean: 5–10 individuals).

  Mixed Vegetable and Animal Proteins Meet Nutritional Protein Requirements Concern has been voiced recently about the type of protein in the American diet. Epidemiologic data and animal studies suggest that consumption of animal protein is  associated with increased incidence of heart disease and various forms of cancer. One could assume that it is probably not the animal protein itself that is involved, but  the associated fat and cholesterol. What sort of protein should we consume? Although the present diet may not be optimal, a strictly vegetarian diet may not be  acceptable to many Americans. Perhaps a middle road is best. Clearly, there are no known health dangers associated with a mixed diet that is lower in animal protein  than the current American standard. An Increase in Fiber from Varied Sources Is Desirable Because of our current knowledge about effects of fiber on human metabolism, most suggestions for a prudent diet recommend an increase in dietary fiber. The main  question is: "How much is enough?" The current fiber content of the American diet is about 14–15 g per day. Most experts feel that an increase to at least 25–30 g  would be safe and beneficial. Since we know that different fibers have different metabolic roles, this increase in fiber intake should come from a wide variety of fiber  sources—including fresh fruits, vegetables, and legumes as well as the more popular cereal sources of fiber (which are primarily cellulose and hemicellulose). Current Recommendations Are for a "Prudent Diet" Several private and governmental groups have made specific recommendations with respect to the ideal dietary composition for the American public in recent years.  This movement was spearheaded by the Senate Select Committee on Human Nutrition, which first published its Dietary Goals for the United States

 

  

Page 1102

in 1977. The Senate Select Committee recommended that the American public reduce consumption of total calories, total fat, saturated fat, cholesterol, simple sugars,  and salt to ''ideal" goals more compatible with good health (Figure 27.2). In recent years the USDA, the American Heart Association, the American Diabetes  Association, the National Research Council, and the Surgeon General all have published similar recommendations, and the USDA has used these recommendations to  design revised recommendations for a balanced diet (Figure 27.3). These recommendations have become popularly known as the prudent diet. How valid is the  scientific basis of the recommendations for a prudent diet? Is there evidence that a prudent diet will improve the health of the general public? These remain  controversial questions. An important argument against such recommendations is that we presently do not have enough information to set concrete goals. We might be creating some problems  while solving others. For example, the goals of reducing total fat and saturated fat in the diet are best met by replacing animal protein with vegetable protein. This might  reduce the amount of available iron and vitamin B12 in the diet. It is also quite clear that the same set of guidelines do not apply for every individual. For example,  exercise is known to raise serum HDL cholesterol and obesity is known to elevate cholesterol and triacylglycerols and reduce glucose tolerance. Thus the very active  individual who maintains ideal body weight can likely tolerate higher fat and sugar intakes than an obese individual. On the "pro" side, however, it clearly can be argued that all of the dietary recommendations are in the right direction for reducing nutritional risk factors in the general  population. Besides, similar diets have been consumed by our ancestors and by people in other countries with no apparent harm. Whatever

Figure 27.2  United States dietary goals.  Graphical comparison of the composition of the current U.S. diet and  the dietary goals for the U.S. population suggested by the Senate Select  Committee on Human Nutrition.  From Dietary Goals for the United States, 2nd ed.  Washington, DC: U.S. Government Printing Office, 1977.

 

  

Page 1103

Figure 27.3  USDA food pyramid.  Graphical representation of USDA recommendations for a balanced diet.  HG Bulletin #252. Washington, DC: U.S. Government Printing Office, 1992.

the outcome of this debate, it will undoubtedly shape much of our ideas concerning the role of nutrition in medicine. Bibliography Protein Energy Malnutrition in Hospitalized Patients The Veterans Affairs Total Parenteral Nutrition Cooperative Study Group. Perioperative total parenteral nutrition in surgical patients. N. Engl. J. Med. 325:525,  1991. Metabolic Consequences of Obesity Hershcopf, R. J., and Bradlow, H. L. Obesity, diet, endogenous estrogens, and the risk of hormone­sensitive cancer. Am. J. Clin. Nutr. 45:283, 1987. Maxwell, M. H., and Waks, A. U. Obesity and hypertension. Bibl. Cardiol. 41:29, 1987. Pi­Sunyer, F. X. Health implications of obesity. Am. J. Clin. Nutr. 53:15955, 1991. Simopoulos, A. P. Obesity and carcinogenesis: historical perspective. Am. J. Clin. Nutr. 45:271, 1987. Metabolic Predisposition to Obesity Bjorntorp, P. Fat cell distribution and metabolism. Ann. N.Y. Acad. Sci. 499:66, 1987. Bray, G. A. Obesity—a disease of nutrient or energy balance? Nutr. Rev. 45:33, 1987. Campfield, L. A., Smith, F. J., Guisez, Y., Devos, R., and Burn, P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central  neural networks. Science 269:546, 1995.

 

  

Page 1104

Dulloo, A. G., and Miller, D. S. Obesity: a disorder of the sympathetic nervous system. World Rev. Nutr. Diet. 50:1, 1987. Halaas, J. L., Gajiwala, K. S., Maffei, M., Cohen, S. L., Chait, B. T., Rabinowitz, D., Lallone, R. L., Burley, S. K., and Friedman, J. M. Weight­reducing effects of  the plasma protein encoded by the obese gene. Science 269:543, 1995. Complex Carbohydrates and Fiber Anderson, J. W. Fiber and health: an overview. Am. J. Gastroenterol. 81:892, 1986. Eastwood, M. Dietary fiber and the risk of cancer. Nutr. Rev. 45:193, 1987. Miller, J. C. B. Importance of glycemic index in diabetes. Am. J. Clin. Nutr. 59(Suppl.):747S, 1994. Shankar, S., and Lanza, E. Dietary fiber and cancer prevention. Hematol. Oncol. Clin. North Am. 5:25, 1991. Wolever, T. M. S., Jenkins, D. J. A., Jenkins, A. L., and Josse, R. G. The glycemic index: methodology and clinical implications. Am. J. Clin. Nutr. 54:846, 1991. Macronutrient Composition and Health Gardener, C. D., and Kraemer, H. C. Monounsaturated versus polyunsaturated dietary fat and serum lipids. Arterioscler. Thromb. Vasc. Biol. 15:1917, 1995. Gorlin, R. The biological actions and potential clinical significance of dietary  ­3 fatty acids. Arch. Intern. Med. 148:2043, 1988. Grundy, S. M. Monounsaturated fatty acids, plasma cholesterol, and coronary heart disease. Am. J. Clin. Nutr. 45:1168, 1987. Grundy, S. M., et al. Rationale of the diet—heart statement of the American Heart Association, news from the American Heart Association. Circulation 65:839A,  1982. Kisselbah, A., and Schetman, G. Polyunsaturated and saturated fat, cholesterol, and fatty acid supplementation. Diabetes Care 11:129, 1988. Kritchevsky, D., and Klurfeld, D. M. Caloric effects in experimental mammary tumorigenesis. Am. J. Clin. Nutr. 45:236, 1987. Rasmussen, O. W., Thomsen, C., Hansen, K. W., Vesterland, M., Winther, E., and Hermansen, K. Effects on blood pressure, glucose, and lipid levels of a high  monounsaturated fat diet compared with a high­carbohydrate diet in NIDDM subjects. Diabetes Care 16:156S, 1993. Simopoulus, A. P. Omega­3 Fatty Acids in Health and Disease and in Growth and Development. Am. J. Clin. Nutr. 54:438, 1991. Welsh, C. W. Enhancement of mammary tumorigenesis by dietary fat: review of potential mechanisms. Am. J. Clin. Nutr. 45:191, 1987. Dietary Recommendations American Heart Association. Recommendations for Treatment of Hyperlipidemia in Adults. Dallas: American Heart Association, 1984. Food and Nutrition Board of the National Academy of Sciences. Towards Healthful Diets. Washington, DC: U.S. Government Printing Office, 1980. National Research Council. Diet, Nutrition and Cancer. Washington, DC: National Academy Press, 1982. Senate Select Committee on Human Nutrition. Dietary Goals for the United States, 2nd ed., Stock No. 052–070–04376–8. Washington, DC: U.S. Government  Printing Office, 1977. Truswell, A. S. Evolution of dietary recommendations, goals, and guidelines. Am. J. Clin. Nutr. 45:1060, 1987. U.S. Department of Agriculture. Nutrition and Your Health, Dietary Guidelines for Americans, Stock No. 017–001–00416–2. Washington, DC: U.S.  Government Printing Office, 1980. U.S. Department of Agriculture. The Food Guide Pyramid, Stock No. HSG­252. Hyattsville, MD: Human Nutrition Information Service, 1992. U.S. Department of Health and Human Services. The Surgeon General's Report on Nutrition and Health, Stock No. 017–001–00465–1. Washington, DC: U.S.  Government Printing Office, 1988. Questions C. N. Angstadt and J. Baggott 1. Of two people with approximately the same weight, the one with the higher basal energy requirement would most likely be: A. taller. B. female if the other were male. C. older. D. under less stress. E. all of the above. 2. Basal metabolic rate: A. is not influenced by energy intake. B. increases in response to starvation. C. may decrease up to 50% during periods of starvation. D. increases in direct proportion to energy expenditure. E. is not responsive to changes in hormone levels. 3. The primary effect of the consumption of excess protein beyond the body's immediate needs will be: A. excretion of the excess as protein in the urine. B. an increase in the "storage pool" of protein. C. an increased synthesis of muscle protein. D. an enhancement in the amount of circulating plasma proteins. E. an increase in the amount of adipose tissue. 4. Which of the following individuals would most likely be in nitrogen equilibrium? A. a normal, adult male B. a normal, pregnant female C. a growing child D. an adult male recovering from surgery E. a normal female on a very low protein diet 5. Vegetarian diets: A. cannot meet the body's requirements for all of the essential amino acids. B. contain only protein that is very readily digestible. C. are adequate as long as two different vegetables are consumed in the same meal. D. would require less total protein than meat proteins to meet the requirement for all of the essential amino acids. E. require that proteins consumed have essential amino acid contents that complement each other. 6. In which of the following circumstances would a protein intake of 0.8 g of protein kg–1 (body weight) day–1 probably be adequate? A. vegetarian diet B. infancy C. severe burn D. about 85–90% of total calories supplied by carbohydrate and fat E. pregnancy 7. Kwashiorkor is: A. the most common form of protein–calorie malnutrition in the United States. B. characterized by a thin, wasted appearance. C. an inadequate intake of food of any kind.

 

  

Page 1105

D. an adequate intake of total calories but a specific deficiency of protein. E. an adequate intake of total protein but a deficiency of the essential amino acids. 8. An excessive intake of calories: A. usually does not have adverse metabolic consequences. B. leads to metabolic changes that are usually irreversible. C. frequently leads to elevated serum levels of free fatty acids, cholesterol, and triglycerides. D. is frequently associated with an increased number of insulin receptors. E. is the only component of obesity. 9. A diet very low in carbohydrate: A. would cause weight loss because there would be no way to replenish citric acid cycle intermediates. B. would result in no significant metabolic changes. C. could lead to a chronic ketosis. D. would lead to water retention. E. would be the diet of choice for a diabetic. 10. Lactase insufficiency: A. is a more serious disease than diabetes mellitus. B. has no clinical symptoms. C. causes an intolerance to glucose. D. causes an intolerance to milk and milk products. E. affects utilization of milk by the liver. 11. Dietary fat: A. is usually present, although there is no specific need for it. B. if present in excess, can be stored as either glycogen or adipose tissue triacylglycerol. C. should include linoleic and linolenic acids. D. should increase on an endurance training program in order to increase the body's energy stores. E. if present in excess, does not usually lead to health problems. 12. Which of the following statements about dietary fiber is/are correct? A. Water­soluble fiber helps to lower serum cholesterol in most people. B. Mucilaginous fiber slows the rate of digestion and absorption of carbohydrates. C. Insoluble fiber increases stool bulk and decreases transit time. D. All of the above are correct. E. None of the above is correct. 13. Which one of the following dietary regimens would be most effective in lowering serum cholesterol? A. restrict dietary cholesterol B. increase the ratio of polyunsaturated to saturated fatty acids C. increase fiber content D. restrict cholesterol and increase fiber E. restrict cholesterol, increase PUFA/SFA, increase fiber 14. Most nutrition experts currently agree that an excessive consumption of sugar causes: A. tooth decay. B. diabetes. C. heart disease. D. permanently elevated triacylglycerol levels. E. all of the above. Refer to the following for Questions 15 and 16: A. 10% of total calories B. 12% of total calories C. 30% of total calories D. 48% of total calories E. 58% of total calories 15. The dietary goal recommended by the Senate Select Committee on Human Nutrition for Polyunsaturated fatty acids. 16. The dietary goal recommended by the Senate Select Committee on Human Nutrition for complex carbohydrates and naturally occurring sugars. 17. A complete replacement of animal protein in the diet by vegetable protein: A. would be expected to have no effect at all on the overall diet. B. would reduce the total amount of food consumed for the same number of calories. C. might reduce the total amount of iron and vitamin B12 available. D. would be satisfactory regardless of the nature of the vegetable protein used. E. could not satisfy protein requirements. Answers 1. A A taller person with the same weight would have a greater surface area. B: Males have higher energy requirements than females. C: Energy requirements decrease  with age. D: Stress, probably because of the effects of epinephrine and cortisol, increase energy requirements (pp. 1088–1089). 2. C This is part of the survival mechanism in starvation. A and B: BMR decreases when energy intake decreases. D: BMR as defined (p. 1088) is independent of  energy expenditure. Only when the exercise is repeated on a daily basis so that lean muscle mass is increased does BMR also increase. E: Many hormones increase  BMR (p. 1089). 3. E Excess protein is treated like any other excess energy source and stored (minus the nitrogen) eventually as adipose tissue fat (p. 1089). A: Protein is not found in  normal urine except in very small amounts. The excess nitrogen is excreted as NH4+ and urea, whereas the excess carbon skeletons of the amino acids are used as  energy sources. B–D: There is no discrete storage form of protein, and although some muscle and structural protein is expendable, there is no evidence that increased  intake leads to generalized increased protein synthesis.

