No title

Current Topics in Membranes and Transport VOLUME 32

Membrane Fusion in Fertilization, Cellular Transport, and Viral Infection

Advisory Board

G . Blobel E. Carafoli 3. S. Cook D.Louvard

Cur rent Top ics in Membranes and Transport Edited by Felix Bronner D i ~ p i r r t t n ~ wot/ BioStrrrctrrrc trnd Firnc.tiori The University of Connecticut Health C r n t e r School of Dental Medicine Farrnington, Connecticut

VOLUME 32

Membrane Fusion in Fertilization, Cellular Transport, and Viral Infection

Guest Editors Nejat Duzguneg

Felix Bronner

Department qf Pharmaceutical Chemistty Cuncer Reseurch Institute and Schools of Medicine and Phnrniacy Uniuersity of Californiu San Francisco. California

Ikpurttnctit

@)

0 1 BioStrrrc.trrri, rrncl Frorc~ti~~ri The University of Connecticut Health Center School of' Dental Medicine Farrnington, Connecticirt

ACADEMIC PRESS, INC Harcourt Brace Jovanovich, Publishers

San Diego New York Berkeley Boston London Sydney Tokyo Toronto

COPYRIGHT

0 1988 BY ACADEMICPRESS, INC.

ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL. INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS, INC. 1250 Sixth Avenue San Diego, California 92101

United Kingdom Edirion published by ACADEMIC PRESS INC. ( L O N D O N ) LTD. 24-28 Oval Road, London NWI 7DX

LIBRARY OF CONGRESS CATALOG

ISBN 0-12-153332-8

CARD

(alk. paper)

PRINTED IN THE UNITED STATES OF AMERICA 8 8 8 9 9 0 9 1

9

8

7

6

5

4

3

2

1

NUMBER: 70-117091

Contents

Contributors, ix Preface, xi Peter Baker, xv Yale Membrane Transport Processes Volumes, xvii

PART I.

MEMBRANE FUSION IN FERTILIZATION AND DEVELOPMENT

Chapter 1. Sperm-Egg Fusion RYUZO YANAGIMACHI

I . Introduction. 4 11. Morphology of Sperm-Egg Fusion, 4 Ill. Specific Sites for Sperm-Egg Fusion, 13

IV. V. VI. VII.

Sperm Movement and Sperm-Egg Fusion, 18 Intermingling of Sperm and Egg Plasma Membrane, 19 Species Specificity of Sperm-Egg Fusion, 20 How Long Does the Egg Plasma Membrane Remain Capable of Fusing with Spermatozoa after the Entry of the First Spermatozoon'?, 24 VI11. Conditions Controlling Sperm-Egg Fusion. 26 IX. Effects of Miscellaneous Substances on Sperm-Egg Fusion, 31 X . Mechanism of Sperm-Egg Fusion. 34 References, 35

Chapter 2. Cortical Exocytosis in the Sea Urchin Egg ROBERT C . JACKSON AND JOSEPH H. CRABB 1. Introduction, 45 11. The Calcium Signal, 51 111. In Virro Models of Exocytosis, 56

IV. Conclusions, 73 V. Addendum. 74 References, 76 V

vi

CONTENTS

Chapter 3. Myoblast Fusion-A

Mechanistic Analysis

MICHAEL J. 0. WAKELAM 1. Introduction, 88 11. Morphological Aspects of Myoblast Fusion, 88

Ill. IV. V. VI. VII. VIII.

Kinetics of Myoblast Fusion, 92 Structure-Function Relationships in Myoblast Plasma Membranes, 94 Fusion-Induced Changes in Membrane Organization, 99 Proposed Mechanisms of Myoblast Fusion, I01 Stimulation of Myoblast Fusion, 106 Conclusions, 107 References, 107

PART 11. CELLULAR TRANSPORT-EXOCYTOSIS

AND ENDOCYTOSIS

Chapter 4. Exocytosis in Electropermeabilired Cells: Clues to Mechanism and Physiological Control PETER F. BAKER

I. Introduction, I I5 11. Morphology of Exocytosis, 117

Ill. Clues to Mechanism, 118 IV. Control of Exocytosis, 120 V. Responding to the Primary Message, 126 V1. Other Control Factors in Exocytosis, 132 References. 134

Chapter 5. Exocytosis and Membrane Recycling JACOPO MELDOLESI AND BRUNO CECCARELLI

I. Introduction, 139 11. Exocytosis, 142 111. Endocytosis, 156 1V. Conclusion, 160 References, 161

Chapter 6. Exocytosis and Endocytosis: Membrane Fusion Events Captured in Rapidly Frozen Cells DOUGLAS E. CHANDLER

I. Exocytosis, 169 11. Endocytosis, 189 111. Concluding Remarks, 194 References, 197

CONTENTS

vii

Chapter 7. Osmotic Effects In Membrane Fusion during Exocytosis KEITH W . BROCKLEHURST AND HARVEY B. POLLARD 1. Introduction. 203 11. Osmotic Properties of Isolated Secretory Granules. 204 111. Osmotic Effects in Secretion from Intact Cells. 211

I V . Osmotic Effects in Secretion from Permeabilized Chrornaffin Cells, 216 V. Conclusions, 219 References. 220

Chapter 8. Polyanionic Agents and Inhibition of PhagosomeLysosome Fusion: Paradox Lost MAYER B. GOREN I . Introduction. 228 I t . The Fluorescent Lysosomal Probes, 231 111. Nonionic Hydrocolloids in Lysosornes: Fusion Inhibitors or Gelatinous Traps?, 238 IV. Fusion Inhibition Is Incompatible with the Cells' Functional Status. 242 V. Some Residual Bodies. 246 VL. Recapitulation and Conclusions, 249 References, 250

PART 111. VIRUS-CELL FUSION Chapter 9. Fusion of Viral Envelopes with Cellular Membranes SHUN-ICHI OHNISHI I . Introduction, 257 11. Membrane Fusion Activity of Enveloped Viruses. 263 111. Mechanism of Fusion, 278

1V. Infectious Cell Entry Mechanisms. 286 References. 288

Chapter 10. Sendai Virus-Mediated Cell Fusion YOSHIO OKADA

I. 11. 111. IV.

Introduction, 297 Critical Problems in Cell-to-Cell Fusion, 298 Structure and Biological Activities of HVJ, 299 Cell-to-Cell Fusion by HVJ. 306 References, 331

viii

CONTENTS

Chapter 11. Fusion Activity of the Hemagglutinin of Influenza Virus MARY-JANE GETHING, JEAN HENNEBERRY, AND JOE SAMBROOK 1. Introduction, 337 11. Influenza Virus-Mediated Fusion: Role of the Hernagglutinin, 339

Ill. Assays for the Fusion Activity of HA, 340 IV. Expression of HA in Cultured Cells from Cloned HA cDNAs, 341 V. Genetic Approaches to Studies of HA-Mediated Membrane Fusion, 34.5 V1. Characterization of the Low pH-Induced Conformational Change in HA. 3.52 VII. Studies o n the Clcavagc Activation of HA. 3 5 3 VIII. Conclusion, 360 References. 360

Index, 365

Contributors

Numbers in parentheses indicate the pages on which the authors’ contributions begin

Peter F. Baker,’ MRC Secretory Mechanisms Group, Department of Physiology, King’s College London, London WC2R 2LS. England ( 1 15) Keith W. Brocklehurst, Laboratory of Cell Biology and Genetics, National Institute of Diabetes. Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892 (203) Bruno Ceccarelli, Center for the Study of Peripheral Neuropathies and Neuromuscular Diseases, University of Milan, 20129 Milan, Italy (139) Douglas E. Chandler, Department of Zoology, Arizona State University, Tempe, Arizona 85287 ( 169) Joseph H. Crabb, Channing Laboratory, Harvard Medical School, Boston, Massachusetts 021 15 (45) Mary-Jane Gething, Department of Biochemistry and Howard Hughes Medical Institute, University of Texas Health Science Center, Dallas, Texas 75235 (337) Mayer B. Goren, Department of Molecular and Cellular Biology, National Jewish Center for Immunology and Respiratory Medicine, Denver, Colorado 80206 (227) Jean Henneberry, Department of Biochemistry, University of Texas Health Science Center, Dallas, Texas 75235 (337) Robert C. Jackson, Department of Biochemistry, Dartmouth Medical School, Hanover, New Hampshire 03756 (45) I

Deceased. ix

X

CONTRIBUTORS

Jacopo Meldolesi, Department of Pharmacology, CNR Center of Cytopharmacology, Scientific Institute Saint Raffaele, Milan, Italy (139) Shun-ichi Ohnishi, Department of Biophysics, Faculty of Science, Kyoto University, Sakyo-ku, Kyoto 606, Japan (257) Yoshio Okada, Institute for Molecular and Cellular Biology. Osaka University, Suita, Osaka 565, Japan (297) Harvey B. Pollard, Laboratory of Cell Biology and Genetics, National Institute of Diabetes, Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892 (203) Joe Sambrook, Department of Biochemistry, University of Texas Health Science Center, Dallas, Texas 75235 (337) Michael J. 0. Wakelam, Molecular Pharmacology Group, Department of Biochemistry, University of Glasgow, Glasgow, G12 8QQ Scotland (87)

Ryuzo Yanagimachi, Department of Anatomy and Reproductive Biology, University of Hawaii School of Medicine, Honolulu, Hawaii 96822 (3)

Preface

Membrane fusion is the first step in fertilization and thus constitutes the initial event in the life of higher organisms. By mediating the cortical reaction that follows after the sperm penetrates the egg, membrane fusion prevents polyspermy and assures orderly development. It also plays a specific role in development, as when myoblasts fuse to form myotubes. As cells mature and differentiate, export of their products requires a well-defined sequence of synthesis. packaging, and routing within and, ultimately, to the outside of the cell. Membrane fusion, between organelles and between organelles and the plasma membrane, plays a prominent role in this process as well. Finally, cellular infection and disease, as brought about by lipid-enveloped viruses, involve fusion of the viral membrane with the plasma membrane or, following endocytosis in coated pits, fusion with the endosome membrane. The development in recent years of the powerful tools of cellular and molecular biology has enabled researchers to observe membrane fusion in detail and to begin studying the underlying steps of this process. Much information has accumulated on this topic, though a definitive understanding still eludes us. The search for this understanding is predicated on knowing what has been done till now, which hypotheses have proved fruitful, and which need to be discarded. The realization that there exist as yei few places where membrane fusion in biology has been described and analyzed systematically prompted us to attempt such a survey. This book is the outcome of that effort. As the title indicates, the book examines three major areas: fertilization and development, cellular transport as exemplified by endocytosis and exocytosis, and the mechanisms by which viruses penetrate cells and cause cell-cell fusion. The first chapter, by Yanagimachi, is a comprehensive survey of sperm-egg fusion in invertebrates, lower vertebrates, and mammals. It describes and analyzes sites for sperm-egg fusion, sperm movement, the intermingling of sperm and egg plasma membranes, the time the egg remains receptive to sperm fusion, and the conditions and mechanism of that event. xi

xii

PREFACE

The second chapter, by Jackson and Crabb, analyzes cortical granule exocytosis. There is now ample evidence that an increase in cytosolic free calcium precedes and is the signal for exocytosis. The calcium signal is therefore described in detail, including the regulatory role of the guanyl nucleotide binding proteins. Finally, the chapter analyzes the use of subcellular fractions of sea urchin eggs with their retained exocytotic apparatus for in uitro studies of the calcium trigger. During development mononucleate myoblasts align and fuse to form the precursors of skeletal muscle fibers. This event is dealt with in the third chapter. Wakelam recognizes that not enough is known to provide a detailed mechanistic description, but his assembly of the relevant findings should permit the reader to form a view of what approaches have or have not been useful. The second section, on the role of membrane fusion in intracellular transport and cellular export, begins with a survey of the nature of exocytosis written by the late Peter Baker. Professor Baker was among the first to have become fascinated by the role played by intracellular calcium in exocytosis and published extensively on this topic. His sudden death at a young age, shortly after he completed Chapter 4, is all the more tragic because it has deprived the scientific world of a perceptive and outstanding scholar. We therefore have asked his long-time colleague, Derek Knight, to help us commemorate him. Meldolesi and Ceccarelli, in Chapter 5, deal with the interesting problem of how membranes of intracellular vesicles that fuse with the plasma membrane are retrieved, rather than lost to the economy of the cell. They draw examples from nerve cells with which they worked extensively, but their overall survey is broad and constitutes a biologically oriented companion to the chapter by Baker. Chapter 6 by Chandler is a detailed structural analysis of the events in endocytosis and exocytosis. Chandler’s contribution is all the more unique because he has accompanied his text by revealing illustrations, some by now classics. Some years ago Pollard and colleagues, basing themselves on the ingenious chemiosmotic mechanism that Peter Mitchell had proposed for the coupling of phosphorylation to electron and hydrogen transfer, proposed an analogous mechanism to explain exocytotic membrane fusion. This hypothesis was fruitful in that it stimulated much careful, experimental work that has greatly broadened our understanding of the fusion process. Brocklehurst and Pollard, in Chapter 7, examine the evidence and conclude that the hypothesis is no longer tenable, at least in its totality. A somewhat similar situation is described in Chapter 8 by Goren. Some ten years ago Goren published findings that indicated that polyanionic

PREFACE

xiii

substances, such as dextran sulfate, when phagocytosed, inhibited the fusion of phagosomes with lysosomes. As it turned out, substances such as acridine orange that had been used as fusion markers had become immobilized, giving the appearance of no fusion, whereas the two vesicles had in fact become fused. Goren reviews the evidence in detail, and thus provides a resolution of what might otherwise have been a scientific cul-de-sac. Ohnishi in Chapter 9 introduces the third section of the book by discussing the mechanisms by which viruses bind and ultimately penetrate target cells. Even when the virus binds to its target cell membrane through interaction of the viral glycoprotein with the cell surface receptor, the two membranes are still far apart. Ohnishi describes how conformational changes in the viral glycoprotein are achieved and how these changes cause the hydrophobic segment of the glycoprotein to approach and interact with target membranes. He also describes how enveloped viruses gain entry into the cell by utilizing the machinery that cells have developed for the uptake and processing of materials useful to them. Okada, one of the first to use the Sendai virus to fuse mammalian cell membranes, reviews this field in detail in Chapter 10. His richly illustrated chapter analyzes the molecular and structural basis of this now widely utilized process. I n the final chapter, Gething, Henneberry, and Sambrook describe cell fusion mediated by influenza virus and describe genetic approaches to this process. The hemagglutinin glycoprotein is the major antigen of the influenza virus. It is the ability of these viruses to synthesize antigenically novel hemagglutinins which explains the difficulty in controlling the disease. At the same time, the advent of recombinant DNA technology, knowledge of the gene encoding the hemagglutinin, and X-ray analysis of the structure of the ectodomain of the hemagglutinin molecule have provided substantial information on its physical domains, location of antigenic and glycosylation sites, its trimeric structure, and orientation with respect to the plasma membrane. The hemagglutinin molecule, therefore, constitutes an excellent model system for the study of membrane fusion. We hope the material contained in this book will lead investigators not only to important practical solutions, such as the design of agents that will prevent entry of virions, but a step closer to understanding a fundamental biological process like fertilization. NEJATDUZGUNES FELIXB R O N N E R

PETERBAKER

Peter Baker (1939-1987)

Professor Peter Baker’s sudden death in March 1987, at the early age of 48, was a shock to those who knew him and to those in the scientific

community who knew and respected his work. Although his career was cruelly curtailed. his life was filled with challenges and was one of purpose. His achievements are now part of the foundation on which others can build. To his colleagues he was a man who succeeded in combining his exceptional scientific gifts with a practical administrative sense to create an environment in which research and achievement were initiated and nourished. To his friends in the scientific community, both at home and abroad, he was a man who carried with him a rare enthusiasm for all that he did, which he conveyed in a way that never failed to stimulate or enlighten those who came in contact with him. His resilience and strength, backed by his unwavering optimism and confidence in the unlimited scope of scientific advancements, served him in his own research and in the wider international field of life sciences. Peter Baker was born in Lincoln, England, on March 1 1 , 1939. He attended Lincoln School and from there won an Open Scholarship to Emmanuel College, Cambridge, where he read Natural Sciences. His research talents were immediately apparent and flourished in his subsequent work with Trevor Shaw and Sir Alan Hodgkin in the Department of Physiology. The most widely recognized aspect of this work was to refine the ionic theory of the nervous impulse by means of performing experiments with squid axon, the protoplasm of which had been replaced by media of accurately defined chemical composition. These studies during the 1960s led to the publication of several significant papers. While at Cambridge, he began the study of the control and maintenance of the ionic components of the intracellular environment, most notably the role of calcium. He characterized the two voltage-sensitive routes of calcium entry into the cell and then proceeded to characterize the component of calcium movement across the plasma membrane associated with the movement of sodium (sodium-calcium exchange). His involvement in xv

xvi

PETER BAKER

this area of research continued to the time of his death, when he was in the midst of organizing the first International Symposium on SodiumCalcium Exchange. His interest in exocytosis also developed at Cambridge where, with Tim Rink, he extended the link between calcium entry into cells and their secretory response. Drs. Baker and Rink provided evidence that strongly suggested that the transient nature of that response resulted from a transient calcium influx into the cell. Thus, they were able to infer that the calcium channels which gave rise to the transient calcium influx had properties similar to the ‘‘late’’ voltage-sensitive calcium channels Dr. Baker had characterized in his earlier studies with the squid axon. Peter Baker accepted the Halliburton Chair at King’s College London in 1975 and was elected a Fellow of The Royal Society a year later. At King’s College he extended his own research into the area of exocytosis, and it was there that I joined him. Together we mounted a direct attack on the problem of the mechanism of secretion by adopting and developing an electropermeabilization technique. This technique breaches the plasma membrane of cells without disturbing the structural integrity of the intracellular organelles. Thus it was possible, by diffusing solutes through these pores in the plasma membrane, to manipulate experimentally the chemical environment of the site of exocytosis. Such studies defined accurately the calcium and nucleotide requirements for secretion and provided a means of directly testing various hypotheses concerned with this mechanism. Peter Baker’s activities, however, were by no means confined to research. He exhibited equal enthusiasm in his other commitments, as a member of various committees of the Medical Research Council, Royal Society, and Agricultural and Food Research Council. He also participated in the editorial board of The Journal of Physiology. Peter Baker’s life contained many challenges. He accepted each with gladness and executed each with hope. His death is a loss, not only to me and those immediately around him, but to all those who know what it means to ‘‘live’’ science. Photo of Peter Baker, working at The Marine Biological Association of the U.K., Plymouth, was taken by T. J. Allen in January 1987.

DEREKKNIGHT Department of Physiology King’s College London

Yale Membrane Transport Processes Volumes

Joseph F. Hoffman (ed.). (1978). “Membrane Transport Processes.” Vol. I . Raven, New York. Daniel C. Tosteson, Yu. A. Ovchinnikov, and Ramon Latorre (eds.). (1978). “Membrane Transport Processes,” Vol. 2. Raven, New York. Charles F. Stevens and Richard W. Tsien (eds.). (1979). “Membrane Transport Processes,” Vol. 3 : Ion Permeation through Membrane Channels. Raven, New York. Emile L. Boulpaep (ed.). (1980). “Cellular Mechanisms of Renal Tubular Ion Transport”: Volume 13 of Cirrrerzf Topics in Membranes and Trunsport (F. Bronner and A. Kleinzeller. eds.). Academic Press, New York. William H. Miller (ed.). (1981). “Molecular Mechanisms of Photoreceptor Transduction”: Volume 15 of Ciirrent Topics in Membranes and Transport (F. Bronner and A. Kleinzeller, eds.). Academic Press. New York. Clifford L. Slayman (ed.). (1982). “Electrogenic Ion Pumps”: Volume 16 of Current Topics in Membranes und Transport (A. Kleinzeller and F. Bronner, eds.). Academic Press, New York. Joseph F. Hoffman and Bliss Forbush 111 (eds.). (1983). “Structure, Mechanism, and Function of the Na/K Pump”: Volume 19 of Cirrrent Topics in Membranes and Transport (F. Bronner and A . Kleinzeller, eds.). Academic Press, New York. James B. Wade and Simon A. Lewis (eds.). (1984). “Molecular Approaches to Epithelial Transport”: Volume 20 of Current Topics in Membranes and Trurzsport (A. Kleinzeller and F. Bronner, eds.). Academic Press, New York. Edward A. Adelberg and Carolyn W. Slayman (eds.). (1985). “Genes and Membranes: Transport Proteins and Receptors”: Volume 23 of Current Topics in Membranes and Transport (F. Bronner and A. Kleinzeller, eds.). Academic Press, Orlando. xvii

xviii

YALE MEMBRANE TRANSPORT PROCESSES VOLUMES

Peter S. Aronson and Walter F. Boron (eds.). (1986). ‘“a’-H’ Exchange, Intracellular pH, and Cell Function”: Volume 26 of Current Topics in Membranes and Transport (A. Kleinzeller and F. Bronner, eds.). Academic Press, Orlando. Gerhard Giebisch (ed.). ( I 987). “Potassium Transport: Physiology and Pathophysiology”: Volume 28 of Current Topics in Membranes und Transport (F. Bronner and A. Kleinzeller, eds.). Academic Press, Orlando.

Part I

Membrane Fusion in Fertilization and Develo p ment

This Page Intentionally Left Blank

CURRENT TOPICS IN MEMBRANES A N D TRANSPORT. VOLUME 32

Chapter I Sperm-Egg Fusion RYUZO YANAGIMACHI Department of Anutotny cind Rcprodiri~tiuc~ Biolox! University of ffiinwii School o f Medic~itw Hotiolirlir , Ha Muii 96822

1.

II.

111.

IV. V. VI. VII.

VIII.

IX.

X.

introduction Morphology of Sperm-Egg Fusion A. Sea Urchins B. Some Other Marine Invertebrdte~ C. Fishes 11. Amphibians E. Birds F. Mammals Specifc Sites for Sperm-Egg Fusion A. Sperm Membrane 8 . Egg Plasma Membrane Sperm Movement and Sperm-Egg Fubion Intermingling of Sperm and Egg Plasma Membrane Species Specificity of Sperm-Egg Fusion How Long Does the Egg Plasma Membrane Remain Capable of Fusing with Spermatozoa after the Entry of the Firct Spermatozoon? Conditions Controlling Sperm-Egg Fusion A. Temperature. pH. and Ionic Composition of Medium B. Electric Potential of the Egg Plasma Membrane Effects of Miscellaneous Substances on Sperm-Egg Fu5ion A. Proteolytic Enzymes B. Antimembrane Antibodies C. Proteinase Inhibitors D. Dithiothreitol E. Erythrosine and Related Dyes F. Cytochalasin G. Other Reagents Mechanism of Sperm-Egg Fusion References

3 Copyright .I” 1988 hy Academic Press. Inc. A l l right\ 01 rrproduction in any form reserved

4

RYUZO YANAGIMACHI

I. INTRODUCTION

I would hope that readers will find sperm-egg fusion (fertilization) a fascinating example of membrane fusion. It is one of the “natural” or “spontaneous” membrane fusions that occur without the deliberate addition of exogenous fusing agents such as viruses or chemicals (Lucy, 1978; Papahadjopoulos et al., 1979; Okada, this volume, Chapter 10). Although fertilization has been studied extensively for over 100 years because of curiosity about the mechanisms that give rise to new generations (Lop0 and Vacquier, 1981), most experimental observations have been concerned with the physiological or developmental events that follow cell fusion. Few experiments have been undertaken with the specific aim to elucidate the membrane events involved in fusion per se (Papahadjopoulos et al., 1979; DuzgiineS, 1985). In this chapter, I shall discuss fertilization in a few selected classes of animals, with major emphasis on the process and mechanism of sperm-egg fusion. For comparative aspects of fertilization in a wide variety of animals, readers are referred to the books edited by Metz and Monroy (1967, 1969, 1985).

II.

MORPHOLOGY OF SPERM-EGG FUSION

A. Sea Urchins

The sea urchin egg, which is about 100 pm in diameter, is surrounded by a very thin (0.01-0.03 pm) vitelline envelope, which is further surrounded by a thick, gelatinous matrix called the jelly coat (Fig. 1A). The vitelline envelope is a glycoprotein (Glabe and Vacquier, 1977, 1978), and the jelly coat contains two major macromolecular components, a fucose sulfate-rich polymer and a sialoprotein (SeGall and Lennarz, 1979). When eggs and spermatozoa are shed into seawater, spermatozoa swarm around the jelly coat, and many of them penetrate it to reach the vitelline envelope (Fig. IB).The head of each spermatozoon binds firmly to the vitelline envelope and rotates (counterclockwise) as the tail beats vigorously (Fig. ID). Then, suddenly one of the spermatozoa (the fertilizing spermatozoon) stops its tail movement (Epel. e? al., 1977; Hinkley et al., 1986), and the vitelline envelope starts to lift up (Fig. 1E). This elevation of the vitelline envelope, due to exocytosis of cortical granules, begins around the fertilizing spermatozoon and propagates in a wavelike fashion from the point of sperm-egg fusion. During elevation of the vitelline envelope, the fertilizing spermatozoon remains almost motionless, but sporadically twitches its tail in such a way that the entire tail becomes incorporated into the perivitelline space (Fig. IF; M. Sugiyama, personal

1. SPERM-EGG FUSION

5

FIG.I . Fertilization (sperm-egg fusion) in the sea urchin. ( A ) Unfertilized egg of Tripw u s f t ~ . sg m r i / / u . surrounded by India ink to show the jelly coat around the egg. ( B ) Seven seconds after insemination. (C) Two minutes after insemination. (D-F) The fertilizing spermatozoon. shown by an arrow in E. stops its vigorous tail movement soon after fusion with the egg. ( G )The sperm head. (H-0) Successive stages of sperm-egg fusion. Abbreviations: a. acrosome; ap, acrosomal process: b. bindin: cg, cortical granules; epm, egg plasma membrane; eve. elevated vitelline envelope: hi. hyaline layer; o m . outer acrosomal membrane; pvs, perivitelline space; spm, sperm plasma membrane; ve. vitelline envelope.

6

RYUZO YANAGIMACHI

communication). The head and the entire sperm tail then gradually sink into the egg cytoplasm. The successive stages of membrane fusion between the spermatozoon and the egg are shown diagramatically in Fig. 1G through 0 (cf. Franklin, 1965; Summers and Hylander, 1974; Schatten and Mazia, 1976; Usui et al., 1980). During the passage of spermatozoa through the jelly coat (or on contact of the spermatozoa with the vitelline envelope), the acrosome opens up as a result of the fusion between the outer acrosomal membrane and the overlying plasma membrane (Fig. IH, I), while the inner acrosoma1 membrane elongates to form a process (Fig. lK, L). These changes are called the acrosome reaction (Dan, 1967). The interaction between the jelly components (or vitelline envelope components) and a sperm membrane protein (Podell and Vacquier, 1985) seems to trigger the reaction. One of the acrosomal components exposed as a result of the acrosome reaction is a protein called “bindin.” It functions in holding the acrosomal process firmly to the vitelline envelope (Fig. l K , L) (Vacquier and Moy, 1977). The mechanism by which the acrosome process penetrates through the vitelline envelope is currently unknown. It could be purely mechanical. Alternatively, it could be aided by a chymotrypsin-like enzyme released from the acrosome during the acrosome reaction (Hoshi, 1985). Membrane fusion begins between the tip of the acrosome process and the egg plasma membrane (Fig. lL, M). It should be noted that it is the acrosomal membrane, not the sperm plasma membrane, that fuses with the egg plasma membrane. Figure 2 shows sea urchin spermatozoa about to fuse (Fig. 2A) or fusing with eggs (Fig. 2B-D). B. Some Other Marine Invertebrates

Early stages of sperm-egg fusion in Hydroides hexagonus (an annelid), Saccoglossus kowalevskii (a hemicordate), Chama macerophylla (a mollusk), and Neanthesjaponica (a polychaete) are shown diagramatically in Fig. 3. Only the anterior region of spermatozoa is shown in the diagrams. The vitelline envelopes of the eggs of these species are much thicker (0.6-0.7 pm) than those of sea urchin eggs. Spermatozoa undergo the acrosome reaction on the vitelline envelope. In Hydroides, the inner acrosoma1 membrane elongates to form many acrosomal processes, each of which fuses with the egg plasma membrane. In Saccoglossus, one long process is formed, the tip of which fuses with the egg plasma membrane. The acrosomal process preexists in the spermatozoa of Chama and Neanfhes. The process, which is exposed after the acrosome reaction, fuses with one of the egg microvilli.

FIG.2 . Scanning electron micrograph5 showing successive stage5 of sperm-egg f'uaion in the w a urchin, S/ro,i~?~/oc,rn/,.c,~lrs prrrp/rr.cctrrs. (From Schatten and Mazia. 1976. with permission of Academic Press.)

8

RYUZO YANAGIMACHI

B Hemlohordate . .

.

C Mollusc

FIG.3. Sperm-egg fusion in four different marine invertebrates showing that spermegg fusion begins between the membrane covering the acrosomal process and the egg plasma membrane. Animal species and references are (A) annelid. Hvdroides lte.rugonrts (Colwin and Colwin, 1964, 1967):(B)hemichordate, Suc~coglossuskonwlruskii (Colwin and Colwin. 1964, 1967); (C) mollusk, Chuma mucerophyllu (Hylander and Summers, 1977); (D)polychaete, Neunthes juponica (Sato and Osanai. 1986). ve, Vitelline envelope.

C. Fishes

The fish egg has a thick and relatively tough vitelline envelope which is commonly called the chorion. Penetration of the spermatozoon into the egg takes place through the micropyle, a specialized narrow channel in the chorion (Fig. 4A). The relatively wide opening of the micropyle may permit the simultaneous entry or two or more spermatozoa, but its narrow bottom allows the passage of only one spermatozoon at a time.

9

1. SPERM-EGG FUSION Micropyle

FIG.4. Sperm-egg fusion in fishes. ( A ) Sagittal section of an egg before (BF) and after ( A F ) fertilization. The insert shows the relative size of the egg and spermatozoon of the herring. Clrrpeu pullasii. (B.C ) Sperm-egg fusion takes place between the plasma membrane of the sperm head and the egg plasma membrane at the bottom of the micropylar canal. Animal species and references are ( B ) carp. Cyprinrrs c,urpio (Kudo. 1980): tC) chum salmon, Oncorkynchrt.~ketu (Kobayashi and Yamamoto. 1981 1987). pvs. Perivitelline space.

Fish spermatozoa, unlike sea urchin spermatozoa, have no acrosome. In fish, it is the plasma membrane covering the sperm head that fuses with the egg plasma membrane. The manner of sperm-egg fusion in carp and salmon is shown diagramatically in Fig. 4B and C, respectively. The sturgeon is unusual among fishes in that its egg has several micropyles in the chorion (Cherr and Clark, 1982, 1985a). It is also unique in that the spermatozoon has an acrosome. The fertilizing spermatozoon appears to undergo the acrosome reaction while passing through the micropyle (Cherr and Clark, 1985b). Therefore, in the sturgeon it must be the inner acrosomal membrane covering the acrosomal process which

10

RYUZO YANAGIMACHI

fuses with the egg plasma membrane ( G . N . Cherr, personal communication). In this respect, the sturgeon is somewhat similar to the lamprey, although the egg of the latter has no micropyle (Kille, 1960). D. Amphibians

The amphibian egg has a relatively thin (3-5 pm) vitelline envelope. The envelope is further surrounded by a thick jelly coat. The probable manner of sperm-egg fusion in the bullfrog (Bufo japonica) is shown diagramatically in Fig. 5 . The acrosome opens on contact of the sperm head with the vitelline envelope (Yoshizaki and Katagiri, 1982). The acrosome is known to contain a lysin that facilitates penetration of the vitelline envelope (Iwao and Katagiri, 1982). Although no one has ever witnessed the moment of sperm-egg fusion in amphibians, it is believed to occur between the inner acrosomal membrane and the egg plasma membrane (C. Katagiri, personal communication). Sperm entry commonly occurs within a 60" circle of the animal pole (Elinson, 1975).

FIG.5. Probable manner of sperm-egg fusion in the frog, Bufo juponiru. as suggested by Dr. C. Katagiri (personal communication: Yoshizawa and Katagiri, 1982). The insert

shows the relative size of the egg and spermatozoon. cg. Cortical granules; ve, vitelline envelope.

11

1, SPERM-EGG FUSION

E. Birds

Figure 6 illustrates the manner of sperm-egg fusion in the domestic fowl (Callus gallus). The egg is surrounded by a thin (-3 p m ) vitelline envelope. The vitelline envelope is further surrounded by another coat (3-8 pm in thickness), the outer envelope. The latter is a secretory product of the oviduct. The fertilizing spermatozoon undergoes the acrosome reaction on the vitelline envelope (Okamura and Nishiyama, 1978a). The proteolytic activity of the acrosomal contents (Yanagimachi and Teichman, 1972) makes a “hole” in the envelope, through which the spermatozoon passes. Fusion begins between the inner acrosomal membrane and the egg plasma membrane (Okamura and Nishiyama, 1978b). F. Mammals

Very little is known about the process of sperm-egg fusion in the Monotremata and Marsupialia. According to Rodger and Bedford ( 1982), the opossum spermatozoon undergoes the acrosome reaction on the vitelline envelope (2-3 p m in thickness), creates a “hole” in the envelope, and then fuses with the egg plasma membrane, perhaps via the inner

I-

iooyn

.

,

.

.

.

.

.

.

. .

...

.

. ,.-

. .

FIG. 6 . Successive stages of sperm-egg fusion in the domestic fowl. G ~ U gcrll//c A (Okarnura and Nishiyarna, 1978a,b). The insert shows the relative size of the egg and spermatozoon. oe. Outer envelope: ve. vitelline envelope.

12

RYUZO YANAGIMACHI

acrosomal membrane. Eggs of eutherian mammals have a relatively thick glycoprotein envelope (usually 10-20 pm in thickness), commonly called the zona pellucida (cf. Dunbar and Wolgemuth, 1984). In most mammals, the zona is further surrounded by the cumulus oophorus at the time of fertilization. The cumulus consists of follicle cells embedded in their gelatinous (hyaluronic acid polymer) matrix. The fertilizing spermatozoon initiates the acrosome reaction (Fig. 7B) in the vicinity of, or on, the zona.

FIG.7. The acrosome reaction (A-B), zona penetration (C), and sperm-egg fusion (DGI in the cutherian mammal. The insert shows the relative size of golden hamster (Mesocric e l u ~rrurutus) and human gametes. (Dthrough G were redrawn from Bedford and Cooper, 1978, with permission of Elsevier/North Holland Biomedical Press.) Abbreviations: ac, acrosomal cap region; CG, cortical granules; eq, equatorial segment of the acrosome; iam,

inner acrosomal membrane: pvs, perivitelline space; zp, zona pellucida.

1. SPERM-EGG FUSION

13

The spermatozoon must be acrosome-reacted to pass through the zona (Fig. 7C). Sperm-egg fusion begins between the egg plasma membrane and the sperm plasma membrane over the equatorial segment of the acrosome (Fig. 7E; Bedford and Cooper, 1978; Bedford et al., 1979). The inner acrosome membrane never fuses with the egg plasma membrane. This membrane is incorporated into the egg cytoplasm in a phagocytic fashion (Fig, 7F, G) and is destined to degenerate within the cytoplasm (Yanagimachi and Noda, 1970a). Figure 8A-C shows golden hamster spermatozoa fusing with eggs. 111.

SPECIFIC SITES FOR SPERM-EGG FUSION

A. Sperm Membrane

In many animals (e.g., sea urchins, frogs, birds, and perhaps sturgeons and opossums), it is the acrosomal membrane that fuses first with the egg plasma membrane. In these animals, the sperm plasma membrane is not capable of initiating fusion with the egg plasma membrane. The fusible acrosomal membrane appears to be protected by the nonfusible plasma membrane until the spermatozoon comes close to the egg (Epel and Vacquier, 1978). In some animals, the acrosomal membrane elongates during the acrosome reaction to become the surface membrane of the acrosomal process (Figs. 1 and 3A, B). It is this surface membrane (the elongated inner acrosomal membrane) that fuses with the egg plasma membrane. It is not known whether the acrosomal membrane can undergo fusion before it elongates. The spermatozoa of some other animals have acrosomal processes even before they undergo the acrosome reaction. The inner acrosomal membrane, which is destined to fuse with the egg plasma membrane, does not elongate during the acrosome reaction (Figs. 3C, D, 5 , and 6). It is not known at present whether the membrane is capable of fusion before the acrosome reaction. Mammalian spermatozoa (with exceptions of marsupialian and perhaps monotreme spermatozoa) fuse with eggs via the head plasma membrane. The inner acrosomal membrane lacks the ability to fuse with the egg plasma membrane (for the characteristics and possible functions of the inner acrosomal membrane of mammalian spermatozoa, readers are referred to Huang and Yanagimachi, 1985). Yanagimachi and Noda (1970b), who examined fusion between zona-free hamster eggs and acrosome-reacted spermatozoa, concluded that the plasma membrane of the postacrosomal region is the first to fuse with the egg plasma membrane. However, careful examination of spermatozoa in the act of fusing with zona-intact eggs has revealed that it is the plasma membrane above the

FIG.8. Scanning electron micrographs of hamster spermatozoa fusing with eggs. [Courtesy of Dr. David M . Phillips, with permission of Plenum Press (A) and Academic Press (B, C). A is from Yanagimachi (1981); B, C are from Shalgi and Phillips (1980).]

15

1. SPERM-EGG FUSION

equatorial segment of the acrosome which fuses first with the egg plasma membrane (Bedford and Cooper, 1978: Bedford et d..1979). It is important to note that the plasma membrane above the equatorial segment can fuse with the egg plasma membrane only after the spermatozoa have undergone the acrosome reaction. The acrosome-intact spermatozoon, regardless of whether capacitated or not, cannot fuse with the egg, even when the zona pellucida is bypassed and the spermatozoon is brought directly in contact with the egg surface (Yanagimachi and Noda, 1970b). This suggests that the plasma membrane over the equatorial segment must undergo change at the same time as (or as the result of) the acrosome reaction (Fig. 9). The possibility that acrosomal material and hydrolytic enzymes released during the acrosome reaction can alter the property of the plasma membrane above the equatorial segment (hypothesis I of Fig. 9B) is unlikely, because acrosome-intact hamster spermatozoa (whether capacitated or not), when treated with crude acrosome extracts, are unable to fuse with zona-free hamster eggs (R. Yanagimachi, unpublished data). Possibly some components of the medium (e.g., monovalent and/or divalent cations) can penetrate the spermatozoon during the acrosome reaction and thereby alter the characteristics of the plasma membrane over the equatorial segment (11 of Fig. 9B). An influx of Ca’’ into acrosome-reactingheacted spermatozoa (Monroy, 1985)could be one of the factors that enables the sperm membrane to fuse with the egg plasma membrane. According to McGrath and Hillman (1980). the spermatozoa of mutant ( r L x l t L y ) mice are fully capable of undergoing the acrosome reaction and of attaching to the egg plasma membrane, yet are totally incapable of fusing with the egg plasma membrane. In this situation, either the sperm plasma membrane above the equatorial segment or the mechanism that enables this membrane to undergo fusion must be defective.

4 / *.\., I l ,. l .\

A

B

C

FIG.9. Spermatozoa before ( A ) , during (B).and after (C) the acrosome reaction. H o w does the plasma membrane over the equatorial segment become capable of fusing with the egg after the acrosome reaction‘! Hypothesis I: Some materials released from the acrosome modify the characteristics of the membrane. Hypothesis 11: Some components of the medium penetrate into the spermatozoon and modify the characteristics of the membrane directly or indirectly. ac, Acrosomal cap region; eq, equatorial segment of the acrosome.

16

RYUZO YANAGIMACHI

Freeze-fracture images of the acrosomal membrane of sea urchin spermatozoa and of the plasma membrane over the equatorial segment of guinea pig spermatozoa, before and after the acrosome reaction, are shown in Fig. 10A-D. Prior to the acrosome reaction, intramembranous particles are seen in the posterior region of the acrosomal membrane of

FIG.10. Freeze-fracture replicas of sperm membranes before and after the acrosome reaction. (A) E-face of the acrosomal membrane of an acrosome-intact sea urchin spermatozoon; note that intramembranous particles are rich in the posterior region of the acrosomal granule (ag). ( B ) The P-face of the acrosomal membrane which covers the acrosomal process (ap) has very few particles. (C)The P-face of the plasma membrane of an acrosome-intact. uncapacitated guinea pig spermatozoon has evenly distributed intramembranous particles. (D)The P-face of the membranes of an acrosome-reacted spermatozoon; note the presence of small, particle-free patches (white asterisks) in the plasma membrane of the equatorial segment. [Courtesy of Drs. Noriko Usui and lchiro Takahashi (A, B) and Fumie Suzuki (C. D).]Abbreviations: acr, plasma membrane of the acrosomal cap region: ag, acrosomal granule; ap, acrosornal process; eq, plasma membrane of the equatorial segment region: iam, inner acrosomal membrane: par, plasma membrane of the postacrosomal region.

1. SPERM-EGG FUSION

17

the sea urchin spermatozoon: the anterior region of the membrane is nearly free of the particles (Fig. IOA). The elongated acrosomal membrane covering the surface of the acrosomal process has sparsely distributed particles (Fig. IOB). According to Mann or ul. (1976), the membrane covering the tip of the acrosomal process, which is destined to fuse with the egg plasma membrane, is virtually free of the particles. In the guinea pig. particle-free patches can be seen in the plasma membrane of the equatorial region of an acrosome-reacted spermatozoon (Fig. lOD), but not in that of an acrosome-intact, uncapacitated spermatozoon (Fig. IOC). These images suggest drastic changes in the physicochemical characteristics of the membrane phospholipids (and perhaps of the cytoskeletal system under the membrane) associated with the acrosome reaction.

6 . Egg Plasma Membrane

In fish, the egg plasma membrane that normally fuses with the spermatozoon is limited to the area of the membrane beneath the micropyle (cf. Fig. 4). However, the plasma membrane in other areas of the egg seems to have the potential to fuse with spermatozoa (Fig. 1 I). In the sea urchin and mammal, the entire area of the egg plasma membrane seems to have the potential to fuse with the spermatozoon (for the mammal, see Blandau and Odor, 1952). In the sea urchin, the spermatozoon fuses with the plasma membrane of small microvilli (papillae; Fig. I). However, there is no evidence that the intermicrovillous region lacks the ability to fuse with the spermatozoon, and some investigators believe that the spermatozoon fuses with the egg plasma membrane more commonly in the intermicrovillous than in the microvillar region (N. Usui, personal communication). In mammals, it is not yet clear whether the spermatozoon fuses with the egg plasma membrane of microvilli or of the intermicrovillar region (for discussion, see Yanagimachi, 1981). The mammalian egg plasma membrane above the meiotic spindle has no microvilli. It is either "smooth," or provided with large cytoplasmic protrusions. Sperm-egg fusion does not occur o r seldom occurs in this region (Johnson el ul., 1975; Ebensperger and Barros, 1984). The reason why this region lacks the ability to fuse with spermatozoa is not clear. According to Wolf and Ziomek (1983), membrane proteins move (diffuse) more freely in this than in other regions. In the frog (Elinson, 1975) and bird (F. Okamura, personal communication), only the egg plasma membrane of the animal hemisphere (frog) o r near the animal pole (bird) appears to have the ability to fuse with the spermatozoon(oa). In the hydrozoa, the region that can undergo fusion

18

RYUZO YANAGIMACHI

seems to be limited to a very small portion of the animal pole where the second polar body was extruded (in this animal, meiosis of the egg is completed before fertilization) (Miller, 1978; Freeman and Miller, 1982). It has been known for more than 50 years that immature sea urchin eggs in the germinal vesicle stage can be penetrated by spermatozoa (Seifriz, 1927). In mammals, the plasma membrane of the immature egg at the germinal vesicle stage has the potential to fuse with spermatozoa, at least at the time when the egg is about to resume meiosis (e.g., Usui and Yanagimachi, 1976). However, it is not known when during egg growth the plasma membrane gains the ability to fuse with spermatozoa.

IV. SPERM MOVEMENT AND SPERM-EGG FUSION

Sperm-egg fusion as such does not appear to require the movement of the spermatozoon. According to Epel et al. (1977), 10 mM sodium azide completely immobilizes the spermatozoa of Lytechinus pictus (a Pacific sea urchin). Immotile spermatozoa can, nevertheless, fertilize eggs. Epel et al. (1977) did not remove the vitelline envelopes from eggs prior to insemination. Immotile spermatozoa, that were carried to the egg surface by currents, must have undergone the acrosome reaction on or near the vitelline envelope; they then must have attached to and dissolved the vitelline envelope before fusing with the egg plasma membrane. It has been shown that tailless (and therefore motionless) sea urchin spermatozoa can fertilize eggs (Vacquier, 1979a). Herring spermatozoa are almost motionless in seawater. When the spermatozoa drift to the surface of the chorion near the micropyle, they become very active and enter the micropyle one by one (Yanagimachi and Kanoh, 1953). The first spermatozoon then fuses with the egg plasma membrane. When an egg is partially denuded and then inseminated, the motionless spermatozoa can be seen to make contact with the “naked” portion of the egg (Fig. IIA) and apparently fuse with the egg membrane (see the legend of Fig. 1 I). Strong sperm movement is required for normal fertilization in mammals. Motionless or weakly motile spermatozoa are unable to pass through egg investments, particularly the zona pellucida. Once the zona is removed, however, motionless or very weakly motile spermatozoa can fuse with the eggs as long as the spermatozoa are alive and have undergone the acrosome reaction (Yanagimachi, 1981). The spermatozoa of humans with Kargagener’s syndrome present an interesting example. They are “normal,” except that they are completely motionless due to the lack of dynein arms (Afzelius et al., 1975). In a medium that allows

19

1. SPERM-EGG FUSION

A

B

C

FIG. I I . Fertilization of a partially denuded herring egg. 'l'he chorion was piirtially removed. at 4 o'clock, with the aid of a pair of iridectoniy scicsorh. By applying gentle preysure. a portion of the egg was squeeked otit of the chorion. When such ?in egg wus inseminated (A). motionless spermaloma were seen to become activated (began to move actively) near the micropyle and entered the micropyle one by one: motionless sperrniiforoa were seen attached to the naked portion of the egg ( A ) . l'he egg was activated by spermatozoa ( 8 ) and cleaved (0.The main body of the egg was fertilized monosperrnically. The protruded portion of the egg, on the other hand. wi15 fertilized polysperniically. (From Y anagimac hi. 19.57.)

capacitation and the acrosome reaction, these spermatozoa fuse with zona-free hamster eggs (Aitken et a/.. 1983). As stated already, the fertilizing spermatozoon of the sea urchin abruptly ceases its active tail movement soon after attachment to the egg surface (e.g., at 16 sec after attachment: Epel ot al., 1977). The cessation of movement is thought to occur some time after sperm-egg fusion (Epel et a / . . 1977). Hinkley el a/. (1986) have estimated that sperm immobilization occurs 4-6 sec after fusion. The fertilizing spermatozoa of mammals also cease their active tail movement almost immediately on contact with the egg plasma membrane (Yanagimachi, 1966; Sato and Blandau, 1979: Gaddum-Rosse, 1985). This can be witnessed easily by inseminating zona-free eggs in vitvo with actively motile. acrosome-reacted spermatozoa (Fig. 12). The reason for this abrupt loss of sperm tail movement is unknown, but it could be caused by an explosive release of intracellular CaITby the egg as observed by Miyazaki rt d.(1986). Although the sperm tail may display sporadic twitching or flexing for some time after the sperm head has become attached to the egg plasma membrane, active tail movement is apparently unnecessary for the migration of the sperm head (nucleus) to a position deep inside the egg.

V. INTERMINGLING OF SPERM AND EGG PLASMA MEMBRANE The sperm plasma membrane intermingles with the egg plasma membrane during fusion. This intermingling can be visualized, for example, by labeling sperm surfaces with FITC or FITC-conjugated antisperm surface

20

RYUZO YANAGIMACHI

FIG. 12. Acrosome-reacted hamster spermatozoa lose their motility soon after contact with the egg plasma membrane. An unfertilized, zona-free egg is approached by an acrosome-reacted spermatozoon (A). The spermatozoon beats its tail vigorously for a while after contact with the egg plasma membrane (B);5-15 sec after contact (C) its beat becomes By 5 min after considerably lower until it becomes motionless 15-25 sec after contact (D). insemination, the egg has many motionless spermatozoa attached to its surface (E).(From Yanagimachi, 1978a. with permission of Academic Press.)

antibodies and examining the surfaces of eggs fertilized by these spermatozoa (Gabel et al., 1979; Gundersen and Shapiro, 1984; Gundersen et al., 1986). Gaunt (1983) produced a monoclonal antibody that binds to the rat

sperm surface antigen (2D6) which is located over the entire surface of the spermatozoon (Fig. 13A, A’). The antigen to the sperm surface spreads gradually (Fig. 13B, B’) until it covers most of the surface of the fertilized egg (Fig. 13C, C’). Only traces of the antigen are detected in the two-cell embryo (Fig. 13D, D’), perhaps because the antigen has become internalized. The antigen, 2D6, is just one of the sperm surface molecules. Other molecules may also persist in the embryo for a long time. In fact, some surface proteins of sea urchin spermatozoa can be detected as late as the late gastrula stage; however, most sperm surface proteins become degraded shortly after fertilization (Gundersen and Shapiro, 1984). VI. SPECIES SPECIFICITY OF SPERM-EGG FUSION A distinct feature of fertilization is its species specificity. Generally speaking, spermatozoa of a given species cannot, or cannot readily, fertil-

FIG. 13. Fate of a rat sperm surface anigen. 2D6. incorporated into the egg plasma membrane (A-D, pha5e contrast; A'-D'. indirect immunofluorescence). ( A ) Spermatomi from the epididymal cauda. ( B ) One recently fertilized egg (center) with two unferlilired eggs. collected from immature females 14 hr after hCG injection and mating. (C) A n egg in the advanced stage of fertilization. 20 hr after hCG injection and mating. (D) A two-cell stage embryo, 40 hr after hCG injection and mating. (Courtesy of Dr. Stephen J. Gaunt. with permission of the Company of Biologists. Ltd.)

22

RYUZO YANAGIMACHI

ize the eggs of other species, although cross fertilization does occur on occasion even between gametes of distant species. An example is the cross-fertilization between sea urchin eggs and mussel or oyster spermatozoa (Longo, 1977; Osanai and Kyozuka, 1982). One of the major barriers for cross-fertilization appears to reside in the vitelline envelope. The spermatozoa of foreign (heterologous) species often fail to undergo the acrosome reaction near or on the envelope. Even if they do, they fail to bind or penetrate the envelope (Summers and Hylander, 1975). The removal of the vitelline coat is known to facilitate fertiliization (sperm-egg fusion) among different species of the sea urchin (Hultin, 1948; Jensen, 1953; Kyozuka and Osanai, 1979). The spermatozoa of most teleostean fishes do not need to undergo the acrosome reaction, as they have no acrosomes; rather their head plasma membrane is ready to fuse with the “naked” egg plasma membrane located at the bottom of the micropyle (cf. Fig. 4). Therefore, as long as the diameter of the bottom end of the micropyle is large enough to allow the passage of one spermatozoon, the chance of cross-fertilization is high. In fact, hybridization in the wild is not very rare, and cross-fertilization in the laboratory is relatively easy (Moenkhaus, 1910; Hubbs, 1955; Kobayashi, 1963; Suzuki, 1964; Nikoljuki, 1965; Ohta, 1985). In mammals, sperm penetration through the zona pellucida and spermegg fusion may occur following cross-insemination between closely related species (cf. Maddock and Dawson, 1974; Roldan e t a l . , 1985). In the case of distant species, spermatozoa usually fail to attach to (or penetrate into) the zona pellucida (Yanagimachi, 1977). Sperm-egg fusion appears to be less species specific than sperm-zona interaction. For instance, mouse spermatozoa, which seldom penetrate the zonae of rat eggs, can readily fuse with zona-free rat eggs (Hanada and Chang, 1972; Pavlok, 1979). Similarly, human spermatozoa, which cannot penetrate the zonae of golden hamster eggs, can fuse with zona-free hamster eggs (Fig. 14; Yanagimachi et al., 1976). These facts, however, should not be taken as implying that the egg plasma membrane is totally lacking in species specificity. The plasma membrane of the mouse, for example, manifests strong specificity and permits the fusion of only mouse spermatozoa (Table I). Although the hamster egg plasma membrane is unique in that it allows fusion of a variety of foreign spermatozoa (Table I), it does show some degree of specificity. For instance, when zona-free hamster eggs are inseminated in a dish containing approximately equal numbers of acrosome-reacted spermatozoa from hamster, guinea pig, and humans, the number of hamster spermatozoa that can later be found within each egg is greater than the number of guinea pig or human spermatozoa (Yanagimachi, 1981). Therefore the affinity of the hamster egg plasma membrane must be greatest

Fiti. 14. Human spermatozoa fused with tona-free hamster egg\. ( A ) Semilongitudinal section through the head of a human spermalozoon undergoing fusion with an egg. (I31 Phase-contrast micrograph of an egg fused with many spermatoroa. about I hr after inseminetion with preincubated (acrosome-reacted) spermatoroa: the egg was compreswd between a slide and coverslip before photography.

24

RYUZO YANAGIMACHI

TABLE 1 PENETRATION O F ZONA-FREE MAMMALIAN EGGS B Y SPERMATOZOA OF HOMOLOGOUS AND HETEROLOGOUS SPECIES" Zona-free eggs Spermatozoa

Golden hamster

Chinese hamster

Golden hamster Chinese hamster Mouse Deer mouse Rat Guinea pig Bat Rabbit Dog Dolphin Pig Bull Sheep Goat Horse Marmoset monkey Cynomolgus monkey Rhesus monkey Bonnet monkey Human

Yes Yesh Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yesf Yes Yes Yes Yes" Yes' Yesf Yes

Yesh Yes

-

-

No

Guinea Mouse

Rat

pig

Rabbit

YesINo Yes No YeslNo No

YesINo

Yes

No Yes Yes N0 Yes

-

-

-

No

-

Yes No Yes No

No

-

Yes

-

No

-

-

From Yanagimachi (1984). except where indicated. Roldan and Yanagimachi (unpublished data). ' Pavlok t i / . (1983); Flechon and Pavlok (1986). ci Hoffman and Curtis (19841. ' Bwatman and Bavister (1984). Warikoo ('I ctl. (1986). "

I,

for the spermatozoa of its own species. Nevertheless, it would be interesting to know what property allows hamster egg plasma membranes to fuse with spermatozoa from a wide variety of species. Clearly the structure of this membrane cannot be unique (Koehler et ul., 1985).

VII. HOW LONG DOES THE EGG PLASMA MEMBRANE REMAIN CAPABLE OF FUSING WITH SPERMATOZOA AFTER THE ENTRY OF THE FIRST SPERMATOZOON?

In some animals (e.g., elasrnobranchs, urodeles, reptiles, and birds), many spermatozoa penetrate a single egg, and their nuclei transform into

1. SPERM-EGG FUSION

25

male pronuclci. However, only one of these pronuclei unites with the female pronucleus. In these animals, the egg plasma membrane must retain the capacity to fuse with spermatozoa for an extended time period after fusion with the first spermatozoon. In the sea urchin, only one spermatozoon fuses with the egg plasma membrane. under normal conditions. All other spermatozoa are prevented from fusing with, or reaching, the egg plasma membrane. The mechanisms responsible for this block are a rapid. transient change in the membrane potential and a relatively slow, but permanent, chemical change that the vitelline envelope undergoes as a result of the action of cortical granule material (Schuel. 1984; Jaffe and Could, 1985). Both of these changes are triggered when the first spermatozoon fuses with the egg. A fertilized sea urchin egg does not allow the entry of excess spermatozoa. even if the vitelline membrane is removed and the egg is reinseminated with large numbers of spermatozoa. Does the egg plasma membrane permanently lose its ability to fuse once fertilization is complete? Sugiyama (1951) was the first to demonstrate that the egg plasma membrane remains potentially capable of fusing with spermatozoa even after the egg has reached the two-cell stage. Fertilized eggs freed from their vitelline envelopes and treated with Ca”- and Mg”-free seawater permit the entry of many spermatozoa (Sugiyama, 1951; Tyler r t al., 1956). This treatment dissolves the hyaline layer that covers the plasma membrane of the fertilized egg. Therefore, it must be the hyaline layer that renders the plasma membrane resistant to further fusion with spermatozoa. In some mammals (e.g., the rabbit), the plasma membrane of the egg becomes “refractile” to excess spermatozoa on activation by the first (fertilizing) spermatozoon (Austin and Walton, 1960; Austin, 1961). The cortical granule material that is exocytosed during activation of the egg is thought to alter the surface properties of the egg plasma membrane (Cooper and Bedford, 1971; Gordon ot d.,1975). Cortical granule material apparently does not change the properties of the zona pellucida of rabbit egg. Spermatozoa continue to penetrate the perivitelline space, often in large numbers, until a mucin coat is deposited around the zona (Austin and Walton, 1960). In some other mammals, as the golden hamster o r the mouse, the proteolytic component of the cortical granule material alters the chemical properties of zona glycoproteins in a way that prevents surplus spermatozoa from binding to, or penetrating, the zona (Gwatkin rt [ J I . , 1973; Wassarman c>t d., 1985). In the mouse, cortical granule proteinases render the egg plasma membrane incapable of fusing with excess spermatozoa (Wolf and Hamada. 1977). The plasma membrane of the hamster egg apparently remains “unaffected” by cortical granule proteinases during fertilization. It can still fuse with spermatozoa even as late as in the four-cell stage (Usui and Yanagimachi, 1976). If an

26

RYUZO YANAGIMACHI

unfertilized zona-free hamster egg is inseminated with acrosome-reacted spermatozoa, the spermatozoa can be seen t o fuse one after the other with the egg (Fig. 12). If the number of spermatozoa surrounding the egg is not limited, an excessively large number of spermatozoa will fuse with the egg. This results inevitably in cytolysis.

VIII.

CONDITIONS CONTROLLING SPERM-EGG FUSION

In experiments that aim at analyzing the mechanism of sperm-egg fusion, it is important to distinguish prefusion events from fusion itself. Consider the following experiment: Eggs with intact investments are washed and inseminated in a medium deficient in ion A and, when examined later, none is found to have been fertilized. It would be hasty to conclude that sperm-egg fusion was inhibited by the absence of ion A. The failure of the eggs to become fertilized might well have been due to the failure of spermatozoa to undergo the acrosome reaction o r to penetrate the egg investment. Sea urchin and mammalian spermatozoa, for example, will never fuse with the egg plasma membrane unless they have undergone the acrosome reaction and have penetrated the egg investments. In the case of mammals, it is advisable to free unfertilized eggs from both the cumulus oophorus and the zona pellucida before inseminating them in uitro with lively, motile spermatozoa that have undergone the acrosome reaction. Under favorable conditions, spermatozoa will fuse with eggs “instantaneously” and “synchronously.” Hinkley et a / . (1986) have reported an ingenious method to distinguish fused from unfused sea urchin spermatozoa. Sea urchin eggs are stained vitally with the DNA-specific fluorochrome Hoechst 33342 and then inseminated with unstained spermotozoa. When fusion occurs, the fluorochrome enters the spermatozoon and stains its DNA, resulting in the appearnance of a brightly fluorescent sperm head on the egg surface. The heads of unfused spermatozoa remain nonfluorescent. This simple method, which allows us to differentiate fused spermatozoa from unfused ones, without laborious electron microscopy, will be very useful in future studies of sperm-egg fusion. A. Temperature, pH, and Ionic Composition of Medium

In the golden hamster, sperm-egg fusion cannot occur at 10°C or below (Hirao and Yanagimachi, 1978a). When zona-free eggs were inseminated in uitro with acrosome-reacted spermatozoa at I0”C or below, motionless spermatozoa were seen attached to egg surfaces. There was no indication

1. SPERM-EGG FUSION

27

of sperm-egg fusion even 17-18 hr after insemination. On transfer to warm (37°C) medium, some of the spermatozoa on the egg surface became motile and sperm-egg fusion took place immediately. Interestingly, postfusion events (e.g., cortical granule exocytosis, sperm nucleus decondensation, resumption of egg meiosis, and pronuclear development) are less sensitive to low temperatures. Once sperm-egg fusion is complete, these events can proceed even at 4°C. Acidic pH of the medium reversibly inhibits sperm-egg fusion (Yanagimachi et ul., 1980). At pH 6.0, acrosome-reacted hamster spermatozoa may bind to the egg plasma membrane. but they fail to fuse with it (Fig. ISA). Fusion takes place upon transfer of eggs to more alkaline (pH 7.3) medium (Fig. ISB). Extracellular Ca2+ is essential for the fusion of mammalian (hamster, guinea pig, and human) spermatozoa with eggs (Yanagimachi, 1978b,

FIG. 15. Electron micrographs showing that acrosome-reacted hamster spermdtoioa are incapable of fusing with the plasma membrane of rona-free hamster eggs in an acid medium ( p H 6.0) ( A ) . but can readily d o so in a more alkaline medium ( p H 7 . 4 ) ( B ) . (From Yanaginiachi 6" ( I / . . 19x0. with permission of the Japanehe Society of Developmental 13iologi\t\.)

28

RYUZO YANAGIMACHI

1982). In Ca’+-deficient media, acrosome-reacted spermatozoa may attach to the egg plasma membrane, but fusion never occurs. Fusion takes place upon transfer of eggs (with attached spermatozoa) to a Ca?+-containing medium. Mg2+,Ba*+,and Sr2+can substitute for Ca?+in triggering sperm-egg fusion, although they are less effective than Ca2+(Fig. 16). In some biological systems, Mg’+ may act antagonistically to Ca?+,but in sperm-egg fusion these cations act synergistically (Yanagimachi, 1978b). Takahashi and Sugiyama (1973), Epel (1982), and Schmidt et a / . (1982) have reported that acrosome-reacted sea urchin spermatozoa can fuse with eggs in the absence of extracellular Ca2+.According to Sano and Kanatani (1980), however, the fusion cannot take place in the complete absence of Ca2+. Unlike the acrosome reaction, which requires a high concentration (5-10 mM) of extracellular Ca2+(Dan, 1954; Collins and Epel, 1977), sperm-egg fusion requires a far lower concentration of Ca’+ (Sano and Kanatani, 1980). Concentrations of Ca2+from 20 to 50 F M are enough to ensure sperm-egg fusion in 25-50% of the eggs (Fig. 17).

Cone. lmMl

FIG. 16. Relative effectiveness of various divalent cations in supporting sperm-egg fusion in the hamster. Zona-free hamster eggs were washed and inseminated with acrosomereacted spermatozoa in Ca”- and Mg’+-free media containing various concentrations of divalent cations (Ca”. Ba”, Sr2+,or Mg’+). They were examined for evidence of fertilization (sperm-egg fusion) 3-3.5 hr later. (From Yanagimachi. 1978b with permission of the Society for the Study of Reproduction.)

29

1. SPERM-EGG FUSION

In mammals, the presence of extracellular Mg?- is beneficial, but not essential, for sperm-egg fusion (Yanagimachi, 197%). In sea urchins, on the other hand, Mg?+ appears to play a significant role in sperm-egg fusion. Unless seawater contains millimolar Mg?', acrosome-reacted spermatozoa rarely fuse with eggs even in the presence of 10 mM Ca2+ (Sano ct (11.. 1980: Mohri cr ul., 1982). Extracellular K' does not appear to be essential for sperm-egg fusion in the mouse, although its presence enhances fusion (Fraser, 19x3). Guinea pig spermatozoa differ from mouse spermatozoa in that exposure to extracellular K' is essential for development of the capacity for fusion. Guinea pig spermatozoa which have been acrosome-reacted in K ' -free media are unable to fuse with the egg plasma membrane unless the insemination medium contains K' (Rogers r t (11.. 1981). Paradoxically, the presence of K' in the insemination media is not essential. Spermatozoa exposed to K ' only during (and shortly after) the acrosome reaction are capable of fusing with eggs in K+-free medium (Table 11). It thus appears that K ' triggers some important physiological change in the plasma membrane of the equatorial segment during or shortly after the acrosome reaction (Fig. 9). Once this change has taken place. extracellular K is no longer needed for sperm-egg fusion. +

loo] 80

0 0

10

20

50 Calcium

300pM

I

!

1

1

2

5

1

lOmM

COnC

Effect of extracellular Ca" concentration on fertiliration of sea urchin eggs by acrosome-reacted spermatozoa. Unfertilized egg\ of ~ ~ , / t t ; ~ , ~ , / 7pirlc r ~ ~hrrrirrrrr ~ 1 ~ / , 5~ were washed with Ca'-- and Mg?'-free seawater and inseminated with acrosome-reacted \perniatozoa in artificial seawater that cont:iined 3X.5 m M Mg?- and variou\ concentrations of CaL-, The percentage fertilization w a s determined 30-60 min after insemination by ohserving the formation of a fertilization envelope (elevated vitelline envelope) and normal cleavage. The value at 0 mM calcium represents that obtained in the Ca?--free medium conraining 1 m M EGTA. (From Sano and Kanatani. 19x0. with permission of Academic Press.) FIG. 17.

30

RYUZO YANAGIMACHI

TABLE 11

EFFECTOF T H E PRESENCE O R ABSENCE OF EXTRACELLULAR K' O N T H E FUSION OF G U I N E A PIC S P E R M A T O Z O A W I T H ZONA-FREE HAMSTEREGGS" Presence

(+)

Capacitation

+ -

-

+

or absence ( - ) of 2.7 mM K+ during Acrosome reaction

+ + +

% Eggs fused

Insemination

+ + +

-

-

-

-

with sperm (range) I00 100

I00 7 2 (44-100) 0 0

Mean number of sperm in each egg fused with sperm 9.2 6.8 6.6 1.8

" From R . Yanagimachi (unpublished data). Guinea pig yxrmatozoa were capacitated by a I-hr incubation in a Ca?+-free m T medium supplemented with 8S pgiml lysolecithin (Fleming and Yanagimachi. 1981). They were washed with albumin-saline \ohtion (0.9% NaCl with 0.1% bovine serum albumin) before transfer to the regular m T medium that contained 2 mM Ca". The majority o f spermatozoa underwent the acrosome reaction within I S min regardless o f the presence or absence o f K ' . Thirty minutes after exposure to Ca?'. the spermatozoa were washed again with the albumin-saline Solution and then mixed with zona-free hamster eggs in Ca?'-containing media with o r without K L .Sperm-egg fusion. as evidenced by the presence of swollen sperm heads or sperm pronuclei in egg cytoplasm. wa\ a\hessed I hr later.

6. Electric Potential of the Egg Plasma Membrane

Jaffe (1976) was the first to report the importance of the electric potential of the egg plasma membrane for sperm-egg fusion in the sea urchin (for reviews, see Gould-Somero and Jaffe, 1984; Jaffe and Gould, 1985; Jaffe and Cross, 1986). The resting membrane potential of the sea urchin (Strongylocentrotus) egg is about -70 mV. Within 3-30 sec after insemination, the egg membrane depolarizes to a plateau at -30 to +20 mV, and about I min later begins to repolarize (Fig. 18). This positive potential shift, triggered by the fusion of the fertilizing spermatozoon with the egg plasma membrane, prevents excess spermatozoa from fusing with the egg plasma membrane during the first minute of fertilization. Permanent mechanical barriers to excess spermatozoa are established by ( I ) transformation of the vitelline envelope to the fertilization envelope and ( 2 ) coating of the plasma membrane surface with the hyaline layer. If the membrane potential of the unfertilized egg is held negative, by applying current, during insemination, polyspermy results. On the other hand, if the potential is held positive, no fertilization takes place. Apparently, the negative resting potential of the egg plasma membrane greatly favors sperm-egg fusion. This seems to be true for other animals such as the starfish, echiuran worm (Urechis),and frog (Jaffe and Gould, 1985). The resting membrane potentials of mouse and hamster eggs are about

1. SPERM-EGG FUSION

31

FIG.IR. Fertilization potential in the w a urchin. .Yrro/i,v,vlol~ ~ n i r o r upsu r p u r o m . (From Jaffe. 1976. with Demission of McMillan Journals. Ltd.)

-40 and -30 mV. respectively. Instead of a positive shift, only either a small oscillation (mouse) or hyperpolarization (hamster) of the membrane potential has been recorded following fusion of sperm with eggs (Miyazaki and Igusa, 1985; Jaffe et d., 1983). The absence of a prominent positive shift in membrane potential in these rodent eggs is not surprising. To inhibit polyspermic fertilization these eggs rely almost exclusively on chemical modification of the zona pellucida (called the zona reaction), rather than on a change in the plasma membrane (cf. Austin and Walton, 1960: Yanagimachi, 1977). Circumstantial evidence strongly suggests that the plasma membrane of rabbit eggs becomes “refractile” to penetration by excess spermatozoa soon after fusion with the first spermatozoon. Some physical or chemical change must take place rapidly in or on the plasma membrane. In this regard. it is somewhat disappointing to find that no dramatic shift in membrane potential has been recorded following insemination of rabbit eggs in uitro (McCulloh ct i i / . , 1983). It is possible that the polyspermy block operating in the egg plasma membrane is not electrically mediated. In the rabbit, excess spermatozoa keep penetrating into the perivitelline space and continue to collide with the plasma membrane of the fertilized egg for many hours. The polyspermy block of the egg plasma membrane appears to be permanent, rather than temporary. Although changes in sialic acid and saccharide residues of the rabbit egg membrane glycocalyx have been implicated in the permanent polyspermy block (Cooper and Bedford, 1971; Gordon et d . , 1975). further studies are needed to elucidate the molecular basis of this block.

IX.

EFFECTS OF MISCELLANEOUS SUBSTANCES ON SPERM-EGG FUSION

A. Proteolytic Enzymes

The plasma membrane of the mouse egg appears to be sensitive to proteolytic enzymes. When unfertilized eggs are freed from zonae pellucidae by treatment with pronase or chymotrypsin, the resulting zona-free

32

RYUZO YANAGIMACHI

eggs are less able to undergo fusion with spermatozoa than are mechanically denuded eggs (Wolf et a / . , 1976; Quinn and Stanger, 1981; Boldt and Wolf, 1985). The egg plasma membrane of the golden hamster, on the other hand, is relatively insensitive to proteolytic enzymes. Even after a 30-min treatment of previously denuded eggs with trypsin or pronase (0.1% solution, at 25-37"C), all of the eggs remain capable of fusing with spermatozoa (Hirao and Yanagimachi, 1978b). This difference between mouse and hamster could be due to inherent differences in the sensitivity of the egg plasma membrane to the cortical granule proteinase (Yanagimachi, 1981). B. Antimembrane Antibodies

Certain antibodies raised against sperm surface components inhibit sperm-egg fusion in the hamster (Tzartos, 1979; Oikawa and Suzuki, 19791, guinea pig (Huang et ul., 1981; Yanagimachi et d., 1981; Primakoff et al., 1985), and mouse (Saling and O'Rand, 1982). Those which block fertilization by inhibiting prefusion events (e.g., sperm capacitation, acrosome reaction, and interaction with egg investments) are not considered here. According to Saling et a/.(1983, 1985), a monoclonal antibody, M29, raised against epididymal mouse spermatozoa binds specifically to the equatorial segment of the mouse acrosome (presumably with the plasma membrane over the equatorial segment). In the presence of this antibody (0.2 mg purified IgM/ml), most spermatozoa are unable to fuse with the egg plasma membrane, although they can firmly attach (bind) to it. According to Matsuda et al. (1985), three monoclonal antibodies raised against hamster eggs bing to the egg plasma membrane and impair spermegg fusion. C. Proteinase Inhibitors

In the sea urchin, the presence of proteinase inhibitors in seawater does not inhibit sperm-egg fusion. However, it renders fertilization polyspermic by disturbing cortical granule exocytosis of the egg (Schuel et al., 1976; Schuel, 1984). Sperm-egg fusion in mammals is certainly possible in the presence of trypsin inhibitors (Saling, 1981; Yanagimachi, 1981; Fraser, 1982), but some investigators have reported that proteinase inhibitors reduce the incidence of successful fusion of acrosome-reacted spermatozoa with the egg plasma membrane (Wolf, 1977; Dravland and Meizel, 1982). According to Van der Ven et al. (1985), fusion of human spermatozoa with zona-free hamster eggs is inhibited by serine proteinase inhibitors. This is probably due, at least in part, to the inhibition of the acrosome reaction by these inhibitors.

1. SPERM-EGG FUSION

33

D. Dithiothreitol

Dithiothreitol (DTT) is a disulfide-reducing and sulfhydryl-maintaining reagent. Yanagimachi o r rrl. (1983) reported that acrosome-reacted guinea pig spermatozoa are unable to bind to the egg plasma membrane in the presence of I mM DTT. Spermatozoa collide with the egg plasma membrane repeatedly. but do not bind to i t . A few spermatozoa may remain on the egg plasma membrane for some time (up to I min), but they eventually detach and swim away. Consequently, none of the eggs are fertilized as long a s DTT is present in the medium. Reduced glutathione and cysteine act similarly, but they are less effective in this respect than DTT. E. Erythrosine and Related Dyes

According to Carroll and Levitan ( 1978a,b),micromolar concentrations oferythrosine B inhibit fertilization in a wide variety of both invertebrates ( e . g . . sea urchins and annelids) and vertebrates (fishes and frogs). These authors surmise that erythrosine B and related dyes prevent sperm-egg fusion by altering the electrostatic properties of the gamete membranes. Although Carrol and Wolf (1979) reported that sperm-egg fusion in the mouse is inhibited by high concentrations of erythrosine B (0.5-1 mM), Yanagimachi (198 I ) has been unable to confirm this in the hamster-. F. Cytochalasin

According to Longo (1978), sea urchin eggs treated with cytochalasin B at concentrations of 1-10 pg/ml (2-20 p M ) are activated by spermatozoa but fail to incorporate sperm heads into the cytoplasm. It is reasonable to assume that sperm-egg fusion has taken place, but, because the cytoskeletal (actin) system has become disorganized, the fertilization cone has failed to form, so that the sperm head is not taken up (Vacquier, 197%). Mouse (Longo, 1978) and hamster eggs (Yanagimachi, 1981) treated with cytochalasin B or D remain capable of fusing with spermatozoa. Perhaps the cytoskeletal system is not directly involved in the sperm-egg fusion per se. It must be pre- or postfusion events that are sensitive to reagents like cyochalasin. G. Other Reagents Other reagents reported to inhibit sperm-egg fusion in mammals include inhibitors of glycoprotein biosynthesis, amphomycin, UDP-glucose, and 2-deoxyglucose (Ahuja, 1985), Ouabain (Talansky e l (il., 1987), lysophosphatidylserine (Fleming and Yanagimachi, 19811, and the basic

34

RYUZO YANAGIMACHI

polycationic pol yamine, compound 48/80 (Fleming and Armstrong, 1985); La3+may possibly also inhibit sperm-egg fusion (Saling, 1982). Pretreatment of egg plasma membranes with plant lectins (Yanagimachi, 1981) or inclusion of Zn2+ in the insemination media (Aonuma et al., 1981) does not disturb sperm-egg fusion. X.

MECHANISM OF SPERM-EGG FUSION

Although purified phospholipid membranes are capable of fusing with one another, fusion of biological membranes appears to be regulated closely by a variety of membrane-associated proteins (Strittmatter et al., 1985). Among these are Ca2+-binding proteins and enzymes such as ATPases, proteinases, and phospholipases. Since we do not know much about the molecular organization of gamete membranes, a molecular model of sperm-egg fusion may be premature; nevertheless, a hypothesis will at least invite debate and stimulate study. For previously published discussions and hypothesis of sperm-egg fusion, readers are referred to Vacquier (1979b), Lop0 (1983), Gould-Somero and Jaffe (1984), Monroy (1985), and particularly Jaffe and Cross (1986). When examined by the freeze-fracture technique, the area of the sperm membrane about to fuse with the egg plasma membrane is free of intramembranous protein particles or has particle-free patches (Fig. 10). This does not mean that this membrane (or particle-free region thereof) is totally lacking in proteins. Surface glycoproteins associated with the outer leaflet of the membrane lipid bilayers are presumably present in this region. The fact that acrosome-reacted (guinea pig) spermatozoa are unable to bind to the egg plasma membrane in the presence of disulfidereducing agents (Yanagimachi et a / . , 1983) would suggest that the tertiary and quaternary structures of the surface proteins of gametes are of critical importance for a close and stable contact between sperm and egg membranes. Divalent cations, particularly Ca2+,aid in establishing close apposition of two membranes by neutralizing negative surface charges of the membranes. Some membrane proteins of the spermatozoon may be inserted in the quiescent egg plasma membrane, thereby perturbing membrane lipids (Fig. 19). Such proteins could be acrosomal bindin (Glabe, 1985) and lysin (Hong and Vacquier, 1986),which have fusogenic properties. They could be enzymes, such as phospholipase which acts on membrane phospholipids to produce membrane-perturbing compounds such as lysophospholipids and fatty acids. Successful insertion of fusogenic proteins may depend on the electric potential of the egg plasma membrane. It is possible that the insertion

35

1. SPERM-EGG FUSION

I +

=f--e -

FUSION

F I G , 19. A model of the mechanism leading to the fu\ion of sperm and egg mcnibrane\. (1) Sperm and egg membranes ( S and E) carry complementary molecules Ic and c ' ) . I n addition. the sperm membrane carries fuwgenic molecule\ tf). ( 1 1 ) I f the potential of fhe egg plasma membrane i s negative and the coniplcment;iry molecule\ of \perm and egg membrane\ fit with each other. the fusogenic \perm molecule\ are in\erted into the "quic\cent" egg plasma membrane. The membrane lipid perturbance caused by the sperm molecule\ and the phase separation o f the membrane lipid\ fhrough the action o l C a 2 - leads 10the furion of the two membranes.

occurs only when the potential is negative (Jaffe and Cross, 1986). Insertion of a fusogenic protein alone may not be enough to induce membrane fusion, however. Certain membrane phospholipids may have to be latcrally separated prior to fusion. Ca?' is known to play a key role in this process (Papahadjopoulos, 1978). In some membrane systems, Mg" is totally ineffective in causing phase separation but can still induce membrane fusion or facilitate fusion induced by Ca?+ (Duzgiine$. 198s). For sperm-egg fusion. at least in mammals, Ca?' and Mg" appear to work synergistically to support the fusion. AC K N O WLEDG M E N TS The preparation o f this chapter was .;upported by National Institutes of Health Grant HD03402. I am grateful to the following persons who provided valuable information about sperm-egg fusion in various animal species: Drs. J. Michael Bedford, Gary N . Cherr. Laurinda A . Jaffe. Chiaki Katagiri. Tadayuki Ohta. Fukashi Koyanagi (Okamoto). Kiyoshi Sano. Masao Sugiyama, Robert G. Summers. Norihiko Uto, and Tadashi S. Yamamoto. I am also grateful to thme who generously supplied the original micrographs for reproduction: D r j . Stephen J. Gaunt, David M. Phillips. Gerald Schatten, Fumie Suzuki. and Noriko Usui. Finally, t h a n k are due to Dr. Ruth G. Kleinfeld. Mr. T. Timothy Smith. and Mrc. Hiroko Yanagimachi for their help i n the preparation of the manuscript.

K E F E R E N C ES Afrelius. B . A . . Eliasson. R.. Johnsen. O., and Lindholmer, C. (1975).Lack ofdynein arms in immotile human spermatozoa. J . Cdl Biol. 66, 225-232. Ahuja. K. K. (1985). Inhibitors o f glycoprotein biosynthesis block fertiliLation in the hamster. G m r r e Re.\. 11, 179-189. Ailken. K . J.. Koss. A . . and Lees. M. M . (1983).Analysis ofsperm function in Kartegener's syndrome. Fcrril. Sreril. 40, 696-698.

36

RYUZO YANAGIMACHI

Aonuma. S . , Okabe, M., Kawaguchi, M., and Kishi. Y. (1981). Zinc effects on mouse spermatozoa and in-uitro fertilization. J . Rtlprod. Fertil. 63, 463-466. Austin. C. R. (1961). ”The Mammalian Egg.” Thomas. Springfield, Illinois. Austin, C. R., and Walton, A. (1960). Fertilisation. I n “Marshall’s Physiology of Reproduction” (A. S. Parkes. ed.), Vol. I , Pt. 2, pp. 310-416. Longmans, London. Bedford, J. M., and Cooper. G . W . (1978). Membrane fusion events in the fertilization of vertebrate eggs. Cell Stir$ Reu. 5, 65-125. Bedford. J . M., Moore, H. D. M.. and Franklin. L. E. (1979). Significance of the equatorial segment of the acrosome of the spermatozoon in eutherian mammals. E.rp. Cell Rr.s. 119, 119-126. Blandau. R. J . , and Odor, L. D. (1952). Observations on sperm penetration into the ooplasma and changes in the cytoplasmic components of the fertilizing spermatozoon in rat ova. Ferfil. Stevil. 3, 13-26. Boatman, D. E.. and Bavister, B. D. (1984). Stimulation of rhesus monkey sperm capacitation by cyclic nucleotide mediators. J . Reprod. Fertil. 71, 357-366. Boldt, J . . and Wolf, D. P. (1986). An improved method for isolation of fertile zona-free mouse eggs. Comere Rrs. 13, 213-222. Carrol. E. J.. and Levitan, H. (l978a). Fertilization in the sea urchin, Stlon~.vloc~r~tirrorrrs piirpurlitus, is inhibited by fluorescein dyes. Deu. Biol. 63, 432-440. Carrol, E. J . . and Levitan. H . (1978b). Fertilization is inhibited in five diverse animal phyla by erythrosin B. Deu. Biol. 64, 329-33 I . Carrol. E. J.. and Wolf, D. P. (1979). Mouse egg penetration is inhibited by erythrosin B. GutncJte Rrs. I , 293-298. Cherr. G. N . , and Clark. W. H. (1982). Fine structure of the envelope and micropyle in the eggs of the white sturgeon, Acipenser troti.stnon/unu.s Richardson. Deu. Crowrh Djffr. 24, 341-352. Cherr, G . N.. and Clark. W. H. (1985a). Gamete interaction in the white sturgeonAc,ipiwser trtin.s/nntitunu.s: A morphological and physiological review. Enuiron. Biol. Fish. 14, I I 22. Cheer. G. N., and Clark, W. H. (198%). An egg envelope component induces the acrosome reaction in sturgeon sperm. J . Exp. Z o o / . 234, 7s-85. Collins, F.. and Epel, D. (1977).The role of calcium ions in the acrosome reaction of the sea urchin sperm. Exp. Cell ROS.106, 21 1-222. Colwin, A. L., and Colwin, L. H. (1964). Role of the gamete membranes in fertilization. I n “Cellular Membranes in Development” (M. Locke, ed.), pp. 233-279. Academic Press, New York. Colwin, L. H.. and Colwin, A. L. (1967). Membrane fusion in relation to sperm-egg association. I n ”Fertilization” (C. B. Metz and A. Monroy. eds.). Vol. I . pp. 295-367. Academic Press, New York. Cooper, G. W., and Bedford. J . M. (1971). Charge density change in the vitelline surfrlce following fertilization of the rabbit egg. J . Reprod. Fertil. 25, 431-436. Dan. J. C. (1954). Studies on the acrosone. 111. Effect of calcium deficiency. B i d . Bull. 107, 335-349. Dan, J . C. (1967). Acrosome reaction and lysins. I n “Fertilization” (C. B . Metz and A. Monroy, eds.). Vol. I , pp. 237-293. Academic Press, New York. Dravland, 3. E., and Meizel, S. (1982). The effect of inhibitors of trypsin and phospholipase A? on the penetration of zona pellucida-free hamster eggs by acrosome-reacted hamster sperm. J . Androl. 3, 388-395. Dunbar, B . S . , and Wolgemuth, D. J . (1984). Structure and function of the mammalian zona pellucida, a unique extracellular matrix. In “Modern Cell Biology” (B. H. Statier, ed.), Vol. 3, pp. 77-111. Liss, New York.

1. SPERM-EGG FUSION

37

DdzguneS. N. ( 1985). Membrane fusion. .S'/rhc.c,//. Bioc,/rc3/tr.11. 195-28'3 Ehensperger. E . . and Uarros. C . (19x4). Changes at the hamster oocyte ~ r f a c efrom the germinal vesicle stage to ovulation. Gttt,ic.tc Rr\. 9, 387-397. Elinson. R. P. (1975). Site o f sperni entry and ;I cortical contraction associated with egg . 257-268. activation i n the frog Rtrt7tr pipiem\. D c v . B i ~ l 47. Epel. D. (19x2). The physiology and chemistry o f calciuni during the fertilization o f eggs. 1)i: "Calcium and Cell Function" ( W . Y. Cheung, e d . ) . pp. 355-383. Academic Press. New York. Epel. D.. and Vacquier. V . D. ( 19781. Membrane fusion event5 during invertehrate fertiliration. Cell S r t r f : R ~ J L5,J . 1-63, Epel, I)., C r o s . N . l,., and Epel. N. (19771. Flagcllar motility i s not involved in the incorporation o f t h e sperm into the egg at fertilization. D c u . Grow//iDi'pr. 119, 15-21. Flechon. J . E.. and Pavlok. A . (1986). Ultrastructural study ofthe interactions and fusion o f rani spermatoroa with zona-free hamster oocyte5. Reprod. Nitrr. Dcu. 26, 999- 1008. Fleming. A . D., and Armstrong. D. T. ( 1 9 x 5 ) . Effects o f polyaniines upon capacitation and 233, 97-100. fertilization in the guinea pig. 1.Exp. Z(JO/. Fleming. A . D.. and Yanagimachi. R. (19x1). Effects o f various lipids on the aci-osome reaction and fertiliLing capacity o f guinea pig spermatozoa. with special reference to the possible involvement o f lysophospholipids in the acrosome reaction. G ( i ) t i e / e Re\. 4, 253-273. Franklin, L. E. (1965). Morphology o f gamete membrane fusion and of sperm entry into oocytes o f the sea urchin. J . Crll Biol. 25, XI-100, Fraser. L. R. ( 1982).p-Aminobenramidine. an aci-osin inhibitor. inhibits mouse sperm penctration o f the zona pellucida but not the acrosome reaction. J . Reprod. Fcrril. 65. 185194. Fraser. L. R. ( 1983). Potassium ions modulate expression o f mouse sperm fertilizing ability. acrosome reaction and hyperactivated motility i/r uitro, J . R ~ p r ( ~Pcrril. d. 69, 539-553. Freeman, G., and Miller. R. L. (1982). Hydroroan egg can only be fertilized at the site of polar body formation. Dcu. Biol. 94, 142-152. Gabel. A,. Eddy. E. M . . and Shapiro. M. (1979). After fertilization. sperm surface components remain a s a patch in sea urchin and mouse embryos. C e l l 18, 207-215. Gaddum-Rosse. P. (19x5). Mammalian gamete interactions: What can be gained from observations on living eggs? A m . J . Anrrt. 174, 347-356. Gaunt. S. J . (1983). Spreading of a sperm surf;ice antigen within the plasma membrane of the egg after fertilization i n the rat. 1.E t t r h r y d , E.Y[J.Morpliol. 75, 259-270. Glabe. C. G. (1985). Interaction of sperm adhe\ive protein. bindin. with phospholipid v e G cles. 11. Bindin induces the fusion o f mixed phase vesicles that contain phosphatidylcholine and phosphatidylserine in u i t m . J . C d l Biol. 100, 800-806. Glabe. C. G.. and Vacquier. V. D. (1977). Isolation and characterization o f the vitelline layer o f sea urchin eggs. J . Cell Biol. 75, 410-421. Glabe. C. G.. and Vacquier. V . D. (1978). Egg surface glycoprotein receptor for sea urchin sperm bindin. Proc,. Ntrtl. Ac.trd. .S(,i. U . S . A . 5, X81-885. Cordon. M.. Fraser. L. R.. and Dandekar. P. V. (1975). The effect of ruthenium red and concanavalin A on the vitelline surface o f fertilized and unfertilized rabbit ova. Antrr. R r t . . 181, 95-1 12. Gould-Somero, M.,and Jaffe. L. A . (1984). Control o f cell fusion at fertilization by mcmbrane potential. In "Cell Fusion: Gene Transfer and Transformation" (K.F. Beers and E. G. Bassett. eds.). pp. 27-38. Raven. New York. Gundersen. G . G . , and Shapiro. B. M. (1984). Sperm surface proteins persist after fertiliiation. J . Cell Riol. 99, 1343-13.53, Gundersen, G . G.. Medili. L.. and Shapiro. B. M . (1986). Sperm surface protein5 are

38

RY UZO YANAG IMACH I

incorporated into the egg membrane and cytoplasm after fertilization. Deu. B i d . 113, 201-2 17. Gwarkin. R . B. L., Williams, D. T., Hartmann. J . F., and Kniazuk, M. (1973). The zona reaction of hamster and mouse eggs: Production in uitro by a trypsin-like protease from cortical granules. J . Reprod. Frrfil. 32, 259-265. Hanada, A, , and Chang, M. C. (1972). Penetration of zona-free eggs by spermatozoa of different species. B i d . Rrprod. 6, 300-309. Hinkley. R . E.. Wright, B. D.. and Lynn. J. W. (1986). Rapid visual detection of sperm-egg fusion using the DNA-specific fluorochrome Hoechst 33342. Dru. B i d . 118, 148-154. Hirao, Y., and Yanagimachi, R . (1978a). Temperature dependence of sperm-egg fusion and post-fusion events in hamster fertilization. J . Exp. Zoo/. 205, 433-438. Hirao, Y., and Yanagimachi, R . (1978b). Effects of various enzymes on the ability of hamster egg plasma membrane to fuse with spermatozoa. Gmnere Rrs. 1, 3-12. Hoffman, M. L.. and Curtis, G . L. (1984). Prevention of monkey sperm penetration of zonafree hamster ova by sperm antibody obtained from vasectomized cynomolgus monkey. F e r / i / . Sterd. 42, 108- I 1 I . Hong, K.. and Vacquier, V . D. (1986). Fusion of liposomes induced by a cationic protein from the acrosomal granule of abalone spermatozoa. B i d m i s f r y 25, 543-549. Hoshi. M. (1985). Lysins. 111 "Biology of Fertilization" (C. B. Metz and A. Monroy, eds.), Vol. 2. pp. 431-462. Academic Press, Orlando, Florida. Huang. T . T . F., and Yanagimachi, R . (1985). Inner acrosomal membrane of mammalian spermatozoa: Its properties and possible functions in fertilization. A m . J . Ancrt. 174, 249-268. Huang, T. T. F . , Tung. K. S. K . , and Yanagimachi, R . (1981). Autoantibodies from vasectomized guinea pig inhibit fertilization in uifro. Science 213, 1267- 1269. Hubbs, C. L. (1955). Hybridization between fish species in nature. S y s f . Zoo/. 4, 1-20, Hultin. T. (1948). Species specificity in fertilization reaction. 11. Influence of certain factors on the cross-fertilization capacity of Arhuciri lixulrr. Ark. Zoo/. Scund. 40A, 1-8. Hylander, B. L . . and Summers, R . G. (1977). An ultrastructural analysis of the gametes and early fertilization in two bivalve molluscs, Chrrrncr mrrc~erop/iy//uand S p i s r h S(J/idksirncr with special reference to gamete binding. Cell T i s s ~ eR P S .182, 469-489. Iwao. Y., and Katagiri, C. (1982). Properties of the vitelline coat lysin from toad sperm. J . Exp. Zoo/. 219, 87-95. Jaffe. L. A. (1976). Fast block to polyspermy in sea urchin eggs is electrically mediated. Nature (London) 261, 68-71. Jaffe, L. A., and Cross, N . L. (1986). Electrical regulation of sperm-egg fusion. Annrr. R c u . Physiol. 48, 191-200. Jaffe, L. A., and Gould. M. (1985). Polysperniy-preventing mechanisms. In "Biology of Fertilization" (C. B. Metz and A. Monroy, eds.). Vol. 3, pp. 223-250. Academic Press, Orlando, Florida. Jaffe, L. A. , Sharp, A. P., and Wolf, D. P. (1983). Absence of an electrical polyspermy block in the mouse. Deu. B i d . 96, 317-323. Jensen. A. B. (1953). The effect of trypsin on the cross-fertilizability of sea urchin eggs. Exp. Cell Res. 5 , 325-328. Johnson, M. H . , Eager, D., Uggleton-Harris, A., and Grave, H. M . (1975). Mosaicism in organization of concanavalin A receptors on surface membranes of mouse eggs. Nrrtrtre (London) 257, 321-322. Kille, R . A. (1960). Fertilization of the lamprey egg. Exp. C r / /Res. 20, 12-27. Kobayashi, H. (1963). Some cytological observations on fertilization in the Ioach (female)funa (male) cross. J p n . J . Genet. 38, 113-122. Kobayashi, W., and Yamamoto, T. S . (1981). Fine structure of the micropylar apparatus of

1. SPERM-EGG FUSION

39

the chum salmon egg, with ii discussion of the mechanism for blocking polyspermy. J , E x p . Zoo/. 217, 265-275. Kobaya5hi. W . . and Yamamoto. T. S. (1987). I*ight and electron microscopic observations of sperm-entry in the chum salmon egg. J . E.vp. Zoo/. 243, 31 1-322. Koehler, J.. Clark. J . M., and Smith, D. (1985). Freeze-fracture observations on mamnialian oocytes. A m . J . Anu!. 174, 317-329. Kudo, S. ( 19x0). Sperm penetration and the formation of a fertilimtion cone in the common carp egg. D P U .Gron.th DiffCJr. 22, 403-414. Kyoruka, K., and Osanai. K. (1979). Heterospermic fertilization of sea urchin eggs pretreated with pancreatin. B i t / / . Mar. Biol. S t n . AS L J ~ ?J M Tohoku .S/I~ Univ. 16, 179- 188. Longo, F. J . ( 1977). An ultrastructural study of cross-fertilization (Arbuciu female x M.~ti/trs male). J . Cell Eiol. 73, 14-26. Longo, F. J. ( 1978). Effects of cytochalasin B on sperm-egg interactions. Deu. Biol. 67, 249-265. Lopo, A. C. (1983). Sperm-egg interactions in invertebrates. I n “Mechanism and Control of Animal Fertilization” (J. F. Hartmann. ed.). pp. 269-324. Academic Press. New York. Lopo, A. C.. and Vacquier, V. D. (1981).Gamete interaction in the sea urchin. I n “Fertilization and Embryonic Development in Virro” ( L . Mastroianni and J . D. Biggers. edh.), pp. 199-232. Plenum. New York. Lucy. J . A. (1978). Mechanisms of chemically induced cell fusion. Cell Surf: R P U .5, 268304. Maddock, M. B . . and Dawson. W . D. (1974). Artificial insemination of deermice with sperm from other rodent species. J . Emhryol. Exp. Morphol. 31, 621-634. McCulloh. D. H.. Rexroad. C. E.. and Levitan. A. (1983). Insemination of rabbit egg in associated with slow depolarization and repetitive diphasic membrane potentials. Dcu. Bin/. 95, 372-377. McGath, J . . and Hillman, N. (1980). Sterility in mutant ( i l “ / i l ‘ )male mice. 111. In uiiro fertilization. J . Emhnol. Exp. Morphol. 59, 49-SX. Mann. S . . Schatten. G . , Steinhardt. R . , and Friend, D.S. (1976). Sea-urchin sperm: oocyte interaction. J. Cell Biol. 70, I10a (Abstr. No. 330). Matsuda. T . , Komori, K . . Funauchi. H.. Shigeta. M.. Koyama. K.. and Isojima. S. (1985). Blocking effect of monoclonal antibodies to oocyte (zona pellucida and plasma membrane) on in uitro fertilization in hamster. Ac.fc4 Ohsre!. Gyrrwol. Jpn. 37, 2615 (Abstr. No. 398). Metz, C. B.. and Monroy. A , , eds. (1967). “Fertilization.” Vol. I. Academic Press, New York. Metz, C . B., and Monroy. A , . eds. (1969). “Fertilization.” Vol. 2. Academic Press. New York. Metz, C. B., and Monroy, A., eds. (1985). ”Biology of Fertilization,” Vols. 1-3. Academic Press, Orlando, Florida. Miller, R. L. (1978). Site-specific sperm agglutination and time related release of a hpermchenio-attractant by the egg of the Leptomedusan, 0rihopy.ri.s (.u/icu/u!u.J . E.rp. Z o o / . 205, 385-392. Miyazaki. S.. and Igusa. Y. (1981). Fertilization potential in golden hamster eggs consists of recurring hyperpolarization. Nurrtrc (London) 290, 702-704. Miyazaki. S . , Hashirnoto. N., Yoshimoto. Y.. Kishimoto, T., Igusa, Y., and Hiramoto. Y. ( 1986). Temporal and spatial dynamics of the periodic increase in intracellular free calcium at fertilization of golden hamster egg. Deu. B i d . 118, 259-267. Moenkhaus, W. J . (1910). Cross-fertilization among fishes. Proc. Indiun Al’Nd.Sc,i. 353-393. Mohri. H., Usui. N . , and Sano, K. (1982). Magnesium ions in fertilization of sea urchins. Cell Difier. 11, 259-260.

40

RYUZO YANAG IMACHI

Monroy, A. (1985). Processes controlling sperm-egg fusion. Errr. J . Biuchem. 152, 5 1-56, Nikoljuki, N. I . (1965). Hybridization. Fish News f n t . 4, 473-476. Ohta, T. (1985). Electron microscopic observations on sperm entry into eggs of the bitterling during cross-fertilization. J . E x p . Zoo/. 233, 291-300. Oikawa, T.. and Suzuki, T. (1979). Demonstration of male reproductive tissue specific antigen and induction of the failure of fusion between gametes in golden hamster. B i d . Reprod. 20, (Suppl. I),41A (Abstr. No. 59). Okamura. F., and Nishiyama, H. (1978a). The passage of spermatozoa through the vitelline membrane in the domestic fowl, Gallrrs gulltrs. Cell Tisstre Res. 188, 497-508. Okamuram F.. and Nishiyama, H. (1978b). Penetration of spermatozoon into the ovum and transformation of the sperm nucleus into the male pronucleus in the domestic fowl, Gallrts gullus. Cell Tissue Res. 190, 89-98. Osanai, K.. and Kyozuka. K. (1982). Cross-fertilization between sea urchin eggs and oyster spermatozoa. Gamete Res. 5 , 49-60. Papahadjopoulos. D. ( 1978). Calcium-induced phase changes and fusion in natural and model membranes. Cell St4rf. Reu. 5 , 765-790. Papahadjopoulos, D., Poste, G.. and Vail, W. J. (1979). Studies on membrane fusion with natural and model membranes. Methods Memhr. B i d . 10, 1-121. Pavlok. A. (1979).Interspecies interaction of Lona-free ova with spermatozoa in mouse. rat and hamster. Anim. Reprod. Sci. 2, 39.5-402. Pavlok. A,, Petelikova, J . , and Flechon. J . E. (1983). Interaction of zona-free hamster eggs with fresh and frozen ram spermatozoa in uitro. In "The Sperm Cell" ( 1 . Andre. ed.). pp. 51-54. Nijhoff, The Hague. Podell, S. B., and Vacquier, V. D. (1985). Purification of the M , 80,000 and M , 210,000 proteins of the sea urchin sperm plasma membrane. J . Biol. Chem. 260, 27152718. Primakoff, P., Hyatt, H., and Myles, D. G. (1985). Guinea pig sperm-egg membrane fusion is blocked by the PH-30 monoclonal antibody which binds to a posterior head sperm surface antigen. J . Cell B i d . 101, 262a (Abstr. No. 1372). Quinn, P., and Stanger. J . D. (1981). Fertilization of pronase-treated mouse ova in uitro. Aust. J. B i d . Sci. 34, 245-248. Rodger. J . C., and Bedford, J . M. (1982). Separation of sperm pairs and sperm-egg interaction in the opossum Didelphis uirginiunu. J . Reprod. Fertil. 64, 171-179. Rogers, B. J., Ueno, M., and Yanagimachi, R. (1981). Fertilization by guinea pig spermatozoa requires potassium ions. Biol. Reprod. 25, 639-648. Roldan, E. R. S . , Vitullo, A. D., Merani, M. S . . and Lawzewitsch, I . V. (1985). Crossfertilization in uiuo and in uitro between three species of Vesper mice, Culom,v.s (Rodentia, Cricetidae). J . E x p . Zoo/. 233, 433-442. Saling, P. M. (1981). Involvement of trypsin-like activity in binding of mouse spermatozoa to zonae pellucidae. Proc. Nutl. Acud. Sci. U . S . A . 78, 623 1-6235. Saling, P. M. (1982). Development of the ability to bind to zonae pellucidae during epididyma1 maturation: Reversible immobilization of mouse spermatozoa by lanthanum. Biol. Reprod. 26, 429-436. Saling, P. M.. and O'Rand, M. G . (1982). Anti-mouse sperm antiserum: Fertilization inhibition in uitrcr and preliminary antigen identification. J . Androl. 3, 434-439. Saling, P. M., Raines, L. M., and O'Rand. M. G . (1983). Monoclonal antibody against mouse sperm blocks a specific event in the fertilization process. J . E x p . Z o o / . 227,481486. Saling, P. M., Irons. G., and Waibel, R. (1985). Mouse sperm antigens that participate in fertilzation. 1. Inhibition of sperm fusion with the egg plasma membrane using monoclonal antibodies. B i d . Reprod. 33, 515-526.

1. SPERM-EGG FUSION

41

Sano. K.. and Kanatani. H. (1980). External calcium ions are requisite for fertilization o f w a urchin eggs by spermatoroa with reacted acrowmes. Drv. Uiol. 78, 242-246. Sano. K., Usui. N.. Ueki. K.. Mohri. T.. and Mohri. H. (1980). Magnesium ion-requiring step in fertilization of sea urchins. DcrI. G r o i i ~ t l iDiffcr. 22, 531-541. Sato. K.. and Blandau. R. J . (1979). Time and process of sperm penetration into cumulusfree mouse eggs fertilized in uitro. G e i m ~ t cRas. 2, 295-304. Sato. M . . and Osanai, K. (1986). Morphological identification of sperm receptor\ above egg microvilli in the polychaete. N W I I / / W ,Sj q ~ o t i i w Doti. . Biol. 113, 263-270. Schatten. G . . and Mazia. D. (1976). The penetration of the spermatozoon through the wii urchin egg surface at fertilimtion. Exp. Cell Rc~s.98, 325-337. Schmidt, T., Patten, C.. and Epel. D. (1982). Is there a role for the Ca'* inHux during fertilization of the sea urchin egg'? Dcu. Riol. 90. 284-290. Schuel, H. (1984).The prevention of polyspermic fertilization in sea urchins. B i d . Bull. 167, 27 1-309. Schuel, H., Longo. F. J . , Wilson. W. L... and Troll. W. (1976). Polyspermic fertilziation of \ea urchin eggs treated with protease inhibitors: Localization of sperm receptor \ires at the egg surface. Deu. Biol. 49, 178-184. SeGall, G . K . . and Lennarz. W. L. (1979). Chemical characterization of the component of thejelly coat from sea urchin eggs responsible for induction of the acrosome reaction. f)<Ju. B i ~ l71, . 33-48. Seifriz. W. ( 1927). Protoplasmic papillae of Ec~/ii~iurrr~./iniit.s oocytes. Pro/oplrr.snicr 1. 1-4. Shalgi. R . , and Phillips, D. (1980). Mechanic5 of sperm entry in cycling hamsters. J . U//ru,stritct. Res. 71, 154-161. Strittmatter. W. J.. Couch. C. B . , and Mundy. 0 . I . (1985). Role of proteins in the fusion of biological membranes. Mev?hr. Flitid. Biol. 4. 259-291. Sugiyama. M . (1951). Re-fertilization of the fertilized eggs of the sea urchin. R i d . Bttll. 101, 335-344. Summers, R. G.. and Hylander. B . L. (1974). A n ultrastructural analysis of early fertiliution in the sand dollar, Echincrrrrc./iniu.s ptrrniti. Cell Tis.\ur Rc,.s. 150, 343-368. Summers. R . G., and Hylander. B. L. (1975). Species-specificity of acrosome reaction and primary gamete binding in echinoids. Exp. C P / /R P S .96, 63-68. Suzuki, R. (1964). Hybridization experiments in cyprinid fishes. VI1. Reciprocal crosses between Psrudogohio esocinrts and B i i i , ~ i ( :i r z r r ~ ~ (Jip. n . J . f c , / i t / i y o / .12, 18-22. Takahashi, Y . M., and Sugiyama. M. (1973). Relation between the acrosome reaction and fertilization in the sea urchin. 1. Fertilization in Ca-free seawater with egg-water-treated spermatozoa. Dru. Gron~rhD(fjrC.r.15, 261 -267. Talansky, B. T . , Berg, P. E., and Gordon, J . E. (1987). Ion pump ATPase inhibitors block the fertilization of zona-free mouse oocytes by acrosome-reacted spermatozoa. J . Reprod. Frrtil. 19, 447-455. Tyler, A , , Monroy. A.. and Metz. C. B. (1956). Fertilization of fertilized sea urchin egg. Biol. B i t / / . 110, 184-195. Tzartoz, S. J . (1979). Inhibition of in uitro fertilization of intact and denuded hamster eggs by univalent anti-sperm antibodies. J . Rrproil. Fertil. 55, 447-455. Usui. N . . and Yanagirnachi, R. (1976). Behavior of hamster sperm nuclei incorporated into eggs at various stages of maturation, fertilization and early development. J . L///rci.s~rircr. Rr.s. 57, 276-288. Usui, N..Sano, K., and Mohri, H. (1980). The surface events at fertilization of the sea urchin egg. I . Events on the surface of the vitelline coat. Dru. Grow//i DIffrr. 22, 461473. Vacquier, V. D. (1979a). The fertilizing capacity of sea urchin sperm rapidly decreases after induction of the acrosome reaction. Deu. Growt/i Differ. 21, 61-69.

42

RY UZO YANAG IMACHI

Vacquier. V. D. (1979b). The interactions of sea urchin gametes during fertilization. A m . Zoo/. 19, 839-849. Vacquier, V. D., and Moy. G. W. (1977). Isolation of bindin: The protein responsible for adhesion of sperm to sea urchin egg. Proc. N u t / . Acud. Sci. U . S . A . 74, 2456-2460. Van der Ven, H. H., Kaminski. J . , Bauer, L . , and Zaneveld, L. J . D. (1985). Inhibition of human sperm penetration into zona-free hamster oocytes by proteinase inhibitors. Feriil. Sieril. 43, 609-616. Warikoo, P. K., Majumdar. S . S . , Allag, I . S . . Das, R. P., and Roy, S. (1986). Interaction5 among motility, fertilizing ability, and testosterone binding on spermatozoa of bonnet monkey (Mucucu rudiuiu). Arch. Androl. 16, 135-141. Wassarman, P. M., Florman, H . M., and Greve. J . M. (1985). Receptor-mediated spermegg interactions in mammals. I n “Biology of Fertilization” (C. B . MetL and A. Monroy, eds.). Vol. 2. pp. 341-360. Academic Press, Orlando. Florida. Wolf, D. E.. and Ziomek, C. A. (1983). Regionalization and lateral diffusion of membrane proteins in unfertilized and fertilized mouse eggs. J . Cell Biol. 96, 1786-1790. Wolf, D. P. (1977). Involvement of a trypsin-like activity in sperm penetration of zona-free mouse ova. J . Exp. Zoo/. 199, 149-156. Wolf, D. P . . and Hamada, M. (1977). Induction of zonal and egg plasma membrane blocks to sperm penetration in mouse eggs with cortical granule exudate. Biol. Reprod. 17, 350354. Wolf, D. P.. Inoue, M.. and Stark, R. A. (1976). Penetration of zona-free mouse ova. Biol. Reprod. 15, 213-22 I . Yanagimachi. R. (1957). Studies of fertilization in Clirpeu pullusii. V . Fertilization of the egg deprived of the membrane. J p n . J . Ichihyol. 6, 41-47. Yanagimachi, R. (1966). Time and process of sperm penetration into hamster ova in uiuo and in uiiro. J . Reprod. Feriil. 11, 359-370. Yanagimachi, R. (1977). Specificity of sperm-egg interaction. I n “lmmunobiology of Gametes” (M. Edidin and M. H. Johnson, eds.). pp. 255-296. Cambridge Univ. Press, London. Yanagimachi, R. (1978a). Sperm-egg association in mammals. Cirrr. Top. Dru. B i d . 12, 83105. Yanagimachi, R. (1978b). Calcium requirement for sperm-egg fusion in mammals. B i d . Reprod. 19, 949-958. Yanagimachi, R. (1981). Mechanisms of fertilization in mammals. I n “Fertilization and Embryonic Development in Viiro” (L. Mastroianni and J . D. Biggers, eds.), pp. 81182. Plenum, New York. Yanagimachi, R. (1982). Requirement of extracellular calcium ions for various stages of fertilization and fertilization-related phenomena in the hamster. Gumefe Rex. 5 , 323344. Yanagimachi, R. ( 1984). Zona-free hamster eggs: Their use in assessing fertilizing capacity and examining chromosomes of human spermatozoa. Gamete Res. 10, 187-232. Yanagimachi, R., and Kanoh, Y. (1953). Manner of sperm entry in herring egg, with special reference to the role of calcium ions in fertilization. J . Fuc. Sci. Hokkaido Uniu. Ser. VI ( Z o o / . )11, 487-494. Yanagimachi, R . , and Noda, Y. D. (1970a). Electron microscopic studies of sperm incorporation into the hamster egg. A m . J . Anul. 128, 429-462. Yanagimachi, R., and Noda, Y. D. (1970b). Physiological changes in the post-nuclear cap region of mammalian spermatozoa: A necessary preliminary to the membrane fusion between sperm and egg cells. J . Ulirasrruci. Res. 31, 486-493. Yanagimachi, R., and Teichman, R. J. (1972). Cytochemical demonstration of acrosomal

1. SPERM-EGG FUSION

43

proteinase in mammalian and avian spermatoroa by ;I silver proteinate method. Biol. R l > [ l i l J d . 6 , 87-97. Yanagimachi. R.. Yanagimachi. H.. and Rogers. B . J . (1976). The use of rona-free animal o v a as a test system for the assessment of the fertiliLing ctipacity of human spermalo/.oil. Biol. Rcprod. 15, 47 1-476. Yanagimachi. R.. Miyashiro, L. H . . and Yanagimachi. H . (1980). Reversible inhibition of ~ i h 22, 281-288. sperm-egg fusion in the hamster by low pH. I1r.u. G i o ~ ~Dlffir. Yanagimachi. R.. Okada. A , . and 'rung. K. S . K. (1981). Sperm autoantigens and fertilization. 11. Effects of anti-guinea pig autoantibodie\ o n sperm-ovum interaction>. Biol. ~ r p ~24, d 5. I 2-5 I 8 . Yanagimachi, R..Huang. T. 'I. F..Fleming. A . D.. Kosower. N . S . , and Nicolson. 6. 12, (1983). Dithiothreitol, a disulfide-reducing agent. inhibits capacitation. iicrosome reaction and interaction with eggs by guinea pig spermatozoa. GcitniJreRrs. 7, 145-154. Yoshizaki. N . . and Katagiri. C. (1982). Acrosonie reaction in sperm of the toad BI&J h j i ) ,jrr/xJnic~cc.(;trtiirri~Rc>.s.6, 343-352.

This Page Intentionally Left Blank

CURRENT TOPICS I N M E M B R A N E S A N D TRANSPORT. VOL.CMC.

i?

Chapter 2 Cortical Exocytosis in the Sea Urchin Egg ROBERT C . JACKSON A N D JOSEPH H . CRABB' Deprirttnttit OJ' Bioclietwistry Dcirttnorrth Mcdicwl School Hutiourr, Nenj Hlitnpshirc, 037.56

I. Introduction A. Exocytosis in Perspective

B. Cortical Exocytosis in the Sea Urchin Egg The Calcium Signal A . Phosphoinositide Metabolism and Signal Transduction B . The Role of G Proteins 111. I,r Vitro Models of Exocytosis A . Permeabilized Cell Sy\tems B . Cell-Free Systems C. Egg Cortex as an ;,I uiiro Model of Ca!--Triggered Exocytosis D. Hypothetical M ~ d e l s 1V. Conclusions V . Addendum A. Sea Urchin Egg B. Other Cell Types References 11.

I.

INTRODUCTION2

The initiation of new life at fertilization is an event of profound significance. for both the individual and the species. For the individual, fertilization marks a beginning; for the species, it represents continuance: for the I Present address: Channing Laboratory, Harvard Medical School. Boston. Mahsachusetts 011 15. Abbreviations: CL, cortical lawn: CSC. cell surface complex: CV. cortical vesicle: DAG, diacylglycerol; FE. fertilization envelope: IP,, inositol 1,4.5-1risphosphate: NEM. N ethylmaleimide: PI, phosphatidylinositol; PIP. phosphatidylinositol 4-phosphate: PIP?. phosphatidylinositol 4.5-bisphosphate; PM. plasma membrane.

45

46

ROBERT C. JACKSON AND JOSEPH H. CRABB

cell biologist, it comprises a unique and often advantageous system for analyzing basic cellular phenomena. Exocytosis, the phenomenon in which secretory proteins and stored effector molecules are delivered to the cell surface and beyond, is a process that can be conveniently studied during fertilization, particularly in organisms (e.g., the sea urchin) that produce an abundance of gametes. In this chapter we focus on the process of cortical exocytosis in the sea urchin egg. Two areas will be examined in detail: (1) the mechanism of generation of the intracellular calcium signal and (2) the use of egg cortex as an in uitro model for investigation of the molecular mechanism of the terminal stages of exocytosis. We have not attempted to present an exhaustive review of the literature on cortical exocytosis, rather we have tried to emphasize results obtained within the last 5 to 7 years in relation to what is known from other exocytotic systems. Other recent reviews have dealt with the biology of fertilization in the sea urchin (Guidice, 1986; Shapiro et al., 1981), and with the structure of cortical secretory vesicles and their role in formation of the fertilization envelope (Epel, 1978, 1982; Epel and Vacquier, 1978; Kay and Shapiro, 1985; Schuel, 1985). A. Exocytosis in Perspective

Exocytosis is, of course, the final step in the secretory pathway; however, it is also representative of the more global phenomenon of vesicular transport, in which molecules entering and exiting the cell, as well as those destined for particular organelles, are transported within membrane-bound vesicles. This process of vesicular transport insures that these molecular “passengers” do not admix with cytoplasmic “residents,” and thus allows remarkably efficient segregation and sorting of incoming (endocytic) and outgoing (exocytic) traffic. Vesicular transport occurs in three stages: budding of a vesicle from the compartment of origin, translocation of that vesicle through the cytoplasm, and fusion of the vesicle with the destination compartment. 1. CONSTITUTIVE VERSUS REGULATEDSECRETION

During exocytosis, the compartment of destination is the extracellular space. Virtually all eukaryotic cells use this mechanism to externalize newly synthesized secretory and plasma membrane (PM) proteins (Kelly, 1985; Tartakoff and Vassalli, 1977, 1978). Secretion from cells on a bulkflow basis, as the proteins are synthesized, is termed constitutive secretion, while exocytosis in direct response to a stimulus at the cell surface is

47

2. CORTICAL EXOCYTOSIS

called regulated secretion. The regulated form of secretion seems to have evolved for the sole purpose of rapidly externalizing large quantities of physiologically important mediators such as hormones. enzymes, neurotransmitters, and immunological effectors in response to specific stimuli. It is marked by the accumulation and storage of secretory vesicles within the cytoplasm. During storage, the secretory vesicles develop a characteristic dense core of secretory product, concentrated some 10- to 200fold over that of constitutive secretory vesicles (Bendayan et d,, 1980; Salpeter and Farquar, 1981). 2. STIM U L US-SECRETION COUPLING The role of Ca2+in regulated secretion was first recognized in the early 1960s by Douglas, who coined the term stimulus-secretion coupling to denote its similarity to the Ca?+-dependentprocess of excitation-contraction coupling (Douglas and Rubin, 1961, 1963). Douglas' description of stimulus-secretion coupling, in conjunction with ultrastructural studies of Palade and co-workers (1975) established the general scheme for regulated secretion shown in Fig. I : Upon stimulation, the cytoplasmic concentration of free calcium ion ([Ca?+],) transiently rises, from resting levels of approximately lo-' M to several micromolar (Rubin, 1982). At about the same time, the secretory vesicles are translocated from the cytoplasm to the plasma membrane and become closely apposed (or at-

@ FIG. I . Regulated secretion. ( I ) Stiumulation or activation through receptor-ligand interaction. (2) Rise in intracellular [Ca?+].(3) Vesicle translocation. (4) Vesicle attachment (close apposition). ( 5 ) Membrane fusion. (6) Discharge and dispersal of vesicle contents.

48

ROBERT C. JACKSON AND JOSEPH H. CRABB

tached) to it. Next, the bilayers of the vesicle and the plasma membrane fuse, establishing continuity between the two membranes and joining the intravesicular and extracellular compartments. This allows the vesicle contents to be discharged and dispersed, while maintaining the membrane topology and the all-important barrier between the cytoplasm and the extracellular milieu. While it is apparent that this process occurs via an interdependent progression, there are basically three discrete stages which can be recognized. These are signal generation, vesicle translocation, and fusion; corresponding to the initial, intermediate, and final stages. It has proved difficult to assign particular biochemical requirements to discrete steps in this pathway, since the stages cannot be independently manipulated in whole cell studies. This is where an in vitro approach can be most useful. In uitro reconstitution of the individual steps along this pathway should permit unambiguous assessment of the requirements and properties of each step. B. Cortical Exocytosis in the Sea Urchin Egg

For discussion purposes, fertilization in the sea urchin can be thought of as a three-step process. The first step consists of sperm activation and gamete binding. Gamete fusion comprises the second step. The third step involves activation of the metabolism of the dormant egg cell. The cortical reaction of the egg occurs immediately after gamete fusion. It is one of two exocytotic events that occur during the fertilization sequence; the other, the acrosome reaction of the sperm, occurs during sperm activation. The unique morphology of the sea urchin egg is the feature that makes it such a useful system for studying the terminal events in exocytosis. The secretory vesicles of most cells must move from the cytoplasm to the plasma membrane before they can fuse and release their contents into the extracellular medium. In the sea urchin egg, as in the egg cells of most other species (Anderson, 1974; Gulyas, 1980), specialized secretory vesicles called cortical vesicles (CVs) become firmly attached to the PM during oogenesis (Anderson, 1968; Detering et al., 1977; Longo, 1981; Vacquier, 1975). This specialization of the egg has two practical consequences: it simplified exocytosis by effectively eliminating the translocation and attachment steps (steps 3 and 4 in Fig. l ) , and it permits the isolation of egg cortex fractions in which the CVs remain bound to the PM and retain their exocytotic capacity. The morphology of the cortical region of a sea urchin egg is shown in Fig. 2. Cortical vesicles in this particular species (Strongylocentrotus prrrpuratus) are approximately 1 p m in diameter and consist of a mem-

FIG.2. Electron micrograph of the cortical region of an unfertilized sea urchin egg (S. p / o p r m u s ) . Cortical vesicles lie in a monolayerjust beneath the plasma membrane and have a characteristic spiral lamellar core. The vitelline layer (arrow)is the thin. fuzzy structure coating the extracellular surface of the plasma membrane. Bar. 0.5 ~ I I I (Reproduced . from Chandler and Heuser, 1979. with permission. )

50

ROBERT C.JACKSON AND JOSEPH H. CRABB

brane-encapsulated spiral lamellar core of condensed proteins and mucopolysaccharides. The entire cytoplasmic surface of the PM is covered with a densely packed array of CVs. This can best be appreciated in light micrographs of fragments of egg cortex (see Fig. 3B in Section lll,C,l). It has been estimated that a single egg cell contains approximately 15,00018,000 CVs and that fusion of the CVs with the PM during cortical exocytosis more than doubles the surface area of the egg (Schroeder, 1979; Vacquier, 1975; Zimmerberg and Whitaker, 1985). The egg PM is comprised of a typical phospholipid bilayer, but it is covered with the vitelline layer, an extracellular glycocalyx, which contains the sperm receptors of the egg (Shapiro et al., 1981). Filamentous structures are abundant on the cytoplasmic surface of the egg cortex, but the specific nature of the linkage between the CVs and the PM is not known (see Section III,C,6). When an egg is penetrated by a sperm at fertilization, the CVs undergo a synchronous Ca*+-regulated exocytotic reaction and release their contents into the perivitelline space (i.e., the narrow channel between the plasma membrane and the vitelline layer). The primary function of the cortical reaction is to produce the fertilization envelope (FE), a tough, extracellular investment that establishes a permanent block to polyspermy and shields the developing embryo from environmental insult. The FE is formed by the combination of CV content proteins with the vitelline layer (Kay and Shapiro, 1985). Cortical vesicles contain a variety of constituents, most of which are thought to participate in FE formation. A serine protease(s) contained within the CV may be responsible for destroying sperm receptors and detaching the vitelline layer from the PM (Alliegro and Schuel, 1985; Carroll and Epel, 1975; Fodor et ul., 1975; Sawada et al., 1984). Proteoliaisin, a recently purified CV protein, is thought to form a major structural component of the FE (Weidman et al., 1984). Ovoperoxidase, a major CV enzyme, cross-links FE proteins via their tyrosine side chains and thereby hardens the so called “soft” FE that is initially formed (Deits et al., 1984; Foerder and Shapiro, 1977; Hall, 1978). Hydration of CV mucopolysaccharides within the perivitelline space is thought to elevate the detached FE from the egg surface (Schuel, 1978; Schuel et al., 1974). Hyalin, a high molecular weight CV glycoprotein, comprises the major constituent of the hyaline layer which forms in the space between the egg and its FE. The hyaline layer is important in blastomer adhesion in the early embryo (Hylander and Summers, 1982; Kane, 1970; McClay and Fink, 1982; Stephens and Kane, 1970). P-Glucanase, a CV enzyme capable of hydrolyzing terminal p-1,3-linked glucosides, has been purified, but

2. CORTICAL EXOCYTOSIS

51

its function remains unknown (Epel et al., 1969; Muchmore er ul., 1969; Talbot and Vacquier, 1982). II. THE CALCIUM SIGNAL There is ample evidence that exocytosis is the direct result of an increase in the cytoplasmic concentration of free Ca2+to the micromolar level. Exocytosis can be initiated by treating intact cells with Ca?? ionophores (Chambers er d., 1974; Forman et ul., 1973; Rasmussen and Goodman, 1977; Steinhardt and Epel. 1974), by the direct intracellular microinjection of CaZ+buffers (Gilkey. 1983; Hamaguchi and Hiramoto, 1981; Hollinger and Schuetz, 1976; Hollinger et a l . , 1979; Kanno et ul.. 1973; Miledi, 1973), or by bathing permeabilized cells in a Ca?+-containing medium (Baker and Knight, 1981; Dunn and Holz, 1983; Knight and Baker, 1982; Knight and Scrutton, 1980; Wilson and Kirshner, 1983). Exocytosis has been shown to be preceded by increases in [Caz+],,and inhibition of the [Caz+],transient by microinjection of EGTA has been shown to block exocytosis in medaka and sea urchin eggs (Gilkey. 1983; Zucker and Steinhardt, 1978). The “wave” of elevated [Ca2+],that accompanies exocytosis was directly detected for the first time by microinjection of the Car+-sensitive photoprotein aequorin into egg cells. In the egg of the medaka fish, the wave of increased [CaZ+], was shown to begin at the point of sperm entry and proceed across the egg within about 2 min (Gilkey et al., 1978). The maximal [Ca”], reached was estimated to be approximately 30 p M . In similar studies with the sea urchin egg. a peak [CaZ+J, of approximately 5 p M was reached within I min after fertilization (Steinhardt el a / . , 1977). In a more recent study, Busa and Nuccitelli (1985) used microelectrode methodology and found that the subcortical [Ca?’], in Xenopus eggs rose from resting levels of 0.40 pM to a peak concentration of 1.2 5 0. I5 p M within 2 min, and returned to prefertilization values within 12 min. In recent years, the development of Ca’+-sensitive fluorescent dyes (quin2 and fura2) that can be loaded into cells as acetoxymethyl esters, and trapped in the cytoplasm by the action of cellular esterases, has revolutionized the measurement of intracellular Car+in smaller cells that cannot be easily microinjected (Grynkiewicz et a l . , 1985; Tsien, 1980, 1983; Tsien et u l . , 1984).This technology has stimulated the accumulation of a wealth of data demonstrating that intracellular Ca?+transients, with peak [Car+],in the micromolar range, are associated with regulated secretion in a wide variety of cells, including platelets, neutrophils, adrenal

52

ROBERT C. JACKSON AND JOSEPH H. CRABB

chromaffin cells, GH3 and primary pituitary cells, PC12 cells, mast cells, and pancreatic cells (Tsien et al., 1984; White et u l . , 1984). The use of quin2 and fura2 in invertebrate eggs has been limited because the relatively low temperatures needed to maintain healthy eggs apparently interferes with the efficient loading of these cells with the “membrane permeant” acetoxymethyl derivatives (Tsien, 1983). Thus, it has been necessary to microinject quin2 and fura2 into eggs. Utilizing this technique Poenie et al. (1985) were able to trace [Caz+Iifluctuations in Lytechinus pictus eggs. They found that the Caz+transient at fertilization peaked at 1.95 -+ 0.16 puM, a value consistent with the earlier aequorin studies of Steinhardt et al. (1977) and the recent microelectrode results of Busa and Nuccitelli ( 1985). A. Phosphoinositide Metabolism and Signal Transduction

An explosion of information has occurred in the last 5 years on the role of a minor class of plasma membrane phospholipid, the phosphoinositides (PIS) [including phosphatidylinositol (PI), phosphatidylinositol 4-phosphate (PIP), and phosphatidylinositol4,S-bisphosphate(PIPI)],in conversion of receptor-mediated signals into intracellular messengers. The PI cycle has been exhaustively reviewed in recent literature (Berridge, 1985; Berridge and Irvine, 1984; Nishizuka, 1984a), and interested readers are directed to these sources. While the details of the PI cycle are beyond the scope of this chapter, it is worth noting that, in brief, PIP and PIP, derive from PI via sequential phosphorylations by specific kinases (Nishizuka, 1984a).Their hydrolysis by receptor-linked activation of PI-specific phospholipase C (also called polyphosphoinositide phosphodiesterase) initiates a metabolic cycle, the products of which serve as intracellular messengers before they are reutilized to replenish PI in the plasma membrane. PIP2,which represents only a few percent of the total PIS in the plasma membrane, is probably the most important of the PIS with regard to intracellular signaling. I . PI-DERIVED SECOND MESSENGERS Receptor-mediated hydrolysis of PIPz by PI-specific phospholipase C generates two bona fide second messengers, inositol 1,4,5-trisphosphate (IPJ and diacylglycerol (DAG). Nishizuka and co-workers (1984b) established that DAG was the endogenous analog of the tumor-promoting phorbol esters and acted intracellularly by activating a Ca2+and phospholipidsensitive protein kinase (protein kinase C). Recent results suggest that

2. CORTICAL EXOCYTOSIS

53

protein kinase C may act synergistically with Ca2+to maximize the exocytotic response (Baker, 1984). In fact, studies combining the use of quin2 and phorbol esters revealed that in certain cells, particularly platelets and neutrophils, treatment with phorbol esters can obviate the requirement for a rise in [Ca2+Jito elicit exocytosis (Baker, 1984; Di Virgilio et al., 1984; Kaibuchi et a / . , 1983; Rink rt a / . , 1983). This indicates that, while elevated [Ca2+]iseems to be sufficient to elicit exocytosis in most cells, it may not be necessary in all cases. Berridge, working on the soluble constituents of PIPz breakdown, first showed that IP3was generated with sufficient rapidity to precede the Ca2+ transient (Berridge. 1983). Next, it was demonstrated that IP3was capable of causing the release of Cali from intracellular stores in permeabilized pancreatic acinar cells and from rat insulinoma microsomes (Prentki r t a l . , 1984; Streb et d., 1983). These results laid the groundwork for subsequent elucidation of the pivotal roles played by PI-derived second messe ngers . The involvement of PIP2 turnover in the generation of the Ca” transient during fertilization was confirmed in a series of reports on activation of sea urchin and Xerioprrs eggs. Turner et a / . (1984) obtained early hints in this regard by demonstrating that the amount of PIP and PIP2increased within IS sec after fertilization, preceding the cortical exocytotic wave. Kame1 et cil. (1985) confirmed Turner’s results, and showed in addition that the amount of PI in an egg declines by about SO95 within 30 sec of fertilization, then rapidly returns to its original value by 2-5 min postinsemination. Concomitant with this decrease in PI. there is a 5-fold increase in [!H]IP3 production (in cells prelabeled with [3H]inositol)during the first 10 min postinsemination. Both sets of results are consistent with the interpretation that PIS are becoming ”primed” for use in egg activation. Whitaker and Irvine (1984) established that IP3 acts as a second messenger in egg activation, by demonstrating that microinjected 1P3 was capable of initiating a cortical exocytotic wave identical to that seen during fertilization. Moreover, they observed that micromolar Ca2+induced PIP, hydrolysis in egg cortex preparations. Clapper and Lee (1985) and Oberdorf et a / . (1986) have since shown that 1P3 is able to stimulate the release of stored Ca2+in egg homogenates and cortex preparations, respectively. Together, these results suggest that the exocytotic wave following sperm binding is the result of an autocatalytic event beginning with receptor-mediated generation of IPj, followed by local Ca” release, further IP3 generation. etc., leading to propagation of the signals around the circumference of the egg.

ROBERT C. JACKSON AND JOSEPH H. CRABB

54

Utilizing double-barreled electrode technology, Busa et al. (1985) were able to directly demonstrate increased [Ca2+]ifollowing IP3 microinjection into Xenopus eggs. Examination of the spatial IP, requirements within the egg indicated that IP3 was most effective when injected into the cortex as opposed to deep within the cytoplasm (Busa et al., 1985). This latter result is consistent with the observation that Xenopus eggs have an extensive cortical reticulum that is thought to contain the Ca2+stores used in cortical exocytosis (Charbonneau and Grey, 1984). Sardet (1984) has observed a similar, though less extensive cortical reticulum in S . purpuratus egg cortex preparations. 6. Role of

G Proteins

While the evidence implicating the PI cycle in intracellular signaling was growing to encompass an increasing variety of cells, independent lines of research were being conducted to determine the link between cell surface receptors and activation of phospholipase C. These studies recognized similarities between PI-mediated signaling and the adenylate cyclaselcyclic AMP system which involves the action of guanyl nucleotidebinding regulatory proteins (G proteins) (Berridge, 1985; Gilman, 1984). In the adenylate cyclase system, there are two G proteins, G, and Gi, which operate through receptors that cause the stimulation and inhibition of adenylate cyclase activity, respectively (Gilman, 1984). These G proteins share many biochemical features. They both consist of three subunits, a,p, and y , with molecular weights of 39,000-45,000, 35,000, and 10,000, respectively. The a subunit is responsible for interaction with adenylate cyclase and contains a high affinity binding site for the guanine nucleotides GTP and GDP as well as a latent GTPase activity. The a subunits also contain sites for NAD-dependent ADP ribosylation by either cholera toxin or islet-activating protein (IAP, a Bordetefla pertussis toxin). The p subunit serves as a regulatory subunit, maintaining the (Y subunit in the resting, or deactivated, state, while the y subunit is uncharacterized at present. The generally accepted chain of events in the cyclase system is as follows. The G proteins contain a tightly bound GDP in the resting state. On receptor stimulation, the a subunit catalyzes the exchange of GTP from the cytosol for bound GDP, and concomitantly releases the p subunit. This a subunit, in the activated conformation, dissociates from the receptor and can then interact with adenylate cyclase and modulate its activity. The latent GTPase activity of a results in hydrolysis of the bound GTP, allowing the p subunit to reassociate, and returning the system to the ground state. Nonhydrolyzable analog of GTP (such as GTPyS,

2. CORTICAL EXOCYTOSIS

55

GppNp. or GppCp) are capable of causing the persistent activation of the G proteins, since the GTPase activity is blocked. Cholera toxin-mediated ADP ribosylation blocks the GTPase activity of G, and results in its persistent activation. IAP-mediated ADP ribosylation of Gi blocks GTP/GDP exchange, thereby blocking activation of Gi and consequently preventing its inhibition of adenylate cyclase (Berridge, 1985). Efforts over the past 2 years have established a role for a G protein in the transduction of Ca?+ signals in a variety of mammalian cell types (reviewed by Gomperts et ml., 1986; Joseph. 1985). The specifics of these studies are beyond the scope of this chapter, but, in brief, the following has been demonstrated: ( I ) GTP analogs (particularly GTPyS) can induce secretion and potentiate exocytosis at low [Ca2‘Ii (Gomperts, 1983; Haslam and Davidson, 1984). (2) Application of GTPyS to plasma membrane preparations has been shown to elicit PIP2.turnover (Cockcroft and Gomperts, 1985; Wallace and Fain, 1985). (3) IAP treatment of intact mast cells resulted in ADP ribosylation of a 41-kDa protein, and blocked agonist-mediated release of histamine, but had no effect on release by A23187 (Nakamura and Ui, 1985). These studies provide strong support for the involvement of a G protein in the P1-linked signal transduction mechanism, but they fall short of demonstrating whether the G protein is authentic Gi or a G protein unique to PI turnover. Is a G protein also involved in the generation of Ca?.’ signals in the sea urchin egg? Turner et uf. (1986) investigated this question and determined that a G protein is indeed involved, at a step prior to IP, production. Using microinjection techniques, they confirmed Whitaker and Irvine’s (1984) result that injection of 1P3resulted in cortical exocytosis (measured as elevation of fertilization envelopes). In addition they demonstrated that injection of EGTA buffers to maintain [Caz+Iiat lo-’ M prevented exocytosis in response to subsequent IP3 injection. This result established that IP3acts at a step prior to generation of the Ca?’ signal. Injection of 28 p M GTPyS also elicited exocytosis at a step prior to Ca?’, generation, since preinjection of EGTA (as above) blocked this response. GDPpS, a competitive inhibitor of GTP exchange in the G proteins, inhibited cortical exocytosis in response to fertilization but not in response to IP3 microinjection. Thus the G protein-dependent step must precede both the 1P3 and Ca2+signals. These results are consistent with the interpretation that a G protein links sperm receptor occupation to activation of phospholipase C and Ca’+ mobilization. In a broader context, they suggest that G proteins may be an integral part of the regulated secretory pathway in invertebrates as well as in vertebrates.

ROBERT C. JACKSON AND JOSEPH H. CRABB

56

111.

In Vitro MODELS OF EXOCYTOSIS

As stated previously, molecular information on the terminal (membrane fusion) stage of exocytosis, is “embarrassingly” sparse (Tartakoff, 1985). The reason is obvious: Membrane fusion is the least accessible step in the exocytotic pathway. It occurs within the cell and is dependent on all previous steps in the process. The difficulty of isolating the fusion stage from the translocation/attachment step is particularly troublesome; for although the cell’s interior can be accessed through microinjection or permeabilization, the fusion stage, in most secretory cells, is absolutely dependent on the translocation/ attachment step. Thus, for example, it is impossible to discern whether the energy requirements observed i n most permeabilized systems (Knight and Baker, 1982) are at the level of translocation/attachment, fusion, or both. A faithful in virro model of exocytosis should surmount this difficulty and yield information on the molecules involved in fusion. To be valid, such an experimental model must meet the morphological and biochemical criteria established for in uiuo exocytosis. So little is known about terminal exocytosis, however, that there are only two well-defined criteria that must be strictly met in order to satisfy the known in viuo requirements. These are ( I ) a [Ca”] threshold in the physiological range, and (2) membrane fusion-mediated vectorial transport of the vesicle contents into a compartment corresponding to the extracellular space. This latter characteristic must be rigorously demonstrated, since release of the vesicle contents can occur merely by vesicle lysis in the absence of any membrane fusion. A. Permeabilized Cell Systems

Permeabilized cell systems represent an intermediate between in uiuo studies and cell-free systems. Since the morphological features of a permeabilized cell remain essentially intact, release of vesicle contents has, in most cases, been assumed to occur by exocytosis without rigorous proof of membrane fusion. However, these important assumptions have been validated by the careful studies of Knight and Baker (1982) on adrenal chromaffin cells that had been permeabilized by the technique of high voltage electric discharge (permeabilized cell systems are reviewed by Baker in Chapter 4, this volume). A particularly important use of the permeabilized cell systems has been

2. CORTICAL EXOCYTOSIS

57

the demonstration of the free calcium concentration required for secretion. Half-maximal exocytosis was found to occur at lCa-”li of I p M in the chromaffin cell (Baker and Knight, 1978). Remarkably similar thresholds have been found in other cell types using this technique (Baker, 1984). This technology has also revealed important requirements regarding metabolic energy and protein involvement in the process. Specifically, Baker’s group has found an absolute requirement for metabolic energy, in the form of Mg-ATP. Omission of Mg-ATP resulted in complete inhibition of exocytosis, even at [Ca2+Iiapproaching 100 p M (Knight and Baker. 1982). Restoration of exocytosis in response to physiologic [Ca2*Ii was achieved by addition of 5 mM Mg-ATP. Other nucleotides could not support the reaction. That the ATP might be needed to maintain an intragranular membrane potential or pH gradient was ruled out, because exocytosis proceeded reasonably well in the presence of trimethyltin, FCCP. and NH,CI, agents that block the granular ATPase and collapse the membrane potential, or pH gradient, respectively. I n a survey of a variety of chemical compounds, Knight and Baker (1982) discovered that N-ethylmaleimide (NEM. a thiol-modifying reagent) at 1OV‘ M abolished exocytosis in response to 10 p M Ca?’. This result provided strong evidence in favor of the involvement of protein(s) in the response to calcium. Moreover, they (and many others) observed inhibition by trifluoperazine. a drug known to interfere with the actions of calmodulin. However, this class of drug inhibits a variety of cellular processes, and is a potent membrane perturbant, so cautious interpretation of this type of data is in order (Mori rt d., 1980; Shier, 1977; Takai rt d., 1981). A variety of other techniques have been developed to transiently or permanently permeabilize cells. These include detergent methods (digitonin or saponin) (Dunn and Holz, 1983: Wilson and Kirshner, 1983).ATP4permeabilization (exploited by Gomperts, 1983. in the GTP studies of mast cell secretion), Sendai virus (at nonfusogenic concentrations) (Barrowman er d.,1986). patch-clamp-type techniques (Lindau and Fernandez. 1986), and staphylococcal a toxin (Ahnert-Hilger rt u1.. 1985). These techniques have their own particular sets of advantages and disadvantages and are reviewed elsewhere (Gomperts and Fernandez. 1985). In short, when these techniques were applied to regulated secretory cells, the requirements observed by Baker’s group (with few exceptions) have been reproduced. The limitations of this general approach are, as discussed previously, that the translocation/attachment step and the fusion step cannot be distinguished.

58

ROBERT C. JACKSON AND JOSEPH H. CRABB

B. Cell-Free Systems

Calcium ion has long been known to fuse liposomes with each other, and with planar phospholipid bilayers (Papahadjopoulos, 1978). These can be considered the simplest in uitro models of membrane fusion, and are oft cited as evidence that exocytosis need not involve protein(s). Unfortunately, the [Ca*‘] requirements for these events are unphysiologically high (typically in the millimolar range), and even the best-tailored systems cannot achieve fusion below 100 pM [Ca2+](Wilschut and Hoekstra, 1984; Diizgiineg 1985). Nevertheless, these studies illustrate an important role that calcium ion may play in exocytosis, i.e., removal of the hydration barrier between phospholipid head groups, thereby promoting close apposition, a prerequisite for fusion (LeNeveu et al., 1976; Parsegian, 1977; Portis et al., 1979; Diizgiineg et ul., 1981; Rand, 1981). There are three or four varieties of in uitro models for exocytosis that utilize cellular fractions: vesicle-vesicle systems (Creutz et al., 1978; Ekerdt et al., 1981), vesicle-plasma membrane systems (Davis and Lazarus, 1979; Konings and DePotter, 1982; Lelkes et al., 1980), vesicleliposome systems (Bental r t ul., 1984), and vesicle lysis, which some consider as a model (Hoffman et al., 1976; Pazoles and Pollard, 1978). The phenomena studied in these systems include Ca?+-inducedvesiclevesicle aggregation (Creutz et a[., 1378), vesicle-vesicle fusion detected by electron microscopy (Ekerdt et al., 1981), vesicle content release on incubation with plasma membranes in the absence (Lelkes et al., 1980) or presence (Davis and Lazarus, 1979; Konings and DePotter, 1982) of Ca*+, and vesicle lysis induced by ATP and anions (Hoffman et al., 1976; Pazoles and Pollard, 1978). The results obtained with these in uitro systems have been generally disappointing. The principal difficulties associated with these systems include (1) inadequate demonstration of membrane fusion-mediated vectorial discharge, (2) Ca2+requirements which are nonexistent of or questionable relevance to the in uiuo situation, and (3) lack of an in uiuo correlate in which to test information derived from the particular cell-free system. The disappointment of investigators with attempts to reconstitute exocytosis in uifro has been succinctly summarized by Burgoyne (1984): “Until an in uitro system is developed which can readily be shown to undergo a Ca2+-dependent release by a demonstrably exocytotic mechanism, the significance of the existing in uitro systems to the in uiuo situation must remain in doubt.”

59

2. CORTICAL EXOCYTOSlS

C. Egg Cortex as an in Vitro Model of Ca2+-TriggeredExocytosis

The principal difficulties associated with the permeabilized cell systems and the aforementioned in v i m models have been largely resolved by the finding that, in sea urchin eggs, subcellular fractions can be obtained which retain their exocytotic apparatus and maintain their capacity to undergo an exocytosis-like reaction in response to applied micromolar Ca? i-

1. EGGCORTEXPREPARATIONS

There are two types of cortex preparation. Cortical lawn (CL) preparations, developed by Vacquier (197.0, consist of fragments of egg cortex 1i.e.. CVs, plasma membrane (PM),and vitelline layer]. A CL is obtained by directing a jet of Ca!+-free buffer across the surface of a polylysinecoated microscope slide that contains a monolayer of adherent egg cells. This procedure shears the eggs and produces a “lawn” of cortical frdgments uniformly oriented with their cytoplasmic face accessible to the medium (Fig. 3). The other preparation is known as the cell surface complex (CSC). This preparation, developed by Detering et ul. (19771, consists of a suspension of purified egg cortex. CSC is prepared by homogenization of a suspension of dejellied eggs, followed by several rounds of differential centrifugation to remove cytoplasmic constituents (Fig. 4).

2 . MORPHOLOGY OF CORTICAL

VESICLE

DISCHARGE

The CL and the CSC preparations are responsive to Ca”. Half-maxima1 release of CV contents, in both preparations, occurs at approximately 3 p M Ca!’ (Moy et a l . , 1983; Whitaker and Baker, 1983) and is accompanied by dramatic morphological changes. The change most readily observed in the light microscope is disappearance of the CVs (compare B and C in Figs. 3 and 4). A fertilization envelope-like structure is concomitantly formed in the CSC preparation. Electron microscopy showed that the morphology of this structure varies to some extent with the buffer composition, but in a buffer closely resembling seawater it is nearly identical to the fertilization envelope of inseminated eggs (Schon and Decker, 1981).

Direct electron microscopic evidence for Ca?+-mediated fusion between CVs and the plasma membrane has been obtained by several investigators. Whitaker and Baker (1983) examined transversely sectioned

60

ROBERT C. JACKSON AND JOSEPH H. CRABB

A

dejellied eggs

I

cortical lawn

(CL1

Fic. 3. Cortical lawn preparation. (A) Procedure for preparing CL. (B) Phase-contrast micrograph of a single cortical fragment in a cortical lawn preparation. (C) Same fragment as in B, after treatment with a buffer containing 12 p M free Ca?'. Bar, 10 prn.

preparations of Echinus esculentus CL that had been bathed in Ca*+containing buffers. They observed that discharge results in the release of CV contents on the extracytoplasmic face of the PM, and the formation of a perivitelline-like space that is both topologically and morphologically equivalent to the perivitelline space of the fertilized egg. Utilizing rotary shadowed replicas of Ca*+-stimulatedS. purpurutus CL, Chandler (1984) and Zimmerberg et al. (1985) observed the formation of structures (membranous domes) indicative of the incorporation of CV membrane into underlying plasma membrane. Both sets of images suggest that CV discharge in CL preparations is the in uirro equivalent of exocytosis. The

2. CORTICAL EXOCYTOSIS

61

cel I surface complex

(CSC) FIG.4. Cell surface complex preparation. ( A ) Procedure for preparing CSC. ( B ) Phasecontrast micrograph of a single cortical fragment in a csc preparation. (c)A different cortical fragment from the same CSC preparation. after treatment with a buffer containing 12 p M free Ca”. Bar. 10 pm.

thin-sectioned samples emphasize the vectorial nature of exocytosis while the rotary shadowed samples emphasize membrane fusion. Two factors complicate electron microscopic observation of CV discharge in these preparations. First, the high density of CVs on the sutface of the CL results in a massive reaction that often obscures the behavior of individual vesicles. Second, because an individual CV must be favorably

62

ROBERT C. JACKSON AND JOSEPH H. CRABB

oriented in time and space in order to visualize fusion, the number of CVs that can be clearly shown to fuse with the PM is a small fraction of the total, Consequently, it is not possible to ascertain, from electron microscopic data alone, whether exocytosis (i.e., CV-PM fusion) is the mechanism by which the bulk of the CVs discharge their contents. That the images observed in the electron microscope are indeed representative of the mechanism by which the bulk of the CVs release their contents has been strengthened by evidence obtained with an immunofluorescence-based vectorial transport assay (Crabb and Jackson, 1985). Analysis of “sparse” CLs (i.e., CLs from which most CVs have been removed by shear, so that discrete CVs can be observed) showed that CV contents are vectorially discharged into a membrane-bounded compartment (beneath the plasma membrane) where they were incapable of combining with antihyalin antibodies (Fig. 5A). Detergent disruption of this plasma membrane-bounded compartment permitted the CV contents to combine with the antibodies (Fig. 5B). The fact that essentially all the CVs in the sparse CL preparations released their contents by the same mechanism ( i s . , CV-PM fusion and vectorial discharge) suggests that the isolated instances of vectorial discharge observed by electron microscopy are representative of the mechanism by which the bulk of the CVs react. Based on these results it seems probable that CV discharge in the CL and CSC preparations is the in uitro equivalent of exocytosis; thus we have chosen to use these terms interchangeably in the remainder of this chapter. However, to prove the equivalence of CV discharge and exocytosis it will ultimately be necessary to identify molecules responsible for the in uitro reaction and to determine whether the same molecules are also essential for exocytosis in uiuo. 3. ASSAYSFOR CORTICAL VESICLEDISCHARGE

The identification of essential molecules will require reliable, quantitative assays for in uitro exocytosis in the CSC and CL system. Several are available. Two groups independently established turbidimetric assays for exocytosis (Haggerty and Jackson, 1983; Sasaki and Epel, 1983). Both variations are based on the observation that CV discharge results in a sharp decrease in the turbidity of a CSC suspension. Since the magnitude of the decrease in turbidity has been shown to be directly proportional to the extent of CV exocytosis, the turbidity assay can be used quantitatively (Haggerty and Jackson, 1983). The turbidity assay is both quick and easy to perform; however, it is compatible only with the CSC preparation, and care must be exercised in its use since conditions that result in aggre-

2. CORTICAL EXOCYTOSIS

63

FIG. 5 . Cortical vesicle contents are vectorially transported across the plasma membrane. Sparse CLs (i.e., CLs from which most CVs had been removed) were treated with a buffer containing 44 @free Ca?' and then fixed with 1% glutaraldehyde. Lawns with intact membranes (A) and lawns whose membranes had been disrupted with 0 5 % ( v h ) Triton X100 after fixation (B)were processed for irnniunofluorescence with a CSC-absorbed polyclonal antihyalin antiserum, at a dilution of 1 : 20. The paired micrographs in A and B are phase-contrast (left) and immunofluorescent (right) images of the same field. Bar. 10 pm. (Reproduced from Crabb and Jackson, 1985.)

gation or fragmentation of CSC (e.g., vigorous shaking) interfere with the assay. CV exocytosis can also be followed by the release of CV enzymes. Two such assays have been developed. In one assay, the amount of released ovoperoxidase (and hence exocytosis) is quantitated by radioiodination of endogenous substrates (Haggerty and Jackson, 1983; Jackson e t al., 1985). In the other, the amount of released p-1,3-glucanaseis quantitated

64

ROBERT C. JACKSON AND JOSEPH H. CRABB

by hydrolysis of an exogenous substrate (Moy et ul., 1983). Although the P-glucanase assay was devised for use with CLs, and the ovoperoxidase assay for use with CSC, both assays could, in principle, be used with either preparation. The enzyme release assays are more tedious than the turbidimetric assay, and substances that inhibit ovoperoxidase or P-glucanase will, of course, interfere. A microscopic assay, devised by Zimmerberg ef d.(1985), can be used to quantitate the extent of CV exocytosis in CLs. The assay is based on the observation that the individual CVs in a CL preparation are essentially the only structures that scatter light when viewed by dark-field microscopy. Fusion of CVs with the plasma membrane reduces the number of scattering centers and decreases the amount of scattered light. The magnitude of the CaZt-triggered decrease in light scattering has been shown to be directly porportional to the extent of CV exocytosis (Crabb and Jackson, 1986). This microscopic assay has two disadvantages: it requires specialized equipment, and it can be very tedious because only one sample can be processed at a time. On the other hand, this is the only available assay that allows simultaneous visualization and quantitation of the sample. 4. INHIBITORS Attempts to unravel the molecular mechanism of exocytosis in the CL and CSC preparations have not yet produced substantial results. A variety of inhibitors have been identified, but the process of elucidating their molecular targets has just begun. u. Thiol Reagents.

NEM and other thiol-modifying reagents have been shown to block exocytosis in the CL and CSC preparations (Haggerty and Jackson, 1983). NEM also blocked the cortical reaction in intact eggs, whereas 5,5’-dithiobis(2-nitrobenziocacid) (DTNB), a membrane-impermeant thiol reagent, did not (K. K. Ward and R . C. Jackson, unpublished). These results demonstrate that cortical exocytosis can be inhibited by modification of a sulfhydryl(s), accessible at the cytoplasmic surface of the CSC. They suggest that the protein containing this sulfhydry1 group(s) may play a role in exocytosis. The mode of inactivation with NEM is interesting. Under mild conditions, NEM inhibits cortical exocytosis by increasing the threshold Ca2+ concentration required for exocytosis, without affecting the magnitude of the response at higher Ca2t concentrations. When the NEM modification conditions are more severe, exocytosis is blocked even at high Ca2+concentrations (Jackson et ul., 1985). This mode of inhibition suggests that

2. CORTICAL EXOCYTOSIS

65

the sulthydryl group(s) responsible for inactivation may be involved in the mechanism that imparts sensitivity to micromolar Ca” .

6 . Protecrses. Like NEM, trypsin and pronase also inhibit exocytosis by progressively increasing the threshold Ca?‘ concentration required to initiate exocytosis (Jackson ef 01.. 1985). Mild tryptic digestions have little effect on both exocytosis and CSC morphology (i.e., CVs are not released from the CSC), whereas prolonged digestions, of the sort required to totally inactivate the CSC, result in the appearance of individual CVs (Jackson et d.,1985). It remains to be determined whether these observations are merely coincidental, or whether the inhibition of exocytosis is the direct result of CV detachment. While investigating the possibility that proteolytic inactivation could be due to digestion of an essential sulfhydryl-containing protein, it was discovered that NEM-treated CSC could be reactivated by mild tryptic digestion (Jackson et d., 1985). Although a straightforward molecular interpretation of this phenomenon is currently not possible. it may prove useful in identifying the NEM target protein. c . Polycutions. Polycations (e.g., polylysine) have recently been shown to inhibit exocytosis in the CL preparation (Crabb and Jackson, 1986). Since polycations inhibit a wide range of cellular processes, this observation does not help elucidate the molecular mechanism of exocytosis. However, it may explain the finding that “aged” CLs (which are prepared on polylysine-coated surfaces) lose their sensitivity to microcmolar Ca” (Moy et al., 1983). Regarding the use of CL preparations. it is important to note that “aging” requires a period of 15-20 min before a noticeable increase in the Ca?’ threshold can be detected (Crabb and Jackson, 1986). Thus, observations made soon after CL preparation should not be affected by the “aging” phenomenon. d . Phenofhiuzines. Several investigators have demonstrated that the phenothiazine drug. trifluoperazine (TFP, a calmodulin antagonist), inhibits exocytosis in the CL and CSC preparations (Baker and Whitaker, 1979: Haggerty and Jackson, 1983; Moy ef ( I / , , 1983). However, as noted above, TFP inhibition data must be interpreted with caution since phenothiazine drugs are potent membrane perturbants (Naccache rf d., 1980; Seeman, 1972) and have been shown to inhibit enzymes that are not calmodulin regulated (Chau and Tai. 1982; Mori ef NI., 1980; Shier, 1977; Takai et a / . , 1981). Additional evidence in support of the calmodulin hypothesis was provided by the report (Steinhard and Alderton, 1982) that exocytosis in CL

66

ROBERT

C.JACKSON AND JOSEPH H. CRABB

preparations can be inhibited by incubation with high concentrations of an affinity-purified anticalmodulin antibody. However, since this work was performed before it was realized that cortical exocytosis is inhibited by the polylysine used in CL preparation (Crabb and Jackson, 1986), the potential for artifact must be reconsidered. We have determined that exocytosis in the CSC preparation is not inhibited by melittin or R 24571 (Crabb and Jackson unpublished), two of the most potent calmodulin antagonists currently available (SellingerBarnette and Weis, 1984; Van Belle, 1981). We have also been unsuccessful in our attempts to inhibit exocytosis in CSC with several samples of anticalmodulin antisera (Stacy and Jackson, unpublished). In addition, Gilkey (1983) has reported that microinjected anticalmodulin IgG failed to inhibit exocytosis in medaka eggs. e. KCI Extraction. Sasaki (1984) has reported that a KCI-extractable proteinaceous factor is required for exocytosis at micromolar Ca2+ in Hemicentrotus pulcherrimus (but not S . purpuratus) CSC. The molecular weight of the crude factor (approximately 100,000, by gel filtration chromatography) suggests that it is not calmodulin. 5. SUBFRACTIONATION OF T H E EGGCORTEX

A second approach to understanding the mechanism of CV discharge entails subfractionating the cortex into its component parts (i.e., plasma membrane and CVs), analyzing each, and reassembling them into active cortex. a . Purified Cortical Vesicles. Several CV purification procedures have been published. Schuel et al. (1972) used zonal centrifugation to prepare CVs from egg homogenates. The purified CV fraction obtained by this method was enriched in /3-1,3-glucanase activity, and it has been used in determining the chemical and enzymatic composition of CV contents (reviewed by Schuel, 1978, 1985). Recent CV purification procedures have utilized egg cortex as starting material. Kopf et al. (1982) prepared CVs from S.purpuratus eggs. Their method involves mechanically dislodging CVs from large cortical lawns (prepared in petri dishes) by repeatedly squirting a jet of a high ionic strength, dissociative buffer across the surface of the lawn. Dislodged CVs are collected by centrifugation. Decker and Kinsey (1983) prepared CVs from Lytechinus variegatus eggs. Their procedure is based on dislodgement of CVs from CSC by gentle homogenization in a low ionic strength, isotonic buffer. CVs are separated from other components of the

67

2. CORTICAL EXOCYTOSIS

homogenate by a two-step procedurc involving ultracentrifugation on a sucrose step graident. I n the procedure reported by Crabb and Jackson (1985). CVs are dislodged from S . pirrpirrutrrs CSC by gentle homogenization in an isotonic buffer with a pH of 9.1, and purified by low speed centrifugation. Marker enzyme analysis indicated that all three procedures yielded CV fractions that were enriched in CV proteoesterase activity and had substantially reduced levels of a plasma membrane marker (Na‘ ,K+-ATPase). Decker and Kinsey (1983) have analyzed the protein and phospholipid Compositions of their CV preparation. CV proteins were found to be rich in high molecular weight glycoproteins. CV lipids were enriched in arachidonic acid and contained 2.3 times as much cholesterol as egg plasma membrane. h. Piirijied P/usmu Membrane. Crabb and Jackson ( 1985)and Kinsey et ul. (1980) have devised plasma membrane purification procedures. The procedure of Crabb and Jackson (1985) is a microprocedure, designed specifically for use in reconstitution experiments (see below). CVs are mechanically disloged from CL preparations with a jet of isotonic buffer. The resulting PM “lawns” are made up of CV-free fragments of PM attached via their vitelline layer to a polylysine-coated slide or coverslip. The procedure of Kinsey et al. (1980) produces a suspension of CV-free plasma membrane sheets. The PM fraction was substantially enriched in Na ,K+-ATPaseand contained reduced levels of CV proteoesterase. The egg PM has an unusually high content of phosphatidylinositol (25%. on a molar basis) and, like the CV membrane, is enriched in arachidonic acid (Decker and Kinsey, 1983). +

6. RECONSTITUTIONOF ACTIVECORTEX PLASMAMEMBRANE

FROM PURIFIED

CVS A N D

Although the CSC and CL preparations are exocytotically competent, their usefulness in investigations of the molecular basis of exocytosis is limited because the CVs and the PM cannot be independently manipulated. Recently, Crabb and Jackson ( 1985) have reported that exocytotically active egg cortex can be reconstituted from purified CVs and egg PM. Reconstituted cortex was prepared by recombining purified CVs with a P M lawn to form a reconstituted lawn preparation (Fig. 6). CV discharge in reconstituted lawn preparations required a 3- to 4-fold higher concentration of Ca’+ than control cortical lawns. Purified CVs that were not attached to a PM lawn were shown to remain intact, even in buffers containing high concentrations of Ca?’ ( I .O mM).

68

ROBERT C. JACKSON AND JOSEPH H. CRABB

FIG.6. Reconstituted cortical lawns and their response to 44 pM free Ca?'. A PM lawn was prepared by dislodging CVs from a CL with a jet of isotonic buffer. A representative field, containing a single PM fragment, was selected and photographed (A). Subsequent steps were performed with the sample in place on the microscope stage so that the same field could be followed throughout the experiment. (B) Reconstituted lawn formed when a suspension of purified CVs was applied to the PM lawn, allowed to bind for 10 rnin, and thoroughly washed with an isotonic buffer. (C, D) Same reconstituted lawn 5 and 60 sec, respectively, after addition of the Ca?'-containing buffer. Bar, I 0 pm. (Reproduced from Crabb and Jackson, 1985.)

Using the immunofluorescence-based vectorial transport assay described above (Section III,C,2), it was shown that CV discharge from a reconstituted lawn resulted in the vectorial transport of hyalin from the cytoplasmic to the extracytoplasmic face of the egg plasma membrane (Fig. 7). These results constitute strong evidence in support of the contention that Ca2+-stimulateddischarge of CV contents in a reconstituted lawn is the in v i m equivalent of exocytosis. Nevertheless, it will be necessary to reconfirm vectorial transport with another technique (ens.,electron microscopy). It will also be necessary to determine whether a CV can fuse

2.CORTICAL EXOCYTOSIS

69

Fic,. 7 . Cortical vesicle contents are vectorially transported across the plasma membrane in reconstituted lawns. Reconstituted lawns were prepared. reacted with it buffer containing 44 wh4 free Ca?*. and tixed with 1% glutaraldehyde. Antihyalin immunofluorescence was performed with CSC-absorbed antihyalin antiserum at a diluation of I : 40. ( A ) Paired phasr-contrast and immunotluorescent images of a reconstituted cortical fragment with intact membranes. (B)Paired phase-contrast and immunofluorescent images of a reconstituted cortical fragment with membranes disrupted by Triton X-100.Bar. 10 pn. (Reproduced from Crabb and Jackson, 198s.)

with any membrane to which it binds, or whether there is specificity in the reassociation and/or Ca”-dependent fusion stages. As a first step in this direction, a method of quantitating CV binding has been devised which was used to show that S. pi4rpirrrrtirs CVs do not reassociate with PM lawns prepared from human red blood cells (R.C. Jackson and P. Modern, unpublished).

70

ROBERT C. JACKSON AND JOSEPH H. CRABB

7. THECORTICAL VESICLE-PLASMA MEMBRANE JUNCT~ON

Elucidation of the structures responsible for binding CVs to the PM should shed light on the function of junctional proteins in exocytosis, and may help unravel the roles of ATP and Ca2+in exocytosis. The possibility that microfilaments or other cytoskeletal structures anchor CVs to the PM is attractive. Actin and other cytoskeletal proteins have been localized to the egg cortex (Bryan, 1982; Burgess and Schoeder, 1977; Otto et al., 1980; Whitaker and Baker, 1983), and filamentous structures approximately 6 nm in diameter have been observed in association with the CV and PM surfaces in both intact eggs and isolated cortex (Chandler, 1984). The similarity of the extraction conditions that have proved effective in releasing CVs from isolated CSC [i.e., 0.6 A4 KI (Vacquier, 1976); elevated pH (Crabb and Jackson, 1985; Hylander and Summers, 1981); low ionic strength sucrose buffers (Decker and Kinsey, 1983)l and the conditions used to disrupt microfilaments (Carraway et al., 1982; Pollard, 1982; Pollard et al., 1974) is also suggestive. On the other hand, it has been reported that cytochalasin B does not disrupt the filamentous connections between CVs and PM (Chandler, 1984), and attempts to inhibit cortical exocytosis in isolated CLs with microfilament and microtubule toxins have been unsuccessful (Whitaker and Baker, 1983). Clearly, it is too early to draw any conclusions regarding the nature or identity of the structures that comprise the CV-PM junction

D. Hypothetical Models

Several hypothetical models for the terminal events in exocytosis have been propsed, but none has gained universal acceptance. Four of the more prominent hypotheses and their relationship to cortical exocy,tosis are considered below.

I . OSMOTIC FORCES The osmotic hypothesis is based on the observation that exocytosis is generally accompanied by a concomitant swelling of the secretory vesicle (reviewed by Finkelstein et al., 1986). The hypothesis suggests that osmotic swelling of the secretory vesicle provides the driving force for vesicle-plasma membrane fusion. Using mammalian cells, Pollard and co-workers have amassed a substantial amount of circumstantial evidence in favor of this hypothesis

2. CORTICAL EXOCYTOSIS

71

(Pollard of a/.. 1979; see also Brocklehurst and Pollard, Chapter 7, this volume). Cohen et t i / . (1982) and Zimmerberg Pf a / . (1980) have presented evidence supporting a role for osmotic forces in the fusion of liposomes with planar phospholipid bilayers. Using sea urchin eggs, Zimmerberg and Whitaker (1985) have shown that CV discharge both in intact eggs and in CL preparations is inhibited by hyperosmotic buffers. Membrane capacitance experiments with intact eggs indicated that increased osmotic strength had inhibited membrane fusion and not just dispersal of CV contents. Since CLs prepared in hyperosmotic, Ca”-containing buffer underwent CV discharge when returned to an isosmolar buffer, (even in the absence of Ca”), it was suggested that Ca?+may produce an irreversible change that primes CVs for exocytosis. Support for this proposal derives from the observation that the CVs in a CL prepared in hyperosmotic, Ca”-containing buffer had a mean vesicular diameter slightly larger than &‘+-free controls (Zirnmerberg and Whitaker, 1985). On the other hand, Holz and Senter ( 1986) have reported that digitonin-permeabilized chrornaffin cells incubated in hyperosmotic Ca?’ -containing buffer do not undergo exocytosis when returned to an isosmotic &?+-free buffer. Furthermore, depsite considerable effort, the search for a mechanism that might be responsible for unleashing intravesicular osmotic forces prior to exocytosis has not been fruitful. In fact, most mechanisms have been eliminated. Thus, in chromaffin cells it has been shown that ( I ) the absence of anions and monovalent cations does not inhibit exocytosis in intact or permeabilized cells, ( 2 ) reagents that collapse the H’ electrochemical gradient across the chromaffin granule membrane are not inhibitory, and (3) hyposrnotic buffers do not induce or enhance exocytosis from intact or permeabilized cells (see Baker and Knight, 1984, and Holz, 1986, for recent critiques of the osmotic hypothesis). In brief, while the osrnotoic hypothesis is attractive, it remains difficult to determine whether hydration and swelling of secretory vesicles occurs before or after membrane fusion. 2 . METABOLIC ENERGY The observation that metabolic poisons inhibit exocytosis in a variety of secretory cells (Becker and Henson, 1973; Feinstein and Schramm, 1970; Jamieson and Palade, 1971; Peterson, 1974) has long suggested that exocytosis is an energy-requiring process. This hypothesis is supported by the fact that most (Bader et a / . , 1986; Barrowman r f u / . , 1986; Dunn and Holz, 1983; Ronning and Martin, 1986; Wilson and Kirshner, 1983), but not all (Ahnert-Hilger et al., 1985; Ruggiero rt al., 1985; Smolen r f u / . ,

72

ROBERT C. JACKSON AND JOSEPH H. CRABB

1986), permeabilized mammalian cell systems require ATP for exocytosis. Exocytosis in the sea urchin egg has also been reported to be inhibited by metabolic poisons (Baker and Whitaker, 1978; Okazaki, 1956); however, unlike most mammalian cells, permeabilized eggs, prepared either from poisoned or control eggs, did not require ATP for exocytosis (Baker and Whitaker, 1978). Likewise ATP is not required for CV discharge in the CL preparation (Baker and Whitaker, 1978). ATP has been reported to slow “aging” and maintain the micromolar Ca2+ sensitivity of CL preparations (Moy er ul., 1983), but as discussed above the “aging” phenomenon is probably caused by the polycations used in preparing CLs and, therefore, may not be physiologically relevant. Inconsistencies among the reported ATP requirements of the various permeabilized mammalian cell types, as well as the “aging” phenomenon, prevent a straightforward assessment of these results. However, there appear to be two paradoxes: (1) Permeabilized mammalian cells seem to require ATP for exocytosis, but permeabilized eggs do not. (2) Permeabilized eggs as well as the CL and CSC fractions do not require metabolic energy for exocytosis, yet intact eggs are inhibited by metabolic poisons. The first of these paradoxes may be the result of the unique morphology of the egg. The vesicle translocation and attachment steps of exocytosis (Fig. 1, steps 3 and 4) are not required for cortical exocytosis in the egg; in mammalian cells they are. If these steps are responsible for the ATP requirement of mammalian cells the paradox can be dispelled. The second paradox could be explained if metabolic poisoning of the egg produced an inhibitory substance that was lost on permeabilization or subfractionation of the egg. Clearly, the role of metabolic energy in exocytosis is an area that warrants further study.

3. THEVIRALMODEL The instance of membrane fusion that is best understood at the molecular level is the fusion of an enveloped virus with its host cell. In the acidic environment within a host cell endosome (pH 5.0-5.5), one of the surface proteins of enveloped viruses undergoes a pH-induced conformational change that exposes a hydrophobic domain, resulting in fusion of the viral and host cell membranes (Skehel et al., 1982; White er al., 1983; see also Chapters 9-1 1, this volume). If exocytotic fusion is comparable to viral fusion, a Ca2+-inducedconformational change in a secretory vesicle or plasma membrane protein might be responsible for membrane fusion. This is an attractive model, but it is also one that is difficult to investigate

73

2. CORTICAL EXOCYTOSIS

experimentally. Thus, there is currently no direct evidence either for or against it. 4. PHOSPHOLIPASES The phospholipase hypothesis suggests that fusogenic molecules generated by Ca”-stimulated phospholipase hydrolysis of membrane phospholipids are required for fusion of secretory vesicles with the plasma membrane. Both phospholipase A? and phospholipase C have properties that are compatible with a role in exocytosis. Phospholipase A: requires Ca?+ for activity (Van Den Bosch, 1980) and produces known fusogens: lysophospholipids (Lucy, 1978) and free fatty acids (Creutz, 1981). Phospholipase A? activity has been detected in egg homogenates, and it has been reported that quinacrine, a phospholipase inhibitor, blocks the cortical reaction in intact eggs (Ferguson and Shen, 1984). As described above, phospholipase C is intimately involved in the production of the Ca2+ signal that is responsible for exocytosis. DAC, a product of phospholipase C hydrolysis of PIP?, is known to satisfy one requirement for membrane fusion, i.e.. it distorts the molecular packing of phospholipid bilayers (Das and Rand, 1984). Another way that phospholipase C activity may promote exocytosis is through reduction of the membrane surface charge. Hydrolytic release of multiply charged inositol phosphate groups from the membrane surface should reduce hydration of the polar phospholipid head groups and thereby decrease the energy barrier to close apposition of lipid bilayers (LeNeveu ef NI., 1976; Parsegian, 1977; Rand, 1981). Whitaker and Aitchison (1985) have shown that &!+-activated CV discharge in CL preparations is accompanied by PI turnover. Both PI turnover and CV discharge had comparable Ca?’ requirements, and neomycin (a polycationic inhibitor of polyphosphoinositide hydrolysis) was shown to inhibit exocytosis in the CL preparation. Additional experiments will be required to determine whether neomycin inhibition results from decreased PI turnover or from other actions of the drug. Nevertheless, the possibility that DAG is involved in exocytosis is attractive: It is known to destabilize membranes and is generated in a timely manner at the appropriate cellular location by a Ca”-stimulated enzyme. IV.

CONCLUSIONS

The fundamental mechanism of Ca?’ signal generation in the sea urchin egg and in mammalian cells appears to be virtually identical: Receptor

74

ROBERT C. JACKSON AND JOSEPH H. CRABB

occupation results in the sequential activation of a G protein and P1specific phospholipase C. IP3 produced by phospholipase C-catalyzed hydrolysis of PIP2acts to release Ca2+from internal stores. The details of the various steps in this sequence (i.e., the linkage between sperm receptor occupation and activation of G protein, the mechanism by which activated G protein stimulates phospholipase C, and the details of Ca2+ gating by IP3) need to be resolved. It is hoped that similarities between signal transduction in the 1P3 and CAMPsystems will continue to facilitate these studies. Studies of the molecular mechanisms of the terminal steps in exocytosis are in their infancy. Several models have been proposed, but the fundamental mechanism of membrane fusion remains unknown. The egg cortex constitutes a particularly advantageous model system for studying exocytotic membrane fusion. The available data strongly suggest that CV discharge in the CSC and CL preparations is the in vitro equivalent of exocytosis in the egg: however, ultimate validation of this in vitro approach will come only through the identification of proteins that are required for exocytosis both in vitro and in vivo. V.

ADDENDUM

Since completing work on this review a number of pertinent articles have been published or have come to our attention. A. Sea Urchin Egg

Turner et al. (1987) have shown that microinjection of cholera toxin into unfertilized sea urchin eggs stimulates cortical exocytosis, via a CAMP-independent mechanism. Cholera toxin treatment of egg CSC resulted in ADP ribosylation of a 47-kD polypeptide; whereas pertussis toxin catalyzed the ADP-ribosylation of a 40-kD polypeptide. These results suggest that the signal transduction mechanism for exocytosis in the egg relies on a G protein that has a different toxin sensitivity than observed in most mammalian secretory cells. Oinuma et al. (1986) have partially purified the pertussis toxin substrate protein from sea urchin eggs, and shown that its 37-kD beta subunit binds antibodies raised to the betdgamma subunits of rat brain G-proteins. A pertussis toxin substrate protein has also been identified in sea urchin sperm (Kopf et al., 1986). Trimmer and Vacquier (1986) have published a review on the activation

2. CORTICAL EXOCYTOSIS

75

of sea urchin gametes that includes sections on exocytosis in both the sperm (acrosome reaction) and egg (cortical reaction). B. Other Cell Types

Based on results obtained with neutrophils (Barrowman e f al., 1986), Gomperts (1986) has proposed that a GTP-binding protein may participate in the terminal stages of exocytosis (i.e.?after the increase in cytoplasmic concentration of Ca?'). Vallar e f ul. (1987) have obtained similar results with an insulin-secreting cell line. Knight et al. (1985) have shown that adrenal chromaffin cells that have been treated for several days with botulinum toxin will not secrete in response to various secretagogues, or when permeabilized and challenged with Ca2+buffers. The long incubation time required for inactivation was apparaently due to the slow rate of entry of the toxin into the cells, since Penner et al. (1986)were able to obtain inactivation within 1 hr by directly microinjecting botulinum or tetanus toxins into cells. The mechanism of action of botulinum and tetanus toxins is not known; however, related bacterial toxins inactivate cellular processes through ADP ribosylation of GTP-binding proteins. Ohashi and Narumiya ( 1987) have observed that treatment of a crude membrane faction from bovine adrenal gland results in ADP-ribosylation of a 21-kD polypeptide. The relationship of this polypeptide to GTP-binding proteins and to exocytosis remains to be determined. Perrin et al. (1987) demonstrated that the exocytotic release of catecholamine from digitonin-permeabilized adrenal chromaffin cells can be partially inhibited by preincubation with anti-fodrin antibodies. Their results suggest that rearrangement of the cytoskeletal network of the cell may be required for efficient exocytosis. With regard to the various models for exocytosis, two groups (Zimmerberg et al., 1987; Breckenridge and Almers, 1987) have presented cogent data in opposition to the osmotic hypothesis. By simultaneously monitoring capacitance and observing the morphology of mast cells from the beige mouse, it was demonstrated that the capacitance increases (i.e., membrane fusion) precede swelling of secretory vesicles. Plattner and colleagues (Vilmart-Seuen e f al., 1986) have divised an in uitro system for investigating the exocytotic release of trichocysts from Parumecirrm. Paramecium is similar to the sea urchin egg in that secretory vesicles are firmly attached to the plasma membrane; hense, Parumecium cortex, like sea urchin egg cortex, is exocytotically active. Evidence from both in uiuo and in uitro experiments (Momayezi e f a l . , 1987)

76

ROBERT C. JACKSON AND JOSEPH H. CRABB

suggest that the exocytotic release of trichocysts may involve a protein phosphatase, perhaps calcineurin. ACKNOWLEDGMENTS We would like to thank Paul Modern and Mary Harrington for helping to prepare the manuscript. The authors' research was supported by Grant GM 26763 from the National Institutes of Health. REFERENCES Ahnert-Hilger, G., Bhakdi. S . . and Gratzl, M . (1985). Minimal requirements for exocytosis. J . B i d . Chern. 260, 12730-12734. Alliegro, M . C.. and Schuel. H. (1985). Characterization of a soybean trypsin inhibitor sensitive protease from unfertilized sea urchin eggs. Biochemistry 24, 3926-393 I . Anderson, E. ( 1968). Oocyte differentiation in the sea urchin, Arhacitr punctitlotu. with particular reference to the origin of cortical granules and their participation in the cortical reaction. J . Cell Biol. 37, 514-539. Anderson, E. (1974). Comparative aspects of the ultrastructure of the female gamete. fnr. Rev. C y t d . Sitppl. 4, 1-70. Bader, M.-F., Thierse. D.. Aunis. D., Ahnert-Hilger, G., and GrdtZl. M. (1986). Characterization of hormone and protein release from ru-toxin-permeabilized chromaffin cells in primary culture. J . Biol. Chetn. 261, 5777-5783. Baker. P. F. (1984). Multiple controls for secretion'? Nuture (London) 310, 629-630. Baker, P. F.. and Knight, D. E. (1978). Calcium-dependent exocytosis in bovine adrenal medullary cells with leaky plasma membranes. Nutitre (London) 276, 620-622. Baker. P. F.. and Knight, D. E. (1981). Calcium control of exocytosis and endocytosis in bovine adrenal medullary cells. Philos. Truns. R . Soc. London Ser. B . 296, 83-103. Baker, P. F . . and Knight, D. E. (1984). Chemiosmotic hypothesis of exocytosis: A critique. Biosci. Rep. 4, 285-298. Baker, P. F., and Whitaker. M. J . (1978). Influence of ATP and calcium on the cortical reaction in sea urchin eggs. Nutiire (London) 276, 513-Sl5. Baker. P. F., and Whitaker. M. J. (1979). Trifluoperazine inhibits exocytosis in sea urchin eggs. J . Physiol. (London)298, 55P. Barrowman, M. M.. Cockcroft. S . , and Gomperts, B. D. (1986). Two roles for guanine nucleotides in the stimulus-secretion sequence of neutrophils. Nuture (London) 319, 504-507. Becker, E. L . , and Henson. P. M. (1973). I n uirro studies of immunologically induced secretion of mediators from cells and related phenomena. Adu. fmmrrnol. 17, 93-193. Bendayan, M., Roth, J . , Perrelet. A,, and Orci, L. (1980). Quantitative immunocytochemical localization of pancreatic secretory proteins in subcellular compartments of the rat acinar cell. J . Histochrm. Cytochem. 28, 149- 160. Bental, M., Lelkes, P. 1.. Schloma. J., Hoekstra, D.. and Wilschut. J. (1984). Ca'+-lndependent, protein-mediated fusion of chromaffin granule ghosts with liposomes. Biochim. Biophys. act^ 774, 296-300. Berridge, M. J. (1983). Rapid accumulation of inositol trisphosphate reveals that agonists hydrolyze polyphosphoinositides instead of phosphatidylinositol. Biochem. J . 212, 849-858.

Berridge, M. J . (1985). The molecular basis of communication within the cell. Sei. A m . 253, 142- 152. Berridge, M. J . , and Irvine, R. F. (1984). Inositol trisphosphate, a novel second messenger in cellular signal transduction. Nuture (London) 312, 315-321,

2. CORTICAL EXOCYTOSIS

77

Breckenridge. L. J . . and Almers. W . (1987).Final steps in exocytosis observed in ii cell with giant secretory granules. /‘roc,. N i i t l . Accctl. S c i . U . S . A . 84, 194.5-1949. Bryan. J . (1982). redistribution of actin and fascin during fertilization in sea urchin eggs. C ~ l lDiffer. . 11, 279-280. Burgess. D. R . . and Schroeder. T. B. (1977). Polarized bundles of actin filaments within microvilli of fertilired sea urchin eggs. ./. Cell Biol. 74, 1032-1037. Burgoyne. R. I>. (1984). Mechanisms of secretion from adrenal chromaffin cells. Biochim. Biopk!.~. Actti 779, 201-216. BUSI. W. B., and Nuccitelli. R. (1985). A n elevated free cytosolic Ca?* wave follows fertilization in eggs of the frog Xer7opirs /irc.ui.\. J . Cell Biol. 100, 1325- 1329. B U S ~ IW, . B.. Ferguson, J. E . , Joseph. S . K.. Williamson, J . K..and Nuccitelli. R . (1985). Activation of frog (Xenopirs /treui.c) eggs by inositol trisphosphate. I . Characterization of Ca” release from intracellular stores. J . Cell Biol. 101, 677-682. Carraway. K . L.. Cerra, R. F., Jung. C . . and Carraway. C. A . C. (1982). Membraneasmciated actin from the microvillar membranes of ascites tumor cells. J . Cell Biol. 94, 624-630. Carroll, E. J . , Jr.. and Epel. D. (1975). Isolation and biological activity of proteases released by sea urchin eggs following fertilization. Dcu. Eic~l.44, 22-32. Chambers, E. L . , Pressman, B. C.. and Rose. B . (1974).The activation of sea urchin eggs . Conrmrtrr. 60, by the divalent ionophores A23 I87 and X-537A. Bioc.liem. B i o p h ~ Res. 126- 132. Chandler. D. (1984). Exocytosis ir7 t ~ i t Ultrastructure ~ : of the isolated sea urchin egg cortex as seen in platinum replicas. J . Ul/rtr.s/rirc./.Rex. 89. 198-21 I . Chandler-. D. E.. and Heuser. J. (1979). Membrane fusion during secretion: Cortical granule exocytosis in sea urchin eggs as studied by quick-freezing and freeze-fracture, J . Cell E i o l . 83, 91-108. Charbonneau, M.. and Grey, R . D. (1984). The onset of activation responsiveness during maturation coincides with the formation of the cortical endoplasmic reticulum in oocytes of X c w p i t s loeuis. Dcu. B i d . 102, 90-97. Chau. L.-Y., and Tai, H.-H. (1982). Resolution into two different forms and study of the properties of phosphatidylinositol-specific phospholipase C from human platelet cytoSOI. B i ~ ~ h iB~i ~i p. k ~ .At./ti \. 713, 344-35 I . Clapper. D. L . . and Lee. H. C. (1985). Inositol trisphosphate induces calcium release from nonmitochondrial stores in sea urchin egg homogenates. J . B i d . Chrn7. 260, I .394713954. Cockcroft. S . . and Comperts. B. D. (1985). Role ofguanine nucleotide binding protein in the activation of polyphosphoinositide phosphodiesterase. Nti/rrr.c.(London)314. 534-5361, Cohen. F.. Akabas. M . H . , and Finkelstein. A . (1982). Osmotic swelling of phospholipid vesicles causes them to fuse with a planar phospholipid bilayer membrane. S i . i r ~ i c ,217, r 458-460. Crabb. J . H.. and Jackson. R . C. (1985). fn uirro reconstitution of exocytosis from plasma membrane and isolated secretory vesicles. J . Cell Biol. 101, 2263-2273. Crahb. J . H.. and Jackson. R. C. (1986). Polycation inhibition of exocytosis from sea urchin egg cortex. J . Memhr. Biol. 91, 85-96. Creutz. C. E. (1981). c.i.s-Unsaturated fatty acids induce the fusion of chroniaffin granules aggregated by synexin. 1.C ~ l Biol. l 91, 247-257. Creutz. C. E., Paroles, C. J., and Pollard. H . H . (1978).Identification and purification of an adrenal medullary protein (synexin) that causes calcium-dependent aggregation of isolated chromaffin granules. J . Biol. C l i ~ m253. . 2858-2866. Das. S . . and Rand. R. P. (1984). Diacylglycerol causes major structural transitions in phosBiop/iy.s. Rcs. Commrtn. 124, 491-496. pholipid bilayer membranes. Bioc~l~cnr.

78

ROBERT C. JACKSON AND JOSEPH H. CRABB

Davis, B., and Lazarus, N . R. (1977). An in uitro system for studying insulin release: Effects of glucose and glucose 6-phosphate. J . Physiol. (London)271, 273-288. Decker, G . L., and Lennarz, W. J. (1979). Sperm binding and fertilization envelope formation in a cell surface complex isolated from sea urchin eggs. J . Cell B i d . 81, 92-103. Decker, S. J . , and Kinsey, W. H. (1983). Characterization of cortical secretory vesicles from the sea urchin egg. Deu. Biol. 96, 37-45. Deits, T., Farrance, M., Kay, E. S . , Medill, L . , Turner, E. E . , Weidman, P. J., and Shapiro, B. M. (1984). Purification and properties of ovoperoxidase, the enzyme responsible for hardening the fertilization membrane of the sea urchin egg. J . Biol. Chem. 259, 1352513533. Detering, N. K., Decker, G . L., Schmell, E. D., and Lennarz, W. J . (1977). Isolation and characterization of plasma membrane-associated cortical granules from sea urchin eggs. J . Cell Biol. 75, 899-914. Di Virgilio, F., Lew, D. P., and Pozzan, T. (1984). Protein kinase C activation of physiological processes in human neutrophils at vanishingly small cytosolic Ca2+levels Nature (London)310, 691-693. Douglas, W. W., and Rubin, R. P. (1961). The role of calcium in the secretory response of the adrenal medulla to acetylcholine. J . Physiol. (London) 159, 40-57. Douglas, W. W . , and Rubin, R. P. (1963). The mechanism ofcatecholamine release from the adrenal medulla and the role of calcium in stimulus-secretion coupling. J . Physiol. (London) 167, 288-3 10. Dunn, L. A., and Holz, R. W. (1983). Catecholamine secretion from digitonin-treated adrenal medullary chromaffin cells. J . Biol. Chem. 258, 4989-4993. DiizgiineS, N. (1985). Membrane fusion. Subcell. Biochem. 11, 195-289. Diizgiines, N . , Wilschut, J., Fraley, R., and Papahadjopoulos, D. (1981). Studies on the mechanism of membrane fusion: Role of head-group composition in calcium- and magnesium-induced fusion of mixed phospholipid vesicles. Biochem. Biophys. Acta 642, 182- 195. Ekerdt, R., Dahl, G., and Gratzl, M. (1981). Membrane fusion of secretory vesicles and liposomes. Two different types of fusion. Biochim. Biophys. Acra 646, 10-22. Epel, D. (1978). Mechanisms of activation of sperm and egg during fertilization of sea urchin gametes. Curr. Top. Deu. Biol. 12, 186-246. Epel, D. (1982). The physiology and chemistry of calcium during the fertilization of eggs. In “Calcium and Cell Function” (W. Y. Cheung, ed.), Vol. 2, pp. 355-383. Academic Press, Orlando, Florida. Epel, D., and Vacquier, V. D. (1978). Membrane fusion events during invertebrate fertilization. I n “Membrane Fusion” (G. Poste and G. L. Nicolson, eds.), pp. 1-63. Elsevier, Amsterdam. Epel, D., Weaver, A. M., Muchmore, A. V . , and Schimke, R. T. (1969). @-I ,3-Glucanase of sea urchin eggs: Release from particles at fertilization. Science 163, 294-296. Feinstein, H., and Schrarnm, M. (1970). Energy production in rat parotid gland. Eur. J . Biochem. 13, 158-163. Ferguson, J. E., and Shen, S. S. (1984). Evidence of phospholipase AZin the seaurchin egg: Its possible involvement in the cortical granule reaction. Gamete Res. 9, 329-338. Finkelstein, A., Zimmerberg, J., and Cohen, F. S . (1986). Osmotic swelling of vesicles: Its role in the fusion of vesicles with planar phospholipid bilayer membranes and its possible role in exocytosis. Annu. Rev. Physiol. 48, 163-174. Fodor, E. J. B . , Ako, H., and Walsh, K. A. (1975). Isolation of a protease from sea urchin eggs before and after fertilization. Biochemisrry 14, 4923-4927. Foerder, C. A., and Shapiro, B. M. (1977). Release of ovoperoxidase from sea urchin eggs

2. CORTICAL EXOCYTOSIS

79

hardens the fertilization membrane with tyrosine cross-links. Proc,. N u t / . Actid. Sci. U . S . A . 74, 4214-4218. Forman, J. C., Mongar, J . L . , and Gomperts. B. D. (1973).Calcium ionophores and movement of calcium ions following the physiological stimulus to a secretory process. Nutrue (London) 245, 249-25 I . Gilkey, J . C. (1983). Roles of calcium and pH in activation of eggs of the medaka fish, Otyzius lutipes. J . Cell Biol. 97, 669-678. Gilkey, J. C.. Jaffe, L. F.. Ridgway, E. B., and Reynolds, G. T. (1978). A free calcium wave traverses the activating egg of the medaka. Oyzius lutipes. J . Cell Eiol. 76, 448-466. Gilman, A. G. (1984). G proteins and dual control of adenylate cyclase. Cell 36, 577-579. Gomperts. B. D. (1983). Involvement of guanine nucleotide-binding protein in the gating of Ca’- by receptors. Nature (London) 306, 64-66. Gomperts. B. D. (1986). Calcium shares the limelight in stimulus-secretion coupling. TIES 11, 290-292. Gomperts, B. D., and Fernandez. J. M. (1985). Techniques for membrane permeabilization. Trends Biochem. Sc,i. 10, 414-418. Gomperts, B. D.. Barrowman. M. M . . and Cockcroft. S . (1986). Dual role for guanine nucleotides in stimulus-secretion coupling. Fed., Fed. A m . Soc. Exp. Biol. Pro(.. 45, 2 156-2 161. Grynkiewicz. G.. Poenie, M., and Tsien, R. Y. (1985). A new generation of Ca?’ indicators with greatly improved fluorescence properties. J . Biol. Chem. 260, 3440-3450. Guidice. G. (1986). “The Sea Urchin Embryo: A developmental Biological System” Springer-Verlag, Berlin. Gulyas, B. J. (1980). Cortical granules of mammalian eggs. fnt. Rev. Cvtol. 63, 357-392. Haggerty, J. G.. and Jackson, R. C. (1983). Release of granule contents from sea urchin egg cortices: New assay procedures and inhibition by sulfhydryl-modifying reagents. J . Biol. C h e m . 258, 1819-1825. Hall, H . G. (1978). Hardening of the sea urchin fertilization envelope by peroxidase-catalyzed phenolic coupling of tyrosines. Cell 15, 343-355. Hamaguchi. R., and Hiramoto, Y. (198 I ) . Activation of sea urchin eggs by microinjection of calcium buffers. Exp. Cell Res. 134, 171-179. Haslam, R. J., and Davidson, M. M. L. (1984). Guanine nucleotides decrease the free [Ca?’] required for secretion of serotonin from permeabilized blood platelets. Evidence of a role for a GTP-binding protein in platelet activation. FEBS L e f t . 174, 90-95. Hoffman, P. G., Zinder. 0.. Nikodijevic. 0.. and Pollard. H. B . (1976). ATP-stimulated transmitter release and cyclic AMP synthesis in isolated chrornaffin granules. J . Suprumol. Struct. 4, 181-184. Hollinger, T. G . , and Schuetz, A. W. ( 1976). “Cleavage” and cortical granule breakdown in Runa pipiens oocytes induced by direct microinjection of calcium. J . Cell Biol. 71. 395401.