 

  

Page 1106

4. A B, C, and D: Although normal, pregnancy is also a period of growth, requiring positive balance, as does a period of convalescence. E: Inadequate protein intake  leads to negative balance (p. 1100). 5. E A–E: It is possible to have adequate protein intake on a vegetarian diet provided enough is consumed (protein content is generally low and may be more difficult  to digest) and there is a mixture of proteins that supplies all of the essential amino acids since individual proteins are frequently deficient in one or more foods (Clin.  Corr. 27.1, p. 1091). 6. D This level of calories from carbohydrate and fat is more than adequate for protein sparing. A: Essential amino acids are low in vegetable protein. B, C, and E:  Periods of rapid growth require extra protein, as does major trauma (p. 1092). 7. D A: The most common protein–calorie malnutrition occurs in severely ill, hospitalized patients who would be more likely to have generalized malnutrition. B and C:  These are the characteristics of marasmus. E: This would lead to negative nitrogen balance but does not have a specific name (p. 1093). 8. C Probably because an increased number and/or size of adipose cells will contain fewer insulin receptors. A: Excess caloric intake will lead to obesity if continued  long enough. B: Fortunately, most of the changes accompanying obesity can be reversed if weight is lost. D: Many of the adverse effects of obesity are associated with  an increased number of adipocytes that are deficient in insulin receptors. E: Inadequate exercise and genetic components also play roles in obesity (pp. 1094–1095). 9. C A: This is a popular myth but untrue because many amino acids are glucogenic. B and C: The liver adapts by increasing gluconeogenesis, fatty acid oxidation, and  ketone body production. D: Low carbohydrate leads to a depletion of glycogen with its stored water, accounting for rapid initial weight loss on this kind of diet (p.  1095, Clin. Corr. 27.4). E: Diabetic diets need to be individualized. There is currently no generalized recommendation for the carbohydrate content of a diabetic diet  (p. 1095 and Clinical Correlation 27.5). 10. D B, D, and E: Lactase insufficiency is an inability to digest the sugar in milk products, causing intestinal symptoms, but is easily treated by eliminating milk products  from the diet. A and C: Diabetes, caused by inadequate insulin or insulin receptors, inhibits appropriate utilization of glucose (p. 1095). 11. C A and C: Linoleic and linolenic acids are essential fatty acids and so must be present in the diet. B and D: Excess carbohydrate can be stored as fat but the  reverse is not true. D: Carbohydrate loading has been shown to increase endurance. E: High­fat diets are associated with many health risks (p. 1096, Clin. Corr.  27.5). 12. D These each illustrate the different properties and roles of the common kinds of fiber (p. 1097). 13. E Any of the measures alone would decrease serum cholesterol slightly, but to achieve a reduction of more than 15% requires all three (pp. 1098–1100). 14. A This is the only direct linkage shown. B and C: There may be an association with these conditions but not a direct cause–effect relationship. D: Transient  elevations may occur on an isocaloric switch from a high­starch to a high­simple­sugar diet but not a permanent elevation (p. 1100). 15. A See Figure 27.2, p. 1102. 16. D See Figure 27.2, p. 1102. 17. C A and C: This would reduce the amount of fat, especially saturated fat, but could also reduce the amount of necessary nutrients that come primarily from animal  sources. B: The protein content of vegetable0s is quite low, so much larger amounts of vegetables would have to be consumed. D and E: It is possible to satisfy  requirements for all of the essential amino acids completely if vegetables with complementary amino acid patterns, in proper amounts, are consumed (p. 1101, Clin.  Corr. 27.1).

 

  

Page 1107

Chapter 28—  Principles of Nutrition II: Micronutrients Stephen G. Chaney

28.1 Overview

1108

28.2 Assessment of Malnutrition

1108

28.3 Recommended Dietary Allowances

1109

28.4 Fat­Soluble Vitamins

1109

Vitamin A Is Derived from Plant Carotenoids

1109

Vitamin D Synthesis in the Body Requires Sunlight

1111

Vitamin E Is a Mixture of Tocopherols

1114

Vitamin K Is a Quinone Derivative

1116

28.5 Water­Soluble Vitamins

1118

28.6 Energy­Releasing, Water­Soluble Vitamins

1119

Thiamine (Vitamin B1) Forms the Coenzyme Thiamine Pyrophosphate  (TPP)

1119

Riboflavin Is Part of FAD and FMN

1121

Niacin Is Part of NAD and NADP

1121

Pyridoxine (Vitamin B6) Forms the Coenzyme Pyridoxal Phosphate

1121

Pantothenic Acid and Biotin Are Also Energy­Releasing Vitamins

1122

28.7 Hematopoietic Water­Soluble Vitamins

1123

(Folic Acid (Folacin) Functions As Tetrahydrofolate in One­Carbon  Metabolism

1123

Vitamin B12 (Cobalamine) Contains Cobalt in a Tetrapyrrole Ring

1125

28.8 Other Water­Soluble Vitamins

1127

Ascorbic Acid Functions in Reduction and Hydroxylation Reactions

1127

28.9 Macrominerals

1128

Calcium Has Many Physiological Roles

1128

Magnesium Is Another Important Macromineral

1129

28.10 Trace Minerals Iron Is Efficiently Reutilized

1130

Iodine Is Incorporated into Thyroid Hormones

1130

Zinc Is a Cofactor for Many Enzymes

1131

Copper Is also a Cofactor for Important Enzymes

1131

Chromium Is a Component of Glucose Tolerance Factor

1132

Selenium Is a Scavenger of Peroxides

1132

Manganese, Molybdenum, Fluoride, and Boron Are Other Trace  Elements

1132

28.11 The American Diet: Fact and Fallacy

1132

28.12 Assessment of Nutritional Status in Clinical Practice

1133

Bibliography

1135

Questions and Answers

1135

Clinical Correlations

   

  

1130

 

28.1 Nutritional Considerations for Cstic Fibrosis

1112

28.2 Renal Osteodystrophy

1113

28.3 Nutritional Considerations in the Newborn

1117

28.4 Anticonvulsant Drugs and Vitamin Requirements

1118

28.5 Nutritional Considerations in the Alcoholic

1120

28.6 Vitamin B6 Requirements for Users of Oral Contraceptives

1124

28.7 Diet and Osteoporosis

1129

28.8 Nutritional Considerations for Vegetarians

1134

28.9 Nutritional Needs of Elderly Persons

1134

Page 1108

28.1—  Overview Micronutrients play a vital role in human metabolism, being involved in almost every known biochemical reaction and pathway. However, the biochemistry of these  nutrients is of little interest unless we also know if dietary deficiencies are likely. The American diet is undoubtedly the best it has ever been. Our current food supply  provides us with an abundant variety of foods all year long and deficiency diseases have become medical curiosities. However, our diet is far from optimal. The old  adage that we get everything we need from a balanced diet is true only if we eat a balanced diet. Unfortunately, most Americans do not consume a balanced diet.  Foods of high caloric density and low nutrient density (often referred to as empty calories or junk food) are an abundant and popular part of the American diet, and  our nutritional status suffers because of these food choices. Obviously then, neither alarm nor complacency is justified. We need to know how to evaluate the  adequacy of our diet. 28.2—  Assessment of Malnutrition There are three increasingly stringent criteria for measuring malnutrition. 1. Dietary intake studies, which are usually based on a 24­hour recall, are the least stringent. The 24­hour recalls almost always tend to overestimate the number of  people with deficient diets. Also, poor dietary intake alone is usually not a problem in this country unless the situation is compounded by increased need. 2. Biochemical assays, either direct or indirect, are a more useful indicator of the nutritional status of an individual. At their best, they indicate subclinical nutritional  deficiencies that can be treated before actual deficiency diseases develop. However, all biochemical assays are not equally valid—an unfortunate fact that is not  sufficiently recognized. Changes in biochemical parameters due to stress need to be interpreted with caution. The distribution of many nutrients in the body changes  dramatically in a stress situation such as illness, injury, and pregnancy. A drop in level of a nutrient in one tissue compartment (usually blood) need not signal a  deficiency or an increased requirement. It could simply reflect a normal metabolic adjustment to stress. 3. The most stringent criterion is the appearance of clinical symptoms. However, it is desirable to intervene long before symptoms became apparent. The question remains: When should dietary surveys or biochemical assays be interpreted to indicate the necessity of nutritional intervention? The following general  guidelines are useful. Dietary surveys are seldom a valid indication of general malnutrition unless the average intake for a population group falls significantly below the  standard (usually two­thirds of the Recommended Dietary Allowance) for one or more nutrients. However, by looking at the percentage of people within a population  group who have suboptimal intake, it is possible to identify high­risk population groups that should be monitored more closely. Biochemical assays can definitely  identify subclinical cases of malnutrition where nutritional intervention is desirable provided (a) the assay has been shown to be reliable, (b) the deficiency can be  verified by a second assay, and (c) there is no unusual stress situation that may alter micronutrient distribution. In assessing nutritional status, it is important for the  clinician to be aware of those population groups at risk, the most reliable biochemical assays for monitoring nutritional status, and the symptoms of deficiencies if they  should occur.

 

  

Page 1109

28.3—  Recommended Dietary Allowances Recommended Dietary Allowances are the levels of intake of essential nutrients considered by the Food and Nutrition Board of the National Research Council to  be adequate to meet the nutritional needs of practically all healthy persons. Optimally, the RDAs are based on daily intake sufficient to prevent the appearance of  nutritional deficiency in at least 95% of the population. This determination is relatively easy to make for those nutrients associated with dramatic deficiency diseases, for  example, vitamin C and scurvy. In other instances more indirect measures must be used, such as tissue saturation or extrapolation from animal studies. In some cases,  such as vitamin E, in which no deficiency symptoms are known to occur in the general population, the RDA is defined as the normal level of intake in the American  diet. There is no set of criteria that can be used for all micronutrients, and there are always some uncertainty and debate as to the correct criteria. The criteria are  constantly changed by new research. The Food and Nutrition Board normally meets every 6 years to consider currently available information and update its  recommendations. RDAs serve as a useful general guide in evaluating adequacy of diets. However, the RDAs have several limitations that should be kept in mind. Important limitations  are as follows: 1. RDAs represent an ideal average intake for groups of people and are best used for evaluating nutritional status of population groups. RDAs are not meant to be  standards or requirements for individuals. Some individuals would have no problem with intakes below the RDA, whereas a few may develop deficiencies on intakes  above the RDA. 2. RDAs are designed to meet the needs of healthy people and do not take into account special needs arising from infections, metabolic disorders, or chronic diseases. 3. Since present knowledge of nutritional needs is incomplete, there may be unrecognized nutritional needs. To provide for these needs, the RDAs should be met from  as varied a selection of foods as possible. No single food can be considered complete, even if it meets the RDA for all known nutrients. This is important, especially in  light of the current practice of fortifying foods of otherwise low nutritional value. 4. As currently formulated, RDAs do not define the "optimal" level of any nutrient, since optimal levels are difficult to define. Because of information suggesting that  optimal intake of certain micronutrients (e.g., vitamins A, C, and E) may reduce heart disease and cancer risk, some experts feel that the focus of the RDAs should  shift from preventing nutritional deficiencies to defining optimal levels that may reduce the risk of other diseases. 28.4—  Fat­Soluble Vitamins Vitamin A Is Derived from Plant Carotenoids The active forms of vitamin A are retinol, retinal (retinaldehyde), and retinoic acid. These substances are synthesized by plants as the more complex carotenoids  (Figure 28.1), which are cleaved to retinol by most animals and stored in the liver as retinol palmitate. Liver, egg yolk, butter, and whole milk are good sources of the  preformed retinol. Dark green and yellow vegetables are generally good sources of the carotenoids. Conversion of carotenoids to retinol is rarely 100%, so that the  vitamin A potency of various foods is expressed in terms of retinol equivalents (1 RE is equal to 1 mg retinol, 6 mg b ­carotene,

 

  

Page 1110

Figure 28.1  Structures of vitamin A and related compounds.

and 12 mg of other carotenoids). b ­Carotene and other carotenoids are major sources of vitamin A in the American diet. These carotenoids are first cleaved to retinol  and converted to other vitamin A metabolites in the body (Figure 28.1).