Hollinger. T. G.. Dumont, J. N . , and Wallace, R. A. (1979). Calcium-induced dehiscence of cortical granules in Xenopus lueuis oocytes. J . Exp. Zoo/. 210, 107-1 16. Holz, R. W. (1986). The role of osmotic forces in exocytosis from adrenal chromaffin cells. Annu. Rev. Physiol. 48, 175-179. Holz. R. W., and Senter. R. A. (1986). Effects of osmolarity and ionic strength on secretion from adrenal chromaffin cells permeabilized with digitonin. J . Neurochem. 46, 18351842. Hylander. B. L., and Summers, R. G. (1981). The effect of local anesthetics and ammonia on cortical granule-plasma membrane attachment in the sea urchin egg. Deu. Biol. 86, I I I.

80

ROBERT C. JACKSON AND JOSEPH H. CRABB

Hylander. B . L., and Summers, R. G . (1982). An ultrastructural immunocytochemical localization of hyalin in the sea urchin egg. Deu. B i d . 93, 368-380. Jackson, R . C., Ward. K. K., and Haggerty, J . G. (1985). Mild proteolytic digestion restores exocytotic activity to N-ethylmaleimide-inactivatedcell surface complex from sea urchin eggs. J . Cell Biol. 101, 6-1 1 (Erratum l o / . 1167). Jamieson, J. D., and Palade. G. E. (1971). Synthesis, intracellular transport, and discharge of secretory proteins in stimulated pancreatic exocrine cells. J . Cell B i d . 50, 135-158. Joseph, S. K. (1985). Receptor-stimulated phosphoinositide metabolism: A role for GTPbinding proteins? Trends Biochem. Sci. 10, 297-298. Kaibuchi, K., Takai. Y.. Sawamura. M., Hoshijima. M., Fujikura, T.. and Nishizuka. Y. (1983). Synergistic functions of protein phosphorylation and calcium mobilization in platelet activation. J . B i d . Chem. 258, 6701-6704. Kamel, L. C., Bailey, J . , Schoenbaum, L., and Kinsey, W. (1985). Phosphatidylinositol metabolism during fertilization in the sea urchin egg. Lipids 20, 350-356. Kane, R. E. (1970). Direct isolation of the hyaline layer protein released from cortical granules of sea urchin eggs at fertilization. J . Cell B i d . 45, 615-622. Kanno, T., Cochrane, D. E . , and Douglas, W. W. (1973). Exocytosis (secretory granule extrusion) induced by injection of calcium into mast cells. C m . J . Phy.siol. Pharmuc~ol. 51, 1001-1004. Kay, E. S . , and Shapiro. B . M . (1985). The formation of the fertilization membrane of the sea urchin egg, In “Biology of Fertilization” (C. B . Metz and A. Monroy, eds.), Vol. 3 , pp. 45-80. Academic Press, Orlando, Florida. Kelly, R. B. (1985). Pathways of protein secretion in eukaryotes. Srienc,e 230, 25-32. Kinsey, W. H., Decker, G . L., and Lennarz, W. J . (1980). Isolation and partial characterization of the plasma membrane of the sea urchin egg. J . Cell Eiol. 87, 248-254. Knight, D. E., and Baker, P. F. (1982). Calcium-dependence of catecholamine release from bovine adrenal medullary cells after exposure to intense electric fields. J . Memhr. Biol. 68, 107-140. Knight, D. E., and Scrutton. M. C. (1980). Direct evidence for a role for Ca?’ in amine storage granule secretion by human platelets. Thromb. Res. 20, 437-446. Knight, D. E., Niggli, V., and Scrutton, M. C. (1984). Thrombin and activators of protein kinase C modulate secretory responses of permeabilized human platelets induced by Ca?’. Eur. J . Biochem. 143, 431-446. Knight, D. E., Tonge. D. A,, and Baker, P. F. (1985). Inhibition of exocytosis in bovine adrenal medullary cells by botulinum toxin type D. Nature (London) 317, 719-721. Konings, F.. and DePotter, W. (1982). A role for sialic acid containing substrates in the exocytosis-like in uitro interaction between adrenal medullary plasma membranes and chromaffin granules. Biochem. Eiophys. Re.s. Commun. 106, 1191-1 195. Kopf, G . S . , Moy, G . W., and Vacquier. V . D. (1982). Purification and characterization of sea urchin egg cortical granules. J . Cell B i d . 95, 924-932. Kopf, G . S . , Woolkalis, M. J., and Gerton, G . L. (1986). Evidence for a guanine nucleotidebinding regulatory protein in invertebrate and mammalian sperm. J . B i d . Chern. 261, 7327-133 I , Lelkes, P. I., Lavie, E., Naquira, D., Schneeweiss, F., Schneider, A. S . . and Rosenheck, K. (1980). Acetylcholine-induced in uitro fusion between cell membrane vesicles and chromaffin granules from the bovine adrenal medulla. FEBS Left. 115, 129-132. Le Neveu, D. M., Rand, R . P., and Parsegian, V . A. (1976). Measurement offorces between lecithin bilayers. Nurure (London) 259, 601-603. Lindau, M., and Fernandez, J . M . (1986). IgE-mediated degranulation of mast cells does not require opening of ion channels. Nature (London) 319, 150-153.

2. CORTICAL EXOCYTOSIS

81

Longo, F. J. ( 19811, Morphological features ofthe surface ofthe %a urchin (Arh(iciopioictul n r t i ) egg: Oolemma-cortical granule association. I>t,u. Biol. 84, 173-182. Lucy. J . A . (1978). Mechanisms o f chemically induced cell fusion. I n ”Membrane Fusion” (G. Poste and G . L. Nicolson. eds.). pp. 268-304. Elsevier. Amsterdam. McClay. D. R.. and Fink. R. D. (1982). Sea urchin hyalin: Appearance and function in development. Deu. B i d . 92, 285-293. M i l d . R . (1973). Transmitter release induced by injection of calcium ions into nerve terminals. /‘roc,. R . Soc. London S e r . B 183, 42 1 - 4 3 , Momayezi. M., Lumpert. C. J., Kersken. H.. Gras, U.. Plattner, H., Krinks. M. H.. and i by exogenous Klee. C. B . ( 1987). Exocytosis induction i n Ptrrcrttieiiirtti / ~ / r ( r r r r e l i (cells phosphoprotein phosphatase in uiuo and in ui/ro: Possible involvement of calcineurin in exocytotic membrane fusion. J . Cc,ll Biol. 105, 181-190. Mori. T., Takai, Y..Minakuchi, R., Yu. B . . and Nishizuka. Y. (1980). Inhibitory action o f chlorpromazine. dibucaine, and other phospholipid-interacting drugs on calcium-activated, phospholipid-dependent protein kinase. J . Biol, Chen7. 255, 8378-8380. Moy. G. W..Kopf, G. S . , Cache, C.. and Vacquier. V. D. (1983). Calcium-mediated release ofglucanase activity from cortical granules of sea urchin eggs. Deu. Biol. 100, 267-274. Muchmore. A. V.. Epel. D.. Weaver. A. M.. and Schimke. R. T. (1969). Purification and properties o f an e x o - p a - I .3-glucanase from sea urchin eggs. Biochitn. Biop1iy.s. Actcr 178. 55 1-560. Naccache. P. H., Molski. T. F. P., Alobaide. T.. Becker. E . L., Showell. H. J . , and Sha’afi. R. 1. (1980). Calmodulin inhibitors block neutrophil degranulation at a step distal from the mobilization o f calcium. Bioc,hrm. Biop/iys. Rc.\. Ccitntiiirn. 97, 62-68, Nakamula. T., and Ui, M. (1985). Simultaneous inhibitions of inositol phospholipid hreakdown, arachidonic acid release. and histamine sxretion in mast cells by islet-activating protein. pertussis toxin. J . Biol. Chetn. 260, 3584-3593. Nishizuka. Y. ( 1984a). Turnover o f inositol phospholipids and signal transduction. .Sc.ii,nc,i, 225, 1365-1370. Nishizuka. Y . (1984b). The role o f protein kinase C in cell surface signal transduction and tumour promotion. Nariir(~( L o n t l o n )308, 693-698. Oberdorf. J. A., Head. J . F.. and Kaminer, B. (1986). Calcium uptake and release by isolated cortices and microsomes from the unfertilized egg o f the sea urchin Sironglyo i ~ r n / r o / i t .droebcrc~hirt7sis. s J . Cell B i d . 102, 2205-2210. Ohashi. Y..and Narumiya. S. (1987). ADP-ribo\ylation o f a M, 21.000 membrane protein by type D botulinum toxin. J . B i d . C/7e,rr. 262, 1430-1433. Oinuma, M., Katada, T.. Yokosawa. H.. and Ui. M. (1986). Guanine nucleotide-binding protein in sea urchin eggs serving as the specific substrate o f islet-activating protein, pertussis toxin. FEBS L e / / . 207, 28-34. Okazaki. R . (1956). On the possible role o f high energy phosphate in the cortical change o f sea urchin eggs. Exp. Cell R r s . 10, 476-504. Otto, J . J . . Kane. R. E.. and Bryan, J. (1980). Redistribution ofactin and fascin i n sea urchin eggs after fertilization. CeII Moril. 1, 31-40. Palade. G . (1975). lntracellular aspects o f the process o f protein synthesis. .Scicv7c,i, 189, 347-358. Papahadjopoulos, D. ( 1978). Calcium-induced phase changes and fusion in natural and model membranes. I n “Membrane Fusion” ( G . Poste and Ci. I-. Nicholson. eds.), pp. 765-790. Elsevier. Amsterdam. Parsegian. V. A . (1977). Considerations in determining the mode of influence o f calcium on vesicle-membrane interaction. Soc. Nrirrosc.i. Sytnp.; Approciclws Cell Biol. Neiiroti,s 2, 161-171.

82

ROBERT C. JACKSON AND JOSEPH H. CRABB

Pazoles, C. J., and Pollard, H. B. (1978). Evidence for stimulation of anion transport in ATP-evoked transmitter release from isolated secretory vesicles. J . B i d . Chem. 253, 3962-3969. Penner, R., Neher. E.. and Dreyer, F. (1986). Intracellularly injected tetanus toxin inhibits exocytosis in bovine adrenal chromaffin cells. Nafure (London) 324, 76-78. Perrin, D., Langley, 0. K., and Aunis, D. (1987). Antifodrin inhibits secretion from permeabilized chromaffin cells. Nature (London) 326,498-501. Peterson, C. (1974). Role of energy metabolism in histamine release: Study on isolated mast cells. Acta Physiol. Scand. 92 (Suppl. 413), 1-34. Poenie, M., Alderton, J., Tsien, R. Y., and Steinhardt, R. A. (1985). Changes offree calcium levels with stages of the cell division cycle. Nature (London) 315, 147-149. Pollard, H. B . , Pazoles, C. J., Creutz, C. E., and Zinder, 0. (1979). The chromaffin granule and possible mechanisms of exocytosis. f n t . Reu. Cvtol. 58, 159-198. Pollard, T. D. (1982). Purification of nonmuscle myosins. I n “Methods in Enzymology” (D. W. Frederiksen and L. W. Cunningham, eds.) Vol. 85, pp. 331-356. Academic Press, New York. Pollard, T. D., Thomas, S. M., and Niederman, R. (1974). Human platelet myosin. I . Purification by a rapid method applicable to other nonmuscle cells. A n d . Biochem. 60, 258-266. Portis, A . , Newton, C., Pangborn, W., and Papahadjopoulos, D. (1979). Studies on the mechanism of membrane fusion: Evidence for an intermembrane calcium complex, synergism with magnesium and inhibition with spectrin. Biochemistry 19, 780790. Prentki, M., Biden, T. J., Janjic, D., Irvine, R. F., Berridge, M. J., and Wollheim, C. B. (1984). Rapid mobilization of Ca2+from rat insulinoma microsomes by inositol 1.4.5trisphosphate. Nature (London) 309, 562-564. Rand, R. P. (1981). Interacting phospholipid bilayers: Measured forces and induced structural changes. Annu. Rev. Biophys. Bioeng. 10, 277-314. Rasmussen, H., and Goodman, D. B. P. (1977). Relationships between calcium and cyclic nucleotides in cell activation. Physiol. Rev. 57, 421-509. Rink, T. J . , Sanchez, A,, and Hallam, T. J . (1983). Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. Nuture (London)305, 317-319. Ronning, S . A., and Martin, T. F. J. (1986). Characterization ofCa?+-stimulatedsecretion in permeable GH3 pituitary cells. J . Biol. Chem. 261, 7834-7839. Rubin, R. P. (1982). “Calcium and Cellular Secretion” Plenum, New York. Ruggiero, M., Zimmerman, T. P., and Lapetina, E. G. (1985). ATP depletion in human platelets caused by permeabilization with saponin does not prevent serotonin secretion induced by collagen. Biochem. Biophys. Res. Commun. 131,620-627. Salpeter, M. M., and Farquar, M. G . (1981). High resolution analysis of the secretory pathway in mammotrophs of the rat anterior pituitary. J. Cell B i d . 91, 240-246. Sardet, C. (1984). The ultrastructure of the sea urchin egg cortex isolated before and after fertilization. J . CeI/ Bio/. 105, 196-210. Sasaki, H. (1984). Modulation of calcium sensitivity by a specific cortical protein during sea urchin egg cortical vesicle exocytosis. Dev. Biol. 101, 125-135. Sasaki, H., and Epel, D. (1983). Cortical vesicle exocytosis in isolated cortices of sea urchin eggs: Description of a turbidimetric assay and its utilization in studying effects of different media on discharge. Dev. Biol. 98, 327-337. Sawada, H., Miura, M., Yokosawa, H., and Ishi, S. (1984). Purification and characterization of trypsin-like enzyme from sea urchin eggs: Substrate specificity and physiological role. Biochem. Biophys. Res. Commun. 121, 598-604.

2. CORTICAL EXOCYTOSIS

83

Schiin. E. A,. and Decker, G. L. (1981). Ion-dependent stages of the cortical reaction in surtiice complexes isolated from Arhtic.ici p~rnc~fiilrifcreggs. J . Ulfrri.vfrirc.l.Res. 76, 191201. Schroeder, T. E. (1979). Surface area change at fertilization: Resorption of the mosaic membrane, l k u . Biol. 70, 306-326. Schuel. H. (1978). Secretory functions of egg cortical granules in fertilization and development: A critical review. Grimere Res. 1, 299-382. Schuel. H . (1985). Functions of egg cortical granules. I n “Biology of Fertilization” ( C . B . Metz and A. Monroy, eds.), Vol. 3 , pp. 1-44. Academic Press. Orlando. Florida. Schuel. H.. Wilson, W. L.. Bressler. R. S . . Kelly. J . W.. and Wilson, J . R. (1972). Purification of cortical granules from unfertilized sea urchin egg homogenates by zonal centrifugation. Dru. B i d . 29, 307-320. Schuel, H.. Kelley, J . W., Berger. E. R.. and Wilson, W. L. (1974). Sulfated acid mucopoly\accharides in the cortical granules of eggs: Effects of quaternary ammonium salts on fertilization. Exp. Cell Rrs. 88, 24-30. Seenian. P. ( 1972). The membrane actions of anesthetics and tranquilizers. Phurmac.ol. Reu. 24, 583-656. Sellinger-Barnette, M., and Weiss. B. ( 1984). Interaction of various peptides and calmodulin. Adu. C.vclic Niic~lrotideRes. 16, 261-276. Shapiro. B. M.. Schackmann, R. W., and Gabel, C. A. (1981). Molecular approaches to the study of fertilization. Annii. R r u . Bioclirrn. 50, 815-843. Shier, W. T. (1977). Inhibition of acyl coenzyme A:lysolecithin acyltransferases by local anesthetics. detergents and inhibitors of cyclic nucleotide phosphodiesterases. Bioi,/ir,tn. Biopliys. Res. Comtniin. 75, 186-197. Skehel, J . J., Bayley. P. M., Brown, E. B . . Martin. S . R.. Waterfield, M. D., White, J . M., Wilson. 1. A.. and Wiley, D. C. (1982). Changes in the conformation of influenza virus hemagglutinin at the pH optimum of virus-mediated membrane fusion. Proc. Nrirl. Aced. S1.i. U . S . A . 79, 968-972. Smolen. J . E.. Stoehr. S. J . . and Boxer, L. A . (1986). Human neutrophils permeabilized with digitonin respond with lysosomal enzyme release when exposed to micromolar levels of free calcium. Biodiini. Biophys. Acfcr 886, 1-17. Steinhardt, R. A,. and Alderton. J . M. (1982) Calmodulin confers calcium sensitivity on secretory exocytosis. Niirrtre (Londoti) 295, 154- 155. Steinhardt. K. A.. and Epel. D. (1974) Activation of sea urchin eggs by a calcium ionophore. Pro(,. Nufl. A<.ud. Sci. U . S . A . 71, 1915-1919. Steinhardt, R., Zucker. R., and Schatten, G . (1977).lntracellular calcium release at fertiliration in the sea urchin egg. Dev. Hiol. 58, 185-196. Stephens. R. E.. and Kane. R. E. (1970). Some properties of hyalin, the calcium insoluble protein of the hyaline layer of sea urchin eggs. J . Cell Biol. 44,611-617. Streb, H., Irvine. R. F . , Berridge. M . J., and Schulz, 1. (1983). Release of Ca?’ from a nonmitochondrial intracellular store in pancreatic acinar cells by inositol 1,4.5-trisphosphate. Nririrrr (London) 306, 67-69. Takai. Y . . Kishimoto. A . . Kawahara. Y . , Minakuchi. R . , Sano, K . . Kikkawa, U., Mori. T,. Y u . B.. Kaibuchi. K., and Nishizuka. Y . (1981). Calcium and phosphatidylinositol turnover as signalling for transmembrane control of protein phosphorylation. Adu. CJIclic Nuclrofide Res. 14, 301-3 13. Talbot. C. F.. and Vacquier, V. D. (1982). The purification and characterization of an exo@)-( I .3)-glucanohydrolase from sea urchin eggs. J . H i d . Cllrni. 257, 742-746. Tartakoff. A. M. (1985). Membrane dynamics and specificity. Trends Biochem. Sci. 10,413. Tartakoff. A. M.. and Vassalli. P. (1977). Plasma cell immunoglubulin secretion. J . Exp. Mrd. 146, 1332-1345.

04

ROBERT C. JACKSON AND JOSEPH H. CRABB

Tartakoff, A. M., and Vassalli, P. (1978). Comparative studies of intracellular transport of secretory proteins. J . Cell Biol. 79,694-707. Trimmer, J. S ., and Vacquier, V. D. (1986). Activation of sea urchin gametes. Annu. Reu. Cell B i d . 2, 1-26. Tsien, R. Y . (1980). New calcium indicators and buffers with high selectivity against magnesium and protons: Design, synthesis and properties of prototype structures. Biochemist~ 19, 2396-2404. Tsien, R. Y. (1983). lntracellular measurements of ion activities. Annu. Reu. Biophy.y. Bioeng. 12, 91-1 16. Tsien, R. Y . , Pozzan, T., and Rink, T. J. (1984). Measuring and manipulating cytosolic Ca?’ with trapped indicators. Trends Biochem. Sci. 9, 263-266. Turner, P. R., Sheetz, M. P., and Jaffe, L. A. (1984). Fertilization increases the polyphosphoinositide content of sea urchin eggs. Nafure (London) 310,414-415. Turner, P. R., Jaffe, L. A., and Fein, A. (1986). Regulation of cortical vesicle exocytosis in sea urchin eggs by inositol I ,4,5-trisphosphate and GTP-binding protein. J . Cell B i d . 102, 70-76. Turner, P. R., Jaffe, L. A . , and Primakoff, P. (1987). A cholera toxin sensitive G-protein stimulates exocytosis in sea urchin eggs. Deu. B i d . 120, 577-583. Vacquier. V. D. (1975). The isolation of intact cortical granules from sea urchin eggs. Calcium ions trigger granule discharge. Dev. B i d . 43, 62-74. Vacquier, V. D. (1976). Isolated cortical granules: A model system for studying membrane fusion and calcium mediated exocytosis. J . Superrnol. Struct. 5 , 27-35. Vallar, L., Biden, T. J . , and Wollheim, C. B. (1987). Guanine nucleotides induce Ca”independent insulin secretion from permeabilized RINm5F cells. J . Biol. Chem. 262, 5049-5056. Van Belle, H. (1981). R24571-A potent inhibitor of calmodulin-activated enzymes. Cell Culcium 2, 483-494. Van Den Bosch, H. (1980). lntracellular phospholipases A. Binchim. Biophys. Actu 604, 19 I -246. Vilmart-Seuwen, J., Kersken, H.. Sturzl, R., and Plattner, H. (1986). ATP keeps exocytosis sites in a primed state but is not required for membrane fusion: An analysis with Purumecium cells in uivo and in uitro. J . Cell Biol. 103, 1279-1288. Wallace, M. A., and Fain, J. N. (1985). Guanosine 5’-O-thiotriphosphate stimulates phospholipase C activity in plasma membranes at rat hepatocytes. J . Biol. Chem. 260,95279530. Weidman, P. J., Kay, E. S . , and Shapiro, B. S . (1984). Assembly of the sea urchin fertilization membrane: Isolation of proteoliaisin, a calcium-dependent ovoperoxidase binding protein. J . Biol. C h e m . 100, 938-946. Whitaker, M., and Aitchison, M. (1985). Calcium-dependent polyphosphoinositide hydrolysis is associated with exocytosis in virro. FEBS Lett. 182, 119-124. Whitaker, M. J., and Baker, P. F. (1983). Calcium-dependent exocytosis in an in uitro secretory granule plasma membrane preparation from sea urchin eggs and the effects of some inhibitors of cytoskeletal function. Proc. R . Soc. London Ser. B 218, 397-413. Whitaker, M . , and Irvine, R. F. (1984). Inositol 1,4.5-trisphosphate microinjection activates sea urchin eggs. Nature (London) 312, 636-639. White, J., Kielian. M., and Helenius. A. (1983). Membrane fusion proteins of enveloped animal viruses. Q. R e v . Biophys. 16, 151-195. White, J. R., Ishizaka, T., Ishizaka, K., and Sha’afi, R. 1. (1984). Direct domonstration of increased intracellular concentration of free calcium as measured by quin-2 in stimulated rat peitoneal mast cell. Proc. Null. A c a d . Sci. U . S . A . 81, 3978-3982.

2. CORTICAL EXOCYTOSIS

a5

Wilschut. I . . and Hoekstra. D. (1984). Membrane fusion: From lipowme\ to biologiciil membranes. 7k,rids Bioc./ic,tn. S c i . 9. 479-4x3. Wilson. S. P.. and Kirshner. N . ( 1983). Calcium-evoked wcretion from digitonin-permwhiIiLed adrenal medullary chromaffin cells. ./. Biol. Clrcm. 258. 4994-5000. Zinimerberg, J . , and Whitaker. M. ( 19851. Irreversible \welling of secretory granule9 during exocytosis caused by calcium. Ntirrrrr ( L t i d w )315. 581-584. Zimmerberg. J.. Cohen, F. S . , and Finkelstein. A . ( I Y X O ) . Micromolar Ca?- stimulates fusion of lipid vesicles with planar hilayers containing a calcium-binding protein. .St.;etic'c 210, 906-908. Zinimerberg. J . , Sardet. C . . and Epel, D. (1985). l-:xocytosi\ of sea urchin cortical vesicle5 irr riirro is retarded by hyperosmotic sucrose: Kinetics of fusion monitored by quantitative light scattering microscopy. J . Cell R i d . 101, 2398-2410. Zimmerberg, J . Curran, M.. Cohen. F. S . , and Brodwick. M. (19871. Simultaneou< electrical and optical measurements show that membrane fusion precede9 secretory granule swelling during exocytosis of beige mouse mast cells. PIW. N o / / . Ac,trd. .Sc,i. U . S . A .84, 1585- 1589. Zucker, R . S . , and Steinhardt, R . A . (1978). Prevention of the cortical reaction in fertilized sea urchin eggs by injection of calcium-chelating ligands. Biot./iirn. Biopliy.~.A(.to 541, 459-466.

This Page Intentionally Left Blank

CURRENT TOPICS IN MEMBRANES AND TRANSPORr. VOLUME 32

Chapter 3 Mechan stic

Myoblast Fus on-A Anal ys is

I. II.

Ill. IV.

V.

VI.

VII. VIII.

Introduction Morphological Aspects of Myoblasl Fusion A. Systems of Study B. Lineage and Cell Specificity Dependence of Myoblast Fusion C. Stages in the Fusion of Myoblasts Kinetics of Myoblsst Fusion A. Role of Metal Ions B. Time Dependence of Fusion Structure-Function Relationships in Myoblast Plasma Membranes A. Proteins of Myoblast Plasma Membranes B. Chrlnges in Proteins C . Antibody Studies D. Lipids of Myoblast Plasma Membranes Fusion-Induced Changes in Membrane Organization A. Ultrastructural Studies B. Fluidity Studies Proposed Mechanisms of Myoblast Fusion A. Changes in Bilayer Structure B. Membrane Forces and Fusion C. Proteins and Fusion D. A Consensus Model for Myoblast Fusion'? Stimulation of Myoblast Fusion Conclusions References

87 Copyrighl c., IYXX hv Academic Pre\\. Inc All rights of reproduction in a n ) form reberved

88

MICHAEL J. 0.WAKELAM

1.

INTRODUCTION

Myoblast fusion is a key event in the differentiation of skeletal muscle. During embryonic development mononucleate myoblasts align and fuse to form multinucleate myotubes which are the biological precursors of skeletal muscle fibers. These fibers are permanent, nonmitotic cells which may contain over a hundred nuclei. The fusion of myoblasts is of considerable interest not only in being a key point in a differentiation pathway, but also because it is an event easily studied and manipulated in tissue culture. It thus provides the potential for a physiologically relevant model of cell and therefore membrane fusion. The transition from mono- to predominantly multinucleated muscle cells can be observed in frozen thin sections of embryonic muscle, but analysis of mechanistic events is impossible in that system. The development by Holtzer et af. (1958) of a primary culture system for the analysis of myoblast fusion was a major advance in the study of muscle differentiation. Fusion studies have since utilized this powerful in uitro model. Myoblast fusion can be studied using either primary cultures, which can be prepared from a range of avian, insect, and mammalian species (see Konigsberg, 1978), or established cell lines. Myoblast fusion has been extensively reviewed previously (Bishoff, 1978; Wakelam, 1985). These reviews cover much background to which readers are referred. This chapter will describe the process of muscle cell fusion, discuss the model systems, but will mainly be concerned with a discussion of membrane union and its regulation.

II. MORPHOLOGICAL ASPECTS OF MYOBLAST FUSION A. Systems of Study

Primary cultures of muscle cells are prepared from embryonic tissues by a combination of mechanical and mild enzymatic disaggregation methods in the absence of divalent cations. The cell suspension is filtered to yield single cells which are plated in an appropriate culture medium that contains serum and embryo extract. The majority of the cells are proliferative myogenic precursor cells, presumptive myoblasts, though some fibroblasts are present. The numbers of fibroblasts can be significantly reduced by preplating techniques. It has been elegantly demonstrated by Puri and Turner (1978) that myoblasts cultured in serum- and embryo extract-free media do not attach to uncoated tissue culture dishes; fibroblasts, in contrast, rapidly attach and flatten out on the dish. If the cell

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

89

suspension is then decanted and plated in the presence of serum onto collagen- or gelatin-coated dishes, an almost pure culture of myoblasts is obtained (Puri and Turner, 1978). Chiquet rr (11. (1979) demonstrated that the attachment-promoting factor present in serum is the glycoprotein fibronectin. The cultured myoblasts undergo one or more rounds of cell division, withdraw from the cell cycle, and then fuse. The fusion process is preceded by alignment, a process that occurs both during and after the proliferative stage. The fusion of avian embryonic myoblasts is depicted in Fig. 1 . Established myogenic cell lines of both rat and mouse origin are available. The most commonly used lines are L6 and L8 and their subclones which were isolated from rat muscle tissue (see Yaffe, 1968; Pearson, 1980). Differentiation in these cells is induced by mitogen depletion; there are, however, great variations in these cells with respect to fusion kinetics, growth factor responsiveness, and indeed morphology, both between the different lines and as compared to primary cultures. Accordingly, information derived from studies on cell lines must be treated with great caution, since it is unlikely to be directly relevant to the physiological situation. Important modifications have been made to the standard primary culture system which have greatly facilitated the direction of the fusion process. The first was related to the finding by Shainberg et al. (1969) that myoblast fusion is Ca'+ dependent. When chick embryonic myoblasts are cultured at a Ca2+concentration of 20 pM (Van der Bosch et al., 1972) or 0.1 pM (Wakelam and Pette. 1982) the cells divide and align, but fail to fuse. Addition of 1 mM Ca2+to such cultures after 50 hr results in rapid, synchronous fusion (see Fig. 2). The second modification, developed by Horwitz's group, is that of myoball culture. In this system the cells are grown in suspension and, following aggregation, they fuse to form myoballs (Knudson and Horwitz, 1977; Neff and Horwitz, 1982). The advantage of this system is that the medium can be easily modified. By assaying changes in cell size a rapid time course fusion assay has been developed (Neff and Horwitz, 1982). 6. Lineage and Cell Specificity Dependence of Myoblast Fusion

Myogenic cells are derived from the embryonic mesoderm. This is also the source of fibroblasts and chondroblasts, and thus raises questions concerning precursor cell heterogeneity. However, Sasse rt ~ i l .( 1984) have shown that early chick limb buds consist of at least two independent subpopulations of committed precursor cells. These workers demon-

90

MICHAEL J. 0. WAKELAM

FIG. I . Muscle cell fusion. (A) Myoblasts 17 hr in culture. Note that cell division is occurring but alignment is under way. (B) After 24 hr, alignment is extensive, and the typical After 70 hr, spindle shape of myoblasts is apparent. (C) After 43 hr, fusion is under way. (D) fusion is complete (see Fig. 2). At this stage cultures are dominated by large, multinucleate cells. Extensive branching, seen in uirm, is not found in uiuo. x 154. (Reproduced from Wakelam, 1985, with permission.)

strated that an antibody, originally isolated for its ability to bind to and detach myogenic cells from extracellular matrices (Neff ef al., 1982), distinguishes between cells of myogenic and chondrogenic lineages. Myoblasts only fuse with cells of their own lineage. Fibroblasts, kidney

3. MYOBLAST FUSION-A

91

MECHANISTIC ANALYSIS

and liver cells, chondrocytes, and smooth and cardiac muscle cells do not fuse either with myogenic cells or myotubes (see Holtzer and Bishoff, 1970). Myoblast fusion is thus cell specific and suggests a unique surface characteristic of the fusing cells (see below). C. Stages in the Fusion of Myoblasts

Myoblast fusion occurs following the alignment of cells. The timing of fusion is dependent on both the density of cells in the culture and on the composition of the medium (Linkhart rt id., 1981). I n general, the fusion of primary embryonic chick myoblasts begins after 40-42 hr in culture and proceeds for the next 18 hr (Fig. 2). As mentioned in Section II,A, removal of Ca?+from the medium can generate a culture of more rapidly fusing cells when Ca?+ is added back to the medium after 50 hr (Fig. 2). Differing models defining the sequence of events in myoblast fusion have been proposed. The simplest of these is that, in the presence of Ca" , fusion occurs directly between apposed regions of cell membrane. This somewhat simplistic proposal is in contrast to the model proposed by Horwitz and co-workers which envisions a sequence of events involving separate recognition, adhesion, and fusion processes. Recognition is probably accomplished during the alignment process. As mentioned in Section II,B, myoblast fusion is cell specific, but since heterotypic fusion between rabbit and rat myoblasts has been observed (Bishoff, 1978) there must be common recognition sites between mamma-

I

.

I

0.

I

24

34 40 50 TIME IN CULTURE (HOURS)

60

FIG. 2 . Time course of chick myoblast fusion. Fusion is scored as the percentage of nuclei in cells that contain three o r more. U, Culture medium of normal composition; A, medium CaL+concentration 0.1 p M ; 0, following culture for 50 hr at a medium Ca?' concentration of 0. I p M , cells were cultured in normal medium, i.e..Ca'- concentration I .4 mM. (Reproduced from Wakelam. 1985, with permission.)

92

MICHAEL J. 0. WAKELAM

lian species. Once recognition is complete the cells aggregate and adhere. The concept of specificity of recognition and aggregation is further strengthened by the observation that when myoblasts are aggregated in suspension cultures, fibroblasts are excluded (Knudson and Horwitz, 1978). This specificity may be due to an enhanced affinity of myogenic cells for laminin (Kiihl rt al., 1986). Irreversible adhesion, which is trypsin sensitive, appears to be obligatory for fusion (Knudson and Horwitz, 1978). It is unclear, however, whether this event is part of, or distinct from, the fusion process per se. Rash and Fambrough (1972) have observed structures similar to gap junctions between aggregated, unfused, cells. Their electrophysiological examination of these cells demonstrated that the cells were electrically coupled, but not fused. A freeze-fracture study of 19-day-old rat fetal muscle (Rash and Staehelin, 1974) provided support for the existence of such coupling. Small, irregular arrays of particles 8-9 nm in size were observed on the cytoplasmic portions of complementary leaflets of myogenic cells. The possibility exists, therefore, that intercellular communication precedes membrane fusion and constitutes the process of irreversible adhesion.

111.

KINETICS OF MYOBLAST FUSION

A. Role of Metal Ions

In common with other examples of membrane fusion events, myoblast fusion is dependent on Ca2+ions. This was first demonstrated by Shainberg et a / . ( 1969), who decreased the medium Ca2+concentration from 1.4 to 0.27 mM and observed an inhibition of fusion, without affecting myoblast proliferation. On restoration of the normal medium Ca2+concentration, rapid fusion was observed. The reduction of medium Ca2+as a technique in the study of myoblast fusion has been subject to criticism. A generalized inhibition of muscle cell differentiation has been observed as a result of Ca2+depletion from the medium (Easton and Reich, 1972; Patterson and Strohman, 1972). This is probably dependent on the extent of Ca2+depletion; indeed, myoblasts divide, align, and achieve fusion competence when cultured at Ca2+concentrations as low as 0.1 pM (Wakelam and Pette, 1982). The effects of Ca2+depletion on myoblast fusion is similar between species. Calf, chick, and rat myoblasts all show approximately 50% fusion at 500 pM Ca2+(Adamo et al., 1976; Schudt el u l . , 1973; Merlie and Gros, 1976; Shainberg et ul., 1969). When myoblasts have been cultured for 50 hr at a Ca2+concentration of 0.1 pM, rapid fusion can be initiated by the addition of 1.4 mM Ca2+(Fig.

3. MYOBLAST FUSION-A

93

MECHANISTIC ANALYSIS

2; Wakelam, 1985). This technique has allowed the accurate determination of the Ca?+ concentration dependence of fusion-1.4 m M (Van der Bosch et a / . , 1972; Schudt and Pette, 1975). This CaZt dependence is depicted in Fig. 3. Other cations-Ba’+, Cuzt, Cd’+, La3+,Li+, Mg?+, Mn?+,and Zn2+-cannot substitute for Ca?“ and are indeed inhibitory at higher concentrations. The exception is Sr7+,which, in addition to being noninhibitory, can substitute for Ca?”(Schudt et al., 1973; Adamo i’t al., 1976). The role of CaZ+in fusion will be discussed in detail below. 100

T T T T 80

60

.-E VI

c

8 40

20

0 0001

001

.01

05

1

25

.5

75

1

25

5

75

10

Calclum concentrailon (mM)

FIG.3. Dependence of myoblast fusion on Ca”. Fusion was scored as in Fig. 2 following 60 hr in culture at the stated Ca2+concentrations. Results are means ? SEM, where n = 12.

94

MICHAEL J. 0.WAKELAM

B. Time Dependence of Fusion

Fusion of myoblasts is not instantaneous even after fusion competence has been achieved. By making use of two complementary fusion assays, Neff et al. (1984) have studied the kinetics of fusion in detail. The first of these is the suspension assay of Knudson and Horwitz (1977) which defines fused myoblasts as those aggregates of myoblasts which cannot be dispersed by treatment with either EDTA or trypsin. The second assay, designed to determine membrane continuity as well as irreversible adhesion, involved the transfer of a fluorescent lipid from labeled to unlabeled cells on mixing and fusion. Neff et al. (1984) demonstrated that, when Ca2+ was added to a suspension of myoblasts from Ca-depleted media, significant membrane continuity could be seen within 20-30 min. The multinucleated morphology typical of a myotube was not observed, however, for a further l hr. This 20 to 30-min period required for membrane continuity to be observed, even in this sensitive assay, implicates a requirement for biochemical changes in membrane structure before membrane union can occur. This is in contrast to the simple model systems of membrane fusion where membrane union is almost instantaneous.

IV.

STRUCTURE-FUNCTION RELATIONSHIPS IN MYOBLAST PLASMA MEMBRANES

The specificity and kinetics of myoblast plasma membrane fusion make it clear that this cell's surface is not only unique, but must undergo biochemical changes before membrane union can occur. This section reviews our knowledge of the changes in the myoblast plasma membrane that relate to fusion. A. Proteins of Myoblast Plasma Membranes

It is probable, but as yet unproved, that glycoproteins are involved in the recognition reactions involved in myoblast fusion; there is no doubt, however, that these proteins are involved in the adhesion process. Following recognition, myoblasts adhere in a trypsin-sensitive manner (Knudson and Horwitz, 1977). There is controversy as to whether proteins play additional roles in the sequence of events leading to myoblast fusion. The primary observation suggesting involvement of surface proteins, in particular glycoproteins, in myoblast fusion was that concanavalin A inhibits fusion in a reversible manner (Den et al., 1975; Sandra et al., 1977;

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

95

Burstein and Shainberg, 1979). This plant lectin binds to surface glycoproteins. Since it is only effective in its tetrameric form, it probably acts by cross-linking glycoproteins. On the other hand, prolonged exposure of muscle cells to concanavalin A leads to cytotoxicity. To resolve this complication, Parfett e f al. (1981) and Cates ef ul. (1984) have isolated L6 myoblast cell lines that are resistant to concanavalin A cytotoxicity by virtue of defects in the synthesis of mannosylated glycoproteins. Since these cells do not fuse, the mannosylated glycoproteins have been assigned a critical role in myoblast fusion (Parfett et ul., 1981; Cates et al., 1984). However, in these studies the authors did not distinguish between inhibition of adhesion and of membrane fusion. In a further attempt to elucidate the role of glycoproteins. Cates c t ul. (1984) isolated two separate classes of concanavalin A-resistant, nonfusing L6 mutants. Although these investigators were unable to determine the defect in one class of cells, they observed a selective loss in the binding of '?51-labeledconcanavalin in A to a protein of M , 46.000 in the second group. This failure to bind was probably due to the absence of the 46-kDa protein, a conclusion strengthened by the observation that somatic hybrids of the two mutant classes regained the ability to fuse. It will be of interest to see if the defect in the second class of mutants is also at the level of glycoprotein synthesis. The concept that glycoproteins are involved in fusion derives support also from the fact that tunicamycin, an inhibitor of glycosylation, inhibits myoblast fusion (Gilfix and Sanwal, 1980). Olden et al. (1981) demonstrated that in quail myoblasts this inhibition could be partially overcome by protease inhibitors. This led them to suggest that glycoproteins may mediate myoblast fusion because their carbohydrate groups stabilize the protein. Cates et ul. (1984) found that proteases were unable to prevent the inhibition of fusion by tunicamycin in L6 cells, but that N-acetylglucosamine could accomplish the reversal. They also demonstrated that, following tunicamycin treatment, the plasma membrane concentration of four glycoproteins ( M , 230,000, 145,000, 119,000, and 46,000) was reduced. Since the 119-kDa protein was the major binding site for Ir51labeled concanavalin A, it is probably a surface protein. Another way to study the role proteins play in myoblast fusion is to follow changes in the levels of particular plasma membrane proteins. For example, the synthesis of myonectin and electronectin increases in L6 cells during the period preceding fusion. Then, as the cells reach confluency and just before fusion takes place, synthesis of these two proteins falls (Gartner and Podleski. 1975, 1976; Podleski et al., 1979; Podleski and Greenberg, 1980). The p-galactoside-binding protein electronectin can exist in both particulate and soluble forms, the activity of the latter being

96

MICHAEL J. 0. WAKELAM

regulated by myonectin. That this may be important is suggested by the finding that an electronectin-like p-galactoside-binding protein, isolated from chick embryonic muscle, can inhibit myoblast fusion (Nowak rt af., 1977; Den and Malinzak, 1977; MacBride and Przybylski, 1980). B. Changes in Proteins

The experiments discussed in the preceding section do not provide information on changes in plasma membrane proteins that occur when fusion is initiated. Cates and Holland (1978) have reported an increase in both the synthesis and accumulation of a 70-kDa protein when fusion begins. Yoshioka and Suroka (1983) found that the relative amount of an M , 200,000-250,000 surface protein increased during L6 myoblast fusion. However, when M3A cells, a nonfusing variant, were examined, this change was not observed, and a 90-kDa protein was detected that was unique to the variant. When the function and properties of these two proteins become better known, the importance of these observations for an understanding of the myoblast cell membrane may become apparent. In an attempt to determine biochemical changes in proteins, Senechal et al. (1982) examined L6 cells for developmentally regulated changes in phosphorylation. Four proteins ( M , 165,000, 105,000, 60,000, and 45,000) were found to be so regulated. Unfortunately this observation alone does not answer the question whether phosphorylation is a consequence of or essential for the fusion process. By making use of changes in the iodination patterns of surface proteins, Moss et al. (1978) and Pauw and David (1979) demonstrated higher levels of low molecular weight and lower levels of high molecular weight proteins. Despite the lack of evidence linking the two types of proteins in a product/precursor relationship, Couch and Strittmatter (1983) have proposed that this is due to limited proteolysis during the fusion process. This important attempt to link structural changes to fusion mechanisms will be discussed further in Section VI,C. C. Antibody Studies

The development of monoclonal antibody technology has led to the use of these reagents in the examination of surface changes during fusion. Monoclonal antibodies have been generated by using as immunogens myoblasts or muscle membranes in various states of differentiation. Further, antibodies raised against other cell types have been found to crossreact with myoblasts (e.g., Wakeshull et al., 1983). Immunofluorescence analysis has demonstrated that quantitatively and topographically distinct

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

97

changes in some cell surface antigens accompany fusion. Such studies have utilized both human and chick myoblasts and cloned cell lines. Walsh et a / . (1984) described an antigen that is unique to prefusion myoblasts in human primary muscle cultures, though its identity is unknown. Grove et ctl. (1981) generated antibodies against two types of embryonic chick muscle cells-midfusion myoblasts and myotubes. The antibodies were then used in an indirect analysis by radioimmunoassay of myoblasts, midfusion myoblasts, and myotubes. The data indicated the existence of determinants unique to each stage of myogenesis and that these determinants undergo significant quantitative changes within a cell stage. The antibodies raised by Kaufmann and Foster (1984) were against the E63 variant of the L8 cell line. They defined at least five discrete temporal classes, including aligned cells and those undergoing fusion. Of particular interest is the observation that there occur transient alterations in surface accessibility to some antigens during fusion. The identity of these antigens will undoubtedly be of great ultimate functional interest. D. Lipids of Myoblast Plasma Membranes

Even though lipids play a fundamental role in the process of membrane union, studies of myoblast lipids are not numerous. Analysis of the content of phospholipids, cholesterol, and fatty acids in muscle cell membranes did not demonstrate any correlative changes in the myoblast to myotube transition (Kent et ul., 1974; Boland et ol., 1977). However, these studies failed to examine the minor phospholipids such as polyphosphoinositides and gangliosides. Whatley et a / . (198I ) demonstrated that the latter indeed undergo changes during fusion. For example, when L6 cells fuse, the GDlA gangliosides increase 3- to 4-fold, but then return to prefusion levels following myotube formation. No such changes were found in a nonfusing L6 variant. While the GDlA ganglioside has been shown to be a fusogenic lipid in other systems (Maggio o f al., 1981). this does not seem true for primary cells. Hence these gangliosides may be involved in prefusion activities. Possibly they, together with glycoproteins, might contribute to the adhesion process. Gross studies of lipid composition cannot provide detailed information concerning the lipid structure of myoblast plasma membranes. Sessions and Horwitz (1981, 1983) studied the asymmetry of phospholipids across the bilayer. They focused on the distribution of phosphatidylethanolamine and phosphatidylserine by utilizing two nonpenetrating amidating reagents. They found that in chick and quail myoblasts 65% of the phosphatidylethanolamine and 45% of the phosphatidylserine were present in the outer leaflet. By comparison, the respective values are 35 and 20%

98

MICHAEL J. 0.WAKELAM

in fibroblasts and 22 and 30% in L6 myoblasts. Although these two aminophospholipids have been shown to be fusogenic in model membrane studies (DuzguneS, 1985), they do not of themselves confer on myoblasts the ability to fuse since these phospholipids are also present in non-fusion competent myoblasts. They do, however, demonstrate the unique nature of the myoblast plasma membrane and may contribute to the fusion process when it is stimulated by other biochemical events. If certain lipids are critical for myoblast fusion, then experimental modification of lipid metabolism should block or promote fusion. Two experimental approaches have been adopted: the first involves manipulation of the culture media and the second the addition of enzymes. Since fusion requires a fluid membrane (see Section VI,B), manipulations of cholesterol content have been studied in some detail. The cholesterol content of the myoblast plasma membrane is 0.33 pmol/mg protein (see Wakelam, 1985). Van der Bosch el ul. (1973) added I mg/ml cholesterol to the culture medium 4 hr before the initiation of fusion and observed inhibition; dipalmitoylphosphatidylcholinehad the same effect. These workers did not determine if the lipids were incorporated into the plasma membrane, so it is not possible to assess if their manipulation affected membrane fluidity. Cornell et ul. (1980) adopted a somewhat more systematic approach and demonstrated inhibition of fusion in cells where the synthesis of cholesterol was inhibited. Whether this effect is indeed due to an inhibition of fusion is unclear since the cells fail to aggregate (see Knudson and Horwitz, 1978). Experiments designed to modify other myoblast lipids have also been performed, utilizing enzymatic modification and culture medium manipulation. Both phospholipase C (Nameroff et ul., 1973; Schudt and Pette, 1976) and phospholipase A (Schudt and Petter, 1976) have been shown to inhibit myoblast fusion when present in the culture medium. Lysolecithin, one of the products generated by phospholipase A, was also found to inhibit fusion (Reporter and Norris, 1973). Kent (1979) has demonstrated that the inhibitory effect of phospholipase C involves the production of diacylglycerol from phosphatidylcholine. This lipid is unchanged during normal fusion (Wakelam and Petter, 1982); persistent generation of this diacylglycerol may down-regulate protein kinase C (see Section VI,C). Several groups have attempted to determine how changes in the lipid content of myoblast culture affect cell fusion. Prives and Shinitzky (1977) found that oleic and linoleic acids facilitated fusion, whereas stearic and elaidic acids caused a delay. Nakornchai et ul. (1981) found a moderate enhancement of fusion with linoleic acid, no effect with arachidic acid, and inhibition with arachidonic, linolenic, myristic, and oleic acids and with glycerol monooleate. The extent of incorporation of these fatty acids

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

99

into membrane lipids is unclear. A further variable is due to differences in lipid composition between horse and fetal bovine sera and between batches from the same species. Interestingly, the addition of lipids thought to be fusogenic, such as lysophosphatidylcholine (Reporter and Norris, 1973),or of other molecules that act as fusogens in other systems, such as dimethyl sulfoxide (Blau and Epstein, 1979; Miranda er al., 1983) and poly(ethy1ene glycol) (Wakelam, 1985), not only does not promote but in fact inhibits myoblast fusion. This type of study has, therefore, failed to provide information on the mechanisms of fusion. One reason for this is that myoblast fusion is an “on-off ’ process, with changes in the membrane transient and occurring only at the time of fusion itself. In an attempt to study membrane lipid changes that may occur at the critical moment of fusion, Wakelam and Pette (1982) inhibited fusion with a low concentration of Ca’’ in the medium and determined the incorporation of [”PIP, into myoblast phospholipids over a 60-min period after fusion was initiated by the addition of millimolar W + . Only phosphatidylinositol incorporated 32P.Phosphatidylserine, phosphatidylethanolamine, phosphatidylcholine, or sphingomyelin did not incorporate the label. This increased labeling is probably a reflection of increased lipid turnover and the breakdown of polyphosphoinositides (Wakelam, 1983).This breakdown is due to a phosphodiester cleavage and generates 1,2-diacylglycerol (Wakelam, 1983) and inosito1 phosphates (Wakelam, 1986). The diacylglycerol is rapidly phosphorylated to form phosphatidic acid (Wakelam, 1983).

V.

FUSION-INDUCED CHANGES IN MEMBRANE ORGANIZATION

A. Ultrastructural Studies

It is extremely difficult to obtain good electron microscope pictures of myoblast fusion. An approach which might conceivably be useful would be to use the suspension culture system described by Neff and Horowitz (1982). Neff et al. (1984) have shown that, in cultures which are inhibited from fusion by low Ca2+,the addition of millimolar Ca’+ results in significant membrane continutity within 20-30 min. Electron microscopy and freeze-fracture studies of such samples may lead to new structural information. Kalderon and Gilula ( I 979), using freeze-fracture techniques, observed particle-free regions in myoblast membranes at fusion sites. They also observed vesicles near the intracellular surface of the plasma membrane. These, they speculated, may play a role in the fusion process. A particle-

100

MICHAEL J. 0. WAKELAM

free region is consistent with the data of Kaufman and Foster (1984) in relation to antigenic site redistribution. Rash and Staehelin (1974) demonstrated that gap junctions link myoblasts and proposed that these could be the site of fusion, but Kalderon et af. (1977) also observed such structures in fusion-arrested cells. Fumagalli et ul. (1981) detected gap junctions, but in the areas of membrane apposition. These workers also showed that membrane fusion and fission do not take place in all areas of apposition, but only restricted sites. Fulton et ul. (1981) observed extensive reorganization of the cytoskeletal structure of detergent-extracted myoblasts on fusion. They detected numerous lacunae which might represent points of fusion that correspond to protein-free areas of the lipid bilayer. These lacunae disappear following fusion. In this context it is perhaps relevant that Bar-Sagi and Prives (1983) have demonstrated inhibition of fusion by calmodulin antagonists, an effect they assign to an inhibition of cytoskeletal reorganization. Even though many other cellular events are also inhibited by calmodulin antagonists, support for specificity comes from findings that both cytochalasin B (Holtzer et al., 1975), an actin polymerization inhibitor, and taxol (Antin et al., 1981), two microtubule disassembly inhibitors, inhibit fusion. B. Fluidity Studies

It is generally agreed that membrane fusion requires a fluid membrane state. The important of this for myoblast fusion is demonstrated by a 15to 20-fold increase in the rate of fusion when the temperature increases from 28" to 40°C (Van der Bosch et af., 1973). This is accompanied by a change in activation energy at about 35°C; at temperatures below 35°C E, = 302-318 kJ/mol. and above it E, = 74.4-92 kJ/mol. The biophysical demonstration of a change in membrane fluidity has been demonstrated by a variety of techniques. Weidekamm et af. (1976) measured fluorescence polarization of a membrane probe in cells cultured at Ca2+concentrations nonpermissive for fusion for 50 hr. The addition of I .4 mM Ca2+to the cells resulted in a sharp rise in polarization within 510 min. The lag in response is indicative of the possibility that biochemical changes must take place before fusion can occur. Indeed the time course bears a striking similarity to the changes in inositol phospholipid metabolism observed in cells cultured under similar conditions (Wakelam, 1983). Studies utilizing the resonance energy transfer technique (Herman and Fernandez, 1978, 1982) have shown that when cells fuse, the greatest changes in fluidity occur in areas of cell contact between fusing cells. It is tempting to consider that these regions of increased fluidity correspond to the particle-free areas reported by Kalderon and Gilula (19791, and to

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

101

have resulted from redistribution of surface antigenic determinants observed by Kaufman and Foster (1984). Further demonstration of the changes in fluidity have been provided by microviscosily (Prives and Shinitzky. 1977). fluorescence polarization recovery, and fluorescence depolarization methods (Elson and Yguerabide, 1979). VI.

PROPOSED MECHANISMS OF MYOBLAST FUSION

Numerous theories have been proposed over the years to explain the fusions not just of myoblasts, but of biological membranes in general. Many of these have invoked changes in the bilayer structure of the plasma membrane. The consensus is that some structure of lower stability than the bilayer is formed as an intermediate, but there is no agreement as to its nature even in model systems. A further factor to be taken into consideration is that there are major force barriers between membrane and indeed between pure bilayer structures. A. Changes in Bilayer Structure

It has been proposed that during membrane fusion there is a transition from a bilayer to a hexagonal 11 ( H i , ) phase. This transition involves the generation of an inverted micellar structure in which the lipids are organized in hexagonally arranged cylinders with the polar lipid head groups surrounding a narrow aqueous channel (see Cullis and Hope. 1978; Verkleij et d.,1979: Cullis and De Kruijff. 1979; Verkleij cf a / . , 1084). This suggestion is based on the fact that Ca?' can trigger the bilayer to HI, transition in mixtures of acidic phospholipids and phosphatidylethanolamine (reviewed in Cullis rr a / . , 1983). These authors propose that, in the presence of Ca'-. the nonbilayer tendencies of endogenous lipid are expressed and thus promote the fusion event. Freeze-fracture electron microscopic studies have provided some support for this proposal, at least in model systems. Incubation with Ca'+ of sonicated vesicles of phosphatid y let hanolamine (PE)-phosphat id y I serine, PE- phosphatid ylgl ycerol , PE-phosphatidylinositol, PE-cardiolipin, and PE-phosphatidic acid results in the formation of larger structures that are suggestive of the occurrence of fusion. The fusion in turn is accompanied by the observation of lipidic particles that are often localized at the fusion interface (Verkleij er id., 1984). The authors thus propose that fusion proceeds via the formation of inverted micellar structures. The involvement of these structures in fusion remains controversial. Lipidic particles are not detected in freeze-fracture electron micrographs of rapidly frozen liposome systems undergoing fusion, but are observed after long incubation times or in the presence of glycerol used as cryopro-

102

MICHAEL J. 0.WAKELAM

tectant (Bearer et al., 1982). Inverted micellar structures have not been clearly demonstrated in cellular systems, and no X-ray diffraction data in support of the hypothesis have yet been presented [for discussion in favor and against this hypothesis, see Verkleij et ul. (1984), Diizgunes (198% and Lucy (1984a,b)l.

B. Membrane Forces and Fusion When two cell membranes approach each other, various interactions can occur. These include long-range attractive and repulsive forces across the aqueous space and also the short-range interactions involving molecular contact between protruding membrane components. The forces have been measured in model systems and interpreted as being the vectorial sum of long-range forces, van der Waals attraction, and electrostatic and hydration repulsion (see Rand ef ul., 1985). In membrane fusion the most important of these is hydration repulsion. This effect keeps hydrophilic surfaces apart, with bilayer vesicles requiring an energy of 10-100 erg cm-2to make contact (Parsegian et al., 1979). These conditions do not favor fusion. If fusion is to occur, the hydration block must be overcome. A possible mechanism is the breakdown of inositol phospholipids. Removal of the polar head group of the inositol phospholipids leaves sn-l,2-diacylglycerol (DAG), a very hydrophobic lipid that is immiscible with water and thus has a profound effect on bilayer structure (see Fig. 4). Das and Rand ( 1984) have demonstrated that incorporation of DAG into phosphatidylcholine or PE vesicles brings about a transition from lamellar to nonlamellar hexagonal structures, as measured by X-ray diffraction. Das and Rand (1986) have also shown that removal of the polar head groups does not result in closer bilayer apposition when excess water is present or when osmotic forces are used to force the bilayer together. DAG effects the lamellar to H1, transition by increasing hydrocarbon volume. This results in membrane curvature with phospholipid spreading (see Fig. 4). When the total hydrocarbon volume exceeds that which can be adequately covered by phospholipids, repacking may occur, generating an HL1phase (see Das and Rand, 1986). Alternatively, the bilayer disruption may result in fusion, with the formation of an inverted micelle as an intermediate (cf. Verkleij et ul.. 1984; see Fig. 4). The contribution to this intermediate from each monolayer involves only about 50 molecules (Siegel, 1984). Thus a single DAG molecule would be equivalent to 2 mol %, a value which approximates that required for a lamellar to HI, transition in egg PE (Das and Rand, 1986). It is thus very tempting to view the production of DAG in myoblast fusion (Wakelam and Pette, 1982; Wakelam, 1983) as required to stimulate fusion, to remove hydration, to

3. MYOBLAST FUSION-A

103

MECHANISTIC ANALYSIS

generate membrane curvature, and also to stimulate the generation of an inverted micellar fusion intermediate (see Fig. 4). C. Proteins and Fusion

An alternative view to the role of Ca2+in fusion processes is that it leads to the stimulation of proteolysis. This could lead to the promotion of fusion in two ways. First, the Ca2+influx observed before the onset of myoblast fusion (David rt ul., 1981) might result in the movement of integral proteins away from fusion sites. This idea is supported by the observation of cytoskeletal reorganization (Fulton r t ul.. 1981) and the requirement for calmodulin in myoblast fusion (Bar-Sagi and Prives, 1983). When human erythrocytes are treated with subtilisin, fusion is stimulated (Ahkong et ul., 1978). A Ca2+-activatedneutral protease has been

n

0

A

O

0

O

O

B

D

E

FIG.4. Model for myoblast membrane fusion. ( A ) Two apposed cells are unable to fuse due to repulsive forces. ( B ) Stimulated inositol phospholipid breakdown generates diacylglycerol in the inner leaflet of each bilayer. ( C ) The diacylglycerol causes deformation of the bilayers. resulting in stretching of the outer leaflets. reducing the forces between the two bilayers. (D)Contact between the two “stretched” bilayers results in the generation of the H11 phase (or inverted micellar intermediate) as a fusion intermediate. (E) The inherently unstable HIIphase breaks down, leading to either the reformation of two independent cells or, as is shown, fusion.

104

MICHAEL J. 0. WAKELAM

shown to increase in activity during the fusion of rat myoblasts; following fusion its activity falls (Kaur and Sanwal, 1981). Couch and Strittmatter (1983) have reported the inhibition of rat myoblast fusion by inhibitors of metalloendoproteases. Couch and Strittmatter (1984) have further demonstrated that the inhibition is in fact of a cytosolic 80-kDa protein. Unfortunately the dependence of this enzyme on Ca2+ concentration and the potential substrates of the enzyme have not been investigated; it is thus difficult to determine at what point in the fusion process these proteases are required. Schollmeyer (1986) has reinvestigated the role of Ca2+-activated proteases in the fusion of L6 cells. She demonstrated that an 80-kDa Ca2+-activatedprotease changes from a random, disperse distribution to a predominantly peripheral disposition, following the transition from proliferative to prefusion myoblasts. She also proposed that this enzyme stimulates the release of fibronectin from the cell surface. It is tempting to speculate that Schollmeyer (1986) and Couch and Strittmatter (1984) have studied the same enzyme. A role for protein phosphorylation, in addition to that supported by the data of Senechal er ul. (1982) (see Section IV,B), is suggested by the work of Wakelam (1983, 1986). The DAG generated by inositol phospholipid breakdown will activate C-kinase (see Berridge and Irvine, 1984). The relevance of this to fusion is suggested by the observation that chronic treatment of myoblasts with the tumor-promoting phorbol esters inhibits fusion (Grove and Schimmel, 1981). Such chronic treatment has been demonstrated to down-regulate C-kinase (Collins and Rozengurt, 1982), and thus to reduce the level of phosphorylation. Indeed, short-term treatment of myoblasts with phorbol esters does not inhibit fusion, whereas incubation with polymyxin B, an inhibitor of C-kinase, results in inhibition of fusion (C. A. Evans and M. J. 0. Wakelam, unpublished experiments). The role of proteins in fusion processes is often thought of as inhibitory, clearance from fusing regions being required for membrane union to occur. However, Lucy (1984a.b) has proposed a more fundamental function. Taking into account the various proteolytic enzymes being both required for and stimulating fusion (see above), he has proposed that proteolysis may release hydrophobic (po1y)peptides which can perturb bilayers and induce fusion. This would be complementary to Sendai virus-induced fusion where the lipid bilayer is penetrated by a fusogenic protein (see Okada, Chapter 10, this volume). It is conceivable that a hydrophobic peptide could have the same effect as proposed for diacyglycerol in Fig. 4. This proposal will, however, remain speculative until such fusogenic, hydrophobic (po1y)peptides can be identified and characterized.

3. MYOBLAST FUSION-A

105

MECHANISTIC ANALYSIS

D. A Consensus Model for Myoblast Fusion?

Considering the different proposals of membrane changes and models of myoblast fusion. it is unfortunate. though perhaps not surprising, that no consensus model exists. Table 1 lists the changes, reported in the literature, upon fusion stimulation and the events that are claimed to be essential. It is tempting to combine some of these, e.g.. to suggest that inositol phospholipid breakdown (Wakelam, 1983) stimulates a rise in intracellular Ca?+concentration (Berridge and Irvine, 1984) as detected in myoblasts just before fusion (David ct ul., 1981). This in turn would lead to Ca?' activation of specific proteases (Schollmeyer, 1986) which could be equivalent to metalloendoproteases (Couch and Strittmatter. 1983). The changes in lipids or proteins could lead to bilayer disruption and therefore to membrane fusion. It is hoped that this proposal will be tested by determining the temporal relationship between the various effects (Table I ) . VII.

STIMULATION OF MYOBLAST FUSION

Not only is there no clear understanding of the membrane events involved in myoblast fusion, there is also no clear understanding of the stimulation of this process. The first attempt to define a biochemical signal

CHANCES

IN

TAB1.E 1 MYOBLAST MEMBR4NES

IN

RELATION

TO

FUSION"

Structural changes Increase in fluidity, stimulated by proteins, lipids Particle-free domains. stimulated by proteins. lipids Cytoskeletal framework, stimulated by proteins. lipids ( ? ) Nonbilayer structures. stimulated by lipids, proteins ('?I Unique myoblast membrane properties Phospholipid asymmetry Ganglioside pattern Stage-specific surface antigens lodination pattern Possible stimulating events Receptor-stimulated inositol lipid breakdown Ca?' entry Ca?+-activatedproteases Metalloendoproteases Cyclic AMP Protein phosphorylation " This table summarizes reported changes in myohla\[ membranes and their possible cause\. Delails of each are discussed i n the text.

106

MICHAEL J. 0.WAKELAM

for the stimulation of myoblast fusion came from the work of Zalin and co-workers, who suggested that the onset of fusion was stimulated by a rise in intracellular cyclic AMP levels. The basis for this proposal was a 10-fold increase in cellular cyclic AMP concentration in myoblasts 5-6 hr before the onset of fusion (Zalin and Montague, 1974); prostaglandin El promoted precocious fusion (Zalin and Leaver, 1975); inhibition of the enzyme cyclooxygenase and of prostaglandin synthesis by indomethacin inhibited myoblast fusion (Zalin, 1979). Support for the proposal has been provided by the studies of David and Higginbotham (1981). However, some studies have demonstrated that cyclic AMP inhibits fusion (Aiu et a/., 1973; Moriyama and Murayama, 1977). Shutzle et ul. (1984) observed a rise in cyclic AMP in fusing myoblasts, but after the onset of fusion. Moreover, inhibition by indomethacin had no effect on fusion; this would rule out a role for prostaglandin stimulating myoblast fusion. The rise in cyclic AMP following fusion may, however, be involved in increasing protein synthesis, especially of muscle-specific proteins (Shutzle et al., 1984; Zalin and Entwistle, 1984). Entwistle et al. (1986) have recently proposed that fusion is promoted by a prostanoid of the one series, but that the effect is independent of cyclic AMP. The general proposal of this group is that eicosatrienoic acid stimulates Ca2+uptake (Entwistle et a/., 1983, 1986). However, eicosatrienoic acid is not found in myoblasts (Kent et al., 1974; Boland et al., 1977). It is possible that linoleic acid is a precursor for the eicosanoid synthesis. According to Entwistle et a / . (1983) and Bevan el a / . (1985), Ca2+entry is regulated by a voltage-dependent K+ channel, a nicotinic acetylcholine receptor, according to Bevan et a / . (1983, though it can also be stimulated by ATP (A. Entwistle, personal communication). It is important in this context that specific ATP-stimulatable cation channels have been detected in fusion-competent myoblasts (Kolb and Wakelam, 1983). The fusion of myoblasts is thus probably a receptor-stimulated event. More direct evidence comes from the demonstration that chick primary myoblast fusion has an absolute requirement for embryo extract (Wakelam and Pette, 1983). The factor involved is likely to be of neuronal origin (Wakelam, 1986), as its action can be mimicked by the neuropeptides angiotensin I1 and vasopressin (Wakelam and Pette, 1983, 1984a,b). These peptides not only stimulate myoblast fusion but also inositol phospholipid breakdown and inositol phosphate generation (Wakelam, 1986). V I vasopressin receptors coupled to inositol phospholipid breakdown 1987). have also been demonstrated on L6 myoblasts (Wakelam et d., Three groups have proposed that transferrinlike molecules of neuronal origin stimulate fusion (Ii et ul., 1982; Oh and Markelonis, 1982; Beach et al., 1983). It is not clear if this has an effect on differentiation or on fusion.

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

107

Indeed. Wakelarn (1986) suggests that transferrin could be involved in alignment and not fusion itself.

VIII.

CONCLUSIONS

Notwithstanding 30 years of extensive study we still do not understand how myoblast fusion occurs or is regulated. However, testable proposals now exist. If this review has stimulated theoretical and experimental analysis of the several mechanisms that have been proposed, it will have served its purpose. ACKNOWLEDGMENTS Work from the author’s laboratory is supported by the Medical Research Council ( U i o . REFERENCES Adamo. S . . Zani, B., Siracusa. G . . and Molinaro. M . (1976). Expression of differentiation traits in the absence of cell fusion during myogenesis in culture. C ~ lDlfer. l 5, 53-67, Ahkong, Q.F.. Blow. A. M. J., Botham. G. M.. Launder. J. M.. Quirk, S. J.. and Lucy, J. A. (1978). Proteinases and cell fusion. FEES Lerr. 95. 147-157. Aiu. E. J., Holt, P. G.. and Simons. P. J . (1973). Myogenesis in u i m . Enhancement by dibutyryl CAMP. E r p . Cell Re.,. 83, 436-438. Antin. P. B.. Forry-Schaudies, S . . Friedman,’T. M.. Tapscoff. S . J.. and Holtzer. H. (1981). Taxol induces postmitotic myoblasts to assemble interdigitating microtubule-myosin / . 300-308. arrays that exclude actin filaments. J . Cell B ~ ( J90, Bar-Sagi, D., and Prives. J. (1983). Trifluoperazine. a calmodulin antagonist. inhibits muscle cell fusion. J . CeII Bifil. 97, 1375-1380. Beach. R . L.. Popiela. H.. and Festoff. B. W . (1983). The identification of neurotrophic factor as a transferrin. FEBS Lcw. 156, 151-156. Bearer. E. L.. Diizgunes, N . . Friend. D. S . . end Papahadjopoulos. D. (1982). Fusion of phospholipid vesicles arrested by quick-freezing. The question of lipidic particles as Actcr 693, 93-98. intermediates in membrane fusion. Biochirn. Bir~p/i\..~. Berridge, M. J.. and Irvine. R. F. (1984). lnositol trisphosphate, a novel second messenger in cellular signal transduction. Ncirrtri~( L o d o n )31, 3 15-39 I . Bevan, S . , Entwistle, A.. Warner, A.. and Zalin. R . (1985). Acetylcholine receptors play a role in the regulation of chick myoblast fusion. J . Physiol. (London)365, 89P. Bishoff. R. (1978). Myoblast fusion. Cell Sit[/: Rtw. 5. 127-179. Blau, H . M..and Epstein, C. J . (1979). Manipulation of myogenesis in uirro: Reversible inhibition by DMSO. Cell 17, 95-108. Boland. R., Chyn. T.. Roufa, D., Reyes, E.. and Martonosi, A . (1977). The lipid composition of muscle cells during development. Biochirn. Biophys. A m 489, 349-359. Burstein. M.. and Shainberg. A . (1979). Concanavalin A inhibit5 fusion of myobla4ts and appearance of acetylcholine receptors in muscle cultures. FEBS L P I I .103, 33-34. Cates, G. A., and Holland. P. C. (1978). Biosynthesis of plasma membrane proteins during myogenesis of skeletal muscle in vitro. Biochem. J . 174, 873-881. Cates, G. A., Brickenden, A. M., and Sanwal. B. D. (1984). Possible involvement o f a cell . h ~ r n259, . surface glycoprotein in the differentiation of skeletal myoblasts. J. B i ~ l C 2646-2650.

108

MICHAEL J. 0. WAKELAM

Chiquet. M., Puri, E. C., and Turner, D. C. (1979). Fibronectin mediates attachment of chicken myoblasts to a gelatin-coated substratum. J . B i d . Chem. 254, 5475-5482. Collins, M. K. L., and Rozengurt. E. (1982). Binding of phorbol esters to affinity sites on murine fibroblastic cells elicits B mitogenic response. J . Cell. Phvsiol. 112, 42-50. Cornell, R. B., Nissley, S. M., and Horwitz, A. F. (1980). Cholesterol availability modulates myoblast fusion. J . Cell B i d . 86, 820-824. Couch, C . B., and Strittmatter, W. J . (1983). Rat myoblast fusion requires metalloendoprotease activity. Cell 32, 257-265. Couch, C. B., and Strittmatter, W. J. (1984). Specific blockers of myoblast fusion inhibit a soluble and not the membrane-associated metalloendoprotease in myoblasts. J . B i d . Chem. 259, 5396-5399. Cullis, P. R., and de Kruijff, B. (1979). Lipid polymorphism and the functional roles of lipids in biological membranes. Biochim. Biophys. Acia 559, 399-420. Cullis, P. R., and Hope, M. J. (1978). Effects of fusogenic agent on membrane structure of erythrocyte ghosts and the mechanism of membrane fusion. Nairrre (London)271,672675.

Cullis, P. R.. de Kruijff, B., Hope, M. J., Verkleij. A. J., Nayar, R . , Farren, S. B., Tilcock. C., Madden, T. D., and Bally, M. B. (1983). Structural properties of lipids and their functional roles in biological membranes. Memhr. Fluid. B i d . 1, 39-81, Das, S . , and Rand, R. P. (1984). Diacylglycerol causes major structural transitions in phospholipid bilayer membranes. Biochem. Biophys. Res. Commun. 124, 491-496. Das, S., and Rand, R. P. (1986). Modification by diacylglycerol of the structure and interaction of various phospholipid bilayer membranes. Biochemistry 25, 2882-2889. David. J. D., and Higginbottom, C.-A. (1981). Fusion of chick embryo skeletal myoblasts: Interactions of prostaglandin El, adenosine 3',5'-monophosphate and calcium influx. Deu. B i d . 82, 308-316. David, J. D., See, W. M., and Higginbottom, C.-A. (1981). Fusion of chick embryo skeletal myoblasts: Role of calcium influx preceding membrane union. Deu. B i d . 82, 297-307. Den, H., and Malinzak, D. A. (1977). Isolation and properties of a p-o-galactoside specific lectin from chick embryo thigh muscle. J . B i d . Chem. 252, 5444-5448. Den. H., Malinzak, D. A., Keating, H. J., and Rosenberg, A. (1975). Influence of concanavalin A, wheat germ agglutinin and soya bean agglutinin on the fusion of myoblasts in uifro.J . Cell Biol. 67, 826-834. Diizgiineg, N . (1985). Membrane fusion. Subcell. Biochem. 11, 195-286. Easton, T. G., and Reich, E. (1972). Muscle differentiation in cell culture. Effects of nucleoside inhibitors and Rous sarcoma virus. J . B i d . Chem. 274, 6420-6431. Elson, H. F., and Yguerabide, J. (1979). Membrane dynamics of differentiating cultured embryonic chick skeletal muscle cells by fluorescence microscopy techniques. J . Supramol. Sfrucf. 12, 47-61, Entwistle, A., Warner, A., and Zalin, R. J . (1983). Is avian myoblast fusion controlled by voltage dependent calcium entry? J . Physiol. (London) 341, 2 IP-22P. Entwistle, A., Curtis, D. H . , and Zalin, R. J . (1986). Myoblast fusion is regulated by a prostanoid of the one series independently of a rise in cyclic AMP. J . Cell Biol. 103, 857-866.

Fulton, A. B., Prives, J., Farmer, S . R.. and Penman, S. (1981). Developmental reorganisation of the skeletal framework and its surface lamina in fusing muscle cells. J . Cell Biol. 91, 103-112.

Funagalli, G.. Brigonzi, A., Tachikawa, T., and Clements, F. (1981). Rat myoblast fusion: Morphological study of membrane apposition, fusion and fission during controlled myogenesis in uiiro. J . Ultrasiruci. Res. 75, 112-125. Gartner, T. K., and Podleski, T. R. (1975). Evidence that a membrane bound lectin mediates fusion of L 6 myoblasts. Biochem. Biophys. Res. Commun. 67, 972-978.

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

109

Gartner. T. K . , and Podleski. T. R . (1976). Evidence that the types and specific activity of lectins control fusion of L6 myoblasts. B ~ o c / I c JHiop/iys. //~. RiJs. Comrnrrn. 70, I 1421149.

Gilfix. B. M.. and Sanwal. B. D. (19x0). Inhibition of myoblast fusion by tunicaniycin and pantomycin. Rioc,/rcrn. Bioplrys. R c s . Co/n/nrtn.96, 1184-1 191. Grove. B. K.. Schwartz. G.. and Stockdale, F. E. (1981). Quantitation of changes in cell surface determinants during skeletal muscle cell differentiation using monospecific antibody. J. Srcprumol. Srritcr. Cell Bioc.henz. 17, 147- 152. Grove. K. I., and Schimmel. S. D. (1981). Generation of 1.2-diacylglycerol in plasma membranes of phorbol ester treated myoblasts. Biochern. Biophys Res. Comrnrtn. 102, 158164. Herman, B. A.. and Fernandez. S. M. (1978). Changes in membrane dynamics associated with myogenic cell fusion. J. CeU. Pliysiol. 94, 253-264. Herman. B. A., and Fernandez. S. M. (1982). Dynamics and topographical distribution of surface glycoproteins during myoblast fusion: A resonance energy transfer study. Biol.hen/i.yZr? 21, 3275-3283. Holtzer. H . , and Bischoff. R. (1970). Mitosis and myogenesis. In "The Physiology and Biochemistry of Muscle as a Food" (E.J. Briskey. K. G . Cassens. and B. B. Marsh. eds.). pp. 29-51. Univ. of Wisconsin Press. Madison. Holtzer. H.. Abbott. J., and Lach. J . (19.58). On the formation of multinucleated myotube5. A M / . R w . 131, 567. Holtzer. H.. Biehl. J.. Yeogh. G., Meganatham. R.. and Ka,ii. A. (1975). Effect ofoncogenic vim\ on muscle differentiation. Prot,. No//.A w d . S ( . i . U . S . A . 72, 4051-4055. li, I.. Kimura. I.. and Ozawa. E. (1982).A myotropic protein from chick embryo extract: Its purification. identity to transferrin, and indispensability for avian niyogenesis. Ueu. ni01. 94, 366-377. Kalderon. N . . and Gilula, N . B. (1979). Membnine events involved in myoblast fusion. J. Cell Biol. 81, 41 1-425. Kalderon, N., Epstein, M. L.. and Gilula. N . B. (1977). Cell-to-cell communication and myogenesis. J. Cpll B i d . 75. 788-806. Kaufnian. S. J.. and Foster, R. F. (1984). Antigenic changes on the myoblast membrane accompany development. Exp. Biol. Mcd. 9, 57-62. Kaur. H.. and Sanwal, B. D. (1981). Regulation of the activity of a calcium-activated neutral protease during differentiation of skeletal myoblasts. Cun. J . Bioc.hem. 59, 743-747. Kent. C. (1979). Stimulation of phospholipid metabolism in embryonic cells treated with phospholipase C. Proc.. Null. Acrid. Sci. U.S.A. 76, 4474-4478. Kent, C.. Schimmel, S. D.. and Vagelas, P. R. (1974). Lipid composition of plasma membranes from developing chick muscle cells in culture. Biochirn. Biophys. Acru 360,317321.

Knudsen. K. A.. and Horwitz, A. F. (1977). Tandem events in myoblast fusion. Dcw. Eiol. 58, 328-338. Knudsen. K. A,. and Horwitz. A. F. (1978). Towards a mechanism of myoblast fusion. J. Siqmimol. Strtrc~.8, 563-568. Kolb. H. A., and Wakelam, M. J. 0. (1983). Transmitter-like action of ATP on patched membranes of cultured myoblasts and myotubes. Ncrrlrrr (London)303, 621-623. Konigsberg. I. R . (1978). Skeletal myoblasts in culture. I n "Methods in Enzymology" (W. B. Jacoby and T. H. Pastan. eds.), Vol. 5 8 . pp. 51 1-527. Academic Press. New York. Kiihl. U . . Ocalan. M . . Timpi. R.. and von der Mark, K . (1986). Role of laminin and fihronectin in selecting myogenic versus fibrogenic cells from skeletal muscle cells in uirro. 1)iw. Biol. 117, 628-635. Linkhart. T. A , . Clegg. C. H . , and Hauschka. S. D. (1981). Myogenic differentiation in

110

MICHAEL J. 0.WAKELAM

permanent clonal mouse myoblast cell lines: Regulation by macromolecular growth factors in the culture medium. Deu. Biol. 86, 19-30. Lucy, J . A. (l984a). Do hydrophobic sequences cleaved from cellular polypeptides induce membrane fusion reactions in viuo? FEBS Lett. 166, 223-231. Lucy, J. A. (1984b). Fusogenic mechanisms. Ciba Found. S y m p . 103,28-39. MacBride, R. G., and Przybylski, R. J. (1980). Purified lectin from skeletal muscle inhibits myotube formation in v i m . J. Cell B i d . 85, 617-625. Maggio, B., Cumar, F. A., and Caputto, R. (1981). Molecular behaviour of glycosphingolipids in interfaces. Possible participation in some properties of nerve membranes. Biochim. Eiophys. Acta 650, 69-87. Merlie, J . P . , and Gros, F. (1976). I n uitro myogenesis. Expression of muscle specific function in the absence of cell fusion. Exp. Cell Res. 97, 406-412. Miranda, A. F., Babiss, I,. E., and Fisher, P. B. (1983). Transformation of human skeletal muscle cells by simian virus 40. Proc. Nut/. Acud. Sci. U . S . A . 80, 6581-6585. Moriyama, Y . , and Murayama, K. (1977). Cyclic nucleotide changes in fusion-arrested chick embryo myoblasts during differentiation in uitro. Cell Strucr. Funct. 2, 339-345. Moss, M., Norris, J. S., Peck, E. J., and Schwartz, E. J. (1978). Alterations in iodinated cell surface proteins during myogenesis. Exp. Ce/l Res. 113, 445-4450. Nakornchai, S., Falconer, A. R., Fisher, D., Goodall, A. H., Hallinan, T., and Lucy, J. A. (1981). Effects of retinol, fatty acids and glycerol monooleate on the fusion of chick embryo myoblasts in uitro. Biochim. Biophys. Acta 643, 152-160. Nameroff, M., Troffer, J. A., Keller, J. M.,and Minar, E. (1973). Inhibition of cellular differentiation by phospholipase C. I. Effects of the enzyme on myogenesis and chondrogenesis in uitro. J . Cell Eiol. 58, 107-1 18. Neff, N. T., and Horwitz, A. F. (1982). A rapid assay for fusion of embryonic chick myoblasts. Exp. Cell Res. 140, 479-483. Neff, N. T . , Lowry, C., Decker, C., Tovar, A,, Damsky, C., Buck, C., and Horwitz, A. F. (1982). A monoclonal antibody detaches embryonic skeletal muscle cells from extracelM a r matrices. J . Cell Biol. 95, 654-666. Neff, N. T., Decker, C., and Horwitz, A. F. (1984). The kinetics of myoblast fusion. Exp. Cell Res. 153, 25-3 1. Nowak, T . P., Kobiler, D., Roel, L. E., and Barondes, S. H. (1977). Developmentally regulated lectin from embryonic chick, pectoral muscle. J . Eiol. Chem. 252,6026-6030. Oh, T. H., and Markelonic, G. J. (1982). Chicken serum transkrrin duplicates the myotrophic effects of sciatin on cultured muscle cells. J . Neurosci. Res. 8, 547-567. Olden, K., Law, J., Hunter, V. A., Romain, R., and Parent, J. B. (1981). Inhibition of fusion of embryonic muscle cells in culture by tunicamycin is prevented by leupeptin. J . Cell Biol. 88, 149-204. Parfett, C . L . J., Jamieson, J. C., and Wright, J. A. (1981). A correlation between loss of fusion potential and defective formation of mannose-linked intermediates in independent concanavalin A-resistant myoblast cell lines. Exp. Cell Rev. 136, 1-14. Parsegian, V. A., Fuller, N. F., and Rand, R. P. (1979). Measured work of deformation and repulsion of lecithin bilayers. Proc. Nut/. Acud. Sci. U . S . A . 76, 2750-2754. Patterson, B., and Strohman, R. C. (1972). Myosin synthesis in cultures of differentiating chick embryo skeletal muscle. Deu. Eiol. 29, 113-138. Pauw, P. G . , and David, J . D. (1979). Alterations in surface proteins during myogenesis of a rat myoblast cell line. Deu. B i d . 70, 27-38. Pearson, M. L. (1980). Muscle differentiation in cell culture: A problem in somatic cell and molecular genetics. In “The Molecular Genetics of Development” (T.Leighton and W. F. Loomis, Jr., eds.), pp. 361-418. Academic Press, New York. Podleski, T. R., and Greenberg, I. (1980). Distribution and activity of endogenous lectin

3. MYOBLAST FUSION-A

MECHANISTIC ANALYSIS

111

during myogenesis as measured with antilectin antibody. Pro(,.Nrrtl. Acud. Sci. U.S.A. 77, 1054-1058. Podleski, T. R., Greenberg, I..and Nichols, S. C. (1979). Studies on lectin activity during myogenesis. Exp. Cell Res. 122, 305-316. Prives, J . , and Shinitzky, M. (1977). Increased membrane fluidity precedes fusion of muscle cells. Natrcre (London) 268, 761-763. Puri, E. C., and Turner, D. C. 11978). Serum-free medium allows chicken myogenic cells to be cultivated in suspension and separated from attached fibroblasts. Exp. Cell Rrs. 115, 159-173. Rand, R. P., Das, S., and Parsegian. V. A. (1985). The hydration force. its character universality and application: Some current issues. Chem. Scripru 25, 15-21. Rash. J . E., and Fambrough, D. (1972). Ultrastructural and electtophysiological correlate\ of cell coupling and cytoplasmic fusion during myogenesis in virro. Deu. B i d . 30, 166186. Rash. J. E., and Staehelin, C. A. (1974). Freeze-cleave demonstration of gap junctions between skeletal myogenic cells in uiuo. D ~ uBid. . 36, 455-461. Reporter. M.. and Norris, G. (1973). Reversible effects of lysolecithin of fusion of cultured rat muscle cells. Differenriurion 1, 83-9.5. Sandra, A., Leon, M. A,, and Przybylski, R. J . (1977). Suppression of myoblast fusion by concanavalin A: Possible involvement of membrane fluidity. J . Cell Sci. 28, 2.51-272. Sasse. J., Horwitz. A.. Pacifici, M., and Holtzer, H. (1984). Separation of precursor myogenic and chondriogenic cells in early limb bud mesenchyme by a monoclonal antibody. J. CeN B i d . 99, 1856-1866. Schollmeyer, J . E. (1986). Possible role of calpain I and calpain I I in differentiating muscle. Exp. Cell Rrs. 163, 413-422. Schudt, C . , and Pette, D. (1975).Influence of the ionophore A23187 on myogenic cell fusion. FEBS Lett. 59, 36-38. Schudt. C., and Pette, D. (1976). Influence of monosaccharides. medium Factors and enzymatic modification on fusion of myoblasts in vitro. Cytobiokopie 13, 74-84. Schudt, C., van der Bosch, J., and Pette, D. (1973). Inhibition of muscle cell fusion in v i t r ~ by Mg?' and K' ions. FEBS Leu. 32, 296-298. Schutzle, U . B., Wakelam, M. J . 0.. and Pette, D. (1984). Prostaglandins and cyclic AMP stimulate creatine kinase synthesis but not fusion in cultured embryonic chick muscle cells. Biochim. Biophys. Acrci 805, 204-210. Senechal, H..Prichard, A. L., Delain, D., Shapira. G., and Wahrmann. J. P. (1982). Changes in plasma membrane phosphoproteins during differentiation of an established myogenic cell line and a non-fusing cu-amanitin resistant mutant. FEBS Lett. 139, 209213. Sessions, A., and Horwitz, A . F. (1981). Myoblast aminophospholipid asymmetry differs from that of fibroblasts. FEBS L e u . 134, 75-78, Sessions, A.. and Horwitz, A. F. (1983). Differentiation-related differences in the plasma membrane phospholipid asymmetry of myogenic and fibrogenic cells. Biochim. Bi<>phy.y.ACIU 728, 103-1 I I . Shainberg, A. Yagil, G., and Yaffe. D. (1969). Control of myogenesis in uitro by Ca?' concentration in nutritional medium. E x p . Cell Res. 58, 163-167. Siegel, D. P. (1984). Inverted micellar structure in bilayer membranes. Formation rates and half-lives. Biophys. J . 45, 399-420. Van Der Bosch, J.. Schudt, C.. and Pette, D. (1972). Quantitative investigation of Ca2+and pH dependence of muscle cell fusion in virro. Biochem. Biophys. Rrs. Commrtn. 48, 326-332. Van der Bosch, J.. Schudt, C., and Pette, D. (1973). Influence of temperature, cholesterol.

112

MICHAEL J. 0. WAKELAM

dipalmitoyllecithin and Ca?' on the rate of muscle cell fusion. Exp. Cell RPS.82, 43343x. Verkleij, A. J . , Mombers, C.. Gerritsen, W. J., Leunissen-Bijvelt. J.. and Cullis. P. R . (1979). Fusion of phospholipid vesicles in association with the appearance of lipidic particles as visualized by freeze fracturing. Bioc,him. B i o p h v s . Ac,/a 555, 358-362. Verkleij, A . J . , Leunissen-Bijvelt, B.. deKruijff, B.. Hope, M.. and Cullis, P. R. (1984). Non-bilayer structures in membrane fusion. Cihn Found. Syrnp. 103, 45-59. Wakelam, M. J. 0. (1983). lnositol phospholipid metabolism and myoblast fusion. Biocht~tn. J . 214, 77-82. Wakelam, M. J . 0. (1985). The fusion of myoblasts. Bioc~hem.J . 228, 1-12, Wakelam, M. J. 0. (1986). Stimulated inositol phospholipid breakdown and myoblast fusion. Bioclwm. Soc. Trcitis. 14, 253-256. Wakelam, M. J. O . , and Pette, D. (19823. The breakdown of phosphatidylinositol in myoblasts stimulated to fuse by the addition of Ca?'. Biocliem. J . 202, 723-729. Wakelam. M. J . O., and Pette, D. (1983). Myoblast fusion, calcium and receptoroccupalion. Absir. FEBS Meet. I S r h S-12; MO-230. Wakelam, M. J. 0..and Pette, D. (1984a). Myoblast fusion and inositol phospholipid breakdown: Causal relationship or coincidence. Cihtr Fortnd. Symp. 103, 100- I18. Wakelam, M. J. O., and Pette. D. (1984b). Myoblast fusion-A receptor mediated event? Exp. B i d . Med. 9, 68-71. Wakelam, M. J . 0.. Patterson, S., and Hanley. M. R. (1987). L6 skeletal muscle cells have functional V,-vasopressin receptors coupled to stimulated inositol phospholipid metabolism. FEBS Lett. 210, 181-184. Wakeshull, E . , Bayne. E. K., Chiquet, M., and Fambrough, D. M. (1983). Characterisation of a plasma membrane glycoprotein common to myoblasts, skeletal muscle satellite cells and glia. Dru. Biol. 100, 464-477. Walsh. F. S., Moore, S. E . , Woodroofe, M. N . , Hurko, O . , Nayak, R . , Brown, S . M.. and Dickson, J . G. (1984). Characterisation of human muscle differentiation antigens. Ekp. B i ~ l Med. . 9, 50-56. Weidekamm, E., Schudt, C . , and Brdiczka, D. (1976). Physical properties of muscle cell membranes during fusion. A fluorescence polarisation study with the ionophore A23187. Biochim. Biophjis. A r i a 443, 169-180. Whatley. R., Ng. S . K.-C.. Rogers, J., McMurray, W. C., and Sanwall, B. D. (1976). Developmental changes in gangliosides during myogenesis of a rat cell line and its drug resislant mutants. Biochem. Biophys. R P S .Commun. 70, 180-185. Yaffe, D. (1968). Retention of differentiation potentialities during prolonged cultivation of myogenic cells. Proc,. Nail. Acud. Sci. U.S.A. 61, 477-483. Yoshioka, M., and Suroka, N . (1983). Cell surface proteins of rat myoblasts. Exp. Cell Re.s. 146, 219-223. Zalin, R. J.. and Entwistle, A. (1984). Two distinct eicosanoid-dependent steps are necessary for chick myoblasts to undergo cytodifferentiation and cell fusion. Exp. Biol. M e d . 9, 22-29. Zalin, R. J., and Leaver, R. (1975). The effect of a transient increase in intracellular cyclic AMP upon muscle cell fusion. FEBS Lett. 53, 33-36. Zalin, R. J . , and Montague, W. (1974). Changes in adenylate cyclase, cyclic AMP and protein kinase levels in chick myoblasts and their relationship to differentiation. Cell 2, 103- 108. Zalin, R. J . (1979). The cell cycle, myoblast differentiation and prostaglandin as a developmental signal. Deu. Biol. 71, 274-288.

Part II

Cellular TransportExocytosis and End ocytos is

This Page Intentionally Left Blank

CURRENT TOPICS I N MEMBRANES A N D TRANSPORT, VOLUME 32

Chapter 4 Exocytosis in Electropermeabilized Cells: Clues to Mechanism and Physiological Control PETER F . BAKER* MRC Secretory Mechunisms Group Depurtment of Physiology King's College London London WC2R 2LS, England

1. 11.

Ill. IV.

V.

VI.

Introduction Morphology of Exocytosis Clues of Mechanism Control of Exocytosis A . General Features B . Studies on Permeabilized Cells C. Generation of the Primary Message D. Toxins and Exocytosis Responding to the Primary Message A. What Is the Role of This Mg-ATP? B . Protein Kinase C and Exocytosis C. Inhibitors of C-Kinase Other Control Factors in Exocytosis References

I.

INTRODUCTION

Exocytosis is the process by which intracellular vesicles fuse with the inner face of the plasma membrane. It is one subset of a host of fusionfission reactions that effect membrane transfer within cells (Palade, 1975; Poste and Nicolson, 1978). Exocytosis serves two functions: to add new * Deceased. 115 Copyrighi I 1988 by Academic Pre% Inc All right\ of reproduction in any form rererved

116

P.

F. BAKER

membrane to the cell surface and to release into the extracellular fluid substances previously stored within the vesicle. It is usual to divide exocytosis into two types: continuous and triggered, both of which are probably present to some extent in most cells. Continuous exocytosis provides for the steady turnover of membrane components and the apparently nonregulated release into the external fluid of proteins such and albumin, immunoglobulins, and extracellular matrix components, while triggered exocytosis permits sudden changes either in membrane composition-such as when antidiuretic hormone acts to increase the water transporting capacity of the distal nephron-or in the release of biologically active molecules including nervous transmitter substances, many hormones, and enzymes. The existence in the same cell of continuous and triggered exocytosis raises important questions about how molecules are targeted to a particular secretory path. The available evidence suggests that each protein destined for exocytosis carries with it an address that determines whether it enters the continuous or triggered pathway (Kelly, 1985). Sorting probably occurs in the Golgi apparatus, and it is not unusual to find a number of different molecules routed to the same secretory pathway. For instance, in the adrenal medulla, secretory vesicles in the triggered pathway contain a number of different proteins (chromogranins, dopamine-/3-hydroxylase, and enkephalins), as well as a variety of molecules of lower molecular weight (catecholamines and ATP) that are accumulated as a result of specific transporters incorporated into the vesicle membrane. Even within the two major categories of exocytosis, there may be a number of variations on the basic theme. This certainly seems the case for triggered exocytosis because many cells contain more than one chemically distinct population of secretory vesicle, the contents of which can be released srlec‘tiurly in response to suitable stimuli (Bainton and Farquhar, 1984).Transmitter release in the 1966; Holmsen rt al., 1981; Knight rt d., nervous system may be a particularly interesting example of this general phenomenon (Lundberg and Hokfelt, 1983; Bartfai, 1985). There is growing evidence that many nerves contain at least two populations of vesicles, one of which often contains a conventional transmitter and the other a peptide, and it seems that the pattern of nerve impulses can determine which subpopulation of secretory vesicle and its chemically distinct contents is released. Exocytosis in the nervous system is of interest for a number of other reasons, especially the speed with which nervous transmitter substances can be released and the continuing debate about whether exocytosis is the route by which this rapid release occurs. At the neuromuscular junction, acetylcholine release is normally restricted to specialized “active zones”

4. CONTROL OF EXOCYTOSIS

117

(Dreyer rt d., 1973; Peper rt ( I / . , 1974). and less than I msec is required between arrival of the nerve impulse and release of transmitter (Katz, 1966). Direct evidence for vesicle fusion within this time period has been obtained by fast freezing the nerve terminal (Heuser cjt a / . , 1979; Chandler and Heuser, 1980; Ceccarelli and Hurlbut. 1980; Torri-Tarelli et ol., 1985). Any chemical reactions involved in exocytosis must be able to operate within this rime scale. One experimental problem in accepting exocytosis as the whole explanation for acetylcholine release at the neuromuscular junction is the appreciable release of acetylcholine that seems not to be in discrete, electrophysiologically detectable, packets (Katz and Miledi, 1977). It has been suggested that this nonquantal release may reflect the opening of some sort of channel (Tauc, 1982); but. for such a mechanism to work, there were need to be a high concentration of free acetylcholine in the cytosol, and this has never been demonstrated convincingly. In addition. if acetylcholine were exiting via a pore, the amount released would be dependent on the electrical driving force, and such a relationship has not been observed. A much more likely explanation is that nonquantal release reflects operation of an acetylcholine transporter that normally transports acetylcholine from the cytosol into the secretory vesicle; but, immediately following exocytosis, while the secretory vesicle is still fused with the plasma membrane this transporter remains active. serving to eject a steady stream of acetylcholine into the extracellular fluid (Edwards (’1 a / . . 1985). If this observation meets with general acceptance. it provides a very satisfactory explanation for the coexistence of quanta1 and nonquantal release at the neuromuscular junction, and one might expect to find equivalent behavior at other synapses.