Figure 28.2  Vitamin A metabolism and function.

Vitamin A serves a number of functions in the body. Only in recent years has its biochemistry become well understood (Figure 28.2). b ­Carotene and some other  carotenoids have recently been shown to have an important role as antioxidants. At the low oxygen tensions prevalent in the body, b ­carotene is a very effective  antioxidant and may be expected to reduce the risk of those cancers initiated by free radicals and other strong oxidants. Several retrospective clinical studies have  suggested that adequate dietary b ­carotene may be important in reducing the risk of lung cancer, especially in people who smoke. However, supplemental b ­carotene  did not provide any detectable benefit and may have actually increased cancer risk in two recent multicenter prospective studies. Retinol is converted to retinyl phosphate in the body. The retinyl phosphate appears to serve as a glycosyl donor in the synthesis of some glycoproteins and  mucopolysaccharides in much the same manner as dolichol phosphate (see p. 738). Retinyl phosphate is essential for the synthesis of certain glycoproteins needed for  normal growth regulation and for mucus secretion. Both retinol and retinoic acid bind to specific intracellular receptors, which then bind to chromatin and affect the  synthesis of proteins involved in the regulation of cell growth and differentiation. Thus both retinol and retinoic acid can be considered to act like steroid hormones in  regulating growth and differentiation. Finally, in the D 11­cis­retinal form, vitamin A becomes reversibly associated with the visual proteins. When light strikes the retina, a number of complex

 

  

Page 1111

biochemical changes take place, resulting in the generation of a nerve impulse, conversion of the retinal to the all­trans form, and its dissociation from the visual protein  (see p. 943). Regeneration of more visual pigments requires isomerization back to the  11­cis form (Figure 28.3).

Figure 28.3  Role of vitamin A in vision.

Based on what is known about the biochemical mechanisms of vitamin A action, its biological effects are easier to understand. For example, vitamin A is required for  the maintenance of healthy epithelial tissue. Retinol and/or retinoic acid are required to prevent the synthesis of high molecular weight forms of keratin and retinyl  phosphate is required for the synthesis of glycoproteins (an important component of the mucus secreted by many epithelial tissues). The lack of mucus secretion leads  to a drying of these cells, and the excess keratin synthesis leaves a horny keratinized surface in place of the normal moist and pliable epithelium. Vitamin A deficiency  can lead to anemia caused by impaired mobilization of iron from the liver because retinol and/or retinoic acid are required for the synthesis of the iron transport protein  transferrin. Finally, vitamin A­deficient animals are more susceptible to both infections and cancer. Decreased resistance to infections is thought to be due to keratinization of  mucosal cells lining the respiratory, gastrointestinal, and genitourinary tracts. Under these conditions fissures readily develop in the mucosal membranes, allowing  microorganisms to enter. Vitamin A deficiency may impair the immune system as well. The protective effect of vitamin A against many forms of cancer probably results  from the antioxidant potential of b ­carotene and the effects of retinol and retinoic acid in regulating cell growth. Since vitamin A is stored in the liver, deficiencies of this vitamin can develop only over prolonged periods of inadequate uptake. Mild vitamin A deficiencies are  characterized by follicular hyperkeratosis (rough keratinized skin resembling "goosebumps"), anemia (biochemically equivalent to iron deficiency anemia, but in the  presence of adequate iron intake), and increased susceptibility to infection and cancer. Night blindness is also an early symptom of vitamin A deficiency. Severe  vitamin A deficiency leads to a progressive keratinization of the cornea of the eye known as xerophthalmia in its most advanced stages. In the final stages, infection  usually sets in, with resulting hemorrhaging of the eye and permanent loss of vision. For most people (unless they happen to eat liver) the dark green and yellow vegetables are the most important dietary source of vitamin A. Unfortunately, these are the  foods most often missing from the American diet. Nationwide, dietary surveys indicate that between 40% and 60% of the population consumes less than two­thirds of  the RDA for vitamin A. Clinical symptoms of vitamin A deficiency are rare in the general population, but vitamin A deficiency is a fairly common consequence of  severe liver damage or diseases that cause fat malabsorption (see Clin. Corr. 28.1). Vitamin A accumulates in the liver and over prolonged periods large amounts of this vitamin can be toxic. Doses of 25,000–50,000 RE per day over months or years  will prove to be toxic for many children and adults. The usual symptoms include bone pain, scaly dermatitis, enlargement of liver and spleen, nausea, and diarrhea. It is,  of course, virtually impossible to ingest toxic amounts of vitamin A from normal foods unless one eats polar bear liver (6000 RE/g) regularly. Most instances of  vitamin A toxicity are due to the use of massive doses of vitamin A supplements. Fortunately, this practice is relatively rare because of increased public awareness of  vitamin A toxicity. Vitamin D Synthesis in the Body Requires Sunlight Technically, vitamin D should be considered a hormone rather than a vitamin. Cholecalciferol (D3) is produced in skin by UV irradiation of 7­dehydrocholesterol  (Figure 28.4). Thus, as long as the body is exposed to adequate sunlight,

 

  

Page 1112

CLINICAL CORRELATION 28.1  Nutritional Considerations for Cystic Fibrosis Patients with malabsorption diseases often develop malnutrition. As an example, let us  examine the nutritional consequences of one disease with malabsorption components.  Cystic fibrosis (CF) involves a generalized dysfunction of the exocrine glands that leads to  formation of a viscid mucus, which progressively plugs the ducts. Obstruction of the  bronchi and bronchioles leads to pulmonary infections, which are usually the direct cause  of death. In many cases, however, the exocrine glands of the pancreas are also affected,  leading to a deficiency of pancreatic enzymes and sometimes a partial obstruction of the  common bile duct. The deficiency (or partial deficiency) of pancreatic lipase and bile salts leads to severe  malabsorption of fat and fat­soluble vitamins. Calcium tends to form insoluble salts with  the long­chain fatty acids, which accumulate in the intestine. While these are the most  severe problems, some starches and proteins are also trapped in the fatty bolus of  partially digested foods. This physical entrapment, along with the deficiencies of  pancreatic amylase and pancreatic proteases, can lead to severe protein–calorie  malnutrition as well. Excessive mucus secretion on the luminal surfaces of the intestine  may also interfere with the absorption of several nutrients, including iron. Fortunately, microsphere preparations of pancreatic enzymes are now available that can  greatly alleviate many of these malabsorption problems. With these preparations, protein  and carbohydrate absorption rates are returned to near normal. Fat absorption is  improved greatly but not normalized, since deficiencies of bile salts and excess mucus  secretion persist. Because dietary fat is a major source of calories, these patients have  difficulty obtaining sufficient calories from a normal diet. This is complicated by increased  protein and energy needs resulting from the chronic infections often seen in these patients.  Thus many experts recommend energy intakes ranging from 120–150% of the RDA. Since inadequate energy intake results in poor growth and increased susceptibility to  infection, inadequate caloric intake is of great concern for cystic fibrosis patients. Thus the  current recommendations are for high­energy–high­protein diets without any restriction of  dietary fat (50% carbohydrate, 15% protein, and 35% fat). If caloric intake from the  normal diet is inadequate, dietary supplements or enteral feedings may be used. The  dietary supplements most often contain easily digested carbohydrates and milk protein  mixtures. Medium­chain triglycerides are sometimes used as a partial fat replacement  since they can be absorbed directly through the intestinal mucosa in the absence of bile  salts and pancreatic lipase. Since some fat malabsorption is present, deficiencies of the fat­soluble vitamins often  occur. Children aged 2–8 years need a standard adult multiple­vitamin preparation  containing 400 IU of vitamin D and 5000 IU of vitamin A per day. Older children,  adolescents, and adults need a standard multivitamin at a dose of 1–2 per day. If serum  vitamin A levels become low, water­miscible vitamin A preparations should be used. For  vitamin E the recommendations are: ages 0–6 mo, 25 IU day–1; 6–12 mo, 50 IU day–1;  1–4 years, 100 IU day–1; 4–10 years, 100–200 IU day–1; and >10 years, 200–400 IU  day–1; in water­soluble form. Vitamin K deficiency has not been adequately studied, but  the current recommendations are: ages 0–12 mo, 2.5 mg week–1 or 2.5 mg twice a week  if on antibiotics; ages >1 year, 5.0 mg twice weekly when on antibiotics or if cholestatic  liver disease is present. Iron deficiency is fairly common in cystic fibrosis patients but iron  supplementation is not usually recommended because of concern that higher iron levels in  the blood might encourage systemic bacterial infections. Calcium levels in the blood are  usually normal. However, since calcium absorption is probably suboptimal, it is important  to make certain that the diet provides at least RDA levels of calcium. Littlewood, J. M., and MacDonald, A. Rationale of modern dietary recommendations in  cystic fibrosis. J. R. Soc. Med. 80(Suppl. 15):16, 1987; and Ramsey, B. W, Farrell, P.  M., and Pencharz, P. Nutritional assessment and management in cystic fibrosis; a  consensus report. Am J. Clin. Nutr. 55:108, 1992.   there is little or no dietary requirement for vitamin D. The best dietary sources of vitamin D3 are saltwater fish (especially salmon, sardines, and herring), liver, and egg  yolk. Milk, butter, and other foods are routinely fortified with ergocalciferol (D2) prepared by irradiating ergosterol from yeast. Vitamin D potency is measured in  terms of milligrams of cholecalciferol (1 mg cholecalciferol or ergocalciferol = 40 IU). Both cholecalciferol and ergocalciferol are metabolized identically. They are carried to the liver where the 25­hydroxy derivative is formed. 25­Hydroxy­ cholecalciferol [25­(OH)D] is the major circulating derivative of vitamin D, and it is in turn converted into the biologically active 1­a ,25­dihydroxychole­calciferol  (also called calcitriol) in the proximal convoluted tubules of kidney (see Clin. Corr. 28.2). The compound 1,25­(OH)2D acts in concert with parathyroid hormone (PTH), which is also produced in response to low serum calcium. Parathyroid hormone  plays a major role in regulating the activation of vitamin D. High PTH levels stimulate the production of 1,25­(OH)2D, while low PTH levels induce formation of an  inactive 24,25­(OH)2D. Once formed, the 1,25­(OH)2D acts

 

  

Page 1113

Figure 28.4  Structures of vitamin D and related compounds.

CLINICAL CORRELATION 28.2  Renal Osteodystrophy In chronic renal failure, a complicated chain of events leads to a condition known as renal  osteodystrophy. The renal failure results in an inability to produce 1,25­(OH)2D, and thus  bone calcium becomes the only important source of serum calcium. In the later stages of  renal failure, the situation is complicated further by increased renal retention of phosphate  and resulting hyperphosphatemia. The serum phosphate levels are often high enough to  cause metastatic calcification (i.e., calcification of soft tissue), which tends to lower serum  calcium levels further (the solubility product of calcium phosphate in the serum is very low  and a high serum level of one component necessarily causes a decreased concentration of  the other). The hyperphosphatemia and hypocalcemia stimulate parathyroid hormone  secretion, and the resulting hyperparathyroidism further accelerates the rate of bone loss.  One ends up with both bone loss and metastatic calcification. In this case, simple  administration of high doses of vitamin D or its active metabolites would not be sufficient  since the combination of hyperphosphatemia and hypercalcemia would only lead to more  extensive metastatic calcification. The readjustment of serum calcium levels by high  calcium diets and/or vitamin D supplementation must be accompanied by phosphate  reduction therapies. The most common technique is to use phosphate­binding antacids  that make phosphate unavailable for absorption. Orally administered 1,25­(OH)2D is  effective at stimulating calcium absorption in the mucosa but does not enter the peripheral  circulation in significant amounts. Thus patients with severe hyperparathyroidism may  need to be treated with intravenous 1,25­(OH)2D. Johnson, W. J. Use of vitamin D analogs in renal osteodystrophy. Semin. Nephrol. 6:31,  1986; McCarthy, J. T., and Kumar, R. Behavior of the vitamin D endocrine system in the  development of renal osteodystrophy. Semin. Nephrol. 6:21, 1986; and Delmez, J. M.,  and Siatopolsky, E. Hyperphosphatemia: its consequences and treatment in patients with  chronic renal disease. Am. J. Kidney Dis. 19:303, 1992.    