II.

MORPHOLOGY OF EXOCYTOSIS

Exocytosis is normally monitored by the appearance in the extracellular fluid of some secretory product; but this does not necessarily provide much information about the underlying fusion process as there may be a delay between fusion and escape of secretory product. More direct information can be obtained either by electron microscopy (Heuser e t a / . , 1979; Chandler and Heuser, 1980) or, in some favorable cases, electrically, by monitoring fluctuations in membrane capacitance (Cole, 193S, Gillespie, 1979; Neher and Marty, 1982; Fernandez rt al., 1984). Electron microscopy, especially of fast-frozen material, can give information about the time course and morphology of fusion, while capacitance, which for a particular membrane system is proportional to membrane area, permits

118

P. F. BAKER

fusion and fission to be monitored directly; under optimum conditions individual fusion and fission events can then be resolved. The striking result from electron microscopy is that the initial fusion event is highly localized (Chandler and Heuser, 1980; Ornberg and Reese, 1981 : Schmidt et al., 1983) and may result from a single molecular interaction. The narrow pore that is generated widens rapidly into the equivalent of the wide-mouthed R profile seen in transmission electron micrographs and may finally flatten out, becoming indistinguishable from the plasma membrane. In the particular case of the adrenal medulla, application to stimulated cells of membrane markers specific to secretory vesicle antigens reveals discrete patches of vesicle membrane incorporated into the plasma membrane (Lingg et al., 1983). These patches are slowly and selectively retrieved by endocytosis. The extent to which vesicles flatten out and the time course and route of membrane retrieval probably differ from cell to cell, but, in general, they are poorly understood. The special form of endocytosis, “triggered endocytosis,” that serves to retrieve membrane added to the cell surface by “triggered exocytosis” seems to differ from the two other well-characterized forms of membrane internalization, fluid-phase and receptor-mediated endocytosis, indicating that for endocytosis, as for exocytosis, there are a number of parallel pathways (see Pastan and Willingham, 1985). 111.

CLUES TO MECHANISM

Although the molecular mechanism of exocytosis remains a mystery, it is worth examining some general principles that may well apply. What is very clear is that most membrane systems are reluctant to fuse spontaneously, and secretory vesicles can be subjected to over 1000 atm pressure without undergoing fusion. This observation makes it very likely that exocytosis is brought about by a specific set of biochemical events. Before discussing the little that is known about these events, it is perhaps worth digressing briefly to consider the one biological fusion about which something is known at the molecular level. This is fusion of certain enveloped viruses with cells. The trick seems to be that the viral membrane has projecting from its surface a special pH-sensitive “spike” protein (White et al., 1981, 1983; see also Chapters 9-1 1, this volume). Viruses attach to the cell surface, become internalized into endocytotic vacuoles, and, as the pH inside the vacuoles falls, the spike protein undergoes a conformational change revealing a hydrophobic sequence that buries itself into the neighboring wall of the vacuole. It is this step that seems to lead to fusion of the viral and vesicular membranes, permitting escape of the viral con-

4. CONTROL OF EXOCYTOSIS

119

tents into the cytosol. Clearly, the experimental evidence does not favor a pH-regulated process in exocytosis; but the dependence on a protein specialized for effecting fusion seems highly likely. Whatever mechanism is used, it must permit both specificity-in that only certain membranes fuse with each other-and control. Two very general solutions can be envisaged (Fig. I ) . By analogy with viral fusion described above, the fusion mechanism could be totally built into one of the reacting membranes (one-sided fusion), or fusion could require a specific molecular contribution from both reacting membranes (key-in-the-lock fusion). In its simplest form, one-sided fusion would not permit of any specificity or control: by blocking the fusion reaction, however, both could be achieved, especially if removal of the block were dependent on interaction with a component of the target membrane. Key-'

fusion'

K 'e-

fusion'

F I G .I . Membrane fusion may depend on specific proteins. Two possible scenarios for fusion of vesicles with the plasma membrane are shown: "one-sided" fusion in which the fusion mechanism is wholly built into one of the reacting membranes and "key-in-the-lock" fusion in which both membranes contribute essential components of the fusogenic process. The left-hand diagrams show the situation in which fusion will occur spontaneously, and the right-hand diagrams illustrate some ways in which spontaneous fusion can be blocked and thereby put under control. Throughout the figure, the location of the reacting molecules is only diagramatic, and their positions could be reversed.

120

P. F. BAKER

in-the-lock fusion, on the other hand, has built-in specificity, but would occur spontaneously in the absence of a blocking reaction. As is clear from the very diagrammatic representation in Fig. I , a single fusion process could be subject to more than one type of block, each of which could be controlled differently. Until more is known of the underlying molecular mechanisms of exocytosis, Fig. 1 is put forward in the hope that it may stimulate new experimental approaches to the problem.

IV.

CONTROL OF EXOCYTOSIS

A. General Features As its name implies, continuous or nonregulated exocytosis has not so far been shown to be subject to specific control, although it is of some interest that fusion of microsomal membranes derived from the liver, a very active secretor, requires low levels of calcium (Judah and Quinn, 1978) and that both continuous and evoked exocytosis virtually cease during mitosis (Warren, 1985). In the terminology of Fig. I,continuous secretion is an obvious candidate for either blocked one-sided fusion with removal of the block at the reacting membrane or unblocked key-in-thelock fusion. The rest of this discussion will focus on regulated or triggered exocytosis which has, until recently, been viewed as a calcium-dependent process. This is particularly true in most nervous and endocrine tissues, although a role for CAMPis strongly implicated in salivary and pancreatic exocrine secretion. The classic view derives from studies of transmitter release at the neurornuscular junction where arrival of the nerve impulse can increase acetylcholine release transiently by 10,000-fold and the whole of this increase is dependent on external calcium (Katz, 1966). Douglas (1968) subsequently widened the range of preparations examined and obtained evidence for a central role for calcium in stimulus-secretion coupling. The weight of experimental evidence was consistent with triggering of exocytosis by a rise in free calcium inside the cell, although a slow rate of secretion is usually possible at the resting level of free calcium. With the development of suitable calcium sensors such as aequorin and quinf, it became possible to test this hypothesis directly (Baker, 1972, 1974). In general, it seems that a rise in free Ca2+from its resting level close to 100 nM into the micromolar range is a sufficient stimulus for most exocytotic systems (Llinas et al., 1981; Knight and Kesteven, 1983; Rink and Hallam, 1984). This view is supported by the finding that microinjection of Ca2+into secretory cells or increasing their permeability to Ca2+by

121

4. CONTROL OF EXOCYTOSIS

exposure to a calcium ionophore such as A23187 or ionomycin, also brings about a rise in intracellular free Ca?' and initiates secretion. However, close inspection shows that secretion evoked by exposure to ionophore may need a somewhat higher free calcium concentration than secretion evoked physiologically; this suggests that factors other than calcium may also be involved. Of course, it is always possible that this discrepancy may arise either because a physiological stimulus brings about a highly localized rise in free calcium or because the ionophore is increasing calcium in the wrong part of the cell; but such arguments are difficult to apply to experiments, such as the one illustrated in Fig. 2a, where exposure of platelets to the physiological stimulus collagen triggers serotonin release with little or no detectable change i n free calcium (Rink et ( I / . , 19831, and in some systems, such as the parathyroid gland, release is associated with a very clear.fcr// in free calcium (Shoback c't a / . , 1984). These experiments suggest that calcium may be only one of a number of factors involved in the control of exocytosis. Other factors probably include cyclic nucleotides, products of phospholipid metabolism, guanosine nucleotides, and messenger molecules yet to be discovered. B. Studies on Permeabilized Cells A major advance in elucidating the relative important of these different systems has been the development of permeabilized preparations in which the intracellular environment of the exocytotic machinery is subDlacylglycerol

r e l e a s e d '0

-/-

0.0

K 51ii

U 0 '

0

a

-0-0-0-6

Collagen

10.0

Cd,*I P M ) FIG.2. Experiments showing apparently calcium-independent secretion and its possible explanation. (a) In intact platelets collagen promote5 secretion of ATP without a concomitant rise in free CaL' measured with quin2. Note: As serotonin (SHT) and ATP are stored in the same population of vesicles, similar results would be seen if serotonin were tn be measured. (From Rink et a / . . 1983.) (b) In electrically permeabilized platelets, diacylglyceol (which is produced following exposure to collagen) brings about a marked increase in sensitivity of the exocytotic machinery for Ca?'. (From Knight and Scrutton. 1984a.)

122

P. F. BAKER

ject to experimental control (Baker and Knight, 1978; Knight and Baker, 1982; Baker et al., 1985; Gomperts and Fernandez, 1985; Baker, 1986). Permeabilization can be achieved by a variety of techniques of which two, electropermeabilization (Baker and Knight, 1978, 1981; Knight and Baker, 1982) and detergent permeabilization (Brooks and Treml, 1983; Dunn and Holz, 1983; Wilson and Kirshner, 1983), have been widely used. In suitable preparations, the cell membrane can also be permeabilied by exposure to certain toxins (Thelestam and Mollby, 19791, viruses (Gomperts et al., 1983), complement (Schweitzer and Blaustein, 1980), or ATP4- (Cockcroft and Gomperts, 1980). Of the two commonest approaches, detergents make large lesions in the membrane, but also rapidly inhibit exocytosis (Baker and Knight, 1981), whereas electropermeabilization makes much smaller lesions (“holes” of effective diameter 4 nm) but has the advantages of being quick, chemically clean, and the “holes” can be very stable (Knight and Baker, 1982). An alternative approach suitable for some systems is to attach a single secretory cell to a patch pipet. Once the membrane under the pipet is broken, the cell interior can be perfused and exocytosis monitored as changes in capacitance (Neher and Marty, 1982; Fernandez rt al., 1984). Cortical plaques generated by breaking sea urchin eggs attached to a suitable substrate provide an even more accessible preparation consisting of portions of plasma membrane with secretory vesicles attached (Vacquier, 1975; Whitaker and Baker, 1983). A major conclusion from studies of these preparations is that calcium seems to be required for the bulk of secretory systems (Baker et al., 1980; Knight and Baker, 1982; Knight and Scrutton. 1980), even including ones in which cAMP seems to play a prominent role (Knight and Koh, 1984). A major exception is the mast cell under whole cell patch clamp, where exocytosis can be extremely difficult to trigger with anything (Neher and Almers, 1986). Other messenger systems seem, in the main, to modify the sensitivity of the exocytotic machinery to calcium. Thus, in the example shown in Fig. 2a, collagen triggers the hydrolysis of phosphatidylinositides, and the diacylglycerol that is liberated increases the apparent affinity of exocytosis for Ca2+ such that secretion can take place at the resting free Ca2+within the cell (Fig. 2b; Knight and Scrutton, 1984a,b). So far, it seems that the primary messengers acting rather directly in exocytosis are Ca2+,which is hydrophilic and presumably cytosolic, and diacylglycerol (DAG), which is highly hydrophobic and almost certainly restricted to membranes. Other messenger systems such as cyclic nucleotides may only be secondary in that they serve to modulate the production of the primary messengers. For instance, cAMP can both increase the

4. CONTROL OF EXOCYTOSIS

123

probability of opening of Ca channels in response to depolarization and alter the rate of DAG production as well; it should be stressed, however, that studies in this area are only in their infancy and much still remains to be discovered. Thus GTP is required for DAG production (Haslam and Davidson, 1984; Berridge, 1984). but it seems likely that its role is best described as permissive rather than as a messenger. C. Generation of the Primary Message Both cytosolic free CaZt and the level of DAG in the plasma membrane are under physiological control, and anything that alters either or both is likely to affect the rate of exocytosis. Cytosolic free Ca2+is determined by the interplay between factors tending to lower Ca’+ and factors tending to elevate it, whereas the membrane concentration of DAG is determined by the relative rates of synthesis and metabolism. The two most important sources of Ca’’ are the extracellular fluid from which CaZt enters the cell via a variety of voltage-sensitive and agonistgated channels and intracellular stores of which the endoplasmic reticulum (ER) seems the most important in terms of calcium mobilization. The voltage-sensitive calcium channels are of particular interest because their opening is usually controlled by other voltage-sensitive processes in the membrane-especially the sodium and potassium conductances-and modification of these other processes can lead to altered Ca?+ entry via the calcium channels. Thus, anything that prolongs the action potential tends to increase Ca?+ entry and transmitter release (Baker, 1972). This can be easily demonstrated by treating nerve terminals with pharmacological agents such as tetraethylammonium ions that block the potassium conductance. A physiologically generated reduction in potassium conductance has been implicated as a mechanism for increasing synaptic efficacy in one experimental model of memory (Siegelbaum et al., 1984). In addition to altering the voltage envelope that controls the opening and closing of calcium channels, there is an ever increasing range of physiological modulators (CAMP, protein kinase C. etc.) and pharmacological agents that can alter Ca2+entry through the calcium channel directly (see Hess et al., 1986). Particularly noteworthy are drugs that increase Ca2+entrythe so-called calcium agonists such as BAY-K 8644-and drugs that reduce or block CaZt entry, the ever increasing spectrum of calcium antagonists of which the dihydropyridines (e.g., nifedipine and nitrendipine) and phenylalkylamines (verapamil and D600) are well-known examples. Already, three different kinds of calcium channel have been recognized in nervous tissue (Nowycky et NI., 1985) and it is very likely that a number

124

P. F. BAKER

of physiologically and pharmacologically distinct calcium channels will prove to be involved in the control of exocytosis, making possible specific control of particular secretory processes. The mechanism of intracellular release of Ca2+from the endoplasmic reticulum involves the water-soluble messenger inositol trisphosphate (IP3)which is generated along with the other primary messenger of exocytosis, DAG, following interaction of various hormones and transmitter substances with receptors at the outer face of the cell membrane (see reviews by Berridge, 1984; Berridge and Irvine, 1984). These receptors activate phospholipase C by a mechanism that seems to involve GTPbinding proteins that can be ribosylated by pertussis toxin with blockage of phospholipase C activity. Phospholipase C can also be stimulated by introducing the GTP analog GTPyS into cells. Phospholipase C acts to promote the breakdown of a number of lipids including phosphatidylinosito1 bisphosphate (PIP2)which it hydrolyzes into the water soluble 1P3and the lipid-soluble DAG. IP3 serves to release Ca2+from the ER-possibly also via a mechanism that requires GTP-and the DAG remains in the membrane until it is converted to phosphatidic acid and other metabolites. From the viewpoint of secretion, both the Ca2+mobilized by IP3 and the membrane-associated DAG seem to be primary messengers for the activation of exocytosis. One implication of activating phospholipase C is that although the production of DAG and IP3 may be maintained, the IP3 can only bring about a transient mobilization of Ca2+from the ER, presumably because the ER has a limited amount of Ca2+to release and the released Ca2+tends to be taken up by other intracellular buffers. It follows that any agonist acting through PIP2 breakdown is likely to generate both Ca2+and DAG initially, but at later times DAG will predominate. This may contribute to the very marked phasic and tonic components of secretion seen in some systems. 0. Toxins and Exocytosis

In many biological systems, specific toxins have provided powerful tools for elucidating mechanism. Toxins that affect exocytosis do exist and can be divided into toxins that affect the generation of the primary message, for instance, omega toxin from mollusks of the genus Conus that blocks some calcium channels (Kerr and Yoshikama, 1984), and toxins that may influence exocytosis more directly. Possible members of this second group are the various botulinum toxins, tetanus toxin, and alatrotoxin (a-LTX). Botulinum toxin from Clostridium species is a potent inhibitor of neuromuscular transmission (Simpson, 1983, whereas a-LTX

125

4. CONTROL OF EXOCYTOSIS

from the venom of the black widow spider is a potent stimulator of exocytosis in the same system (see Meldolesi r t ( I / . , 1986). Particular interest centers on the botulinum toxins because of their potency. At the neuromuscular junction, it appears that these toxins bind to specific acceptors on the outer surface of the nerve terminal from where it is presumed that they gain entry into the cytosol (Dolly et a / . , 1984). Recently, an action of botulinum toxin type D has been demonstrated on cultured bovine adrenal medullary cells (Knight et a / . , 1985). Inhibition of catecholamine release is not associated with block of Ca2+entry and persists in permeabilized cells which display a lack of response to calcium (Fig. 3). The curious finding that only type D toxin is active may reflect failure of the other toxin types to penetrate the adrenal cell. This has received dramatic confirmation in a very recent series of experiments in which various toxins were introduced directly into the interior of the bovine adrenal medullary cells by perfusion through a patch pipet (Penner er a / . , 1986). In these experiments, a range of botulinum toxin types and also tetanus toxin were very effective blockers of the capacitance increase that follows introduction of 10 p M free CaZTinto the cell interior. The molecular target of these toxins is not yet known.

30

0

I

v

a m

c

v

r

,

L 0 -8

I

I

I

-7

-G

-5

1-og [ a 2 + ( M I FIG.3 . Inhibition of calcium-dependent exocytosis by pretreatment of cultured bovine adrenal medullary cells with botulinum toxin type D. Cells were subsequently rendered permeable and exposed to calcium: 0. control: 0 botulinum treated (From Knight P I d,, 1985.)

126

P. F. BAKER

V. RESPONDING TO THE PRIMARY MESSAGE How do Ca2+ and DAG alter the rate of exocytosis? It is still not possible to given even a remotely satisfactory answer to this crucial question, but in the terminology of Fig. 1, one possibility might be that they are involved in generating a fusogenic process. They might, for instance, be involved in the creation of a fusogenic molecule or in the removal of a block from an underlying fusogenic molecule. Permeabilized cells have permitted some features of exocytosis to be elucidated under conditions where the normal physiological controls have been by passed. Only a limited number of secretory systems have been investigated, but there are some common features: (1) Exocytosis is activated in the micromolar range of calcium concentrations. (2) It is largely unaffected by a wide variety of agents that interact specifically with the cytoskeleton. (3) It can usually be inhibited by agents that also inhibit both calmodulin-dependent processes and protein kinase C. (4) It is usually inhibited by raising the osmotic pressure of the medium by addition of sucrose. Systems differ, however, in other features especially their sensitivity to removal of ATP. The cortical reaction in the sea urchin egg persists for at least 1 hr after complete removal of ATP from the perfusion fluid (Baker and Whitaker, 1978), whereas secretion from adrenal medullary cells, human platelets, mast cells, exocrine and endocrine pancreas ceases within minutes of exposure to an ATP-free medium. Where it exists, this requirement for ATP is very specific, and the preferred substrate is MgATP; this suggests that ATP may be involved in phosphorylation. Those cells that require Mg-ATP also usually display some sensitivity to phorbol esters and analogs of diacylglycerol, as well as GTPaS which probably acts, at least in part, by stimulating production of endogenous DAG. Although the sea urchin egg may have a tightly bound store of ATP, the simplest hypothesis consistent with the data is that ATP is not essential for exocytosis in this tissue and that secretory control is exerted directly via a calmodulin-like molecule. This seems not to be the case in the various mammalian systems examined where there is a rather specific requirement for millimolar amounts of Mg-ATP (Fig. 4). A. What Is the Role of This Mg-ATP?

Although a number of proteins are phosphorylated in a calcium-dependent fashion, no one has discovered a substrate that is phosphorylated in a manner uniquely associated with exocytosis in eukaryotes; in view of the many possibilities inherent in Fig. 1, however, this may be too restric-

127

4. CONTROL OF EXOCYTOSIS

19'

O 0 - b Q

1 mM M p - A l p

0

LOG Ca2+ ( M ) FIG.4. Dependence on Mg-ATP of catecholamine release from permeable bovine adrenal medullary cells. Cells were suspended in potassium glutamate medium and rendered permeable by exposure to ten 2-kV pulses T - 200 Fsec (see Knight and Baker, 1982). Note that lowering Mg-ATP reduces the extent of secretion but not its apparent affinity for Ca?+.

tive a criterion. The failure to find a protein that is phosphorylated (or dephosphorylated) in parallel with exocytosis may also have a purely technical explanation in that minor phosphorylated species are very hard to detect. For instance, if only one phosphorylation occurred per vesicle, the total amount of phosphate incorporated would be extremely difficult to detect in a permeabilized preparation which is effectively a whole cell, and exocytosis has not so far been reconstituted reproducibly in simpler systems. One phosphoprotein, synapsin I. seems to be generally distributed in the mammalian nervous system where it apparently coats the cytosolic faces of synaptic vesicles and may play some role in exocytosis (Navone r t af.,1984; Llinas el al., 1985).And in certain protozoans, there is very clear evidence for dephosphorylation associated with exocytosis of trichocysts (Zieseniss and Plattner, 1985). Where Mg-ATP is essential, one experimental approach is to examine possible ATP-dependent reactions that may be involved in secretion. Two stand out: (1) as substrate for the proton pump that generates the acid internal environment and membrane potential of the secretory granule and (2) as substrate for the enzyme protein kinase C which is known to be activated both by Ca2+ and DAG (Nishizuka, 1984), the two major primary messengers for exocytosis.

128

P. F. BAKER

The first possibility, activation of the vesicular proton pump, is of some theoretical interest as one might envisage the proton gradient and potential developed by the pump as possible sources of energy for exocytosis. Specific models have been proposed on this basis such as the suggestion that when the vesicle and plasma membrane come into close contact, presumably via a calcium-dependent step, an anion channel may be generated that permits chloride to enter the vesicle from the extracellular fluid, the entry being driven by the positive internal potential of the vesicle (Pollard et nl., 1979). This specific idea can be tested in permeabilized cells; but the experimental data give no support because exocytosis can be triggered apparently normally in a chloride-free sucrose medium. Although other possible mechanisms utilizing these gradients can be envisaged, they are rather unlikely to be of importance physiologically because the pH gradient and potential can be collapsed, either separately or together, in a variety of ways without affecting the ability of the vesicles to participate in calcium-dependent exocytosis (Knight and Baker, 1985). The second possibility seems much more interesting, especially in view of the finding that most exocytotic systems are affected by analogs of DAG (Knight and Baker, 1983; Di Virgilio et al., 1984; Peterfreund and Vale, 1983; Jones et al., 1985) and can be inhibited by a variety of compounds that inhibit protein kinase C, albeit rather nonspecifically. Protein kinase C requires Ca2+,a phospholipid, and DAG for maximal activity. It utilizes Mg-ATP as substrate and is activated in the low millimolar range. DAG increases the apparent affinity of the enzyme for Ca2+and can be replaced in this action by certain phorbol esters including 12-0-tetradecanoyl phorbol-13-acetate (TPA). So far, protein kinase C is the only known substrate for these phorbol esters and, from the viewpoint of exocytosis, it is particularly interesting that protein kinase C becomes strongly associated with the plasma membrane in the presence of TPA and DAG (Kraft and Anderson, 1983). Secretory systems do not all respond to phorbol esters in the same way. Figure 5 summarizes the three main types of response that have been observed in permeabilized cells. Type I exocytosis is characterized by a rather small leftward shift in the calcium-activation curve following addition of the phorbol ester TPA or analogues of DAG. Type I1 has a very large leftward shift such that, at high concentrations of TPA or DAG, secretion seems to become independent of Ca2+,and in type 111 exocytosis phorbol esters do not alter the affinity for Ca2+but increase the extent of secretion. Some examples are type I, catecholamine release from the adrenal medulla (Knight and Baker, 1983); type 11, serotonin release from human platelets (Knight and Scrutton, 1984a,b) and possibly histamine release from mast cells (White et nl., 1984); type 111, P-N-acetylgluco-

129

4. CONTROL OF EXOCYTOSIS C

d , -8

-7

-6

-5

-8

L o g Ca

-7

-6

‘+/M

FIG.5 . Modification of calcium-dependen! secretion in “leaky“ cells by exposure to the phorbol ester TPA and other agents. Examples of the three classes of response that have been observed are shown. (A) Small leftward shift seen in catecholamine secretion from bovine adrenal medullary cells in the presence of3 nM TPA (+). I(Kl pM dioctanoylglycerol (A).(From Knight and Baker. 1983.) ( B ) Marked leftwird shift seen in serotonin secretion from human platelets in the presence of 30 nM T’PA (+I. 20 pM OAG (A).or 5 0 pM G W y S ( 0 ) tC) . Lack of any shift, only an increase in extent of exocytosi\ of N-acetylgluco~aminidase. also from platelet5 in the pre\ence of 10 mginil TPA (0). 30 &nl DAG (A). o r 0 . 5 U/ ml thrombin (A).(From Knight c f d., 1984.)

saminidase release from human platelets (Knight et u / . , 1984) and insulin release from pancreatic /3 cells (Jones 01 d., 1985). In the platelet there are at least two distinct populations of secretory granules, one of which exhibits type 11 and the other type 111 exocytosis. The coexistence of different types of exocytosis in the same cell may provide a mechanism for effecting differential release of secretory products. Thus, at low cytosolic free Ca?+a rise in DAG will provoke type 11 exocytosis but will have little or no effect on type Ill until the free Ca?’ is also elevated. 6. Protein Kinase C and Exocytosis

Comparisons between the properties of protein kinase C determined in the test tube and exocytosis in “leaky” cells are fraught with problems, two of the most important being lack of knowledge of the physiological substrates of the enzyme (it is conventionally assayed by phosphorylation of histone) and the very real possibility that its properties may change during isolation or when it no longer has membranes with which to associate. In addition, diacylglycerol is a generic term, and it is possible that different forms of C-kinase prefer different members of the diacylglycerol family. As the enzyme requires both Ca”’ and DAG for activation. it is of

-5

130

P. F. BAKER

interest to consider what would happen if the enzyme has a preferred order of binding of these two substrates. If we neglect the binding of phospholipid and Mg-ATP, there are three possibilities, which are shown in Scheme 1. The calcium dependence of exocytosis predicted by these

E

y

ECa

A

DAG

ECa

three reactions (Fig. 6) are quite different. Equation ( 1 ) shows a small, but limited, leftward shift on adding DAG; Eq. ( 2 ) a larger shift that continues to increase as the DAG concentration is increased such that at high DAG concentrations secretion will seem to be independent of Ca2+;and Eq. ( 3 ) shows no shift at all, only an increase in the extent of secretion. The striking parallel between the calcium dependencies of Fig. 6 and the three major patterns observed in permeable cells may be more than a coincidence (see Baker, 1986). Thus Eq. (1) is remarkably similar to type I, Eq. (2) to type 11, and Eq. ( 3 ) to type 111. It is certainly a very interesting point that these apparently different kinetics can all be generated via the same enzyme simply by specifying different orders of preferred substrate binding. It may be pertinent that protein kinase C exists in more than one form (Knopf et al., 1986; Ohno et al., 1987). C. Inhibitors of C-Kinase

In view of the data presented so far, it is important to estiblish whether protein kinase C is essential for exocytosis, or whether it merely plays a modulatory role. This is a difficult question to answer unequivocally because there are no really specific inhibitors of the kinase. The isolated kinase shares with calmodulin sensitivity to trifluoperazine and other socalled calmodulin antagonists and can also be inhibited by high concentrations of amiloride and polymyxin B. Exocytosis is also sensitive to these agents, but none of the inhibitors is specific enough to clinch the argument.

131

4. CONTROL OF EXOCYTOSIS 1

100

'/.-'-

/

/

i

/'

.C

0 -9

-8

./' -7

/'

0.1

a -6

-5

Log Cazt/

-4

M

FIG.6. Calcium dependence o f secretion calculated on the assumption of ordered binding o f DAG and Ca2+to protein kinase C. Nurnhers on the curves refer to DAG concentrations (pLM).

E -

DAG

EDAG7

'b

Ca KCa

DAG ECa

DAG

1 (A)

=

ECa Ca

DAG

Ca

DAG

ECa'C--LECa

'b

Kca

7 EDA+

DAG

AECa

Ca

For ease of calculating the curves, the affinity constants Kcd and K D for Ca?- and DAG. respectively, have been set at I pM. (From Baker, 1986.)

It is particularly striking that no inhibitor has been found which can remove the TPA shift while leaving calcium-dependent exocytosis otherwise unaffected. Agents that inhibit exocytosis all seem to leave a residual TPA shift, consistent with the view that the kinase is directly involved in the machinery of exocytosis.

132

P. F. BAKER

In conclusion, for those systems that are sensitive to both Calf and DAG, the attraction of the protein kinase C hypothesis is that one single molecule serves as receptor for Ca2+,DAG, and Mg-ATP; but this economy should be viewed with caution. It does, however, seem a reasonable working hypothesis that at least one subset of exocytotic reactions may utilize protein kinase C as part of the control process. But not all exocytotic systems are sensitive to phorbol esters, and it seems quite possible that other subsets of exocytotic reactions may utilize other calcium receptors such as calmodulin (Baker and Whitaker, 1980; Steinhardt and Alderton, 1982; Trifdro and Konigsberg, 1983), calelectrin (Sudhof et ul., 1983, synexin (Creutz et al., 1978), and the chromobindins (Creutz et al., 1983). Even where protein kinase C may be the main channel through which Ca acts, we do not know how it brings about exocytosis-especially the nature of the substrates, if any, that must be phosphorylatedand how fusion itself is generated. VI.

OTHER CONTROL FACTORS IN EXOCYTOSIS

Table I lists a number of agents that can inhibit calcium-dependent exocytosis in permeable adrenal medullary cells. They include chaotropic anions, high osmotic pressures, GTPyS and botulinum toxins. In no instance is the mechanism of inhibition known, but each may provide important clues to the mechanism of exocytosis. Thus chaotropic anions may serve to dissociate some macromolecular assembly that is essential for exocytosis, and inhibition by high osmotic pressures may reflect the importance of osmotic factors at some stage in the exocytotic process (see reviews by Baker and Knight, 1984; Holz, 1986). One possibility is that once a point fusion between vesicle and plasma membrane has been generated, entry of water through this microchannel may be essential to enlarge and complete the fusion process. Such water flow may be prevented or reversed by high external osmotic pressures permitting the micropore to reseal. Inhibition by GTPyS and botulinum toxin are quite different. That GTPyS should inhibit at all is surprising because it might be expected to stimulate phospholipase C and thereby raise the endogenous level of DAG. Whether or not this occurs, the net effect of GTPyS is strongly inhibitory, and inhibition persists in the presence of high concentrations of TPA that should saturate the TPA/DAG binding site on protein kinase C. The implication is that GTPyS is exerting a direct inhibitory effect on exocytosis quite distinct from its actions on phospholipase C and the endogenous level of DAG. A rather similar conclusion, but favoring a

133

4. CONTROL OF EXOCYTOSIS

TABLE I PROPERTlES OF CA'*-DEPENDFNT CATFCHOLAMINF RELEASF M F D U L L A RCY ~ L L ~

FROM

"LEAKY"A D R E N A L

Activation half-maximal at a Ca?- concentration of I pM Requirement for Mg-ATP is very specific: half-maximal activation requires I mM Unaffected by Agonists and antagonists of acetylcholine receptor\ Ca?*-channel blocker D 600 (100 pM) Agents that bind to tubulin (colchicine. vinblastinc. 100 pM) Cytochalasin B ( Im M ) Inhibitors of anion permeability (SITS. DIIIS. 100 phf) Protease inhibitors T L C K ( I mM), leupeptin ( I mM) Cyclic neucleotides (CAMP. cGMP. 1 mM) S-Adenosylmethionine ( 5 mM) Phalloidin ( I mM) Vanadate (10 M ) Leu and Met enkephalins. substance P (I00 pM) Somatostatin ( I p M ) NHJCl (30 m M ) Trimethyltin (0.2 m M ) Inhibited by Chaotropic anions, SCN 1 Br > CI Detergents (complete inhibition after a 10-niin incubation with 10 pgiml of digitonin. Brij 58. or saponin) Trifluoperazine (complete inhibition with 20 pg/ml) High Mg" concentration: small increase i n apparent K , for Ca?- accompanies large reduction in V,,,, High osmotic pressure; large reduction in L',,,;,, but no significant changes in the affinity for Ca?Carbonyl cyanide p-trifluoromethoxyphenyhydrazone (FCCP) (45% inhibition by 10 p M ) N-Ethylmaleimide ( N E M ) (100% inhibition at M) Neomycin ( I m M ) Ameloride ( I m M ) GTPyS (SO% inhibition at 5 p M ) Botulinum toxin types A, B. and D Tetanus toxin

stimulating role for GTP, has been reached by Barrowman Pt ul. (1986) from their studies of histamine secretion in permeabilized mast cells. The nature of the GTPyS binding site is unknown; but it is an attractive possibility that G proteins may be involved directly in exocytosis. Speculating a little, it is possible that when protein kinase C becomes associated with the plasma membrane, it also becomes subject to modulation by inhibitory (or stimulatory) G proteins, and, if this were true, it would not be a very large step to hypothesize that G proteins might mediate transmem-

134

P. F. BAKER

brane control of protein kinase C and exocytosis. A direct effect of this kind might, for instance, underlie a number of presynaptic control mechanisms. One specific case where this might apply is the inhibition of insulin release from pancreatic /3 cells following sympathetic nerve stimulation. This effect persists following application of noradrenaline to permeabilized /3 cells, and a direct effect on exocytosis looks a real possibility (Jones ef ul., 1986). Inhibition by botulinum toxin is particularly exciting as it persists for many days and may lead ultimately to isolation of one of the components of the secretory machinery. So far, the mechanism by which botulinum toxin acts has resisted exposure, but many related toxins act via ADP ribosylation of different GTP-binding proteins. A fascinating feature of botulinum action is that, after exposure to the toxin, adrenal medullary cells accumulate catecholamines and behave in all respects as if they have “forgotten” how to secrete despite exposure to the appropriate messages. Endogenous agents having actions similar to botulinum toxin could play a key role in the long-term alteration of synaptic efficacy, opening up the possibility that direct effects on the machinery of exocytosis may contribute to the long-term regulation of synaptic function. REFERENCES Bainton, D. F., and Fdrquhar, M. G . (1966). Origin of granules in polymorphonuclear leukocytes. Two types derived from opposite faces of the Golgi complex in developing granulocytes. J . Cell B i d . 28, 277-301. Baker, P. F. (1972). Transport and metabolism of calcium ions in nerve. Prog. Biophys. M o l . Biol. 24, 177-223. Baker, P. F. (1974). Excitation-secretion coupling. Recenr Adu. Physiol. 9, 51-86. Baker, P. F. (1986). Protein kinase C and exocytosis. Prog. Zoo/. 33, 265-274. Baker, P. F. and Knight, D. E. (1978). Calcium-dependent exocytosis in bovine adrenal medullary cells with leaky plasma membranes. Nature (London)276, 620-622. Baker, P. F. and Knight, D. E. (1981). Calcium control of exocytosis and endocytosis in bovine adrenal medullary cells. Philos. Trans. R . Soc. London Ser. B 296, 83-103. Baker, P. F. and Knight, D. E. (1984). Chemiosmotic hypotheses ofexocytosis: A critique. Biosci. R e p . 4, 285-298. Baker, P. F. and Whitaker, M. J . (1978). Influence of ATP and calcium in the cortical reaction in sea urchin eggs. Nature (London) 276, 513-515. Baker, P. F., and Whitaker, M. J. (1980). Trifluoperazine inhibits exocytosis in sea urchin eggs. J . Physiol.(London) 298, 5SP. Baker, P. F. Knight, D. E., and Whitaker, M. J. (1980). The relation between ionized calcium and cortical granule exocytosis in eggs of the sea urchin Echinus e.sculentus. Proc. R . Soc. London Ser. B 207, 149-161. Baker, P. F., Knight, D. E., and Umbach, J . A. (1985). Calcium clamp of the intracellular environment. Cell Calcium 6 , 5-14. Barrowman, M. M., Cockcroft, S . , and Gornperts, B. D. (1986). Two roles for guanine nucleotides in the stimulus-secretion sequence of neutrophils. Nature (London) 319, 506-507.

4. CONTROL OF EXOCYTOSIS

135

Bartfai. T. (1985). Presynaptic aspects of the co-existence of classical neurotransmitters and peptides. Trends Pharmacol. Sci. 6, 33 1-334. Berridge, M. J. (1984). Inositol trisphosphate and diacylglycerol as second messengers. Biochem. J . 220, 345-360. Berridge, M. J . , and Irvine, R. F. (1984). Inositol trisphosphate: A novel second messenger in cellular signal transduction. Nuticrc4London) 312, 315-321. Brooks. J . C.. and Treml. S. (1983). Catecholamine secretion by chemically-skinned cultured chromaffin cells. J. Neurochem. 40,468-473. Ceccarelli, B., and Hurlbut, W. P. (1980). Vesicle hypothesis of the release of quanta of acetylcholine. Physiol. Rev. 80, 396-441. Chandler, D. E. and Heuser, J . E. (1980).Arrest of membrane fusion events in mast cells by quick freezing. J. Cell Biol. 86, 666-674. Cockcroft, S. and Gomperts, B. D. (1980). The ATP- receptor of rat mast cells. Biochem. J . 188, 789-798. Cole, K. S . (1935). Electric impedance of Hipponoe eggs. J . Gen. Physiol. 18, 877-887. Creutz, C. E., Pazoles, C. J., and Pollard. M. B . (1978). Identification and purification of an adrenal medullary protein (synexin) that causes calcium-dependent aggregation of isolated chromaffin granules. J. B i d . Chem. 253, 2858-2866. Creutz. C. E . , Dowling, L. G., Sands J . J . . Vilan-Palasi, C.. Whipple, J . H.. and Zaks. W. J . (1983). Characterization of trans-lindins: Soluble proteins that bind to the chromaffin granule membrane in the presence of Ca”. J . B i d . Chem. 258, 14664414674, Di Virgilio, F., Lew, D. P.. and Pozzan. T. (1984). Protein kinase C activation of physiological processes in human neutrophils at vanishingly small cylosolic Ca” levels. Nature (London) 310, 691-693. Dolly. J . 0.. Black, J.. Williams, R. S. and Melling, J . (1984). Acceptors for botulinum neurotoxin reside on motor nerve terminals and mediate its internalization. Narure (London) 307,457-460. Douglas, W. W. (1968). Stimulus-secretion coupling: The concept and clues from chromaffin and other cells. Br. J . Pharmcicol. 34, 451-474. Dreyer, F., Peper. K . , Akert, K., Sandri. C.. and Moor. H. (1973). Ultrastructure of the active zone in the frog neuromuscular junction. Bruin Res. 62, 373-381. Dunn. L. A., and Holz, R. W. (1983). Catecholamine secretion from digitonin-treated adrenal medullary chromaffin cells. J. B i d . Chem. 258, 4989-4993. Edwards, C., Dolezal. V., Tucek, S., Zemkova. H.. and Vyskocil, F. (1985). Is an acetylcholine transport system responsible for non-quanta1 release of acetylcholine at the mouse myoneural junction’?Pruc. N u t . A u l d . Sci. U . S . A . 82, 3514-3519. Fernandez. J. M., Neher. E. and Gomperts, B. D. (1984).Capacitance measurements reveal stepwise fusion events in degranulating mast cells. Natirre (London) 312, 453-455. Gillespie, J . 1. (1979). The effect of repetitive stimulation on the passive electrical properties of the presynaptic terminal of the squid giant synapse. Proc. R . SOC. London. S e r . B 206, 293-306. Gomperts. B. D., and Fernandez. J . M. (19x5). Techniques for membrane permeabilization. Trends Biochem. Sci. 10, 414-417. Gomperts. D. D.. Baldwin. J . M., Micklem. K . J. (1983). Rat mast cells permeabilized with Sendai virus secrete histamine in response to Ca?* buffered in the micromolar range. Biochem. J . 210, 737-745. Haslam, R. J., and Davidson. M. M. L. (1984). Receptor induced diacylglycerol formation in permeabilised platelet; possible role for a GTP binding protein. J. Receptor Res. 4,605629. Hess, P., Fox, A. P., Lansman, J. B., Nilius. B., Nowycky, M. C., and Tsien, R . W. (1986).

136

P.

F. BAKER

Calcium channel types in cardiac, neuronal and smooth muscle-derived cells: Differences in gating, permeation and pharmacology. I n "Ion Channels in Neural Membranes," pp. 227-252. Liss, New York. Heuser, J. E., Reese, T. S . , Dennis, M. J., Jan, Y.. Jan, L.. and Evans. L. (1979). Synaptic vesicle exocytosis captured by quick-freezing and correlated with quanta1 transmitter release. J . Cell B i d . 81, 27.5-300. Holmsen, H . , Dangelmaier. C. A,. and Holmsen, H.-K. (1981). Thrombin-induced platelet responses differ in requirement for receptor occupancy. Evidence for tight couplings of occupancy and compartmentalized phosphatidic acid formation. J . Eiol. Cheni. 256, 9393-9396. Holz, R. W. (1986). The role of osmotic forces in exocytosis from adrenal chromaffin cells. Annu. Rev. Physiol. 48, 175-189. Jones. P. M., Stutchfield. J . , and Howell, S . L. (1985). Effects ofCa?' and a phorbol ester on insulin secretion from islets of Langerhans permeabilized by high-voltage discharge. FEBS L e f f .191, 102-106. Jones. P. M., Fyles, J . M.. and Howell, S . L. (1986). Regulation of insulin secretion by CAMP in rat islets of Langerhans permeabilized by high voltage discharge. FEBS Lett. 205, 205-209. Judah, J. D., and Quinn. P. S . (1978). Calcium ion-dependent vesicle fusion in the conversion of proalbumin to albumin. Nuturc (London)271, 384-385. Katz, B. (1966). "Nerve. Muscle and Synapse." McGraw-Hill. New York. Katz, B., and Miledi, R. (1977). Transmitter leakage from motor nerve endings. Pror. R . Soc. London. Ser. B 196, 59-72. Kelly, R. B. (1985). Pathways of protein secretion in eukaryotes. Science 230, 25-31. Kerr, L. M., and Yoshikama, D. (1984). A venom peptide with novel presynaptic blocking action. Nurrrre (London) 308, 282-284. Knight, D. E., and Baker, P. F. (1982). Calcium-dependence of catecholamine release from bovine adrenal medullary cells after exposure to intense electric fields. J . M e m h r . Eiol. 68, 107-140. Knight, D. E., and Baker, P. F. (1983). The phorbol ester TPA increases the affinity of exocytosis for calcium in "leaky" adrenal medullary cells. FEES Lett. 160, 98-100. Knight, D. E.. and Baker, P. F. (1985). The chromaffin granule proton pump and calciumdependent exocytosis in bovine adrenal medullary cells. J . Membr. Biol. 83, 147-156. Knight. D. E., and Kesteven. N. T. (1983). Evoked transient intracellular free Ca?' changes and secretion in isolated bovine adrenal medullary cells. Proc. R . Soc. London, Ser, E 218, 177-199. Knight, D. E.. and Koh. E. (1984). Ca?' and cyclic nucleotide-dependence of amylase release from isolated rat pancreatic acinal cells rendered permeable by intense electric fields. Cell Calcium 5, 401-418. Knight, D. E., and Scrutton, M. C. (1980). Direct evidence for a role for Ca2* in amine storage granule secretion by human platelets. Throtnh. Res. 20, 437-446. Knight, D. E.. and Scrutton, M. C. (1984a). The relationship between intracellular second messengers and platelet secretion. Eiochem. Sot,. Trcins. 12, 969-972. Knight, D. E.. and Scrutton, M. C. (1984b). Cyclic nucleotides control a system which regulates Ca?+-sensitivityof platelet secretion. Nrttrtre (London) 309, 66-68. Knight. D. E., Niggli, V . , and Scrutton. M. C. (1984). Thrombin and activators of protein kinase C modulate secretory responses of permeabilized human platelets induced by CaZT.Eur. J . Biorhem. 143, 437-446. Knight, D. E . , Tonge, D. A., and Baker, P. F. (1985). Inhibition of exocytosis in bovine adrenal medullary cells by botulinum toxin type D. Nutitre (London) 317, 719-721. Knopf, J . L., Lee, M.-H., Sultzman, L. H.. Kriz, R. W., Loomis, C. R., Hewick, R. M.,

4. CONTROL OF EXOCYTOSIS

137

and Bell. K. M. (1986). Cloning and expression of multiple protein kinase C cL>NAs. C’o// 46. 491-502. Kraft. A . S.. and Anderson. W. U . (1983). Phorbol e\ters increase the amount of C:i>*. phospholipid-dependent protein kinase associated with pla\ma membrane. Ntrrrrw (LOt7dOtJ)301,

621-623.

Lingg. C.. Fischer-Colbrie. K.,Schmidt. W. and Winkler. H . (19x3). Expo5ure o f an antigen of chromaftin granules in cell surface during eaocytosis. Ntrrrrre (Lotrdori)301, 610-61 I . Llinas, R . , Steinberg. 1. Z., and Walton. K . (19x1). Pre\ynaptic calcium currents i n squid giant synapse. Biop/r,v.\. J . 33, 289-32 I . Llinas. K.. McGuineah. T. L. M.. Leonard. c‘. S..Sugimori. M . . and Greengard, P. ( 19x5). lntraterminal injection o f synapsin I or calciumic;rlmodulin-dependent protein kinase I I alters neurotransmitter release at the squid giant synapse. Proc.. N o r / . A ( rrd. .Sr.i. U. S . A .82. 3035-3039. Lucy. 1 . A . . and Ahkong. 0 . F. ( 19x6). An osmotic model tor the fusion of hiological membranes. FEBS L r / r . 199, 1 - 1 I . Lundberg. J . M . . and Hokfelr. T. ( 1983). (‘0-existence of peptide\ and classical neurotransi. .6, Y ~375-333. mitter\. 7wrrd.c N c u ~ ( J Meldolesi. J.. Scheer. H.. Madeddu. L . . and Wank. E. (1986). On the mechanism of action of cu-latrotoxin. the presynaptic stimulator)) toxin o f the black widow spider venom. Trt,trtl.r P/rtrr.trlrrc~o/.sc.i. 7, 151-155. Navone. F.. Greengard. P . . and l>e Caniillo. 1’. (19x4). Synap\in I in nerve terminals: Selective aswciation with \mall synaptic vesicles. .Sr.ietii.c, 226, 1209- 12 I I. Neher. E. and Almers. W . (1986). Fast calcium transients in rlit peritoneal mast cells are not sufficien~t o trigger- e x o c y t o k . E M B O J . 5, 5 1-53, Neher. E. and Marty, A . (1982). Discrete changes o f c e l l membrane capacitance ohserved under conditions of enhanced secretion in bovine adrenal medullary cells. Proc . Ntrrl. A<.t/tl..Si.i. U.S.A. 79, 6712-6716. Nishizuka. Y , (19x4). The role o f protein kinase C in c e l l surface signal transduction and tumour promotion. Nrr/riw (Loridotr) 308, 693-698. Nowycky. M. C.. Fox. A . P. and T‘sien, K. W. (19x5). Three types of neuronid ciilciuni channel with different calcium agonist sen5itivity. Nrrrrtw (Lontlotr) 316, 440-443. Ohno. S.. Kawasiki, H.. Imajoh. S . , Suzuki. K.. Inagaki. M . . Yokokura. H.. Sakoh. I... and Hidaka. H. (1987). Tissue-specific expression of three distinct types of rabbit protein kinase C. Nrrrrrr.~( L o t i d o t / )325, Ihl-166. Ornberg, K. L.. and Keese. 1’. S. (1981). Beginning of exocytosis captured by rapid freezing o f Lirnri/rc.\ amoehocyte5. J . Cell Biol. 90, 40-54. Palade. G. E. (1975). lntracellular aspects o f the process o f protein secretion. .S(,ic,ttc c 189, 347-358. Pabtan. I . and Willingham. M. C . , eds. (1985). Pathways of endocytosis. Irr “Endocytosis.” pp. 15-28. Plenum. New York. Penner. R.. Neher. E.. and Dreyer. F. (19x6). In1r;icellularly injected tetanus toxin and its fragment B inhibit exocytosis in bovine adrenal chromaffin cells. Nrrrirrt, (Lonclot7) 324, 76-78. Peper. K.. Dreyer. F . . Sandri, C., Akert. K . . and Moore. H. (1974). Structure and ultrastructure o f t h e frog motor endplate. A freeze etching study. Cell Tis.sIrt, Re.\. 149, 437455. Peterfreund. R . A , , and Vale. W. W. (1983). Phorbol diesters stimulate somatostatin secretion in cultured brain cells. E~idoc~vit7o/og~ 20, 300-708. Pollard. H . B . . Pazoles. C . J.. Creutz. C. E.. and Zinder. 0. (1979). The chromaffin granule and possible mechanisms o f exocytosis. /ti/. R e v . C y / o / . 58, 159. Poste, G., and Nicolson. G. L.. eds. (1978). Membrane fusion. Ci4l Sit
138

P. F. BAKER

Rink, T. J. and Hallam, T. (1984). What turns platelets on. Trends Biochem. Sci. 12, 215219. Rink, T. J., Sanchez, A., and Hallam, T. (1983). Diacylglycerol and phorbol esters stimulate secretion without raising cytoplasmic free calcium in human platelets. Natctre (London) 305, 317-319. Schmidt, W., Patzak, A., Lingg, G., Winkler, H., and Plattner, H. (1983). Membrane events in adrenal chromaffin cells during exocytosis: A freeze-etch analysis after rapid cryofixation. Eur. J. Cell Biol. 32, 31-37. Schweitzer, E. S . , and Blaustein, M. P. (1980). The use of antibody and complement to gain access to the interior of presynaptic nerve terminals. Exp. Brain ReS. 38, 443-453. Shoback, D. M., Thatcher, J., Leombruno. R. and Brown, E. M . (1984). Relationship between parathyroid hormone secretion and cytosolic calcium concentration in dispersed parathyroid cells. Proc. Nail. Acud. Sci. U . S . A . 81, 3113-31 17. Siegelbaum, S . A., Camardo, J. S., and Kandel, E. R. (1984). Serotonin and cyclic AMP close single K' channels in Aplysia sensory neurones. Nurure (London)299, 413-417. Simpson, L. L. (1985). Molecular pharmacology of botulinum toxin and tetanus toxin. Annu. Rev. Pharmacol. Toxicol. 25, 155-188. Steinhardt, R. A., and Alderton, J. M. (1982). Calmodulin confers calcium sensitivity on secretory exocytosis. Nature (London) 295, 154- 155. Siidhof, T. C., Walker, J. H., and Fritsche, U. (1985). Characterization of calelectrin, a Ca2+-bindingprotein isolated from the electric organ of Turpedo marmoru/u. J . Neurochem. 44, 1302-1307. Tauc, L. (1982). Non-vesicular release of transmitter. Physiol. Rev. 62, 857-893. Thelestam, M., and Mollby, R. (1979). Classification of microbial, plant and animal cytolysins based on their membrane-damaging effects on human fibroblasts. Biochirn. Biophys. Arta 557, 156-169. Torri-Tarelli, F.. Grohovaz, F., Fesce, R., and Ceccarelli, B. (1985). Temporal coincidence between synaptic vesicle fusion and quanta1 secretion of acetylcholine. J . Cell B i d . 101, 1386-1399. Trifaro, J. M., and Konigsberg, R. L. (1983). Microinjection of calmodulin antibodies into chromaffin cells provides direct evidence for a role of calmodulin in the secretory process. Fed. Proc., Fed. Am. Soc. Exp. B i d . 42, 456. Vacquier, V. D. (1975) The isolation of intact cortical granules from sea urchin eggs: Calcium ions trigger granule discharge. Deu. B i d . 43, 62-74. Warren, G. B. (1985). Membrane traffic and organelle divison. Trends Biochem. Sri. 10, 439-443. Whitaker, M. J., and Baker, P. F. (1983). Calcium-dependent exocytosis in an in uirro secretory granule plasma membrane preparation from sea urchin eggs and the effects of some inhibitors of cytoskeletal function. Proc. R. Soc. London, Ser. B 218, 397-413. White, J., M a t h , K . , and Helenius, A. (1981). Cell fusion by Semliki Forest, influenza and vesicular stomatitis viruses. J . Cell B i d . 89, 674-679. White, J., Kielan, M., and Helenius, A. (1983). Membrane fusion proteins of enveloped animal viruses. Q. Rev. Biophys. 16, 151-196. White, J. R., Ishizaka, T., Ishizaka, K.and Sha'afi, R. I. (1984). Direct demonstration of increased intracellular concentration of free calcium as measured by quin-2 in stimulated rat peritoneal mast cell. Proc. Nail. Acad. Sci. U.S.A. 81, 3978-3982. Wilson, S. P., and Kirshner, N. (1983). Calcium-evoked secretion from digitonin-permeabilized adrenal medullary chromaffin cells. J . B i d . Cham. 258, 4994-5000. Zieseniss, E. and Plattner, H. (1985). Synchronous exocytosis in Parumecium cells involves very rapid reversible dephosphorylation of a 65-kD phosphoprotein in exocytosis-competent strains. J. Cell B i d . 101, 2028-2035.

CURRENT TOPICS IN MEMBRANES AND TRANSPORT, VOLUME 32

Chapter 5

Exocytosis and Membrane Recycling JACOPO MELDOLESI"9 AND BRUNO CECCARELLP f * Department of Pharmacology CNR Center of C.vt(~pharmric'oloKy $ Scient$c Institute S . Rafiuele and t Center .for the Studv of Peripheral Nerrroputhies and Nertromusculur Diseuses University of Milan 20129 Milan, Italy

1. Lntroduction 11.

Exocytosis A . Second Messenger Contrr. of Regulate Exocytosis 6 . Membrane Fusion-Fission in Exocytosis C. Toxins

111.

Endocytosis A. Membrane Sorting in Endocytosis 6. Regulation of Endocytosis

IV. Conclusion References

1.

INTRODUCTION

Exocytosis, in the past also referred to as emyocytosis or reverse endocytosis, is the process by which the content of specific membrane-bound secretory organelles (named, depending on their size, granules or vesicles) is released in bulk to the extracellular medium. The mechanism that underlies the exocytotic process is the fusion-fission of the limiting membrane of the organelle with the plasma membrane (or a specialized portion of the latter). Over the years the meaning of the term exocytosis has broadened to include not only release, but also the causative fusionfission event (for reviews, see Palade, 1975; Ceccarelli and Hurlbut, 139 Copyright 1988 by Acddernic Pre% InL All rights ot rupruduc(ion In m y form r c w v e d

140

JACOPO MELDOLESI AND BRUNO CECCARELLI

1980a; Meldolesi and Ceccarelli, 1981; De Lisle and Williams, 1986; Baker and Knight. 1986; Meldolesi et ul., 1987). The initial identification of exocytosis was based exclusively on the morphological observation of R , or flask-shaped, infoldings of the plasma membrane, characterized by the same size and, in many cases, by a content corresponding in texture and density to a partially dissolved secretory organelle. This type of observation (made first in the acinar cells of the exocrine pancreas by Palade. 1959, and later reported in a variety of other secretory systems) still remains an important criterion for the recognition of the process. Beginning already in the 196Os, however, different secretory systems were found to be characterized by markedly variable frequency of exocytotic images, irrespective of their relative ability to secrete. For example, it proved to be much more difficult to obtain exocytotic images at stimulated nerve terminals, where vesicles fuse very rapidly, than at glandular cells, such as pituitary mammotrophs or acinar cells of the parotid gland, where the rate of discharge is much slower. The reason is that the ability to obtain exocytotic images depends not only on the frequency of fusion, but also and especially on the average residence time of fused, but recognizable, secretory organelles at the cell surface (see Ceccarelli and Hurlbut, 1980a.b). Hence morphological studies as those carried out by conventional thinsection electron microscopy cannot provide quantitative information on exocytosis rates. Instead it is necessary to use information from biochemical and/or electrophysiological experiments, in which the rate or release of specific products in response to appropriate stimuli is evaluated. In doing so it is necessary to take into account the complex composition of the secretory granule content. If release occurs by exocytosis, then all constituents of the granule should appear in the extracellular fluid at the same time and in the correct stoichiometric ratio. Otherwise, release is likely to have occured by a nonexocytotic mechanism. An important limitation of the biochemical study of exocytosis is its usually poor resolution. In most experiments, samples are collected over periods of several seconds, usually from a number of cells. Results based on such studies therefore represent averages of processes that are not necessarily synchronous and of identical magnitude in individual cells. In addition, it is often difficult in biochemical investigations to study the release of secretion products at defined sites of the plasma membrane located strategically in relation to the directionality of the signal. the geometry of the extracellular space, or other comparable parameters. Neither of these limitations applies when exocytosis is studied by electrophysiological approaches. These studies record electrical events that have been induced in a target cell by specific products released by an

5. EXOCYTOSIS AND MEMBRANE RECYCLING

141

adjacent secretory system, usually a nerve terminal at a synapse. This approach appears ideal in many respects: It is the only one that permits the direct observation of individual secretory events in real time: the results permit a relatively high degree of resolution in space and can be quantitated. On the other hand, the electrophysiological approach cannot be applied to all secretory systems. So far, detailed studies have been carried out primarily at the neuromuscularjunction, where it is possible to evaluate in a quantitative fashion changes of the postsynaptic potentials caused by the release of acetylcholine (ACh) quanta and the ensuing activation of the nicotinic receptor. A completely different development of the electrophysiological studies on exocytosis is based on the measurement of stepwise increases of the plasma membrane capacitance due to subsequent fusions of individual granules. Measurements of this kind have been made on secretory cells such as chromaffin cells (Neher and Marty, 1982) and mast cells (Fernandez er d . , 1985) studied by the patch clamp technique. After exocytosis, the fused membrane of secretory granules and vesicles does not remain permanently integrated in the plasma membrane, but is removed and reused in successive secretory cycles (reviews: Douglas, 1974; Meldolesi and Ceccarelli, 1981; Herzog, 1981; Farquhar, 1982, 1985: Meldolesi ef d.,1987). The concept of membrane retrieval (hereafter also referred to as exo-endocytotic coupling) was initially put forth based on the recognition of ( 1 ) the transient nature of the surface enlargement that follows a burst of exocytosis; ( 2 ) the increased labeling of cytoplasmic vesicles in cells and synapses stimulated to secrete in the presence of an extracellular marker; and (3) the difference in chemical composition of the plasma membrane as compared to that of the granule (or vesicle) membranes. These results suggested the existence of an efficient mechanism(s) of endocytosis which matches precisely, both in terms of quantity and quality, the insertion of the membrane that occurs at the cell surface during exocytosis. That exo-endocytic coupling is indeed the case was later demonstrated by turnover experiments, which revealed the much longer (usually 10-100 times) intracellular half-life of the integral proteins of the membranes of secretory organelles as compared to that of the corresponding secretory proteins. Further morphological studies demonstrated. first by freezefracture (De Camilli ef d . .1976) and more recently by immunocytochemical reactions studied at the light and electron microscope levels (Phillips ez ul., 1983; Patzak rt ul., 1984; Patzak and Winkler, 1986), the specific removal of granule (or vesicle) membrane components from the plasma membrane immediately after their exocytotic insertion. Biochemical (Thilo, 1985) and electrophysiological (Fernandez er id., 1985) results that

142

JACOPO MELDOLESI AND BRUNO CECCARELLI

support exo-endocytotic coupling have also been reported, so that at the present time the concept has gained general acceptance. Because of its complexity and its heterogeneous features in various secretory systems, exocytosis has not yet been adequately explained in molecular terms (see Geisow and Fisher, 1986; Baker and Knight, 1986), and some aspects, like mechanisms of control, have defied molecular analysis. On the other hand, significant progress has occurred in the understandingof some aspects, usually as a consequence of the introduction of new experimental approaches and techniques. The aim of this chapter is to provide a current (although by no means exhaustive) account of the state of the art in the field. Because previous experience and present research interests of the authors have primarily been focused on neurons and neurosecretory cells, major attention has been paid on these systems.

11. EXOCYTOSIS There exists at least two types of secretory systems, those capable of regulated, intermittent discharge of their secretion products, and those that appear to release their products continuously. The former include endocrine, neuronal, and some other secretory cells. The latter include plasma cells and other cells whose secretory activity is more limited and concerns primarily the proteins and proteoglycans of the extracellular matrix. Recent work by Kelly and associates (1985) suggests that nonregulated (or constitutive) secretion exists in cells also capable of regulated release. In other words, the regulated pathway has developed in addition to the constitutive pathway. Both types of secretion are believed to utilize exocytosis. The constitutive pathway, however, does not require a large storage compartment. The typical granules and vesicles of secretory cells and synapses can therefore be assigned to the regulated pathway. Whether a given protein is targeted for one or the other pathway appears encoded in their structure. Current evidence favors the existence of only one constitutive pathway, whereas the situation of the regulated pathway seems more complicated. Thus, in the case of blood polymorphonuclear leukocytes, two populations of releasable granules were demonstrated. One of these, the azurophilic granules, was lysosomal in nature. In addition there exists another class of secretory organelles, the secretory vesicles, which are completely separate from the azurophilic and the so-called specific granules and which discharge gelatinase, an enzyme specific to these vesicles (Dewald et al., 1982). Exocytosis of each of the three classes of secretory organelles is regulated independently (Lew et al., 1986). Similarly, glan-

5. EXOCYTOSIS AND MEMBRANE RECYCLING

143

dular cells contain different classes of secretory organelles, each containing different proteins. An example is the thyroid gland. Work on the bovine pituitary gland from our laboratory has recently led to the identification of mixed acidophilic cells, where growth hormone and prolactin are preferentially segragated into separate secretion granules (Fumagalli and Zanini, 1985). An additional class of granules, containing another secretory protein, secretogranin 11, has also been identified in these cells immunocytochemically (Hashimoto t t a/., 1987) (Fig. I ) . As yet regulation of hormone and secretogranin discharge from mixed cells has not been investigated. The secretory cell where coexistence of multiple regulated secretion pathways has been investigated in detail is the neuron. Many neurons can release two types of secretion products independently, a classical neurotransmitter and one or more peptides each stored in organelles of different morphology. The two pathways appear to operate in parallel (Lundberg and Hokfelt, 1983). One, analogous to the pathway in endocrine cells, is responsible for peptide neurotransmitters; the second, for classical neurotransmitters, is peculiar inasmuch as it can be filled locally in the nerve terminal and is characterized by fast exo-endocytotic cycling. Interestingly, the release organelles of the two regulated pathways of neurons might possess specific membrdne-associated components. All the vesicles containing classical neurotransmitters contain at their cytosolic surface a peripheral phosphoprotein, synapsin I (De Camilli et ul., 1983; Navone r>t al., 1984; Valtorta et al., 1988; Fig. 2), that might be involved at some initial step in the regulation of the release process (Llinas el a / . , 1985; see also Section 11,B). The mechanism(s) by which classical neurotransmitters, in particular, acetylcholine (ACh), are released from nerve terminals has attracted interest since the classic studies of Katz and associates, who proposed the synaptic vesicle as the structural correlate of the ACh quantum whose discharge induces the unitary postsynaptic event, the miniature end plate potential (Katz, 1966). Subsequent studies have lent considerable support to this concept, but alternative views have also been proposed (Tauc, 1982). The recent demonstration (by rapid freezing electron microscopy) of a temporal coincidence between the appearance of vesicle fusion and the end plate potential evoked after a single stimulus of the nerve at a frog neuromuscular junction (Torri-Tarelli rt d., 1985; Fig. 3) rules out an important argument raised in the past against the classic interpretation of Katz. Nonexocytotic, molecular release of ACh has been shown also to occur at the neuromuscular junction, and to predominate under resting conditions (Katz and Miledi, 1977; Gorio ef al., 1978). At catecholaminergic celfs and synapses, exocytosis is the process responsible for quanta1 release (Douglas, 1974; Smith and Winkler. 1972; Thureson-

FIG. 1. lmmunocytochemical demonstration of secretory proteins segregated within different granules in an ultrathin cryosection of a single bovine pituitary cell. Small gold particles reveal growth hormone, preferentially located within a granule of large size. whereas large gold particles reveal secretogranin 11. Granules unlabeled by either type of gold particle contain another hormone, prolactin (P). ~ 8 4 , 9 7 0(From . S. Hashimoto and J . Meldolesi, unpublished results.)

5. EXOCYTOSIS AND MEMBRANE RECYCLING

145

FIG.2. High power electron micrograph of a portion of a frog neuromuscular junction revealing the specific immunolocalization of synapsin I. A rabbit antiserum and ferritinconjugated F(ab)?goat anti-rabbit IgCs were used to indirectly localize synapsin I. Notice the specific association of the dark ferritin particles with the outer membrane of synaptic vesicles. Mitochondria, axolemma. and muscle fiber components are not labeled. X96.600. (Original from A. Villa, F. Valtorta. P. De Camilli, and B. Ceccarelli.)

Klein, 1983). Nonexocytotic background release also exists in these systems, and it may become important when drug treatment is initiated. Acetylcholine release at the neuromuscular junction can be readily studied quantitatively when secretion occurs at low rates. It can also be quantitated when secretion is vigorous and sustained, evoked by electrical stimulation of the nerve (synchronous ACh secretion; Ceccarelli er al., 1973). When, however, vigorous quanta1 secretion occurs asynchronously (as is the case with various secretagogues, e.g., a-latrotoxin, oua-

FIG.3. Electron micrograph of a neuromuscular junction from a cutaneous pectoris nerve-muscle preparation of frog quick-frozen 2.5 msec after a single stimulus while bathed in I mM Caminopyridine. At the level of an active zone, different degrees of association between synaptic vesicles and the prejunctional membrane are evident in this cross section. Arrowheads indicate clear vesicle openings. Many images suggestive of intermediate steps between fusion and fission are also evident. The time elapsing between the stimulus and the evoked postsynaptic event recorded at the edge of the muscle where the specimens were collected was about 3 msec. Thus vesicle openings were caught by this physical fixation precisely at the time that quanta of ACh were released. Asterisks indicate a large vesicular structure. ~44.520.(Reproduced with permission from Torri-Tarelli er a/., 1985.)

5. EXOCYTOSIS AND MEMBRANE RECYCLING

147

bain, La3+,high K + , the miniature end plate potential frequency becomes so high that individual events can no longer be resolved. Even under these conditions, however, accurate estimates can be obtained by applying a statistical approach based on the principle of noise analysis (Segal et al., 1985). This allows deducing the amplitude and rates of occurrence of elementary events. Specifically, the procedure derives the rate, amplitude, and waveform of miniature end plate potentials from the power spectrum, variance, and skew of the fluctuations of the muscle membrane potential recorded at the end plate region (Fesce rt a / . , 1986a,b). The relevance of this procedure for obtaining information on exocytosis-coupled endocytosis is discussed in Section II1,B. An example of results obtained at the frog neuromuscular junction is shown in Fig. 4. A. Second Messenger Control of Regulated Exocytosis

Until recently, the various secretory systems were widely believed to be under the control of distinct regulatory mechanisms. Examples are the neuromuscular junction and the parotid acinar cells. In the former, the key role played by the cytosolic Ca?+concentration, [Ca2+],,was initially inferred from the strict dependency of the ACh quanta1 release evoked by electrical stimulation on extracellular Ca?+.This was later confirmed by a large body of additional experimental evidence (see Katz, 1966; Silinsky, 1983. In contrast, secretion of salivary enzymes from parotid acinar cells was shown to be under the control of CAMP (Schramm and Selinger, 1975). Even in these two systems, however, evidence began to accumulate suggestive of a greater complexity of the exocytotic control, with possible involvement of multiple second messengers. Thus, at the neuromuscular junction several methods caused exocytosis even in Ca2+-free media (Hurlbut and Ceccarelli, 1979; Cinsborg and Jenkinson, 1976; Reichardt and Kelly, 1983); in the parotid acinar cells some release was obtained even after application of agents that induce the rise not of CAMP but of [Ca”], (Schramm and Selinger, 1975; Putney. 1986). During the past 5 years, the concept of multiple mechanisms regulating exocytotic discharge has gained considerable momentum. This has been the result of a number of important contributions. For the first time it has become possible to evaluate separately the role of various second messengers. In addition, comparison of the results obtained in different secretory systems has yielded interesting clues. Among the major advances in the field, we wish to mention the following: ( I ) the recent understanding of the role played by a receptor-triggered metabolic reaction, the hydrolysis of polyphosphoinositides-such a reaction is now known to be responsible both for the redistribution of Ca2+from a “microsomal” store to the

A

B

A

1'";

130 min

170 min

Time in ouabain (minl

FIG. 4. (A) Intracellular recording of miniature end plate potential (mepps) obtained during the effect of 50 p M ouabain in Ringer's solution modified by the addition of 4 mM Mg*+.After a long delay the frequency of mepps increased, reached a peak in about 20 min. and then subsided to low values in about 40 min. The upper record was taken I5 min before the onset of the increase in mepp frequency, whereas the lower record was taken at the peak of mepp frequency. It is clear from this lower record that mepps overlap so extensively that their individual features cannot be resolved. (B) Time courses of the changes in mepp and number of quanta secreted (m) at a junction bathed in 50 p M ouabain in frequency (0) Ca2+-freesolution (1 mM EGTA and 4 mM Mg2+).The parameters of mepps were deduced from the variance, skew, and power spectra of the end plate recordings by applying a recently described (Segal e l a / . , 1985; Fesce et d.,1986b)modification of the classic fluctuation analysis. (C. D)Electron micrographs of cross-sectioned frog neuromuscularjunctions. (C) Resting muscle soaked for 3 hr in modified Ringer's solution (4 mM Mg2+).(D)Muscle fixed 60 min after the peak frequency induced by 100 pM ouabain applied in the same modified Ringer's solution. The depletion of synaptic vesicles and some infoldings of the axolemma (arrowheads) are evident. x 16,500. (Reproduced with permission from Haimann ef a / . , 1985.)

5. EXOCYTOSIS AND MEMBRANE RECYCLING

149

cytosol and for the activation of an important phosphorylating enzyme. protein kinase C (Berridge and Irvine. 1984: Nishizuka. 1986); (2) the introduction into secretion studies of the techniques of cell permeabilization that offer the unique opportunity of direct access to the cytosol from the extracellular medium (Baker and Knight, 1981: Knight and Baker, 1982: Baker, Chapter 4, this volume): ( 3 ) the development, from both oocytes and protozoa, of cell-free secretory systems, composed of granules already docked at the plasma membrane, where the requirements of the fusion event can be directly and conveniently investigated (Whittaker and Baker, 1983; Vilmart-Seuwen et (11.. 1986); (4) the development of ingenious techniques for measuring the average [Ca’? ]i in cells of any size by means of trapped fluorescence indicators (quin2 and, more recently. fura2)-these same indicators can be used to clamp [Ca”], at defined levels, for example. in the 10-X-lO-h M range, i.e., from 0. I - to I0-fold the resting [Ca?’], level in most cells (Tsien rt a / . , 1982. and 1984; Grynkyevicz r t d.,1985): ( 5 ) the introduction of the patch clamp technique by which a wealth of information can be obtained concerning. for example. the conductance of single channels, the currents carried by individual ions. and the changes of plasma membrane capacitance underlying the addition and removal of vesicle membranes (Sakmann and Neher, 1983: Neher and Marty, 1982: Fernandez rt d . , 1984). The information on regulation of exocytosis that has been gathered in a variety of cell types, following the technical and intellectual achievements described above. confirms that in most secretory systems changes in [Ca” J i are the signal to trigger exocytotic responses. [Ca2+],transients have been accurately measured in permeabilized as well as intact cells, and were found to vary in the 0.3-1 pM range (see, inter uliu, Knight and Baker, 1982, 1983; Knight and Scrutton, 1985; Pozzan ct ul., 1984; Di Virgilio er ul., 1987). An unexpected finding was the transiency of the release responses induced by the elevation of [Car+],in both permeabilized and intact cells. In intact cells, this could be due to the uneven distribution of Ca” in the cytosol after stimulation. In particular. localized Ca2+gradients in the cytosolic regions immediately adjacent to the plasmalemma may play a key role in the triggering of exocytotic responses. This problem, previously analyzed only theoretically (for a comprehensive review, see Silinski, 1985). is now beginning to be investigated experimentally (Williams rt ul., 1985: Keith cr ul., 1985; Connor, 1986). Studies with electrically permeabilized chromaffin cells and platelets have revealed a requirement for ATP, together with Ca?+.for regulated exocytosis (Baker and Knight, 1981: Knight and Scrutton, 1985). These results appear to correlate well with the known energy requirement for exocytosis reported in some, but not all, intact secretory systems. In

150

JACOPO MELDOLESI AND BRUNO CECCARELLI

permeabilized PC12 cells (a tumor line derived from a pheochromocytoma), however, the requirement for ATP was not found (Ahmert-Hilger et ai., 1985), while results in a paramecium-derived cell-free system suggest that ATP may be needed to prime exocytosis rather than to bring about membrane fusion (Vilmart-Seuwen et al., 1986). In parotid cells stimulation of CAMP, as with isoproterenol, is more effective in causing exocytosis than a direct change in intracellular calcium (Schramm and Selinger, 1975; Putney, 1986). In pancreatic acinar cells, however, the effect of the two second messengers is reversed (Gardner and Jensen, 1986). It should be noted, however, that the effect of CAMP on secretion is not always stimulatory, but can be inhibitory, for example, in mast cells. In contrast, no inhibition of exocytosis has ever been observed after the application of activators of protein kinase C, such as diacylglycerol and phorbol esters. By themselves these agents are without effect on [Ca2+]iin most cells. In a variety of secretory systems [e.g., platelets (Rink et ul., 1983),pancreatic acinar cells (Gunther, 1981); PC12 (Pozzan et ui., 1984)],protein kinase C activators trigger exocytotic responses even when administered alone. In granulocytes, platelets, and, to a lesser extent, in PC12 cells, this stimulation is maintained even when [Ca2+]iis clamped to very low levels, i.e., below lo-# M (Rink ef a / . , 1982; Di Virgilio et ul., 1984; Pozzan ef ui., 1984). In contrast, in other systems [e.g., bovine chromaffin cells (Knight and Baker, 1983),frog neuromuscularjunctions (Haimann et al., 1987)],the effect of protein kinase C activators alone was negligible, but large, synergistic effects were observed when these activators were administered together with agents that increase [Ca2+Ii,for example, Cazt ionophores. These synergistic responses were not only greater, but also more persistent than those elicited by [Ca2+]i-increasingagents alone (Pozzan et al., 1984; Kolesnik and Geshenghorn, 1985). In other words, the activators of protein kinase C appear to prevent the inactivation of the stimulatory effect of [Ca2+],on exocytosis. Activation of the receptors followed by polyphosphoinositide hydrolysis leads to a rise in [Ca2+]iand the generation of diacylglycerol, the physiological activator of protein kinase C. The study of the release responses elicited by the activation of these receptors has yielded interesting results. For example, in pancreatic acini stirnulation of the muscarinic receptor results in a rapid but transient (-5 min) rise of [Caz+]i;and in a much more prolonged ( > I hr) stimulation of exocytosis. The latter response is entirely dependent on the activation of the receptor, inasmuch as it can be quickly shut off by the application of the muscarinic antagonist, atropine (Pandol et ul., 1985). Thus the two second messengers generated by receptor activation appear to play coordinated roles. Raised

5. EXOCYTOSIS AND MEMBRANE RECYCLING

151

[Ca' +Ii triggers the response, and stimulated protein kinase C maintains it for long periods of time. The results that we have summarized emphasize the complexity of the interaction between the two intracellular signals, Ca!' and protein kinase C. As discussed by Baker (1984) and Baker and Knight (1986), various models can be envisaged to account for the observed results. The two signals could work separately, by triggering parallel stimulatory processes; alternatively, one of them could play a facilitatory role with respect to the other (see also Section 11,B). Evidence suggestive of an even greater complexity of the intracellular control processes has begun to accumulate. Some of the results now reported (see. for example, Vara and Rozengurt, 1985; Cooke and Hallet, 1985; Hallam ef al.. 1985) cannot be readily accounted for by the mechanisms discussed so far (as well as by cGMP, a second messenger system now enjoying a wave of interest: Houslay, 1985). It is likely, therefore, that new messengers will be discovered or that new functions will be assigned to already known signaling pathways. B. Membrane Fusion-Fission in Exocytosis

The time between the application of the stimulus and the appearance of exocytosis varies in different cell systems depending also on the second messenger that is involved. At the neuromuscular junction and other synapses the delay between the opening of voltage-gated, presynaptic Ca?' channels and the evoked postsynaptic responses is only a fraction of a millisecond (Llinas ef a/., 1981). In that time the rate of quanta1 transmitter release can increase by more than 1000-fold (Torri-Tarelli et d., 1985). The time involved is probably too short to invoke an enzymecatalyzed reaction (Reichardt and Kelly, 1983). The nature of the nonenzymatic CaZ+effect is still undefined. If negative fixed charges were screened in at strategic sites of the cytosolic surface of vesicle or plasma membranes, this arrangement might act to reduce the barrier of electrostatic energy that exists at the membrane and thus facilitate membrane fusion (Blioch et al., 1968). To overcome the even more significant hydration energy barrier (Bass and Moore, 1966; Silinski, 1983, Ca2+could cause a conformational change(s) in one or more membrane proteins. By analogy with the better known instance of membrane fusion-fission mediated by the influenza virus hemagglutinin, the conformational change responsible for exocytosis may involve the exposure of a hydrophobic domain at the surface of fusogenic protein(s) (reviews: Reichardt and Kelly, 1983; Ohnishi, Chapter 9, this volume). A role for a Ca*+-binding protein(s) appears quite likely (Hong ef d.,

152

JACOPO MELDOLESI AND BRUNO CECCARELLI

1982a,b; Geisow and Walker, 1986; Meers et al., 1987), but so far no fusogenic proteins have been identified with certainty. A calmodulin-like protein has been shown to be an integral component of brain synaptic vesicles (Hooper and Kelly, 1984). Indeed, anticalmodulin drugs and antibodies were reported to block exocytosis (Baker and Knight, 1981; Kenigsberg et al., 1983; Kenigsberg and Trifaro, 1985). However, in view of the poor specificity of these drugs (which inhibit a variety of other enzymes as well, including protein kinase C) and of the central role played by calmodulin in functions other than exocytosis, these observations appear still inconclusive. Another Caz+-binding protein, synexin, studied in chromaffin cells, might be involved more in the attachment to than in fusion of membranes (Pollard et al., 1981). Of potentially greater interest is the case of two other proteins. One of these, named chromobindin 9, identified in chromaffin cells by its ability to bind to granule membranes in a Ca2+-dependent fashion, has been found to be a substrate for protein kinase C both in uitro and in intact cells (Michener ef al., 1986). The second, synapsin 1, a phosphoprotein phosphorylated by both Caz+-calmodulin and CAMP-dependent protein kinases, appears to be involved in the regulation of neurotransmitter release at a step(s) preceding exocytosis. An example is the movement of vesicles that contain neurotransmitters from the deep cytoplasm of the nerve terminal to regions near the specific sites of fusion on the presynaptic membrane (Navone et al., 1984; Llinas et al., 1985). Finally, recent indirect evidence obtained in permeabilized chromaffin and insulinoma cells, and with granulocytes, suggests the involvement in exocytotic fusion-fission of a GTP-binding protein that might be localized in either one of the interacting membranes (Knight and Baker, 1985a,b; Barrowman et al., 1986; Vallar e l al., 1987). To account for membrane fission in exocytosis it has been proposed that the close apposition (or fusion) of granule and plasma membranes triggers an influx of small solutes. This in turn leads to swelling of the granule and to fusion of the apposed membranes (for recent reviews, see Finkelstein et al., 1986; Brocklehurst and Pollard, Chapter 7, this volume). This model is supported by results with artificial membranes, and by some indirect evidence in intact cells and partially reconstituted cell-free systems. Other results, however, appear incompatible with this osmotic model of exocytosis (see Knight and Baker, 1985b; Holz, 1986; Brocklehurst and Pollard, Chapter 7, this volume). The rates of exocytosis induced by agents that are believed to work through the activation of protein kinases (CAMP-dependent protein kinase; protein kinase C) are much slower than those induced by Ca2+. Plausible hypotheses are that phosphorylation of proteins by these enzymes increases the probability of fusion, possibly by increasing affinity

5. EXOCYTOSIS AND MEMBRANE RECYCLING

153

for Ca2+ (Knight and Baker, 1983; Baker and Knight, 1986). Alternatively, a smaller number of calcium ions may be needed to activate the process (Haimann et ul., 1987). These models readily explain the stimulation of exocytosis without there occurring a rise in [Ca2+]i.They also explain the synergism of the responses induced by the application of a protein kinase C activator together with a [Ca2+]iraising agent (see Section 11,A). Some experimental results, however, are difficult to reconcile with this unifying hypothesis. Among these are the stimulation of exocytosis by either protein kinase C activators or GTP analogs at very low [Ca2+Ii, well below the Kd for Ca2+of any known Ca2+-regulatedproteins (Di Virgilio et ul., 1984; Vallar et al., 1987);the greater efficacy of CAMP with respect to [Ca2+Iirises in the parotid acinar cells (Selinger and Schramm, 1985); and the dissociation between the stimulatory effect of protein kinase C activators and the inhibition brought about by Ca2+in parathyroid cells (Shobak et ul., 1984: Brown rt ul., 1984: Nemeth and Scarpa, 1986). As far as the substrates of protein kinases go, increased phosphorylation has indeed been observed following stimulation in a variety of cell systems (reviews: De Lisle and Williams, 1986; Baker and Knight, 1986; Meldolesi et a ( . , 1987). In n o case, however, has there been a convincing demonstration of a direct involvement of phosphorylation in the modulation of exocytosis. C. Toxins

The use of natural toxins, with their ability to interact with specific molecules of physiological importance, has led to advances in a variety of biological fields. Many toxins exist that in one system or another, most frequently in neurons, modulate exocytosis. The mechanism of their actions differs widely. Most do not act on exocytosis per se, but modify the generation of the intracellular signals that control exocytosis. A general discussion of these toxins is beyond the scope of this chapter: the interested reader is referred to reviews by Ceccarelli and Clementi (19759, Howard and Gundersen (1980~Hucho and Ovchinnichov (1983), and Harris ( 1986). 1 . CLOSTRIDIUM TOXINS: INHIBITORS OF EXOCYTOSIS

Neurological syndromes as diverse as tetanus and the paralysis due to botulinum toxin are probably the result of the blockade of exocytosis in different synapses (Sellin, 1985; Simpson, 1986). Botulinum and tetanus toxins are two similar proteins that consist of two chains joined by a

154

JACOPO MELDOLESI AND BRUNO CECCARELLI

disulfide bond, 100,000 and 50,000 in molecular weight. Both toxins exert their effect after a delay of at least 30 min, the period required for binding and transmembrane translocation. CIostridium toxins bind with high affinity to various gangliosides (GDlb,GTlb, GDla,and possibly others) and to a discrete number of specific surface receptors. As discussed by Montecucco (1986), the gangliosides may act to concentrate the toxin molecules at the external surface of the target membranes, thereby increasing their affinity for the specific protein receptors. Recent evidence (Dolly et al., 1984; Black and Dolly, 1986a,b)indicates that, at the neuromuscluar junction, the botulinum toxin receptor complex is redistributed from the surface to an intraterminal, endosomal compartment. Thus, the uptake of the toxin proceeds by receptor-mediated endocytosis. The low pH of the endosomal compartment could promote conformational rearrangement of the heavier toxin chain, with generation of a pore in the endosomal membrane, through which the lighter chain is translocated (Black and Dolly, 1986a,b; Hoch et al., 1985; Simpson, 1986). Interestingly, many synapses of the central nervous system bind botulinum toxin with high affinity but lack the ability to translocate the lighter subunit efficiently. These synapses are affected only by high concentrations of the toxin (Black and Dolly, 1986b). Similarly, bovine chromaffin cells become blocked when cultured for days in the presence of high concentrations of type D botulinum toxin, with no concomitant inhibition of Na+ and Ca*+ fluxes (Knight et al., 1985). Such inhibition was seen also when toxin-poisoned cells were permeabilized and exposed to a Ca2+concentration as high as 10 p M . Additional, even stronger, evidence is provided by recent results which show that both tetanus and botulinum type A toxins, when directly microinjected into chromaffin cells, inhibit exocytosis (Penner el al., 1986). It can thus be concluded that many (possibly all) synapses and neurosecretory cells represent potential targets of tetanus and botulinum toxins, but that both the expression of the specific receptors, and the presence of effective transmembrane transport processes, represent the determinants of the toxin target specificity. The mechanism whereby Clostridium toxins, once translocated to the cytosol, inhibit exocytosis remains undefined. Several hypotheses have been put forth. By analogy with the other toxins (e.g., diphtheria, cholera, pertussis toxins) that share similar binding-internalization processes, tetanus and botulinum toxin lighter chains may possess enzyme activity (ADP-ribosylase) or may affect the activity of enzymes in nerve terminals. Indeed, ADP-ribosylase activity has recently been demonstrated for a botulinum toxin (Aktories el af., 1986) that is devoid of neurotoxic activity. Alternative1y, the toxins could bind to, and inactivate, a strategic site(s) important for exocytosis (Simpson, 1986; Sellin, 1985). In any event, the block of exocytosis by Clostridium toxins appears to occur at a

155

5. EXOCYTOSIS AND MEMBRANE RECYCLING

step or steps distal to the opening of Ca?+channels that is induced by an action potential and occurs in toxin-poisoned terminals. Increasing the [Ca”], signal (e.g., by the use of 4-aminopyridine) results in a partial relief of the block (Sellin, 1985: Simpson, 1986). In addition, the stimulation of exocytosis brought about by a-latrotoxin is almost unaffected by pretreatment with Closrridium toxins (Cull-Candy et al., 1976). This aspect is further discussed below. 2. LATR TROT OX IN

AND

CONGENERS: S T I M U L A T O R S

OF

ExocYTosls

a-Latrotoxin is a high molecular weight ( 1 30,000) protein contained in the venom of the black widow spider. At the frog neuromuscularjunction, the action of this toxin consists of a massive stimulation of asynchronous quanta1 acetylcholine release, i.e., synaptic vesicle exocytosis, leading to complete vesicle depletion in the nerve terminal within 20-30 min. (reviews: Hurlbut and Ceccarelli, 1979; Meldolesi et a l . , 1986). The depleting effect is due to blockade of endocytosis, also induced by the toxin (see Section i11,B). Subsequent studies revealed that sensitivity to a-latrotoxin is a property not only of the motor end plate but of all synaptic terminals of vertebrates. Among neurosecretory cells, those of the lines PC12 and AtT20 are sensitive, whereas rat chromaffin cells are unaffected by the toxin in the intact animal and become responsive when cultured in the presence of nerve growth factor (for review, see Meldolesi (’1 a / . , 1986). The mechanism of a-latrotoxin action has been unraveled in part. The initial step involves the binding of the toxin to a specific, high-affinity receptor (a high molecular weight, integral membrane protein) that can now be considered as a surface marker of nerve terminals (Valtorta iJt al., 1984; Scheer and Meldolesi, 1985). The receptor is also expressed in the small number of other targets of the toxin. a-Latrotoxin is not internalized, but remains bound to its receptor at the external surface of the presynaptic membrane (Valtorta et al., 1984; Meldolesi et a / . , 1986). In PC12 cells the binding interaction activates a small (- I5 psec) nonclosing cation channel (Wanke et al., 1986). and this entails depolarization, increased Ca2+influx, and persistent elevation of the [Ca?’], (Grass0 et ul., 1980; Meldolesi et ul., 1984).These phenomena have been recently reproduced also in a reconstituted system composed of liposomes bearing the purified receptor and exposed to a-latrotoxin (Scheer et al., 1986). The nature of the channel (whether part of the receptor, or a hydrophobic domain of the toxin molecule after the latter has been inserted across the membrane) remains to be clarified (Meldolesi et al., 1986). The activation of the cation channel represents only part of the alatrotoxin action. In fact, the toxin can stimulate release of neurotransmit-

156

JACOPO MELDOLESI AND BRUNO CECCARELLI

ters even when applied in a Ca2'-free (but Mg2+-containing)medium, an effect that is particularly striking at the frog neuromuscular junction (Hurlbut and Ceccarelli, 1979). In this system, the toxin-induced responses are massive, whether or not Ca2+ is present. However, differences exist in the pattern and time course of the release. Specifically, with Ca2+the pattern of the release is bursting, and the reponse is more prolonged (Fesce et d.,1986a). The possibility that the Ca2+-freeeffect of alatrotoxin is only apparent, due to redistribution of CaZ+from the stores to the cytosol, has not been substantiated by direct experiments with the fluorescent [Ca2+]iindicator, quin2 (Meldolesi et ul., 1984). Exocytosis thus is stimulated by a-latrotoxin through the activation of at least two parallel mechanisms: one dependent, the other independent of [Ca*+]i. The activation of this second as yet undefined mechanism may account for the unique ability of a-latrotoxin to surmount the blocking effects of Clostridium toxins at both synapses and PC12 cells (see Hurlbut and Ceccarelli, 1979; Figliomeni and Grasso, 1985). 111.