  

Page 1114

alone as a typical steroid hormone in intestinal mucosal cells, where it induces synthesis of a protein, calbinden, required for calcium transport. In the bone 1,25­(OH) D and PTH act synergistically to promote bone resorption (demineralization) by stimulating osteoblast formation and activity. Finally, PTH and 1,25­(OH)2D inhibit  2 calcium excretion in the kidney by stimulating calcium reabsorption in the distal renal tubules. The overall response of calcium metabolism to several different  physiological situations is summarized in Figure 28.5. The response to low serum calcium levels is characterized by elevation of PTH and 1,25­(OH)2D, which act to  enhance calcium absorption and bone resorption and to inhibit calcium excretion (Figure 28.5a). High serum calcium levels block production of PTH. The low PTH  levels allow 25­(OH)D to be metabolized to 24,25­(OH)2D instead of 1,25­(OH)2D. In the absence of PTH and 1,25­(OH)2D bone resorption is inhibited and  calcium excretion is enhanced. High levels of serum calcium and phosphate increase the rate of bone mineralization (Figure 28.5b). Thus bone is a very important  reservoir of the calcium and phosphate needed to maintain homeostasis of serum levels. When vitamin D and dietary calcium are adequate, no net loss of bone calcium  occurs. However, when dietary calcium is low, PTH and 1,25­(OH)2D will cause net demineralization of bone to maintain normal serum calcium levels. Vitamin D  deficiency also causes net demineralization of bone due to elevation of PTH (Figure 28.5c). The most common symptoms of vitamin D deficiency are rickets in young children and osteomalacia in adults. Rickets is characterized by continued formation of  osteoid matrix and cartilage, which are improperly mineralized, resulting in soft, pliable bones. In the adult demineralization of preexisting bone takes place, causing the  bone to become softer and more susceptible to fracture. This osteomalacia is easily distinguishable from the more common osteoporosis, by the fact that the osteoid  matrix remains intact in the former, but not in the latter. Vitamin D may be involved in more than regulation of calcium homeostasis. Receptors for 1,25­(OH)2D have  been found in many tissues including parathyroid gland, islet cells of pancreas, keratinocytes of skin, and myeloid stem cells in bone marrow. The role of vitamin D in  these tissues is the subject of active investigation. Because of fortification of dairy products with vitamin D, dietary deficiencies are very rare. The cases of dietary vitamin D deficiency that do occur are most often seen  in low­income groups, the elderly (who often also have minimal exposure to sunlight), strict vegetarians (especially if their diet is also low in calcium and high in fiber),  and chronic alcoholics. Most cases of vitamin D deficiency, however, are a result of diseases causing fat malabsorption or severe liver and kidney disease (see Clin.  Corr. 28.1 and 28.2). Certain drugs also interfere with vitamin D metabolism. For example, corticosteroids stimulate the conversion of vitamin D to inactive  metabolites and have been shown to cause bone demineralization when used for long periods of time. Vitamin D can also be toxic in doses 10–100 times the RDA. The mechanism of vitamin D toxicity is summarized in Figure 28.5d. Enhanced calcium absorption and  bone resorption cause hypercalcemia, which can lead to metastatic calcification. The enhanced bone resorption also causes bone demineralization similar to that seen in  vitamin D deficiency. Finally, the high serum calcium leads directly to hypercalciuria, which predisposes the patient to formation of renal stones. Vitamin E Is a Mixture of Tocopherols For many years vitamin E was described as the ''vitamin in search of a disease." While vitamin E deficiency diseases are still virtually unknown, its metabolic role in  the body has become better understood in recent years. Vitamin E occurs in the diet as a mixture of several closely related compounds, called tocopherols.

 

  

Page 1115

Figure 28.5  Vitamin D and calcium homeostasis.  Dominant pathways of calcium metabolism under each set of metabolic  conditions are shown with heavy arrows. The effect of various hormones  on these pathways is shown by red arrows for stimulation or blue arrows  for repression. PTH, parathyroid hormone; D, cholecalciferol; 25­(OH)D,  25­hydroxycholecalciferol; and 1,25­(OH) D,  2

1­a,25­dihy­droxycholecalciferol.

 

  

Page 1116

a ­Tocopherol is the most potent and is used as the measure of vitamin E potency (1 a ­tocopherol equiv = 1 mg a ­tocopherol). First and foremost, vitamin E is an important naturally occurring antioxidant. Due to its lipophilic character it accumulates in circulating lipoproteins, cellular  membranes, and fat deposits, where it reacts very rapidly with molecular oxygen and free radicals. It acts as a scavenger for these compounds, protecting unsaturated  fatty acids (especially those in the membranes) from peroxidation reactions. Vitamin E appears to play a role in cellular respiration, either by stabilizing coenzyme Q or  by helping transfer electrons to coenzyme Q. It also appears to enhance heme synthesis by increasing the levels of  ­aminolevulinic acid (ALA) synthetase and ALA  dehydratase. Most of these vitamin E effects are thought to be an indirect effect of its antioxidant potential, rather than its actual participation as a coenzyme in any  biochemical reactions. For example, an important role of vitamin E in humans is to prevent oxidation of LDL, since it appears to be the oxidized form of LDL that is  atherogenic. Finally, neurological symptoms have been reported following prolonged vitamin E deficiency associated with malabsorption diseases. Studies on the recommended levels of vitamin E in the diet have been hampered by the difficulty of producing severe vitamin E deficiency in humans. In general, it has  been assumed that the vitamin E levels in the American diet are sufficient, since no major vitamin E deficiency diseases have been found. However, vitamin E  requirements increase as intake of polyunsaturated fatty acids (PUFAs) increases. While the recent emphasis on high PUFA diets to reduce serum cholesterol may be  of benefit in controlling heart disease, the propensity of PUFA to form free radicals on exposure to oxygen may lead to an increased cancer risk. Thus it appears only  prudent to increase vitamin E intake in high PUFA diets. Premature infants fed on formulas low in vitamin E sometimes develop a form of hemolytic anemia that can be corrected by vitamin E supplementation. Adults suffering  from fat malabsorption show a decreased red blood cell survival time. Hence vitamin E supplementation may be necessary with premature infants and in cases of fat  malabsorption. In addition, recent studies have suggested that supplementation with at least 100 mg day–1 of vitamin E may decrease the risk of heart disease. This is  well above the current RDA and is far greater than can be obtained from even a very well balanced diet. These findings have rekindled the debate as to whether  dietary recommendations should consider optimal levels of nutrients rather than the levels needed to prevent deficiency diseases. As a fat­soluble vitamin, E has the  potential for toxicity. However, it does appear to be the least toxic of the fat­soluble vitamins. No instances of toxicity have been reported at doses of 1600 mg day–1  or less. Vitamin K Is a Quinone Derivative Vitamin K is found naturally as K1 (phytylmenaquinone) in green vegetables and K2 (multiprenylmenaquinone), which is synthesized by intestinal bacteria. The body  converts synthetically prepared menaquinone (menadione) and a number of water­soluble analogs to a biologically active form of vitamin K. Dietary requirements are  measured in terms of micrograms of vitamin K1 with the RDA for adults being in the range of 60–80  g day–1.

Figure 28.6  Function  of  vitamin K.

Vitamin K1 is required for the conversion of several clotting factors and prothrombin to the active state. The mechanism of this action has been most clearly  delineated for prothrombin (see p. 970). Prothrombin is synthesized as an inactive precursor called preprothrombin. Conversion to the active form requires a vitamin  K­dependent carboxylation of specific glutamic acid residues to g ­carboxyglutamic acid (Figure 28.6). The g­carboxyglutamic acid residues are good chelators  and allow prothrombin to bind calcium. The prothrombin­Ca2+ complex in turn binds to the phospholipid membrane, where

 

  

Page 1117

proteolytic conversion to thrombin can occur in vivo. The mechanism of the carboxylation reaction has not been fully clarified but appears to involve the intermediate  formation of a 2,3­epoxide derivative of vitamin K. Dicumarol, a naturally occurring anticoagulant, inhibits the reductase, which converts the epoxide back to the  active vitamin. Recently, vitamin K has been shown to be essential for the synthesis of g­carboxyglutamic acid residues in the protein osteocalcin, which accounts for 15–20% of the  noncollagen protein in the bone of most vertebrates. As with prothrombin, the g­carboxyglutamic acid residues are responsible for most of the calcium­binding  properties of osteocalcin. Because osteocalcin synthesis is controlled by vitamin D and osteocalcin is thought to play an important role in bone remodeling, vitamin K  may be important for bone formation. The only readily detectable symptom of vitamin K deficiency in humans is increased coagulation time, but some studies have suggested that vitamin K deficiency may  be a factor in osteoporosis as well. Since vitamin K is synthesized by bacteria in the intestine, deficiencies have long been assumed to be rare. However, recent  studies have suggested that intestinally synthesized vitamin K may not be efficiently absorbed and marginal vitamin K deficiencies may be more common than originally  thought. The most common deficiency occurs in newborn infants (see Clin. Corr. 28.3), especially those whose mothers have been on anticonvulsant therapy (see Clin.  Corr. 28.4). Vitamin K deficiency also occurs in patients with obstructive jaundice and other diseases leading to severe fat malabsorption (see Clin. Corr. 28.1)  and patients on long­term antibiotic therapy (which may destroy vitamin K­synthesizing organisms in the intestine). Finally, vitamin K deficiency is sometimes seen in  the elderly, CLINICAL CORRELATION 28.3  Nutritional Considerations in the Newborn Newborn infants are at special nutritional risk. In the first place, this is a period of very  rapid growth, and needs for many nutrients are high. Some micronutrients (such as  vitamins E and K) do not cross the placental membrane well and tissue stores are low in  the newborn infant. The gastrointestinal tract may not be fully developed, leading to  malabsorption problems (particularly with respect to the fat­soluble vitamins). The  gastrointestinal tract is also sterile at birth and the intestinal flora that normally provide  significant amounts of certain vitamins (especially vitamin K) take several days to become  established. If the infant is born prematurely, the nutritional risk is slightly greater, since the  gastrointestinal tract will be less well developed and the tissue stores will be less. The most serious nutritional complications of newborns appear to be hemorrhagic  disease. Newborn infants, especially premature infants, have low tissue stores of vitamin  K and lack the intestinal flora necessary to synthesize the vitamin. Breast milk is also a  relatively poor source of vitamin K. Approximately 1 out of 400 live births shows some  signs of hemorrhagic disease. One milligram of the vitamin at birth is usually sufficient to  prevent hemorrhagic disease. Iron is another potential problem. Most newborn infants are born with sufficient reserves  of iron to last 3–4 months (although premature infants are born with smaller reserves).  Since iron is present in low amounts in both cow's milk and breast milk, iron  supplementation is usually begun at a relatively early age by the introduction of iron­ fortified cereal. Vitamin D levels are also somewhat low in breast milk and  supplementation with vitamin D is usually recommended. However, some recent studies  have suggested that iron in breast milk is present in a form that is particularly well utilized  by the infant and that earlier studies probably underestimated the amount of vitamin D  available in breast milk. Other vitamins and minerals appear to be present in adequate  amounts in breast milk as long as the mother is getting a good diet. Recent studies have  suggested that in situations in which infants must be maintained on assisted ventilation with  high oxygen concentrations, supplemental vitamin E may reduce the risk of  bronchopulmonary dysplasia and retrolental fibroplasia, two possible side effects of  oxygen therapy. Studies have also suggested that anemia of prematurity may respond to  supplemental folate and vitamin B12. In summary, most infants are provided with supplemental vitamin K at birth to prevent  hemorrhagic disease. Breast­fed infants are usually provided with supplemental vitamin D,  with iron being introduced along with solid foods. Bottle­fed infants are provided with  supplemental iron. If infants must be maintained on oxygen, supplemental vitamin E may  be beneficial. Barness, L. A. Pediatrics. In: H. Schneider, C. E. Anderson, and D. B. Coursin (Eds.),  Nutritional Support of Medical Practice, 2nd ed. New York: Harper & Row, 1983,  pp. 541–561; Huysman, M. W., and Sauer, P. J. The vitamin K controversy. Curr.  Opin. Pediatr. 6:129, 1994; Worthington­White, D. A., Behnke, M., and Gross, S.  Premature infants require additional folate and vitamin B12 to reduce the severity of anemia  of prematurity. Am. J. Clin. Nutr. 60:930, 1994; and Mueller, D. P. R. Vitamin E  therapy in retinopathy of prematurity. Eye 6:221, 1992.    