ENDOCYTOSIS

This section will focus on the form of endocytosis that follows exocytosis in secretory cells (triggered endocytosis, as defined by Baker and Knight, 1986). This process differs from both adsorptive (or receptormediated) and fluid-phase endocytosis, because it appears to consist of the recycling to the cytoplasm of the same membranes inserted in the cell surface during exocytosis. Also, the regulation and the intracellular route of the vesicles originated by exocytosis-coupled endocytosis appear different from those of the other forms of endocytosis. In glandular cells, the vesicles that recycle granule (or vesicle) membranes are preferentially directed to the Golgi complex, whereas the vesicles of the other forms of endocytosis are first targeted to the endosome and lysosome compartments (Herzog, 1981; Farquhar, 1982, 1985). A. Membrane Sorting in Endocytosis

The specific composition of the recycled membrane with respect to the plasma membrane raises the problem of the sorting mechanism(s). Essentially two models can be envisaged. In the first, the components of the two membranes, granule-vesicle and plasma membranes, remain segregated, at least for some time after exocytosis. Freeze-fracture and immunocytochemical studies of De Camilli et ul. (1976) and Patzak and Winkler (1986)

5. EXOCYTOSIS AND MEMBRANE RECYCLING

157

have documented that, in stimulated cells, components of the granule/ vesicle membrane remain segregated in patches at the cell surface. Rapid endocytosis restricted to these patches could easily preserve the molecular identity of the two participating membranes. I n the second model a certain degree of intermixing occurs, but it is corrected for by a process of “molecular filtration” by the retrieval organelles, i.e., the coated pits. Such a process would consist in the exclusion from the pits of the components specific for the plasma membrane, with concomitant accumulation of those specific for the granule/vesicle membrane (Bretscher c’t a l . , 1980). Studies of the frog neuromuscular junction suggest that these two models of endocytosis can coexist under certain conditions (Meldolesi and Ceccarelli, 1981; Torri-Tarelli r t a / . , 1987). As long as secretion evoked by electrical stimulation is maintained at rates up to 400 quanta/ sec, the incorporation of vesicles in the presynaptic membrane and the ensuing recycling are kept in balance, i.e., so significant decrease in vesicle number or enlargement of the presynaptic surface area takes place at the nerve terminal. In these terminals, the number of coated pits or vesicles is only slightly increased (Ceccarelli et ul., 1973; Ceccarelli and Hurlbut, 1975, 1980a; Torri-Tarelli r f d., 1987). Thus, when the rate of secretion does not lead to nerve exhaustion, the residence time of the vesicle membrane at the terminal surface is very short, possibly because each vesicle undergoes a quick exo-endocytotic cycle without even flattening down in the presynaptic membrane. As discussed in detail elsewhere (Ceccarelli and Hurlbut, 1980a).endocytosis without previous vesicle flattening would be energetically favorable because it does not require work for membrane invagination (for alternative interpretations of secretion at the frog neuromuscular junction, see Heuser and Reese, 1973; Miller and Heuser, 1984). Only when the secretion rate is high, or when treatments are used that impair recycling (see Section III,B), does the presynaptic surface area increase (Fig. 5 ) . Under these conditions the number of coated pits and vesicles increases, but only in proportion to the increase of the terminal surface area, i.e., the concentration of coated structures per area of the terminal plasmalemma remains unchanged (Torri-Tarelli et ul., 1987).The lack of correlation between the number of quanta secreted and the number of coated structures accumulated suggests that, at the neuromuscular junction, the sorting mechanism might be based primarily on the rapidity of the exo-endocytotic cycle. The membrane filtering ability of coated pits (Bretscher et a / . , 1980) would become important only when the exocytosis rate exceeds the potential of quick recycling of the terminal. Thus, the coated pit mechanism might serve more as a device to return to equilib-

158

JACOPO MELDOLESI AND BRUNO CECCARELLI

FIG.5 . Electron micrograph of a frog neuromuscular junction taken in a preparation exposed for 2 hr to a-latrotoxin at about 1°C in Ringer’s solution modified by the addition of 4 m M Mg2+.Notice the almost complete depletion of synaptic vesicles, the development of infoldings of the axolemma, and the presence of numerous coated vesicles and coated pits (arrowheads). X27.000. (Original from B. Ceccarelli, W.P. Hurlbut, and N. Iezzi; see also Ceccarelli et a / . , 1988.)

rium a system outstripped by exhaustive stimulation than as a rapid recycling mechanism. An alternative interpretation of the results by TorriTarelli ef al. (1987) would assign a central role in vesicle recycling to coated structures. This would be possible only if it is assumed that the rate constant of formation and/or the lifetime of these structures has changed following stimulation. Kinetics of this kind would certainly be unusual. Experimental findings in bovine chromaffin cells have led to the suggestion that fast and slow mechanisms of endocytosis can coexist. Membrane

5. EXOCYTOSIS AND MEMBRANE RECYCLING

159

FIG.6. Exo-endocytotic coupling in mouse pancreatic acinar cells. The freeze-fracture picture to the left shows two exocytoses, one caught at an initial stage, with a very small tubular connection between the granule content and the acinar lumen (arrow), the other showing the classic, R-shaped appearance (asterisk). Notice the vesicle in the process of budding to the cytoplasm from the latter exocytosis image (arrowhead). I n the thin section (right), numerous coated pits and vesicles are seen recycling from the acinar surface at infoldings corresponding to previously discharge zymogen granules (asterisks). Notice in particular the five coated structures aligned in row around the infolding toward the top of the figure. AL, Acinar lumen. x 19,430 and 21,440. (From H . Koike and J . Meldolesi, unpublished results.)

capacitance measurements by Neher and Marty (1982) have shown that endocytosis can follow exocytosis within seconds. However, if the cells are stimulated strongly, the residence time of the specific granule components at the cell surface appears to be much longer, with a fl/? of the order of 10 min (Patzak et a / . , 1984). In that condition membrane recycling seems to occur by means of coated vesicles. a type of recycling that predominates in exocrine gland cells (Fig. 6) and in mast cells (Meldolesi and Ceccarelli, 1981; Herzog, 1981; Thilo, 1985). However, in the latter case quick membrane recapture may also occur (Fernandez et d., 1985). Although it is obvious that our knowledge of the molecular mechanisms that underlie endocytosis is quite limited, the suggestion that, depending on the nature and intensity of the stimuli, exocytosis can involve different rates or mechanisms should be treated with caution. This is particularly true because in some experiments cells were stimulated to exhaustion. B. Regulation of Endocytosis

Although recycling of the granulehesicle membrane from the cell surface can occur at different rates in some secretory systems, this does not mean that a given type of endocytosis is regulated independently of second messenger mechanism(s). At the neuromuscular junction, and at other synapses and neurosecretory cells, exocytosis and endocytosis are

160

JACOPO MELDOLESI AND BRUNO CECCARELLI

coupled fairly tightly until the secretion rate becomes high, when coupling becomes loose. When that occurs the V,,, of exocytosis can exceed that of (rapid) endocytosis. The enlargement of the surface area that follows the uncoupling of the two processes may constitute the trigger of the slow, coated vesicle-mediated recycling (Ceccarelli and Hurlbut, 1980a; Meldolesi and Ceccarelli, 1981). Up to now, the pharmacology of endocytosis has attracted only limited attention. Treatment with some drugs has led to depletion of synaptic vesicles at the neuromuscular junction. In some cases depletion was achieved only after the original vesicle pool had turned over several times (Segal et al., 1985). Under these conditions vesicle depletion may have resulted from the uncoupling of exo- and endocytosis. However, with two consecutive treatments, vesicles are rapidly depleted, the nerve terminals swell, and the number of quanta secreted equals the number present in the terminal at rest; in other words, endocytosis is blocked (Haimann et ul., 1985; Fesce et ul., 1986a; Ceccarelli et al., 1988; Fig. 5). Conceivably the treatment-induced increase in intracellular Na+ that can be expected to occur may be responsible for the observed blockade. Studies on frog neuromuscular junctions exposed to low concentrations of a-latrotoxin (Ceccarelli and Hurlbut, 1980b) and parotid acinar cells stimulated with isoproterenol (Koike and Meldolesi, 1981) have shown that endocytosis is impaired when Ca2+is omitted from the incubation medium. However, in other conditions, e.g., when neuromuscular junctions were treated with La3+,the presence of Ca2+was much less important than with a-latrotoxin (Segal et al., 1985; Fesce et al., 1986a). Also, accurate morphometric studies by von Grafstein et al., (1987), carried out in permeabilized chromaffin cells, failed to reveal that [Ca2+]iregulates the rate of membrane recycling which occurs after stimulation of exocytosis. The role of Ca2+remains uncertain and may not have general significance. IV. CONCLUSION

Secretion of large hydrophilic molecules is an almost universal property of eukaryotic cells. Exocytosis, the process developed by cells to carry out this essential function, has some distinct advantages over simple transmembrane transport processes. Three in particular can be listed: (1) exocytosis permits cells to store large quantities of secretory products and to release them as required without the need to expend large amounts of energy; (2) release can take place at specific, strategically located portions of the plasmalemma (polarized release), so that high concentrations of secretory products can be achieved in the extracellular space immedi-

5. EXOCYTOSIS AND MEMBRANE RECYCLING

161

ately adjacent to the release sites: and ( 3 ) exocytosis permits rapid and efficient regulation of release. It should be emphasized, however, that while exocytosis follows the same general scheme in all cells, it is clearly not identical in the various secretory systems. In other words. the specific mechanism of exocytosis needs to be adapted to functional needs. Thus, the principal features of exocytosis in a given secretory system can be looked at as variations (often of great bilogical importance) around a common theme. One feature that exists in many secretory cells and may therefore constitute a general characteristic is the tight coupling to endocytosis. This coupling enables cells to maintain their population of storage organelles and to conserve the size and identity of their plasma membrane. This chapter has attempted to summarize the many important advances in our knowledge of exo- and endocytosis that have occurred in the field. However, perhaps because of its complexity, involving not only the granule and plasma membranes but also second messengers, their intracellular targets, and various other cytosolic components, exocytosis has so far escaped an appropriate molecular description. In particular, the attempts that have been made to reconstruct the process in uitro, starting from completely disassembled components, have not been successful (see De Lisle and Williams, 1986). This failure has prevented a direct investigation of various aspect5 of exocytosis, such as the molecular mechanisms that control fusion. Almost 30 years after the discovery of exocytosis, our understanding of the process remains therefore incomplete, and much further work appears necessary in order to clarify questions that are still obscure or completely mysterious. ACKNOWLEDGMENTS This review was prepared while the authors were recipients of grants from the CNR Strategic Project Molecular Biology (J.M.) and the Muscular Dystrophy Association of America (B.C.) REFERENCES Ahnert-Hilger, G., Bhakdi, S . , and Gratzl, M. ( 1985). Minimal requirements for exocytosis. A study using PCI2 cells permeabilized with staphylococcal toxin. J . Biol. Clierri. 260, 12730-12734. Aktories, K . . Barmann, M., Ohishi, I.. Tsuyama, S . . Jakobs, K . H., and Habermann. E. (1986). Botulinum C2 toxin ADP-ribosylates actin. Nurrtre (London) 322, 390-392. Baker, P. F. (1984). Multiple controls for secretion? Nuture (London) 310, 619-620. Baker. P. F.. and Knight, D. E. (1981). Calcium control of exocytosis and endocytosis in bovine adrenal medullary cells. Philos. Tron.v. R . Soc. London. Ser. B 296, 83-104. Baker. P. F.. and Knight, D. E. (1986). Exocytosih. control by calcium and other factors. B r . M e d . Bull. 42, 399-404. Barrowman, M. M . , Cockcroft. S . . and Gomperts, B. D. (1986). Two roles for guanine

162

JACOPO MELDOLESI AND BRUNO CECCARELLI

nucleotides in the stimulus-secretion sequence of neutrophils. Nature (London) 319, 504-507. Bass, L., and Moore, W. J. (1966). Electrokinetic mechanism of miniature postsynaptic potentials. Proc. Natl. Acad. Sci. U . S . A . 55, 1214-1217. Berridge, M . J., and Irvine, R . F. (1984). Inositol trisphosphate, a novel second messenger in cellular signal transduction. Nature (London) 312, 315-320. Black, J. D., and Dolly, J. 0. (1986a). Interaction of Wlabeled botulinum neurotoxins with nerve terminals. I . Ultrastructural autoradiographic localization and quantitation of distinct membrane acceptors for type A and B on motor nerves. J . Cell Eiol. 103, 521534. Black, J . D., and Dolly, J. 0. (1986b). Interaction of 1251-labeledbotulinum neurotoxin with nerve terminals. 11. Autoradiographic evidence for its uptake into motor nerves by acceptor-mediated endocytosis. J . Cell B i d . 103, 535-544. Blioch, Z. L., Glagoleva, 1. M . , Liberman, E. A , , and Nenashev, V. A . (1968). A study of the mechanism of quanta1 transmitter release at a chemical synapse. J . Physiol. (London) 199,, 11-35. Bretscher, M. S., Thompson, J . N . , and Pierce, B. M. F. (1980). Coated pits act as molecular filters. Proc. Natl. Acad. Sci. U . S . A . 77, 4156-4159. Brown, E. M . , Redgrave, J., and Thatcher, J. (1984). Effect of phorbol ester TPA on PTH secretion. Evidence for a role for protein kinase C in the control of PTH release. FEBS Lett. 175, 72-75. Ceccarelli, B., and Clementi, F. (1979). “Neurotoxins, Tools in Neurobiology.” Raven, New York. Ceccarelli, B., and Hurlbut, W. P. (1975). The effects of prolonged repetitive stimulation in hemicholinium on the frog neuromuscular junction. J . Physiol. (London) 247, 163-188. Ceccarelli, B., and Hurlbut, W. P. (1980a). Vesicle hypothesis of the release of quanta of acetylcholine. Physiol. Rev. 60, 396-441. Ceccarelli, B . , and Hurlbut, W. P. (1980b). Ca2+-dependentrecycling of synaptic vesicles at the frog neuromuscular junction. J . Ce11 Biol. 87, 297-303. Ceccarelli, B., Hurlbut. W. P., and Iezzi, N. (1988). Effect of a-latrotoxin on the frog neuromuscular junction at low temperature. J . Physiol. (London), in press. Ceccarelli, B., Hurlbut, W. P., and Mauro. A. (1973). Turnover of transmitter and synaptic vesicles at the frog neuromuscular junction. J . Cell B i d . 57, 499-524. Connor. J . A. (1986). Digital imaging of free Ca?’ changes and of spatial gradients in growing processes in single mammalian central nervous system cells. Proc. Natl. Acad. Sci. U . S . A . 83, 6279-6183. Cooke. E . , and Hallett, M. B. (1985). The role of C-kinase in the physiological activation of the neurtrophil oxidase. Biochem. J . 232, 323-327. Cull-Candy, S. G . , and Lundh, H., and Thesleff, S. (1976). Effects of botulinum toxin on neuromuscular transmission in the rat. J . Physiol. (London) 260, 177-203. De Camilli, P., Peluchetti, D., and Meldolesi, J. (1976). Dynamic changes of the lumenal plasmalemma in stimulated parotid acinar cells. J . Cell Biol. 70, 59-74. De Camilli, P., Harris, S . M., Jr., Huttner, W. B., and Greengard, P. (1983). Synapsin I (Protein I), a nerve terminal-specific phosphoprotein. 11. Its specific association with synaptic vesicles demonstrated by immunocytochemistry in agarose-embedded synaptosomes. J . Cell B i d . 96, 1355-1373. De Lisle, R. C., and Williams, J . A. (1986). Regulation of membrane fusion in secretory exocytosis. Annu. Rev. Physiol. 48, 225-238. Dewald, B., Bretz, U . , and Baggiolini, M. (1982). Release of gelatinase from a novel secretory compartment of human neurophils. J . Clin. Invest. 70, 518-525. Di Virgilio, F., Lew, D. P., and Pozzan, T. (1984). Protein kinase C. activation of physiolog-

5. EXOCYTOSIS AND MEMBRANE RECYCLING

163

ical processes in human neutrophils at vanishingly small cytosolic Ca?* levels. Nurrtre (London) 310, 619-693. Di Virgilio, F., Milani, D., Leon, A. Meldolesi. J . , and Pozzan, T. (1987).voltage dependent activation and inactivation of calcium channels in PC12 cells. Correlation with neurotransmitter release. J . B i d . Chem. 262, 9188-9195. Dolly, J. O., Haliwell, J . V., Black, J. D., Williams, R. S.. Pelchen-Mattews, A., Breeze. A. L., Mehraban, F.. Othman, I . B., and Black, A. R. (1984). Botulinum neurotoxin and dendrotoxin as probes for studies o n transmitter release. J . P/zv.riol. (Puri.?) 79, 280303. Douglas, W. W. (1974). Exocytosis and the exocytosis-vesiculation sequence: With special reference to neurohypophy chromaffin and mast cells, calcium and calcium ionophores. In “Secretory Mechanisms in Exocrine Glands” N . A. Thorn, and 0. H. Petersen, (eds.), pp. 116-136. Munskgaard. Copenhagen. Farquhar, M. G. (1982). Membrane recycling in secretory cells: pathways to the Golgi complex. In “Membrane Recycling” (D. Evered and G . M. Collins, eds.), pp. 157-183. Pitman, London (CIEA Found Symp 92). Farquhar, M. G. (1985). Progress in unraveling pathways of Golgi traffic. Annrr. Reu. Cell Biol. 1, 447-488. Fernandez, J . M., Neher. E., and Gompertb. B. D. (1985). Capacitance measurements reveal stepwise fusion events in degranulating mast cells. Nuture (London) 312, 453455. Fesce. R., Segal, J. R.. Ceccarelli, B., and Hurlbut, W. P. (1986a). Effects of black widow spider venom and Ca?’ on quanta1 secretion at the frog neuromuscularjunction. J . Gcw. Physiol. 88, 59-8 1 . Fesce. R.. Segal. J . R., and Hurlbut, W. P. (1986b). Fluctuation analysis of non-ideal shot noise. Application to the neuromuscular junction. J . C e n . Phvsiol. 88, 25-57. Figliomeni, B., and Grasso, A. (1985). Tetanus toxin affects the K+-stimulated release of catecholamines from nerve growth factor-treated PC12 cells. Biochcm. Biophys. R e s . Cotnmun. 128, 249-256. Finkelstein. A., Zimmerberg, J., and Cohen. F. S. (1986). Osmotic swelling of vesicles. Annrr. R e v . Physiol. 48, 163-174. Fumagalli, G., and Zanini, A. (1985). In cow anterior pituitary growth hormone and prolactin can be packed in separate granules of the same cell. J . Cell Biol. 10, 20192024. Gardner. J. D.. and Jensen, R . T. (1986). Receptors and cell activation associated with pancreatic enzyme secretion. Annrc. Reu. Phvsiol. 48, 103-1 17. Geisow, M. J., and Fisher, D. (1986). Molecular mechanisms of membrane fusion. Biochem. Soc. Trans. 14, 241-242. Geisow. M. J., and Walker, J. H. (1986). New proteins involved in cell regulation by Ca?’ and phospholipids. Trends Biochem. Sci. 11, 420-423. Ginsborg, B. L., and Jenkinson, D. H. (1976). Transmission of impulses from nerve to muscle. In “Neuromuscular Junction” ( E . Zaimis, ed.), pp. 229-364. Springer-Verlag, Berlin and New York. Gorio, A,, Hurlbut, W. P., and Ceccarelli, B . (1978). Acetylcholine compartments in mouse diaphragm: A comparison of the effects of black widow spider venom, electrical stimulation and high concentrations of potassium. J . Cell B i d . 78, 716-733. Grasso. A , , Alema, S . , Rufini, S.. and Senni, M. I . (1980). Black widow spider toxininduced calcium fluxes and transmitter release in a neurosecretory cell line. Nuture (London) 283, 774-776. Gunther, G . R. (1981). Effect of 12-0-tetradecanoyl phorbol-13-acetate on Ca?‘ efflux and protein discharge in pancreatic acini. J . Biol. C h e m . 256, 12040-12045.

164

JACOPO MELDOLESI AND BRUNO CECCARELLI

Grynkyevicz. G., Poenie, M., and Tsien, R. Y . (1985). A new generation of Ca?’ indicators with greatly improved fluorescence properties. J . Biol. C h e m . 260, 3440-3450. Haimann, C . , Torri-Tarelli. F., Fesce, R., and Ceccarelli, B . (1985). Measurement of quantal secretion induced by ouabain and its correlation with depletion of synaptic vesicles. J . Cell B i d . 101, 1953-1965. Haimann, C . , Meldolesi. J.. and Ceccarelli, B. (1987). The phorbol ester enhances the evoked quanta1 release of acetylcholine at frog neuromuscular junction. PJkgers Arch. 408, 27-31. Hallam, T. J., Daniel, J. L . , Kendrick-Jones, J., and Rink, T. J . (1985). Relationship between cytoplasmic free calcium and myosin light chain physphorylation in intact platelets. Biochem. J . 232, 373-377. Harris, T. B. (1986). Proc. World Congr. Anim, Plant. and Microb Toxins, 8th. Hashimoto. S . , Fumagalli, G., Zanini, A.. and Meldolesi, I. (1987). Sorting of growth hormone, prolactin and secretogranin 11 to separate granules in bovine pituitary cells. J . Cell B i d . 105, 1579-1586. Herzog, V. (1981). Endocytosis in secretory cells. Philos, Trrrns. R . Soc. London, Ser. B 269, 67-72. Heuser, J . E., and Reese, T. S. (1973). Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction. J . Cell B i d . 57,315-344. Hoch, T., Romero-Mira. M.. Ehrich, B. E., Finkelstein, A,, Das Gupta, B. R., and Simpson, L. L. (1985). Channels formed by botulinum, tetanus and diphtheria toxins in planar lipid bilayers. Relevance of translocation of proteins across membranes. Pruc. Null. A c a d . Sci. U . S . A . 82, 1682-1696. Holz, R. W. (1986). The role of osmotic forces in exocytosis from adrenal chromaffin cells. Annu. R e v . Physiol. 48, 175-190. Hong. K., DuzguneS. N.. Papahadjopoulos. D. (1982a). Modulation of membrane fusion by calcium binding proteins. Biophys. J . 37, 297-305. Hong. K., DuzgiineS, N . , Ekerdt, R., and Papahadjopoulos, D. (l982b). Synexin facilitates fusion of specific phospholipid vesicles at divalent cation concentrations found intracellularly. Pruc. Nutl. A c u d . Sci. U . S . A . 79, 4642-4644. Hooper. J. E., and Kelly, R. B . (1984). Calmodulin is tightly associated with synaptic vesicles independent of calcium. J . Biol. Chem. 259, 148-153. Houslay, M. D. (1985). Renaissance of cyclic GMP. Trends Biochem. Sci. 10, 465-466. Howard, B. D., and Gundersen, C. B. (1980). Effects and mechanisms of polypeptide neurotoxins that act presynaptically. Annu. Rev. Pharmac~ol.Tuxicol. 20, 307-366. Hucho, F., and Ovhinnichov, Y. A. (1983). “Toxins as Tools in Neurochemistry.” De Gruyter, Berlin. Hurlbut, W. P., and Ceccarelli, B. (1979). Use of black widow spider venom to study the release of neurotransmitters. I n “Neurotoxins, Tools in Neurobiology” (B. Ceccarelli. and F. Clementi, eds.), pp. 87-115. Raven, New York. Katz, B. (1966). “Nerve, Muscle, and Synapse.” McGraw-Hill, New York. Katz, B . , and Miledi, R. (1977). Transmitter leakage from motor nerve endings. Proc. R . Soc. London, Ser. B 196, 59-72. Keith, C. H . , Ratan, R., Maxfield. F. R., Bajer, A,, and Shelanski, M. L. (1985). Local cytoplasmic calcium gradients in living mitotic cells. Nature (London) 316, 848-850. Kelly, R. B . (1985). Pathways of protein secretion in eukaryotes. Science 230, 25-31. Kenigsberg, R. L., and Trifaro, J. M. (1985). Microinjection of calmodulin antibodies into cultured chromaffin cells blocks catecholamine release in response to stimulation. Neuroscience 14, 335-347. Kenigsberg, R. L., Cote, A., and Trifaro, J . M . (1982). Trifluoperazine. a calmodulin inhibitor, blocks secretion in cultured chromaffin cells at a step distal from calcium entry. Neuroscience 7, 2277-2286.

5. EXOCYTOSIS AND MEMBRANE RECYCLING

165

Knight. D. E.. and Baker. P. F. (1982). Calcium dependence of catecholamine release from bovine adrenal medullary cells after exposure to intense electric field. J . Mrmhr. Biol. 68, 107-140.

Knight. D. E., and Baker, P. F. (1983). The phorbol ester TPA increases the affinity of exocytosis for calcium in leaky adrenal medullary cells. FEBS Lett. 160, 98-100. Knight, D. E., and Baker, P. F. (198Sa). Guanine nucleotides and Ca:--dependent exocytosis. Studies on two adrenal cell preparations. FEBS Lett. 189, 345-349. Knight, D. E.. and Baker, P. F. (I985b). The chromaffin granule proton pump and calcium dependent exocytosis in bovine adrenal medullary cells. J . Memhr. B i d . 83, 147- 156. Knight. D. E.. and Scrutton, M. C . (1985). The relationship between intracellular second messengers and platelet secretion. E i o c ~ l r ~ n Soc. r. Truns. 12, 969-971. Knight, D. E.. Tonge. D. A., and Baker. P. F. (19x5). Inhibition of exocytosis in bovine adrenal medullary cells by botulinum toxin type D. Narcire (London) 317. 719-711. Koike, H., and Meldolesi. J. (1981 ). Post-stimulation retrieval of lurnenal surface membrane in parotid acinar cells is Ca?’-dependent. E s p . Cell Res. 134, 377-388. Kolesnick. R. N., and Geshenghorn, M. G. (1985). Direct evidence that burst but not subtained secretion of prolactin stimulated by TRH is dependent on elevation of cytoplasmic Ca?’. J . B i d . Chem. 260, 5217-5220. Lew. P. D.. Monod, A,, Waldovogel, F. A.. Dewald. B., Baggiolini, M.. and Pozzan, T. ( 1986). Quantitative analysis of rhe cytosolic free calcium dependency of exocytosis from three subcellular compartments in intact human neutrophils. J. Cell Biol. 102,, 2 197-2204. Llinas. R., Steinberg, Z.. and Walton, K. (1981). Relationship between calcium current and postsynaptic potential in squid giant 5ynapse. B i o p h y . ~J. . 33, 323-352. Llinas. R . . McGuiness. T. L. M . . Leonard. C . S.. Sugimori, M . . and Greengard. P. ( 198.5). lntraterminal injection of synapsin 1 or calcium/calmodulin-dependent protein kinase I I alters neurotransmitter release at the squid giant synapse. Pro(.. Nurl. Acud. Sci. U.S.A. 82, 3035-3039. Lundberg, J . M.. and Hokfelt. T. ( 1983). Coexistence of peptides and classical neurotransmitters. Trends Neiirosci. 6, 375-333. Meers. P.. Ernst, J . D., DiizguneS, N . , Hong, K., Fedor, J., Goldstein, I . M . , and Papahadjopoulos, D. (1987). Synexin-like proteins from human polymorphonuclear leukocytes. Identification and characterization of granule-aggregating and membrane-fusing activities. J . B ~ o ( Chem. . 262, 7850-7858. Meldolesi, J . , and Ceccarelli. B. (1981 ). Exocytosis and membrane recycling. Philos. Truns. R. Soc. London. S r r . B 296, 55-65. Meldolesi, J . , Huttner, W.B.. Tsien, R. Y..and Pozzan. T. (1984). Free cytoplasmic Ca” and neurotransmitter release: Studies on PCI2 cells and 5ynaptosomes exposed to alatrotoxin. Proc. N u t / . A w d . Sci. U.S.A. 81, 620-624. Meldolesi, J . , Scheer, H.. Madeddu. L.. and Wanke. E. (1986). On the mechanism of action of a-latrotoxin. the presynaptic stimulatory toxin of the black widow spider venom. Trends Phurmurol. Sci. 7, 151-155. Meldolesi. J., Pozzan. T.. and Ceccarelli, B. ( 1987). Exo-endocytosis. Mechanisms of drugs and toxin action. In “Handbook of Experimental Ca2+and Drug Action Pharmacology” (P. F. Baker, ed.), pp. 339-359. Springer-Verlag, Berlin. Michener, M . L.. Dawson, L. B., and Creutr. C. E. (1986). Phosphorylation o f a chrornaffin granule binding protein in stimulated chromaffin cells. J . B i d . Chem. 261, 6548-6555. Miller, T. M.. and Heuser, J . E. (1984). Endocytosis of synaptic vesicle membrane at the frog neurornuscular junction. J. C d Biol. 98, 685-698. Montecucco, C. (1986). How do tetanus and botulinum toxin bind to neuronal membranes? Trmds Eiochem. Sci. 11, 314-316.

166

JACOPO MELDOLESI AND BRUNO CECCARELLI

Navone, F.. Greengard, P., and De Camilli, P. (1984). Synapsin 1 in nerve terminals: Selective association with small synaptic vesicle. Science 226, 1209-121 I . Neher, E., and Marty, A. (1982). Discrete changes of cell membrane capacitance observed under conditions of enhanced secretion in bovine adrenal chromaffin cells. Proc. Nut/. Acud. Sci. U . S . A . 79, 6712-6716. Nemeth, E . F., and Scarpa, A. (1986). Cytosolic Ca2+ and the regulation of secretion by parathyroid cells. FEBS Lett. 203, 15-19. Nishizuka, Y. (1986). Studies and perspectives of protein kinase C. Science 233, 305-312. Palade, G. (1959). Functional changes in the structure of cell components. In “Subcellular Particles” (T. Hayashi, ed.), pp. 64-80. Ronald, New York. Palade, G. E. (1975). lntracellular aspects of the process of protein secretion. Science 189, 347-358. Pandol, S . J . , Schoeffield, M. S., Sachs, G., and Muallem, S. (1985). Role of free Ca*+in secretagogue stimulated amylase release from dispersed acini from guinea pig pancreas. J . B i d . Chem. 260, 10081-10086. Patzak, A., and Winkler, H. (1986). Exocytotic exposure and recycling of membrane antigens of chromaffin granules: Ultrastructural evaluation after immunolabeling. J . Cell Biol. 102, 5 10-5 15. Patzak, A,, Bock, B., Fischer-Colbrie, R., Schanchenstein, K., Schmidt, W., Lingg, G., and Winkler, H. (1984). Exocytotic exposure and retrieval of membrane antigens of chromaffin granules: Quantitative evaluation of immunofluorescenceon the surface of chromaffin cells. J. Cell Biol. 98, 1817-1824. Penner, R., Neher, E., and Dreyer, F. (1986). Intracellularly injected tetanus toxin inhibits exocytosis in bovine chromaffin cells. Nature (London) 324,76-78. Pollard, H . B., Pazoles, C. J., and Creutz, C. E. (1981). Mechanism of calcium action and release of vesicle-bound hormones during exocytosis. Recent Prog. Horm. Res. 37, 299-332. Pozzan, T., Gatti, G., Dozio, N., Vicentini, L. M., and Meldolesi, J. (1984). Ca2’-dependent and -independent release of neurotransmitters from PC12 cells. A role for protein kinase C activation? J . Cell Biol. 99, 628-638. Putney, J. W. (1986). Identification of cellular activation mechanisms associated with salivary secretion. Annu. Rev. Physiol. 48, 75-88. Reichardt, L. F., and Kelly, R. B. (1983). A molecular description of nerve terminal function. Annu. Rev. Biochem. 52, 871-926. Rink, T. J., Smith, S. W., and Tsien, R. Y. (1982). Cytoplasmic free Ca2+in human platelets: Ca2+ thresholds and Ca2+-independent activation for shape changes and secretion. FEBS L e f t . 148, 21-26. Rink, T. J., Sanchez, A., and Hallam, T. J. (1983). Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. Nature (London) 305, 317-319. Sakmann, B., and Neher, E. (1984). Patch clamp techniques for studying ionic channels in excitable membranes. Annu. Rev. Physiol. 46, 455-472. Scheer, H., and Meldolesi, J. (1985). Purification of the putative a-latrotoxin receptor from bovine synaptosomal membranes in an active binding form. EMBO J. 4, 323327. Scheer, H., Prestipino, H., and Meldolesi. J. (1986). Reconstitution of the purified a-latrotoxin receptor in liposomes and planar lipid membranes. Clues to the mechanism of toxin action. EMBO J. 5 , 2643-2648. Schramm, M . , and Selinger, Z. (1975). The functions of cyclic AMP and calcium as alternative second messengers in parotid gland and pancreas. J . Cyclic Nucleofide Res. 1, 181192.

5. EXOCYTOSIS AND MEMBRANE RECYCLING

167

Segal, J . R.. Ceccarelli, B., Fesce. R.. and Hurlbut, W. P. (1985). Miniature end plate potential frequency and amplitude determined by an extension of Campbell's theorem. Biophys. J. 47, 183-202. Sellin, L. C. (1985). The pharmacological mechanism of botulism. Trend.s Pharmucol. Sci. 6, 80-83. Shoback. 1). M., Thatcher. J., Leombruno. R., and Brown, E. M . (1984). Relationship between parathyroid hormone secretion and cytosolic calcium concentration in dispersed parathyroid cells. Proc. Ncrtl. Accid. Sci. U . S . A . 81, 3113-31 17. Silinski. E. M . (1985). The biophysical pharmacology of calcium-dependent acetylcholine secretion. Phurmuid. Rcw. 37, 81-132. Simpson, L. L. (1986). Molecular pharmacology of botulinuni toxin and tetanus toxin. Annu. Rev. Phnrtnarol. Toxicol. 26, 427-453. Smith. A. D., and Winkler. H . (1972). Fundamental mechanisms of release of catecholamines. Handh. Exp. Phartnoi~ol.33, 53X-617. Tauc. L. ( 1982). Non-vesicular release of neurotransmitter. Physiol. Rev. 62, 857-893. Thilo, L. (1985). Selective internalization of granule membrane after secretion in mast cells. Proi.. Nutl. Acod. Sci. U . S . A . 82, 1716-1720. Thureson-Klein, A. (1983). Exocytosis from large and small dense-cored vesicles i n noradrenergic nerve terminals. Nertro.sciencr 10. 245-259. Torri-Tarelli. F., Grohovaz, F., Fesce. R., and Ceccarelli, B. (1985). Temporal coincidence between synaptic vesicle fusion and quantal secretion of acetylcholine. J . Cell Eiol, 101, 1386- 1399.

Torri-Tarelli, F., Haimann. C., and Ceccarelli, B. (1987). Coated vesicles and pits during enhanced quantal release of acetylcholine at the neuromuscular junction. J . N(~rtrocyto/. 16, 205-214. Tsien, R. Y., Pozzan, T.. and Rink, T. J. (1982). Calcium homeostasis in intact lymphocytes: Cytoplasmic free calcium monitored by a new, intracellularly trapped fluorescent indicator. J. Cell Biol. 94, 325-334. Tsien, R . Y.,Pozzan, T., and Rink, T. J . (1984). Measuring and manipulating cytosolic Ca" with trapped indicators. Trends Biochem. Sci. 9, 263-266. Vallar, L.. Biden, T. J.. and Wollenheim, C. B. (1987). Guanine nucleotides induce Ca2+independent insulin secretion from permeabilized RlNmSF cells. J . Biol. Chrnr. 262, 5049-5056.

Valtorta, F., Madeddu. L., Meldolesi, J., and Ceccarelli, B. (1984). Specific localization of the a-latrotoxin receptor in the nerve terminal plasma membrane. J. Cell B i d . 9, 124132. Valtorta F.. Villa A,, Jahn R., De Camilli P.. Greengard P.. and Ceccarelli B. (1988). Localization of synapsin I at the frog neuromuscular junction. Nerrro.sc~ience24, 593603.

Vara. F., and Rozengurt, E. (1985). Stimulation of Na-/H+ antiport activity of epidermal growth factor and insulin occurs without activation of protein kinase C. Biochem. Biophys. RPS.Commun. 130, 646-653. Vilmart-Seuwen. J., Kersken, H . , Sturzl. R., and Plattner. H. (1986). ATP keeps exocytosis sites in a primed state, but is not required for membrane fusion: An analysis with Priramecium cells in viuo and in uitro. J . Cell B i d . 103, 1279-1288. von Grafstein. H.. Roverts, C. S., and Baker. P. F. (1987). The kinetics of the exocytosisendocytosis secretory cycle in bovine adrenal medullary cells. J . Cell Biol. (in press). Wanke, E., Ferroni, A . , Gattanini, F., and Meldolesi. J. (1986). a-latrotoxin of the black widow spider venom opens a small, nonclosing cation channel. Biochem. Biophys. Res. Commun. 134, 320-325. Whittaker, M. J . , and Baker, P. F. (1983). Calcium-dependent exocytosis in an it? virro

JACOPO MELDOLESI AND BRUNO CECCARELLI

168

plasma membrane preparation from sea urchin eggs and the effects of some inhibitors of cytoskeletal function. Proc. R . Soc. London. Ser. B 218, 397-413. Williams, D. A,, Fogarty, K . E., Tsien, R. Y . , and Fay, F. S. (1985). Calcium gradients in single smooth muscle cells revealed by the digital imaging microscope using Fura-2. Nature (London) 318, 558-561.

NOTEADDED I N PROOF.Since the submission of the present chapter (February 1987). many important developments have been reported in the field of exo-endocytosis; among these we would like to emphasize the following: ( I ) The demonstration by Breckenridge and Almers (1987a) that in mast cells exocytosis is initiated by the formation of an ion-sized connection (the “fusion pore”) between the granule interior and the extracellular space. Immediately afterwards an increase in the whole-cell capacitance developes over tens of microseconds. Its time course has been interpreted as reflecting the widening of the “fusion pore” (Zimmerberg et ol.. 1987; Breckenridge and Almers, 1987b). Capacitance flickers are observed before the increase in capacitance becomes permanent, suggesting that the ”fusion pore” opens and closes several times before the true exocytotic opening stabilizes and enlarges. These data imply the involvement of specific channel-proteins localized in both granules and plasma membrane. Their initial arrangement in the first step of the fusion process might be quite similar to the arrangement of the proteins in the gap junction monomer, the connexon. (2) The suggestion that in some systems the metabolic reaction directly controlling exocytosis might not be a phosphorylation, but the dephosphorylation of a specific substrate. Such an event could in turn be controlled by Ca2*and other processes, such as phosphorylation(s) by protein kinase C. Evidence in favor of this model has been reported in Paramecia (Stecher et a / . , 1987) and data consistent with this view have been also obtained in mast cells (Gomperts et a / . , 1988). Breckenridge, L. J.. and Almers, W. (1987a). Currents through the fusion pore that form during exocytosis of a secretory vesicle. Nature (London) 328, 814-817. Breckenridge, L. J., and Almers, W. (1987b). Final steps in exocytosis observed in a cell with giant secretory granules. Proc. N u t / . Acad. Sci. U . S . A . 84, 1945-1949. Gomperts, B., Cockcroft, S . , Howell, T. W., and Tatham, P. E. R. (1988). Intracellular Ca2+,GTP and ATP as effectors and modulators of exocytotic secretion from rat mast cells. In “Molecular Mechanisms in Secretion.” ( N . A. Thorn, M. Treiman, 0. H. Petersen, and J. H. Thysen. eds.), Benzon Symp. 25. Munskgaard, Copenhagen, in press. Stecher, B.. Hohne. B., Gras, U., Momayezi, M., Glas-Albrecht, R., and Plattner, H . (1987). Involvement of a 65 K Da phosphoprotein in the regulation of membrane fusion during exocytosis in Paramecium cells. FEBS L e u . 223, 25-32. Zimmerberg, J., Curran, M., Cohen, F. S., and Brodwick, M. (1987). Simultaneous electrical and optical measurements show that membrane fusion precedes secretory granule swelling during exocytosis of beige mouse mast cells. Proc. N u t / . A w d . Sci. U . S . A . 84, 1585-1589.

CURRENT TOPICS IN MEMBRANES A N D TRANSPORT. VOLUME 32

Chapter 6 Exocytosis and Endocytosis : Membrane Fusion Events Captured in Rapidly Frozen Cells DOUGLAS E. CHANDLER Depurtment of Zoology AriTonrr State University Tempe, Arizona 85287

Exocytosis A . Observations in Chemically Fixed Cells B. Observations in Rapidly Frozen Cells 11. Endocytosis 111. Concluding Remarks References 1.

1.

EXOCYTOSIS

The concept that membranes undergo fusion during bulk movement of material into and out of cells, by processes we now call endocytosis and exocytosis, was first presented by Palade (1959). His suggestion was based on the fact that secretory cells contain membrane-bound organelles that release their contents into the extracellular space during periods in which one can chemically assay release of secretions (Jamieson and Palade, 1967; Jamieson, 1972). Exocytosis requires that the secretory granule and plasma membranes come in contact and form a pore which joins the granule interior with the extracellular space. The basic mechanisms of exocytosis have been firmly established: ( 1 ) a transient increase in intracellular free calcium initiates exocytosis (Douglas, 1968, 1974; Rasmussen, 1981; Berridge and Irvine, 1984; Rubin et ul., 19851, (2) antibodies directed against secretory granule membrane proteins are found on the cell surface after but not before secretion, 169 Copyright Q 1988 by Academic Press. Inc All rightb of reproduction in any form rexrved

170

DOUGLAS E. CHANDLER

indicating that continuity has been established between these membranes (Von Wedel et al., 1981; Lingg et al., 1983; Stenberg et al., 1985; Patzak and Winkler, 19861, (3) cinematography at the light microscope level shows that granules approach the plasma membrane and release their contents into the extracellular space within 1-3 msec (Douglas, 1974; Holstein and Tardent, 1984), (4) electrophysiological measurements of membrane capacitance resolve discrete increases in cell surface area during exocytosis (Jaffe et al., 1978; Neher and Marty, 1982; Fernandez et al., 1984), and (5) electron microscopy has documented the steps that occur during fusion of secretory granule and plasma membranes. A. Observations in Chemically Fixed Cells

Until 1979, ultrastructural studies of exocytosis relied on chemical fixation to stop biological processes. Aldehyde fixation resulted in many new discoveries that form a foundation for our thoughts about the structural transitions in membranes during exocytosis. Palade and Bruns (1968), in their study of a vesicle shuttling system in capillary endothelial cells, outlined three basic steps in the exocytosis process. First, the secretory granule or vesicle approaches the plasma membrane, and granule and plasma membranes become closely apposed with no intervening cytoplasm (between arrows, Fig. 1). Based on the three-layered staining pattern of biological membranes, this area of apposition has been termed the “pentalaminar structure.” Second, the pentalaminar structure is transformed into a trilaminar structure: a single bilayer in continuity with both the plasma membrane and the granule membrane (between arrows, Fig. 2). Third, the trilaminar structure disintegrates to form an aqueous pore connecting the granule interior with the extracellular space (between arrows, Fig. 3). Trilaminar and pentalaminar structures are present in many secretory cells (Kim et af., 1972; Lagunoff, 1973; Heuser et af., 1974; Berger et al., 1975; Palade, 1975; Neutra and Schaeffer, 1977; Lawson et al., 1977; Pinto da Silva and Nogueira, 1977; Burwen and Satir, 1977; Tilney et af., 1979; Specian and Neutra, 1980) and in other cells undergoing membrane fusion (Kalderon and Gilula, 1979). Tandler and Poulson (1976) provide an excellent example in their study of the cat submandibular gland. Upon cholinergic stimulation, mucous droplets are seen fusing their membranes with the plasma membrane; both trilaminar structures (single arrow, Fig. 4) and pentalaminar structures (double arrows, Fig. 4) are present. Figure 5 shows that the pentalaminar structure results from close apposition of granule and plasma membranes over a comparatively large area. A number of reviews have outlined the three-step fusion process as seen in

6 . MEMBRANE FUSION DURING EXOCYTOSIS

171

chemically fixed cells (Palade, 1975: Orci et a / . , 1981: Plattner, 1981; Diizgunes, 1985). The freeze-fracture technique, with its panoramic and three-dimensional view of membrane structure, has made important contributions to what we know about exocytosis. Early studies on Tetrahymenu (Satir et al.. 1973), Paramecium (Plattner et al., 1973). and the frog neuromuscular junction (Heuser et al., 1974) have shown that secretory vesicles are docked at plasma membrane sites that are distinguished by intramembrane particle (IMP) arrays: a rosette of particles in the protozoa and a double ridge of particles in the neuromuscular junction. These IMPs are instrumental for exocytosis in protozoa (Beisson et al., 1976; Plattner et al., 1980; Lefort-tran et al., 1981). possibly representing stimulus-gated calcium channels (Satir and Oberg, 1978), or sites of Ca2+-dependent ATPase activity (Plattner et a / . , 1980). Organized particle arrays have not been observed in other secretory cells, however, indicating that this is not a generalizable feature of exocytosis. Freeze-fracture has revealed structures that appear to be analogs of the pentalaminar and trilaminar formations seen in thin section. P-face fractures of the plasma membrane visualize large areas that are cleared of IMPS and positioned over secretory granules exactly where one would expect to see pentalaminar structures in thin sections (Fig. 6). These IMPfree areas have been observed in numerous secretory cells (Chi et d., 1976; Friend et ul., 1977; Lawson et d.,1977; Orci et al., 1977; Amherdt et a / . , 1978; Peixoto de Menezes and Pinto da Silva, 1978; Theodosis et a / . , 1978; Zerban and Franke, 1978; Aunis et d., 1979; Tilney et al., 1979) but are not seen in every cell type (De Camilli et a / . , 1976; Tanaka et a / . , 1980). In addition, freeze-fracture illustrates the presence of single bilayer diaphragms, an equivalent of the trilaminar structure (Pinto da Silva and Nogueira, 1977; Pinto da Silva et al., 1980). As shown in Fig. 7 (arrows), these diaphragms are circular, recessed areas of membrane that are not etchable and have been cleared of IMPs; they appear to be continuous with both the plasma membrane and the granule membrane (arrow, Fig. 8). Often, a cluster of diaphragms separated by a network of cytoplasm-filled tunnels join the membrane of a single granule to the plasma membrane (Pinto da Silva and Nogueira, 1977). These diaphragms are thought to disintegrate, forming aqueous, etchable pores (Figs. 7 and 8). The fact that both pentalaminar and trilaminar structures are free of IMPs led to the hypothesis that fusion between membranes involves microdomains that are protein depleted and lipid rich. The presence of microdomains was given further impetus from methods which visualize lipid heterogeneity in membranes (Elias et a/., 1979). In mammalian sperm, for example, the plasma membrane overlying the acrosomal granule was

FIGS.1-3. Electron micrographs showing sequential stages of fusion between a vesicle and the plasma membrane of a capillary endothelial cell. Seen between the arrows are a pentalaminar structure (Fig. I), a trilaminar structure (Fig. 2). and a narrow pore leading into

6. MEMBRANE FUSION DURING EXOCYTOSIS

173

found to be high in cholesterol and anionic lipids (Friend and Bearer, 1981 ; Bearer and Friend, 1982; Friend, 1982). This membrane undergoes fusion during the acrosomal reaction, and its specialized lipid composition could serve to prepare it for fusion. With the advent of rapid freezing (quick-freezing)as a practical alternative to chemical fixation for preservation of cell structure (Heuser et u l . , 1979; Plattner and Bachmann, 1982; Gilkey and Staehelin, 1986). it became clear that the outline described above for exocytosis was in need of revision. Rapid freezing differs from conventional fixation procedures because it is fast (<2 msec, Heuser rt d., 1979) and well suited for capturing intermediate steps in processes as brief as exocytosis. In addition, it avoids exposure of cells to nonphysiological chemicals such as fixatives and dehydrating reagents. Comparison of quick-frozen cells with chemically fixed cells reveals a number of artifacts. First, the IMP-free areas corresponding to pentalaminar structures are much too large, sometimes covering an area almost as large as a granule diameter (Fig. 6). These areas are seen in glutaraldehyde-fixed cells where secretory granules bulge against the piasma membrane but are seldom seen in quick-frozen cells. Second, single bilayer diaphragms, in certain situations, prove to be artifacts of dehydration (Chandler and Heuser, 1979; Chandler, 1979). Such diaphragms are not seen in quick-frozen sea urchin eggs, and their occurrence in glutaraldehyde-fixed cells increases in frequency as the concentration of cryoprotective glycerol is increased. What these artifacts mean is that dehydration by glycerol can induce movement of lMPs and membrane fusion ujter glutaraldehyde fixation. In retrospect, this is not surprising. Glutaraldehyde is excellent at crosslinking proteins but is very poor at fixing lipids (Jost et ul., 1973). Glycerol and glycerol derivatives are well known as fusogenic agents (Quirk et a / . . 1978; Knutton, 1979; Aldwinckle et ul., 1982; Hui ef a l . , 1985) and are capable of causing IMP movements in unfixed cells (McIntyre et u l . , ~~

the vesicle (Fig. 3). v, Vesicle; er, endoplasmic reticulum. ~381.300(Figs. I and 3) and X353.400 (Fig. 2). Bars, 30 nm. [Reproduced from Palade and Bruns, J . Cell B i d . 37,633649 (1%8), by copyright permission from The Rockefeller University Press.] Fitis. 4 and S. Electron micrographs showing the presence of trilaminar (single arrow) and pentalaminar (double arrows) structures between granule and plasma membranes after stimulation of mucus-secreting cells in the cat submandibular gland. At high magnification (Fig. 5) one can see that penlalaminar structure consists of plasma and granule membranes tightly apposed with no intervening cytoplasm. MD. Mucous droplet; LU. lumen. x74.400 (Fig. 4) and x 167,400 (Fig. 5). Bars. 100 nm (Fig. 4) and 50 nm (Fig. 5). [Reproduced from Tandler and Poulson. J . Cell Biol. 68, 775-781 (1976). by copyright permission from The Rockefeller University Press.]

FIG.6. Freeze-fracture replica showing a large intramembrane particle (IMP)-free area on the P-face of a mast cell 30 sec after stimulation with polymyxin B. Note that the underlying granule bulges against the plasma membrane; there appears to be a small pore

6 . MEMBRANE FUSION DURING EXOCYTOSIS

175

1974). Evidently what is happening is that aldehyde fixation leaves membrane lipids quite mobile and subsequent addition of a fusogen initiates fusion events that did not take place in the living cell. Finally, it is known that aldehyde fixation produces blebs and vesicular structures (Hasty and Hay, 1978; Bretscher and Whytock, 1977). Such blebs can be seen exactly at the site of exocytosis in sea urchin eggs and mast cells (Lawson et a / . , 1977; Chandler, 1979, 1984a,b), and during membrane fusion in other systems (Kalderon and Gilula, 1979; Tanaka er al., 1980). It has been suggested that formation of vesicles and blebs may be a natural event during the course of membrane fusion. In our hands, however, these structures are fixation artifacts not seen in rapidly frozen cells (Chandler, I 984a,b) . What once were considered to be intermediate stages of membrane fusion in chemically fixed cells must now be viewed as artifacts in some cells. It is not safe to generalize that these structures are artifacts in every system: indeed, pentalaminar structures and IMP-free areas have been reported in rapidly frozen goblet cells (Neutra and Schaeffer, 1977) and sperm (Friend and Heuser, 1981). It is important to stress, though, that chemically induced artifacts are common and rapid freezing is essential to evaluate the structures seen during membrane fusion. 6. Observations in Rapidly Frozen Cells

The series of events that we and others find when cells are rapidly frozen during exocytosis can be described using mast cells as an example. Prior to secretion, histarnine-containing granules fill the cell; granules closest to the plasma membrane are separated from it by a 50- to 150-nmthick layer of cytoplasm (Fig. 9). Small filaments extend from the cytoplasm to contact the granule membrane. Upon stimulation, a small, 50nm-diameter pore is formed leading from the extracellular space into the granule (Fig. 10). This pore is the result of a highly localized fusion event within the IMP-free area. ~75,000.Bar, 100 nm. [Reproduced from Chandler and Heuser. J . Cell Biol. 86, 666-674 (1980), by copyright permission from The Rockefeller University Press.] FIG.7. Freeze-fracture replica showing IMP-free areas (arrows) on the plasma membrane of a sea urchin egg during cortical granule exocytosis. These areas are single bilayer diaphragms created artifactually by glycerination after fixation. X62.000. Bar, 200 nm. [Reproduced from Chandler and Heuser, J . Cell B i d . 83, YI-108 (1979), by copyright permission from The Rockefeller University Press.] FIG.8. Replica showing a single bilayer diaphragm (arrow) connecting granule and plasma membranes in a sea urchin egg. ~ 6 7 , 0 0 0 Bar, . 200 nm. [Reproduced from Chandler (1979) by copyright permission from Raven Press.]

6. MEMBRANE FUSION DURING EXOCYTOSIS

177

that takes place in the absence of any expansive granule-plasma membrane contact. The plasma membrane has dipped inward to accomplish the fusion event. The pore enlarges, and at a diameter of 100 nm it is etchable, indicating that it is an aqueous channel leading into the granule interior (Fig. 11). Subsequently, the exocytic pocket widens, and its contents are expelled (Fig. 12). Simultaneously, interior granules contact the pocket wall (arrows, Fig. 12). At higher magnification (Fig. 13), one can see that these contacts are plateaulike, 50 nm in diameter, and have an unetched top. This suggests that the earliest contact between the two membranes is pentalaminar or trilaminar but that it occurs over only a very restricted area. The localized nature of this event would account for why extensive IMP-free areas of membrane are not observed in quick-frozen cells. Figure 13 also demonstrates that membranes can contact and fuse at multiple sites. In mast cells we have observed up to five fusion points clustered together linking one granule membrane with another (Chandler and Heuser, 1980). Ornberg and Reese (1981) have made observations on Limulus amebocytes which complement our work on mast cells. Immediately after stimulation, they see pedestal-like indentations of the plasma membrane which come in close contact with the secretory granule (Fig. 14). As viewed in stereo micrographs, the pedestal tops are concave, matching the curvature of the granule membrane (Fig. 15). it is in this localized area of contact that a minute, 10- by 40-nm pore forms (Fig. 16). The pore is etchable, indicating that a narrow, aqueous communication has opened up between the granule interior and the extracellular space. No IMP-free areas of membrane are seen during pedestal or pore formation. Finally, as the pore widens, the rest of the granule remains well separated from the plasma membrane (Fig. 17). The work described above has been extended by a number of other studies using rapid freezing to arrest membrane fusion. Schmidt er ul. (1983), Olbricht et al. (1984), and Chandler (1984b) document the early stages of exocytosis in chromaffin cells. protozoa, and sea urchin eggs, respectively. They show that in these systems too, fusion begins with the formation of a pore as small as 10 nm in diameter which subsequently FIGS.9-1 1. Early events during fusion of secretory granule and plasma membranes in a mast cell, The two membranes are separated by a generous layer of cytoplasm in the unstimulated cell (Fig. 9). Fusion begins with formation of a narrow (SO nm diameter) pore (Fig. 10). As the pore enlarges, its etched appearance shows that it is an aqueous channel leading from the extracelIular space into the granule interior (Fig. I I ) . x 165,000 (Fig. 9) and x 185,000 (Figs. 10 and 1 I ) . Bars, SO nm. [Reproduced from Chandler and Heuser, J . Cell B i d . 86, 666-674 (1980), by copyright permission from The Rockefeller University Press.]

6 . MEMBRANE FUSION DURING EXOCYTOSIS

179

FIG.13

enlarges. At no point in the process are IMP-cleared areas of membrane seen, either in the plasma or granule membranes. Furthermore, R. L. Ornberg (personal communication) has observed that exocytosis in chromaffin cells is preceded by a pedestallike deformation of the plasma membrane extending inward to contact the granule membrane. The fact that membrane fusion begins in a localized area has suggested that fusion may be initiated at point defects in the contact region. Several studies have pointed out that “lipidic” particles (inverted micelles within the bilayer) represent such a defect (Verkleij e t d.,1979, 1980). These defects are seen as 10-nm-diameter deformations in freeze-fracture replicas of artificial lipid vesicles undergoing calcium-dependent membrane fusion, and the manner in which they might be involved in membrane FIGS.I ? and 13. Opening up of an exocytic pocket in a mast cell. Etching reveals that the granule contents have pulled away from the cross-fractured pocket wall and are being expelled. A more interior granule has begun to fuse with the pocket wall, as evidenced by two pedestallike contacts (arrows). At higher magnification (Fig. 13). it can be seen that these pedestals appear to represent highly localized pentalaminar contacts (arrows). Also seen is a small cytoplasmic vesicle that has fused with the pocket wall (asterisks). x65,000 (Fig. 12) and x 120,000 (Fig. 13). Bars. 100 nm (Fig. 12) and 100 nm (Fig. 13). [Reproduced from Chandler. Biochrm. Soc. Trrins. 12,961-963 (1984). by copyright permission from The Biochemical Society.]

180

DOUGLAS E. CHANDLER

FIGS.14 and IS. Formation of pedestallike contacts between plasma and granule membranes in endotoxin-stimulated Limulus amebocytes. In Fig. 14, three pedestal-shaped protrusions are seen on the E-face of the plasma membrane. The stereo pair of micrographs in Fig. I5 shows a pedestal that has contacted the granule membrane, forming a pentalaminarlike structure that is slightly concave to match the curvature of the granule. ~49,000 (Fig. 14) and ~ 3 8 , 0 0 0(Fig. 15). Bars, 200 nm. [Reproduced from Ornberg and Reese, J . Cell B i d . 90, 40-54 (1981). by copyright permission from The Rockefeller University Press.]

fusion has been discussed by Plattner (1981) and Verkleij (I984). Rapid freezing studies have not provided a definitive answer on the importance of this mechanism. Hui et al. (1981) have shown that fusion of artificial lipid vesicles induced by freezing and thawing results in point contacts between membranes that resemble lipidic particles. In contrast, Rand et al. (1981) and Bearer et al. (1982) report that point deformations are not present during calcium-induced fusion of lipid vesicles and instead appear to be artifacts of glycerination. Furthermore, Rand et al. (1981) observe that membrane fusion in rapidly frozen lipid vesicles involves formation of localized contacts and small pores, a process similar to that described above for exocytosis. Regardless of what mechanisms actually produce the first tiny pore, several observations suggest that the lipid bilayers at the point of fusion

6 . MEMBRANE FUSION DURING EXOCYTOSIS

181

FIG. IS

are highly mobile. First, in glutaraldehyde-fixed cells, there is a tendency for blebs to be formed exactly at the point of membrane fusion. These blebs, quite large and composed of membrane that is continuous with the plasma or granule membranes, indicate that there has been a substantial movement of phospholipids into the area where fusion is about to take place. Our interpretation is that chemical fixation is a relatively slow, diffusion-limited process requiring several seconds to halt biological processes. In contrast, the entire exocytosis sequence is quite rapid, requiring a few milliseconds to a few tenths of a second. It seems reasonable to believe, then, that these artifacts occur as exocytosis is attempted in a cell that is already partially fixed. Chemical cross-linking of some components, such as the granule contents or the cytoskeleton, may slow or halt granule opening, while mechanisms that increase the fusibility of these bilayers just before exocytosis are free to act for an abnormally long

182

DOUGLAS E. CHANDLER

FIGS. 16 and 17. Pore formation during exocytosis in endotoxin-stimulated Limulus amebocytes. In Fig. 16, a 10- by 40-nm etchable pore has opened up at a pedestallike contact. In Fig. 17, the pore has enlarged to about 300 nm in diameter and clearly represents a narrow aqueous channel leading from the extracellular space into the granule interior. x60,OOO (Fig. 16) and X50,OOO (Fig. 17). Bars, 200 nm. [Reproduced from Ornberg and Reese, J . CellBiol. 90,40-54 (1981), by copyright permission from The Rockefeller University Press.]

period of time. Formation of blebs, as well as extensive IMP-free areas, may be an exaggeration of a physiological process which under normal cellular control occurs only transiently and only in a highly localized area so as to facilitate the initial fusion event. A second observation that reflects the increased fusibility of these membranes is the presence of small vesicles at the entrance to exocytic pockets in quick-frozen sea urchin eggs (arrow, Fig. 18). These membrane-bound vesicles appear to bud off the lip of the exocytic pore (Fig. 19). This could mean that enlargement of the exocytic pore is accomplished by additional fusion events that form vesicles. Retraction of cytoskeletal filaments from a localized area may leave granule and plasma membranes unsupported, highly distensible, and readily capable of fusion. Regions cleared of filaments have been seen between fusing membranes in mast cells (Lawson et al., 1977), and red blood cell membranes when stripped of their underlying cytoskeleton become highly susceptible to vesiculation (Elgsaeter et al., 1976; Allan et al., 1980, 1982). In some cells, small cytoplasmic vesicles appear to act as initiators of

6. MEMBRANE FUSION DURING EXOCYTOSIS

183

fusion between larger granules. Chemotactic peptide-initiated release of lysosomal enzymes from neutrophils results in fusion of granules to form fingerlike invaginations (Chandler rt ul., 1983; Chandler and Kazilek. 1986). Seen at the growing tips of invaginations and attached to single granules are small vesicles apparently in the process of fusion. Fusion of cytoplasmic vesicles with secretory granules is also seen in basophils (Dvorak el d.,1981). Likewise, in quick-frozen mast cells we see evidence of small vesicles that have fused with granule membranes either at the exocytic pore (not shown) or at points adjacent to sites of ongoing membrane fusion (asterisk, Fig. 13). Further studies are needed to determine whether small cytoplasmic vesicles can act to facilitate fusion between larger granules just as virus particles act to induce fusion between cells. The increased mobility of plasma and granule membranes before or during fusion contrasts with an apparent decrease in mobility after the exocytic pore opens. In quick-frozen mast cells (Chandler and Heuser, 19801, sea urchin eggs (Chandler and Heuser, 1979). and nerve terminals (Heuser and Reese, 1981). the E-face of secretory granule or vesicle membranes have a low density of large IMPS while the plasma membrane P-face has a high density of smaller IMPs. These differences in IMP distribution can be used as a way to monitor the intermixing of granule and plasma membrane constituents during fusion. This intermixing does not occur as fast as one might imagine based on diffusion. Sharp boundaries in IMP distribution are seen at the lip of exocytic pockets well after they have completely opened up (Chandler and Heuser, 1979, 1980). The wavelike sweep of cortical granule exocytosis in the sea urchin egg, traveling at 5 pmlsec, bas allowed us to estimate that about 5 sec elapse between the first sign of granule fusion at the front of the wave and the final smoothing out of exocytic pockets at the rear of the wave. Intermixing of plasma membrane and granule membrane IMPs requires 2-3 sec of this period. In the neuromuscular junction, exocytosis of each synaptic vesicle is accompanied by addition of several large IMPs to the plasmalemma. These markers of vesicle membrane remain clustered for at least I sec, a long time considering that exocytosis itself takes less than 5 msec (Heuser and Reese, 1981). Subsequently, these particles are taken up by endocytosis, which reaches a peak at 30 sec after stimulation (Miller and Heuser, 1984). In thin sections of quick-frozen Liiiiulirs amebocytes one can see a boundary between the plasma and granule membranes during exocytosis (Ornberg and Reese, 1981). The granule membrane is thinner than the adjoining plasma membrane and their differences in thickness persist after

6. MEMBRANE FUSION DURING EXOCYTOSIS

185

fusion. This boundary does not disappear until the exocytic pocket is completely open, a process that takes 0.5 sec as estimated from cinematographic data. A similar observation has been made based on distribution of cholesterol in membranes. Filipin binds stoichiometrically to P-hydroxysterols and in doing so produces 20-nm-diameter protuberances that are easily seen in freeze-fracture replicas (Elias et a / . , 1979). Orci et al. (1981) have shown that fusion of cholesterol-rich granule membranes with cholesterol-poor plasma membranes results in a mosaic membrane in which filipin complexes are seen clustered together in exocytic pockets and in patches on the cell surface. Frequently, there remains a sharp demarcation between cholesterol-rich and cholesterol-poor domains for several seconds after granule-plasma membrane fusion. Although these observations were made in chemically fixed cells, Friend and Bearer (1981) indicate that filipin can be used in conjunction with rapid freezing to visualize cholesterol-rich microdomains in capacitated guinea pig sperm. Future ultrastructural studies need to utilize these techniques to study cholesterol distribution during exocytosis in other secretory cells. Membrane proteins and lipids play a crucial role in the fusion process, but additional interest has focused on the role of osmotic forces in driving membrane fusion and expulsion of granule contents. What is known is that entrance of water and ions into the granule results in swelling of the secretory granule during exocytosis. Such swelling in cortical granules from sea urchin eggs has been quantitated at the light microscope level by Zimmerberg and Whitaker (1985) and can be qualitatively recognized in electron micrographs of quick-frozen cells. The exocytic pocket in Fig. 18, for example, is about twice the diameter of a cortical granule before fusion. Studies with both artificial lipid vesicles (Cohen et al., 1980, 1984) and cortical granules (Zimmerberg and Whitaker, 1985) suggest that swelling may take place before membrane fusion has ocurred, indicating that granule constituents have become osmotically active and water has entered. Ornberg and Reese (1981) have shown in the amebocyte that dissolution of granule contents is first seen as a separation between granule membrane and contents just under the point of granule-plasma memFIGS. 18 and 19. Electron micrographs of a quick-frozen and freeze-substituted sea urchin egg showing an exocytotic pocket from which granule contents are being expelled. As the mouth of the pocket enlarges, small vesicles appear to pinch o f ffrom the lips (arrow, Fig. 18). At higher magnification (Fig. 19), these apparently membrane-bound vesicles are spherical or tubular in shape. X50,OOO (Fig. 18) and x 160,000 (Fig. 19). Bars, 200 nm (Fig. 18) and 50 nm (Fig. 19). [Reproduced from Chandler. J . Cell Sci. 72, 23-36 (1984), by copyright permission from The Company of Biologists, Ltd.]

186

DOUGLAS E. CHANDLER

brane contact. Delamination of the granule membrane from the core spreads quickly, and at the same time the granule core begins to dissociate into columns of 50- to 100-nm densely stained particles. Dissolution of granule contents moves as a wave from the pore toward the interior of the granule. Wherever a pore has formed, dissolution has already begun; however, the small size of the initial pore makes it impossible to determine whether dissolution can precede pore formation (Ornberg and Reese, 1981). An alternative experimental approach has been to use hyperosmotic conditions to inhibit exocytosis. Increasing the osmotic strength of the extracellular medium should prevent the movement of water into the granule. Indeed, hyperosmotic conditions do inhibit exocytosis in sea urchin eggs (Zimmerberg and Whitaker, 198% adrenal chromaffin cells (Hampton and Holz, 1983; Pollard et al., 1984), and neutrophils (D. E. Chandler, unpublished observations). Using sea urchin egg cortices, Zimmerberg et al. (1985) have shown that low molecular weight osmoticants, such as sucrose and stachyose, reduce calcium-initiated swelling of cortical granules and slow the rate of exocytosis. Future ultrastructural studies utilizing quick-freezing should help visualize granule swelling and preliminary steps in membrane fusion that have been stabilized during hyperosmotic inhibiton. Whitaker and Zimmerberg (1986) further show that dextrans of low molecular weight (<3000) inhibit exocytosis, as expected from the osmotic strength of dextran solutions. Higher molecular weight dextrans, however, inhibit at lower osmotic strengths than expected, and inhibition is greater in the presence of divalent cations. These investigators postulate that high molecular weight dextrans are unable to enter the meshlike superstructure that holds the granule contents together. Exclusion of dextran from the granule core is thought to prevent water movement into the granule after pore formation; divalent cations are thought to stabilize the granule core and prevent it from dissolving. Ultrastructural studies are needed to confirm these predictions, but, if true, they would suggest that hydration and dissolution of granule contents immediately after fusion is an important force in widening the pore and expelling the secretory products. Such a hypothesis is supported by recent studies of trichocyst discharge in Paramecium (Bilinski et al., 1981; Gilligan and Satir, 1983). Membrane fusion is followed by expansion of the trichocyst, forming a greatly elongated matrix. Matrix expansion requires extracellular calcium, and in its absence the trichocyst is not propelled into the extracelluIar space even though membrane fusion has occurred. Bilinski et al. (1981) have postulated that calcium precipitates the high amount of phos-

187

6.MEMBRANE FUSION DURING EXOCYTOSIS

phate found in unreleased trichocysts and thereby neutralizes phosphatematrix charge interactions. In addition, Gilligan and Satir (1983) have shown in mutant paramecia that triggering of trichocyst discharge results in partial matrix expansion even in the t i h s ~ n c eof membrane fusion. This suggests that ions and/or water may be entering the trichocysts during or even before the abortive attempt at membrane fusion. Figure 20 summarizes what we currently know about exocytosis. In Fig. 20A, an unfused secretory granule lies just below the plasma membrane. Fusion is preceded by reorganization of the cytoskeleton between the granule and plasma membrane so as to allow one membrane to approach the other in a highly fusible state. Figure 20B shows that a pedestallike deformation in the plasma membrane extends into the cell and contacts the membrane of the secretory granule over a limited region. The pedestal forms a small pentalaminar contact with the granule membrane, and at that contact a pore, 10-50 nm in diameter, is formed connecting the B A

FIG.20. Steps in membrane fusion as seen in cells quick-frozen during exocytosis. ( A ) Rearrangement of cytoskeletal or docking proteins between plasma and granule membranes has created a highly localized and differentiated region in which membrane fusion can take place. (B) In this region, the plasma membrane forms a pedestal-shaped protrusion which extends inward to contact the granule membrane. (C) Membrane fusion at this contact creates a small pore and subsequently a narrow. tubular neck, SO-I00 nm in diameter. between the plasma and granule membranes; it represents an aqueous channel joining the extracellular space with the granule interior. (D)The exocytic pocket has now opened up, and small vesicles are budding from the lip of the expanding pore. The exocytic pocket. swollen during B and C, now has a noticeably larger diameter than that of an unreleased granule. An interior granule has begun to fuse its membrane with that of the exocytic pocket by forming multiple pedestallike contacts. Small cytoplasmic vesicles also fuse with the exocytic pocket.

FIGS.21-23. Electron micrographs showing the sequence of events during endocytosis of yolk protein in hen oocytes. In Fig. 21, a concave invagination has formed in the plasma membrane which has an electron-dense coating on its cytoplasmic surface. In Fig. 22, invagination has progressed to form a “coated pit.” Finally (Fig. 23), a spherical vesicle has

6. MEMBRANE FUSION DURING EXOCYTOSIS

189

granule interior with the extracellular space (Fig. 20C). At some point during this process (Figs. 20B, C) the granule begins to swell due to movement of water into the granule. Finally the pore widens as the granule swells and contents dissipate (Fig. 20D). Tiny vesicles bud off from the receding lip of the pore and are now seen in the extracellular space. Interior granules begin to fuse at sites where multiple, pedestallike deformations of their membrane contact the exocytic pocket. Also seen are small vesicles in the cytoplasm that have just fused with the pocket wall and which may act to promote fusion between granules. II. ENDOCYTOSIS

The role of endocytosis has been expanded in recent years to include membrane retrieval, entry of receptor-bound ligands, receptor cycling, and entry of fluid- and solid-phase materials from the extracellular space. It is the subject of a number of recent reviews (Farquhar, 1983; Helenius et ul., 1983; Goldstein et ul., 1985; Pastan and Willingham, 1985; Wileman et u l . , 1985). Our treatment here will be limited to the structural features of endocytosis as seen by the quick-freeze, deep-etch, rotary-shadow method (Heuser and Salpeter, 1979; Chandler, 1986). In this method cells are aldehyde fixed or detergent extracted (or both), quick-frozen, then extensively etched. Sublimation of ice exposes biological structures on the external and internal surfaces of membranes that are then visualized by showering them with platinum/carbon at a low angle. As viewed in thin section (Figs. 21-23), endocytosis begins as a shallow concavity in the plasma membrane. At this point, the membrane is coated on its cytoplasmic surface with a layer of densely staining material that is formed which is about to pinch off at the narrow neckjoining it to the cell surface. x 101,370. Bars, 100 nm. [Reproduced from Perry and Gilbert, J . Cell Sci. 39, 257-272 (1979), by copyright permission from The Company of Biologists, Ltd.] FIGS.24 and 25. Platinum replicas of deep-etched and rotary-shadowed fibroblasts illustrating the structure of endocytic pits and coated vesicles. The endocytic invagination in Fig. 24 is coated with a basketlike network of filaments seen in cross section. Endocytosis results in formation of a coated vesicle such as that seen in Fig. 25. In both figures the vesicle coat appears to be linked to elements in the cytoplasm. x 192,510. Bars, SO nm. [Reproduced from Heuser, J . Cell Bio/. 84,560-583 (1980), by copyright permission from The Rockefeller University Press.] FIG. 26. Stereo electron micrographs showing an “aerial” view of an endocytic invagination in fibroblasts. Low density lipoprotein particles are clustered at the pit, probably bound to receptors that are being moved into the invagination. ~90,210.Bar, 100 nm. [Reproduced from Heuser, J . Cell Biol. 84, 560-583 (1980). by copyright permission from The Rockfeller University Press.]

6. MEMBRANE FUSION DURING EXOCYTOSIS

191

known to contain clathrin. The electron-dense material seen within the invagination is yolk protein being taken up by hen oocytes (Perry and Gilbert, 1979). As the invagination deepens, the curvature of the membrane increases (Fig. 22) until a nearly spherical vesicle is formed having a small pore connecting its interior with the extracellular space (Fig. 23). Membrane fusion takes place at this pore, completely sealing off the vesicle and allowing it to move into the cell interior. In deep-etched cells (Fig. 24), this pore is seen as a narrow, tubular neck joining the clathrincovered vesicle with the cell surface above. The clathrin coat appears to have linkages to the cytoskeleton both before (Figs. 24) and after (Fig. 25) the vesicle has pinched off. A Stereo view of the fibroblast surface during endocytosis of low density lipoprotein particles (Fig. 26) demonstrates the narrow entrance to the vesicle during invagination. The structure of the clathrin basket was originally deduced from thin sections of “coated” vesicles from brain (Kanaseki and Kadota. 1969) and has since been elegantly visualized in deep-etched fibroblasts by Heuser (1980). As shown in Fig. 27, Heuser was able to reconstruct the sequence of structural changes that the clathrin basket goes through during pinching off of an endocytic vesicle. Before endocytosis begins, the inner side of the fibroblast plasma membrane is covered with large, planar arrays of hexagonal meshwork (Fig. 27a). Invagination of the plasma membrane is accompanied by increasing curvature of this basket until a geodesic dome-like structure is formed (Figs. 27b-d). Finally, a bulbous protrusion is squeezed out from the plasma membrane (Figs. 27e-g), and a mushroomlike invagination is formed (Fig. 27h). Endocytic invaginations in many cells have a tip that is covered with clathrin, but the neck is usually bare (Heuser, 1980; D. E. Chandler, unpublished observations). The neck of the invagination as well as the clathrin basket can be linked to cytoskeletal filaments. In Fig. 28 (stereo), endocytic invaginations in rabbit neutrophils are covered with clathrin and in addition are surrounded by a collar of cytoskeletal filaments. In some cases the filaments appear as if they were guy wires lending support. At baskets that we presume are in the process of rounding up (Fig. 29, stereo. viewed from the side), we see a number of cytoskeletal filaments that extend to the basket and contact it. Reorganization of the basket structure may provide the force necessary to pull surface membrane into the cell. As illustrated in Fig. 30, the normal clathrin basket consists of a planar network of hexagons fitting together much like tiles on a kitchen floor (Heuser, 1980). Curvature of FIG.27. Sequential sieps in endocytosis as viewed in quick-frozen, deep-etched, and rotary-shadowed fibroblasts. x 176.000. Bar, 100 nm. [Reproduced from Heuser, J . Cell B i d . 84, 560-583 (19801, by copyright permission from The Rockefeller University Press.]

192

FIG.28. Stereo electron micrographs of the cytoplasmic aspect of the neutrophil plasma membrane. Note several coated invaginations at various stages of formation. All seem to have a collar of filaments that extend from the filament network in contact with the inner surface of the plasma membrane. The specimen was fixed with glutaraldehyde. quick-frozen. freeze-dried, and rotary-shadowed with platinum. x 162,750. Bar, 50 nm. FIG.29. Stereo electron micrographs of a partially formed endocytic invagination in the neutrophil plasma membrane. This figure must be rotated 90" to be viewed properly. Note cytoskeletal filaments that extend to contact the clathrin coat. The specimen was prepared as described in Fig. 28. ~205,000.Bar, 50 nm.

193

194

DOUGLAS E. CHANDLER

the basket during endocytosis is invariably accompanied by dislocations in this pattern. At each dislocation, two adjacent hexagons rearrange to form a heptagon and a pentagon (7 and 5, Fig. 30). These dislocations serve to make parallel rows of hexagons converge, and thereby produce curvature in the basket. Similar baskets are seen on the inner side of plasma membrane being internalized by rnacrophages, suggesting that these same mechanisms may be at work during phagocytosis (Aggeler and Werb, 1982; Aggeler et a/., 1983; Takemura et al., 1986). Clathrin is the major constituent of these baskets (Pearse, 1976, 1978; Kirchhausen and Harrison, 1981). Ungewickell and Branton (1981, 1982) have shown that the shape of this protein is consistent with its ability to form such a network. Purified clathrin, visualized by low-angle shadowing on a mica surface (Fig. 31), is a pinwheellike trimer having three “heavy chains” extending from a central focus. Each chain has a well defined kink about midway in its length. These protein building blocks, illustrated diagramatically in Fig. 32A, can be arranged to form a hexagonal meshwork (Fig. 32B) much like that seen in baskets of deep-etched cells. Each trimer is centered at one apex in the hexagon, and each side of the hexagon consists of heavy chains from four clathrin molecules. In addition, there are smaller peptides, the “light chains,” associated with each clathrin trimer shown in Fig. 32A but omitted from Fig. 32B for clarity. Dislocations within such an arrangement, to form an adjacent heptagon and pentagon (Fig. 32C), are possible without changing the basic arrangement of the trimers within the basket, The mechanisms that produce dislocation are not yet known. The coat is associated with two protein kinase activities (Bar-Zvi and Branton, 1986) and it is known to transiently associate with a clathrin-dependent ATPase that is involved in removal of clathrin coats from vesicles (Schlossman et al., 1984; Schmid et al., 1984). Whether these or other enzymatic activities regulate basket curvature, or whether the cytoskeleton plays an important role in this process, is a subject for future study. 111.

CONCLUDING REMARKS

Exocytosis and endocytosis present some mechanistic problems that are solved in a similar manner. Both utilize membrane fusion events that occur in highly localized environments. During exocytosis, membrane fusion creates an exceedingly small pore that leads to formation of a narrow, tubular neck. During endocytosis formation of a narrow tubular neck is required before fusion will take place (see Fig. 23). This means that fusion between cell membranes requires conditions that are realized

FIG.30. The lattice structure of the clathrin coat surrounding endocytic invaginations. Most of the basket is composed of hexagonal units. At certain points there are dislocations consisting of adjacent heptagons (7) and pentagons ( 5 ) that require curvature in the basket. ~345,600. Bar, 30 nm. [Reproduced from Heuser, 1. Cell Biol. 84, 560-583 (19801, by copyright permission from The Rockefeller University Press.] FIG.3 I , Structure of individual clathrin trirners visualized by low-angle rotary shadowing. x 174,720. Bar. 50 nm. [Reproduced from Ungewickell and Branton. Trends Biochrm. Sci. 7, 358-361 (1982). by copyright permission from Elsevier Biochemical Press.]

196

DOUGLAS E. CHANDLER

A

Y

FIG.32. Diagramatic representation showing how individual clathrin molecules might fit into the basketlike structure which coats the endocytic invagination. (A) clathrin trimer consisting of three heavy chains that splay out radially from the center, each having a kink about midway, and light chains found close to the center. (B) Hexagonal unit cells of the basket have a single trimer centered at each apex and sides consisting of heavy chains from four separate clathrin trimers. (C) Heptagodpentagon dislocations in the basket could in theory have a similar arrangement of clathrin molecules as in B while forcing the basket to have curvature (arrows).

only within a small domain for a brief moment. This does not mean that only preordained sites on membranes can engage in fusion (for which there is evidence only in particular cases) but that a transient and presumably unstable state of high fusibility can be induced only over an area of small physical dimensions. In addition, formation of a narrow neck during endocytosis or enlargement of the pore during exocytosis requires rapid movements of phospholipids into or out of the region. In this region the bilayer must be highly mobile, a characteristic that may be of importance in the fusion event itself. It is also clear that exocytosis and endocytosis are fundamentally different. Exocytosis involves release of potential energy that is stored in the highly organized structure of the granule core. The contents of nearly all secretory granules are packaged so as to limit hydration and to maximize interaction between constituents. In some cases, the granule membrane is designed to maintain ion gradients between the interior of the granule and the cytoplasm. Swelling and hydration of the contents due to movement of ions and water into the granule during exocytosis may be a powerful force either for driving membrane fusion or for opening up the pore after fusion. In contrast, endocytosis cannot use such potential energy as a resource and instead must rely on specially designed structures, the clathrin coat or the cytoskeleton (in the case of uncoated vesicles), to induce membrane curvature to the point at which membrane fusion can occur.

6. MEMBRANE FUSION DURING EXOCYTOSIS

197

ACKNOWLEDGMENTS Some of the studies described were supported by grants from the National Science Foundation (DCB-8407152) and the National Institutes of Health (HD00619). I thank Charles Kazilek for preparing the publication prints and Greg Hendricks for drawing Fig. 20 and 32. REFERENCES Aggeler, J., and Werb, Z. (1982). Initial events during frustrated phagocytosis by macrophages viewed from outside and inside the cell: Membrane particle interactions and clathrin. J . Cell B i d . 94. 613-623. Aggeler, J . , Takemura. R., and Werb, Z. (1983). High-resolution views of membrane-associated clathrin and cytoskeleton in critical point dried macrophages. J . Cell Biol. 97, 1452-1458. Aldwinckle. T. J., Ahkong, Q. F., Bangham. A. D., Fisher, D.. and Lucy, J . A . (1982). Effects of poly(ethy1ene glycol) on lipostrmes and erythrocytes. Permeability changes and membrane fusion. Biochim. Biophys. Acrcr 689, 548-560. Allan. D.. Thomas, P., and Limbrick. A. R. (1980). The isolation and characterization of 60 nm vesicles ('nanovesicles') produced during ionophore A23 187-induced budding of human erythrocytes. Biocliem. J . 188, 8X 1-887. Allan, D.. Thomas, P., and Limbrick, A . R. (1982). Microvesiculation and sphingomyelinase activation in chicken erythrocytes treated with ionophore A23 I87 and Ca?*.Biochim. Biophys. Acta 693, 53-67. Amherdt, M., Baggiolini, M., Perrelet, A , , and Orci, L. (1978). Freeze fracture of membrane fusions in phagocytosing polymorphonuclear leukocytes. Lab. Inuesf. 39, 398-404. Aunis. D., Hesketh, J . E . , and Devilliers, G. (1979). Freeze-fracture study of the chromaffin cell during exocytosis: Evidence for connections between the plasma membrane and secretory granules and for movements of plasma membrane-associated particles. Cell Tissue Res. 197, 433-441. Bar-Zvi, D., and Branton, D. (1986). Clathrin-coated vesicles contain two protein kinase activities. Phosphorylation of clathrin p-light chain by casein kinase 11. J . Biol. C h r m . 261, 9614-9621, Bearer. E. L., and Friend, D. S. (1982). Modifications of anionic-lipid domains preceding membrane fusion in guinea pig sperm. J . Cell Biol. 92, 604-615. Bearer. E. L.. DuzgiineS, N., Friend, D. S . . and Papahadjopoulos, D. (1982). Fusion of phospholipid vesicles arrested by quick-freezing. The question of lipidic particles as intermediates in membrane fusion. Biochim. Biophvs. Acra 693, 93-98. Beisson, J . , Lefort-tran, M.. Pouphile, M.. Rossignol, M . , and Satir, B. (1976). Genetic analysis of membrane differentiation in Porarnecirrm. Freeze-fracture study of the trichocyst cycle in wild-type and mutant strains. J . Cell B i d . 69, 126-143. Berger. W., Dahl, G . , and Meisner, H.-P. (1975). Structural and functional alterations in fused membranes of secretory granules during exocytosis in pancreatic islet cells of the mouse. Cytohiologie 12, 119-139. Berridge, M. J.. and Irvine, R. F. (1984). lnositol trisphosphate, a novel messenger in cellular signal transduction. Nature (London) 312, 315-321. Bilinski. M., Plattner, H., and Matt, H. (1981). Secretory protein decondensation as a distinct, calcium-mediated event during the final steps of exocytosis in Paratnecium cells. J . Cell Biol. 88, 179-188. Bretscher, M. S . , and Whytock, S . (1977). Membrane-associated vesicles in fibroblasts. J . Ullrasrruct. Res. 61, 215-2 17.

198

DOUGLAS E. CHANDLER

Burwen, S. J . , and Satir, B. H. (1977). A freeze fracture study of early membrane events during mast cell secretion. J. Cell Biol. 73, 660-671. Chandler, D. E. (1979). Quick-freezing avoids specimen preparation artifacts. In “Freeze Fracture: Methods, Artifacts, and Interpretations” ( J . E. Rash and C. S. Hudson. eds.). pp. 81-87. Raven, New York. Chandler, D. E. (1984a). Exocytosis involves highly localized membrane fusions. Biochem. Soc. Truns. 12, 961-963. Chandler, D. E. (1984b). Comparison of quick-frozen and chemically fixed sea urchin eggs: Structural evidence that cortical granule exocytosis is preceded by a local increase in membrane mobility. J . Cell Sci. 72, 23-36. Chandler, D. E. (1986). Rotary shadowing with platinum-carbon in biological electron microscopy: A review of methods and applications. J . Electron Microsc. Tech. 3, 305335. Chandler, D. E., and Heuser, J . (1979). Membrane fusion during secretion. Cortical granule exocytosis in sea urchin eggs as studied by quick-freezing and freeze-fracture. J . Cell Biol. 83, 91-108. Chandler, D. E., and Heuser, J . (1980). Arrest of membrane fusion events in mast cells by quick-freezing. J . Cell Biol. 86,666-674. Chandler, D. E., and Kazilek. C. J . (1986). Chemotactic peptide-induced exocytosis in neutrophils: Granule fusion patterns depend on the source of messenger calcium. J . Cell Sci. 83, 293-313. Chandler, D. E., Bennett, J . P., and Gomperts. B. D. (1983). Freeze-fracture studies of chemotactic peptide-induced exocytosis in neutrophils: Evidence for two patterns of secretory granule fusion. J . Ul~rtrs~rucr. Res. 82, 22 1-232. Chi, E. Y., Lagunoff, D., and Koehler, J . K. (1976). Freeze fracture study of mast cell secretion. Proc. Nurl. Acud. Sci. U . S . A . 73, 2823-2827. Cohen, F. S.. Zimmerberg, J., and Finkelstein, A. (1980). Fusion of phospholipid vesicles with planar phospholipid bilayer membranes. 11. Incorporation of a vesicular membrane marker into the planar membrane. J . Gen. Physiol. 75, 251-270. Cohen, F. S., Akabas, M. H.. Zimmerberg, J . , and Finkelstein, A. (1984). Parameters affecting the fusion of unilamellar phospholipid vesicles with planar bilayer membranes. J . Cell Biol. 98, 1054-1062. De Camilli, P., Peluchetti, D.. and Meldolesi. J . (1976). Dynamic changes of the luminal plasmalemma in stimulated parotid acinar cells. J . Cell Biol. 70, 59-74. Douglas, W.W. (1968). Stimulus-secretion coupling-The concept and clues from chromaffin and other cells. Br. J . Phurmucol. 34, 451-474. Douglas, W. W . (1974). Involvement of calcium in exocytosis and the exocytosis-vesiculation sequence. Biochern. Soc. Symp. 39, 1-28. DiizguneS, N. (1985). Membrane fusion. Subcell. Biochern. 11, 195-286. Dvorak, A. M., Galli, S. J., Morgan, E., Galli, A. S . , Hammond, M. E., and Dvorak, H. F. (1981). Anaphylactic degranulation of guinea pig basophilic leukocytes. I . Fusion of granule membrane and cytoplasmic vesicles: Formation and resolution of degranulation sacs. Lab. Inuesr. 44, 174-191. Elgsaeter, A., Shotton, D. M., and Branton. D. (1976). lntramembrane particle aggregation in erythrocyte ghosts. 11. The influence of spectrin aggregation. Biochim. Biophy.r. Acra 426, 101-122.

Elias, P. M., Friend, D. S., and Goerke, J . (1979). Membrane sterol heterogeneity. Freezefracture detection with saponins and filipin. J . Hisrochem. Cyrochem. 27, 1247-1260. Farquhar, M. G. (1983). Multiple pathways of exocytosis, endocytosis, and membrane recycling: Validation of a Golgi route. Fed. Proc.. Fed. Am. Soc. E x p . Biol. 42, 24072413.

6 . MEMBRANE FUSION DURING EXOCYTOSIS

199

Fernandez, J. M.. Neher. E . , and Gomperts, B. D. (1984). Capacitance measurements reveal stepwise fusion events in degranulating mast cells. Nature. (London) 312, 453455. Friend. D. S . (1982). Plasma membrane diversity in a highly polarized cell. J . Cell B i d . 93, 243-249, Friend. D. S . , and Bearer, E. L. (1981). 8-Hydroxylsterol distribution as determined by freeze-fract ure cytochernistry . Hisrochem. J . 13, 535-546. Friend. D.S., and Heuser. J . E. (1981). Orderly particle arrays on the mitochondria1 outer membrane in rapidly frozen sperm. Awut. Rec. 199, 159-175. Friend. D. S., Orci. L.. Perrelet. A., and Yanagirnachi, R . (1977). Membrane particle changes attending the acrosorne reaction in guinea pig spermatozoa. J . Cell Eiol. 74, 561-577. Gilkey, J. C.. and Staehelin. L. A. (1986). Advances in ultrarapid freezing for the preservation of cellular ultrastructure. J . Eleciron M i i w s c . Tech. 3, 177-210. Gilligan. D. M., and Satir, B. H . (1983). Stimulation and inhibition of secretion in Purumecium: Role of divalent cations. J. Cell Biol. 97, 224-234. Goldstein, J. L.. Brown, M. S., Anderson, R . G. W., Russel, R. W.. and Schneider. W. J . (1985). Receptor-mediated endocytosis. A W ~ IReo. I . Cell Biol. 1, 1-40. Hampton, R . Y . , and Holz, R. W . (1983). Effects of changes in osmolality on the htability and function of cultured chromaffin cells and the possible role of osmotic forces in exocytosis. J. Cell Eiol. 96, 1082-1088, Hasty, D. L., and Hay, E. D. (1978).Freeze-fracture studies ofthe developing cell surface. 11. Particle-free membrane blisters on glutaraldehyde fixed corneal fibroblasts are artefacts. J . Cell Biol. 78, 756-768. Helenius, A., Mellman, I., Wall, D., and Hubbard. A. (1983). Endosomes. Trends Biochem. Sci. 8, 245-250. Heuser. J. E. ( 1980). Three-dimensional visualization of coated vesicle formation in fibroblasts. J . Cell Biol. 84, 560-583. Heuser. J . E., and Reese, T. S. (1981). Structural changes after transmitter release at the frog neuromuscular junction. J . Cell B i d . 88, 564-580. Heuser. J . E . , and Salpeter, S . R. (1979).Organization of acetylcholine receptors in quickfrozen, deep-etched, and rotary-replicated Torpedo postsynaptic membrane. J. CelI B i d . 82, 150-173. Heuser. J. E.. Reese, T. S.. and Landis. D. M. D. (1974). Functional changes in frog neuromuscular junctions studied with freeze-fracture. J. Nerrrocyfol. 3, 109- 13 1. Heuser, J. E., Reese, T. S., Dennis, M.J.. Jan, Y.. Jan, L., and Evens, L. (1979). Synaptic vesicle exocytosis captured by quick-freezing and correlated with quanta1 transmitter release. J . CeN Biol. 81, 275-300. Holstein, T., and Tardent, P. (1984). An ultrahigh speed analysis of exocytosis: Nematocyst discharge. Science 223, 830-833. Hui. S.W., Stewart, T. P., Boni, L. T., and Yeagle, P. L. (1981). Membrane fusion through point defects in bilayers. Science 212, 92 1-923. Hui. S . W . , Isac, T., Boni, L. T.. and Sen, A . (1985). Action of polyethylene glycol on the fusion of human erythrocyte membranes. J. Memhr. B i d . 84, 137-146. Jaffe, L. F., Hagiwara, S., and Kado, T. (1978). Time course of cortical granule fusion in sea urchin eggs observed as membrane capacitance changes. Deu. Biol. 67, 243-248. Jamieson, J. D. (1972). Transport and discharge of exportable proteins in pancreatic exocrine cells: I n uirro studies. Curr. Top. Memhr. Trump. 3, 273-338. Jamieson, J. D., and Palade, G . E. (1967). lntracellular transport of secretory proteins in the pancreatic exocrine cell. 11. Transport to condensing vacuoles and zymogen granules. J . cell B i d . 34, 597-615.

200

DOUGLAS E. CHANDLER

Jost, P., Brooks, U . J., and Griffith, 0. H. (1973). Fluidity of phospholipid bilayers and membranes after exposure to osmium tetroxide and glutaraldehyde. J . Mid. B i d . 76, 3 13-3 18. Kalderon, N., and Gilula, N. B. (1979). Membrane events involved in myoblast fusion. J . Cell B i d . 81,411-425. Kanaseki, T., and Kadota, K. (1969). The “vesicle in a basket.” A morphological study of the coated vesicle isolated from the nerve ending of guinea pig brain, with special reference to the mechanism of membrane movements. J . Cell Biol. 42, 202-220. Kim, S . K., Nasjleti, C. E., and Han, S . S . (1972). The secretion processes in mucous and serous secretory cell of the rat sublingual gland. J . Ultrastrucr. Res. 38, 371-389. Kirchhausen, T., and Harrison, S. C. (1981). Protein organization in clathrin trimers. Cell 23,755-761. Knutton, S. (1979). Studies of membrane fusion. 111. Fusion of erythrocytes with polyethylene glycol. J . Cell Sci. 36, 61-72. Lagunoff, D. (1973). Membrane fusion during mast cell secretion. J . Cell Biol. 57, 252-259. Lawson, D., Raff, M. C., Gomperts, B. D., Fewtrell, C., and Gilula, N. B. (1977). Molecular events during membrane fusion. A study of exocytosis in rat peritoneal mast cells. J . Cell Biol. 72, 242-259. Lefort-tran, M., Aufderheide, K., Pouphile. M., Rossignol, M.. and Beisson, J. (1981). Control of exocytic processes: Cytological and physiological studies of trichocyst mutants in Paramecium tetraurelia. J . Cell Biol. 88, 301-31 I . Lingg, G., Fischer-Colbrie, R., Schmidt, W., and Winkler, H. (1983). Exposure of an antigen of chromaffin granules on the cell surface during exocytosis. Nature (London) 301,610-61 I . McIntyre, J. A., Gilula, N. B., and Karnovsky, M. J. (1974). Cryoprotectant induced redistribution of intramembranous particles in mouse lymphocytes. J . Cell Biol. 60, 192-203. Miller, T. M., and Heuser, J. E. (1984). Endocytosis of synaptic vesicle membrane at the frog neuromuscular junction. J . Cell Biol. 98, 685-698. Neher, E., and Marty, A. (1982). Discrete changes of cell membrane capacitance observed under conditions of enhanced secretion in bovine adrenal chromaffin cells. Pvoc. Narl. Acad. Sci. U . S . A . 79, 6712-6716. Neutra, M. R., and Schaeffer, S. R. (1977). Membrane interactions between adjacent mucous secretion granules. J . Cel1 B i d . 74, 983-991. Olbricht, K., Plattner, H., and Matt, H. (1984). Synchronous exocytosis in Paramecium cells. 11. Intramembranous changes analyzed by freeze-fracturing. Exp. Cell Res. 151, 14-20, Orci, L . , Perrelet, A., and Friend, D. S. (1977). Freeze-fracture of membrane fusions during exocytosis in pancreatic p-cells. J . Cell Biol. 75, 23-30. Orci, L., Amherdt, M., Montesano, R., Vassalli, P., and Perrelet, A. (1981). Topology of morphologically detectable protein and cholesterol in membranes of polypeptide-secreting cells. Philos. Truns. R . SOC.London, Ser. B 296, 47-54. Ornberg, R. L., and Reese, T. S. (1981). Beginning of exocytosis captured by rapid-freezing of Limulrts amebocytes. J . Cell Biol. 90, 40-54. Palade, G . E. (1959). Functional changes in the structure of cell components. I n “Subcellular Particles” (T. Hayashi, ed.), pp. 64-80. Ronald Press, New York. Palade, G. E. (1975). Intracellular aspects of the process of protein synthesis. Science 189, 347-358. Palade, G . E., and Bruns, R. R. (1968). Structural modulations of plasmalemmal vesicles. J . Cell Biol. 37, 633-649.

6 . MEMBRANE FUSION DURING EXOCYTOSIS

201

Pastan, I., and Willingham, M. C. (1985). The pathway ofendocytosis. I n “Endocytosis” (I. Pastan and M. C. Willingham, eds.). pp. 1-44. Plenum, New York. Patzak. A., and Winkler, H. (1986). Exocytotic exposure and recycling of membrane antigens of chromaffin granules: Ultrastructural evaluation after immunolabeling. J . Cell E i ~ l 102, . 510-515. Pearse, B. M. F. (1976). Clathrin. A unique protein associated with intracellular transfer of membrane by coated vesicles. Proc. Nad Amd. Sci. U.S.A. 73, 1255-1259. Pearse, B. M. F. (1978). On the structural and functional components of coated vesicles. J. M o l . B i d . U6, 803-812. Peixoto de Menezes. A.. and Pinto da Silva, P. (1978). Freeze fracture observations of the lactating rat mammary gland. J . Cell Biol. 76, 767-778. Perry, M. M., and Gilbert, A. B. (1979). Yolk transport in the ovarian follicle of the hen (Gallus domesticits): Lipoprotein-like particles at the periphery of the oocyte in the rapid growth phase. J . Cell Sci. 39, 157-272. Pinto da Silva, P., and Nogueira. M. L. (1977). Membrane fusion during secretion. A hypothesis based on electron microscope observation of Phyfophthoru palmiuoru zoospores during encystment. J. Cell B i d . 73, 161-181. Pinto da Silva. P., Peixoto de Menezes, A.. and Mather, I. H . (1980). Structure and dynamics of the bovine milk fat globule membrane viewed by freeze fracture. Exp. Cell Res. 125, 127-139. Plattner, H. (1981). Membrane behaviour during exocytosis. Cell B i d . I n t . Rep. 5, 435459. Plattner, H., and Bachmann, L. (1982). Cryofixation. A tool in biological ultrastructural research. I n t . Rev. Cytol. 79, 237-304. Plattner, H., Miller, F., and Bachmann, L. (1973). Membrane specializations in the form of regular membrane-to-membrane attachment sites in Paramecium. A correlated freezeetching and ultrathin-sectioning analysis. J . Cell Sci. 13, 687-7 19. Plattner, H..Reichel, K., Matt, H . . Biesson, J . , Lefort-tran, M., and Pouphile, M. (1980). Genetic dissection of the final exocytosix steps in Puramec,ium terruurelia cells: Cytochemical determination of ca’+-ATPase activity over preformed exocytosis sites. J . Cell Sci. 46, 17-40. Pollard, H.B . . Pazoles. C. J., Creutz. C. E.. Scott, J . H., Zinder, O., and Hotchkiss. A. (1984).An osmotic mechanism for exocytosis from dissociated chromaffin cells. J . Biol. Chem. 259, 1114-1121. Quirk, S . J.. Ahkong, Q. F., Botham, G. M.. Vos. J., and Lucy, J. A. (1978). Membrane proteins in human erythrocytes during cell fusion induced by oleoglycerol. Biochrm. J. 176, 159-167. Rand, R. P.. Reese, T. S., and Miller, R . G. (1981). Phospholipid bilayer deformations associated with interbilayer contact and fusion. Nature (London) 293, 237-238. Rasmussen, H.(1981). “Calcium and Cyclic AMP as Synarchic Messengers.” Wiley, New York. Rubin, R. P., Weiss, G. B., and Putney, J. W., Jr. (1985). “Calcium in Biological Systems.” Plenum, New York. Satir, B., and Oberg, S. G. (1978). Paramecium fusion rosettes: Possible function as calcium gates. Science 199, 536-538. Satir, B., Schooley, C., and Satir, P. (1973). Membrane fusion in a model system. Mucocyst secretion in Tefrahymenu. J. Cell Biol. 56, 153-176. Schlossman, D. M., Schmid, S. L., Braell, W. A., and Rothman, J. E. (1984). An enzyme that removes clathrin coats: Purification of an uncoating ATPase. J . CrN B i d . 99,723733.

202

DOUGLAS E. CHANDLER

Schmid, S . L., Braell, W. A., Schlossman, D. M., and Rothman. J. E. (1984). A role for clathrin light chains in the recognition of clathrin by "uncoating" ATPase. Nainre (London) 311, 228-23 I . Schmidt, W.. Patzak, A., Lingg, G., Winkler. H.. and Plattner, H. (1983). Membrane events in adrenal chromaffin cells during exocytosis: A freeze-etching analysis after rapid cryofixation. Eur. J . Cell Eiol. 32, 31-37. Specian, R. D., and Neutra. M. R. (1980). Mechanism of rapid mucus secretion in goblet cells stimulated by acetylcholine. J . Cell Eiol. 85, 626-640. Stenberg, P. E., McEver, R. P., Shuman, M. A,, Jacques, Y . V . , and Bainton. D. F. (1985). A platelet a-granule membrane protein (GMP-140) is expressed on the plasma membrane after activation. J . Cell Eiol. 101, 880-886. Takemura. R . , Stenberg, P. E., Bainton, D. F.. and Werb, Z. (1986). Rapid redistribution of clathrin onto macrophage plasma membranes in response to Fc receptor-ligand interaction during frustrated phagocytosis. J . Cell B i d . 102, 55-69. Tanaka, Y., De Camilli. P., and Meldolesi. J . (1980). Membrane interactions between secretion granules and plasmalemma in three exocrine glands. J . Cell Eiol. 84, 438-453. Tandler, B., and Poulsen, J. H. (1976). Fusion of the envelope of mucous droplets with the lurninal plasma membrane in acinar cells of the cat submandibular gland. J . Cell Eiol. 68, 775-78 I . Theodosis, D. T., Dreifuss, J . J., and Orci, L. (1978). A freeze-fracture study of membrane events during neurohypophysial secretion. J . Cell Biol. 78, 542-553. Tilney, L. G . , Clain, J. G., and Tilney, M. S. (1979). Membrane events in the acrosomal reaction of Limrrlrcs sperm. Membrane fusion, filament-membrane particle attachment, and the source and formation of new membrane surface. J . Cell B i d . 81, 229-253. Ungewickell, E., and Branton, D. (1981). Assembly units of clathrin coats. Nature (London) 289, 420-422. Ungewickell, E., and Branton, D. (1982). Triskelions: The building blocks of clathrin coats. Trends Eiochem. Sci. 7, 358-361. Verkleij, A. J . (1984). Lipidic intramembrane particles. Eiochim. Eiophys. Actrr 779, 43-63. Verkleij, A. J., Mombers, C., Gerritsen, W. J., Leunissen-Bijvelt, L., and Cullis, P. R. (1979). Fusion of phospholipid vesicles in association with the appearance of lipidic particles as visualized by freeze fracturing. Eiochirn. Eiophys. Acia 555, 358-361. Verkleij, A. J . , Van Echteld, C. J. A., Gerritsen, W. J., Cullis, P. R., and De Kruijff, B. (1980). The lipidic particle as an intermediate structure in membrane fusion processes and bilayer to hexagonal H,,transitions. Eiochim. Eiophys. Acia 600, 620-624. Von Wedel, R. J., Carlson, S . S . , and Kelly, R. B. (1981). Transfer of synaptic vesicle antigens to the presynaptic plasma membrane during exocytosis. Proc. Nail. Acad. Sci. U . S . A . 78, 1014-1018. Whitaker, M., and Zimmerberg, J. (1987). Inhibition of secretory granule discharge during exocytosis in sea urchin eggs by polymer solutions. J . Physiol. (London) 389, 527-539. Wileman, T., Harding, C., and Stahl, P. (1985). Receptor-mediated endocytosis. Biochem. J . 232, 1-14. Zerban, H., and Franke, W. W. (1978). Milk fat globule membranes devoid of intramembranous particles. Cell B i d . In?. Rep. 2, 87-98. Zimmerberg, J., and Whitaker, M. (1985). Irreversible swelling of secretory granules during exocytosis caused by calcium. Nature (London) 315, 581-584. Zimmerberg, J., Sardet, C., and Epel, D. (1985). Exocytosis of sea urchin egg cortical vesicles in vitro is retarded by hyperosmotic sucrose: Kinetics of fusion monitored by quantitative light-scattering microscopy. J . Cell Eiol. 101, 2398-2410.

CURRENT TOPICS I N M E M B R A N E S A N D TRANSPORT. VOLUME 32

Chapter 7

Osmotic Effects in Membrane Fusion during Exocytosis KEITH W . BROCKLEHURST AND HARVEY B . POLLARD Luhorutory of Cell Biology und Genc4c.s Nutionul Institrite of’Diribetes, Digrstiut,mu’ Kidney Diseases National Institutes oj‘Heulth Betliesdii, Muryland 20892

I. II.

111.

IV.

V.

Introduction Osmotic I-’roperties of I s i h t c d Secretory Ciranule\ A . Chromaffin Granules U . Other Scci-ctory Ur;inulc.s Osmotic EfVects in Secretion from I n t x t <’ell\ A . Chromaftin Cells H. Other Cell ‘Types Osmotic Effects in Secretion From I’ermc;ihili/ed C‘hrornaftin Cells Conclusions

References

I. INTRODUCTION Secretory cells release the contents of their secretory granules into the extracellular environment by the process of exocytosis, which culminates in the fusion of the granule membrane with the plasma membrane and the liberation of secretory product. The molecular events associated with this membrane fusion are far from understood as indeed are most aspects of the exocytotic process. One hypothesis that has been proposed to explain the membrane fusion event in exocytosis is the “chemiosmotic hypothesis of secretion” (Pollard et al., 1979a). The original chemiosmotic hypothesis was developed by Mitchell (1961) to explain the coupling of mitochondria1 electron transport to ATP synthesis. The term was subsequently applied to secretion after the discovery of proton-translocating ATPases in the membranes of certain se203 Copyright t’ IYSX hy Acadernlc Pre\r. Inc. All rights of reproduction in any form rererved.

204

KEITH W. BROCKLEHURST AND HARVEY 6.POLLARD

cretory vesicles. These enzymes utilize the energy released on ATP hydrolysis, in the presence of Mg2+,to translocate protons into the vesicle interior, thus establishing an electrochemical proton gradient across the vesicle membrane. Pollard et al.’s proposal was that the resulting chemiosmotic properties of secretory vesicles might play an important role in the mechanism of exocytosis. The chemiosmotic hypothesis of secretion has received a great deal of attention. This chapter aims to assess this hypothesis in the light of the properties exhibited by isolated secretory granules and by intact and permeabilized secretory cells. II. OSMOTIC PROPERTIES OF ISOLATED SECRETORY GRANULES A. Chromaffin Granules

1. CHARACTERIZATION OF THE PROTON-TRANSLOCATING ATPASE

The discovery of a proton-translocating ATPase in secretory granules of adrenal medullary chromaffin cells resulted from studies of the mechanism by which the chromaffin granule accumulates large amounts of catecholamine that are destined for secretion (for review, see Pollard et al., 1985). This work led to the conclusion that the ATPase of the chromaffin granule pumped protons into the intragranular space and that the resulting proton gradient across the membrane of the granule was responsible for coupling ATP hydrolysis to catecholamine uptake. The ATPase associated with chromafh granule membranes was found to be similar to the mitochondrial F,Fo-ATPase (Apps and Schatz, 1979). Others, however, suggested that the ATPase activity in chromaffin granules was due to mitochondrial contamination and that a separate anionsensitive ATPase activity was associated with the granule membrane (Cidon and Nelson, 1983; Cidon et al., 1983). More recently Percy et al. (1985) have shown that granule membrane preparations contain, in addition to small amounts of mitochondrial F1-ATPase, two ATPase activities termed ATPase I and ATPase 11, which can be separated and which exhibit different inhibitor and substrate specificities. ATPase I has an M , of 400,0OO,is inhibited by trialkyltin, quercetin, and alkylating agents, and hydrolyzes both ATP and ITP; ATPase 11, which has an M, of 140,000, is inhibited by vanadate and is specific for ATP. Evidence that ATPase I is the proton-translocating ATPase comes from the observation that proton translocation by granule ghosts is supported by the hydrolysis of ATP or ITP and is inhibited by quercetin or alkylating agents, but not by vana-

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

205

date. ATPase I was found to contain up to five different types of subunits that include a low molecular weight hydrophobic dicyclohexylcarbodiimide (DCCDbreactive polypeptide which may form the protonconducting channel. The subunits of ATPase I were unrelated to those of the mitochondria1 FIFo-ATPaseas judged by size and immunoreactivity. 2. THE ELECTROCHEMICAL PROTON GRADIENT The electrochemical proton gradient (A&+ ) across the granule membrane established by the proton-translocating ATPase has two components: a pH gradient (ApH) with the granule interior having a pH lower than the outside, and a transmembrane electrical potential (A$), with the granule interior positive with respect to the outside. Measurements of the pH of the chromaffin granule interior, obtained from the distribution of methylamine across the granule membrane and by ) ' P N M R , have shown that the resting intragranular pH is approximately 5.7 and is relatively independent of the extragranular pH (Casey ct ul., 1977; Johnson and Scarpa, 1976a; Njus et a l. , 1978: Pollard et d., 1976b. 1979b). Addition of ATP in the presence of a permeant anion caused the pH to drop by about 1979b). Lipid-soluble ions 0.3 units (Casey et al., 1977; Pollard rt d., (Holz, 1978, 1979; Johnson and Scarpa. 1979; Pollard et ul., 1976b) and a voltage-sensitive dye (Ogawa and Inouye, 1979: Salama et ul., 1980) have been used to measure the A$ of isolated chromaffin granules. With permeant anions absent the addition of ATP to granules (whose resting potential at pH 6.9 is negative inside) generated a membrane potential of +50 to +70 mV inside positive. In the presence of a permeant anion, ATP induces a proton influx into the granules that is electrically neutralized by anion influx. The proton/ anion influx results in a drop in intragranular pH (Casey et ul., 1977). but causes only a small change in membrane potential (Johnson and Scarpa, 1979). Similarly, when granules are incubated in the absence of a permeant anion, ATP induces an inside positive membrane potential without a concomitant change in the intragranular pH (Casey et ul., 1977; Holz, 1978: Johnson and Scarpa, 1979). The intragranular pH does not change under these conditions, presumably because relatively few positive charges enter the granule before the A$ electrically limits further proton entry, and because the granule interior has a large buffering capacity. Chrornaffin granule ghosts exhibit similar chemiosmotic properties, although their internal buffering capacity is much reduced (Apps ef a l . , 1980; Johnson ef al., 1979, 1981; Knoth et ul., 1980; Phillips and Allison, 1978).

206

KEITH W. BROCKLEHURST AND HARVEY 6.POLLARD

3. IONICPERMEABILITIES A N D OSMOTIC LYSIS Chromaffin granules and granule ghosts exhibit a relative impermeability to cations (e.g., Nat, K+,Mg2+, Ca2+,and H+) and a greater but selective permeability to anions (Johnson and Scarpa, 1976b; Phillips, 1977). Isolated chromaffin granules incubated under isotonic conditions with ATP release their contents in the presence of C1- or other permeant anions (Lishajko, 1969; Oka et al., 1965; Poisner and Trifaro, 1967; Pollard et al., 1976a,b). Release could be inhibited by increasing the osmotic strength of the medium with either salt or sucrose (Casey et al., 1976; Pollard et al., 1976b). The addition of ATP to chromaffin granules resulted in the entry of 36CI- (Pazoles and Pollard, 1978), and CI-,ATP-induced granule lysis was inhibited by mitochondria1 uncouplers (Casey et al., 1976). The granule membrane ATPase may therefore catalyze the inward translocation of protons as counterions to C1-, thus raising the osmotic content of the granule and causing lysis (Casey et al., 1976). One problem with this proposed mechanism of ATP-induced granule lysis is that nonhydrolyzable ATP analogs such as App(NH)p caused changes in the transmembrane electrical potential of isolated chromaffin granules (Pollard et al., 1976b) and supported granule lysis (Hoffman et al., 1976). The V,,, for lysis by the analog was approximately half that for ATP and the apparent KIhfor App(NH)p was approximately four times greater than that for ATP (Hoffman et al., 1976). Also, ADP was about half as effective as ATP in inducing both changes in A$ (Pollard rr a / . , 1976b) and lysis (Izumi et al., 1977), yet did not cause any detectable change in intragranular pH as measured by )‘P NMR (Pollard et al., 1979b). Although granule preparations may contain myokinase activity, a situation that could explain the ADP effects, the ATP-induced changes in A$ of granules were also detected at 2°C (Pollard ct ul., 1976b). a temperature at which enzyme reactions are slow. It is thus possible that some granule functions may be mediated by ATP binding rather than hydrolysis, or that only very small amounts of ATP hydrolysis may be necessary to generate changes in transmembrane A+. In the exocytotic process, after secretory granules have fused with the plasma membrane of the cell forming a bilayer diaphragm (alternatively named, “semifusion”; Duzgunes, 1985), the barrier separating the granule interior from the extracellular medium undergoes breakage, or fission (alternatively named “fusion”; See Chandler, Chapter 6, this volume). This fission step has been modeled in vitro by studying the ATP-driven osmotic lysis of chromaffin granules in the presence of permeant anions, on the grounds that granule components are involved in both events and that both events seem to involve a break in the granule membrane. Since

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

207

permeant anion concentration gradients exist across most cell membranes, they should also exist across the membrane barrier that separates the granule interior from the extracellular fluid at the time of fusion (or “semifusion”). For example, the extracellular CI- concentration is approximately 120 mM, whereas the intragranular CI- concentration has been estimated at 15-40 mM, approximately the same as the cytoplasmic CI- concentration (Pazoles and Pollard, 1978). The first indication that anion entry may be of importance to granule lysis and exocytosis came from experiments in which the potassium ionophore valinomycin induced granule lysis in the absence of ATP (Dolais-Kitabgi and Perlman, 1975). This was interpreted to mean that valinomycin allowed the electrogenic entry of K + into the granule because the intragranular K + concentration was low (Johnson and Scarpa, 1976b). When the K + salt of an impermeant anion was used in this system, little K t entry resulted because of the rapid buildup of a K tdiffusion potential, positive inside. If the K’ salt used were that of a permeant anion, however, anion entry accompanied K + entry; this in turn resulted in electroneutral cotransport of ions and in granule lysis. Based on these experiments the following anion permeability series was defined: SCN-, I-, Br- > CI- > acetate, F-, isethionate. This series is similar to that obtained for the passive anion permeabilities of chromaffin granule ghosts, i.e., SCN- > I- > trichloroacetate > Br- > CI- > SO:> acetate, HCO;, F-, PO:- (Phillips, 1977). Certain anions supported ATP-induced granule lysis better than others: SCN-, I-, and Br- were better than C1-; in turn, C1- was better than acetate, PO:- or isethionate (Casey et al., 1976; Hoffman er al., 1976: Taugner, 1972). This led Casey et ul. (1976) to suggest that ATP-driven proton translocation was analogous to valinomycin-mediated K’ movement in inducing net uptake of permeant anions. Anion transport across the granule membrane thus appeared to be controlled by a selective mechanism. Further evidence was provided by observations that the rate of ATP-induced granule lysis was a saturable function of Cl-concentration and that isethionate, an impermeant anion, acted as a competitive inhibitor of lysis with respect to Cl(Pazoles and Pollard, 1978). A biophysical model has been constructed to describe the behavior of isolated chromaffin granules on undergoing CI-,ATP-induced lysis (Creutz and Pollard, 1980). The model makes the assumption that the granule release reaction represents osmotic lysis due to the ATPase-dependent influx of protons and osmotically active CI- ions. The consequences of this influx were predicted from osmotic fragility curves determined by suspending granules in hypotonic media, and turbidity measurements of granule suspensions were used to fit the model parame-

208

KEITH W. BROCKLEHURST AND HARVEY 8. POLLARD

ters. The model successfully describes the time course, C1- dependence, ATP dependence, and the osmotic strength suppression of the release reaction, as measured either by turbidity changes or epinephrine release. In the erythrocyte membrane, anion transport in the form of anion exchange is mediated by the membrane protein band 111. This transport system is competitively inhibited by several drugs including 4-acetamido4’-isothiocyanostilbene 2,2’-disulfonic acid (SITS), probenecid, and pyridoxal phosphate (Cabantchik ef al., 19781, all of which are relatively impermeant aromatic anions. These compounds also block ATP-induced granule lysis in the presence of CI-. This inhibition is competitive with C1- and is analogous to the competitive effect by isethionate on CI-supported release (Pazoles and Pollard, 1978). The Ki values of SITS, probenecid, and pyridoxal phosphate for granule lysis were 40 pM, 125 pM,and 3.6 mM, respectively, values that are similar to those reported for the erythrocyte system. SITS was also found to inhibit ATP-induced uptake of 3hC1-into granules. It was therefore concluded that the mechanism of CI-,ATP-induced lysis of chromaffin granules probably involved anion entry through a site functionally analagous to band 111 in erythrocytes. The anion transport blockers have since been used to characterize different mechanisms of anion transport in chromaffin granules. Three groups of anions have been distinguished (Pazoles, 1982; Pazoles et al., 1980), based on their ability to support granule lysis under different conditions. Class I consists of CI-, Br-, and I-, all relatively permeant anions; class I1 includes PO:- and isethionate, which are impermeant anions; and class 111 includes F-, SO:- and acetate, the permeabilities of which depend on experimental conditions. Granule lysis mediated by the ionophore nigericin or by NHd is independent of anion permeation, as all three anion classes were permissive and lysis was not inhibited by anion transport-blocking drugs. Nigericin causes granule lysis in the presence of K + by allowing the exchange of intragranular osmotically inactive protons for medium K+,whereas N&+ mediates lysis by permeating the granule membrane as NH1 and subsequently becoming protonated and trapped within the granule. Both valinomycin,K+-induced lysis and ATP-induced lysis were dependent on the presence of class I permeant anions, but the former was not, while the latter was inhibited by the anion transport blockers. Although class I1 impermeant anions inhibited ATP-induced lysis in the presence of class 1 anions, this was not true for valinomycin,K+-induced release. These findings suggest the existence of two distinct types of anion transport site with similar anion selectivity, but with differential sensitivity to inhibitors.

7.

OSMOTIC EFFECTS IN MEMBRANE FUSION

209

Granule lysis was also induced by imposing an artificially induced, inwardly directed proton gradient across the granule membrane by incubating granules in a medium of low pH (pH < 5.5). As with ATP-induced lysis, class I anions stimulated lysis in a blocker-sensitive manner, while class I1 anions were ineffective in supporting lysis, but inhibited lysis in the presence of class I anions. Thus, anion permeation induced by low external pH and ATP appeared to involve the same anion transport site. Class 111 anions also proved to be effective at inducing lysis at low external pH, but this effect was not blocker sensitive. It is possible that this class of anions permeates the granule membrane as an uncharged protonated species and then becomes deprotonated and trapped within the granule. A possible analogous situation involves the permeant anion SCN-, which can support granule lysis whenever anion permeability is necessary, as in lysis induced by valinomycin,K+, ATP, or low external pH. However SCN--supported lysis with ATP, or at low pH, is not blocker sensitive, and so it is possible that the lipophilic nature of SCNallows it to permeate the granule membrane without the need of a specific transport site. These studies led the authors to conclude that only CI-, Br-, and I - use the blocker-sensitive anion transport site and that this site is only used in conjunction with proton entry driven by either ATP hydrolysis or an artificially imposed proton gradient across the granule membrane. It is possible that the proton transport site and the blocker-sensitive anion transport site are intimately releated and may be part of a single macromolecular complex that includes the ATPase. This idea is supported by the finding that chromaffin granule ATPase activity, even in detergentsolubilized form, is potentiated by permeant but not impermeant anions, and is inhibited by SITS and pyridoxal phosphate (Pazoles et al., 1980). 6. Other Secretory Granules

The general properties of chromaffin granules appear to be shared by other types of secretory vesicles. For example, the interior of insulin secretory granules of pancreatic /3 cells may be more acidic than the cytosol in as much as granule-enriched subcellular fractions from pancreatic islets have been shown to accumulate the fluorescent dye 9-aminoacridine (Abrahamsson and Gylfe, 1980). Insulin granules can accumulate the base 5-hydroxytryptamine in vivo (Ekholm et ul., 1971; Hellman ef ul., 1972), and, when purified from a transplantable rat insulinoma, such granules possess an inwardly directed proton-translocating ATPase activity (Hutton and Peshavaria. 1982). The latter is not simply due to mitochondrial contamination as judged by the distribution of marker proteins

21 0

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

on density gradients and the sensitivity of the enzyme to drugs. On the other hand, the properties of the insulin granule proton-translocating ATPase activity were indistinguishable from those of the activity present in the chromaffin granule membrane, and the two types of granules showed similar levels of activity. The resulting chemiosmotic properties of the insulin granule were also very similar to those of the chromaffin granule (Hutton, 1982). The capacity of the insulin granule ATPase for ATP hydrolysis far exceeded the energetic requirement for amine transport, as the amine content of insulin granules is much lower than that of chromaffin granules. However, the intragranular pH was found to correspond to the optimal condition for crystallization of zinc-insulin hexamers (Hutton. 1982). Purified neurosecretory vesicles from bovine neurohypophyses also exhibited chemiosmotic properties similar to those of chromaffin granules and insulin granules; again, this is consistent with the existence of an inwardly directed proton-translocating ATPase (Russell, 1984; Russell and Holz, 1981; Scherman et al., 1982). Indeed, the neurosecretory vesicles appear to possess two ATPase activities with properties similar to those of the chromaffin granule enzymes reported by Percy el al. (1989, only one of which appears to be responsible for proton translocation across the vesicle membrane (Russell, 1984). In the case of the neurosecretory vesicle the acidic intragranular pH corresponds to the pH of maximal stability of the hormone-neurophysin complex. In contrast, a recent paper by Saermark et al. (1986) suggests that secretory vesicles isolated from guinea pig neurohypophysis do not contain a proton pump, even though coated microvesicles from the same tissue do possess such an activity. Accordingly, the secretory vesicles in their preparation did not exhibit some of the chemiosmotic properties reported for bovine neurosecretory vesicles. Isolated secretory vesicles from the rat parotid gland also lack a proton-translocating ATPase (Arvan et al., 1984), the internal pH of these granules being approximately 6.8. However, the latter are osmotically responsive and exhibit selective permeability to anions, as certain anions can cause lysis. Secretory granules isolated from porcine anterior pituitary have an anion-sensitive ATPase activity (Lorenson et al., 1981), and dense granules from the bovine gland apparently contain a proton-translocating ATPase responsible for the generation of ApH and A$ gradients across the granule membrane (Carty et al., 1982). The existence of a transmembrane ApH in the dense granules of porcine platelets has also been reported (Johnson et al., 1978), with the inside of the granules being acidic. This observation was later confirmed when it was shown that the dense granules transported 5-hydroxytryptamine in response to the ApH across the granule

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

21 1

membrane; similar results were obtained with ghosts prepared from the dense granules (Wilkins and Salganicoff, 1981). It was also shown that an ATPase activity was present in the ghost membrane and that inhibition of the ATPase led to a corresponding decrease in 5-hydroxytryptamine transport. More recent findings have provided evidence for a protontranslocating ATPase in the platelet granule membrane (Carty et ul., 1981; Dean et ul., 1984). They have also shown 5-hydroxytryptamine transport across the membrane is responsive to both ApH and the A$ across the granule membrane (Carty et ul., 1981). In another amine storage granule, the synaptic vesicle, a pH gradient across the vesicle membrane induced by an ATPase activity appears to be the main driving force for catecholamine uptake and storage (Toll and Howard, 1978). Thus, the secretory granules of many different cell types exhibit very similar chemiosmotic properties. The maintenance of a AGH+ across the granule membrane may simply serve specific functions such iis amine transport in amine storage granules. or the provision of an optimal chemical environment for the storage and/or processing of the secretory product in peptide secretory granules. However, chemiosmotic events may also play a more general role in the exocytotic process. This possibility will now be analyzed on the basis of secretion studies in intact and permeablized cells.

111.

OSMOTIC EFFECTS IN SECRETION FROM INTACT CELLS

A. Chromaffin Cells

If lysis of chromaffin granules represents the membrane fission step in exocytosis, then certain predictions can be made about the exocytotic process in chromaffin cells on the basis of experimental results obtained with isolated granules. When secretory granules form a fusion complex with the plasma membrane of the cell then, according to the chemiosmotic model (Pollard et al., 1979a), an anion transport site derived from the granule membrane might span the membrane complex which now separates the intragranular space from the extracellular medium. Anions would then enter the granule, down their concentration gradient. When coupled to the proton-translocating ATPase in the granule membrane, this anion transport could then provoke fission of the fused membrane complex by osmotic lysis, It is relevant to note that the fusion of liposomes with planar lipid bilayers in an artificial system requires an osmotic gradient across the vesicle membrane to allow the entry of HzO, leading to vesicle swelling (Cohen et ul., 1980, 1982; Zimmerberg et a l . . 1980). In

21 2

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

intact cells, therefore, exocytosis would be inhibited by (1) anion transport-blocking drugs, (2) the absence of specific permeant anions from the bathing medium, (3) hypertonic media, and (4) protonophores. These postulates have been tested by Pollard ef al. (1984), who used bovine chromaffin cells that had been freshly isolated by collagenase digestion of adrenal medullae. Relatively high concentrations of the anion transport blockers probenecid and pyridoxal phosphate were needed to inhibit veratridine induced secretion. However, in contrast to the effect on the chromaffin granule lysis reaction, SITS and other stilbene disulfonates were inactive. As for the second prediction, epinephrine secretion was found to occur only in the presence of a permeant anion such as C1-; veratridine-induced secretion was inhibited when medium NaCl was replaced by sodium isethionate. Veratridine supported epinephrine release according to the activity series Br-, I - , NO; > methyl sulfate, SCN- > Cl- > acetate >> isethionate. A somewhat similar series was obtained when the Ca2+ ionophore A23187 was used as secretogogue. However, although the anion series for ATP-dependent granule lysis was qualitatively comparable, the quantitative parallelism between anion dependence of chemiosmotic granule lysis and secretion from intact cells was poor. Third, increasing the osmotic strength of the medium with either NaCl or sucrose led to suppression of veratridine- or acetylcholine-induced epinephrine secretion. Quantitatively, this suppression was almost identical with the osmotic suppression of C1-, ATP-induced lysis of isolated chromaffin granules. Finally, the uncoupler carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) was found to inhibit veratridine-induced secretion from chromaffin cells in a concentration range previously shown to block CI-,ATP-driven granule lysis. Other agents that act on mitochondria1 energy metabolism, such as CN-, azide, and rotenone, however, also inhibited secretion, whereas they were without effect on Cl-,ATP-driven granule lysis. The inhibition of epinephrine secretion by FCCP therefore cannot be attributed unequivocally to an effect on granules. Thus while the properties of secretion from chromaffin cells resembled those of Cl-, ATP-induced epinephrine release from isolated secretory granules, a number of discrepancies were found. The effects of changes in medium osmolality on the stability and function of cultured bovine chromaffin cells have also been investigated in detail (Hampton and Holz, 1983). Measurements of cell volume in media of differing osmolality showed that the cells behaved as osmometers, with the intracellular osmolality equilibrating rapidly with that of the external medium. Hyperosmotic solutions inhibited both nicotinic agonist-stimulated and K+-stimulated catecholamine secretion, but were without effect

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

213

on nicotinic agonist-induced Ca'+ uptake and only weakly inhibited K+induced Ca2+uptake. Thus, increased osmolality inhibited both nicotinic agonist- and K+-induced secretion at a step distal to Ca2+entry into the cell. This study also showed that the osmotic fragility of chromaffin granules in uitro is greater than that of granules in situ. Reducing the osmolality of the medium in which isolated chromaffin granules were suspended from 310 to 210 mOs caused the release of more than 75% of their catecholamine content. By contrast, reduction of the medium osmolality to below 165 mOs was necessary before substantial intracellular granular lysis became apparent. Thus what proves to be true of isolated granules may not be true of granules in their intracellular environment. Cultured bovine chromaffin cells have also been used to study the possible involvement in exocytosis of a proton electrochemical gradient across the granule membrane (Holz et al., 1983).Although treatment with NH;, methylamine ions, or nigericin reduced the pH gradient across the granule membrane. as judged by cell/medium methylamine concentration ratios or by 71P-NMRmeasurements of intragranular ATP, these treatments were without effect on veratridine-induced catecholamine secretion. Moreover, nigericin did not alter secretion induced by nicotinic receptor stimulation. FCCP, at a concentration (1 pM) that appeared to uncouple mitochondria within chromaffin cells, did not inhibit carbacholinduced secretion. Finally, dicyclohexylcarbodiimide (DCCD) did not alter K+-induced catecholamine release under conditions that blocked coupled transport of [7H]norepinephrine into chromaffin granules. DCCD is an irreversible inhibitor of proton-translocating ATPases and inhibits the chromaffin granule membrane ATPase, the ATP-induced increase in granule membrane potential, and ATP-induced catecholamine uptake in isolated chromaffin granules (Bashford et c i l . , 1976). The authors concluded that these results make it unlikely that a proton-translocating ATPase, extracellular anions, and an electrical potential across the granule membrane are involved in the mechanism of exocytosis. Although these data are fairly convincing, they are subject to the following criticisms: first, NHB, methylamine ions, and nigericin reduce the ApH across the chromaffin granule membrane, but do not reduce the A$. Second, since FCCP is without effect on the internal pH of isolated granules, a pH gradient may still exist between the granules and the cytosol (or extracellular medium). Third, the effect of DCCD was judged by its effect on coupled [3H]norepinephrine uptake into granules isolated from DCCD-treated cells. However, part of the inhbition of ATP-induced catecholamine uptake may be attributable to a direct effect on the catecholamine transporter (Schuldiner et al., 1978). Also, as the authors point out, under certain conditions methylamine. nigericin, and FCCP would all inhibit catecholamine secretion: high methylamine concentrations inhib-

214

KEITH W. BROCKLEHURST AND HARVEY 6.POLLARD

ited secretion but also reduced the cytosolic ATP concentration; increasing the nigericin concentration from 1 to 10 pM caused partial inhibition of secretion, but this was not correlated with a substantial change in the intragranular pH; and FCCP concentrations greater than 1 p M inhibited secretion but were larger than necessary to uncouple intracellular mitochondria maximally. Therefore, inhibition of secretion under these conditions may have resulted from effects other than on the electrochemical proton gradient. 6. Other Cell Types

The chemiosmotic hypothesis as applied to secretion from intact cells has also been investigated in cell types other than the chromaffin cell. Indeed, the predictions of the hypothesis were first tested on human platelets (Pollard et al., 1977).The secretion of 5-hydroxytryptamine from human platelets, stimulated either by thrombin or A23187, was inhibited by the anion transport blockers SITS, pyridoxal phosphate, probenecid, and suramin. However, when either sucrose or sodium isethionate replaced NaCl in the medium, secretion was not inhibited. Rather reduction in the medium pH inhibited secretion, and inhibition by the anion transport blockers was competitive with respect to OH-. The Ki values for SITS, probenecid, pyridoxal phosphate, and suramin were 28, 335, 56, and 0.9 p M , respectively, with respect to OH-. Thus the activity series for inhibition of platelet secretion was suramin >> SITS, pyridoxal phosphate >> probenecid with isethionate being inactive, whereas the corresponding series for inhibition of chromaffin granule lysis was SITS > probenecid > suramin > pyridoxal phosphate >> isethionate. Conceivably, the differences in activity of the drugs in the two systems relate to anion specificity. 5-Hydroxytryptamine release from platelets was suppressed by increasing extracellular osmotic strength, and the relationship between suppression and osmotic strength was quantitatively similar to that observed for the inhibition of chromaffin granule lysis. FCCP also inhibited platelet secretion. Pollard et al. (1977) therefore concluded that platelets and chromaffin granules were similar in terms of the osmotic basis of the release reaction, but differed with respect to the anions involved. The chemiosmotic hypothesis of secretion was also investigated in dissociated parathyroid cells (Brown et al., 1978) from which parathyroid hormone (PTH) secretion could be elicited by exposure to a low extracelMar Ca2+concentration (Brown et al., 1976). Both SITS and probenecid were found to inhibit PTH secretion almost completely, whereas replacement of NaCl by either sucrose or sodium isethionate caused a 70% inhibition of secretion. The inhibition of secretion by SITS and probene-

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

215

cid was competitive with respect to CI-, with each drug having a K , between 400 and 600 pM. Increasing the osmotic strength of the medium completely inhibited secretion. Various cation replacements for Na' had no effect on PTH release; FCCP blocked secretion. The observation that anion transport blockers inhibited PTH secretion almost totally, whereas omision of C1- or its replacement with isethionate only inhibited release by 70%. suggested the possibility that another anion beside CI- might also play a role. Lowering the medium pH inhibited PTH secretion, and probenecid was found to be a competitive inhibitor of release with respect to O H - . Thus, OH- might be another permeant anion in this system. Lysosomal enzyme release from human neutrophils stimulated with immune complexes was inhibited by SITS and 4,4'-diisothiocyano-2,2'stilbene disulfonic acid (DIDS), as was A23 187-induced release (Korchak et ul., 1980). Neither the nature of the permeant anion(s) nor the role of anion influx in degranulation was identified, however. Influx of CI-, OH-, or PO:- did not seem to be important. Permeant anions supported antigen-induced histamine release from human basophils in the following order: acetate > Br-, I- > CI- (Hook and Siraganian, 1981). lsethionate and SO:- did not support histamine release. SITS and probenecid did not inhibit IgE-mediated histamine release from these cells, and isosmotic solutions of several sugars were capable of supporting antigen-or antiIgE-induced histamine release in the presence of Ca2+.Increasing the osmolarity of the medium by adding more NaCI/KCI actually enhanced antigen- or anti-IgE-induced histamine release. These results do not support the hypothesis that exocytosis from basophils depends on anions or results from osmotic lysis. Finally, predictions based on the chemiosmotic hypothesis have been applied to the process of insulin secretion from pancreatic islet cells. Insulin release induced by glucose or a-ketoisocaproate from isolated rat pancreatic islets was inhibited when extracellular CI- was replaced with isethionate or SO:-; when the extracellular osmotic strength was raised by the addition of sucrose; and when the islets were exposed to probenecid or DIDS (Orci and Malaisse, 1980; Pace and Smith, 1983; Somers ef a / . , 1980). The inhibition of glucose-induced insulin release by C1- substitution was associated with a small decrease in glucose oxidation but no significant change in glucose-stimulated net 45Ca2tuptake by the islets (Somers e f a / . , 1980). In the isolated perfused rat pancreas the isethionate- or sucrose-induced inhibition of glucose-stimulated insulin release was shown to be a rapid and rapidly reversible phenomenon (Somers P t a / . , 1980). However, the substitution of C1- by isethionate in this system inhibited the second phase of the glucose-induced secretory response more than the first phase; it also failed to inhibit the insulin response to gliclazide. Similarly, in perfused rat islets, CI- substitution by isethionate

216

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

did not inhibit the first phase of glucose-induced insulin secretion nor did it inhibit tolbutamide-stimulated secretion (Tamagawa and Henquin, 1983). Thus, although the chemiosmotic hypothesis for exocytosis may apply to the process of insulin secretion, further evidence is required to substantiate this mechanism. IV. OSMOTIC EFFECTS IN SECRETION FROM PERMEABILIZED CHROMAFFIN CELLS

In an effort to gain direct access to the secretory machinery of the cell and thus bypass the permeability barrier imposed by the plasma membrane, chromaffin cells have been permeabilized by subjecting them to high-voltage discharge. Catecholamine release from such cells can be induced by low Ca2+concentrations (in the low micromolar range) and requires ATP (Baker and Knight, 1978, 1980, 1981; Knight and Baker, 1982). The standard medium used to study catecholamine release from these cells contains glutamate as the principal anion. When glutamate was replaced with CI-, Ca2+-dependentrelease was inhibited. The effectiveness of different anions at inhibiting Ca2+-dependentrelease was in the order SCN- > Br- > CI- > acetate > glutamate, a sequence that follows the lyotropic series. Inhibitory anions may therefore bind to and disrupt some part of the release machinery which may regulate exocytosis in uiuo. Neither SITS nor DIDS (at concentrations of 0. I mM) had an effect on Caz+-evokedcatecholamine release. Glutamate (in the form of sodium glutamate) supported catecholamine release from intact chromaffin cells approximately 75% as well as CI- in the form of NaCl; glutamate also supported ATP-dependent chromaffin granule lysis, but significantly less well than C1- (Pollard ef al., 1984). Ca2+-activationof release from permeabilized cells was essentially normal in buffered isotonic sucrose. When the osmotic pressure of the medium was increased by raising the sucrose concentration, Ca2+-dependentrelease was inhibited. This observation may reflect the involvement of an osmoticially active step in the secretory process. The internal pH of chromaffin granules in electrically permeabilized cells was estimated to be 5.8, based on [*4C]methylamineuptake by the cells (Knight and Baker, 1985). Methylamine accumulation was greatly reduced in the presence of NH: or monensin (a protodcation exchanger), but was unaffected by the proton pump blocker trimethyltin or FCCP. SI4CN- accumulation was also used as a measurement of the granule membrane potential. The mean membrane potential was reduced in the absence of exogenous ATP and in the presence of trimethyltin or FCCP.

7. OSMOTIC EFFECTS IN MEMBRANE

FUSION

217

Neither the intragranular pH nor the granule membrane potential was affected by Ca?' over the range of free Ca2+concentrations that activate secretion, except when large amounts of catecholamine were released. In that case both the methylamine and SCN- spaces were reduced. This is consistent with accumulation of these compounds in secretory granules and their release on exocytosis. Exposure of leaky cells to NH: concentrations that alkalinize the granule core failed to block Ca2t-dependent release, although there was a small reduction in this release at NHd concentrations that greatly reduced the pH gradient between the vesicle core and the cytosol. Both Ca2+-dependent release and S14CN- accumulation were dependent on ATP, but the latter was more sensitive than the former. There were also differences in the nucleotide specificity of these two processes, CaZ'-dependent release being very specific for ATP, and SI4CN- accumulation being almost equally activated by ATP, GTP, UTP, and ITP. Ca2+Dependent release still occurred when the granule proton pump and SCN- accumulation were inhibited by trimethyltin, although high trimethyltin concentrations did cause a small reduction in Ca?'-dependent release. Ca?+-Dependent release also persisted at FCCP concentrations that collapsed the granule membrane potential; however, very high FCCP concentrations caused a modest inhibition of Ca2+-dependent release. There were no significant alterations in the Ca2+activation curve of catecholamine release under conditions where either the vesicle pH gradient or the membrane potential was collapsed. Ca?+-dependentrelease also took place when both the granule pH gradient and membrane potential were collapsed with NHd and trimethyltin or FCCP, respectively. Therefore, Ca?+-dependent release from high-voltage-permeabilized cells appears to be largely unaffected by the pH gradient across the secretory vesicle or its membrane potential. However, as the authors point out. this conclusion is heavily dependent on the interpretation of SI4CN- and [14C]methylaminespaces in terms of the membrane potential and internal pH of the secretory vesicles. Although contributions to these spaces from other compartments such as lysosomes cannot be ruled out, it is likely that chromaffin granules make a significant contribution to these spaces. As Cat+-dependent release persists with both methylamine and SCNspaces close to the 3H20 space, it would seem that the pH gradient and membrane potential of the chromaffin granule are not necessary for this process. However, the reduction in Ca?+-dependentrelease seen at high concentrations of agents used to collapse the pH gradient and membrane potential suggests that these two processes may play a small role in exocytosis. It must be remembered, however, that in permeabilized cells neither a membrane potential nor ion gradients exist across the plasma

218

KEITH W. BROCKLEHURST AND HARVEY 8. POLLARD

membrane. Therefore, the release event in these cells may not be representative of what occurs in intact cells. The possibility that an “osmotic event” plays a role in the exocytotic process has also been investigated with chromaffin cells that were permeabilized with the detergent digitonin. As is true of electrically permeabilized cells, digitonin-permeabilized cells require Ca2+and ATP for catecholamine secretion (Dunn and Holz, 1983; Wilson and Kirshner, 1983). Glutamate and acetate were most effective in supporting Ca2+dependent secretion in these cells, with Br- and CI- less so (Wilson and Kirshner, 1983). Sucrose and glycine support Ca2+-evokedcatecholamine release at about 80% of that seen with sodium acetate (Wilson and Kirshner, 1983); potassium isethionate also supports release (Holz and Senter, 1985). Evidence from studies of [3H]noradrenaline uptake into digitonin-permeabilized cells suggests that an ATP-induced proton electrochemical gradient exists across the membrane of chromaffin granules within these cells (Holz and Senter, 1985). The observation that digitonin-treated cells, unlike intact cells, undergo little or no cell shrinkage in solutions made hyperosmotic with potassium glutamate or various carbohydrates indicates that molecules as large as tetrasaccharides have free access to the cell interior. Also, experiments in which the cells are placed in hyposmotic solutions have shown that chromaffin granules inside digitonin-permeabilized cells have osmotic stability similar to that of intact cells (Holz and Senter, 1985). The inhibition by hyperosmotic solutions of catecholamine secretion from digitonin-permeabilized chromaffin cells has been investigated by Holz and Senter (1986). These cells were much more resistant to increases in medium osmolality due to sucrose or other carbohydrates than were intact chromaffin cells. On the other hand, the permeabilized cells proved to be more sensitive to hyperosmotic solutions of potassium glutamate and sodium isethionate than to hyperosmotic carbohydrate solutions. Increases in salt concentrations inhibited secretion even if osmolality was kept constant with sucrose. The inhibition of secretion from intact cells by hyperosmotic solutions may therefore be mediated by increased intracellular ionic concentrations resulting from cell shrinkage. The fact that the sensitivity of electrically permeabilized chromaffin cells to hyperosmotic solutions seems to fall between that of intact and digitoninpermeabilized cells may be attributed to the differences in permeability characteristics, the digitonin procedure apparently allowing a more rapid equilibration of sucrose across the plasma membrane. Thus, a component of the hyperosmotic inhibition of Ca2+-dependentsecretion from electrically permeabilized chromaffin cells could be due to cell shrinkage and increased intracellular ionic strength. However, this interpretation as-

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

219

surnes that chromaffin granules in digitonin-perrneabilized cells are irnperrneable to the osmoticants used and that significant granule shrinkage can occur under hyperosmotic conditions. While these findings do not rule out the possibility that an osmotically regulated process is involved in exocytosis, they place real constraints on such a process. Again, questions can be raised regarding the validity of using.permeabilized cells as a model system. In the course of membrane fusion during exocytosis in intact cells, ion gradients separate the granule interior from the extracellular medium at the site of fusion, and other ion gradients separate the granule interior from the cytosol. In permabilized cells, on the other hand, these two sets of ion gradients are essentially the same.

V.

CONCLUSIONS

The chemiosmotic hypothesis of secretion was formulated following the discovery of the chemiosmotic properties of isolated chromaffin granules. However, subsequent results obtained with intact and permeabilized cells revealed large differences between the predicted and observed secretory properties of these cells. An interesting finding to emerge from these studies was that isolated chromaffin granules display properties that differ from those exhibited by granules present in intact or perrneabilized cells. Perhaps granules undergo a “preparation catastrophe” during their isolation which causes them to express characteristics normally revealed only under stress. Studies are needed to elucidate the reasons why isolated granules show properties different from granules in their natural environment. Although the chemiosmotic hypothesis as originally proposed is no longer valid, several alternative hypotheses have been advanced according to which the osmotic swelling of secretory vesicles constitutes the force that drives the fission step in exocytosis. One such hypothesis, proposed by Geisow and Burgoyne (1982), is based on the observation that the exposure of isolated chrornaffin granules to the cation ionophore monensin in Na+- or K+-containing isotonic media resulted in granule lysis. Lysis occurred equally well in media containing the K’ or Na+ salt of the irnpermeant anion isethionate or in media containing the salt of the permeant anion C1-. Lysis was prevented by hyperosmotic media or by lowering the medium pH to 5.5. Since monensin also causes intracellular lysis of chrornaffin granules, the authors proposed that in the cell monensin probably exchanges cytosolic K+ ions for intragranular protons. The efflux of protons from the granule may cause the dissociation of intragran-

220

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

ular impermeant weak acid species; the resulting K+ salt formation would lead to entry of H 2 0 into the granule. Grinstein et al. (1982) have proposed a model of secretory granule swelling during exocytosis in which the countertransport of H+/alkali metal cation across the granule membrane is mediated by an exchanger. More recently, Zimmerberg and Whitaker (1985) presented data that are consistent with the notion that a secretory granule must swell in order to fuse with the plasma membrane during the exocytosis of cortical granules in eggs of the sea urchin, Lytechinus pictus (see Jackson and Crabb, Chapter 2, this volume). Their experiments showed that granules shrink on exposure to the high osmolality stachyose medium and appear to swell slightly when exposed to Ca2+.Thus, it was proposed that under normal conditions Ca2+causes granule swelling which leads to fusion, and, in the presence of solutions of high osmolality, the Ca*+-induced swelling is not sufficient to result in fusion. The effects of Ca2+,however, are thought to be irreversible such that when the granule is returned to a solution of normal osmolality swelling continues and fusion occurs. The mechanism by which Ca2+promotes granule swelling is not known. Evidence that such a mechanism does not appear to be involved in exocytosis from the chromaffin cell, however, has been presented by Holz and Senter (1986). Using digitonin-permeabilized chromaffin cells, it was shown that pretreatment of cells (in which secretion was inhibited) with Ca2+in a solution made hyperosmotic with stachyose did not result in enhanced catecholamine secretion when cells were returned to normal osmolality. Finally, a model for exocytosis based on the opening of Ca2+-activated K+ channels in secretory vesicle membranes has been proposed (Stanley and Ehrenstein, 1985). The opening of these channels coupled to anion transport across the vesicle membrane could result in the influx of K+ and anions and subsequent osmotic lysis of the vesicles. At the moment, however, there is little evidence to support such a model. Thus, the idea that the fission step of exocytosis follows osmotic swelling of the secretory granule still appears to be accepted by many. However, the mechanism by which osmotic swelling is achieved, if indeed it occurs, remains to be elucidated. REFERENCES Abrahamsson, H . , and Gylfe, E. (1980). Demonstration of a proton gradient across the insulin granule membrane. Acra Physiol. Scand. 109, 113-114. Apps, D., and Schatz, G. (1979). An adenosine triphosphatase isolated from chromaffingranule membranes is closely similar to Ft-adenosine triphosphatase of mitochondria. Eur. J . Biochem. 100,411-419.

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

221

Apps. D. K.. Pryde. J . G . . and Phillips, J . H. (1980). Both the transmemhrane pH gradient and the membrane potential are important in the accumulation of amines by resealed chromaffin-granule “ghosts.” FEES Lett. 111, 396-390. Arvan. P., Rudnick, G . , and Castle. J . D. (1984). Osmotic properties and internal pH of isolated rat parotid secretory granules. J . B i d . C‘hPm. 259, 13.567- 13572. Baker. P. F . , and Knight. D. E. (1978). Calcium-dependent exocytosis in bovine adrenal medullary cells with leaky plasma membranes. Nature. ( h n d o w ) 276, 620-622. Baker. P. F.. and Knight, D. E. (1980). Gaining access to the site of exocytosis in bovine adrenal medullary cells. J . Plrysiol. (Prrris) 76, 497-504. Baker, P. F.. and Knight, D. E. 11981). Calcium control ofexocytosis and endocytosis in bovine adrenal medullary cells. Philo.\. Trun.~.R . SOC. London. S e r . B 296, 83-103. Bashford, C. L.. Casey. R. P., Radda. G. K . , and Ritchie. G . A. (1976). Energy-coupling in adrenal chromaffin granules. Nf,ro.osc.ic.,rc,c,I, 399-412. Brown. E. M.. Hurwitz, S., and Aurhach, G. D. (1976). Preparation of viable isolated bovine parathyroid cells. E n d ~ c r i n ~ l ~99, ~ g1582y 1588. Brown. E. M.. Pazoles. C. J . , Creutz. C. E.. Aurbach. G . D.. and Pollard. H. B. (1978). Role of anions in parathyroid hormone release from dispersed bovine parathyroid cells. PI.OC.Nut/. Acrid. Sci. U . S . A . 75, 876-880. Cabantchik, Z. I., Knauf. P. A,, and Rothstein. A. (1978).The anion transport system of the red blood cell. The role of membrane protein evaluted by the use of “probes.” Biochim. Biophys. Actrr 515, 289-302. Carty, S. E . . Johnson, R. G . , and Scarpa. A. (1981). Serotonin transport in isolated platelet granules. Coupling to the electrochemical proton gradient. J . Biol. Chetn. 256, 11244I 12.50. Carty. S. E., Johnson. R. G . , and Scarpa, A. (1982). Electrochemical proton gradient in dense granules from anterior pituitary. J . Biol. Cltrm. 257, 7269-7273. Casey, R. P., Njus. D., Radda, G. K.. and Sehr. P. A. (1976). Adenosine triphosphateevoked catecholamine release in chromaffin granules. Osmotic lysis as a consequence of proton translocation. Biochrm. J . 158, 583-588. Casey, R. P., Njus, D.. Radda, G . K., and Sehr, P. A . (1977). Active proton uptake by chromaffin granules: Observation by amine distribution and phosphorus-3 I nuclear magnetic resonance techniques. Biocliemisrry 16, 972-977. Cidon. S., and Nelson, N. (1983). A novel ATPase in the chromaffin granule membrane, J . B i d . Chem. 258, 2892-2898. Cidon. S.. Ben-David. H., and Nelson, N . (1983). ATP-driven proton flux across membranes of secretory organelles. J. Biol. Clietn. 258, 11684-1 1688. Cohen, F. S., Zimmerberg, J.. and Finkelstein. A. (1980). Fusion of phospholipid vesicles with planar phospholipid bilayer membranes. 11. Incorporation of a vesicle membrane marker into the planar membrane. J . Grn. Physiol. 75, 251-270. Cohen, F. S., Akabas, M. H., and Finkelstein. A. (1982). Osmotic swelling of phospholipid vesicles causes them to fuse with a planar phospholipid bilayer membrane. Science 217, 458-460, Creutz, C. E., and Pollard, H . B. (1980). A biophysical model of the chromaffin granule. Accurate description of the kinetics of ATP and CI dependent granule lysis. B i c p h y ~ . J . 31, 25s-270. Dean, G. E.. Fishkes, H., Nelson, P. J., and Rudnick, G . (1984).The hydrogen ion-pumping adenosine triphosphatase of platelet dense granule membrane. Differences from FIFO. 259, 9569-9574. and phosphoenzyme-type ATPases. J . B i ~ l Chem. Dolais-Kitabgi, J . , and Perlman, R. L. (1975).The stimulation of catecholamine release from chromaffin granules by valinomycin. Mol. Pharmnc,ol. 11, 745-750.

222

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

Dunn, L. A., and Holz, R. W. (1983). Catecholamine secretion from digitonin-treated adrenal medullary chromaffin cells. J . Biol. Chem. 258,4989-4993. DiizgiineS. N. (1985). Membrane fusion. Subcell. Biochern. 11, 195-286. Ekholm. R., Ericson, L. E., and Lundquist, 1. (1971). Monoamines in the pancreatic islets of the mouse. Subcellular localization of 5-hydroxytryptamine by electron microscopic autoradiography. Diubetologiu 7 , 339-348. Geisow, M. J., and Burgoyne, R. D. (1982). Effect of monensin on chromaffin cells and the mechanism of organelle swelling. Cell Biol. I n r . Rep. 6, 933-939. Grinstein, S.. Meulen, J. V., and Furuya, W. (1982). Possible role of H+-alkali cation countertransport in secretory granule swelling during exocytosis. FEBS L e f t . 148, 1-4. Hampton, R. Y., and Holz, R. W. (1983). Effects of changes in osmolality on the stability and function of cultured chromaffin cells and the possible role of osmotic forces in exocytosis. J . Cell B i d . 96, 1082-1088. Hellman, B., Lernmark, A., Sehlin, J . , and Taljedal, I.-B. (1972). Transport and storage of 5-hydroxytryptamine in pancreatic p-cells. Biochem. Pharmucol. 21, 695-706. Hoffman, P. G., Zinder, 0..Bonner, W. M., and Pollard, H. B. (1976). Role of ATP and fl,y-iminoadenosine triphosphate in the stimulation of epinephrine and protein release from isolated adrenal secretory vesciles. Arch. Biochem. Eiophys. 176, 375-388. Holz, R. W. (1978). Evidence that catecholamine transport into chromaffin vesicles is coupled to vesicle membrane potential. Proc. Nurl. Acad. Sci. U.S.A. 75, 5190-5194. Holz, R. W. (1979). Measurement of membrane potential of chromaffin granules by the accumulation of triphenylmethylphosphonium cation. J. Biol. Chem. 254, 6703-6709. Holz, R. W., and Senter, R. A. (1985). Plasma membrane and chromaffin granule characteristics in digitonin-treated chromaffin cells. J . Neurochem. 45, 1548-1557. Holz, R. W., and Senter, R. A. (1986). Effects of osmolality and ionic strength on secretion from adrenal chromaffin cells permeabilized with digitonin. J . Neurochem. 46, 18351842. Holz, R. W., Senter, R. A., and Sharp, R. R. (1983). Evidence that the H+ electrochemical gradient across membranes of chromaffin granules is not involved in exocytosis. J . Biol. Chem. 258, 7506-7513. Hook, W. A . , and Siraganian, R. P. (1981). Influence of anions, cations and osmolarity on IgE-mediated histamine release from human basophils. Immunology 43, 723-73 1. Hutton, J. C. (1982). The internal pH and membrane potential of the insulin-secretory granule. Biochem. J . 204, 171-178. Hutton, J. C., and Peshavaria, M. (1982). Proton-translocating Mg2+-dependent ATPase activity in insulin-secretory granules. Biochem. J . 204, 161-170. Izumi, F., Kashimoto, T.. Miyashita, T., Wada, A., and Oka, M. (1977). Involvement of membrane associated protein in ADP-induced lysis of chromaffin granules. FEBS Letr. 78, 177-180. Johnson, R. G . , and Scarpa, A. (1976a). Internal pH of isolated chromaffin vesicles. J . Biol. Chem. 251, 2189-2191. Johnson, R. G., and Scarpa, A. (1976b). Ion permeability of isolated chromaffin granules. J. Cen. Physiol. 68, 601-631. Johnson, R. G., and Scarpa, A. (1979). Protonmotive force and catecholamine transport in isolated chromaffin granules. J . Biol. Chem. 254, 3750-3760. Johnson, R. G., Scarpa, A,, and Salganicoff, L. (1978). The internal pH of isolated serotonin containing granules of pig platelets. J . Eiol. Chem. 253, 7061-7068. Johnson, R. G., Pfister, D., Carty, S. E.,and Scarpa, A. (1979). Biological amine transport in chromaffin ghosts. Coupling to the transmembrane proton and potential gradients. J. Biol. Chem. 254, 10963-10972. Johnson, R. G., Carty, S. E., and Scarpa, A. (1981). Proton: substrate stoichiometries

7. OSMOTIC EFFECTS IN MEMBRANE FUSION

223

during active transport of biogenic amines in chromitffin ghosts. J . Biol. Chem. 256, 5773-5780. Knight, D. E., and Baker, P. F. (1982). Calcium-dependence of catecholamine release from bovine adrenal medullary cells after exposure to intense electric fields. J . Mm7hr. &id. 68, 107-140. Knight, D. E., and Baker, P. F. (1985). The chromaffin granule proton pump and calciumdependent exocytosis in bovine adrenal medullary cells. J . Mrmhr. B i d . 83, 147-156. Knoth, J . . Handloser. K.. and Njus, D. (1980). Electrogenic epinephrine transport in chromaffin granule ghosts. Bioc,hemis/ry 19, 2938-2942. Korchak. H. M.. Eisenstat. B. A., Hoffstein. S . T.. Dunham. P. B., and Weissmann, G . ( 1980). Anion channel blockers inhibit lysosomal enzyme secretion from human neutrophils without affecting generation of superoxide anion. Proc.. Nut/. A c a d . Sci. U . S . A . ?7, 2721-2725. Lishajko, F. (1969). Influence of chloride ions and ATP-Mg?’ on the release of catecholamines from isolated adrenal medullary granules. A c f a Phvsiol. Scand. 75, 255-256. Lorenson, M. Y., Lee, Y.-C., and Jacobs. L. S. (1981). Identification and characterization of an anion-sensitive Mg”-ATPase in pituitary secretory granule membranes. J . B i d . Ckem. 256, 12802-12810. Mitchell, P. (1961). Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nrrturt (London) 191, 144-148. Njus. D., Sehr, P. A., Radda, G . K., Ritchie, G . A., and Seeley. P. J . (1978). Phosphorus-31 nuclear magnetic resonance studies of active proton translocation in chromaffin granules. Biochemistry 17, 4337-4343. Ogawa, M . . and Inouye. A. (1979). Responses of the transmembrane potential coupled to the ATP-evoked catecholamine release in isolated chromaffin granules. Jpn. J . Phy.\iol. 29, 309-325. Oka, M.. Ohuchi, T., Yoshida, H . , and Imaizumi, R . (1965). Effect of adenosine triphosphate and magnesium on the release of catecholamines from adrenal medullary granules. Biochim. Biophvs. Acfu 97, 170-171. Orci. L., and Malaisse. W. (1980). Single and chain release of insulin secretory granules is related to anionic transport at exocytotic sites. Dicrheres 29, 943-944. Pace, C. S . . and Smith, J . S . (1983). The role of chemiosmotic lysis in the exocytotic releare of insulin. Endocrinology 113, 964-969. PaLoles, C. J . (1982). Anion and proton transport in chromaffin granules. Fed. Proc., Fed. Atn. Soc. Exp. Biol. 41, 2769-2774. Pazoles, C. J . , and Pollard, H. 8 . (1978). Evidence for stimulation of anion transport in ATP-evoked transmitter release from isolated secretory vesicles. J . Eiol. C h e m . 253, 3962-3969. Pazoles. C. J.. Creutz. C. E., Ramu. A.. and Pollard. H. B. (1980). Permeant anion activation of Mg-ATPase activity in chromaffin granules. Evidence for direct coupling of proton and anion transport. J . Bio(. C h e m . 255, 7863-7869. Percy, J. M., Pryde, J . G., and Apps. D. K. (1985). Isolation of ATPase I, the proton pump of chromaffin-granule membranes. BiochcJm.J . 231, 557-564. Phillips. J . H. (1977). Passive ion permeability of the chromaffin-granule membrane. B ~ o chem. J . 168,289-297. Phillips, J . H., and Allison, Y. P. (1978). Proton translocation by the bovine chromaffingranule membrane. Biochem. J . 170, 661-672. Poisner, A. M., and Trifaro, J. M. (1967). The role of ATP and ATPase in the release of catecholamines from the adrenal medulla. I . ATP-evoked release of catecholamines. ATP, and protein from isolated chromaffin granules. Mot. Phrumacol. 3, 561-571. Pollard, H. B., Zinder. O . , and Hoffman, P. G. (1976a). Occurrence and properties of

224

KEITH W. BROCKLEHURST AND HARVEY B. POLLARD

chromaffin granules. I n "Biological Handbook, I: Cell Biology" (P. L. Altman and D. D. Katz, eds.). pp. 358-362. Fed. Am. SOC.Exp. Biol., Bethesda, Maryland. Pollard. H.B., Zinder, O., Hoffman, P. G., and Nikodijevik, 0. (1976b). Regulation of the transmembrane potential of isolated chromaffin granules by ATP, ATP analogs, and external pH. J. Eiol. Chem. 251, 4544-4550. Pollard, H. B., Tack-Goldman, K.. Pazoles, C. J., Creutz, C. E., and Shulman, N. R. (1977). Evidence for control of serotonin secretion from human platelets by hydroxyl ion transport and osmotic lysis. Proc. Nut/. Acud. Sci. U.S.A. 74, 5295-5299. Pollard, H . B . , Pazoles, C. J . , Creutz, C. E., and Zinder, 0. (1979a). The chromaffin granule and possible mechanisms of exocytosis. I n t . Rev. Cytcd. 58, 159-197. Pollard. H . B., Shindo. H., Creutz, C. E., Pazoles, C. J.. and Cohen, J. S. (1979b). Internal pH and state of ATP in adrenergic chromaffin granules determined by "P nuclear magnetic resonance spectroscopy. J. B i d . Chem. 254, 1170-1 177. Pollard. H . B., Pazoles, C. J . , Creutz, C. E., Scott, J. H.. Zinder, O., and Hotchkiss, A. (1984). An osmotic mechanism for exocytosis from dissociated chromaffin cells. J. B i d . Ch6T.m. 259, 1114-1121. Pollard. H. B., Ornberg, R., Levine, M.. Kelner, K.. Morita, K., Levine, R., Forsberg, E., Brocklehurst, K. W., Duong, L., Lelkes, P. I.. Heldman, E., and Youdim, M. (1985). Hormone secretion by exocytosis with emphasis on information from the chromaffin cell system. Vitum. Horm. 42, 109-196. Russell, J. T. (1984). ApH, H' diffusion potentials, and Mg2*-ATPase in neurosecretory vesicles isolated from bovine neurohypophyses. J . B i d . Chem, 259, 9496-9507. Russell, J . T., and Holz, R. W. (1981). Measurement of ApH and membrane potential in isolated neurosecretory vesicles from bovine neurohypophyses. J . B i d . Chem. 256, 5950-5953. Saermark, T.. Andersen. N . M., Atke, A., Jones, P. M., and Vilhardt, H. (1986). Processing and secretion in the neurohypophysis. Stability of isolated secretory vesicles and role of internal pH. Biochem. J . 236, 77-84. Salama, G.. Johnson, R. G., and Scarpa, A. (1980). Spectrophotometric measurements of transmembrane potential and pH gradients in chromaffin granules. J . Gen. Physiol. 75, 109-140. Scherman. D., Nordmann, J., and Henry, J.-P. (1982). Existence of an adenosine 5'-triphosphate dependent proton translocase in bovine neurosecretory granule membrane. Eiochemistry 21, 687-694. Schuldiner, S., Fishkes, H., and Kanner, B. I. (1978). Role of a transmembrane pH gradient in epinephrine transport by chromaffin granule membrane vesicles. Proc. Nurl. Acad. Sci. U.S.A. 75, 3713-3716. Somers, G., Sener, A., Devis, G., and Malaisse, W. J . (1980). The stimulus-secretion coupling of glucose-induced insulin release. XLV. The anionic-osmotic hypothesis for exocytosis. Pfluegers Arch. 388, 249-253. Stanley, E. F., and Ehrenstein, G. (1985). A model for exocytosis based on the opening of calcium-activated potassium channels in vesicles. Life Sci. 37, 1985-1995. Tamagawa, T., and Henquin, J.-C. (1983). Chloride modulation of insulin release, nhRb+ efflux, and 4sCa2+fluxes in rat islets stimulated by various secretagogues. Diabetes 32, 4 16-423. Taugner, G. (1972). The effects of univalent anions on catecholamine fluxes and adenosine triphosphatase activity in storage vesicles from the adrenal medulla. Biochem. J. 130, 969-973. Toll, L., and Howard, B. D. (1978). Role of MgZ+-ATPaseand a pH gradient in the storage of catecholamines in synaptic vesicles. Biochemistry 17, 25 17-2523.

7.OSMOTIC EFFECTS IN MEMBRANE FUSION

225

Wilkins. J. A., and Salganicoff, L. (1981). Participation of a transmembrane proton gradient in 5-hydroxytryptamine transport by platelet dense granules and dense-granule ghosts. Biochem. J . 198, 113-123. Wilson, S. P., and Kirshner, N. (1983). Calcium-evoked secretion from digitonin-permeabilized adrenal medullary chromaffin cells. J . Biol. Chrm. 258, 4994-5000. Zimmerberg, J., and Whitaker, M. (1985). Irreversible swelling of secretory granules during cxocytosis caused by calcium. Nrrture (London) 315, 581-584. Zimmerberg, J.. Cohen, F. S., and Finkelstein. A. (1980). Fusion of phospholipid vesicles with planar phospholipid bilayer membranes. I . Discharge of vesicle contents across the planar membrane. J . Gen. Physiol. 75. 241-250.

This Page Intentionally Left Blank

CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 32

Chapter 8 Polyanionic Agents and Inhibition of Phagosome-Lysosome Fusion: Paradox Lost M A YER B . GOREN' Department o j Molecular and Celliilar Biology Notional Jewish Center for Immunology cind Reqiratory Medicine Denvm, Colorudo 80206 and Department of Microbiology und 1rnmiinolog.v University of Colorado Health Sciences Center Denver. Colorado 80220 I.

11.

111.

1V.

V.

VI.

Introduction A . Historical Background B. Why Question the Polyanionics Hypothesis? C. Procrustes and an Alternative Interpretation The Fluorescent Lysosomal Probes A. Concerning Acridine Orange B. The Sulfonated Fluors: Mobile but Not Fickle Nonionic Hydrocolloids in Lysosomes: Fusion Inhibitors or Gelatinous Traps? A . Developing the Gelatinous Trap Model B. Differential Delivery of Lysosomal Constituents C. Differential and Sequential Transfer of Two Fluorescent Lysosornal Markers Fusion Inhibition Is Incompatible with the Cells' Functional Status A. Indigestion? Not at All B. Microbicidal Activities Some Residual Bodies A. Concerning Mechanisms: In Search of a Phenomenon'? B. Some Vesicles Are More Equal than Others Recapitulation and Conclusions References

'

Visiting Scientist, Department of Biophysics, The Weizmann Institute of Science, Rehovot, Israel.

227 Copyright 8 IYRB hy Academic P r e w Inc. All nghts of reproduction i n any form r e w v e d .

228

MAYER B. GOREN

When two texts contradict each other, the meaning run be determined only when u third texi is found which harmonizes them. Rabbi Ishmael, Introduction to Sifru

This anecdotal and editorial account is concerned with a hypothesis advanced by myself and colleagues from Mill Hill, London, about the antagonistic effects of endocytosed polyanionic substances on phagosome-lysosome fusion in cultured macrophages (Goren et al., 1976). It is a hypothesis that has been embraced by others and extensively studied. Still, I am now convinced that it is invalid. This chapter is an analysis of a host of studies-principally those of others-which, in our interpretation, support this conviction. The bulk of experimental evidence from our own studies is detailed in recent reports (Goren et al., 1987a,b). Because of its special relevance to the sulfatides of Mycobacterium tuberculosis, the “polyanionics hypothesis” is one to which I had an understandably strong commitment (Goren, 1977; Goren and Brennan, 1979). However, when I was unable to confirm the results of the earlier collaborative studies, the complication led us to recognize certain important artifacts of methodology. Based on these findings, I expressed my early doubts and apprehensions about polyanionic agents and fusion inhibition at the Third Leiden Conference on Mononuclear Phagocytes (Goren et al., 1980), but I believe the weight of unquestionably impressive contradictory opinions. also presented at the conference, was persuasive (Hart and Young, 1980; Kielian and Cohn, 1980a). Indeed, ultimately we were able to bring our results into agreement with those of others. But our more recent findings are for us convincing that the “polyanionics hypothesis” is flawed and invalid. For a participant in the creation and development of an attractive and seemingly fruitful concept, there is little satisfaction to be derived from its dismantling. That is a “pleasure”-a Schadenfreude-usually reserved for others! For me the dismantling is doubly disappointing, for it potentially denies the fusion-inhibiting activity of the sulfatides (“sulfolipids”) of Mycobacterium tuberculosis (see Goren and Brennan, 1979) and revokes their role as “virulence agents,” enjoyed for a brief period when they seemed to be implicated in protecting the intracellular pathogen from assault by the macrophage lysosomal hydrolases (Armstrong and Hart, 1971). For the present, the agent, if any, that mediates this behavior (see below) remains unknown. 1.

INTRODUCTION

A. Historical Background Certain viable intracellular pathogens, acting from within phagosomes, can prevent the fusion of the entrapping phagosorne with lysosomes. The

8 . POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

229

phenomenon has been viewed as a potentially powerful mechanism in pathogenesis (Armstrong and Hart, I97 I ; Davis-Scibienski and Beaman, 1980: Friis, 1972; Horwitz, 1983; Jones and Hirsch, 1972; Wyrick and Brownridge, 1978; also see reviews by Densen and Mandell, 1980; Edelson, 1983; Goren, 1977; Goren c f al., 1980; Horwitz, 1982). Interest in identifying specific substances that might be implicated in antagonizing fusion led to the recognition of several structurally defined lysosomotropic (de Duve et a / . , 1974) polyanionic compounds that appeared to impose such a block to phagosome-lysosome (P-L)’ fusion from the lysosomal side. The original studies, exciting at the time, showed that when an active polyanionic such as suramin (Hart and Young, 1975),mycobacterial sulfolipids, or dextran sulfate (Goren r t a / . , 1976) was incorporated into macrophage secondary lysosomes, together with specific markers that also accumulate in these organelles (fluorescent acridine orange or electron-opaque colloidal ferritin), transfer of the lysosomal labels to newly formed phagosomes was largely prevented as compared with their prompt and abundant delivery to phagosomes in control cells. Viable Baker’s yeast, Saccharornyces crreuisiae, was a phagocytosis target of choice. Absence of the markers in the phagosomes after yeast uptake was logically viewed as evidence of P-L fusion inhibition. The resulting “polyanionics hypothesis” (Goren et d . , 1976; Goren, 1977) that linked polyanionic structure with an ability to induce the fusion dysfunction was subsequently “confirmed” by other investigators using similar techniques. And it was strengthened and broadened further in perhaps a score of publications that followed (Draper et a/., 1979; Geisow et d . , 1980, 1981; Hart, 1982; Hart and Young, 1978, 1979, 1980; Hart e l a / . , 1983; Kielian and Cohn, 1980a,b, 1982; Kielian ct a / . , 1982). But our own studies confronted us with a disturbing paradox. We consistently confirmed polyanion-induced fusion inhibition as judged from the behavior of the lysosomal acridine orange (AO) marker, whose transfer to yeast-containing phagosomes was almost abolished. But this was consistently denied by electron microscopic (EM) evidence: transfer of the particulate colloidal marker (Thorotrast) was izot inhibited-a finding also reported by Pesanti (1978), who used ferritin. The dilemma led us to recognize and describe serious artifactual limitations of the A 0 methodology stemming principally from the dye’s weakly basic properties (see below). We therefore largely abandoned A 0 as a lysosomal marker and concentrated on EM methodology and on fluorescent labeling with highly

’

Abbreviations: P-L phagosome-lysosome: AO, acridine orange: LR. lissamine rhodamine: S R , sulforhodamine; EM, electron microscopy: DS, dextran sulfate; FD, fluoresceinated dextran; SL, mycobacterial sulfolipids; FITC, fluorescein isothiocyanate: LY, lucifer yellow; TCA, trichloroacetic acid; KLH, keyhole limpet hemocyanin; HRP, horseradish peroxidase.

230

MAYER 8.GOREN

ionized “nonpermeant” markers such as lissamine rhodamine, lucifer yellow, and sulforhodamine (see below) (Goren, 1983; Goren et al., 1984a). Our recent studies relieved, but still did not explain, the earlier dilemma of the contradictory A 0 versus EM results (Goren et a[., 1987a), but they allowed us to reestablish that polyanionic substances indeed can prevent transfer of both A 0 and the electron-opaque Thorotrast to phagosomes. In the main, we accepted this behavior as confirmation of the original hypothesis that we had long sought, and it brought our results into agreement with those of others. Nevertheless, 1 was not able to relieve a persisting suspicion that polyanion-induced fusion inhibition may be more artifactual than real. 8. Why Question the Polyanionics Hypothesis?

Is not the concordant evidence from both vital fluorescence acridine orange studies and from electron microscopy with markers such as ferritin or Thorotrast adequate to affirm that polyanions in fact antagonize P-L fusion? Certainly the apparent inhibition documented by both methods seems undeniable. We suggest, however, that most of the data heretofore reported for “pol yanion” cells as obtained with acridine orange may legitimately be questioned (see Section III,A) The fickle behavior of this mobile fluor has too often been ignored in a surrender to the dye’s seductive aesthetic allure. On the other hand, it seems reasonable that the transfer of Thorotrast from secondary lysosomes to phagosomes, whether in control or polyanion-containing cells, is probably a valid indication that the labeled lysosomes have participated in some kind of fusion process to effect delivery. But absence of the particulate marker from phagosomes is not unequivocal evidence that lysosomal fusion has been abolished. It implicates only a failure of delivery of a specific electron-opaque marker from lysosomes to phagosomes and provides no evidence about fusion behavior (a membrane phenomenon) or about the transfer, or not, of other lysosomal contents. Our conviction that the “polyanionics hypothesis” is invalid rests principally on the argument that no important function of macrophages seems to be disturbed by the supposed fusion inhibition, surely a strange finding if the activity of such a prominent cytoplasmic compartment is abolished in cells in which it must find almost constant use. This then is the “text” that contradicts the text of “no marker delivery, therefore no fusion.” C. Procrustes and an Alternative Interpretation

Studies in macrophages of phenomena that would be expected to be notably affected by an inhibition of phagolysosome formation have con-

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

231

tinued to uncover puzzling features-ever more difficult to reconcile with a profound lysosomal dysfunction. In my opinion, this has required invoking increasingly tenuous and convoluted interpretations in order that a growing number of disparate and paradoxical experimental findings can all be accommodated into a polyanionic “Procrustean bed,” another recurrent theme. Our analysis suggests that a conjunction of methodologic and physicochemical pitfalls transformed the “bed” into an “iron maiden” in which all of us who have utilized the established techniques for studying the phenomena were embraced. Almost all of the accumulated evidence may be accounted for by the alternative interpretation that we propose, namely, that fusion is not inhibited, in contradiction of conclusions deduced solely from the retention of specific markers in lysosomes. We suggest instead that the polyanionic agents accumulate in lysosomes as gelatinous, sluggishly moving hydrosols that p h y s i c d v entrap electron-opaque markers and ionicul/y trap the usual alternative marker, the weakly basic acridine orange, in a matrix that may not transfer to phagosomes for many hours. But there are good reasons to believe, as described in succeeding sections, that more mobile lysosomal contents are still delivered to the fusion “partners,” after which the partly depleted lysosome probably pinches off and reseals itself. Evidence supporting this interpretation is presented later. In Section I1 (admittedly pedantic), the behavior of various fluorescent lysosomal labels will be examined in order to simplify analysis of phenomena that are observed by these techniques. Taken together with our model of lysosomal polyanionic “gels” that trap both EM markers and A 0 but do not inhibit P-L fusion, a plausible interpretation of earlier unexplainable A 0 data is then developed. The behavior of nonionic hydrocolloids as “fusion inhibitors” is then discussed, followed by sections that review and buttress arguments for rejecting the “polyanionics hypothesis.” The fertility of the alternative model is illustrated by the valid predictions and elementary interpretations that it allows.

II. THE FLUORESCENT LYSOSOMAL PROBES

A. Concerning Acridine Orange

An elementary analysis of the properties of acridine orange (Fig. I ) and its behavior as a lysosomal marker will provide an understanding of why it may so rarely be used with confidence. Its movement can be especially misleading in the presence of polyanionic agents. For additional, more detailed analyses, see de Duve et a / . , (1974), Allison and Young (1969), and Goren et al. (1984a).

MAYER B. GOREN

232

FIG. I .

Structure of acridine orange, a weakly basic permeant lysosomotropic dye.

1. PERMEATION OF MEMBRANES

Acridine orange is a weak base that in its uncharged free base form is sufficiently lipid soluble to traverse biological membranes fairly freely. As depicted in Fig. 2 (after de Duve et al., 1974), in lysosomes, of considerably lower pH than the cytoplasm, the base becomes protonated and trapped within these organelles because the membrane is a more formidable barrier to passage of the charged salt form: R R

I I

R-N:+HA

K

I I

L R - N + : H

A

R

Free base form Protonated salt form (traverses membrane freely) (traverses membrane poorly)

Incubation of normal macrophages with a weak base of appropriate pK ( A 0 in this instance) results in concentrating the dye within lysosomes (pH -4.5) up to several hundred-fold over its concentration in the cytoplasmic or extracellular space (pH -7.2). This concentration process is promoted by proton pumping at the lysosomal membrane, (de Duve et al., 1974; Ohkuma and Poole, 1978). The process is not instantaneous; achieving the final concentration within lysosomes may require several hours (de Duve et al., 1974). When a macrophage monolayer is exposed to 5 pg AO/ml in Hanks buffer for about 15 min, the lysosomes are usually adequately labeled to provide good transfer of fluor to a phagocytosed target such as a yeast cell. B Y ANIONIC SUBSTANCES 2. COMPLEXING

Acridine orange, a basic dye, complexes with and is bound by acidic substances: the mycobacterial sulfolipids (SL), cation-exchange resins (Lerman, 1961), microorganisms with acidic surfaces (i.e., tubercle bacilli), dead microorganisms (Strugger, 1947; Freidlin et d., 1977; Pantazis and Kniker, 1979; Goren et al., 1984a),and DNA. Accumulation of A 0 in macrophage lysosomes is powerfully reinforced when these contain pol yanionic substances such as suramin, SL, dextran sulfate, or polyglu-

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

pH4.5

233

2

W

FIG.2. Uptake of acridine orange by a macrophage. Protonation of the free base inside lysosomes traps the dye and concentrates it in these organelles. (Afterde Duve rr ul.. 1974.)

tamic acid. From our observations, a semiquantitative order of affinity of these substances or domains for A 0 is as follows: Dowex-50 or equivalent Polyanions within lysosomes Sulfolipids of M. tuberculosis Viable or killed tubercle bacilli Normal lysosomes; macrophage nuclei Killed and digesting microorganisms (yeasts, Escherichia coli) Viable microorganisms (yeasts) Accordingly, in a closed system containing elements as enumerated above, an excess of acridine orange will provide dye to all of the components. But if A 0 is available in only limited amounts, only the elements of highest affinity for the dye may be colored.

3 . L’ACRIDINA k MOBILE Because of the equilibrium between protonated and free base forms of AO, mobile free base is always present even if only in minute amounts and is free to traverse membranes. Therefore, in a closed system, A 0 from any compartment can distribute to any other compartment where it is more tightly bound. When a macrophage monolayer labeled with A 0 is exposed in medium to sufficient Dowex-SO (Na+.form), essentially all of the dye is transferred to the extracellular cation-exchange resin (Fig. 3) Goren, 1983; Goren r t a / . , 1980, 1984a). A second monolayer with lysosomes containing sulfoli-

MAYER 8.GOREN

234

HA=(

Dowex 50

\

FIG.3. Redistribution of acridine orange by membrane permeation from lysosomes to an external trapping agent, Dowex-SO.

pids or dextran sulfate may be substituted for the Dowex-50 beads. From the above affinity scale, one would infer that, if there is insufficient A 0 to saturate all domains, dye present in a compartment or organelle of lower affinity can redistribute to a domain of higher affinity, for example-and notably-to lysosomes sequestering polyanions (even in neighboring macrophages), if they are not saturated with the dye. 4. ARTIFACTSAND

A PARADOX

In employing acridine orange as the lysosomal label for assessing P-L fusion, most investigators expose the cells to about 5 pg AO/ml for about I5 rnin (Hart and Young, 1975; Kielian and Cohn, 1980a,b, 1982; Goren et al., 1984a,b). As noted before, this seems adequate for appropriate labeling and for subsequent delivery of the dye to (yeast) phagosomes (Fig. 4a). Penetration of the phagosome by A 0 to give a uniform bright green, yellow, or orange color is good evidence that the yeast or other target has been either severely injured or killed and is undergoing digestion (reviewed by Goren ef d . , 1984a). The macrophage nuclei are also bright green with acquired dye. Curiously, this is not a trivial observation. We have noted and described before (Goren et al., 1980, 1984a) that when “polyanion” macrophages sequestering, e.g., sufficient rnycobacterial sulfatides, dextran sulfate, or polyglutamic acid are similarly exposed FIG. 4. (a) Normal control macrophage prelabeled by exposure to 5 pg AO/ml and allowed to phagocytose viable yeasts. The yeast phagosomes (Y)show a variety of colors. Uniform yellow to orange colorations indicate the yeasts to be at least severely injured or dead, and undergoing digestion. The macrophage nuclei (N)are green with acquired dye, a typical fusion pattern, Bar, 10 pm. (b) Macrophages with mycobacterial sulfatide sequestered in lysosomes after exposure to 5 pg AO/ml for 15 min. The dye is entirely trapped in lysosomes (L) where the complexing capacity for A 0 has not been reached. The nuclei are devoid of dye. Bar, 5 Fm.

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

235

for 10-15 min to 5 Fg AO/ml, the lysosomes become brilliant. But in contrast with the aspects of Fig. 4a, the nuclei are essentially devoid of dye (Fig. 4b). When such cells ingest yeasts, the phagosomes remain dark against a background of brilliantly fluorescent lysosomes. The nuclei still remain dark (Fig. Sa). Kielian and Cohn (1980a) reported the same behavior (see also Fig. 2b of Kielian ef u l . , 1982). This is a typical “nonfusion” pattern, for absence of dye from the phagosomes has been taken as definitive evidence that fusion has been suppressed. The apparent fusion inhibition is also supported by EM evidence: invariably, transfer of the Thorotrast lysosomal marker to the phagosomes is also highly suppressed. Surely this concordance would appear to be unequivocal evidence for fusion inhibition. Instead, we suggest on the basis of the alternative “gelatinous trap” interpretation that the “unequivocal evidence” is only a compounding of two artifacts: trapping of the markers and a deficiency of acridine orange.

5. PARADOX RESOLVED? We earlier described, but were not able to explain, how labeling of “polyanion” cells with IS-30 pg AO/ml (instead of 5 pg) for the usual IS min, followed by yeast phagocytosis, changes the “nonfusion” pattern to one of abundant fusion (Goren ef al., 1980, 1984a). Figure Sb shows such a polyanion monolayer (that was also prelabeled with Thorotrast) after a IS-min exposure to 25 p g AO/ml followed by yeast phagocytosis. The macrophage nuclei are green. The phagosomes are now aglow with dye. The yeasts are very likely dead and being digested, undeniable evidence of P-L fusion and delivery of both lysosomal A 0 and lysosomal hydrolases. But, paradoxically, the secondary lysosomes are evidently still unchanged-engorged and also brilliant with dye; this is ordinarily taken as evidence that fusion is inhibited (Goren rt al., 1980. 1984a). And this interpretation is still supported by EM evidence: the exposure to the higher concentration of acridine orange did not result in increased release FIG.5 . (a) Macrophage as in Fig. 4b 2 hr after ingestion of viable yeasts. Phagosomes

(Y)remain dark against a background of engorged brilliant lysosornes. The nucleus (N)is also dark because the A 0 exposure was inadequate to saturate the ”polyanion” lysosornes. Bar. 10 p n . (b) “Polyanion” macrophages labeled by exposure to 25 pg AOlml for 15 min. Viable yeasts were fed after the cells were washed and chased. Reversal of the inhibition of dye transfer seen in Fig. 5a is apparent. All of the phagosomes are brilliant with dye; the nuclei are green. But the polyanion lysosornes still retain the engorged status, with dye complexed by gelatinous polyanions. Thorotrast in these cells was not transferred to the phagosomes. Bar, 10 pm.

236

MAYER 6.GOREN

of thoria marker to the phagosomes (M. B. Goren and A. E. Vatter, unpublished). This situation can be resolved merely by invoking a “Procrustean bed” interpretation, namely, that the undeniable intrusion of lysosomal contents into the yeast phagosomes stems solely from the activities of primary lysosomes, for the secondary lysosomes are still clearly intact based on both EM and visual evidence. We suggest that contributions from primary lysosomes would indeed be expected, and will be discussed later. However, the “gelatinous trap” model taken together with the very high affinity of the polyanions for A 0 allows for a more convincing explanation of this curious behavior. Although the lysosomes of the “polyanion” cells exposed to 5 pg AO/ ml (Figs. 4b and 5a) have acquired abundant dye, the nuclei remain unilluminated. Since the polyanion lysosomes represent the domain of greatest affinity for A 0 in this system, it is evident then that the lysosomal complexing capacity has not been satisfied during the exposure to the usual 5 pg AO/ml. The appearance suggests that the gelatinous polyanion traps not only the Thorotrast but evidently all of the dye present in the cells. None is free to equilibrate with nuclei; none is free to be delivered to phagosomes. However, the capacity of the polyanion for weak bases is not infinite. It is evidently saturated by exposure to the higher concentration of A 0 (or might be by longer exposure to lower concentrations), and the evidence for this is found in the nuclei (Fig. 5b): they too have now acquired the fluor. Free dye is also accumulated by the lysosomes, probably by the usual passive diffusion and protonation, aided by proton pumping. It is largely this increment of free AO, then, that is delivered to phagosomes. The gelatinous polyanion with its complexed dye and entrapped thoria remains in the secondary lysosomes. And so the illusion of inhibited fusion of secondary lysosomes as judged by both EM and fluorescence microscopy is sustained-by a dual deception. B. The Sulfonated Fluors: Mobile but Not Fickle

From considering the sulfonated heterocyclic structures shown in Fig. 6, i.e., highly ionized, polar fluors of relatively low molecular weight, it may be inferred that lissamine rhodamine B (LR) (Goren, 1983; Goren et a!., 1984a), lucifer yellow (LY), and sulforhodamine (SR) should behave as impermeant fluorescent labels for macrophage secondary lysosomes (see also Goren et al., 1987a). Indeed, we found them to be taken up adequately from the culture medium by pinocytosis during prolonged exposure (usually more than 24 hr). Thus, the mechanism of uptake is

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

237

so, H FIG. 6. Structure of the sulfonated heterocyclic fluors: ( a ) lissarnine rhodarnine. (b) lucifer yellow. and ( c ) sulforhodamine. These are strongly ionized in solution, mobile, irnperrneant, and accumulate in lysosomes. They are not cornplexed by polyanionic substances.

clearly different from the rapid passive diffusion, and trapping by protonation, that governs labeling by acridine orange. After uptake, all of the sulfonated probes showed a quite similar punctate-to-vacuolar appearance characteristic of secondary lysosomes, and these exhibited dramatic saltatory motions when viewed by means of time-lapse video recordings. Wang and Goren (unpublished) found no evidence in such time-lapse recordings that the saltatory motions were inhibited by either sequestered dextran sulfate or by 10 mM NH4CI in the medium. (Compare Hart er ul., 1983). The lysosomal localization of these fluors has since been well established for LY (Miller et al., 1983; Swanson et a / . , 1985) and for LR and SR (Wang and Goren, 1987).

238

MAYER B. GOREN

Following phagocytosis of heat-killed yeasts, all of these dyes are delivered with about the same facility to the phagosomes and there penetrate and color the targets. Since we established that the behavior of viable and of heat-killed yeasts is identical in polyanion-containing macrophages (Goren rt al., 1984b, 1987a), the killed yeasts are legitimate targets for PL fusion studies using these highly ionic probes as lysosomal labels. When viable yeasts are used as targets, killing and especially digesrion evidently do not proceed rapidly enough for significant penetration of the targets by these fluors in less than about 6 hr (Goren et al., 1984a). They are mobile but, in our experience, not so freely so as acridine orange. Neither are they fickle: we found no evidence of any redistribution behavior such as we earlier documented for AO. This is in accord with their structures: because they are ionized salts of strong (sulfonic) acids, they are not complexed by lysosomally sequestered polyanionic substances or by macrophage nuclei, so there is no “demand” for these fluors in these domains. Neither is it plausible that they can be delivered from lysosomes to phagosomes (or elsewhere) by a facile traversal of biological membranes. As described before, various intracellular organelles can be essentially depleted of the permeant A 0 by, e.g., extracellular Dowex-SO. We found, however, that SR, as an example, although strongly adsorbed from solutions by anion-exchange resins, is not subject to such “robbing” from lysosomes by either strong acid or strongly basic ion-exchange resins. It is therefore impermeant. Hence polar fluors are not complexed and trapped by either polyanionic (DS, polyglutamic acid) or by nonionic hydrocolloids (Goren, 1983; Goren et al., 1987a). Although the monochromatic fusion figures that the sulfonated fluors exhibit are not so psychedelically exhilarating as those obtained with AO, the polar probes are judged to be much more reliable and trustworthy in their behavior. Use of FITC-labeled heat-killed yeasts as targets (Goren, 1983) enhances the aesthetic qualities and provides a palette of fusion figures (with SR lysosomes) from pure green (unfused) to orange-red that unequivocally demonstrates that P-L fusion is quantitatively the broad spectrum that would be expected for a biological system-not all black or white, but rather including many shades of gray. 111.

NONlONlC HYDROCOLLOIDS IN LYSOSOMES: FUSION INHIBITORS OR GELATINOUS TRAPS?