  

Page 1118

CLINICAL CORRELATION 28.4  Anticonvulsant Drugs and Vitamin Requirements Anticonvulsant drugs such as phenobarbital or diphenylhydantoin (DPH) present an  excellent example of the type of drug–nutrient interactions that are of concern to the  physician. Metabolic bone disease appears to be the most significant side effect of  prolonged anticonvulsant therapy. Whereas children and adults on these drugs seldom  develop rickets or severe osteomalacia, as many as 65% of those on long­term therapy  will have abnormally low serum calcium and phosphorus and abnormally high serum  alkaline phosphatase. Some bone loss is usually observed in these cases. While the cause  of the hypocalcemia and bone loss is thought to be an effect of the anticonvulsant drugs  on vitamin D metabolism, not all of the studies have shown decreased levels of 25­(OH) D and 1,25­(OH)2D in patients on these drugs. However, supplemental vitamin D in the  range of 2000–10,000 units per day appears to correct both the hypocalcemia and  osteopenia. Anticonvulsants also tend to increase needs for vitamin K, leading to an  increased incidence of hemorrhagic disease in infants born to mothers on anticonvulsants.  In addition, anticonvulsants appear to increase the need for folic acid and B6. Low serum  folate levels are seen in 75% of patients on anticonvulsants and megaloblastic anemia may  occur in as many as 50% without supplementation. By bio­chemical parameters, 30–60%  of the children on anticonvulsants exhibit some form of B6 deficiency. Clinical symptoms  of B6 deficiency are rarely seen, however. From 1 to 5 mg of folic acid and 10 mg of  vitamin B6 appear to be sufficient for most patients on anticonvulsants. Since folates may  speed up the metabolism of some anticonvulsants, it is important that excess folic acid not  be given. Moslet, U., and Hansen, E. S. A review of vitamin K, epilepsy and pregnancy. Acta  Neurol. Scand. 85:39, 1992: Rivery, M. D., and Schottelius, D. D. Phenytoin­folic acid:  a review. Drug Intelligence Clin. Pharm. 18:292, 1984; and Tjellesen, L. Metabolism  and action of vitamin D in epileptic patients on anticonvulsant treatment and healthy  adults. Dan. Med. Bull. 41:139, 1994.   who are prone to poor liver function (reducing preprothrombin synthesis) and fat malabsorption. Clearly, vitamin K deficiency should be suspected in patients  demonstrating easy bruising and prolonged clotting time. 28.5—  Water­Soluble Vitamins Water­soluble vitamins differ from fat­soluble vitamins in several important aspects. Most are readily excreted once their concentration surpasses the renal threshold.  Thus toxicities are rare. Deficiencies of these vitamins occur relatively quickly on an inadequate diet. Their metabolic stores are labile and depletion can often occur in a  matter of weeks or months. Since the water­soluble vitamins are coenzymes for many common biochemical reactions, it is often possible to assay vitamin status by  measuring one or more enzyme activities in isolated red blood cells. These assays are especially useful if one measures both the endogenous enzyme activity and the  stimulated activity following addition of the active coenzyme derived from that vitamin. Most of the water­soluble vitamins are converted to coenzymes, which are utilized either in the pathways for energy generation or hematopoiesis. Deficiencies of the  energy­releasing vitamins produce a number of overlapping symptoms. In many cases the vitamins participate in so many biochemical reactions that it is impossible to  pinpoint the exact biochemical cause of any given symptom. However, it is possible to generalize that because of the central role these vitamins play in energy  metabolism, deficiencies show up first in rapidly growing tissues. Typical symptoms include dermatitis, glossitis (swelling and reddening of the tongue), cheilitis at  the corners of the lips, and diarrhea. In many cases nervous tissue is also involved due to its high energy demand or specific effects of the vitamin. Some of the  common neurological symptoms include peripheral neuropathy (tingling of nerves at the extremities), depression, mental confusion, lack of motor coordination, and  malaise. In some cases demyelination and degeneration of nervous tissues also occur. These deficiency symptoms are so common and overlapping that they can be

 

  

Page 1119

considered as properties of the energy­releasing vitamins as a class, rather than being specific for any one.

Figure 28.7  Structure of thiamine.

28.6—  Energy­Releasing Water­Soluble Vitamins Thiamine (Vitamin B1) Forms the Coenzyme Thiamine Pyrophosphate (TPP) Thiamine (Figure 28.7) is rapidly converted to the coenzyme thiamine pyrophosphate (TPP), which is required for the key reactions catalyzed by pyruvate  dehydrogenase complex and a ­ketoglutarate dehydrogenase complex (Figure 28.8). Cellular energy generation is severely compromised in thiamine deficiency. TPP is  also required for the transketolase reactions of the pentose phosphate pathway. While the pentose phosphate pathway is not quantitatively important in terms of energy  generation, it is the sole source of ribose for the synthesis of nucleic acid precursors and the major source of NADPH for fatty acid biosynthesis and other biosynthetic  pathways. Red blood cell transketolase is also the enzyme most commonly used for measuring thiamine status in the body. TPP appears to function in transmission of  nerve impulses. TPP (or a related metabolite, thiamine triphosphate) is localized in peripheral nerve membranes. It appears to be required for acetylcholine synthesis  and may also be required for ion translocation reactions in stimulated neural tissue. Although the biochemical reactions involving TPP are fairly well characterized, it is not clear how these biochemical lesions result in the symptoms of thiamine  deficiency. The pyruvate dehydrogenase and transketolase reactions are the most sensitive to thiamine levels. Thiamine deficiency appears to selectively inhibit  carbohydrate metabolism, causing an accumulation of pyruvate. Cells may be directly affected by lack of available energy and NAPDH or

Figure 28.8  Summary of important reactions involving thiamine  pyrophosphate.  The reactions involving thiamine pyrophosphate are  indicated in red.

 

  

Page 1120

may be poisoned by the accumulated pyruvate. Other symptoms of thiamine deficiency involve the neural tissue and probably result from the direct role of TTP in  nerve transmission. Loss of appetite, constipation, and nausea are among the earliest symptoms of thiamine deficiency. Mental depression, peripheral neuropathy, irritability, and fatigue  are other early symptoms and probably directly relate to the role of thiamine in maintaining healthy nervous tissue. These symptoms of thiamine deficiency are most  often seen in the elderly and low­income groups on restricted diets. Symptoms of moderately severe thiamine deficiency include mental confusion, ataxia (unsteady  gait while walking and general inability to achieve fine control of motor functions), and ophthalmoplegia (loss of eye coordination). This set of symptoms is usually  referred to as Wernicke–Korsakoff syndrome and is most commonly seen in chronic alcoholics (see Clin. Corr. 28.5). Severe thiamine deficiency is known as  beriberi. Dry beriberi is characterized primarily by advanced neuromuscular symptoms, including atrophy and weakness of the muscles. When these symptoms are  coupled with edema, the disease is referred to as wet beriberi. Both forms of beriberi can be associated with an unusual type of heart failure characterized by high  cardiac output. Beriberi is found primarily in populations relying exclusively on polished rice for food, although cardiac failure is sometimes seen in alcoholics as well. The thiamine requirement is proportional to caloric content of the diet and is in the range of 1.0–1.5 mg per day for the normal adult. This requirement should be raised  somewhat if carbohydrate intake is excessive or if the metabolic rate is elevated (due to fever, trauma, pregnancy, or lactation). Coffee and tea CLINICAL CORRELATION 28.5  Nutritional Considerations in the Alcoholic Chronic alcoholics run considerable risk of nutritional deficiencies. The most common  problems are neurologic symptoms associated with thiamine or pyridoxine deficiencies  and hematological problems associated with folate or pyridoxine deficiencies. The  deficiencies seen with alcoholics are not necessarily due to poor diet alone, although it is  often a strong contributing factor. Alcohol causes pathological alterations of the  gastrointestinal tract that often directly interfere with absorption of certain nutrients. The  liver is the most important site of activation and storage of many vitamins. The severe liver  damage associated with chronic alcoholism appears to interfere directly with storage and  activation of certain nutrients. Up to 40% of hospitalized alcoholics are estimated to have megaloblastic erythropoiesis  due to folate deficiency. Alcohol appears to interfere directly with folate absorption and  alcoholic cirrhosis impairs storage of this nutrient. Another 30% of hospitalized alcoholics  have sideroblastic anemia or identifiable sideroblasts in erythroid marrow cells  characteristic of pyridoxine deficiency. Some alcoholics also develop a peripheral  neuropathy that responds to pyridoxine supplementation. This problem appears to result  from impaired activation and increased degradation of pyridoxine. In particular,  acetaldehyde (an end product of alcohol metabolism) displaces pyridoxal phosphate from  its carrier protein in the plasma. The free pyridoxal phosphate is then rapidly degraded to  inactive compounds and excreted. The most dramatic nutritionally related neurological disorder is Wernicke–Korsakoff  syndrome. The symptoms include mental disturbances, ataxia (unsteady gait and lack of  fine motor coordination), and uncoordinated eye movements. Congestive heart failure  similar to that seen with beriberi is also seen in a small number of these patients. While this  syndrome may only account for 1–3% of alcohol­related neurologic disorders, the  response to supplemental thiamine is so dramatic that it is usually worth consideration.  The thiamine deficiency appears to arise primarily from impaired absorption, although  alcoholic cirrhosis may also affect the storage of thiamine in the liver. While those are the most common nutritional deficiencies associated with alcoholism,  deficiencies of almost any of the water­soluble vitamins can occur and cases of alcoholic  scurvy and pellagra are occasionally reported. Chronic ethanol consumption causes an  interesting redistribution of vitamin A stores in the body. Vitamin A stores in the liver are  rapidly depleted while levels of vitamin A in the serum and other tissues may be normal or  slightly elevated. Apparently, ethanol causes both increased mobilization of vitamin A  from the liver and increased catabolism of liver vitamin A to inactive metabolites by the  hepatic P450 system. Alcoholic patients have decreased bone density and an increased  incidence of osteoporosis. This probably relates to a defect in the 25­hydroxylation step  in the liver as well as an increased rate of metabolism of vitamin D to inactive products by  an activated cytochrome P450 system. Dietary calcium intake is also often poor. In fact,  alcoholics generally have decreased serum levels of zinc, calcium, and magnesium due to  poor dietary intake and increased urinary losses. Iron­deficiency anemia is very rare  unless there is gastrointestinal bleeding or chronic infection. In fact, excess iron is a more  common problem with alcoholics. Many alcoholic beverages contain relatively high iron  levels, and alcohol appears to enhance iron absorption. Hayumpa, A. M. Mechanisms of vitamin deficiencies in alcoholism. Alcohol. Clin. Exp.  Res. 10:573, 1986; and Lieber, C. S. Alcohol, liver and nutrition. J. Am. Coll Nutr.  10:602, 1991.    

  

Page 1121

contain substances that destroy thiamine, but this is not a problem for individuals consuming normal amounts of these beverages. Routine enrichment of cereals has  assured that most Americans have an adequate intake of thiamine on a normal mixed diet. Riboflavin Is Part of FAD and FMN Riboflavin is the precursor of the coenzymes flavin adenine dinucleotide (FAD) and flavin mononucleotide (FMN), both of which are involved in a wide variety of  redox reactions. The flavin coenzymes are essential for energy production and cellular respiration. The most characteristic symptoms of riboflavin deficiency are  angular cheilitis, glossitis, and scaly dermatitis (especially around the nasolabial folds and scrotal areas). The best flavin­requiring enzyme for assaying riboflavin status  appears to be erythrocyte glutathione reductase. The recommended riboflavin intake is 1.2–1.7 mg day–1 for the normal adult. Foods rich in riboflavin include milk,  meat, eggs, and cereal products. Riboflavin deficiencies are quite rare in this country. When riboflavin deficiency does occur, it is usually seen in chronic alcoholics.  Hypothyroidism has recently been shown to slow the conversion of riboflavin to FMN and FAD. It is not known whether this affects riboflavin requirements, however. Niacin Is Part of NAD and NADP Niacin is not a vitamin in the strictest sense of the word, since some niacin can be synthesized from tryptophan. However, conversion of tryptophan to niacin is  relatively inefficient (60 mg of tryptophan is required for the production of 1 mg of niacin) and occurs only after all of the body requirements for tryptophan (protein  synthesis and energy production) have been met. Since synthesis of niacin requires thiamine, pyridoxine, and riboflavin, it is also very inefficient on a marginal diet. Thus  most people require dietary sources of both tryptophan and niacin. Niacin (nicotinic acid) and niacinamide (nicotinamide) are both converted to the ubiquitous  oxidation–reduction coenzymes NAD+ and NADP+ in the body. Borderline niacin deficiencies are first seen as a glossitis (redness) of the tongue, somewhat similar to riboflavin deficiency. Pronounced deficiencies lead to pellagra,  which is characterized by the three Ds: dermatitis, diarrhea, and dementia. The dermatitis is characteristic in that it is usually seen only in skin areas exposed to sunlight  and is symmetric. The neurologic symptoms are associated with actual degeneration of nervous tissue. Because of food fortification, pellagra is a medical curiosity in  the developed world. Today it is primarily seen in alcoholics, patients with severe malabsorption problems, and elderly on very restricted diets. Pregnancy,  lactation, and chronic illness lead to increased needs for niacin, but a varied diet will usually provide sufficient amounts. Since tryptophan can be converted to niacin, and niacin can exist in a free or bound form, the calculation of available niacin for any given food is not a simple matter.  For this reason, niacin requirements are expressed in terms of niacin equivalents (1 niacin equiv = 1 mg free niacin). The current recommendation of the Food and  Nutrition Board for a normal adult is 13–19 niacin equivalents (NE) per day. The richest food sources of niacin are meats, peanuts and other legumes, and enriched  cereals. Pyridoxine (Vitamin B6) Forms the Coenzyme Pyridoxal Phosphate

Figure 28.9  Structures of vitamin B6.

Pyridoxine, pyridoxamine, and pyridoxal are all naturally occurring forms of vitamin B6 (Figure 28.9). All three forms are efficiently converted by the body to  pyridoxal phosphate, which is required for the synthesis, catabolism, and interconversion of amino acids. The role of pyridoxal phosphate in amino

 

  

Page 1122

acid metabolism has been discussed previously (see p. 449). While pyridoxal phosphate­dependent reactions are legion, there are a few instances in which the  biochemical lesion seems to be directly associated with the symptoms of B6 deficiency (Figure 28.10). Pyridoxal phosphate is essential for energy production from  amino acids and can be considered an energy­releasing vitamin. Thus some of the symptoms of severe B6 deficiency are similar to those of the other energy­releasing  vitamins. Pyridoxal phosphate is also required for the synthesis of the neurotransmitters serotonin and norepinephrine and for synthesis of the sphingolipids  necessary for myelin formation. These effects are thought to explain the irritability, nervousness, and depression seen with mild deficiencies and the peripheral  neuropathy and convulsions observed with severe deficiencies. Pyridoxal phosphate is required for the synthesis of  ­aminolevulinic acid, a precursor of heme. B6  deficiencies occasionally cause sideroblastic anemia, which is characteristically a microcytic anemia seen in the presence of high serum iron. Pyridoxal phosphate is  also an essential component of glycogen phosphorylase; it is covalently linked to a lysine residue and stabilizes the enzyme. This role of B6 may explain the decreased  glucose tolerance associated with deficiency, although B6 appears to have some direct effects on the glucocorticoid receptor as well. Vitamin B6 is also required for the  conversion of homocysteine to cysteine, and hyperhomocysteinemia appears to be a risk factor for cardiovascular disease. Finally, pyridoxal phosphate is one of the  cofactors required for the conversion of tryptophan to NAD. While this may not be directly related to the symptomatology of B6 deficiency, a tryptophan load test is a  sensitive indicator of vitamin B6 status (see Clin. Corr. 28.6, p. 1124).

Figure 28.10  Important metabolic roles ofpyridoxal phosphate.  Reactions requiring pyridoxal phosphate are  indicated with red arrows. ALA,  ­aminolevulinic  acid; aKG, a­ketoglutarate; GPT,  glutamate pyruvate aminotransferase; and GOT,  glutamate oxaloacetate aminotransferase.

The requirement for B6 in the diet is roughly proportional to the protein content of the diet. Assuming that the average American consumes close to 100 g of protein per  day, the RDA for vitamin B6 has been set at 1.4–2.0 mg day–1 for a normal adult. This requirement is increased during pregnancy and lactation and may increase  somewhat with age as well. Vitamin B6 is fairly widespread in foods, but meat, vegetables, whole­grain cereals, and egg yolks are among the richest sources. Evaluation of B6 nutritional status has become a controversial topic in recent years. Some of this controversy is discussed in Clin. Corr. 28.6. It has usually been  assumed that the average American diet is adequate in B6 and it is not routinely added to flour and other fortified foods. However, recent nutritional surveys have cast  doubt on this assumption. A significant fraction of the survey population was found to consume less than two­thirds of the RDA for B6. Pantothenic Acid and Biotin Are Also Energy­Releasing Vitamins Pantothenic acid is a component of coenzyme A (CoA) and the phosphopantetheine moiety of fatty acid synthase and thus is required for the metabolism of all fat,  protein, and carbohydrate via the citric acid cycle. More than 70 enzymes have been described to date that utilize CoA or its derivatives. In view of the importance of  these reactions, one would expect pantothenic acid deficiencies to be a serious concern in humans. This does not appear to be the case for two reasons: (1)  pantothenic acid is very widespread in natural foods, probably reflecting its widespread metabolic role, and (2) most symptoms of pantothenic acid deficiency are  vague and mimic those of other B vitamin deficiencies. Biotin is the prosthetic group for a number of carboxylation reactions, the most notable being pyruvate carboxylase (needed for synthesis of oxaloacetate for  gluconeogenesis and replenishment of the citric acid cycle), acetyl­CoA carboxylase (fatty acid biosynthesis), and propionyl­CoA carboxylase (methione, leucine, and  valine metabolism). Biotin is found in peanuts, chocolate, and eggs and is synthesized by intestinal bacteria.

 

  

Page 1123

Figure 28.11  Structure of folic acid and N5­methyltetrahydrofolate.

28.7—  Hematopoietic Water­Soluble Vitamins Folic Acid (Folacin) Functions As Tetrahydrofolate in One­Carbon Metabolism The simplest form of folic acid is pteroylmonoglutamic acid. However, folic acid usually occurs as polyglutamate derivatives with from 2 to 7 glutamic acid residues  (Figure 28.11). These compounds are taken up by intestinal mucosal cells and the extra glutamate residues are removed by conjugase, a lysosomal enzyme. The free  folic acid is then reduced to tetrahydrofolate by the enzyme dihydrofolate reductase and circulated in the plasma primarily as the free N5­methyl derivative of  tetrahydrofolate (Figure 28.11). Inside cells, tetrahydrofolates are found primarily as polyglutamate derivatives, and these appear to be the biologically most potent  forms. Folic acid is also stored as a polyglutamate derivative of tetrahydrofolate in the liver. Various one­carbon tetrahydrofolate derivatives are used in biosynthetic reactions (Figure 28.12). They are required, for example, in the synthesis of choline, serine,  glycine, purines, and dTMP. Since adequate amounts of choline and the amino acids can usually be obtained from the diet, the participation of folates in purine and  dTMP synthesis appears to be metabolically the most

Figure 28.12  Metabolic roles of folic acid and vitamin B12 in one­carbon metabolism.  The metabolic interconversions of folic acid and its derivatives are indicated  with black arrows. Pathways relying exclusively on folate are  shown with red arrows. The important B12­dependent reaction converting  N5­methyl H4folate back to H4folate is shown with a blue arrow. The box encloses  the "pool" of C1 derivatives of H4folate.

 

  

Page 1124

CLINICAL CORRELATION 28.6  Vitamin B6 Requirements for Users of Oral Contraceptives The controversy over B6 requirements for users of oral contraceptives best illustrates the  potential problems associated with biochemical assays. For years, one of the most  common assays for vitamin B6 status had been the tryptophan load assay. This assay is  based on the observation that when tissue pyridoxal phosphate levels are low, the normal  catabolism of tryptophan is impaired and most of the tryptophan is catabolized by a minor  pathway leading to synthesis of xanthurenic acid. Under many conditions, the amount of  xanthurenic acid recovered in a 24­h urine sample following ingestion of a fixed amount of  tryptophan is a valid indicator of vitamin B6 status. When the tryptophan load test was  used to assess the vitamin B6 status of oral contraceptive users, however, alarming reports  started appearing in the literature. Not only did oral contraceptive use increase the  excretion of xanthurenic acid considerably but the amount of pyridoxine needed to return  xanthurenic acid excretion to normal was 10 times the RDA and almost 20 times the level  required to maintain normal B6 status in control groups. As might be expected, this  observation received much popular attention in spite of the fact that most classical  symptoms of vitamin B6 deficiency were not observed in oral contraceptive users. More recent studies using other measures of vitamin B6 have painted a slightly different  picture. For example, erythrocyte glutamate pyruvate aminotransferase and erythrocyte  glutamate oxaloacetate aminotransferase are both pyridoxal phosphate­containing  enzymes. One can also assess vitamin B6 status by measuring the endogenous activity of  these enzymes and the degree of stimulation by added pyridoxal phosphate. These types  of assays show a much smaller difference between nonusers and users of oral  contraceptives. The minimum level of pyridoxine needed to maintain normal vitamin B6  status as measured by these assays was only 2.0 mg day–1, which is slightly greater than  the RDA and about twice that needed by nonusers. Why the large discrepancy? For one thing, it must be kept in mind that enzyme activity  can be affected by hormones as well as vitamin cofactors. Kynureninase is the key  pyridoxal phosphate­containing enzyme of the tryptophan catabolic pathway. The activity  of kynureninase is regulated both by pyridoxal phosphate availability and by estrogen  metabolites. Even with normal vitamin B6 status most of the enzyme exists in the inactive  apoenzyme form. However, this does not affect tryptophan metabolism because  tryptophan oxygenase, the first enzyme of the pathway, is rate limiting. Thus the small  amount of active holoenzyme is more than sufficient to handle the metabolites produced  by the first part of the pathway. However, kynureninase is inhibited by estrogen  metabolites. Thus with oral contraceptive use its activity is reduced to a level where it  becomes rate limiting and excess tryptophan metabolites are shunted to xanthurenic acid.  Higher than normal levels of vitamin B6 overcome this problem by converting more  apoenzyme to holoenzyme, thus increasing the total amount of enzyme. Since the estrogen  was having a specific effect on the enzyme used to measure vitamin B6 status in this assay,  it did not necessarily mean that pyridoxine requirements were altered for other metabolic  processes in the body. Does this mean that vitamin B6 status is of no concern to users of oral contraceptives?  Oral contraceptives do appear to increase vitamin B6 requirements slightly. Several  dietary surveys have shown that a significant percentage of women in the 18­24­year age  group consume diets containing less than 1.3 mg of pyridoxine per day. If these women  are also using oral contraceptives, they are at some increased risk for developing a  borderline deficiency. Thus, while the tryptophan load test was clearly misleading in a  quantitative sense, it did alert the medical community to a previously unsuspected  nutritional risk. Bender, D. A. Oestrogens and vitamin B6—actions and interactions. World Rev. Nutr.  Diet. 51:140, 1987; and Kirksey, A., Keaton, K., Abernathy, R. P., and Grager, J. L.  Vitamin B6 nutritional status of a group of female adolescents. Am. J. Clin. Nutr. 31:946,  1978.   significant of those reactions. In addition, tetrahydrofolate and vitamin B12 are required, along with vitamin B6, for the conversion of homocysteine to methionine. As  mentioned earlier, this may also be significant because hyperhomocysteinemia appears to be a risk factor for cardiovascular disease. Methionine, of course, is also  converted to S­adenosylmethionine, which is used in many methylation reactions. The most pronounced effect of folate deficiency is inhibition of DNA synthesis due to decreased availability of purines and dTMP. This leads to arrest of cells in S  phase and a characteristic ''megaloblastic" change in size and shape of nuclei of rapidly dividing cells. The block in DNA synthesis slows down maturation of red blood  cells, causing production of abnormally large "macrocytic" red blood cells with fragile membranes. Thus a macrocytic anemia associated with megaloblastic changes  in the bone marrow is characteristic of folate deficiency. In addition, hyperhomocysteinemia is fairly common in the elderly population and appears to be due to  inadequate intake and/or decreased utilization of folate, vitamin B6, and vitamin B12. Elevated homocysteine levels usually respond to supplementation with RDA levels  of those vitamins.

 

  

Page 1125

There are many causes of folate deficiency, including inadequate intake, impaired absorption, increased demand, and impaired metabolism. Some dietary surveys  have suggested that inadequate intake may be more common than previously supposed. However, as with most other vitamins, inadequate intake is probably not  sufficient to trigger symptoms of folate deficiency in the absence of increased requirements or decreased utilization. Perhaps the most common example of increased  need occurs during pregnancy and lactation. As the blood volume and the number of rapidly dividing cells in the body increase, the need for folic acid increases. By  the third trimester the folic acid requirement has almost doubled. In the United States almost 20–25% of otherwise normal pregnancies are associated with low serum  folate levels, but actual megaloblastic anemia is rare and is usually seen only after multiple pregnancies. However, recent studies have shown that inadequate folate  levels during the early stages of pregnancy increase the risk for neural tube defects, a type of birth defect. Normal diets seldom supply the 400  g of folate needed  during pregnancy, so most physicians routinely recommend supplementation for women during the child­bearing years. Folate deficiency is common in alcoholics (see  Clin. Corr. 28.5). Folate deficiencies are also seen in a number of malabsorption diseases and are occasionally seen in the elderly, due to a combination of poor  dietary habits and poor absorption. There are a number of drugs that also directly interfere with folate metabolism. Anticonvulsants and oral contraceptives may interfere with folate absorption and  anticonvulsants appear to increase catabolism of folates (see Clin. Corr. 28.4). Oral contraceptives and estrogens also appear to interfere with folate metabolism in  their target tissue. Long­term use of any of these drugs can lead to folate deficiencies unless adequate supplementation is provided. For example, 20% of patients using  oral contraceptives develop megaloblastic changes in the cervicovaginal epithelium, and 20–30% show low serum folate levels. Vitamin B12 (Cobalamine) Contains Cobalt in a Tetrapyrrole Ring Pernicious anemia, a megaloblastic anemia associated with neurological deterioration, was invariably fatal until 1926 when liver extracts were shown to be curative.  Subsequent work showed the need for both an extrinsic factor present in liver and an intrinsic factor produced by the body: vitamin B12 was the extrinsic factor.  Chemically, vitamin B12 consists of cobalt in a coordination state of six—coordinated in four positions by a tetrapyrrole (or corrin) ring, in one position by a  benzimidazole nitrogen, and in the sixth position by one of several different ligands (Figure 28.13). The crystalline forms of B12 used in supplementation are usually  hydroxycobalamine or cyanocobalamine. In foods B12 usually occurs bound to protein in the methyl or 5 ­deoxyadenosyl forms. To be utilized the B12 must first be  removed from the protein by acid hydrolysis in the stomach or trypsin digestion in the intestine. It then must combine with intrinsic factor, a protein secreted by the  stomach, which carries it to the ileum for absorption.