A fortuitous conjunction of circumstances led to an explosive multiplication in the number of documented polymeric materials that antagonize

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

239

transfer of lysosomal EM markers to phagosomes and so would be judged to inhibit P-L fusion. We asked (probably naively) if the quadrivalent positive thorium oxide marker might become cross-linked with adsorbed polyanionics into a hydrocolloidal gelatinous matrix that was effectively “immobilized” within lysosomes even if they were not inhibited in fusion with phagosomes. For us, the answer to this question provided the elusive harmonizing “third text” to which Rabbi Ishmael referred. A. Developing the Gelatinous Trap Model

The resulting search, detailed in recent publications (Goren rt d., 1987a,b), led to the surprising finding that a variety of nonionic waterdispersible polymers behaved remarkably like dextran sulfate (DS)in macrophage lysosomes. Exposure of cells to much higher concentrations was required, for unlike DS the neutral polymers are not endocytosed by an accelerated uptake that can concentrate the polyanions in lysosomes to surprisingly high levels (Cohn and Parks, 1967; Kielian et al.. 19821, and it is likely that the polyanions within lysosomes are much more viscous than the counterpart nonionic polymers (see below). The effects of the nonionics on secondary lysosomes were much the same as induced by DSmorphologically, and in interfering with transfer of EM marker to phagosomes (Goren ~t a l . , 1987a). However, in direct contradiction of the fusion inhibition implicit in the EM results,.pirorescent Iysosomul mcirkcrs used for simultaneous labeling were readily delivered to the phagosomes. not only the sulfonated fluors of the preceding section, but, of course, acridine orange as well. None of these are complexed by the nonionic polymers. A “Procrustean bed” analysis nevertheless would argue that the neutral nonionic polymers cilso must inhibit P-L fusion. and would dismiss the fluorescent marker transfer as being due to membrane permeation. It is of interest that Kielian et a / . (1982) and Kielian and Cohn (1980b), from comparing neutral dextran with the polyanionic dextran sulfate. concluded that cells exposed to 30 mg dextran/ml (instead of 10-20 pg DS/rnl) were nor inhibited in P-L fusion. But, as judged from their description, the behavior was evidently assessed only by fluorescence microscopy with A 0 and not by E M , and, as we noted, A 0 would not be trapped by dextran. Clearly the “Procrustean” interpretation embracing all of the neutral polymers as inhibitors would require a considerable dismantling of various hypotheses (described later) offered to account for the fusion-inhibiting behavior of the polyunions. We suggest instead that all of the polymers described, whether neutral, nonionic, or polyanionic, form “hydrosols” in water (Weiser, 1939)and in sufficiently high concen-

240

MAYER B. GOREN

trations will acquire a gelatinous or jellified structure that moves only sluggishly within (or out of) lysosomes. Indeed, fluorescence polarization measurements with fluoresceinated dextran (molecular weight 40,000; FD 40) indicate that the polymer behaves in lysosomes as if dissolved in 90% glycerol. Thus, its “motion in the normal intralysosomal environment was relatively restricted by comparism with free solution” (Geisow et al., I981). We would predict that lysosomal fluorescinated chlorite-oxidized amylose (“COAM”) (Geisow et ul., 1980) would exhibit an even more aberrant fluorescence polarization than FD 40 because of its much higher molecular weight and anionic (carboxylic) qualities. It has been described as a potent inhibitor of P-L fusion. B. Differential Delivery of Lysosomal Constituents We further deduce that the hydrosols are viscous and trap the colloidal EM markers physically, and we have presented EM evidence that supports this conviction (Goren et ul., 1987a). The neutral hydrosols cannot ionically trap any of the fluorescent mobile markers, which are therefore delivered to phagosomes with relative facility. This behavior is thus viewed as a diffPrentia/ delivery of lysosomal constituents. Polyanions, on the other hand, should trap only acridine orange ionically but should not trap the sulfonated probes. In accord with this expectation, when “dextran sulfate cells” were labeled with both Thorotrast and any one of the ionic sulfonated fluors, the dyes were transferred to phagosomes, whereas Thorotrast was not. This is still another example of differential delivery-and we cannot account for this by other than a fusion process. “Inhibition” of movement of EM markers must be understood not as an absolute abolition of marker transfer, but rather as a shift toward much lower amounts of label found in phagosomes as compared with those seen in control macrophages. The difference has been assessed by stereologic analysis (Kielian and Cohn, 1980a,b; Kielian et a/., 1982) or by visual “scoring” (scale 0-4) of the amount of marker delivered (Goren et ui., 1980, 1984b, 1987a). We have suggested, however, that because of the swelling of hydrocolloid lysosomes and distortions in distribution of the EM markers even these methods of comparing marker delivery in control and in experimental cells, although “sensitive” (Kielian et ul., 1982) may not be entirely valid (Goren et al., 1987a). As noted before, the ionic fluors are not as freely mobile as AO. Their transfer in the presence of DS requires about 3-4 hr for abundant delivery as compared with 1-2 hr in the presence of the neutral hydrocolloids. This probably reflects the consequences of a “cross-linking” of DS by bivalent cations and even, to an extent, by multiple cationic sites on proteins-to

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

24 1

increase the effective molecular weight of polymer and to promote its gelatinous qualities. “DS” cells that allow considerable transfer of sulforhodamine during 3-4 hr may still trap Thorotrast marker even after 24 hr (Goren et a/., 1987a). C. Differential and Sequential Transfer of Two Fluorescent Lysosomal Markers The differential transfer of the impermeant sulfonated fluors from lysosomes in which Thorotrast is presumed to be “immobilized” in a gelatinous nonionic hydrosol logically leads to predictions about derivative systems that allow for testing of the “gelatinous trap” model. If dextran is the polymer in the “jelly,” and is also rendered fluorescent, as by fluoresceination, then the green gel should be visihly immobilized in the lysosoma1 compartment. And it would be anticipated that if lysosomes are labeled with both the red mobile fluor and the green dextran then the red fluor of low molecular weight should transfer preferentially to phagosomes. We found the behavior to be essentially as predicted when cells were simultaneously labeled with sulforhodamine and fluoresceinated dextran (FD), molecular weight 40,000. The SR was delivered almost to the point of disappearing from the lysosomal domain during about 3 hr, to leave only the green FD behind. But afterward, this hydrophilic polymer of modest molecular weight, not surprisingly, then became transferred slowly to phagosomes over a period that lasted about 24 hr, when, for the most part, it was found in the same phagosomes to which SR had been delivered earlier. We have interpreted this behavior as supporting a piecemeal or “nibbling” process to describe the interactions of phagosomes with these particular kinds of lysosomes, i.e., containing components of grossly different mobilities (Wang and Goren, 1987). This is an important result: the FD 40 behavior is the same if it is used alone to label the macrophages. It also trapped Thorotrast, but only transiently (M. B. Goren, unpublished), probably because of insufficiently high molecular weight. Accordingly, if the functional status of such cells is judged only from the delivery of the FD 40 to phagosomes during a 4-5 hr period, then the cells would be considered “inhibited,” when in fact they are not! (see Goren et ul., 1984a, for an example). Based on these studies we anticipate that FD of considerably higher molecular weight or perhaps even DS conjugated with an appropriate fluor would remain immobilized in lysosomes for much longer periods, while mobile markers of low molecular weight are transferred differentially. Fluoresceinated “COAM,” an oxidized starch described as a remarkably potent “inhibitor of fusion” by Geisow et al. (1980), might

242

behave in degraded. molecular prolonged

MAYER B. GOREN

such a fashion for a prolonged period if it is not lysosomally It is a true polyanionic substance and of considerably higher weight than FD 40. Both characteristics should contribute to lysosomal immobilization.

IV. FUSION INHIBITION IS INCOMPATIBLE WITH THE CELLS’ FUNCTlONA L STATUS

As noted earlier, it is strange that no important function of macrophages seems to be disturbed by the supposed fusion inhibition, e.g., phagocytosis, microbicidal activity, but especially paradoxical is the welldocumented unaltered intracellular digestion. A. Indigestion? Not at All!

Data from our own fairly extensive quantitative studies, published in part (Goren, 1983; Goren et al., 1985), as well as more limited data from others indicate that the course of intracellular digestion is unaffected in “polyanion” macrophages. Hart and Young (1980) found digestion of S. cereuisiue “to proceed” in such cells and reconciled fusion inhibition with this behavior by invoking the participation of primary lysosomes. Indeed, they stated that lysosomal acid phosphatase was delivered “in only slightly reduced amounts into yeast-containing phagosomes” (in polyanion-inhibited cells). Kielian et al. (1982) reported that the enzymatic half-life of endocytosed horseradish peroxidase (HRP) was the same in control and in dextran sulfate (DS)-inhibited cells. Choosing among several interpretations, these investigators postulated that the normal digestion might be attributed to “a population of primary lysosomes and/or secondary lysosomes that have not yet taken up very much DS and that are able to fuse and degrade pinocytosed HRP.” In examining EM profiles of Thorotrast-labeled “polyanion” cells or cells in which the marker was trapped by a nonionic polymer, we have regularly recognized yeasts digesting in phagosomes in which no trace of marker was discernible (see Fig. 7). And we earlier described the entirely unaltered digestion of various radiolabeled soluble and particulate substances by “polyanion” cells. The surprising results were also interpreted as possibly being due to the activities of primary lysosomes or to incompletely inhibited secondary lysosomes (Goren, 1983). As stated earlier, participation of primary lysosomes would be expected: by definition they could not have accumulated any polyanionic substances. More recently we compared the digestion of yet another target--”SI-labeled heat-killed

FIG.7. Macrophage labeled with a colloidal gold preparation stabilized with gelatin and given heat-killed yeast cells to ingest. The gelatin hydrosol traps the gold and prevents transfer of the label to phagosomes. Nevertheless, the yeasts show signs of extensive digestion. Bar. 0.5 Fm.(Electron micrograph courtesy of Dr. N . Mor.)

244

MAYER B. GOREN

yeasts-in control cells and in monolayers exposed to more than adequate dextran sulfate to inhibit Thorotrast transfer to yeast phagosomes. The release of trichloroacetic acid (TCA)-soluble radiolabel from these cells over 10 hr is depicted in Fig. 8 (Goren et al., 1985, 1987b). The conclusions are the same as reached earlier: digestion is essentially unaffected by the polyanionics residing in secondary lysosomes. To account for these kinds of data we suggested that perhaps secondary lysosomes of resident macrophages may be less functional than would have been thought (Goren et al., 1985). However, we abandoned this interpretation for it seemed weak and unconvincing (Goren et al., 1987b). 6. Microbicidal Activities

At one time the presumed fusion inhibition seemed to be a potential tool for dissecting the role of P-L fusion in the microbicidal activities of

\,. \*-

Residual IZsl

r

0

*

10

-

Normal Cells 0---0 Dextran Sulphate Cells &A

1

5

I 10

How$ FIG.8. Digestion of '2sI-labeled heat-killed yeasts by control macrophages compared with cells that have accumulated dextran sulfate in more than sufficient quantities to inhibit transfer of Thorotrast marker to the yeast-containing phagosomes. Monolayers in quadruplicate were pulsed with 1 x 106 opsonized radioiodinated yeasts for 45 min then washed free of uningested yeasts, and the cells were incubated in culture medium during the succeeding 10 hr. At the indicated intervals, the supernatants were collected, and their total radioactivity and TCA-soluble activity were assessed. The rate of release of soluble label during 10 hr was substantially identical for the two kinds of cells (data courtesy of Dr. W. J. Bruyninckx). (Reproduced from Goren et al., 1987b with permission of Alan R. Liss, Inc.)

8.POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

245

mononuclear phagocytes. And yet no convincing evidence has emerged that polyanion-induced P-L fusion “dysfunction” in macrophages antagonizes fungicidal or bactericidal behavior for S. cweuisiue, Li.stc.riu monocytogenes, Staphylococcus uiireiis, or Proteirs miruhilis, or affected even related activities, e.g. phagocytosis (Pesanti, 1978; Hart and Young, 1980; Geisow et ul., 1981; Kielian et al., 1982; Warr and Jakab, 1983). Our studies also showed that microbicidal activity for E . c d i or for Streprococciis fuecalis was the same in control and in “dextran sulfate” macrophages (Goren et al., 1985). The ambitious early study by Pesanti was stimulated by an extremely important goal: “If phagolysosome formation could be blocked, then the role of lysosomes in macrophage antimicrobial activity could be more accurately assessed.” Pesanti found that suramin pretreatment of the cells “inhibited transfer of acridine orange to phagosomes in macrophages that had interiorized either Snc.charornyces or L . tnonoc~ytogetie.s” (virulent or avirulent for mice). “However, electron micrographs of ferritin-loaded macrophages indicated that . . . there was fusion of lysosomes with phagosomes following ingestion of living Listeriu of either strain. All interiorized organisms in both control and suramin-treated macrophages werc contained within vacuoles in which ferritin was present . . . . ,. Pesanti simply accepted the EM evidence with L . monocytogc’nrs as definitive-for uninhibited fusion. However, Pesanti’s results can be explained away so as not to violate the “polyanionics hypothesis.” To account for the unrestricted fusion of phagosomes containing L. monocytogenes (in the presence of lysosomal suramin), Hart and Young (1980; see also Draper et a / ., 1979) invoked an overwhelming “fusion-promoting power of the intracellular targets” which surpassed the fusion-inhibiting power of the suramin; they dismissed the disparate A 0 data as being “due to optical limitations of the system.” Some data from Alexander (1981) suggest that “inhibition” of P-L fusion by poly-D-glutamic acid promoted the survival of Leishmania mexicuna mexicanu amastigotes and promastigotes in infected macrophages, and that chloroquine caused a marked reduction in parasite growth. It has been claimed that this and other lysosomotropic weak bases accelerate the P-L fusion process and also relieve the fusion block imposed by polyanions (Hart and Young, 1978, 1979, 1980). Although the evidence from Alexander is impressive, only a circular argument can link the effects with fusion “inhibition” on one hand and “fusion promotion” on the other. If there are components in macrophage lysosomes or in nonlysosomal granules that are cidal or inhibitory for Leishmuniu, might they not be inactivated or antagonized by polyanions? There is abundant evidence that the latter can inhibit a variety of enzymes (for review see Ginsburg and Sela, 1976). Considering the obverse side of

246

MAYER B. GOREN

the coin, the lysosomotropic chloroquine will surely reach phagosomal targets either by permeation or via P-L fusion. But the rise in pH that this base elicits dramatically antagonizes the activity of the lysosomal acid hydrolases; this is reflected in an equally dramatic fall in digestive processing (Seglen el d., 1979; Ziegler and Unanue, 1982). Is it not then more likely that chloroquine by itself is the specific agent that damages intraphagosomal Leishmania rather than that the damage is due to a facilitated but probably ineffectual P-L fusion? The same analysis can be applied to the intracellular behavior of virulent M . tuberculosis. The normal nonfusion pattern for this pathogen is described as being reversed by “coating” the target with either normal or immune rabbit serum before phagocytosis. But this had no effect on the pathogen’s intracellular survival (Armstrong and Hart, 1975). The nonfusion pattern may also be reversed with chloroquine, but now with a concomitant intracellular killing or stasis of the pathogen (Hart and Young, 1980). If relief of the fusion block by (digestible) serum coating-which cannot have long lasting effects in protecting tubercle against any hydrolytic onslaught-fails to alter the intraphagosomal fate of the bacilli, how can a putative relief of the block by chloroquine, which simultaneously deranges lysosomal hydrolase activity, serve to kill or suppress the bacteria? Again, it seems probable that the behavior is due to a direct action of the amine on the pathogen. The facility of some weak bases to promote P-L fusion or to relieve a polyanionic “block” is not unequivocal. Kielian et a / . (1982) found that chloroquine “failed to modify inhibited P-L fusion.” Our (limited) tests with chloroquine on polyanion cells labeled with acridine orange gave similar negative results (Goren, 1983). Still, it seems plausible that chloroquine could cooperate with A 0 to saturate the complexing capacity of lysosomal polyanions for weak bases. An increment of complexed A 0 might then be released and transferred to phagosomes. Hence the chloroquine would substitute for the additional A 0 of Fig. 5b.

V.

SOME RESIDUAL BODIES

The resolution of the acridine orange paradoxes, the correct anticipation concerning the differential deliveries of any of the mobile fluorescent probes in the presence of nonionic hydrosols and of delivery of the sulfonated probes from dextran sulfate lysosomes, and the valid prediction of the behavior of colocalized low molecular weight red sulforhodamine and polymeric yellow-green fluorescein-dextran all seem to presage a utility of the “gelatinous trap” model to provide simple answers to some

8 . POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

247

complex questions. There are. of course, exceptions-“residual bodies”-that have so far defied dissolution. The suggested antagonism between polyanions and the weak bases (above) is one example. Another might be the classification of microorganisms as “fusers” or “nonfusers” on the basis of their appearance within polyanion cells labeled for electron microscopy. But the assignment to the “fuser” category in effect endows the microorganism with an ability to influence from within the phagosome the response to the lysosoma1 polyanionic “inhibitors” (see Hart and Young, 1980; Draper et al., 1979). It is a facility that defies prediction and which can be assigned only after the fact. If EM marker delivery is recognized with a given target in cells that are “inhibited” when tested with viable yeasts, then the target in question becomes a “fuser”: otherwise it is a “nonfuser.” Current EM studies by N . Mor and M. Goren (in preparation) suggest that these “fuser”-“nonfuser” assignments are probably not warranted. However, the phenomenon alluded to earlier, of fusion inhibition from the phagosome by a limited selection of viable intracellular pathogens in normul macrophages. seems established, although the mechanisms are uncertain.

A. Concerning Mechanisms: In Search of a Phenomenon?

1 . LYSOSOMAL pH; MEMBRANE FLUIDITY The search for a convincing mechanistic interpretation to account for the “fusion-inhibiting” qualities of the polyanions has not yielded any profound insights. It has been neither a road to Damascus nor an Archimedes submerged. In a study by Kielian and Cohn (19821, “fusion inhibition” by dextran sulfate could not be correlated with a slight lowering of intralysosomal pH (from about 4.76 to about 4.36). and “increasing the lysosomal pH with . . . chloroquine failed to modify the inhibition of fusion.” But Geisow et al. (1981) found the intralysosomal pH of cells powerfully “inhibited’ with the polyanionic “COAM” to be increased to about 5.0. It is, therefore, unlikely that P-L fusin could be inhibited by an increased lysosomal pH on one hand and also by a decreased pH on the other. Fluorescence polarization studies with isolated DS-containing lysosomes labeled withp-parinaric acid, a membrane probe, led to the inference that direct interaction of lysosomal membranes with polyanions decreases the fluidity of the membranes and thereby antagonizes fusion (Kielian and Cohn, 1982). The postulated direct interactions are, of course, ionic. Accordingly, since the nonionic hydrocolloids mimic the behavior of DS in inhibition of Thorotrast transfer, then the modest influence of DS on membrane fluidity would appear to be of marginal significance.

248

MAYER B. GOREN

2. CALCIUM ION The inhibition of P-L fusion by polyanions is curiously insensitive to manipulations of the internal Ca2+concentration. It is well known that this ion has important roles in many membrane fusion processes (for review see Papahadjopoulos, 1978). Therefore, an appealing mechanistic interpretation for the presumed inhibition would take into account the capacity that the polyanionics surely have for complexing calcium. Moderate affinity of “COAM” for divalent cations has been described. Therefore, a relief of the block might be expected if a putative Ca2+deficiency were alleviated. However, according to Geisow et ul. (1980), although levels of both Ca2+and Mg2+in macrophages were artificially controlled with the ionophore A23187, neither high nor low concentrations of the cations appeared to affect P-L fusion in “COAM” cells that were fusion inhibited. T h i s j n d i n g should not be surprising. If, as we suggest, P-L fusion in these cells is actually not inhibited, but only marker is not transferred to phagosomes, then changes in intracellular Ca2+concentration would have no effect on the appearance of the yeasts. Marker would not be delivered in any event. In the absence of convincing alternatives, most investigators have concluded that the basis for “polyanion-induced fusion dysfunction” probably lies in membrane perturbations caused by interaction between the polyanion and multiple cationic sites in membrane proteins. Even this interpretation seems to be denied by the evidence. It is relevant that either sequestration of calcium or membrane perturbations might be fulfilled by particles of cation-exchange resins, which are, of course, particulate insoluble polyanions. It seemed to us, therefore, that these should be able to impose the fusion block from within the phagosomal confines. Our early studies in acridine orange-labeled cells with strong acid ionexchange particles served only to reveal important pitfalls of the A 0 technique (Goren et al., 1980b, 1984a). In recent EM studies we found that neither carboxylated nor sulfonated microscopic cation-exchange particles prevent P-L fusion: phagocytosis of these substances elicits a prompt and abundant fusion response (Goren et a l . , 1987a). B. Some Vesicles Are More Equal than Others

If polyanions indeed block fusion of biological membranes, then it is paradoxical that they accumulate in macrophage secondary lysosomes. Should not the polyanions inhibit fusion of the pinosome that carries it into the cell? Perhaps the pinosomal concentration is insufficient for this.

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

249

But how can the delivery then be maintained in the face of a growing concentration of polyanions in the lysosomes that ultimately renders the latter “incapable” of fusing with phagosomes? Lysosomes heavily laden with dextran sulfate still undergo fusion with pinosomes that carry either horseradish peroxidase (HRP) or Thorotrast markers into the cells (Kielian et al., 1982). C. L. Swendsen and M. B. Goren (unpublished data) also have documented the labeling of “polyanion” cells with Thorotrast and have interpreted the results to mean that pinosomes are truly free to fuse with ‘‘inhibited” lysosomes. Accordingly, at the time we abandoned soluble materials (iodinated KLH) in favor of particulate targets (KLH-anti-KLH; radiolabeled yeasts, zymosan, etc.) as substrates for intracellular digestion studies in “polyanion” cells (Goren, 1983; Goren et al., 1985, 1987a).How than can we account for what is surely an awkward dilemma if the lysosomes are presumed to be unable to fuse? We now believe that no unusual properties of the pinosomal membrane need be invoked (see Kielian ef al., 1982)to relieve this paradox. Accepting unhindered fusion of polyanion lysosomes with both types of endocytic vesicles, we suggest that only “one-way” traffic of marker is allowed: HRP and Thorotrast are surely free to move from ordinary pinosomes into pol yanion-containing lysosomes following fusion, but lysosomal Thorotrast, trapped in a matrix of gelatinous dextran sulfate (polyanion), is not free to move out. VI.

RECAPITULATION AND CONCLUSIONS

In summary, no important macrophage function has been discerned that is antagonized by lysosomal accumulation of polyanions: phagocytosis, intracellular digestion, microbicidal activity. The apparent block to fusion is indifferent to Ca2+concentrations and cannot be imposed from the phagosomal compartment by particles of cation-exchange resins. If the “phenomenon” depends on perturbations in vesicle membranes, then pinosomal membranes would appear to be embarrassing exceptions to the inhibition; instead they are very likely behaving like phagosomal membranes, both allow fusion with “polyanion” lysosomes-but with oneway blocks to traffic. Neutral hydrocolloids mimic the behavior of polyanions with Thorotrast markers, but still allow delivery of permeant or nonpermeant mobile lysosomal fluors to phagosomes. We have documented similar behavior in “dextran sulfate” cells doubly labeled with mobile ionized fluors and Thorotrast; that is, the fluors are transferred to phagosomes, but the Thorotrast, and presumably the polyanion, are not. In addition, we have recognized a differential and sequential delivery that

250

MAYER B. GOREN

distinguishes between mobile impermeant fluors and fluoresceinated dextran of even moderate molecular weight. Finally, the gelatinous trap model allows understanding of the behavior of acridine orange in the presence of polyanions and why it can be manipulated. These findings justify the conclusion that polyanionic agents do not inhibit phagosomelysosome fusion. ACKNOWLEDGMENTS Original studies from my laboratory described here were supported by Grant A1 17509 from the National Institutes of Health and by Grant A1 08401 from the U.S.-Japan Cooperative Medical Science Program administered by the National Institute of Allergy and Infectious Diseases. I thank Nadia destackelburg for illustrations and Shirley Downs and Marjorie McCormick for careful preparation of the manuscript. M.B.G. is the Margaret Regan Investigator in Chemical Pathology, National Jewish Center for Immunology and Respiratory Medicine. REFERENCES Alexander, J. (1981). Leishmania mexicana: Inhibition and stimulation of phagosome-lysosome fusion in infected macrophages. Exp. Parasirol. 52, 261-270. Allison, A. C., and Young, M. R. (1969). Vital staining and fluorescence microscopy of lysosomes. In “Lysosomes in Biology and Pathology” (J. T. Dingle and H. B. Fell, eds.), Vol. 2, pp. 600-628. Elsevier, Amsterdam. Armstrong, J . A., and Hart, P. D. (1971). Response of cultured macrophages to Mycobacterium tuberculosis. with observations on fusion of lysosomes with phagosomes. J . Exp. Med. 134, 713-740. Armstrong, J. A., and Hart, P. D. (1975). Phagosome-lysosome interactions in cultured macrophages infected with virulent tubercle bacilli. Reversal of the usual non-fusion pattern and observations on bacterial survival. J . Exp. Med. 142, 1-16. Cohn, Z. A., and Parks, E. (1967). The regulation of pinocytosis in mouse macrophages. 11. Factors inducing vesicle formation. J . Exp. Med. 125, 213-232. Davis-Scibienski, C., and Beaman, B. L. (1980). Interaction of Nocardia asteroides with rabbit alveolar macrophages: Association of virulence, viability, ultrastructural damage, and phagosome-lysosome fusion. Infect, Immun. 28, 610-619. de Duve, C., de Barsy, T., Poole, B., Trouet, A,, Tulkens. P., and Vanhoof. F. (1974). Lysosomotropic agents. Biochem. Pharmacol. 23, 2495-253 I . Densen, P., and Mandell, G. L. (1980). Phagocyte strategy vs. microbial tactics. Rev. Infect. Dis. 2, 817-838. Draper, P., Hart, P. D., and Young, M. R. (1979). Effects of anionic inhibitors of phagosome-lysosome fusion in cultured macrophages when the ingested organism is Mycobacterium lepraemurium. Infect. Immun. 24, 558-561. Edelson, P. J . (1982). lntracellular parasites and phagocytic cells: Cell biology and pathophysiology. Rev. Infect. Dis. 4, 124-135. Freidlin, I. S ., Khavkin, T. N., Artemenko, N. K., and Sakharova, I. Ya. (1977). Vital fluorescence microscopy of lysosomes in cultured mouse peritoneal macrophages during their interactions with microorganisms and active substances. 11. Interactions of macrophages with a non-pathogenic strain of Escherichia coli. Acta Microbiol. Acad. Sci. Hung. 24, 293-302.

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

25 1

Friis, R. R. (1972). Interaction of L cells and Chlarnydiu p.yittu(.i-Entry of the parasite and host responses to its development. J . BLic,tcriol. 110, 706-721. Geisow. M. J.. Beaven. G . H., Hart, P. D., and Young. M. R . (1980). Site of action of a polyanion inhibitor of phagosome-lysosome fusion i n cultured macrophages. E x p . Cell Res. 126, 159-165. Geisow, M. J . , Hart, P. D., and Young, M. R. (1981). Temporal changes of lysosome and phagosome pH during phagolysosome formation i n macrophage$: Studies by Huore+ cence spectroscopy. J . Cell Biol. 89, 645-652. Ginsburg. I., and Sela. M. (1976). The role of leukocytes and their hydrolases in the persistence, degradation, and transport of bacterial constituents in tissues: Relation to chronic inflammatory processes in staphylococcal. streptococcal, and mycobacterial infections and in chronic periodontal disease. Crit. Rev. Mic,robiol. 4. 249-332. Goren. M. B. ( 1977). Phagocyte lysosomes: Interaction with infectious agents, phagosomes. and experimental perturbations in function. Annu. Reu. Micmhiol. 31, 507-533. Goren, M. B. (1983). Some paradoxes of macrophage functions. I n “Host Defenses to Intracellular Pathogens” (T. K. Eisenstein. P. Actor. and H . Friedman. eds.), pp. 3150. Plenum. New York. Goren, M. B., and Brennan, P. J. (1979). Mycobacterial lipids: Chemistry and biologic activities. In “Tuberculosis” (G. P. Youmans, ed.), pp. 64-193. Saunders, Philadelphia. Goren, M. B . . Hart, P. D.. Young, M. R.. and Armstrong. J . A . (1976). Prevention of phagosome-lysosome fusion in cultured macrophages by sulfatides of Myc~ohoc~teriurn tuhrrculosis. Proc. N a t l . Acud. Sci. U . S . A . 73, 2510-2514. Goren, M. B., Swendsen, C. L., and Henson, J. (1980). Factors modifying the fusion ot phagosomes and lysosomes: Art, fact, and artefact. I n “Mononuclear Phagocytes: Functional Aspects, Part 11” (R. van Furth. ed.). pp. 999-1038. Martinus Nijhoff. The Hague. Goren, M. B.. Swendsen. C. L., Fiscus. J.. and Miranti. C. (1984a). Fluorescent markers for studying phagosome-lysosome fusion. J . Lcrrkocvrc, Biol. 36, 273-292. Goren. M. B.. Vatter. A. E., and Fiscus, J . (1984b). Mycobacterial sulfolipids and polyanionics as inhibitors of phagosome-lysosome fusion: Resolution of an enigma. I n “Proceedings, Joint Tuberculosis and Leprosy Symposium,” pp. 33-42. US-Japan Cooperative Medical Science Program. Tokyo. Goren, M . B., Bruyninckx. W . J . , Leung. K.-P., Swendsen, L. S.. Heifets, L., and Fiscus. J . ( 1985). Functionality of secondary lysosomes in murine resident peritoneal macrophages. I n “Proceedings, Joint Leprosy and Tuberculosis Symposium.” pp. 9-30. U.S.-Japan Cooperative Medical Science Program. Bethesda. Maryland. Goren, M . B . , Vatter, A . E., and Fiscus, J . (1987a). Polyanionic agents as inhibitors of phagosome-lysosome fusion in cultured macrophages: Evolution of an alternative interpretation. J . Leukocyte Biol. 41, I 1 1-121. Goren. M . B.. Vatter, A. E., and Fiscus. J . (1987b). Polyanionic agents do not inhibit phagosome-lysosome fusion in cultured macrophages. J . Leukocyte B i d . 41, 122- 129. Hart, P. D. (1982). Lysosome fusion responses of macrophages to infection: Behavior and significance. In “Phagocytosis-Past and Future” (M. L. Karnovsky and L. Bolis, eds.), pp. 437-447. Academic Press. New York. Hart, P. D., and Young, M. R . (1975). Interference with normal phagosome-lysosome fusion in macrophages using ingested yeast cells and suramin. Nature (London) 256, 47-49. Hart, P. D., and Young, M. R. (1978). Manipulations of the phagosome-lysosome fusion response in cultured macrophages. Enhancement of fusion by chloroquine and other amines. Exp. Cell Res. 114, 486-490.

252

MAYER B. GOREN

Hart, P. D., and Young, M. R. (1979). The effect of inhibitors and enhancers of phagosomelysosome fusion in cultured macrophages on the phagosome membranes of ingested yeasts. Exp. Cell. R e $ . 118, 365-375. Hart, P. D., and Young, M. R. (1980). Manipulation of phagosome-lysosome fusion in cultured macrophages: Potentialities and limitations. In “Mononuclear Phagocytes: Functional Aspects. Part 11’’ (R. van Furth, ed.), pp. 1039-1055. Martinus Nijhoff, The Hague. Hart, P. D., Young, M. R., Jordan, M. M., Perkins, W. J., and Geisow, M. J . (1983). Chemical inhibitors of phagosome-lysosome fusion in cultured macrophages also inhibit saltatory lysosomal movements. A combined microscopic and computer study. J . Exp. M e d . 158, 477-492. Horwitz, M. A. (1982). Phagocytosis of microorganisms. Reu. Infect. Dis. 4, 104-123. Horwitz, M. A. (1983). The legionnaires’ disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J . Exp. Med. 158,2108-2126. Jones, T. C., and Hirsch, J . G. (1972). The interaction between Toxoplasmn gondii and mammalian cells. 11. The absence of lysosomal fusion with phagocytic vacuoles containing living parasites. J . Exp. M e d . 136, 1173-1 194. Kielian, M. C., and Cohn, Z. A. (1980a). Phagosome-lysosome fusion. Characterization of intracellular membrane fusion in mouse macrophages. J . Cell Biol. 85, 754-765. Kielian, M. C., and Cohn, Z. A. (1980b). Determinants of phagosome-lysosome fusion in mouse macrophages. In “Mononuclear Phagocytes: Functional Aspects. Part 11” (R. van Furth, ed.), pp. 1077-1095. Martinus Nijhoff, The Hague. Kielian. M. C., and Cohn, Z. A. (1981). Phorbol myristate acetate stimulates phagosomelysosome fusion in mouse macrophages. J . Exp. Med. 154, 101-1 1 I . Kielian, M. C., and Cohn, Z. A. (1982). Intralysosomal accumulation of polyanions. 11. Polyanion internalization and its influence on lysosomal pH and membrane fluidity. J . Cell Biol. 93, 875-882. Kielian, M. C., Steinman, R. M., and Cohn, Z. A. (1982). Intralysosomal accumulation of polyanions. 1. Fusion of pinocytic and phagocytic vacuoles with secondary lysosomes. J . Cell Biol. 93, 866-874. Lerman, L. S. (1961). Structural considerations in the interaction of DNA and acridines. J . Mol. Biol. 3, 18-30. Miller, D. K., Griffiths, E., Lenard, J., and Firestone, R. A. (1983). Cell killing by lysosomotropic detergents. J . Cell Biol. 97, 1841-1851. Ohkuma, S., and Poole, B. (1978). Fluorescence probe measurement of the intralysosomal pH in living cells and the perturbation of pH by various agents. Proc. Narl. A c a d . Sci. U . S . A . 15, 3327-3331. Pantazis, C. G., and Kniker, W. T. (1979). Assessment of blood leukocyte microbial killing by using a new fluorochrome microassay. J . Reticuloendothel. SOC.26, 155-170. Papahadjopoulos, D. (1978). Membrane fusion. Calcium-induced phase changes and fusion in natural and model membranes. Cell Surf. R e v . 5, 766-790. Pesanti, E. L. (1978). Suramin effects on macrophage phagolysosome formation and antimicrobial activity. Infecf Immun. 20, 503-51 1. Seglen, P. E., Grinde, B., and Solheim, A. E. (1979). Inhibition of the lysosomal pathway of protein degradation in isolated rat hepatocytes by ammonia, methylamine, chloroquine and leupeptin. Eur. J . Biochem. 95, 215-225. Strugger, V. S. (1947). Die Vitalfluorochromierung des Protoplasmas. Naturwissenschaften 34, 267-273. Swanson, J. A,, Yirinec, B . D., and Silverstein, S. C. (1985). Phorbol esters and horseradish peroxidase stimulate pinocytosis and redirect the flow of pinocytosed fluid in macrophages. J . Cell Biol. 100, 851-859.

8. POLYANIONIC AGENTS AND PHAGOSOME-LYSOSOME FUSION

253

Wang, Y.-l., and Goren, M . B. (1987). Differential and sequential delivery of lysosomal fluorescent probes into phagosomes in mouse peritoneal macrophages. J . Cell Biol. 104, 1749-1754. Warr. G. A . , and Jakab, G. J . (1983). Lung macrophage defense responses during suramininduced lysosomal dysfunction. E x p . M o l . Parhol. 38, 193-207. Weiser. H. B. (1939). “Colloid Chemistry.” Wiley. New York. Wyrick. P. B., and Brownridge, E. A. (1978). Growth of Chltrmydiu p . s i r / ~ c in i macrophages. Infecr. I m m u n . 19, 10.54-1060. Ziegler. K.. and Unanue, E. R. (1982). Decrease in macrophage antigen catabolism caused by ammonia and chloroquine is associated with inhibition of antigen presentation to T cells. Pvac. Narl. Acud. Sci. U . S . A .79, 17.5-178.

This Page Intentionally Left Blank

Part Ill

Virus-Cell Fusion

This Page Intentionally Left Blank

CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 32

Chapter 9

Fusion of Viral Envelopes with Cellular Membranes SH UN-ICHI OHNISHI Depctrtment of Biophysics Fucirlty of Science Kyoto University Sak-vo-kid. Kyoto 606, Jnpcin

Introduction Membrane Fusion Activity of Enveloped Viruses A . HVJ or Sendai Virus B . Influenza Virus C. Semiliki Forest Virus D. Vesicular Stomatitis Virus 111. Mechanism of Fusion A . Binding and Close Apposition B. Interaction of the Hydrophobic Segment with the Target Cell Membrane Lipid Bilayer LV. lnfectious Cell Entry Mechanisms References I.

11.

1.

INTRODUCTION

Many families of viruses have an envelope wrapping their genome, RNA o r DNA. The envelope is a lipid bilayer membrane with the virusspecific glycoproteins spanning it. The bilayer originates from the host cell membrane and has a lipid composition and transbilayer distribution quite similar to the host's. The viral glycoproteins have the functions of binding to the target cell surface and fusion with the cell membranes (see Table I). The two functions are carried by a single glycoprotein in influenza virus (HA), VSV (G), and SFV (E).' In HVJ, the functions are 257 Copyright 0 19811 by Academic Pre\\. Inc All rights of reproduction in any form reserved.

258

SHUN-ICHI OHNlSHl

TABLE I VIRALENVELOPE GLYCOPROTEINS

Virus Hemagglutinating virus of Japan (HVJ) Influenza virus Semliki Forest virus (SFV) Vesicular stomatitis virus (VSV) Mouse mammary tumor virus (MMTV) La Crosse virus Mouse hepatitis virus

Molecular weight (K) Glycoprotein

HN

61

Function

Refs.u

Binding and neuraminidase

I

F2 + FI 52, I I HA1 + HA2 44, 30 48-63 NA E3, E2, 1I, 52, El 51 G 61

Fusion Binding and fusion Neuraminidase Binding and fusion

2, 3 4, 5

Binding and fusion

8

gp52 + gP36 G I , G2 E2 + 90A + 90B

52, 36

Fusion

9

120, 34 120 90, 90

Binding and fusion (?) (GI)

6 7

10

II Binding and fusion (90A)

Key to references: (I)Kohama er 01. (1978). (2) Homma and Ohuchi (1973). (3) Scheid and Choppin (1974). (4) Lazarowitz and Choppin (1975). ( 5 ) Klenk el ul. (1975). (6) Bucher and Pdlese (1975). (7) Garoff el u!. (1980). (8) Rose and Gallione (1981). (9) Redmond and Dickson (1983). (10) GonzalezScarano (1985). (11) Sturman ef ul. (1985).

carried by separate glycoproteins, HN for binding and F for fusion. When viruses encounter target cells, they first bind to the cell surface through interaction of the viral glycoprotein with receptors (Fig. la). Sialoglycoproteins andlor sialoglycolipids are known to be the receptors for H N in HVJ and HA in influenza virus. Fusion of the virus envelope with target cell membranes is an essential initial step in infection since the virus can transfer its genome into the cytoplasm by this event (Fig. lc). Envelope fusion is induced by the action of the virus glycoprotein on target membranes. Some fusion proteins are initially produced in precursor forms and then cleaved posttranslationally by proteolytic enzymes. F in HVJ and HA in influenza virus are such examples. They are produced as precursor forms FO and HA0 and then cleaved into F2 and F1 (Homma and Ohuchi, 1973; Scheid and Choppin, 1974) and HA1 and HA2 (Klenk et al., 1975; LazaroI Abbreviations: HVJ, hemagglutinating virus of Japan, a synonym of Sendai virus; SFV, Semliki Forest virus; VSV, vesicular stomatitis virus; MMTV, mouse mammary tumor virus; HA, hemagglutinin; BHA, HA released from influenza virus by bromelain treatment; IMPS, intramembrane particles; PC, phosphatidylcholine; PE, phosphatidylethanolarnine; PS, phosphatidylserine.

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

259

"4 ff

t

FIG. I . (a) Binding of an enveloped virus to a target cell membrane through interaction of the viral glycoproteins with receptors. (b) Close apposition of the two membranes and disturbance in the target cell membrane lipid bilayer by the hydrophobic segment of the virus fusion glycoprotein. ( c ) Fusion of the two membranes. resulting in release of the viral nucleocapsid into the target cell cytoplasm. The nucleocapsid is a complex of virus genome. RNA or DNA, with proteins. Matrix proteins underneath the envelope probably bind to both the membrane and nucleocapsid.

witz and Choppin, 19751, linked by disulfide bonds between the subunits, respectively. The precursors are inactive, but the cleaved forms are active in both fusion and infectivity, strongly suggesting a causal relationship between them. Other examples are gPr73 in MMTV and E2 in mouse hepatitis virus which are cleaved into gpS2 and gp36 (Redmond and Dickson, 1983) and 90A and 90B (Sturman et al., 1983, respectively, to become active. E in S F V is cleaved into E3, E2, and El (Garoff et d.,1980). On the other hand, G in VSV is not subject to cleavage. The amino-terminal segment of HA2, F1, and gp36 consists of 20 or more hydrophobic amino acid residues (Table 11). That of HA2 also contains two or three acidic residues. These segments are produced on posttranslational proteolytic activation. They are assumed to be responsible for fusion activity because of the hydrophobicity and also because of the conservation of sequence among various strains of HVJ and influenza virus (Gething el a!., 1978). However, the sequence homology is not always observed; for example, influenza C virus and pneumovirus have sequences different from other viruses except for the hydrophobicity. E l in SFV and G in VSV do not contain such amino-terminal hydrophobic segments but have internal hydrophobic stretches. Residues 80-100 in El and residues 100-131 in G, which have sequence homology among the strains, may be such stretches though not strongly hydrophobic (Table 11). They also contain a few acidic residues within the sequence. In the fusion reaction, the two membranes should come close together,

TABLE I1 HYDROPHOBIC SEGMENTSOF VIRALFUSIONGLYCOPROI-EINS"

fu

8

Pararny xoviridae Paramyxovirus F1 HVJ' sv5=

NDV' Pneurnovirus FI RS3 Ort hornyxoviridae Influenza virus HA2 A/PR/8/34 (H I )4 NJapan1305157 (HW A/Aichi/2/68 (H3F NFPV/Rostock/34 (H7)' BIL.ee/W Influenza C virus C/CaI/l 89

FFGAVIGTIALGVATSAQITAGIALAEARFAGV VIGLAALGVATAAQVTFIGAIIGGVALGVATAAQITFLGFLLGVGSAIASGVAVSK-

GLFGAIAGFIEGGWTGMIDGWYGYHGLFGAIAGFIEGGWQGMVDGWYGYH-

GLFGAIAGFIENGWEGMIDGWYGFRGLFGAIAGFIENG WEGLVDGWYGFRGFFGAIAGFLEGGWEGMIAGWHGTY-

IFGIDDLIIGLLFVAIVETGIGGYLLGSR-

Togaviridae Alphavirus El SFV (79-1 10)"' Sindbis (79-1 Rhabdoviridae Vesiculovirus G VSV Ind (100-132)" VSV NJ (100-132)'3 VSV Ind ( 1 74-200)12 VSV NJ ( 174-200)'2 Retroviridae Type B oncovirus MMTV gp3614 Type C oncovirus MoMLV plSE' Akv p15Ei6 F-MUIVp15E'ATLV p20E18

-KVYTUVYPFMWGGAYCFCDSENTQLSEAYVDR-KVFGGVY PFMWGGAQCFCDSENSQMSEAYVEL-

-KQGTWLNPG FPPQSCGYATVTDAEAVIVQVTPH-

-KDGVSFNPGFPPQSCGYGTVTDAEAHIVTVTPH-KGLCDSNLISMDITFFSEDGELSSLGK-ESVCSQLFTLVGGIFFSDSEEITSMGL-

FVAAIl LGlSALl AIITSFAVATTALVK-

EPVSLTLALLLGGLTMGGIAAGIGTGTTALMATQQFQQLQAA VQDDLREPVSLTLALLLGGLTMGGlAAGVGTGTTALVATQQFQQLQ.4AMHDL)LKEPVSLTLALLLGGLTMGG1AAGVGTGTTALVATQQFQ~LHAAVQL)DLKAVPVAV WLVSALAMGAGVAGGITGSMSLASGK-

" Key 10 superscript reference numbers: I I ) Blumberg e f ctl. (19851: Hsu and Choppin (1984). ( 2 ) Richardson

r r / . (IYXO). ( 3 1Collins (1984).(4) Winter rf ul. (1981). ( 5 ) Gething " f ul. (1980). (6) Verhoeyen e r d . (1980). (71 Porter C I d . (197Y). ( X I Kr! stal c*/ d.(1982). (9) Nakada ef ul. (1984). (10) Garoff el 01. (1980). ( 1 1 ) Rice and Strauss (19x1). (12) Rose and Gallione 11981). (13) Gallione and Rose (1983). (14) Redmond and Dickson (1983). ( 1 0 Shinnick ef crl. 11981). (16) Lenz ef ol. 119821. (17) Koch c r ol. (19831. (181Seki c d . (1983). (21

ef ul.

(21

262

SHUN-ICHI OHNlSHl

and the putative fusogenic segment should be able to interact with the target membrane, inducing some disturbance eventually leading to fusion (Fig. Ib). Generally, proteins may have similar hydrophobic segments in them. Even acqueous enzymes have such segments: two in porcine trypsin and one in hepatic alcohol dehydrogenase (Asano and Asano, 1984). Virus fusion proteins can also have more than one hydrophobic segment. For example, residues 175-199 in G is another hydrophobic sequence (Table 11). These hydrophobic segments can be fusogenic only when they approach and interact with target membranes. Envelope fusion was first shown to occur for HVJ by electron microscopy (Howe and Morgan, 1969). Such fusion was not clearly demonstrated for other viruses, however, and, instead, uptake of viruses into intracellular coated vesicles and smooth vesicles was observed. This raised a question of the mechanism of virus genome transfer into the target cell cytoplasm for these viruses. Ten years later, the induction of envelope fusion activity in mildly acidic media was discovered for SFV (VMnanen and Kaariainen, 1979, 1980), influenza virus (Maeda and Ohnishi, 1980; Huang ef al., 1981; Lennard and Miller, 19811, and later VSV (White et al., 1981; Mifune et al., 1982; Matlin et al., 1982). The pH dependence of fusion was markedly different; while HVJ can fuse at neutral as well as acidic pH values, SFV, influenza virus, and VSV can fuse only at acidic pH (Fig. 2). A new virus entry mechanism was proposed for these viruses on the basis of these findings, SFV by Helenius et ul. (1980), influenza virus by Maeda and Ohnishi (1980), Matlin et al. (19811, and Yoshimura et al. (1982), and VSV by Matlin et al. (1982). After uptake into intracellular vesicles, these viruses fuse with vesicle membranes when the intravesicular lumen becomes acidic and release their genome into the cytosol. At first, lysosomes were proposed as the acidic compartment, since they are well known to have a pH of 4.8 (Ohkuma and Poole, 1978). After the proposal, however, a rapid acidification of prelysosomal endocytic vesicles (endosomes) was discovered in 1982 by Tycko and Maxfield (1982) and van Renswoude et al. (1982). Genome transfer by fusion with endosomes was shown shortly afterward by Marsh et al. (1983b) for SFV and Yoshimura and Ohnishi (1984) for influenza virus. In this chapter, I first review some characteristic features of membrane fusion activity for each virus and then discuss the mechanisms of membrane fusion, especially low pH-induced membrane fusion. I concentrate on the interaction of the hydrophobic segment with the target cell membrane lipid bilayer and suggest the entrance of the segment into the lipid bilayer hydrophobic core as a key step in fusion. For the hydrophobic segments containing a few acidic residues, I emphasize protonation of those residues as a requirement for the entrance. Finally, I briefly review

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

~ H V J

:,

-a-."

__---

v

263

___---

o

b IFV

c SFV

d VSV

PH Fic. 2. pH dependence of membrane fusion activity of enveloped viruses: envelope fusion (13). hemolysis ( 0 )and . cell fusion (A). Curves were derived from the following data: inlluenra virus envelope fusion and cell f u \ i o n with MDCK cells IYoshimur'u c'f ( I / . . 19x2): SFV envelope fusion with BHK-21 cells (White c! o / , , 1980) and fusion with BHK-21 cell\ (White c ~ f [ I / . . 19x1): VSV envelope fu5ion and hernolysis with lrypsinized erythrocytes (Yamada and Ohnishi. 1986) and fusion with MDCK cells (White et ( I / . , 1981). (Modified from Ohnishi and Yoshimura. 19x4.)

the entry pathway of virus into cells leading to infection. Several review articles have been published on virus membrane fusion activity (White r t a / . ,1983; Ohnishi and Yoshimura, 1984; Asano and Asano, 1984; Ohnishi, 1985a) and virus entry mechanisms (Helenius t r f i l . , 1980: Dimmock, 1982; Marsh, 1984). II. MEMBRANE FUSION ACTIVITY OF ENVELOPED VIRUSES

A. HVJ or Sendai Virus

I . ENVELOPE FUSION Fusion of HVJ with erythrocyte membranes was first observed by electron microscopy as a dispersal of viral antigen into the cell membrane (Howe and Morgan, 1969). Virus-induced hemolysis has been used as a

264

SHUN-CHI OHNlSHl

rapid, sensitive, and convenient assay for envelope fusion. However, this assay is indirect since it measures a result of envelope fusion. A different approach based on intermixing of viral lipids with target cell membrane lipids was developed (Maeda et al., 1975; Ohnishi, 1985a). Virus is incubated with spin-labeled phospholipids to incorporate them into its envelope. The spin-labeled virus is then incubated with target cell membranes. The ESR peak height increases on fusion due to dilution of spin-labeled phospholipids with cell membrane lipids. Assays of fusion using fluorescent probes based on the same principle were also developed (Wyke at al., 1980; Struck et al., 1981). A recent example is to use octadecyl rhodamine B chloride loaded into virus envelopes and to measure the relief of fluorescence quenching on fusion (Hoekstra at al., 1984). These spectroscopic methods provide rapid, continuous, and quantitative information on fusion. A drawback is the difficulty in discriminating fusion from exchange of lipids between two membranes without fusion, if it occurred. Envelope fusion is a rapid reaction reaching a saturation level within 12 min at 37°C. It does not require Ca2+or Mg2+.It has a characteristic temperature dependence; practically no fusion below 20°C and progressively faster fusion at higher temperatures. The temperature dependence was correlated with the onset of the segmental and rotational motion of the viral glycoproteins in the envelope as measured by the decay of transient dichroism of eosine triplet probes (Lee et al., 1983). A discontinuity in the fluidity of target erythrocyte membranes was also observed near 20°C using spin-labeled PC (Tanaka and Ohnishi, 1974). The fusion reaction follows first-order kinetics. The rate constant of fusion with erythrocyte membranes was obtained as 0.84 min-' at 37°C by the spin-label assay (Kuroda et al., 1985). Kinetic analysis suggested that not all virus particles bound on cells fuse at this rate constant but only about 1-2 particles per cell do. Kinetic analysis of the fluorescence dequenching data based on the mass action law gave 2.4-4.2 min-' for the rate constant (Nir et al., 1986b), which is around 3-5 times greater than that obtained by the spin-label method. The pH dependence of fusion is rather broad in the range from pH 8 to 5 (Fig. 2a). The fluorescence assay also showed a broad pH dependence, although the rate constant as well as the efficiency of fusion are considerably low on either sides of neutral pH (Hoekstra et al., 1985). HVJ can fuse with liposomes containing as well as not containing receptors (glycophorin or gangliosides). Fusion with liposomes containing anionic phospholipids is more efficient. The effect of cholesterol on fusion is controversial. While Haywood and Boyer (1984) observed no effect of cholesterol, Hsu et al. (1983) showed a requirement in fusion and Kundrot

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

265

et a / . (1983) demonstrated a requirement in virus-induced lysis of lipo-

somes containing glycophorin. The role of cholesterol in viral fusion activity has been reviewed recently (DuzgiineS, 1988). Liposomes reconstituted with the viral glycoproteins HN and F have hemolytic and fusogenic activities (Hosaka and Shimizu, 1972; Volsky and Loyter, 1978a). Ozawa and Asano (1981) showed that cholesterol was required for the functional reconstitution.

2. HEMOLYSIS HVJ causes hemolysis. The virus appears to have an inherent defect in the envelope membrane which creates pores in the cell membrane after fusion. Accelerated water inflow causes cell swelling and lysis after a certain threshold. Pore formation is observed as the permeability to low molecular weight (<1,000) compounds increases (Pasternak, 1984) and the membrane potential or resistance decreases (Okada et al., 1975). Membrane damage in cultured cells is repaired on further incubation at 37°C (Okada et a/., 1975). Ca2+inhibits pore formation, thus inhibiting virus-induced hemolysis (Pasternak, 1984). Virus harvested I day after infection of embryonated eggs (one cycle of reproduction) is nonhemolytic, in marked contrast to hemolytic viruses obtained after 3 days (Homma ei ul., 1976). In electron micrographs, uranyl acetate stain does not penetrate the early harvested viral envelope, whereas it penetrates the late harvested viruses (Shimizu et ul., 1976). The early harvests have fusion activity as well as infectivity, in spite of the lack of hemolytic activity. The virus therefore appears to have less damage in the envelope and cannot produce pores in the cell membranes. Interestingly, early harvests become hemolytic on freeze-thawing, incubation at 36"C, or sonication (Homma et a / . , 1976). A morphological study was carried out to obtain the structural basis for the difference (Kim et al., 1979). Both early and late harvests have closely packed spikes on the envelope consisting of HN and F. However, there is a marked difference between t em in freeze-fracture electron micrographs. While late harvests have large IMPs with diameters of 15 nm on the E-face, early harvested viruses have no visible particles on either fracture face. Large IMPs appeared on the E-face on incubation at 37°C. Nucleocapsid strands are regularly folded under the envelope in early harvests but become irregular and detached from the envelope on incubation. The invisible to visible transformation of IMPs could be due to aggregation of viral glycoproteins in the envelope. The peripheral matrix (M)proteins may also be involved in the transformation in this case.

P

266

SHUN-ICHI OHNlSHl

3. CELLFUSION Cell fusion induced by HVJ was discovered by Okada (1958) (see Okada, Chapter 10, this volume). Cell fusion is initiated by viral envelope fusion with the cell membrane, but requires factors, such as ATP, in addition to the latter process. It is inhibited by cytochalasins and high concentrations of mono- and disaccharides (Maeda et a / . , 1977), whereas envelope fusion is not affected by these agents. Envelope fusion produces intercellular cytoplasmic bridges, and the ensuing osmotic swelling of cells causes expansion of the locally fused sites to form spherical polykaryons (Knutton, 1979). This sequence was clearly shown using nonhemolytic viruses. The fusion reaction stopped at the initial stage with these viruses because of the lack of cell swelling. When the cells were subsequently swollen by lowering the osmolarity of the medium, they formed spherical polyerythrocytes (Knutton, 1979; Knutton and Bachi, 1980). Polyerythrocyte formation is very efficient with erythrocytes but extremely inefficient with erythrocyte ghosts because of the lack of osmotic swelling in the latter. Sekiguchi and Asano (1978) have succeeded in attaining a large enhancement of polyerythrocyte formation by preloading bovine serum albumin (5%) into ghosts. 4. EFFECTON TARGET MEMBRANES

HVJ causes clustering of IMPs with the accompanying formation of a naked lipid bilayer area in erythrocyte (Bachi et al., 1973) as well as in cultured cell membranes (Kim and Okada, 1981). The clustering in cultured cell membranes disappears on continued incubation at 37°C but persists in erythrocyte membranes. The clustering is dependent on the temperature at which the membrane specimen was frozen. It is observable when quenched from lower temperatures (0-4°C) but not on quenching from higher temperatures (28-37°C) (Volsky and Loyter, 1978b; Kim and Okada, 198I). This is essentially a reversible thermotropic phenomenon triggered by mobilization of the membrane proteins by the action of the virus (see Section II,B,3). The clustering of IMPs is well correlated with cell fusion, and it is not observed under conditions where the fusion of cultured cells is inhibited (Kim and Okada, 1981). For erythrocytes, this correlation is further confirmed by an experiment using ghosts loaded with bovine serum albumin. HVJ caused efficient fusion of these ghosts, but, when antispectrin antibody was additionally loaded in ghosts, both cell fusion and IMP clustering were inhibited (Sekiguchi and Asano, 1978). This result also indicates a correlation between IMP clustering and the cytoskeletal spectrin meshwork as well.

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

267

5 . AMINO-TERMINAL HYDROPHOBIC SEGMENT OF Fl

The amino-terminal segment of Fl consists of about 20 hydrophobic amino acids, 26 hydrophobic residues with an anionic Asp at position 27 and a positive Arg at 29 in HVJ, and 19 hydrophobic residues with a positive Lys at 20 in respiratory syncytial virus (a pneumovirus) (see Table 11). The sequence is well conserved among paramyxoviruses, but the homology with pneumovirus is poor. These hydrophobic segments should be able to approach target membranes in the fusion reaction. Partial exposure of this segment to external media was shown by using various proteolytic enzymes (Asano er al., 1980). Treatment of HVJ with aminopeptidase M resulted in a change of the F1 amino terminus from Phe to Ala, yet in no apparent change in the mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), suggesting that 3 (or 9) amino-terminal residues were cleaved off. Since the treated virus retained infectivity, these terminal amino acids may not be essential for fusion activity. Presence of receptors for the amino-terminal segments in target membranes has been suggested from studies on inhibition of virus replication by small peptides with amino acid sequences similar to that of the viral amino terminus (Richardson et al., 1980; Richardson and Choppin, 1983). and Z-D-PheCarbobenzoxy (Z)-D-Phe-L-Phe-Gly-D-Ala-D-Val-D-Ile-G1y L-Phe-Gly are the most potent inhibitors of measles virus replication. Other peptides with 1-3 amino acids are also inhibitory, though weaker. These peptides bind to and express their inhibitory activity on cells but not on virus. However, the recognition is limited to only 2-4 residues. Even Z-D-Phe and Z-1.-Phe had inhibitory activity. Peptides with natural amino termini have much weaker activity, 1/10 to 1/10,000 of that of peptides with the artificial carbobenzoxy group. D-Amino acids are not contained in the viral peptides. The results therefore do not appear to prove the presence of a specific receptor. Moreover, the results are not compatible with the observation that the cleaving off of afew residues from the amino terminus does not affect infectivity (Asano et al., 1983). Asano and Asano (1985) proposed a specific binding of the amino-terminal segment of F1 to cholesterol in target membranes. B. Influenza Virus

1. MEMBRANE FUSION ACTIVITY a. Virus. Low pH-induced envelope fusion, hemolysis, and cell fusion were observed by Maeda and Ohnishi (1980), Huang et al. (19811, Lennard and Miller (1981), White et al. (1981), and Yoshimura et ul. (1982).

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

269

Fusion is activated at pH values lower than 6 and becomes optimum at pH S (Fig. 2b). An electron micrograph showing fusion with erythrocytes at acidic pH is given in Fig. 3B. Proteolytic cleavage of HA0 into HA1 and HA2 is required for fusion activity (Huang et a/., 1981; Maeda et al., 19811, indicating an essential role for the low pH fusion activity in infection. Envelope fusion is a rapid reaction, reaching a saturation level of 6070% fusion in 1-2 min at 37°C at pH 5.2. Temperature dependence of the hemolytic activity correlated well with that of the rotational mobility of HA glycoproteins. The mobility thus appears to be required for the molecular rearrangements necessary for fusion activity (Junanker and Cherry. 1986). Not only various strains of type A and B influenza viruses (Shibata et d.,1982) but also type C viruses (Ohuchi r't a/., 1982; Kitamae ef a / . , 1982) have low pH-induced fusion activity. The pH dependence is a little different among them. As a conventional measure to show the difference in the pH dependence, the pH at half-maximal fusion, pH,,?. is taken (Table 111). It ranges from 5.1 to 5.7 for most strains. 6 . Variants and Miifants. Wild-type viruses may contain variants with different pH characteristics. Such variants were isolated from the X-3 I strain. Variants with a common substitution at position 17 in HAI, Arg for His, showed a large shift (+0.6 unit) in the pHli2value (Rott et uf., 1984) (see Table 111). Another variant showed a smaller shift ( + 0 . 3 unit). A substitution at position 132 in HA2, Asn for Asp, is responsible for this shift (Doms et d . , 1986). Daniels rt al. (1985) selected mutant viruses with altering pH dependence by growing virus in cells treated with amantadine chloride, which raises the endosomal pH. Thirty-eight mutants had pHli2values from 0.1 to a maximum of 0.6 unit higher than the parent viruses, X-31 and

Pi(;. 3. Interaction of inHuenra vii-us with erythrocyte\ ( A - F j and MDCK cells ( C i , H ) . ( A . 13) Envelope fusion occurs at ncitlic ptl hut not :it neutral pH. Virus wits ahsorheil onto erythrocyte\ and incubated for 10 nlin ;it 37°C ;It pH 7.2 ( A ) or 5.2 I R ) . The viral glycoproteins were visunlized using ferritin-conjug~ttccl :intibody. tC-F) Virus affects IMP distrihution in the E-fiicc. Virii\ ( D . E) or H A rosette\ (I-) wci-e adsorhcd onto erythrocyte\ and incuhated at pH 5 . 2 at 37°C for I min ( I ) ) or 10 min ( E . F ) , A control (c')wa\ incuhatcd at acidic pH without virus. ( G . H ) Endocytosis o f vii-us occurs in MDC'K cells. Virus wa\ adsorhed onto cells and incuhated at neutral p t i ; i t 37°C for 4 min ( C i ) or iit 20°C' for 30 min ( H ) . Virus in ii coaled pit. coated vesicles. itnil endowmcs ;ire ohscrvcd. Under thew incubation conditions, most virus particles are not yet transported into secondary lysosomes, but their genome has already been released into the cytoplasm to initiate reproduction of viral materials. Bar, 0.1 p m . (Photographs A-F are rrom Yoshimura C I o l . , 1985. and G from Yoshimura 6'1 d., 1982.)

LAW

T A B L E 111 pH-INDUCEDMEMBRANEFUSION IN

Influenza virus Type A PK/X/34 H I N I WSN H l N l Japan/305/57 H2N Mutant (HA2:4G+E) Mutant (HA2:I IE+G) Mulant (HA2:4G+E. I I E+G) Asia/M/57 H2N2 Hong Kong/l/hX H3N I X31 H3N2 Variant ( H A I : I44G-D. 2ISP-.L. HA2:132D-.N) Variant (HAI:17H+R) M i t tant ( H A 2:9 F- 1. ) FPV/Kostock/34 H I N I Chicken/Cerniany/49 H2N I Chicken/Germany/34 H7N I Mutant (HA1:221P+S, HA2:114E-K) Equine/Miami/l/6X HZN? Swine/l976/31 H l N l

ENVEI.OPED

VIRUSES

S.6(h. c). 5.7 5.5. 5.4 (el. 5.3 S.I(C). 5 . 2 ( C ) 5.5 ( c ) 5.1 ( c ) 5.5 ( c )

5.5 5.2 5.3. 5.7, 5.8 5.6 6.4 6.4 6. I 5.4 (el. 5.5 ( c ) .

I. 2 3. 3. 4 5. 6 6 6 6 2

2 7, 8. 9 7

8 9 2, 10. 5 II

5.5 ( I ) 5.4 5.7 6.2 5.6 5.7

9 9

5.2 6.2 6.2 6.2

12 12 12 12

5.6. 5.7. 5.9 6.4, 6.2 (el, 6.0 ( c ) , 6.3 (c), 6.0 (1) 5.1 (c)

13 14. 15. 5. 16. 17

9

2

2

Type H 1.ee/40 KagoshimaihX Hong Kimg/73 Amegusa/78 Influenza C virus, JJ/50 Semliki Forest virus

Mutant (fus-I) Sindhis virus Yellow fever virus West Nile virus Vesicular stomatitis virus. New Jersey Rabies virus Mouse mammary tumor virus IA Crcwse virus

16

5.6 ( C )

IX

5.6 6.X (1) 6.2, 6.3 (h,c). 6.1 (e), 6.1 (C). 6.2-6.35 (I) 5.8 5.5 ( c ) 6.4 ( c )

19

20 21, 22. 10. 5. 22 21 23 24

“ The p H VIIIUL‘ at which a half maximal fusion occurs. Values with letlcrs e . c . h. and I in parenthesis ai-c for envelope fu\ion. cell fwion. hemoly\is. ;ind fusion with liposonie\. re\pcctivcly. Other v ~ i l t i c \itrc

for. hcmol y s i b . Key 10 rcferencc\: ( 1 1 Maed;t and Ohnishi (1980).( 2 ) Huang cf t i / . ( 1981). ( 3 ) Yoshimtiria c’/ rrl. (19x2). d. (19x2). ( 5 ) While c’f crl. (19x1). (6)Gething c’f t r / . (19x6). (7) hims i’t rrl. (IY86). ( 8 )Rot1 i’f crl. (19x4). (9) Daniels e’r ell. (19x5). (10) Matlin c’t crl. (19x2). ( I I)White ef id. (19x21. (12) Shihata c’f id. (1982). (13) Ohuchi c’f ti/. (19x2). (14) Vgiiniincn and Kiiiirii$intn (I97Y). ( I S ) White r’f c d . (19x0). (16) Kielian e f d.(IYX4).(17) White and Helcnius (IYX(1). ( I X ) Fdward und Brown (19861. (19)Cammack and Gould (19X.S). (20) Gollinh and I’orturfirld (IYXh). (21) Mifune ef rd. (lYX2l. ( 2 2 ) Yainada i d Ohni\hi (1986). (23) Kedmond c’l d . (19x4). (24) Gonzales-Scarano t’f ti/. (1984). ”

(4) Kitdniilc c/

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

271

Chicken/Germany/34 strains. Mutants with the highest pHliz value (6.4 and 6.2) are listed in Table 111. Destabilization of the location of the HA2 amino-terminal segment and alteration of the intersubunit contacts are related to the altered pH dependence. Gething et al. (1986) constructed three mutants by site-specific mutagenesis that introduced amino acid changes in the HA2 amino-terminal segment (see Table 111). A mutant substituting Glu for Gly at position I I had the same fusion activity as the parent virus, Japan/305/57. Substitution of Gly for Glu at position 4 raised the threshold pH by 0.3 unit and decreased fusion efficiency. Substitution of Gly for Glu at position I destroyed fusion activity down to pH 4.8. c. HA and BHA. HA solubilized from the virus with detergent forms rosettelike micelles on removal of the detergent. HA rosettes bind to the cell surface receptor and induce hemolysis at acidic pH in a similar manner to the parent virus (Sato et al., 1983). They can also cause fusion of erythrocyte membranes at low pH, when the membranes are brought into close contact by poly(ethy1ene glycol) (5%) (S. B. Sato et al., unpublished), as well as fusion of liposomes (Wharton et al., 1986). Recently, it has been shown that a synthetic 20 amino acid peptide with the same sequence as that of HA2 N-terminal segment induced fusion of egg PC vesicles at acidic pH (Murata et al., 1987b; see Section III,B,2,b). HA is a member of a family of glycoproteins which contain a covalently linked fatty acid (Schmidt, 1983). A possible involvement of the fatty acid in the hemolytic or fusion activity was shown by Schmidt and Lambrecht (1985). These authors observed loss of hemolytic activity of influenza virus and HA rosettes on removal of fatty acid with hydroxylamine. Bromelain treatment of the virus releases the ectodomain of HA by cleaving the peptide bond at residue 175 in HA2. BHA is soluble in water but aggregates to form micelles in acidic media. BHA binds to the cell surface receptor but cannot induce hemolysis at acidic pH (Sato et al., 1983),

d . Interaction of HA with Liposomes. Interaction of HA with receptor (gang1ioside)-containingliposomes was studied by measuring the fluorescence of Trp residues (Ohnishi, 198%; Kobayashi et al., unpublished). After allowing the binding of HA to liposomes and washing, the liposome suspension was acidified, neutralized, and treated with semialkali protease. HA1 was digested and HA2 remained bound. Liposomes were isolated, subjected to further extensive proteolysis by bromelain or proteinase K, and centrifuged in a sucrose density gradient. Isolated liposomes

272

SHUN-ICHI OHNlSHl

still contained a small peptide(s) which gave fluorescence due to the Trp residue at 337 nm. The fluorescence was not quenched by the aqueous quencher acrylamide, but was greatly quenched by spin-labeled stearates incorporated into liposome membranes. These results suggest that a peptide portion of HA2 containing a Trp residue(s) has entered into the lipid bilayer at acidic pH. e . HA-Reconstituted Vesicles. HA reconstituted in PCkholesterol (2 : 1) vesicles can have the same fusion activity as the parent virus (Kawasaki et al., 1983). Fusion efficiency was dependent on the spike density

in the reconstituted membranes. Preparations at appropriate protein-lipid ratios produced vesicles with a high spike density that fused with erythrocyte membranes at pH 5.2 as rapidly and efficiently (63-66%) as the parent virus.

f. Fusion with Liposomes. Influenza virus can fuse with liposomes, either containing or not containing receptors, in the same way as with cell membranes. Fusion does not require specific classes of phospholipids, although fusion is more efficient with PS-containing liposomes than with PC alone. Cholesterol does not significantly affect fusion (Maeda et ai., 1981; White et al., 1982). Fusion with cardiolipin liposomes was more efficient than that with PS (Stegmann et al., 1985). The fusion rate constant was obtained as 1 sec-' at 37°C at pH 5.0 from the analysis of fluorescence dequenching data (Nir et al., 1986a). Liposomes made only of cardiolipin could be considered to be artificial and to have different characteristics than natural membranes as target membranes for fusion (Stegmann et al., 1986). CONFORMATIONAL CHANGE I N HA 2. Low PH-INDUCED The three-dimensional structure of BHA at neutral pH was determined to a resolution of 5 A (Wilson et ul., 1981)(see Fig. 4). HA molecules form trimers and extend 13.6 nm from the envelope surface. The receptor binding site is located in HA1 near the upper surface of the molecule. The amino-terminal hydrophobic segment in HA2 is located in a different domain near the envelope surface and hidden inside the trimer. When the virus binds to the cell surface receptor, the two membranes are still far apart. A large conformational change is therefore required at acidic pH to bring the two membranes into close contact and to make the hydrophobic segment exposed and approach the target membrane. Such a structural change has been observed by various methods.

FIG.4. Three-dimensional structure of influenza viru4 glyco-protein H A . The protciw form a trimer in the envelope. The receptor binding site ( ) and the putative fusogenic hydrophobic segment (HA2 amino terminus) are indicated. (Adapted by permission from Wilson P I ul., 1981.)

274

SHUN-ICHI OHNlSHl

BHA and HA are resistant to trypsin at neutral pH but become susceptible after low pH treatment. The pH dependence of the susceptibility agrees well with that of envelope fusion (Skehel et al., 1982; Sat0 et al., 1983). The cleaved sites were determined to be at Lys 27 and Lys 224 in the HA1 component (Skehel et al., 1982). The conformational change appears to expose the HA2 amino-terminal hydrophobic segment. Aggregation of BHA molecules to form micelles at acidic pH is likely to take place via hydrophobic association, placing the amino-terminal segments near the center. Treatment with thermolysin removed the amino-terminal 23 residues and caused resolubilization (Daniels et al., 1983). BHA binds to phospholipids or detergents at acidic pH when these amphiphilic molecules coexist, probably by association with the exposed hydrophobic segment. The pH dependence of the binding is the same as that of envelope fusion. Electron microscopic observation of the isolated virus and of HA showed a thinning and marked elongation of HA in acidic media. Trypsin treatment reversed the elongation, resulting in a shortening of HA (Ruigrok et al., 1986). Low pH treatment of BHA did not signficantly affect the CD spectrum in the far UV region, indicating a very small, if any, effect on the secondary structure. Therefore, the structural change may not involve a gross denaturation but a relative movement of domains which maintain their individual structures (Skehel et al., 1982). The low pHinduced phenomena are apparently irreversible on restoring the pH to neutral. HA molecules form micelles at neutral pH. This association is probably due to hydrophobic clustering of the membrane-spanning hydrophobic segments near the HA2 carboxy terminus. Low pH causes aggregation of these micelles, probably via hydrophobic association of the HA2 aminoterminal segments. HA molecules interact strongly with target membranes probably via the exposed hydrophobic segment, and cause lysis and fusion. Entrance of a Trp-containing peptide segment in HA2 into the lipid bilayer at acidic pH has been suggested (Section II,B, I ,d).

MEMBRANES 3. EFFECTON TARGET Influenza virus causes IMP clustering in erythrocyte membranes when incubated at pH 5.2 but not at pH 7.2 (see Figs. 3C-E) (Yoshimura et al., 1985). Clustering was dependent on the temperature of quenching, as in the case of HVJ (Section II,A,4). At 37"C, protein distribution was more random. The clustering is a reversible thermotropic aggregation of membrane proteins, and the effect is similar to that of HVJ, but much more

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

275

extensive (see, for example, Fig. 3E). The difference is not due to the incubation of erythrocytes at acidic pH, since the clustering induced by HVJ at pH 5.0 was similar to that at pH 7.4. HA rosettes also cause IMP clustering in erythrocyte membranes at pH 5.2, although the degree of clustering is not so large (Fig. 3F), similar to that caused by the virus after incubation for a short period, e.g., I min (Fig. 3D). IMP clustering is probably a result of virus-induced mobilization of target membrane proteins as observed by fluorescence photobleaching recovery measurements (Yoshimura et ul., 1985). Lateral mobility of band 3 proteins in erythrocyte membranes is largely restricted under physiological conditions; the diffusion constant is of the order of cm? sec-I. Incubation of erythrocytes with influenza virus or HA rosettes at pH 5.2 caused a large mobilization of band 3 proteins, the diffusion constant increasing 50- or 20-fold, respectively. No such mobilization was observed when the cells were incubated at pH 7.4 or when ghosts were incubated at pH 5.2 without virus. The mobilization may be related to modification of the cytoskeletal meshwork underneath the membrane, consisting of spectrin-actin-band 4.1 peripheral proteins. A diffuse fibril structure was converted into aggregated dense spots, and membrane regions lacking the meshwork were produced after incubation with virus at pH 5.2. A model for the restriction by the cytoskeletal network has recently been presented (Tsuji and Ohnishi, 1986). There are two populations of band 3 proteins. About lO-I5% of them, calculated as band 3 monomers, are anchored to the cytoskeletal network via ankyrin and are immobile. Ankyrin-free band 3 proteins are mobile, but the mobility depends on the spectrin associated state (tetramer or dimer). Band 3 cannot move when its cytoplasmic domain is surrounded by the tetrameric spectrin network, but it can cross the network when spectrins are in dissociated dimers. The virus-induced mobilization could therefore arise from modification of the spectrin association state by direct or indirect action. C. Semliki Forest Virus Low pH-induced hemolytic activity of SFV was first observed by Vaananen and Kaariainen (1979; 1980). Envelope fusion with various cultured cells was very rapid (within 5 sec) and efficient (7040%) at pH 5.5 at 37°C (White ef d.,1980). Virus-induced cell fusions were also observed (White et ul., 1981). The pH dependence of fusion activity is shifted a little higher than that of influenza virus (Fig. 2c). The pHI,2 value ranges from 6.0 to 6.4 (Table 111). A fusion mutant had a lower value of

276

SHUN-ICHI OHNlSHl

5.1. Sindbis, another alphavirus, had a pHI12of 5.6. Low pH-induced fusion activity was also observed for yellow fever virus (Cammack and Gould, 1985) and West Nile virus (Gollins and Porterfield, 1986), both belonging to a different genus, Flauiuirus (Table 111). SFV can fuse with simple liposomes in a way similar to fusion with cell membranes. In these fusions, cholesterol in the target membranes is required, a characteristic difference from many other viruses (White and Helenius, 1980). Maximum fusion was observed with more than 33 mol % cholesterol. The phospholipid requirement was not strong, although fusion with PC was less efficient compared with other classes of phospholipids such as PE and PS. Ca2' or Mg2+was not required for fusion (White and Helenius, 1980). Isolated E glycoproteins form octameric micelles which, however, do not have hemolytic activity (Vaananen and Kaariainen, 1979). The glycoproteins reconstituted into liposomes showed low pH-induced hemolytic and fusion activities (Marsh et al., 1983a). The fusion efficiency was lower (25%) than that of the parent virus, possibly due to lower spike density on the reconstituted membranes. Cholesterol was not required for reconstitution. Of the three components of E, El was shown to be active for hemolysis and hemagglutination. Yamamoto et al. (1981) solubilized E glycoproteins from Western equine encephalitis virus, an alphavirus, and isolated the components. Reconsituted vesicles containing E I alone but not those containing E2 alone showed fusion activity. It is suggested that E2 may cooperate with El in the hemagglutination activity. Both El and E2 span the viral envelope membrane, with the hydrophobic segment near their carboxy termini. Another hydrophobic segment near but not at the amino terminus of E l , residues 80-100 or 80-109, may be fusogenic (Table 11). There are a few acidic residues in the segment. Sindbis El has a quite similar sequence at the same site. A low pHinduced conformational change in Sindbis El was detected by the change in susceptibility to tryptic cleavage (Edwards et al., 1983).

D. Vesicular Stomatitis Virus Low pH-induced hemolysis and cell fusion were observed in VSV by White et al. (1981) and Mifune et al. (1982). The pH range for fusion activity is similar to SFV (Fig. 2d), the pHlI2value ranging from 6.1 to 6.4 (Table 111). A characteristic of VSV different from other viruses is that protease pretreatment of target cells does not destroy but rather enhances virus binding and fusion. The human erythrocyte is an extreme example: the virus cannot agglutinate nor fuse with intact cells but shows enhanced

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

277

hemagglutination and fusion activities to trypsin-treated cells (Mifune el al.. 1982). Fusion with trypsinized erythrocytes is very rapid (within 1-2 min) and efficient (-80% fusion) at pH 5.5 at 37°C (Yamada and Ohnishi, 1986). Pretreatment of L cells with trypsin or neuraminidase also results in increased binding and plaque formation (Schloemer and Wagner, 1975). These results strongly suggest that sialoglycoproteins and sialoglycolipids as well as proteins are not receptors. Instead, lipids may be the direct target site. Inhibition of virus binding and infection by PS liposomes supports this idea (Schlegel et d.,1983). VSV can bind and fuse with simple liposomes (Yamada and Ohnishi, 1986). Fusion with liposomes is quite similar to that with trypsinized erythrocyte membranes, in the rate, efficiency, and pH dependence. Binding to liposomes is also dependent on pH; the threshold pH (7.5) is about I unit higher than that for fusion. Binding and fusion with liposomes did not decrease at lower pH values, while those with cells decreased. Binding to liposomes was dependent on the head groups of the phospholipids, being stronger to phosphoserine than to phosphocholine. On the other hand, the head group requirement was not strong for fusion, but the presence of cis-unsaturated fatty chains in phospholipids was required. For example, efficiency of fusion with cis-unsaturated dioleoyl PC was 5296, but that with saturated dimyristoyl PC was negligibly small, 8%. Cholesterol enhanced fusion further but was not an absolute requirement. Efficient fusion with various natural phospholipids and lipids is probably due to the presence of cis-unsaturated fatty acids. lsolated G glycoproteins do not have hemolytic activity by themselves; however, G glycoproteins reconstituted into vesicles have fusion activity (Eidelman et a / ., 1984). Reconsituted vesicles prepared at high protein/ lipid ratios (-0.3 mol 96 protein) and by slow removal of detergent had high fusion activity. Fusion with liposomes containing PS or PE are much faster than that with PC alone. Inclusion of cholesterol did not affect the fusion. However, it is noteworthy that the pH range for fusion shifted largely to lower values, with a threshold at pH 5.0. A n alternative reconstitution procedure has been used recently to produce reconstituted vesicles whose fusion activity has a pH dependence similar to that of the intact virus (Metsikko et al., 1986). The amino-terminal segment of G protein is not hydrophobic, containing 3 basic Lys and 2 His residues. Residues 102-131 have sequence homology between lndiana and New Jersey serotypes and appear more hydrophobic (Table 11). However, a thermodynamic estimate shows that they are not hydrophobic as a whole (Section IIl,B,2). Residues 175-199 can be hydrophobic at acidic pH (see Table IV). Peptides of 26 amino acid residues corresponding to the amino-terminal

278

SHUN-ICHI OHNlSHl

segment of G can cause hemolysis at low pH, with a pH dependence similar to that for the virus (Schlegel and Wade, 1984, 1985). Even much smaller peptides (6 amino acid residues) caused more efficient hemolysis. However, the activity was independent of pH. Therefore, the activity may not reflect that of the parent virus. Low pH-induced fusion activity has also been observed for other enveloped viruses, a retrovirus (MMTV) and a bunyavirus (La Crosse) (Table 111). A low pH-induced conformational change in La Crosse virus GI glycoprotein has been reported (Gonzales-Scarano, 1985).

111.

MECHANISM OF FUSION

A. Binding and Close Apposition

In the fusion reaction, virus first binds to a target cell membrane through interaction of the viral glycoprotein with the receptor (Fig. la). The virus and cell membranes would still be far apart since the viral glycoproteins usually extend externally some 10 nm or more (Fig. 4). Closer apposition of the two membranes is required for further interactions. In influenza and other viruses, the close apposition may be achieved by a conformational change of the viral glycoprotein induced at acidic pH. The conformational change would cause mobilization and rearrangement of target cell membrane proteins to produce naked lipid bilayer domains assisting the close apposition. The conformation change would also cause exposure of the hydrophobic segment in the glycoprotein to make it approach and interact with the lipid bilayer in target membranes. (Fig. Ib). Morphological observation of isolated influenza virus and HA showed a thinning and marked elongation of HA at low pH (Ruigrok et al., 1986). However, this is the change induced between HA spikes in the absence of binding to target membranes, and change in HA when bound to target membranes may be different. The conformational change may be triggered by protonation of some acidic amino acids. The conversion of charged residues into protonated neutral ones would greatly alter interresidue interactions. The pK, value of mildly acidic residues in proteins lies in a range 5.6-7.0 for His, 3.0-4.7 for Asp (P-carboxyl), -4.5 for Glu (y-carboxyl), and 3.0-3.2 for the carboxy-terminal a-carboxyl group (Dawes, 1980). The pK, value may deviate for some specific residues. For example, in lysozyme, Glu 35 has a much higher value of 6-6.5 and Asp 66 a much lower value, 1.5-2, while the other Asp residues 52 and 101 have ordinary values, 3-4.6 and 4.2-

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

279

4.7. respectively (Imoto ef ul., 1972). Since the conformational change of viral glycoproteins occurs at pH values lower and higher than 6.0, Asp, Glu, and His residues are possible sites for the protonation. Viruses can also bind and fuse with simple liposomes lacking receptors. The mode of binding should be different from that of binding to the receptor-containing membranes. For example, the site in the viral glycoprotein used for the binding should be different. The pH dependence may therefore be different between these two types of binding. Such differences have been demonstrated: while binding of influenza virus to cell membranes occurs in a wide range of pH values from 8 to 5 ( M a t h et a l . , 1981 : Yoshimura et al., 1982), binding to liposomes occurs only below pH 6.2, in parallel to the fusion activity (Doms et al., 1986). The binding to liposomes may involve interaction of the HA2 amino-terminal hydrophobic segment, exposed at the low pH, with the lipid bilayer. Both events are thus effected by the same cause. On the other hand, binding to receptors on target membranes occurs at neutral pH and also at acidic pH so long as the conformational change in HA does not destroy the binding site in HA1 . No gross denaturation of HA at low pH was indicated by CD measurements (Skehel et a / . , 1982). Binding of HVJ to liposomes may be ascribed to the partially accessible hydrophobic segment in F1. 8. Interaction of the Hydrophobic Segment with the Target Cell Membrane Lipid Bilayer 1. LIPIDBILAYER DOMAIN AS T H E TARGET

The hydrophobic segment in virus fusion proteins probably attacks the lipid bilayer domain in the target cell membrane. Such domains may be formed as a result of virus-induced mobilization of target membrane proteins (Yoshimura et d.,1985). Formation of lipid domains has been detected by freeze-fracture electron microscopy. Ability of viruses to fuse with liposomes either containing or not containing receptors supports the idea, provided that such fusions mimic those with cell membranes. The presence of a specific receptor to the amino-terminal segment has been suggested but not fully proved (Section I1 ,A,5). 2. ENTRANCE OF

THE

HYDROPHOBIC SEGMENT INTO THE LIPID

BILAYER The hydrophobic segment may enter the target lipid bilayer hydrophobic core. The free energy for transfer of the segment from aqueous to lipid

280

SHUN-ICHI OHNlSHl

bilayer phases will be negative. An estimate for the free energy value was made as shown in Table IV. This analysis gave large negative values for many viral hydrophobic segments. A segment containing charged residues is not favorable for transfer because of the work necessary to bring a charge into hydrocarbon media (Parsegian, 1969). When these charged residues are neutralized by protonation or deprotonation, transfer would be much easier. It has been shown that a single substitution of a hydrophobic amino acid residue (e.g.,Ala) for a charged residue (e.g., Glu) in the signal peptide of secretory proteins blocks transfer of the proteins across the inner membrane of mutant E. coli (Bedouelle et al., 1980; Emr er al., 1980).

TABLE IV

FREEENERGIES FOR TRANSFER OF HYDROPHOBIC SEGMENTS FROM AQUEOUS TO

LIPID

BILAYER PHASES"

Free energy (kJlmol) Viral fusion protein HVJ

F1

Influenza virus HA2 AIPW8134 A/Japan/305/57 A/Aic hi12168 B/Lee/40 CICalIl8 SFV E l VSV ind G MMTV gp36

Residues

At neutral pH

At acidic pH

2-28 1-28

-74 - 66

-90 -70

2-24 1-24 2-24 1-24 2-24 1-24 2-2 1 1-21 2-28 1-28 80- 100 80- 109 101-131 175-1 99 2-27 1-27

- 52 -25 -29 -2.8 -23 +3.0 -65 -39 -51 - 36 + 24 +82 + 84 + 83 -152 - 143

-86 -48 -63 -25 -73 - 35 -97 -59 - 103 -77 -9.6 + I4 + 50 -3.3 -152 - 132

" Free energies were calculated for transfer of a-helical segments, using parameter values assigned to each amino acid residue by von Heijne f 1981). The values for transfer at neutral pH include the work to protonate acidic residues at neutral pH, while the work is omitted from the values at acidic pH since protonation occurs spontaneously. For the amino-terminal hydrophobic segments, free energy values are given for transfer of the internal segment (e.g.. 2-24) and the whole segment (e.g., 1-24), corresponding to models A and B in Fig. 5 . respectively. In the latter, a contribution from the work to deprotonate the aminoterminal charge at the respective pH values is included.

28 1

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

a . Protonution of Acidic, Residues in the Hydrophobic Segment. The hydrophobic segment in viral glycoproteins which have fusion activity only at acidic pH often contains a few acidic residues. We have proposed that the low pH is required for protonation of the acidic residues, as well as for the conformational change, to neutralize the charge so that the hydrophobic segment may interact more easily with the target lipid bilayer (Maeda and Ohnishi, 1980; Ohnishi, 198Sa).The protonation occurs spontaneously at acidic pH, but requires work at neutral pH since the latter is higher than the pK, value of the acidic residues. This extra work, 15-18 kJ/mol per acidic residue, is included in the free energy for transfer at neutral pH but omitted at acidic pH. The free energy values at acidic pH are therefore much lower (30-50 kJ/mol) than those at neutral pH (Table IV), although the latter values are also negative in most cases. The hydrophobic segment in gp36 of MMTV, another low pH-active virus, does not contain acidic residues, however. In this case, the low pH may be required only for the conformational change to expose the hydrophobic segment which is ready for interaction without protonation. The free energy value for transfer is largely negative, independent of pH (Table IV). Our point is that when the segment has acidic residues, they should be neutralized for the interaction. This leads to a hypothesis that the hydrophobic segment in viruses with fusion activity at neutral pH should not contain acidic residues, which so far holds. There are two possiblities for the entrance of the neutralized hydrophobic segment into the lipid bilayer (Fig. 5 ) : entrance of the internal hydrophobic segment between two charged residues (A) or entrance of the whole hydrophobic segment (B). Extra work is required to neutralize the aminoterminal charge in B , when it has a charge, and the free energy value for transfer is correspondingly high (Table IV). The internal segments in SFV El and VSV G proteins listed in Table I1 are not so hydrophobic (Table A

t

t

e

t

FIG.5. Model for the interaction of the amino-terminal hydrophobic segment of the viral fusion glycoprotein with a lipid bilayer: entrance of an internal (A) or a whole ( B ) segment into the lipid hydrocarbon layer.

282

SHUN-ICHI OHNlSHl

IV). The segment 80-109 in El between two positively charged residues has a positive free energy value at neutral and acidic pH. A shorter segment (80-100) can have a negative value at acidic pH, though small. The well-conserved sequence 101-131 in VSV G protein is not hydrophobic. Another segment (175-199) can have a negative free energy value at acidic pH. Whether these segments are actually involved in fusion of these viruses requires investigation. b. pH-Dependent Fusion Activity of a Peptide with the Same Sequence as That of the HA2 Amino-Terminal Segment. In order to investigate that the putative fusogenic segment actually has membrane fusion activity and also that protonation of the acidic residues is required for activity, we synthesized a 20-amino-acid peptide with the same sequence as that of the HA2 amino-terminal segment and studied its fusion activity (Murata et al., 1987b). The synthetic peptide caused a rapid and efficient fusion of egg PC vesicles at acidic pH, but not at neutral pH, in a manner quite similar to that of the parent virus (Fig. 6A). The threshold pH was around 6.0, and the pHllz value was 5.6. Fusion efficiency was dependent on the peptide to phospholipid ratio in the mixture, increasing with increases in the ratio. The low pH-induced fusion can be stopped immediately by shifting the pH to neutral and started again by readjusting to acid. The peptides with acetylated or succinylated amino termini also had similar fusion activity. The pH range was, however, shifted about 1 pH unit to the acidic side. These results clearly demonstrate a direct involvement of the fusogenic segment in fusion and strongly suggest the requirement of protonation for activity since there are no other acid-sensitive groups in this simplified fusion system.

c. pH-Dependent Fusion Activity of Succinylated Melittin. Hydrophobic peptides can have membrane fusion activity. The natural hydrophobic peptide bee venom melittin is one such example. This peptide actually caused fusion of egg PC vesicles at neutral as well as acidic pH (Fig. 6B) (Murata et al., 1987a). A succinylated derivative of this peptide was synthesized in which all four amino groups, one a-amino group and three E amino groups at lysines, were converted to carboxyl groups. This derivative caused fusion of PC vesicles only at acidic pH, in marked contrast to the parent peptide (Fig. 6B). The threshold pH for fusion was 5.2 and the pHllz5.15. The pH characteristics of fusion can thus be modified to acid sensitive by introducing acidic groups. The low pH-induced fusion by succinylated melittin can be stopped immediately by shifting the pH to neutral and started again by readjusting to acid. Protonation of the introduced carboxyl groups was studied by 13C-NMR

283

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

E

H$G L FG A IAG F I GGWTGM I 6 G-CO;

A

B ai

I G A V L ~ V L T T G L P A LI SWI kitAAaa-cotw,

H&G

bt MY-GIGAVLVVLTTGLPA L 1 SWI K & K ~ ~ Q - c o N H ~ co CO $0 co

g'E

%

=

z

& UI v) 3 ? 50

z i s o z

4

5

6

1

P"

8

4

5

6

7

PH

8

9

FIG.6. Membrane fusion activity of hydrophobic peptides: (A) 20-amino-acid peptide with the same sequence as that ofthe HA2 amino-terminal segment of influenza virus A/PK/ 8/34 and ( B ) intact (a) and succinylated (b) melittin. Fusion of sonicated egg yolk PC vesicles was assayed by the spin-label method. The niolar ratio of phospholipid to peptide in the reaction mixture was 60 ( A ) or 1.55 ( B ) . The percent fusion after 10 min at 23°C is plotted against pH.

spectroscopy using melittin succinylated with [ I ,4-i3C]succinicanhydride. Protonation in the presence of PC vesicles occurred in the same pH range as that for fusion. The pK, value for the four carboxyl groups was obtained as 5.19, in close agreement with the pHlizvalue. The agreement strongly indicates that fusion activity is induced by protonation of acidic residues. The pK, value was higher than that in the absence of vesicles, 4.8. The upward shift will occur when the protonated form is stabilized in the presence of vesicles. A probable model is entrance of the protonated hydrophobic segment into the lipid bilayer as B in Fig. 5, since it has no charge at the amino terminus when protonated. Such stabilization will cause a shift of the acid-base equilibrium in favor of the protonated form. In the I3C-NMR spectrum of succinylated melittin both in the presence and absence of vesicles, only a single peak was observed for each carboxyl group during the titration. This indicates fast exchange between the protonated and deprotonated forms, faster than 3 msec. If the pro-

284

SHUN-ICHI OHNlSHl

tonated segment entered into the lipid bilayer, then the fast exchange would include the entrance into and returning back to the surface of bilayer membranes. Such a dynamic nature of interaction can well explain the rapid on/off switching of fusion on shifting pH to neutral and acid. d . Primury Sequence versus Hydrophobicity. The above discussions are solely thermodynamic, considering only the hydrophobicity of the segment which does not require any specific primary sequence. Is the primary sequence also essential for the fusion function? Needless to say, the primary sequence determines the three-dimensional structure and the conformational change of HA at low pH. The HA2 amino-terminal segment is also involved in secondary structure formation; the terminal nitrogen and the amide nitrogen of residues 4,5, and 6 form hydrogen bonds to oxygen atoms of residue 112 (Asp) in the long helix of HA2 (Wilson et al., 1981). Mutations at these sites should alter interresidue interactions to affect stability of the structure and pH for the conformational transition (Daniels et al., 1985). However, the primary sequence may not be important for interaction with the lipid bilayer. For example, influenza C virus has an amino-terminal sequence with very little homology with A and B viruses, but has a common feature of hydrophobicity.

3. FUSIONMECHANISM a . Membrane Perturbation. Studies on the fusion of phospholipid vesicles by hydrophobic peptides containing acidic residues indicated that fusion activity was induced by protonation of the acidic residues (Section III.B,2). At neutral pH. these peptides bind to the vesicle surface but do not cause aggregation of vesicles. At acidic pH, when charges in the hydrophobic segment are neutralized by protonation, vesicles may be brought into close contact owing to the increased surface hydrophobicity, which could overcome the surface dehydration energy (Ohki, 1987). Vesicles aggregate under such conditions. Moreover, the neutralized hydrophobic segment may enter into the lipid bilayer. The segment would then compress the lipid molecules surrounding it. Since the interaction can be dynamic, the rapid entrance into and returning back to the surface would cause density fluctuations in the membrane. Such destabilization would trigger fusion of two apposed bilayer membranes. The hydrophobic segment may produce some perturbed structures in the target lipid bilayer membrane by binding to lipid molecules. We have recently attempted to detect the possible intermediate structure in liposomes incubated with influenza virus at acidic pH using the quick-frozen replica technique. On replicas made 30 sec after acidification at 23°C of

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

285

receptor (g1ycophorin)-containingegg PC liposomes bound with influenza virus, we observed very close apposition of the virus particle and liposomes and also small 8-10 protrusions with a diameter of about 20 nm on the convex fracture face, with a complementary pit on the opposite concave face, of liposomes. No close apposition nor perturbed structures were observed on replicas obtained after incubation at neutral pH. The protrusiordpit structure may be the fusion point because of disturbance due to the high curvature (Kawasaki et a / . , 1987). Cullis and Hope ( 1978) have proposed inverted micelles and other nonbilayer structures as the intermediate in membrane fusions. Phospholipids with larger tail to head volume ratios such as PE or cardiolipin spontaneously form nonbilayer structures under appropriate conditions (Cullis and de Kruijff, 1979). Even PC can have a nonbilayer structure under specific conditions (Gruner et ul., 1985). It is interesting to note that, in fusion of VSV with liposomes, the presence of cis-unsaturated fatty acyl chains in target phospholipids was required (Yamada and Ohnishi, 1986). Also, phospholipids used in fusion studies with various enveloped viruses have been those from natural sources which contain large fractions of cisunsaturated fatty acyl chains. Cis-unsaturated phospholipids have larger tail to head volume ratios. h. p H Dependence of Virus Memhriinr Fusion. The pH characteristics of virus membrane fusion activity will be determined by the pH dependencies of the conformational change of the fusion protein and the neutralization of the fusogenic segment. The two processes may occur in the same or different pH ranges. If they are different, the process occurring at lower pH would determine the overall pH profile of fusion. For example, the pHlizfor MMTV may represent that of the conformational change of gp52 + gp36. Its fusogenic segment does not contain acidic residues and can interact with target membranes at neutral as well as acidic pH. Influenza virus variants and mutants have been isolated and studied (Section II,B,I ,a). In these mutants, modification of fusion activity can be caused by changes in the pH ranges both for the conformational change and also for neutralization of the fusogenic segment. For example, a mutant with Gly at positon I 1 for Glu in the HA2 amino-terminal segment showed the same fusion activity as the wild type (Gething et al., 1986). In this case, the fusogenic segment can interact with the target lipid bilayer at the higher pH because of the lack of one acidic residue in the arm, but the conformational change of the mutant HA was shown to occur in the same pH range as the wild type. Substitution of Gly for Glu at position 1 in the arm destroyed fusion activity down to pH 4.8 (Gething et a / . , 1986). However, the pH range for fusion activity of the mutant segment may

286

SHUN-ICHI OHNlSHl

have shifted lower than 4.8. The HA2 amino-terminal peptide with a chemically modified amino terminus had such a shifted fusion activity (Section III,B,2,b). IV.