Figure 28.13  Structure of vitamin B12 (cobalamine).

In humans there are two major symptoms of B12 deficiency (hematopoietic and neurological), and only two biochemical reactions in which B12 is known to participate  (Figure 28.14). Thus it is very tempting to speculate on exact cause and effect mechanisms. The methyl derivative of B12 is required for conversion of homocysteine to  methionine and the 5­deoxyadenosyl derivative is required for the methylmalonyl­CoA mutase reaction (methylmalonyl CoA   succinyl CoA), which is a key step in  the catabolism of some branched­chain amino acids. The neurologic disorders seen in B12 deficiency are due to progressive demyelination of nervous tissue. It has been  proposed that the

 

  

Page 1126

Figure 28.14  Metabolism of vitamin B12.  Metabolic interconversions of B12 are indicated with light arrows, and B12­requiring  reactions are indicated with red arrows. Other related pathways are  indicated with a blue arrow.

methylmalonyl CoA that accumulates interferes with myelin sheath formation in two ways. 1. Methylmalonyl CoA is a competitive inhibitor of malonyl CoA in fatty acid biosynthesis. Since the myelin sheath is continually turning over, any severe inhibition of  fatty acid biosynthesis will lead to its eventual degeneration. 2. In the residual fatty acid synthesis, methylmalonyl CoA can substitute for malonyl CoA in the reaction sequence, leading to branched­chain fatty acids, which might  disrupt normal membrane structure. There is some evidence supporting both mechanisms. Megaloblastic anemia associated with B12 deficiency is thought to reflect the effect of B12 on folate metabolism. The B12­dependent homocysteine to methionine  conversion (homocysteine + N5­methyl THF   methionine + THF) appears to be the only major pathway by which N5­methyltetrahydrofolate can return to the  tetrahydrofolate pool (Figure 28.14). Thus in B12 deficiency there is a buildup of N5­methyltetrahydrofolate and a deficiency of the tetrahydrofolate derivatives needed  for purine and dTMP biosynthesis. Essentially all of the folate becomes "trapped" as the N5­methyl derivative. Vitamin B12 also may be required for uptake of folate by  cells and for its conversion to the biologically more active polyglutamate forms. High levels of supplemental folate can overcome the megaloblastic anemia associated  with B12 deficiencies but not the neurological problems. Hence caution must be taken in using folate to treat megaloblastic anemia. Vitamin B12 is widespread in foods of animal origin, especially meats. Liver stores up to a 6­year supply of vitamin B12. Thus deficiencies of B12 are extremely rare.  They are occasionally seen in older people due to insufficient production of intrinsic factor and/or HCl in the stomach. B12 deficiency can also be seen in patients with  severe malabsorption diseases and in long­term vegetarians.

 

  

Page 1127

28.8—  Other Water­Soluble Vitamins Ascorbic Acid Functions in Reduction and Hydroxylation Reactions Vitamin C or ascorbic acid is a six­carbon compound closely related to glucose. Its main biological role is as a reducing agent in several important hydroxylation  reactions in the body. Ascorbic acid is required for the hydroxylation of lysine and proline in protocollagen. Without this hydroxylation protocollagen cannot properly  cross­link into normal collagen fibrils. Thus vitamin C is obviously important for maintenance of normal connective tissue and for wound healing, since the connective  tissue is laid down first. Vitamin C is also necessary for bone formation, since bone tissue has an organic matrix containing collagen as well as the inorganic, calcified  portion. Finally, collagen appears to be a component of the ground substance surrounding capillary walls, so vitamin C deficiency is associated with capillary  fragility. Since vitamin C is concentrated in the adrenal gland, especially in periods of stress, it may be required for hydroxylation reactions in synthesis of some corticosteroids.  Ascorbic acid has other important properties as a reducing agent, which appear to be nonenzymatic. For example, it aids in absorption of iron by reducing it to the  ferrous state in the stomach. It spares vitamin A, vitamin E, and some B vitamins by protecting them from oxidation. Also, it enhances the utilization of folic acid, either  by aiding the conversion of folate to tetrahydrofolate or the formation of polyglutamate derivatives of tetrahydrofolate. Finally, vitamin C appears to be a biologically  important antioxidant. The National Research Council has recently concluded that adequate amounts (RDA levels) of antioxidants such as b ­carotene and vitamin C in  the diet reduce the risk of cancer. The data for other naturally occurring antioxidants such as vitamin E and selenium are not yet conclusive. Most of the symptoms of vitamin C deficiency can be directly related to its metabolic roles. Symptoms of mild vitamin C deficiency include easy bruising and  formation of petechiae (small, pinpoint hemorrhages in skin) due to increased capillary fragility and decreased immunocompetence. Scurvy is associated with  decreased wound healing, osteoporosis, hemorrhaging, and anemia. Osteoporosis results from the inability to maintain the collagenous organic matrix of the bone,  followed by demineralization. Anemia results from extensive hemorrhaging coupled with defects in iron absorption and folate metabolism. Since vitamin C is readily absorbed, deficiencies almost invariably result from poor diet and/or increased need. There is uncertainty over the need for vitamin C in  periods of stress. In severe stress or trauma there is a rapid drop in serum vitamin C levels. In these situations most of the body's supply of vitamin C is mobilized to  the adrenals and/or the area of the wound. Does this represent an increased demand for vitamin C, or merely a normal redistribution to those areas where it is needed  most? Do the lowered serum levels of vitamin C impair its functions in other tissues in the body? The current consensus seems to be that the lowered serum vitamin C  levels indicate an increased demand, but there is little agreement as to how much. Smoking causes lower serum levels of vitamin C. In fact, the 1989 RDAs recommend that smokers consume 100 mg of vitamin C per day instead of the 60 mg day–1  needed by nonsmoking adults. Aspirin appears to block uptake of vitamin C by white blood cells. Oral contraceptives and corticosteroids also lower serum levels  of vitamin C. While there is no universal agreement as to the seriousness of these effects, the possibility of marginal vitamin C deficiencies should be considered with  any patient using these drugs over a long period of time, especially if dietary intake is less than optimal. The most controversial question surrounding vitamin C is its use in megadoses to prevent and cure the common cold. Ever since this use of vitamin C was first  popularized by Linus Pauling in 1970, the issue has generated

 

  

Page 1128

considerable controversy. However, some double­blind studies have suggested that while vitamin C supplementation does not appear to be useful in preventing the  common cold, it may moderate its symptoms. The mechanism by which vitamin C ameliorates the symptoms of the common cold is not known. It has been suggested  that vitamin C is required for normal leukocyte function or for synthesis and release of histamine during stress situations. While megadoses of vitamin C are probably no more harmful than the widely used over­the­counter cold medications, some potential side effects of high vitamin C  intake should be considered. For example, oxalate is a major metabolite of ascorbic acid. Thus high ascorbate intakes could theoretically lead to the formation of  oxalate kidney stones in predisposed individuals. However, most studies have shown that excess vitamin C is primarily excreted as ascorbate rather than oxalate.  Pregnant mothers taking megadoses of vitamin C may give birth to infants with abnormally high vitamin C requirements. Earlier suggestions that megadoses of vitamin  C interfered with B12 metabolism have proved to be incorrect. 28.9—  Macrominerals Calcium Has Many Physiological Roles Calcium is the most abundant mineral in the body. Most is in bone, but the small amount of calcium outside of bone functions in a number of essential processes. It is  required for many enzymes, mediates some hormonal responses, and is essential for blood coagulation. It is also essential for muscle contractility and normal  neuromuscular irritability. In fact, only a relatively narrow range of serum calcium levels is compatible with life. Since maintenance of constant serum calcium levels is so  vital, an elaborate homeostatic control system has evolved (see pp. 862 and 1112). Low serum calcium stimulates formation of 1,25­dihydroxycholecalciferol, which  enhances calcium absorption. If dietary calcium intake is insufficient to maintain serum calcium, 1,25­dihydroxychole­calciferol and parathyroid hormone stimulate bone  resorption. Long­term dietary calcium insufficiency, therefore, almost always results in net loss of calcium from the bones. Dietary calcium requirements, however, vary considerably from individual to individual due to the existence of other factors that affect availability of calcium. For  example, vitamin D is required for optimal utilization of calcium. Excess dietary protein may upset calcium balance by causing more rapid excretion of calcium. Exercise  increases the efficiency of calcium utilization for bone formation. Calcium balance studies carried out on Peruvian Indians, who have extensive exposure to sunlight, get  extensive exercise, and subsist on low­protein vegetarian diets, indicate a need for only 300–400 mg calcium day–1. However, calcium balance studies carried out in  this country consistently show higher requirements and the RDA has been set at 800–1200 mg day2+. The chief symptoms of calcium deficiency are similar to those of vitamin D deficiency, but other symptoms such as muscle cramps are possible with marginal  deficiencies. A significant portion of low­income children and adult females in this country do not have adequate calcium intake. This is of particular concern because  these are the population groups with particularly high needs for calcium. For this reason, the U.S. Congress has established the WIC (Women and Infant Children)  program to assure adequate protein, calcium, and iron for indigent families with pregnant/lactating mothers or young infants. Dietary surveys show that 34–47% of the over­60 population consumes less than one­half the RDA for calcium. This is the group most at risk of developing  osteoporosis, characterized by loss of bone organic matrix as well as progressive demineralization. Causes of osteoporosis are multifactorial and

 

  

Page 1129

largely unknown, but it appears likely that part of the problem has to do with calcium metabolism (see Clin. Corr. 28.7). Recent studies have also suggested that  inadequate intake of calcium may result in elevated blood pressure. Although this hypothesis has not been conclusively demonstrated, it is of great concern because  most low­sodium diets (which are recommended for patients with high blood pressure) severely limit dairy products, the main source of calcium for Americans. Magnesium Is Another Important Macromineral Magnesium is required for many enzyme activities and for neuromuscular transmission. Deficiency is most often observed in conditions of alcoholism, use of certain  diuretics, and metabolic acidosis. The main symptoms of magnesium deficiency are weakness, tremors, and cardiac arrhythmia. There is some evidence that  supplemental magnesium may help prevent the formation of calcium oxalate stones in the kidney. CLINICAL CORRELATION 28.7  Diet and Osteoporosis The controversies raging over the relationships between calcium intake and osteoporosis  illustrate the difficulties we face in making simple dietary recommendations for complex  biological problems. Based on the TV ads and wide variety of calcium­fortified foods on  the market, it would be easy to assume that all an older woman needs to prevent  osteoporosis is a diet rich in calcium. However, that may be like closing the barn door  after the horse has left. There is strong consensus that the years from age 10 to 35, when  the bone density is reaching its maximum, are the most important for reducing the risk of  osteoporosis. The maximum bone density obtained during these years is clearly  dependent on both calcium intake and exercise and dense bones are less likely to become  seriously depleted of calcium following menopause. Unfortunately, most American  women are consuming far too little calcium during these years. The RDA for calcium is  1200 mg day–1 (4 glasses of milk per day) for women from age 11 to 24 and 800 mg  day–1 (2 glasses of milk per day) for women over 24. The median calcium intake for  women in this age range is only about 500 mg day–1. Thus it is clear that increased  calcium intake should be encouraged in this group. But what about postmenopausal women? After all, many of the advertisements seem to  be targeted at this group. Do they really need more calcium? The 1994 NIH consensus  panel on osteoporosis recommended that postmenopausal women consume up to 1500  mg of calcium per day, but this recommendation has been vigorously disputed by other  experts in the field. Let's examine the evidence. Calcium balance studies have shown that  many postmenopausal women need 1200–1500 mg of calcium per day to maintain a  positive calcium balance (more calcium coming in than is lost in the urine), but that does  not necessarily mean that the additional calcium will be stored in their bones. In fact, some  recent studies have failed to find a correlation between calcium intake and loss of bone  density in postmenopausal women while others have reported a protective effect. All of  those studies have been complicated by the discovery that calcium intake may have  different effects on different types of bones. Calcium intakes in the range of 1000–1500  mg day–1 appear to slow the decrease in density of cortical bone, such as that found in the  hip, hand, and some parts of the forearm. Similar doses, however, appear to have little or  no effect on loss of density from the trabecular bone found in the spine, wrist, and other  parts of the forearm. At least some of the confusion in the earlier studies appears to have  resulted from differences in the site used for measurement of bone density. Thus the effect  of high calcium intakes alone on slowing bone loss in postmenopausal women remains  controversial at present. It is clear that elderly women should be getting at least the RDA  for calcium in their diet. With the recent concern about the fat content of dairy products,  calcium intakes in this group appear to be decreasing rather than increasing. Furthermore,  even with estrogen replacement therapy, calcium intake should not be ignored. Recent  studies have shown that with calcium intakes in the range of 1000–1500 mg day–1, the  effective dose of estrogen can be reduced significantly. While the advertisements and much of the popular literature focus on calcium intake, we  also need to remember that bones are not made of calcium alone. If the diet is deficient in  other nutrients, the utilization of calcium for bone formation will be impaired. Vitamin C is  needed to form the bone matrix and the macrominerals magnesium and phosphorus are  an important part of bone structure. Recent research has also shown that vitamin K and a  variety of trace minerals, including copper, zinc, manganese, and boron, are important for  bone formation. Thus calcium supplements may not be optimally utilized if the overall diet  is inadequate. Vitamin D is important for absorption and utilization of calcium. It deserves  special mention since it may be a particular problem for the elderly (see Clin. Corr. 28.9).  Finally, an adequate exercise program is just as important as estrogen replacement  therapy and an adequate diet for preventing the loss of bone density. Schaafsma, G., Van Berensteyn, E. C. H., Raymakers, J. A., and Dursma, S.A.  Nutritional aspects of osteoporosis. World Rev. Nutr. Diet. 49:121, 1987; Heaney, R.  P. Calcium in the prevention and treatment of osteoporosis. J. Intern. Med. 231:169,  1992; and National Institutes of Health. Optimal calcium intake. NIH Consens.  Statement, 12 (Nov. 4), 1994.    

  

Page 1130

28.10—  Trace Minerals Iron Is Efficiently Reutilized Iron metabolism is unique in that it operates largely as a closed system, with iron stores being efficiently reutilized by the body. Iron losses are minimal (f, 575­584, 1139 Variable (V) regions, 90 Variable surface glycoprotein, 401 Variegate porphyria, 1012t, 1016 Vasoactive intestinal polypeptide, 932t, 1062 actions, 848t as secretagogue, 1061t, 1062 source, 848t structure, 1062t Vasoconstriction, in blood coagulation, 960 Vasopressin, 842, 844t actions, 847t, 848t, 880­883, 882f, 883t, 884t secretion, 881­882, 882f source, 848t structure, 858, 858t synthesis, 851, 851f Vectors, 780, 783 Vegetarian diet(s), 1091 nutritional considerations with, 1133, 1134cc and protein­energy requirements, 1091, 1091cc serum cholesterol and, 1100 and vitamin B12 (cobalamine) deficiency, 1126 Velocity profile of reaction. See Reaction(s), velocity Versican, 356 Very low density lipoprotein(s), 58, 379, 529. See also Lipoproteins in obesity, 1095 Vesicle­associated membrane protein, 927 Vibrio cholerae, diarrhea caused by, 1068cc Vibrio vulnificus, 1003cc VIP. See Vasoactive intestinal polypeptide Viruses, in expression vectors, 785 Vision, 932­946 genes for, chromosomal location, 944­945, 946cc loss, 481cc signal transduction in, 936­937 Visual cycle, 943, 943f, 944t Visual pigments, 937, 944 absorption spectrum, 941f amino acid sequences, 944, 945f genes, 944­945 Visual proteins, 1110 Vitamin(s) A, 938, 1109­1111 deficiency, 1111 dietary sources, 1111 mechanism of action, 1111 metabolism, 1110, 1110f structure, 1109, 1110f, 1145­1146 toxicity, 1111 in vision, 1110­1111, 1111f absorption, 1082 B1 (thiamine), 1119 deficiency, 1119­1120 requirements, 1120­1121 structure, 1119, 1119f therapy with, 479cc B6 (pyridoxine), 142, 1121. See also Pyridoxal phosphate deficiency, 450, 476, 1122 requirements, 1122 and anticonvulsant therapy, 1118cc for oral contraceptive users, 1124cc status, evaluation, 1122, 1124cc structure, 1121, 1121f B12 (cobalamine), 1125 deficiency, 1125, 1126 dietary sources, 1126 metabolic role, in one­carbon metabolism, 1123f metabolism, 1125­1126, 1126f structure, 1125f, 1125­1126 C (ascorbic acid), 536, 1127 and common cold, 1127­1128 deficiency, 1127 functions, 1127­1128 megadoses, 1128 precursors, 344 requirements, 1127 serum levels, factors affecting, 1127 transport, in mammalian cells, 211t D in calcium homeostasis, 1112­1114, 1115f deficiency, 1114 dietary sources, 1112, 1114 endocrine system, 906f, 907 parathyroid hormone and, 1112­1114 requirements and anticonvulsant therapy, 1118cc in newborn, 1117cc structure, 1113f synthesis, 420, 1111­1114 toxicity, 1114 D2 (ergocalciferol), 420, 1112 D3 (cholecalciferol), 420 actions, 907 metabolism, 894, 992 photochemical conversion of 7­dehydrocholesterol to, 419f structure, 1145 synthesis, 907­908, 1111 deficiency, in alcoholism, 1120cc E, 1114 as antioxidant, 1116 deficiency, 1114 requirements, 1116 structure, 1145­1146 supplementation, 1116 in newborn, 1117cc tocopherols in, 1114­1116 toxicity, 1116 fat­soluble, 1109­1118 structure, 1145­1146 K (phytonadione), 1116, 1116­1118 deficiency, 1117, 1117cc, 1117­1118 epoxide, 971 function, 1116, 1116f in protein glutamyl carboxylation reactions, 970f, 970­971 requirements and anticonvulsant therapy, 1118cc in newborn, 1117cc synthesis, 1117 K1 (phytylmenaquinone), 1116 K2 (multiprenylmenaquinone), 1116, 1145­1146 requirements, anticonvulsants and, 1118cc, 1125 water­soluble, 1118­1119 coenzyme function, 1118 deficiencies, 1118 energy­releasing, 1118­1122 versus fat­soluble vitamins, 1118 hematopoietic, 1123­1126

 

  

Page 1186

Vitreous humor, 932, 933 VLDL. See Very low density lipoprotein(s) Vmax, 137­138, 152 Voltage­gated channels, 201, 921, 922, 923 disorders, 956cc­957cc Von Gierke's disease, 317cc von Willebrand factor, 967 VSG. See Variable surface glycoprotein vWF. See von Willebrand factor W Warfarin, 971 Water, 4, 4f in cellular environment, 4­12 diffusion, through membranes, 196 molecules, 4f, 4­5 reactions with sodium lactate, 5f solvent properties, 5, 5 as weak electrolyte, 6 Waxes, 1143 Wear and tear pigment (lipofuscin), 19 Werner's syndrome, 639cc Wernicke­Korsakoff syndrome, 1120, 1120cc WIC (Women and Infant Children) program, 1128, 1130 Wilson's disease, 1132 Wobble hypothesis, 719, 719t Wolman's disease, 19cc Women, nutritional considerations for, 1133 X Xanthine, 492 Xanthine oxidase, 502 Xenobiotic regulatory elements, 986 Xenobiotic­metabolizing enzyme(s), 993, 994t Xenobiotics, 992, 992, 994t Xeroderma pigmentosum, 636, 638cc X­linked spinal and bulbar muscular atrophy, 602cc X­ray crystallography, 106t, 109f, 116, 897f X­ray diffraction, 76, 102f XREs. See Xenobiotic regulatory elements Xylitol dehydrogenase deficiency, 345cc Xylose, UDP­, 344 Xylose 5­phosphate, 338, 338­340 Xylulose, structure, 1140 Y YAC. See Yeast artificial chromosomes Yeast(s), 613, 821 Yeast artificial chromosomes, 780­781 Z Z line, 948 Z­DNA, 571f, 573t, 574 ZDV (zidovudine). See Azidothymidine Zellweger syndrome, and peroxisome function, 20cc Zero­order reactions, 135 Zinc absorption, 1131 as cofactor, 1131 deficiency, 1131 as Lewis acid, 144, 145, 145f Zinc finger(s), 108, 109f, 110f, 913, 913f Zinc metalloenzymes, 1071t, 1072, 1131 Zinc proteases, 144 Zona fasciculata cells, in steroid synthesis, 898 Zona glomerulosa cells, in steroid synthesis, 898 Zona reticularis cells, in steroid synthesis, 898 Zonal centrifugation, 585, 585­586 Zwitterion form, 33, 33­34, 1146 Zymogen granules, 1060 Zymogens, 98, 743­744, 1060, 1071­1072

 

  

Page 1187 NORMAL CLINICAL VALUES: BLOOD* INORGANIC SUBSTANCES    Ammonia  Bicarbonate  Calcium  Carbon dioxide  Chloride  Copper  Iron  Lead  Magnesium  Pco2   

pH  Phosphorus  Po2   

Potassium  Sodium      ORGANIC MOLECULES    Acetoacetate  Ascorbic acid  Bilirubin  Direct  Indirect  Carotenoids  Creatinine  Glucose  Lactic acid  Lipids  Total  Cholesterol  Phospholipids  Total fatty acids  Triglycerides  Phenylalanine  Pyruvic acid  Urea nitrogen (BUN)  Uric acid  Vitamin A    PROTEINS    Total  Albumin  Ceruloplasmin  Globulin  Insulin    ENZYMES    Aldolase  Amylase  Cholinesterase  Creatine kinase (CK)  Lactic dehydrogenase  Lipase  Nucleotidase  Phosphatase (acid)  Phosphatase (alkaline)  Transaminase (SGOT)    PHYSICAL PROPTERTIES    Blood pressure  Blood volume  Iron binding capacity  Osmolality  Hematocrit

    12–55 µmol/L  22–26 meq/L  8.5–10.5 mg/dl  24–30 meq/L  100–106 meq/L  100–200 µg/dl  50–150 µg/dl  10 µg/dl or less  1.5–2.0 meq/L  35–45 mmHg  4.7–6.0 kPa  7.35–7.45  3.0–4.5 mg/dl  75–100 mmHg  10.0–13.3 kPa  3.5–5.0 meq/L  135–145 meq/L          negative  0.4–15 mg/dl    0–0.4 mg/dl  0.6 mg/dl  0.8–4.0 µg/ml  0.6–1.5 mg/dl  70–110 mg/dl  0.5–2.2 meq/L    450–1000 mg/dl  120–220 mg/dl  9–16 mg/dl as lipid P  190–420 mg/dl  40–150 mg/dl  0–2 mg/dl  0–0.11 meq/L  8–25 mg/dl  3.0–7.0 mg/dl  0.15–0.6 µg/ml        6.0–8.4 g/dl  3.1–4.3 g/dl  23–43 mg/dl  2.6–4.1 g/dl  0–29 µU/ml        0–7 U/ml  4–25 U/ml  0.5 pH U or more/h  40–150 U/L  110–210 U/L  2 U/ml or less  1–11 U/L  0.1–0.63 Sigma U/ml  13–39 U/L  9–40 U/ml        120/80 mmHg  8.5–9.0% of body weight in kg  250–410 µg/dl  280–296 mOsm/kg H O  2

37–52%

NORMAL CLINICAL VALUES: URINE* Acetoacetate (acetone)  Amylase  Calcium  Copper  Coproporphyrin  Creatine  Creatinine  5­Hydroxyindoleacetic acid  Lead  Phosphorus (inorganic)  Porphobilinogen  Protein (quantitative)  Sugar  Titratable acidity  Urobilinogen  Uroporphyrin

0  24–76 U/ml  0–300 mg/d  0–60 µg/d  50–250 µg/d  under 0.75 mmol/d  15–25 mg/kg body weight/d  2–9 mg/d  120 µg/d or less  varies; average 1 g/d  0  less than 165 mg/d  0  20–40 meq/d  up to 1.0 Ehrlich U  0–30 µg/d

*Selected values are taken from normal reference laboratory values in use at the Massachusetts General  Hospital and published in the New England Journal of Medicine 314:39, 1986 and 327:718, 1992. The reader  is referred to the complete list of reference laboratory values in the literature citation for references to  methods and units. dl, deciliters (100 ml); d, day.

 

  

View more...

Comments

Copyright ©2017 KUPDF Inc.
SUPPORT KUPDF