INFECTIOUS CELL ENTRY MECHANISMS

The response of cells to viruses is quite similar to that in the receptormediated endocytosis of physiological ligands such as low density lipoprotein, asialoglycoprotein, and epidermal growth factor (for review, see Goldstein et al., 1986). A characteristic difference is that the viruses fuse with cell membranes to release their genome into the cytoplasm during endocytic processing. The fusion site depends on the pH characteristics of the virus membrane fusion activity. Paramyxovirus can fuse with the cell surface plasma membrane because of its ability to undergo fusion at neutral pH. On the other hand, most other viruses fuse with the membrane of acidic vesicles after entering the cell because of their restriction to fusion at acidic pH. Various types of viruses bind to the target cell surface and coated pits, often near the foot of microvilli, and are taken up into coated and smooth vesicles after brief warming (see Figs. 3G, H for influenza virus) (for reviews, see Marsh, 1984; Dimmock, 1982). Uptake of HVJ by endocytosis was also observed, but the number of virus particles in intracellular vesicles was smaller, less than half as compared with a fusion-inactive cell-growth HVJ (Yasuda et a [ . , 1981). On further incubation at 36"C, the endocytosed viruses are transported into secondary lysosomes and subject to degradation by the action of various lysosomal hydrolytic enzymes. Excretion of the degraded materials into the external medium is observed after 20-30 min. The degradation is inhibited in the presence of lysosomotropic reagents (Helenius et al., 1980, for SFV; Matlin et al., 1981, and Yoshimura et al., 1982, for influenza virus; Matlin et al., 1982, for VSV). The intracellular fusion site for these viruses was initially assigned to lysosomes because of the well-known acidity of the lysosome lumen with a pH value of 4.8 (Ohkuma and Poole, 1978), which induces envelope fusion (SFV, Helenius et al., 1980; influenza virus, Matlin et al., 1981, and Yoshimura et al., 1982; VSV, Matlin et al., 1982). Inhibition of virus replication by "lysosomotropic" weak bases, which raise the lysosomal pH quickly, supports the idea. These reagents do not affect the binding and uptake of viruses. They inhibit virus replication when added in the early phases of infection, up to 10 min after infection, but do not inhibit when added after 30 min. The relationship between the lysosomal pH and virus replication has been studied. The intralysosomal pH can be adjusted

[Ki!l‘

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

287

8.5

t3

6.0 i; 3t

P

5.5

0.01Chloroquine 0.1

ml

5.0

O+i---%

NH&I m l

FIG.7. Influenza virus reproduction in the presence of various concentrations of acidotropic reagents. NH,CI or chloroquine. After incubation for 4 hr at 37°C in the presence of the reagents at neutral pH, the medium was replaced with one lacking the reagents, and and lysosornal pH ( 0 ) incubation was continued for a further 6 hr. Virus reproduction (0) were measured using dextran tagged with fluorescein isothiocyanate according to Ohkurna and Poole 11978). (From Ohnishi and Yoshimura, 1984.)

by changing either the concentration of the reagents or the pH of the extracellular medium, as shown in Fig. 7. Influenza virus replication was assayed under various lysosomal pH conditions. Inhibition of the replication was observed when the lysosomal pH was greater than 6.0 but not at pH values below 6.0 (Fig. 7) (Yoshimura r t al., 1982; Ohnishi and Yoshimura, 1984).The pH dependence agrees well with that of envelope fusion. Just after the publication of these studies, a rapid acidification of prelysosomal endocytic vesicles was discovered by Tycko and Maxfield ( 1982) and van Renswoude et al. (1982). The acidification is caused by a proton pump on the vesicle membrane, and the pH value can be as low as 5.0. Envelope fusion is thus possible when the viruses arrive in endosomes before reaching the secondary lysosomes. The release of the virus genome into the cytoplasm by fusion in endosomes has been shown for SFV by Marsh et al., (1983b) and for influenza virus by Yoshimura and Ohnishi (1984). These authors showed that the virus replication had already started after 5-7 min at 37”C, or after 1 hr at 2 0 T , under which conditions the viruses are still in endosomes and not yet transported into secondary lysosomes, as confirmed by biochemical assays. Yoshimura and Ohnishi (1984) measured the pH of the virus environment using HA tagged with fluorescein isothiocyanate. The pH was lowered to 5.1-5.2 after 10 min at 37OC, or after I hr at 20”C, while the viruses were in endosomes. The pH was raised to 6.7 by addition of a “lysosomotropic” reagent, NH,CI (20 mM), in the medium. Virus replication was inhibited in the presence of the reagent. Previous data on the inhibition by “lysosomotropic” reagents are not inconsistent with genome release from endosomes, since the reagents rapidly penetrate and raise the endosomal pH as

288

SHUN-ICHI OHNlSHl

well as the lysosomal pH as shown above. The pH value in the ordinate in Fig. 7 may well represent the endosomal pH. The lysosomotropic reagents are also endosomotropic, and thus are more generally termed acidotropic reagents (de Duve et d.,1974). Some populations of viruses are transported to secondary lysosomes where they can also fuse to release the viral genome, in competition against degradation. However, fusion in endosomes would contribute more to infection simply because it occurs earlier. HVJ fuses with the plasma membranes, but it can also fuse with endosomes after endocytosis since it can fuse at acidic pH as well. Other viruses may also take the intracellular uncoating route. For MMTV, La Crosse virus, the West Nile virus, the low pH-induced fusion activity and the inhibition of virus replication by acidotropic reagents (Andersen and Nexo, 1983, for murine type C retrovirus; Gollins and Porterfield, 1985, for WNV) have been observed, supporting the intracelMar route. Enveloped viruses thus utilize the host cell machinery, developed for the uptake and processing of biological materials, for their entry. The viral genome enters the cytoplasm from the acidic endosomes where the cells separate ligands from receptors and send these materials to their respective destinations. Not only enveloped viruses but also naked viruses may enter cells from acidic endosomes, as suggested for adenovirus (FitzGerald e t a / . , 1983; Yoshimura, 1985) and poliovirus (Madshus et al., 1984). A polypeptide toxin, diphtheria toxin, also enters the cytoplasm through acidic endosomes (Sandvig and Olsnes, 1980). Membrane fusions and splittings occur frequently in endocytic processing and in the transport of newly synthesized proteins, lipids, and other materials to organelles, the plasma membrane, and the external media. These membrane fusions may be mediated by some specific proteins. It would not be unreasonable to imagine that the virus fusion proteins originate from such cellular fusion proteins. ACKNOWLEDGMENTS

I am indebted to Dr. Akira Asano for his comments on the manuscript and to my colleagues and graduate students for their cooperation. REFERENCES Andersen, K . 9..and Nexo, 9. A. (1983). Entry of murine retrovirus into mouse fibroblasts. Virology 125, 85-98. Asano, A . , and Asano, K. (1984). Molecular mechanisms of virus entry to target cells. Tumor Res. (Sapporo) 19, 1-20. Asano, K . , and Asano. A. (1985). Why is a specific amino acid sequence of F glycoprotein required for the membrane fusion reaction between envelope of HVJ (Sendai virus) and target cell membranes? Biochenr. I n r . 10, 115-122.

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

289

Asano, K., Murachi, T.. and Asano, A . (1983). Structural requirements for hemolytic activi t y of F-glycoprotein of HVJ (Sendai virus) studied by proteolytic dissection. J. Bio(.hem. 93, 733-741. Bachi, T.. Aguet. M., and Howe. C. 11973). Fusion of erythrocytes by Sendai virus studied by immuno-freeze-etching. J . Virol. 11, IOW-1012. Bedouelle, H.. Bassford. P. J., Fowler. A. V.. Zabin. I . , Beckwith, J . , and Hoffnung. M . (1980). Mutations which alter the function of the signal sequence of the maltose binding protein of Esc/ieric,hia w l i . Ntr/rrrc, (London) 285. 78-81. Blumberg, B. H., Giorgi. C., Rose. K.. and Kolakofsky. D. (1985). Sequence determination of the Sendai virus fusion protein J. Gen. Virol. 66, 3 17-33 I . Bucher, D.. and Palese, P. ( 1975). The biologically active proteins of influenza virus: Neuraminidase. In “The Influenza Viruses and Influenra” (E. Kilbourne. ed.). pp. 83-123. Academic Press. New York. Camrnack, N.. and Gould, E. A . (1985). Conditions for haemolysis by flaviviruses and characterization of the haemolyh. J . Gen. Virol. 66, 2291-2296. Collins, P. L.. Huang. Y. T., and Wertz. Ci. W. (1984). Nucleotide sequence of the gene encoding the fusion (F) glycoprotein of human respiratory syncytial virus. Proc. Null. AcctJ. Sci. U.S.A. 81, 7683-7687. Cullis, P. K . , and de Kruijff. 8. (1979). Lipid polymorphism and the functional roles of lipids in biological membranes. B i o d i i t n . Biop embrane structure of Cullis, P. R., and Hope. M. J. (1978). Effects erythrocyte ghosts and the mechanism of membrane fusion. N i i / w r (London) 271,672674. Daniels, R. S.. Douglas. A. R . , Skehel. J . J.. Waterfield. I . A.. and Wiley. D. C. (1983). Studies of the influenza virus haemagglutinin in the pH S conformation. I n “The Origin of Pandemic Influenza Viruses” (W. G . Laver. ed.), pp. 1-7. Elsevier, Amsterdam. Daniels, K. S., Downie, J . C.. Hay, A . J . , Knossow. M . . Skehel. J . J.. Wang, M . L.. and Wiley. D. C. (1985). Fusion mutants of the influenza virus hemagglutinin glycoprotein. Cell 40, 43 1-439. Dawes, E. A. (1980). “Quantitative Problems in Biochemistry.” 6th Ed. p. S7. Churchill Livingstone. Edinburgh. de Duve. C.. de Barsy, T., Poole, B., Trouet. A , . Tulkens. P.. and van Hoot, F. (1974). Lysosomotropic agents. Biocliern. Pharrnc~ol.23, 2495-253 I . Dimmock, N . J . (1982). Initial stages in infection with animal viruses. J . Geri. Virol. 99, I22. Doms. R. W . , Gething, M.-J., Henneberry. J.. White. J . , and Helenius. A. (1986). Variant influenza virus hemmagglutinin that induces fusion at elevated pH. J . Vircil. 57, 603613. Diizgiineg, N. (1988). Cholesterol and membrane fusion. I n “Cholesterol in Biological Systems” (P. L. Yangle, ed.), in press. CRC Press, Boca Raton. Florida. Edwards. J.. and Brown, D. T. (1986). Sindbis virus-mediated cell fusion from without is a two step event. J . Gen. Virol. 67. 377-380. Edwards. J., Mann. E.. and Brown. D. T. (1983). Conformational changes in Sindbis virus envelope protein accompanying exposure to low pH. J . Virol. 45, 1090-1097. Eidelman. 0 . .Schlegel. R., Tralka, T. S., and Blumenthal. R. (1984). pH-dependent fusion induced by vesicular stomatitis virus glycoprotein reconstituted into phospholipid vesicles. J . B i d . Cliem. 259, 4622-4628. Emr. S. D., Hedgpeth. J., Clement, J.-M.. Silhavy. T. J . . and Hofnung, M.(1980). Sequence analysis of mutations that prevent export of A receptor, an Escherichiu coli outer membrane protein. Noturc (London) 285, 82-85. FitLGerald, D. J.. Padmanabhan, R..Pastan. I.. and Willingham, M. C. (1983). Adenovirus-

290

SHUN-ICHI OHNlSHl

induced release of epidermal growth factor and pseudomonos toxin into the cytosol of KB cells during receptor-mediated endocytosis. Cell 32, 607-617. Gallione, C. J . , and Rose, J . K. (1983). Nucleotide sequence of a cDNA clone encoding the entire glycoprotein from the New Jersey serotype of vesicular stomatitis virus. J . Virol. 46, 162-169. Garoff, H., Frischauf, A.-M., Simons, K., Lehrach, H., and Delius, H. (1980). Nucleotide sequence of cDNA coding for Semliki Forest virus membrane glycoprotein. Nutrrrr (London) 288, 236-241. Gething, M.-J., White, J. M., and Waterfield, M. D. (1978). Purification of the fusion protein of Sendai virus: Analysis of NH2-terminal sequence generated during precursor activation. Proc. N d . Acud. Sri. U.S.A. 75, 2737-2740. Gething, M.-J., Bye, J . , Skehel, J . , and Waterfield, M. (1980). Cloning and DNA sequence of double-stranded copies of haemagglutinin genes from H2 and H3 stains elucidates antigenic shift and drift in human influenza virus. Noture (London) 287, 301-306. Gething, M.-J., Doms, R. W., York. D., and White, J . (1986). Studies on the mechanism of membrane fusion: Site-specific mutagenesis of the hearngglutinin of influenza virus. J . Cell Biol. 102, I 1-23. Goldstein, J. L., Brown, M. S., Anderson, R. G., Russell, D. W., and Schneider, W. J. (1985). Receptor-mediated endocytosis: Concepts emerging from the LDL receptor system. Annu. Rev. Cell B i d . 1, 1-39. Gollins. S. W., and Porterfield, J . S. (1984). Flavivirus infection enhancement in macrophages: Radioactive and biological studies on the effect of antibody on viral fate. J. Gen. Virol. 65, 1261-1272. Gollins, S. W., and Porterfield. J . S. (1985). Flavivirus infection enhancement in macrophages: An electron microscope study of viral cellular entry. J. Cen. Virol. 66, 19691982. Gollins, S. W., and Porterfield, J. S. (1986). pH-dependent fusion between the flavivirus West Nile and liposomal model membranes. J. Gen. Virol. 67, 157-166. Gonzalez-Scarano, F. (1985). La Crosse virus G 1 glycoprotein undergoes a conformational change at the pH of fusion. Virology 140, 209-216. Gonzalez-Scarano, F., Pobjecky, N., and Nathanson, N. (1984). La Crosse bunyavirus can mediate pH-dependent fusion from without. Virology 132, 222-225. Gruner, S. M., Cullis. P. R.,Hope, M. J., and Tilcock, C. P. S. (1985). Lipid polymorphism: The molecular basis of nonbilayer phases. Annu. Rev. Biophys. Chem. 14, 211-238. Haywood, A. M., and Boyer, B. P. (1984). Effect of lipid composition upon fusion of liposomes with Sendai virus membranes. Biochemistry 23, 4161-4166. Helenius, A., Marsh, M., and White, J . (1980). The entry of viruses into animal cells. Trends Biochem. Sci. 5 , 104-106. Helenius, A., Kartenbeck, J., Simons, K., and Fries, E. (1980). On the entry of Semliki Forest virus into BHK-21 cells. J . Cell B i d . 84, 404-420. Hoekstra, D., de Boer, T., Klappe, K., and Wilschut, J. (1984). Fluorescence method for measuring the kinetics of fusion between biological membranes. Biochemistry 23,56755681. Hoekstra, D., Klappe, K., de Boer, T., and Wilschut, J. (1985). Characterization of the fusogenic properties of Sendai virus: Kinetics of fusion with erythrocyte membranes. Biochemistry 24, 4739-4745. Homma, M., and Ohuchi, M. (1973). Trypsin action on the growth of Sendai virus in tissue culture cells. J . Virol. 12, 1457-1465. Homma, M., Shimizu, K.,Shimizu, Y. K., and Ishida, N. (1976). On the study of Sendai virus hemolysis I. Complete Sendai virus lacking in hemolytic activity. Virology 71,4147.

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

291

Hosaka. Y . , and ShiniiLu, Y. K . (1972). Artificial assembly of envelope particles of HVJ t Sendai virus) 1. Assambly of hemolytic and fusion factor\ from envelopes solubilized by Nonidet P40. L'irolog? 49, 627-639. Howe, C.. and Morgan, C. (1969). Interaction hetween Sendai virw and human erythrocytes. J . C'irol. 3, 70-8 I . Hsu. M.-C.. and Choppin. P.W. (1984). Analysis of Sendai virus mRNA with cDNA clones o f viral genes and sequence of biologiciilly imporlanl regions of the fusion protein. Proc,. N u / / .Ai,ail. Sc.i. U . S . A . 81, 7732-7736. Hsu. M.-C.. Scheid, A , , and Choppin. P. W . (19x3). Fusion of Sendai virus with lipowmes: Dependence on the viral fusion protein tF) and the lipid composition of liposomes. Virology 126, 361-369. Huang. R. T . C.. Rott. R.. and Klenk, H.-D. ( I Y X I ) . lnfleunza viruses cause hemoly\is and fusion of cells. Virology 110, 243-247. Imoto, T., Johnson. La. N., North, A. C. T.. Phillips. 0.C., and Rupley, J. A. (1972). Vertebrate lysozymes. IN "The Enzymes" (1'. D. Boyer. ed.). Vol. 7. 3rd Ed.. pp. 665868. Academic Press. New York. Junankar, P., and Cherry, R. J . (1986).Temperature and pH dependence of the haemolytic activity of influenza virus and of the rotational mobility of the spike glycoprotein. Biochim. Biuphys. Acru 854, 198-106. Kawasaki. K . , Sato. S. B., and Ohnishi. S . (19x3). Membrane fusion activity of reconbtiluted vesicles of influenza virus hemagglutinin glycoprotein. Bioc%irn. Biophy.r. Acirr 733, '86-290. Kawasaki. K.. Murata. M., Ohnishi. S.. Ikeuchi. Y.. &L Kanaseki. T. (1987). Membrane fusion of influenza virus and liposomes studied by quick-frozen replica techniques. . 12, 717. (Abstr.) Cdl S r r c ~ f Frmcf. Kielian. M. C.. Keranen, S., KBiriiiinen. L.. and Helenius. A. (1984). Membrane fusion mutants of Semliki Forest virus. J . Cell B i d . 98, 139-145. Kim. J . . and Okada. Y . (1981). Morphological chmges in Ehrlich ascites tumor cells during the cell fusion reaction with HVJ (Sendai virus) 11. Cluster formation of inIramembrane particles in the early stage of cell fusion. Exp. C d l Hrs. 132, 125-136. Kim. J.. Hama. K.. Miyake, Y . , and Okada. Y . (1979). Transformation of intramembrane particles of HVJ (Sendai virus) envelopes from an invisible to visible form on aging of virions. V i m h g y 95, 523-535. Kitamae, F.. Sugawara. K.. Ohwada, K.. and Homma, M . (19x2). Proteolytic activation of hemolysis and fusion by influenza C virus. Arch. Virol. 73, 357-361. Klenk, H.-D., Rott, R . . Orlich. M.. and Blodorn, J. (1975).Activation ofinfluenza A viruses by trypsin treatment. Viro/og.v 68, 426-439. Knutton, S. (1979). Studies of membrane fusion. V. Fusion oferythrocytes with non-haemolytic Sendai virus. J . C d l S c i . 36, 85-96, Knutton. S.. and Bachi. T. (19x0). The role of cell swelling and haemolysis in Sendai virusinduced cell fusion and in the diffusion of incorporated viral antigens. J . Cell Sci. 42, 153- 167. Kohama. T.. Shimizu, K . , and Ishida, N . (1978).Carbohydrate composition of the envelope glycoproteins of Sendai virus. Virology 90, 226-234. Krystal, M., Elliot. R.. Benz. E., Young, J., and Palese. P. (1982). Evolution of influenza A and I3 viruses: Conservation of structural features in the hemagglutinin gene>. Prioc.. N t d . A c d . Sci. U.S.A. 79, 4800-4808. Kundrot, C. E.. Spangler. E. A.. Kendall. D. A . , MacDonald, R. C . . and MacDonald, R. I. ( 1983). Sendai virus mediated-lysis of liposomes requires cholesterol. Proc. Nu//. Actid. Sc?. U . S . A . 80, 1608-1612. Kuroda, K . , Kawasaki, K., and Ohnishi. S. (19XS). Kinetic analysis of fusion of hemaggluti-

292

SHUN-ICHI OHNlSHl

nating virus of Japan with erythrocyte membrane using spin-labeled phosphatidylcholine. Biochernisrry 24, 4624-4629. Lazarowitz, S., and Choppin, P. W. (1975). Enhancement of the infectivity of influenza A and B viruses by proteolytic cleavage of the hemagglutinin polypeptide. Virolog? 68, 440-454.

Lee, P. M.. Cherry, R. J., and Blchi. T. (1983). Correlation of rotational mobility and flexibility of Sendai virus spike glycoproteins with fusion activity. Virology 128, 65-76. Lenard, J., and Miller, D. K. (1981). pH-dependent hemolysis by influenza. Semliki Forest virus, and Sendai virus. Virology 110, 479-482. Lenz. J., Crowther, R.,Straceki, A., and Haseltine, W. (1982). Nucleotide sequence of the AKV enu gene. J . Virol. 42, 519-529. Madshus, 1. H., Olsnes, S., and Sandvig, K. (1984). Mechanism of entry into the cytosol of poliovirus Type 1: Requirement of low pH. J . Cell Bid. 98, 1194-1200. Maeda, T.. and Ohnishi, S. (1980). Activation of influenza virus by acidic media causes hemolysis and fusion of erythrocytes. FEES Lett. 122, 283-287. Maeda, T.. Asano, A.. Ohki, K., Okada, Y . , and Ohnishi, S. (1975). A spin-label study on fusion of red blood cells induced by hemagglutinating virus of Japan. Biochemistry 14, 3736-3741.

Maeda, Y., Kim. J., Koseki, I., Mekada, E., Shiokawa, Y.. and Okada, Y. ( 1977). Modification of cell membrane with viral envelopes during fusion of cells with HVJ (Sendai virus). Ill. Effects of mono- and disaccharides on cell fusion and membrane movement of fused cells. Exp. Cell Res. 108, 95-106. Maeda, T.. Kawasaki, K., and Ohnishi. S. (1981). Interaction of influenza virus hemagglutinin with target membrane lipids is a key step in virus-induced hemolysis and fusion at pH 5.2. Proc. Natl. Acud. Sci. U.S.A. 78, 4133-4137. Marsh, M. (1984). The entry of enveloped viruses into cells by endocytosis. Biochem. J. 218, 1-10.

Marsh, M., Bolzau, E., White, J., and Helenius, A. (1983a). Interactions of Semliki Forest virus spike glycoprotein rosettes and vesicles with cultured cells. J. Cell B i d . 96, 45.5461.

Marsh, M.. Bolzau, E., and Helenius, A. (1983b). Penetration of Semliki Forest virus from acidic prelysosomal vacuoles. Cell 32, 93 1-940. Matlin, K. S . , Reggio, H., Helenius, A., and Simons, K. (1981). Infectious entry pathway of influenza virus in a canine kidney cell line. J. Cell &a/. 91, 601-613. Matlin, K. S . . Reggio, H.. Helenius, A.. and Simons, K. (1982). Pathway of vesicular stomatitis virus entry leading to infection. J . M o l . &I/. 156, 609-63 I . Metsikko, K., van Meer. G., and Simons, K. (1986). Reconstitution of the fusogenic activity of vesicular stomatitis virus. EMBO J. 5, 3429-3435. Mifune. K., Ohuchi, M.. and Mannen, K.(1982). Hemolysis and cell fusion by rhabdovirus. FEBS L e t t . 137, 293-297. Murata, M., Nagayama, K., & Ohnishi. S. (1987a). Membrane fusion activity of succinylated melittin is triggered by protonation of its carboxyl groups. Biochemistry 26, 40564062.

Murata, M., Sugahara, Y.,Takahashi, S., & Ohnishi, S . (1987b). pH-Dependent membrane fusion activity of a synthetic twenty amino acid peptide with the same sequence as that of the hydrophobic segment in influenza virus hemagglutinin. J . Biochem. 102, 957962.

Nakada, S., Creager, R. S., Krystal, M., Aaronson, R. P.. and Palese, P. (1984). Influenza C virus hemagglutinin: Comparison with influenza A and B virus hemagglutinins. J . Virol. 50, 118-124. Nir. S., Stegmann, T.. and Wilschut, J. (1986a). Fusion of influenza virus with cardiolipin

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

293

liposomes at low pH: Mass action analysi\ of kinetics and extent. Biochcmistr? 25, 257266.

Nir. S.. Klappe. K..and Hoekstra. [I. (1986b). Kinetics and extent offusion between Sendai virus and erythrocyte ghosts: Application of ti mas\ action kinetic model. Bioclwrnisrry 25, ? I 5 5 - 2 l 6 l . Ohki. S. (IY87). Surface tension. hydration energy and membrane fusion. In “Molecular Mechanisms of Membrane Fusion” (S. Ohki. D. Doyle, T. Flanagan. S. W. Hui. and E. Mayhaw. eds.), in press. Plenum. New York. Ohkuma. S.. and Poole. B. (1978). Fluorescence probe measurement of the intl-alysosomal pH in living cells and the perturbaof pH by various agents. Proc. Null. Acrrd. Sci. U . S . A . 75, 3317-3331. Ohnishi. S . (l985a). Membrane fwion: Mechanism and role in cellular response to enveloped virus. In “Biomolecules: Electronic Aspects” tC. Nagata. M. Hatano, J. Ttinaka. and H. Suruki. eds.). pp. 227-252. Japan Scientific Societies Press. Tokyo; Elsevier. Amsterdam. Ohnishi, S . ( I985b). Interaction of hemagglutinin with target membranes. V a c c i n e 3, 104. Ohnishi. S.. and Yoshimura. A. (1984). Infectious cell entry mechanism of enveloped v i ruses. Virus (Jpn.) 34, 11-24. Ohuchi. M., Ohuchi. R.. and Mifune, K . (1982). Demonstration of hemolytic and fusion activities of influenza C virus. J . Virid. 42, 1076-1079. Okada, Y. ( 1958). The fusion of Ehrlich’s ascites tumor cells caused by HVJ virus in u i ~ r o . ~ i / i J ~. 1, , ~103-110. Okada, Y . . Koseki, I . . Kim. J., Maeda, Y., Hashimoto. T.. Matsui. Y . , and Kanno. Y. (1975). Modification of cell membranes with viral envelopes during fusion of cells with HVJ (Sendai virus) I . Interaction between cell membranes and virus in the early stage. k i p . C P / /R C S .93, 368-378. Ozawa, M., and Asano. A. ( 198 I).The preparation of cell fusion-inducing proteoliposomes from purified glycoproteins of HVJ (Sendai virus) and chemically defined lipids. J . B i d . C’licrn. 256, 5954-5956. Parsegian. A. ( 1969). Energy of an ion cros\ing B low dielectric membrane: Solutions to four relevant electrostatic problems. Narrrre ~Lonrlr~n) 221, 844-846. Pahternak. C. A. ( 1984). How viruses damage cells: Alterations in plasma membrane function. J . Biosci. 6, 569-583, Porter. A. G . . Barber. C.. Carey, N. H . . Hallewell. K. A., Threlfall. G . . and Emtage, J. S. ( 1979). Complete nucleotide sequence of an influenza virus haemagglutinin gene from cloned DNA. Ncrlure (Lonclon)282, 471 -477. Kedmond. S. M. S., and Dickson, C. (1983). Sequence and expression of the mouse mammary tumor v i r w enu gene. E M B O J . 2, 125-131. Kedmond, S., Peters, G., and Dickson, C. (1984). Mouse mammary tumor virus can mediate cell fusion at reduced pH. Virology 133, 393-402. Rice. C. M.. and Strauss. J. H. (1981). Nucleotide sequence of the 26s mRNA of Sindbis virus and deduced sequence of the encoded virus structural proteins. Proc. Nirtl. Acod. 3’c.i. U.S.A. 78, 2062-2066. Richardson, C. D.,and Choppin, P. W.(1983). Oligopeptides that specifically inhibit membrane fusion by paramyxoviruses: Studies on the site of action. Virology 131, 518-532. Richardson. C. D., Scheid. A,, and Choppin. P. W. (1980). Specific inhibition of paramyxovirus and myxovirus replication by oligopeptides with amino acid sequences similar to those at the N-termini of the F, or H A 2viral polypeptides. Virology 105, 205-222. Kose. J . K., and Gallione. C. J. (1981). Nucleotide sequences of the mRNA’s encoding the vesicular stomatitis virus G and M proteins determined from cDNA clones containing the complete coding regions. J . Virol. 39, 519-529.

294

SHUN-ICHI OHNlSHl

Rott. R., Orlich, M.. Klenk, H.-D., Wang, M. L., Skehel, J. J., and Wiley. D. (1984). Studies on the adaptation of influenza viruses to MDCK cells. EMEO J. 3, 3329-3332. Ruigrok, R. W. H.,Wrigley, N. G . , Calder, L. J . , Cusack. S . , Wharton, S . A., Brown, E. B.. and Skehel, J. J. (1986). Electron microscopy of low pH structure of influenza virus huemagglutinin. EMBO J . 5, 41-49. Sandvig, K.. and Olsnes, S . (1980). Diphtheria toxin entry into cells is facilitated by low pH. J . Cell Bird. 87, 828-832. Sato, S. B.. Kawasaki, K.. and Ohnishi, S . (1983). Hemolytic activity of influenza virus hemagglutinin glycoproteins activated in mildly acidic environments. Proc. Null. Arud. Sci. U . S . A . 80, 31534157. Scheid. A., and Choppin, P. W. (1974). Identification of biological activities of paramyxovirus glycoproteins. Activation of cell Fusion, hemolysis and infectivity by proteolytic cleavage of an inactive precursor protein of Sendai virus. Virufogy57,475-490. Schlegel, R., and Wade, M. (1984). A synthetic peptide corresponding to the NH2 terminus of vesicular stomatitis virus glycoprotein is a pH-dependent hemolysin. J. B i d . Chem. 2S9,469 1-4694.

Schlegel, R..and Wade. M. (1985). Biologically active peptides of the vesicular stomatitis virus glycoprotein. J . Virol. 53, 319-323. Schlegel. R.. Tralka, T. S., Willingham, M . C., and Pastan, I. (1983). Inhibition of VSV binding and infectivity by phosphatidylserine: Is phosphatidylserine a VSV binding site? Cell 32, 639-647. Schloemer, R. H., and Wagner, R. R. (1975). Cellular adsorption function of the SkdlOglyCOprotein of vesicular stomatitis virus and its neuraminic acid. J . Virol. 15, 882-893. Schmidt, M. F. G. (1983). Fatty acid binding: A new kind of posttranslational modification of membrane protein. C w r . Top. Microhio/. fmmunol. 102, 101-129. Schmidt, M. F. C . , and Lambrecht, B. (1985). On the structure of the acyl linkage and the function of fatty acyl chains in the influenza virus haemagglutinin and the glycoproteins of Semliki Forest virus. J . Cen. Virrd. 66, 2635-2647. Seki, M., Hattori. S . , Hirayama, Y.,and Yoshida, M. (1983). Human adult T-cell leukemia virus: Complete mucleotide sequence of the provirus genome integrated in leukemia cell DNA. Proc. Natl. Acud. Sci. U.S.A. 80, 3618-3622. Sekiguchi, K.,and Asano, A. (1978). Participation of spectrin in Sendai virus-induced fusion of human erythrocyte ghosts. Pror. Null. Aciid. Sri. U.S.A. 75, 1740-1744. Shibata, M., Maeno, K., Tsurumi, T.. Aoki, H.,Nishiyama, Y., Ito, Y.,Isomura. S.. and Suzuki, S. (1982). Role of viral glycoproteins in hemolysis by influenza B viruses. J. Gen. Virol. 59, 183-186. Shimizu, Y. K., Shimizu, K., lshida, N., and Homma. M. (1976). On the study of Sendai virus hemolysis 11. Morphological study of envelope fusion and hemolysis. Virology 71, 48-60.

Shinnick, T. M., Lerner, R. A., and Sutcliffe, J. G. (1981). Nucleotide sequence of Moloney murine leukemia virus. Nufure ( h n d O n ) 293, 543-548. Skehel, J. J.. Bayley, P. M.,Brown, E. B.,Martin. S. R., Waterfield, M. D., White, J. M.. Wilson, 1. A,. and Wiley, D. C. (1982). Changes in the conformation of influenza virus haernagglutinin at the pH optimum of virus-mediated membrane fusion. Proc. Nod. A c u ~ Sci. . U.S.A. 79, 968-972. Stegmann, T.,Hoekstra, D., Scherphof. G., and Wilschut. J. (1985). Kinetics of pH-depen, dent fusion between influenza virus and liposomes. Biochemistry 79,968-972. Stegmann, T.. Hoekstra. D.,Scherphof, G.,and Wilschut, J. (1986). Fusion activity of influenza virus. A comparison between biological and artificial target membrane vesicles. J. B i d . Chem. 261, 10966-10969.

9. FUSION OF VIRAL ENVELOPES WITH CELLULAR MEMBRANES

295

Struck, I). K., Hoekstra. 0.. and Papano. P. E. (1981). Use of resonance energy tr;insfer t o monitor membrane fusion. Bioc./ierni.trr.y 20, 4093-4099. Sturman, L. S.. Kicard, C . S..and Holmeh. K. V . (1985). Proteolytic cleavage of the E? glycoprotein of murine Coronavirus: Activation of cell-fusing activity of virions by trypsin and separation o f two different YOK cleavage fragments. J . V i r ( ~ /56, . 904911. 'l'anaka. K.. and Ohnishi, S. ( 1974). Heterogeneity in the tiuidity of intact erythrocyte membrane and i t s homogenization upon hemolysih. Bioclrirt?. B i c ~ p hA~ i.m 426, 218-

231. Tsci.ji. A , . and Ohnishi. S . (1986). ReWiction o f t h e liitcriil motion o f hand 3 in the erythrocyte membrane by the cytoskcletiil network: Dependence on speclrin association state. BicJf'/iCr!li.Y/r\ 25, 6 133-6 139, Tycko, B., and Maxfield. F. R. ( 19x2). Rapid iiciditication of endocytic vesicles containing a~-inacroglobulin.Ci4/ 28, 6 4 - 6 5 I . Viiinanen, P.. and Kiiirilinen. L. ( 1979). Haemolysis hy two iilphaviruses: Semliki Forest and Sindhis virus. J . Giw. Viral. 43, 593-601. Viiiiniinrn. P.. and Kiiiiriiiinen. L . i 19x0). Fuhion hind haeniolysih of erythrocytes caused by three togaviruses: Semliki Forest Sindbih, ;ind rubella. J . Gou. V i r o / . 46, 467-475. van Kenswoude. J . . Bridges. K . R..Harford. J . B . . and Klausner. K. I). (19x2). Receptoi-mediated endocytosis o f transferrin and the uptake of Fe in KS62 cell\: Identification of a nonlysosomal acidic compartment. Pro(.. Nor/. Actrd. S c i . U.S.A. 79. 6186-6190. Verhoeyen, M.,Fang. K..Min Jou. W., Devih, K., Huylehroeck. D.. Saman. E.. and Fiers, W. ( 1980). Antigenic drift between the heemagglutinin of the Hong Kong influenza striiins AiAichii2i68 and AiVictoriai3i75, Nirrrrrc. ( L o t i d o n ) 286, 771-776. Vtrlsky. D. J.. and Loyler, A . i 197Ra). An cflicient method for r e a w m h l y o f fusogenic Sendni virus envelopes after solubili/ation of intact virions with Triton X- 100. FERS LtJII.92, 190-194. Volsky. I).J . . and Loyter, A. (1978h). Role ofC;i'* in virus-induced membrane fu\ion. Ca" accumulation and ultrastructural changes induced by Sendai viru\ in chicken erythrocytes. J . c 1 4 / B i o / . 78, 465-47'). von Heijne. G . (1981). On the hydrophobic nature of signal sequence. Ero.. 1.Bioc./it,/n. 116, 419-422. Wharton, S. A., Skehel. J. J . . and Wiley. 1). C. ( IYX6). Studies ofinlluenza haemagglutininmediated membrane fusion. V ~ ~ ( J / O149, R X 27-35. White. J . . and Helenius. A. (1980). pH-dependent fusion between the Semliki Forest virus membrane and liposomes. /'roc. N d . A m / . .Y(.i. U.S.A. 77, 3273-3277. White, J . . Kartenbeck. J . . and Heleniw. A. (19x0). Fusion olSemliki Forest virus with the plasmit membrane can be induced by low pH. 1.Coil E i ~ l 87, . 264-272. White. J., Matlin, K.. and Heleniu5. A . (1981). Cell fuhion by Semliki Forest. influenzit kind vesicular stomatitis viruh. J . Cr//Biol. 89, 674-679. White, J.. Kartenbeck. J . , and Heleniub, A. (1982). Membrane fusion activity of influenza virus. EMBO J . I , 217-222. White, J . , Kielian. M.,and Helenius, A. (1983). Membrane fusion protein5 of enveloped animal viruses. 0. Rev. Bioplrv.\. 16, 151-195. Wilson, I . A , . Skehel. J . J . , and Wiley. I). C. (19x1). Structure o f the haemagglutinin memhrane glycoprotein of influenza virus at 3 A resolution. Nurrtrr~( f m i c l m i )2H9, 366373. Winter. G . , Fields, S.. and Brownlee, G . G . (1981). Nucleotide sequence o f the haemagglutinin gene of a human influenza virus H I subtype. Ntrrrrrr (fAJlidOt/)292, 72-75. Wyke, A . M.. Impraim. C. C.. Knutton. S..iind Pasternak, C. A. (1980). Components

296

SHUN-ICHI OHNlSHl

involved in virally mediated membrane fusion and permeability changes. B i o c h m . J. 190, 625-638. Yamada. S.. and Ohnishi. S . (1986). Vesicular stomatitis virus binds and fuses with phospholipid domain in target cell membranes. BiocRemistrV 25, 3703-3708. Yamamoto, K., Suzuki, K., and Shimizu, B. (1981). Hemolytic activity of the envelope glycoproteins of Western equine encephalitis virus in reconstitution experiments. Virology 109,452-454. Yasuda, Y., Hosaka, Y., Fukami, Y., and Fukai, K. (1981). Immunoelectron microscopic study on interactions of noninfectious Sendai virus and murine cells. J . Virol. 39, 273281. Yoshimurd, A . ( 1985). Adenovirus-induced leakage of co-endocytosed macromolecules into the cytosol. Cell Strcrc/. Fun<,[.10, 391-404. Yoshimura. A , . and Ohnishi, S. (1984). Uncoating of influenza virus in endosomes. J . V i r d . 51,489-496. Yoshimura, A.. Kuroda. K.. Kawasaki. K.. Yarnashina, S.. Maeda, T.. and Ohnishi. S. (1982). Infectious cell entry mechanism of influenza virus. J. Viral. 43, 284-293. Yoshimura. A., Yamashina, S., and Ohnishi, S. (1985). Mobilization and aggregation of integral membrane proteins in erythrocytes induced by interaction with influenza virus at acidic pH. Exp. Cell Res. 160, 126-137.

CURRENT TOPICS IN MEMBRANES AND TRANSPORT, VOLUME 32

Chapter I0

Sendai Virus-Mediated Cell Fusion YOSHIO OKADA lnstitutc~jbr Molecrrlnr und Crilirlur Biology Osuku University Siritrr, Osukri 565, Juputr

I. Introduction 11. 111.

IV.

Critical Problems in Cell-to-Cell Fusion Structure and Biological Activities of HVJ A. Structure of HVJ B. Cell Fusion Activity, Hemolytic Activity, and Infectivity of HVJ C. Envelope Structure of HVJ D. Viral Factors Required for Fusion of the Envelope with the Target Cell Membrane Cell-to-Cell Fusion by HVJ A. HVJ Concentration and pH Range Required for Cell-to-Cell Fusion B. Cell Lysis and Cell Fusion C. Virus-Cell Interaction at Low Temperatures D. Early Events Observed on Incubation at 37°C E. Alteration of Cytoplasmic Organelles and Their Reversion F. Separation of Viral Envelope Fusion and Cell Fusion G. Perturbation of Cell Membrdne Structure by HVJ H. Evidence for Close Attachment of Cell Membranes Induced by HVJ I . Discussion and Summary of Ehrlich Ascites Tumor Cell Fusion by HVJ References

I. INTRODUCTION

Developments in biotechnology since the early 1970s have been based on progress in both gene engineering and the manipulation of biomembranes, and especially of cell membranes. HVJ (Sendai virus') is the first I The virus was isolated from laboratory mice over the period 1953-1955 in several different laboratories in Japan during attempts 10 isolate the causative agents of some infectious diseases of man. All the isokdtes had the same antigenicity, but the virus was given a (footnote continued on page 298)

297 Copyright r r 1988 by Academic Pres. Inc All rights of reproduction in any form reserved

298

YOSHIO OKADA

agent found to be effective for manipulation of mammalian cell membranes in the living state. The demonstration of its cell fusion activity (Okada et al., 1957; Okada, 1958, 1962a,b; Okada and Tadokoro, 1962) was an important step in the new field of “somatic cell genetics,” established in the late 1960s, and resulted in development of not only cell hybridiziation procedures but also several refined cell engineering techniques, such as reconstitution of cells and introduction of macromolecules into cells (Furusawa et af., 1974; Veomett et al., 1974; Uchida et al., 1979; Ueda et al., 1981; Loyter and Volsky, 1982; Sugawa et a / . , 1985). HVJ cannot, however, be used to fuse cells that do not have HVJ receptors, such as cells of invertebrates and plants. Consequently, it has been used less since the discovery of a new fusogen, polyethylene glycol (Kao and Michayluk, 1974; Pontecorvo, 1975; Davidson and Gerald, 1977), which can be used for fusion of cells from all kinds of organisms. Subsequently, several other techniques for fusion have been reported and used for various cell membranes. But HVJ was the first fusogen discovered, and its discovery opened up many experimental fields. Moreover, unlike chemical fusogens or mechanical fusion techniques, it is unique in that it causes dissociation of fusogenic compartments in such a way that they readily reassemble. Thus, it is useful not only for analysis of the cell fusion mechanism but also for developing new technologies in cell biology and medicine. HVJ may be the most moderate fusogen known, and under suitable conditions it is not toxic. For instance, McGrath and Solter (1983) reported a high frequency of development of reconstituted eggs formed by the introduction of nuclei into enucleated eggs using HVJ. This low toxicity may be because the fusion reaction involves a part of the natural way in which HVJ infects host cells: low damage of the host cell is essential for growth of the virus. The fusion mechanism may be a model of fusion of organelles in cells. II. CRITICAL PROBLEMS IN CELL-TO-CELL FUSION

Some parts of the cell surface are hydrophilic domains formed of glycoproteins and glycolipids, and these interfere with cell-cell fusion, for different name in each laboratory. The Society of Japanese Virologists proposed the name “Hemagglutinating Virus of Japan (HVJ)” for these isolates in 1955. Independently, a strain isolated at Tohoku University was sent to the United States and provisionally named Sendai-type pneumonitis virus and then “Sendai virus.” The latter name has become the common name for this virus internationally.

10. SENDAI VIRUS-MEDIATED CELL FUSION

299

which close contact of lipid bilayers on the membranes of the fusing cells is essential. Another problem in cell fusion is the presence of hydrophobic domains of membrane proteins integrated into lipid bilayers, since these proteins may be associated with cytoskeletal systems inside the bilayer and so interfere with fusion. These characteristics of the cell membrane are in fact probably useful in living organisms for maintaining the integrity of the cells and preventing spontaneous cell fusion. In contrast, naked lipid vesicles produced by sonication or obtained by treatment with detergent fuse spontaneously in water. This kind of spontaneous fusion is also observed in protoplasts of pollen. Ito (1973) reported that protoplasts formed from pollen cells of liliaceous plants by treatment with cellulase fuse when they collide; forming giant protoplasts containing multiple nuclei; when those protoplasts are cultured, their cell fusion capacity decreases with time. The perturbation of two lipid bilayers in contact is essential for fusion. Fusion of vesicles depends on the properties of the lipids constituting these vesicles, the ionic environment, and osmotic mechanisms. This step may also be considered as a physical one that reduces the potential energy of the two biological lipid vesicles in water. Thus, the important function of a fusogen for fusion of living cells is to establish conditions permitting close contact of two lipid bilayers. This function of HVJ in inducing cell fusion is considered in this chapter.

111.

STRUCTURE AND BIOLOGICAL ACTIVITIES OF HVJ

A. Structure of H V J HVJ, a member of the Paramyxouirrrs genus of paramyxoviruses, is an enveloped virus of about 200 nm diameter containing a negative strand of RNA of about l S kb. The virus is constituted from six of the proteins shown in Table I. On the envelope, F and HANA (HN) proteins are exposed as spikes, and in the inside, there is a left-handed helical nucleocapsid of about 1 p n length and 17-18 nm width, which consists of single-stranded RNA and nucleocapsid proteins (NP). L and P proteins, also present in the nucleocapsid structure, may be involved in RNAdependent RNA polymerase activity. Guanylate cyclase activity, derived from host cells, is detectable inside of virions, and it coprecipitates with nucleocapsids on sucrose density gradient centrifugation (Kirnura rl al., 1981). The nucleocapsid and the inner suiface of the envelope are connected by M protein, which may play a role in the maturation of the virus, including budding of newly developed virus from the surface of the host

300

YOSHIO OKADA

TABLE I PROTEINS OF HVJ Designation L P HANA (HN) NP

FO FI

F2

M C

Molecular weight

Presence of carbohydrate

224,000 79.000 72,000 60,000 64,700 5 1,500 1 1,300 34,000 22,000

-

-

+ + + + -

Reference Shioda ef al. (1986) Kingsbury (1974) Kingsbury (1974) Kingsbury ( 1974) Gething ef ( I / . (1978) Gething ef a / . (1978) Gething rf a/. (1978) Kingsbury (1974) Lamb et a/. (1976)

cell, and in maintaining the rod-shaped morphology of HVJ particles (Kim and Okada, 1979). C protein is not an integral part of the virus (Lamb et ul., 1976). B. Cell Fusion Activity, Hemolytic Activity, and Infectivity of HVJ

HVJ has both hemolytic (Fukai and Suzuki, 1955)and cell fusion activities (Okada et ul., 1957), which appear to be closely related (Okada and Tadokoro, 1962). Morgan and Howe (1968) reported the important finding that the viral envelopes fuse with host cell membranes in the infection step. The next important finding was that when the envelope structure is solubilized with detergent, the hemolytic and cell fusion activities both disappear and reappear when the structure is reassembled on removal of the detergent (Hosaka and Shimizu, 1972a). Production of noninfectious and nonhemolytic virus progeny has been observed in many kinds of cell strains in culture other than in embryonated eggs (Ishida and Homma, 1961; Matsumoto and Maeno, 1962). The virus is not infectious to cultured cells in uitro, but is infectious when injected into the chorioallantoic cavity of embryonated eggs. This phenomenon was originally called a host-controlled modification2 and was examined by Homma (1971, 1972). He found that treatment with a low This kind of modification is not seen in other paramyxoviruses such as NDV and SV5: their progeny has active F. Nagai ef a / . (1976) demonstrated that an avirulent NDV strain propagated in cultured cells had inactive Fo like HVJ. In 1984, Paterson ef a/. explained these findings by comparing the nucleotide sequences of the HVJ and SV5 genomes. At the cleavage site of Fo,SV5 has a sequence containing five arginine residues whereas in HVJ the sequence has only a single arginine. Thus, SV5 is more sensitive than HVJ to proteolytic clearage.

10. SENDAI VIRUS-MEDIATED CELL FUSION

301

concentration of trypsin activates the virus which then becomes infectious to cultured cells. He also demonstrated the same level of proteolytic activity in chorioallantoic fluid. This proteolytic activity causes cleavage of F glycoprotein (Homma and Ohuchi, 1973: Scheid and Choppin, 1974). F is synthesized as inactive F0 in host cells and is cleaved to the active forms F, and F2 by a proteolytic enzyme as a posttranslational modification. In 1978, Gething c’t a / . sequenced the IS amino acid residues of the amino-terminal region of F, and showed that this region is composed of hydrophobic or intermediate residues, with no ionic residues. The aminoterminal sequence of 20 amino acids is well conserved in FI of paramyxoviruses such as SV5 and NDV (Richardson c’t d..1980). Noninfectious virus progeny propagated in cultured cells have Fo and show no hemolytic or cell fusion activity. Thus, the envelope structure containing spikes of HANA and F,,?is important for infection. and infection is believed to result from the introduction of the nucleocapsid into the cytoplasm by fusion of the viral envelope with the cell membrane. The hemolytic and cell-to-cell fusion activities both seem to be closely related with the fusion reaction. C. Envelope Structure of HVJ

The envelope of HVJ is composed of a lipid bilayer, the lipid components of which may be derived from host plasma membranes (Klenk and Choppin, 1970). and many surface spikes of HANA and F proteins. HANA is a glycoprotein showing hemagglutinating and neuraminidase activities (Scheid et C J / . , 1972; Tozawa p t al., 1973). The viral receptor consists of sialoglycoproteins and sialolipids, and HANA contains both receptor binding and receptor destroying activities. The minimal structure recognized by HANA is the sequence NeuAca2. 3Gal, and NeuAca2,8NeuAca2,3Gal shows maximal binding capacity (Markwell et ul., 1981). HANA proteins are present as oligomers connected by interpeptide disulfide bonds, and their reductive cleavage is accompanied by loss of biological activity (Ozawa et ul., 1976). F is also a glycoprotein participating in the interaction between the viral envelope and the lipid bilayer of host cell membranes. Recently, the complete base sequence of HANA cDNA of the Harris strain of Sendai virus was reported by Blumberg et ul. (198Sa),and that of the Z strain of HVJ by Miura et CJI. (1985a) and Shioda et ul. (1986). The sequence and the hydrophobic profile of the protein of HVJ Z strain are shown in Figs. 1 and 2. As deduced from the base sequence, the peptide consists of 575 amino acids and has an M, of 63,433. It contains five potential acceptor sites for N-linked carbohydrates (Am-X-Thr/Ser).The

302

YOSHIO OKADA l-AGGGTGAAAGTGAGGTCGCGCGGTACTTTAGCTTTCACCTCAAACAAGCACAGATCATG GAT GGT GAT AGG GGC AAA CGT GAC TCG TAC TGG TC T met a s p gly asp a r g g l y l y s a r g a s p ser t y r trp ser -11

96-

ACT TCT CCT ACT GGT AGC ACT ACA AAA TTA GCA TCA GGT TGG GAG AGG TCA ACT AAA GTT GAC ACA TGG TTG C TG A T 1 CTC TCA t h r rer pro ser g l y s e r thr t h r l y s leu a l a rer g l y t r p g l u a r g ser i e r l y s val asp l e u l e u i l e l e u ser -41

180-

TTC ACC CAG TGG GCT TTG TCA A T 1 GCC ACA GTG ATC ATC TGT ATC L T A A T 1 TCT GCT AGA CAA GGG T A T AGT ATG AAA GAG TAC phe t h r gln t r p a l a l e u s i r i l e a l a thr val 11e i l e r y s i l e I l e i l e ser a l a a r g gln g l y t y r i e r met 1ys glu t y r -69

264-

TCA ATG ACT GTA GAG GCA TTG AAC ATG AGC AGC AGG GAG GTG AAA GAG TCA CTT ACC ACT CTA ATA AGG C A I GAG GTT ATA GCA ser met t h r val glu a l d l e u l a r n m e t ] s e r arg glu val l y s g l u ser l e u thr ser l e u i l e a r g gln g i u va1 i l e a11 -97

348-

AGG GCT GTC AAC A T 1 CAG AGC TCT GTG C A I ACC GGA ATC CC9 GTC TTG TTG AAC AAA AAC AGC AGG GAT GTC ATC CAG ATG A T 1 a r g a l a Val d i n I l e g l n ser s e r val g l n t h r g l y i l e Pro val l e u l e u a m l y r as" rer a r g asp val i l e gln met i l e -125

432-

GAT AAG TCG TGC AGC AGA CAA GAG CTC ACT LAG CAC TGT GAG ACT ACG ATC GCA GTC CAC C AT GCC GAG GGA A T 1 GCC CCA CTT dsp l y s ser cys s e r a r g g l n g l u leu t h r gln h i s c y i glu ser thr l l e 111 val h i s h i s a l a glu g l y i l e a l a p'o l e u -153

51 6-

GAG C C A C A T AGT TTC TGG AGA TCC C C T GTC GGA GAA CCG TAT C T T AGC T C A GAT c u GAL AT[ TCA TTG CTG C C T GGr CCG AGE g l u pro h i s ser phe t r p a r g cys pro Val g l y glu pro tyr l e u ser ser asp pro glu i l e ser l e u l e u pro ply pro ser -181

600-

T T G T T A TCT GGT TCT ACA ACG ATC TCT GGA

TGT GTT AGG CTC CCT TCA CTC TCA A T 1 GGC GAG GCA ATC TAT GCC TAT TCA TCA

leu l e u rer g l y rer thr thr i l e ser ply c y s vdl a r g l e u pro ser leu rer i l e g l y glu a l a i l e t y r d l d t y r rer ser -209 684-

ART CTC A T 1 ACA C I A GGT TGT GCT GAC ATA GGG AAA TCA T A T CAG GTC CTG CAG C TA GGG TAC ATA TCA C l C AAT TCA GAT ATG leu i l e t h r g l n g l y c y s d l a asp i l e g l y l y s ser t y r g l n val l e u gln l e u g l y t y r 1 1 e rer leu a m rer asp met -217

dsn

768-

ATC CCT GAT CTT AAC CCC GTA GTG TCC CAC ACT TAT GAC ATC AAC GAC AAT CGG AAA TCA TGC TCT GTG GTG GCA ACC GGG ACT i l e PIO asp leu asn pro val Val ser h i s t h r t y r d i p l l e asn a s p dsn d r g l y r ser c y r i e r val val a l a thr g l y t h r -265

852-

AGG GGT TAT CAG CT T TGC TCC ATG CCG ACT GTA GAC GAA AGA ACC GAC TAC TCT AGT GAT GGT ATC GAG GAT CTG GTC C T l GAT a r g g l y t y r g l n leu c y s ser met pra t h r ual asp g l u d i g t h r asp t y r ser ier a s p g l y l l e g l u asp l e u Val leu asp -293

916-

GTC CT G GAT CTC AAA GGG AGA ACT RAG TCT CAC CGG T A T CGC AAC AGC GAG GTA GAT CTT GAT CAC CCG TTC TC T GCA CTA TAC va1 leu asp l e u l y s g l y d i g t h r l y i ser h i s a r g t y r a r g d i n ser g l u ual asp leu a r p his pro phe ser a l a leu t y r -321

1020-

CCC ACT GTA GGC AAC GGC A T 1 GCA ACA GAA GGC TCA TTG ATA TTT CTT GGG TAT GGT GGA CTA ACC ACC CCT CTG CAG GGT GAT pro ser w.1 g l y asn g l y i l e a l a t h r g l u g l y ser l e u i l e phe leu g l y t y r g l y g l y l e u t h r thr pro leu g l n g l y dsp -149

11 04-

ACA AAA TGT AGG ACC CAR GGA TGC CAA CAG GTG TCG CAA GAC ACA TGC AAT GAG GCT CTG AAA A T 1 ACA TGG C TA GGA GGG AAA thr l y s cys a r g t h r g l n g l y c y s g l n g l n Val ser g l n a s p f h r c y r d i n g l u a l a l e u l y s Ile t h r t r p leu p l y g l y l y s -377

1 1 88- CAG GTG GTC AGC GTG ATC ATC CAG GTC AAT GAC TAT CTC TCA GAG AGG CCA AAG ATA AGA GTC ACA ACC A T 1 CCA ATC ACT CAA g l n r a l val ser vat i l e i l e g l n val a r n asp t h r l e u ser g l u a r g pro l y s i l e a r g Val t h r t h r nle pro i l e t h r g l n -405 1272-

AAC T A T CTC GGG GCG GAA GGT AGA TTA TTA AAA TTG GGT GAT CGG GTG TLC ATC T A T ACA AGA TCA TCA GGC TGG CAC TCT CAA t y r l e u g l y a l a g l u g l y d q leu l e u l y s l e u g l y a s p a r g V a l t y r i l e tyr t h r a r g ser ser g l y t r p h i s ser g l n -433

dsn

1156-

CTG CAG ATA GGA GTA C TT GAT GTC AGC CAC CCT TTG ACT A TL AAC TGG ACA CCT C AT GAA GCC TTG TCT AGA C C A GGA AAT AAA leu gln i l e g l y Val l e u a s p Val ser h i s pro leu t h r i l e l a s n t r p thrlpro h i s g l u 11I l e u ser a r g pro g l y a s n l y s -461

1440-

GAG TGC ART TGG TAC AAT AAG TGT CCG RAG GAA TGC A TA TCA GGC GTA TAC ACT GAT GCT TAT CCA TTG TCC CCT GAT GCA GCT g l u c y s asn t r p t y r asn l y r c y s pro l y r g l u c y s , l e ser g l y val t y r t h r asp d l a t y r pro l e u ser pro asp a l a a l a -489

1524-

AAC GTC GCT ACC GTC ACG CTA TA T GCC RAT ACA TCG CGT GTC ARC CCA ACA A l C ATG TAT TCT AAC ACT ACT AAC A T 1 ATA AAT a r n val a l a Lhr val t h r l e u t y r a l a [ f h r ] a r g v a l l a r n pro t h r l i l e met t y r rerlarn t h r t h r j a m i l e i l e a s n -517

1608-

A T G T T A AGG A T A AAG GAT GTT CAA TTA GAG GCT GCA TA T ACC ACG ACA TCG TGT ATC ACG CAT TTT GGT AAA GGC TAC TGC TTT met l e u arg i l e l y s asp val g l n l e u g l u a l a a l a t y r t h r t h r f h r ser c y s i l e t h r h i s phe p l y l y s g l y t y r c y s phe -545

1692-

CAC ATC ATC GAG ATC RA T CAG AAG AGC CTG A A T ACC T T A CAG CCG ATG CTC TTT AAG ACT AGC ATC CCT AAA TTA TGC AAG GCC h i s i l e i l e g l u i l e asn g l n l y r set' l e u asn t h r l e u g l n pro met l e u phe l y s t h r s e r i l e pro l y s l e u c y s l y s a l l - 5 7 1

1776-

GAG 1 C l TAAATTTAACTGACTAGCAGGCTTGTCGGCCTTGCTGACACTAGAGTCATCTCCGAACATCCACAATATCTCTCAGTCTCTTRCGTCTCTCACAGTATTAAG-1883 g l u ser ***

FIG. 1 . Nucloetide sequence of the cDNA for HANA mRNA and the predicted amino acid sequence. The amino-terminal hydrophobic domain is doubly underlined. The positions of potential asparagine-linked acceptor sites for carbohydrate are marked with boxes. (From Miura er a / . , 1985b).

amino terminus of the HANA protein (amino acids 35-60) is the only strongly hydrophobic region, and may be the signal sequence for both the transfer of the protein across the endoplasmic reticulum (ER) membrane and its anchorage in the lipid bilayer. The base sequence of Fo cDNA has also been reported (Blumberg et al., 1985b; Miura et al., 1985b; Shioda et al., 1986). The base sequence and the proposed amino acid sequence are presented in Fig. 3. Figure 4 shows the hydrophobicity profile of the peptide, which has 565 amino acids and

303

10. SENDAI VIRUS-MEDIATED CELL FUSION

>t

-

2

G -

I

I

1

a 0 u 0 n > I

-1

2

3 0

1 s

158

-m

258

rn

m

,

LBB

----A_--

1%

rn

%m

AMINO ACID SEQUENCE NUMBER Fic. 2. Hydrophobicity profile of the H A N A protein. Numbers under the horizontal axis are positions from the amino-terminal methionine. The proposed site of the transmembrane anchor is marked TM. (From Miura ct 01.. 198385b.3

an M , of 61,665. Three regions are very hydrophobic and could interact directly with membranes: these are the amino-terminal putative signal peptide (amino acids 1 1-23), the carboxy-terminal putative membrane anchorage domain (amino acids 500-5231. and the amino-terminal region of the F, polypeptide (residues 117-142). The carboxy-terminal region, which consists of 42 amino acids and is moderately hydrophilic, may remain in the cytoplasm or viroplasm.

D. Viral Factors Required for Fusion of the Envelope with the Target Cell Membrane A unique technique has been established for analysis of the factors required for fusion (Uchida r t a / ., 1979).This is the formation of artificial lipid vesicles with viral spikes on their surface and fragment A of diphtheria toxin molecules inside. Diphtheria toxin consists of fragment A (22 kDa) and fragment B (40 kDa). Fragment B is the domain that binds with its receptor on the surface of susceptible cells. Fragment A has the N A D elongation factor 2 (EF-2) ADP-ribosyltransferase activity and blocks protein synthesis in eukaryotic cells. It is not toxic to cells when added to

cis

YOSHIO OKADA

304

I - U r O A T A M G m C T T C T C A G T G ' 3 7 G A T T C C W C X T C ~ C A l G ACA GCA TAT ATC CAG AGA TCA CIC T7X ATC 'PCI ACA 1u Met T h r A l a Tyr I l e G I " A l q Ser Gln C I S I I L Set T h r Ser -14

0 96-

CTA CTG G T I CTT CTC ACC ACA TF2 GTC TCC TGT CAG ATT CCC AGO GAT AGG CTC TCT M C ATA GGC CTC ATA GTC GAT G M G% Leu Leu Val Val Leu Thhr T h r Leu Val-Sel C y s G i n I l e P r o Arq A s p Arq Leu Ser &En IIC G l y Val Ile Val Asp G I Y G l Y - 4 2

0

180-

AM TCA CTG M G ATA GCT GGA TCC CAC G M TCG AGG TAC ATA GTA CTG hGT CTL 0°F CCG GOD GTA GAC TIT GAC M T u% TGC Lya Str Leu Lys I l e Aka G l y Ser HIS GI" S t r I r q Tyr Ile Val Leu Ser Leu Val Pro G l y V a l Aap P h r GI" Asn G l y C B

264-

GGA ACA

318-

c1G ATA KT Gn' ACC M T GAT Ux: ACA C M M T UIC Om oc7 CCA CAG X G Iw TTC Tn: UIT OCT GlG A m ffiT ACT ATC CCA Leu I l e T h r V a l T h C L l s n P v T h l Gln A*" A I a G l Y A I a P r o G I n S P Z A r J z h e P h c C l y A I d V a l I l e G l y T h r l i e AIa -126

01-

CTT GGA G%

70

li-

G X UD GTT

ATC CAG T I C M G A K CTA CTC M C AGG CTG TTA hTC CCA T% AGG GAT GCC TTL GAT CTT CAD GAG GCT C I y Thr A 1 8 G I " Val I l t G l n Tyr LyS Jer Leu Leu A m A l g Leu Leu I l e Pro Leu Arg Asp A l a Leu Asp Leu G l n GIU AIa -98

GCG ACA X A GCA C M lTC Icc GCA GCG A T T GCA CTA GCC W GCG AGG GAG CCC AM AGA GAC ATA BX X I2 . ATC A I d ILYA l a G l u A I d Arq G I " A l a Lya Acg Asp I l r Ale Leu I l c -154

Leu G l y Val A11 T h r Ser A l a O l n I t 1 T h l A l a G l y I l e 516-

AM O M TCG A T 0 ACA MA ACA CAC MG TCT ATA G M CTC CTG C M M C GCT CTG GGG GAA C M ATT CTT GCT CTA &G ACA CTC Lys G l u Sac I4+t ThC Lye T h c His Ly9 Set Ile G l u Leu Leu G l n A m A l d Val G l y G l u C l n I l e Leu A l a Leu Lys T h r Leu - 1 8 2

600-

CAG GAT GTG M T GAT GAG A X MA CCC GCA ATA AGC G M T T A GU: TGT GAG ACT GCT GCC TTA AGA CTG GGT ATA MA TPG Gln hsp P h e Val As" Asp Glu Ile Lys P f o A I a Ile Sex G l u Leu G l y C s G I 0 T h r A l a A I a Leu A r q Leu G l y Ile Lys Leu -210

684-

ACA CAG CAT TAC TCC GAG CTG TTA ACT GCG TTC GCC XG MT m GGA ACC ATC GGI GAG AAG AGC CTC ACG CTG CAG U'G CTG T h r G l n H L S Tyyr SPr G I Y Leu Leu T h r & l a Phe G l y Scr A m Phc G l y Thr I I I G l y G l u Ly6 Scr Leu T h r Mu G l n A l a Leu -218

168-

n'T V2.L

852-

O M C f f i AW MA f f i A LCG GTG ATk GAT

K"

CR TAC n'T GCT MC Am ACT GAG A T T ATG MX ACA A X MG ACA GGG CAG TCT MC ATC TAT GAT G K A T T TAT ACA Ser Ser Leu Tyr Ser A l a l A ~ n Ile T h r l G l u l l e Uet T h r T h r IIC Lyb T h r G l y G l n Ser AS" I I e T y r A s p V a l I l e W l T h r -266 Glu Gln

I l c Lyr

GFO GAT C T A GAC AGA TAC A T 0 G K ACC c1G TCT GTG M G A X CCT A m CTT TCP G M G l y T h r Val Ile A s p V a l Asp Leu G l u Arq Tyr Met V a l T h r L e u Set Val Lys Ile P r o I l e Leu S e l C l u - 2 9 4

916-

G X CCA GGT G1C CTC ATA CAC M G GCA TCA TCT ATT TCT T I C M C ATA GAC CCG GAG O M TGG TAT GTG ACT GTC CCC AGC CAT Val Pro G l y YaI Leu Ile H l e Lys A1a Ser Scc Ile %I T y r Am I I a ABP G l y G l u G i u T r p Tyr Val TO1 Val P r o Ser H i s -122

1020-

ATA C K AGT CGT GCT TCT I T 2 TTA OOG GGT GCA CAC AT1 ACC GAT TCT GTT GAG TCC AGA YT.2 ACC TAT A T A TCC CCC AGO GAT I1c Leu SCC A l q A I a Ser P h e Leu C l y G l y A I I Asp I l e T h r Asp C 9 Val G I " S e l Acq Leu T h c Tyr l l e C L PIO hrg Asp -150

1104-

CCC &?A C M C l O ATA CCT WU: AGC C I G C M A X K T ATC CTG CGG G I C ACA ACA AW* TGT CCT GTC IrCA MA GTT GTG W C AGC P r o A l a G l n LCY I l e Pro A s p Ser G l n G i n LYS C ?I I l e Leu G l y A s p T h r Thr Arg Pro Vat T h r Lys Val V a l Asp Ser -178

1188-

c71 ATC CCc MG TIT CCT TI7 GTC M T GDG ax' GTT GII' GCT MC W ATA CCA l V C ACA Leu I l e P r o LyS P h c l \ l d P h e Val AS" G I Y C l y V a l V a l A l a A m C S l l e A l a Ser T h r C

1172-

CCA 1TC LOF CAG GAT CCC W T AM GOT CTA GTA lTC CTR ACC CAT GAC M C 'FGT OOP CTP ATA GGT GTC M T GGG m A G M Tn P w tle Ser G l n A s p A l g Scr LYs C\Y V a l V a l Phe Leu T h r H i s Asp A m C y s Gly Leu I I c G l y Val Asn G l y Val G I " Leu - 4 1 4

I J56-

I;

I;

B

B

S

ACC T7X Dui WCI CDC CGA Iw I h r C E G l y T h r G l y A r q A c 9 -406

8 ' B

0

TAT GCT M C CGG AGA GGG CAC GAT GCC ACT TGG GGC GTC CAC M C TTG ACA GTC GCT CCT GCA ATT GCT ATC ACA CCC A T T GAT A19 Air9 C l Y 116 AaP A!. m< T l p G l Y V a l G l n [ m ] V a l G l y P r o A l a I l e A l d Ile A i q P r o I l e Asp -462

h r A I a ASn 1440-

A T T lT'f C X AM W G C T GAT GCT ACG M T TTC TPG C M GAC TCT M C CCT GAG C V GAG AM GCA CGG AM A T C CTC TCG GAG I l c Ser Leu A m Leu A I * ASP A I a Thr Asn P h e Leu G l n Asp %r LYS A1a G I u Leu G l u Lys A I a hrg Lys l i e Leu Ser GI" -190

1324-

CTA DGT AGA TOC TIC MC TCA nwL GAG ACT cn; ATI LCC AX ATA GTA GFT AX GTC GTA ATA m m mc am ATA GTO ATC V a l G l Y Arq T:p Tyr A m Ser 1 1 4 G l u T h l V a l I l e Wr l i e I I c Val V a l Met Val V a l 11. ~ e uV a l Val l i e i l e V a l l i e - 5 1 8

1608-

ATC ATC G7G Ile I l C V I I

1692-

CCG AAG AV2 AGA CAT ATG T I C ACA MC GOT GGG W G A T GCA AlG GCT GAG M A AGA T O A T C A C G A C C A V A K A G * ~ T C Y T . 2 T ~ A ~ PCD Lya I l e Arq 11. h t T y r T h c A m G l y G l y P h e Asp A l a Met A1a G l u t y s A q 4.

1784-

CCA~WTA~CCG~AGAT~TCTATATMT

CTP TAT AW CTC 1GA LOG WA Leu Tyr A r 9 Leu A t 9 A t 9 Set

A l G CTA A% Om AAT CCA GAT W C CGT ATA cct AGO GAC AC1 FAC ACA T I A WIG Met Leu U e C G l y A m P r o Asp Asp A r g I l r P r o A r g Asp T h i Tyr T h i Leu G I " - 5 4 6

FIG. 3. Nucleotide sequence of the cDNA for Fo mRNA and the predicted protein sequence. The amino-terminal and carboxy-terminal hydrophobic domains are underlined. The hydrophobic domain at the predicted amino terminus of FI is marked with double underlining. The arrow marks the predicted site of proteolytic cleavage that activates fusion activity. The positions of cysteine residues (circles) and potential asparagine-linkedacceptor sites for carbohydrate (boxes) are indicated. (From Miura er al., 1985a.)

the culture medium alone, because without attached fragment B it cannot penetrate the cell membrane, but when introduced artificially into the cytoplasm of cells, one molecule is enough to kill a cell (Yamaizumi et al., 1978). Thus, when the artificial liposomes fuse with cells, fragment A is introduced into the cytoplasm and kills the cells. Liposomes containing viral spikes effectively killed cells, but liposomes without spikes showed no

10. SENDAI VIRUS-MEDIATED CELL FUSION

305

V

AMINO ACID SEQUENCE NUMBER FIG.4. Hydrophobicity profile of the unprocessed F protein. Numbers under the horizontal axis are positions from the amino-terminal methionine. The deduced positions of ( I ) the amino-terminal signal peptide. (7) the proposed amino terminus of the F , . and (3) the transmembrane anchor are shown. The site for proteolytic cleavage IS marked with an arrow. Potential acceptor sites for N-linked carbohydrate are marked with open triangles. (From N . Miura and Y. Okada. unpublished data.)

toxicity. Liposomes containing either HANA or F (active form) alone also showed no fusion. Thus, integration of both glycoproteins into the liposomes is essential for their fusion. Further analysis indicated the neceshity for a critical molar ratio of F to HANA of 2 : 1, shown in Fig. 5 (Nakanishi ef ul., 1982). The ratio of F to HANA in the intact virions was found to be about 1.7 by measurement of the intensities of bands on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE). These observations suggest that an optimum conformation of the two envelope glycoproteins is important for efficient fusion. Analysis with a monoclonal antibody against HANA also indicated that HANA protein participates in the cell fusion reaction as does F protein (Miura er d.,19821. The antigenic determinant site of the monoclonal antibody is different from the sites of hemagglutination and neuraminidase activities in the HANA molecule, and binding of the antibody does not inhibit either activity. HVJ treated with this monoclonal antibody can

306

YOSHIO OKADA

A

R at i o of proteins FIG.5 . Toxicity to L cells of liposornes containing fragment A. Liposornes were reconstituted with 0.6 rng of F protein, 0-0.9 mg of HANA protein, and 2 rng of lipids (A), or with 0.3 mg of HANA protein, 0-1.8 rng of F protein, and 2 rng of lipids (B). They were purified, and their toxicity for L cells was measured. Each sample contained about 2 pg/ml of fragment A. Control dishes contained (A) 119 colonies; (B) 140 colonies. (From Nakanishi er a / . , 1982.)

be adsorbed to cells and agglutinate them as effectively as native virus, but cannot cause cell-to-cell fusion. Treatment with antibody also inhibits the hemolytic activity and fusion of the viral envelope with the cell membrane. In addition, using a hybrid envelope formation technique, Ozawa et al. (1979a) observed that for virus-induced hemolysis, HANA of HVJ could not be replaced by HA glycoprotein (hemagglutinin) of influenza virus. These observations indicate that HANA protein is important in the fusion process as well as in receptor binding and receptor destruction.

IV. CELL-TO-CELL FUSION BY HVJ

This section is mainly concerned with the fusion of Ehrlich ascites tumor (EAT) cells by HVJ. EAT cells have certain very suitable characteristics for fusion studies, such as a high cell fusion capacity. Unlike cells

10. SENDAI VIRUS-MEDIATED CELL FUSION

307

FIG.6 . Time course of EAT cell fusion by HVJ, observed under a dark-field, phasecontrast microscope. (a) EAT cells: ( b ) cell aggregates with HVJ at 0°C: (c-f) cells incubated at 37°C for 5 , 10. I S . and 35 min. respectively. Arrows show a degenerated cell which is dissociated from fusion of cells in a cell aggregate.

growing attached to a substratum they have no extracellular matrix substances, and can be harvested in large amounts from mouse abdomen. Indeed, the first observation of cell fusion was that of fusion of EAT cells in mouse abdomen (Okada c't al., 1957). In Fig. 6, the time course of the cell fusion reaction is shown in a series of photographs. Cell aggregates formed by adsorption of HVJ onto cell surfaces at 4°C promptly fuse when the aggregates are shifted to 37"C, and formation of giant polynuclear cells is completed within 30 min. The initial cell aggregate in Fig. 6b consists of 13 intact cells and one degenerated cell (arrow). Figure 6f shows that these intact cells fused and only the degenerated cell was excluded from the polykaryocyte.

308

YOSHIO OKADA

A. HVJ Concentration and pH Range Required for Cell-to-Cell Fusion

Fusion of one virus envelope with a host cell membrane is enough for infection, but multiple virus particles are required for fusion of cells. Figure 7 shows the relation of the number of virus particles added per cell to the fusion index (FI = initial number of cells/number of cells after the cell fusion reaction - 1). From Fig. 7, the minimal number of particles for fusion was calculated to be about 250 HVJ particles per EAT cell and 120 particles per KB cell. Under the conditions used, about 80% adsorption of added virus onto cells would be expected. Thus, no fusion of EAT cells was detected when less than 200 virus envelopes were fused to a given cell membrane. When more than the minimum number of HVJ are added, cell fusion efficiency increases as the virus titer increases. The fusion capacities of different strains vary and depend also on the conditions of the cells, as described in the next section. Since multiple virus particles are required, the virus may be necessary for both agglutination of cells and modification of cell membranes for fusion.

P

/

KB CELLS

5

I 0

15

2b

2'5

3.0

35 x i 0 0

Virus p a r t i c l e N o . / c e l l FIG. 7. Dose-response curves for fusion of KB and EAT cells by HVJ. Results with three lines of EAT cells are shown ( A , B, and C). Cells A which had been preincubated under aerobic conditions had the maximum fusion capacity. B cells were preincubated under less severe conditions, and C cells were not preincubdted. (From Okada and Murayama. 1968.)

309

10. SENDAI VIRUS-MEDIATED CELL FUSION

As shown in Fig. 8, the cell fusion reaction by HVJ proceeds well in a neutral pH range, and the optimum pH is about 7.6 (Okada, 1962a). This pH range is characteristic for fusion by paramyxoviruses and is in clear contrast to the low pH range of cell fusion by influenza virus or VSV. The cell-to-cell fusion by HVJ decreases sharply on lowering the pH, and at pH 6.0 no cell fusion is observed.

6. Cell Lysis and Cell Fusion

HVJ shows lytic activity to cultured cells corresponding to its hemolytic activity. There is a close correlation between cell lysis and cell fusion induced by HVJ (Okada, 1958, 1962a.b: Okada and Tadokoro, 1962). When EAT cells and HVJ are suspended in a balanced salt solution and incubated at 37°C under anaerobic conditions, cell fusion is inhibited and the cells die within 30 min. On the other hand, when the cells are incubated under aerobic conditions, they fuse without undergoing lysis. Cell lysis and inhibition of cell fusion by HVJ are also demonstrable on addition of 2,4-dinitrophenol (Fig. 9). NaNl and NaCN show the same effect

5.0-

4.0.-

x Q,

D 3.0.-

.-C C

.-0 v)

2.0.-

1 LA 1

.o--

0.04

t-

6.0

7.0

8.0

PH FIG.8. pH dependence of the fusion reaction. 0. Fusion index; A, degenerated cells. (From Okada and Tadokoro, 1962.)

31 0

YOSHIO OKADA

P aJ

CI

!l

X

aJ C aJ 0,

aJ

U

.-C

aJ D

c

.-0

b b

0

wt

0

:

0.5

4

1.1

2.1

4.2

8.5

17

Final concentration o f DNP ( ~ 1 0 - 4 ) ~

FIG.9. Effect of 2.4-dinitrophenol on the fusion reaction. 0 , Fusion index of the test sample (left-hand ordinate); A,degenerated cells in the test sample (right-hand ordinate); A, degenerated cells in a control sample without virus. (From Okada, 1962b.3

as 2,4-dinitrophenol; however, when glucose is added to the medium, cell lysis induced by these reagents is inhibited and cell fusion occurs (Yanovsky and Loyter, 1972). Therefore, the ATP-generating system of the cells supports cell fusion, and suppression of ATP generation inhibits cell fusion and induces cell lysis (Okada et al., 1966). Homma et al. (1976) showed that hemolytic activity is high in aged and low in non-aged HVJ particles immediately after they bud from the host cell. The non-aged HVJ particles have cell fusion activity and can kill cells in the presence of NaN3 (Y.Okada, unpublished data). A correlation of cell lysis and cell fusion has also been observed under conditions with or without Ca2+ions(Okada and Murayama, 1966). When Ca2+ ions are removed from the reaction mixture, cell lysis becomes predominant and cell fusion is suppressed. Ca2t ions at a concentration of 420 p M are necessary for fusion and can be replaced by Sr, Ba, or Mn, but not by Mg ions. As shown in Fig. 10, the sensitivity of cells to Ca?+ions appears promptly when the cells are shifted to 37"C, it decreases rapidly

31 1

10. SENDAI VIRUS-MEDIATED CELL FUSION

1 looo~\ @

\

0

0

5

0

10

15

30( m i n )

T i m e of addition of E D T A Fic. 10. Effect of EDTA on the cell fusion reaction. Volumes of 0.5 ml of a 10% suspension of EAT cells and 0.5 ml of 3,000 H A Uiml of HVJ. both suspended in BSS containing 420 p M of CaCL. were mixed in a series of test tubes. The tubes were kept at 4°C for 10 min and then incubated at 37°C with shaking. During incubation, 0.5-mI volumes of 3 mM EDTA were added to the tubes at different times indicated on the abscissa, and then the mixtures were incubated for a further 60 min at 37°C. 0 . Total number of cells observed; 0. number of intact cells. (From Okada and Murayama, 1966.)

during incubation. The time course of appearance of sensitivity of ATP is similar to that of sensitivity of Ca?' ions. For electric pulse-induced fusion, cell lysis and cell fusion have also been reported to be correlated in the presence of divalent cations (Ohno-Shosaku and Okada, 1985).

312

YOSHIO OKADA

C. Virus-Cell Interaction at Low Temperatures

On incubation of a cell-virus mixture at low temperatures such as 015"C, no cell-cell or cell-viral envelope fusion is observed. However, there is some interaction between the virus and the cells: HVJ particles are adsorbed to the cell surface, and huge cell aggregates are produced at 0°C. When erythrocytes are added to the sample, they are adsorbed onto the cell aggregates, that is, the surfaces of the initial cell aggregates are hemadsorption positive. But, on incubation at WC,the aggregates become hemadsorption-negative and partial damage of the ion barrier of the cell membranes occurs. Electron microscopic observations have shown that immediately after formation of cell aggregates, the viral particles, which are localized on the free cell surfaces but do not bridge two cells, are adsorbed onto the cell at one side of the envelope, while the other side is free. Red blood cells can then attach to the unoccupied side of the virus (hemadsorption positive). When the cell aggregates become hemadsorption negative, almost all the envelope surfaces of the virus particles become closely attached to the cell surfaces, and are engulfed by the cell membrane, as shown in the model in Fig. I I . When observed under a dark-field, phase-contrast microscope, the initial cell aggregates were seen to be composed of bright spherical cells. On incubation at O-IS"C,the cells in the aggregates became slightly darker and swollen (especially at ISOC)and lost their hemadsorption activity. On loss of hemadsorption, a decrease in potential across the cell membrane was detected: as shown in Table I1 and Fig. 12, the potential of native cells of - 18.2 2. I mV decreased to -6.1 2 2.5 mV. The mernbrane resistance also decreased from 491 to 162 ohm cm2. These results

*

HA- p o s i t i v e

HA-negative

FIG.1 1 . Model of appearance of hemadsorption (HAhegative stage in cell aggregates with HVJ on incubation at temperatures below 15°C.

313

10. SENDAI VIRUS-MEDIATED CELL FUSION

suggest that the cell membrane structure has been partially damaged and can no longer serve as a complete ion barrier. This kind of damage was not seen in cell aggregates produced by adsorption of the inactive form of HVJ that contains Fo glycoprotein (Okada, 1972; Okada rt 01.. 1975). It is reasonable to think that the terminal region of the active form of F protein is inserted into the lipid bilayer of cell membranes, resulting in the distortion of their structure, although F is a hydrophilic glycoprotein and thus should have difficulty in approaching the lipid bilayer. The high hydrophobicity in the amino-terminal sequence of FI (Gething e t d., 1978) may overcome this barrier. Asano and Asano (1984) demonstrated directly the insertion of the terminal sequence of Fl into the lipid bilayer of human erythrocytes at 0°C by utilizing photoaffinity labeling. On incubation at low temperatures, HVJ particles are thought to approach the lipid bilayer of cell membranes in stepwise fashion by binding first to sialoproteins then to sialolipids by HANA activity, and finally the amino-terminal portion of Fl is inserted into the lipid bilayer. This process seems to be more rapid at IS" than at 0°C. The finding that the surfaces of HVJ particles become covered with cell membranes (loss of hemadsorption activity) seems to indicate a high binding force of the virus to cell membranes. This in turn may be due largely to the interaction between HANA and the receptor. Insertion of FI into the lipid bilayer may also

'TABLE I I CHANGES I N POTENTIAI A N D MLMBRANE RESISTANCE OF FL CFI I S DUHINL FUSION BY HVJ,' Treatment of cell3 Untreated At 35°C At 10°C

With HVJ At 10°C At 37°C For 5 min

Potential (-mV)

Membrane resistance (ohm cm!)

22.0 18.2

_t

63

2

2.1

437 491

2 2

71 149

6.1

?

2.5

162

2

84

3.5

2 1.6

92

2

46

For 10 min

10.5

?

6.7

-

For 15 min

20.5

2

7.3

-

For 30 min

23.2 2 6.3

" Modified from Okada (1972)

513

-t

192

314

YOSHIO OKADA

?6

60..

IO"C

at

5040-

30..

20.. 10.

0%

10

15

.

20

.

25

100.. 9 0.

37°C

at

80.

for

5min

7 0. 60 5 04 0-

30.

20. 10..

0-

L 10

15

20

25

30

35

40

-mV

Potent i a I FIG.12. Decrease of potential across the cell membrane induced in FL cells by HVJ containing active F,,z (filled columns), but not by HVJ containing inactive F,, (open columns), on incubations at 10" and 37°C for 5 min. (From Y . Okada, unpublished.)

contribute. This high binding force is effective for close contact of the two cell membranes and will provide a suitable arrangement for cell-to-cell fusion (Figs. I t and 22). D. Early Events Observed on Incubation at 37°C

When the hemadsorption-negative cell aggregate is incubated at 37"C, it promptly becomes hemadsorption positive, the dark cells become bright,

315

10. SENDAI VIRUS-MEDIATED CELL FUSION

P

E I

0

40..

30.-

200

to-

0 1 :

:

:

30

;

i

:

60

;

90

120

rnin

The time determined P o t e n t i a l FIG.13. Potential of FL cells (treated with HVJ at 37°C for S min) determined at 10°C and then at 3S"C. (From Okada. 1972.)

and then fuse. During these events. the potential across the membrane first decreases more than that of cells incubated at low temperature and then increases sharply to levels above that of native cells (Fig. 13). Cells incubated for 5 min at 37°C show minimum levels of both potential ( - 3 . 5 2 1.6 mV) and membrane resistance (92 ohm cm? (Table 11). Cells treated with inactive Fo type HVJ show no change in potential, with the potential remaining at the level found in native cells during incubation at either low temperature or 37°C. The appearance of a minimal potential at 37°C thus seems to correlate with the unique amino-terminal structure of FI. With regard to the damage of the ion barrier observed at low temperatures and 37"C, Yamaizumi et ul. (1979) showed that the damage at low temperature did not permit the penetration of macromolecules into the cytoplasm, whereas the damage induced at 37°C did. They used fragment A of diphtheria toxin as a probe. Fragment A does not show any toxicity to native cells when included in the medium. Concomitant treatment with both UV-inactivated HVJ and fragment A at 0°C also had no killing effect, but cells incubated at 37°C with both materials were killed because protein synthesis was inhibited. These results showed that fragment A could

316

YOSHIO OKADA

penetrate the cytoplasm through the cell membranes and inactivate EF-2. The penetration efficiency was highest immediately after the temperature was raised to 37°C and decreased over the course of incubation at 37°C. This phenomenon was utilized to insert UV-specific endonuclease V (about 16 kDa, derived from T4 bacteriophage) into cultured xeroderma pigmentosum cells. The insertion proved successful, and the hereditary dysfunction of the cells (unscheduled DNA synthesis) was overcome (Tanaka et al., 1975). These observations suggest that the disorder appearing in cell membranes at low temperatures permits only an ion leak, whereas the disorder produced at 37°C permits macromolecules to diffuse through the cell membrane. In turn, this is followed by cell death when calcium ions are absent or when the ATP-generating system of the cells is inhibited as described above. The disorder can be repaired on further incubation at 37°C under suitable conditions, as shown by the restoration of the potential across the membrane. This restoration cannot be explained by the repair of the membrane disorder only, and must be associated with the energy-requiring higher functions of the ion pumps in the cell membranes. The mechanism by which cell membranes are damaged by the active form of F protein at 37°C is unknown. Asano and Asano (1984, 1985) found that specific binding of cholesterol to purified F glycoprotein occurs extensively on incubation at 37°C: such binding is poor at 20°C (Fig. 14).

01 0

L

200

F protein

400

( p rno~es/qop~)

FIG.14. Binding of cholesterol to F glycoprotein. (From Asano and Asano, 1985.)

10. SENDAI VIRUS-MEDIATED CELL FUSION

317

Asano and Asano constructed Stuart-type models of FI and examined the possibilities of a three-dimensional formation of a complex with cholesterol. If it is assumed that a hydrogen bond is formed between the 3phydroxyl residue of cholesterol and the amide nitrogen bond of Gln 20, and if, in addition, a hydrophobic interaction is assumed to occur between the conserved region of the amino terminus and a cholesterol molecule, it becomes easy to construct the complex. These authors therefore concluded that this segment could be the cholesterol binding site and that reorganization of the segment to its cholesterol-binding form might be a temperature-dependent step. These findings suggest that the marked damage of cell membranes observed at 37°C may be induced by the binding of the amino terminus of FI to cholesterol in the lipid bilayer of cell membranes. The damage may be followed by cell lysis; alternatively, it may be repaired and the ionic balance in the cytoplasm may be restored in the presence of Ca2+ions and ATP. At this stage of marked damage, Ohki et al. (1975, 1978) observed an increase in cyclic AMP as a function of the viral dose. There seems to be a correlation of cyclic AMP levels with cell fusion, because the concentration of cyclic AMP was increased by raising the virus dose in the range that exhibited a dose-dependent response in cell fusion. The concentration reached a maximum (about 2-fold) on incubation for 2 min at 37°C and then decreased to the initial level in 20 min. Moreover, addition of inhibitors of cyclic nucleotide phosphodiesterase, such as theophylline, increased the frequency of cell fusion 2-fold. The stimulatory effect of theophylline was dependent on the extracellular concentration of Ca? ions, with optimal concentration at 0.5 mM. +

E. Alteration of Cytoplasmic Organelles and Their Reverslon

Transformation of organelles appears abruptly, the degree depending on cell membrane damage (Kim and Okada, 1980). As shown in Figs. 15 and 16, when cells were treated with HVJ at 37°C for only 3 min, the mitochondria became condensed with enlarged intracristal spaces, the cisternae of the ER and Golgi stacks became highly distended, and the inner space of the nuclear envelopes became enlarged. The arrangement of 100-Afilaments also became disordered. On further incubation at 37"C, all of these morphological changes were reversed within 30 min, except those of the mitochondria. Incubation for about 60 min at 37°C was required for restoration to the typical orthodox form of mitochondria (Fig. 17). In cells treated with HVJ at o"C, an intermediate form of transformation

318

YOSHIO OKADA

FIG.15. Ultrastructure of an intact EAT cell. x 12,300. Asterisk shows the Golgi area. Bar, 1 prn. (From Kim and Okada, 1980.)

of organelles gradually appeared during incubation. On incubation for 5 min, some cells showed no change, but in others, the orthodox form and slightly changed mitochondria were mixed, although no distinct alterations of the ER, nuclear envelopes, or 100-A filaments were detectable. On further incubation at 0°C the proportion of cells with mitochondria in intermediate stages increased, but the extent of transformation never became as great as that at 37°C. These morphological changes of cytoplasmic organelles are probably side effects resulting from damage to the ion barrier of the cell membrane

10. SENDAI VIRUS-MEDIATED CELL FUSION

31 9

FIG. 16. An EAT cell treated with HVJ at 37°C for 3 min. Remarkable alterations are seen in most of the cytoplasmic organelles. x 12.600. Arrows mark endoplasmic reticulum, and the asterisk indicates the Golgi area. Bar. 1 gm. (From Kim and Okada. 1980.)

induced by HVJ at low temperatures and at 37°C. But at 37°C the spherical morphologies of transformed organelles, such as mitochondria, ER, and Golgi apparatus, resemble those generally observed on isolation of cell components; this suggests that a cytoplasmic apparatus that supports the normal morphologies of organelles in native cells-likely a cytoskeletal component-may be disordered on disruption of the ion barrier of the

320

YOSHIO OKADA

FIG.17. Typical features of the endoplasmic reticulum (top) and of mitochondria (bottom) at each step of the cell fusion reaction. The morphological transformations of the two orgenella and their reversions are synchronous, except that recovery of the normal mitochondrial matrix takes longer. (a) Intact cell; (b-d) cells incubated with HVJ at 37°C for 3, 30, and 60 min, respectively. Endoplasmic reticulum (a, c, d) x40,000, (b) ~ 2 0 , 0 0 0mito; chondria (a-d) x37,500. Bars, I p m . (From Kim and Okada, 1980.)

cell membrane. The cell membrane is also an organelle associated with the cytoskeleton. Under conditions inducing the transformation of organella, the cell membrane structure may show some changes, such as clustering of intramembrane particles (see below). F. Separation of Viral Envelope Fusion and Cell Fusion

Virus-induced cell fusion was found to be strongly inhibited by cytochalasin D (Asano and Okada, 1977) and cytochalasin B (Pasternak and

10. SENDAI VIRUS-MEDIATED CELL FUSION

321

Micklem, 1973). The cell-to-cell fusion induced by HVJ was inhibited completely by 5 pglml of cytochalasin D, but fusion of viral envelopes with cell membranes was not inhibited. In the presence of cytochalasin, the amount of viral envelope fusion was calculated to be 97.696 of that in its absence (Miyake et d., 1978). In the presence of cytochalasin D, the huge cell aggregates induced by virus adsorption at 0°C become dissociated on incubation at 37°C into unfused, single cells. Except for cell-cell fusion, almost all of the predicted interactions between viral envelopes and cell membranes occurred. On incubation at 0°C the hemadsorption-negative stage appeared on the surface of the cell aggregates, while at 37°C hemadsorption reappeared: transformation of organelles was observed; the stage of sensitivity to ATP was induced; and viral envelopes fused with the cell membranes. Detailed examination showed that all these events appeared a little later and to a lesser extent than in the absence of cytochalasin D. This separation of two kinds of fusion was also observed on addition of high concentrations of saccharides, such as 0.38 M glucose, mannose. or galactose or 0.25 M sucrose (Maeda ~t a / . . 1977). The addition of saccharides did not have the effect of cytochalasin in that an ATP-sensitive stage at 37°C did not appear and mixing of viral materials with cells due to viral envelope fusion was strongly inhibited. With saccharides, the extent of viral envelope fusion was calculated to be 82% of that without saccharides, Even though the intermediate step of viral envelope fusion was seen after incubation at 37°C for 30 min, the fused virus still retained its specific structure on the cell surface (Fig. 18). In the control without saccharides or with cytochalasin, the structure of fused envelopes disappeared completely, the area of fusion was only detectable antigenically, and no nucleocapsids were detected in the cytoplasm near the area of fusion. The result of a spin-label study on phospholipid molecules of viral envelopes mixed with EAT cells is shown in Fig. 19. Virus particles containing spin-labeled phosphatidylcholine (PC*) were mixed with unlabeled EAT cells ( A . Asano et ul., unpublished data). Mixing did not occur at 0°C but under standard conditions promptly appeared on incubation at 37°C increased for 15 min, and then decreased sharply after 20 min, probably due to the disappearance of spin-labeled PC" by exposure to the reductive conditions of the cells. This decrease was not seen in the fusion of viral envelopes with human erythrocytes (Maeda et al., 1975). With cytochalasin mixing occurred, but the reaction was slightly slower and less than that without cytochalasin. In the presence of saccharide, the mixing was incomplete and remarkably slow. In the case of saccharides, the slow rate of the interactions between the viral envelopes and cell membranes at 37°C under such hyperosmotic

FIG.18. Electron micrograph of an EAT cell treated with HVJ at 37°C for 30 min in a medium containing 0.5 M glucose. Fusion of a viral envelope with the cell membrane is seen, but the specific structure of the envelope remains intact and the nucleocapsid is still present inside the envelope. Bar, 0.1 km. (From Maeda er a / . , 1977.)

10. SENDAI VIRUS-MEDIATED CELL FUSION

323

conditions may be the reason why cell-to-cell fusion was separable from viral envelope-cell membrane fusion. This separation with cytochalasin D may be different from what is the case with saccharides. Since cytochalasin D is known to be a microfilament inhibitor, microfilaments probably participate in HVJ-induced fusion of EAT cells. Ohki et 01. ( 1978) reported that an increase in extracellular calcium concentration above 0.5 mM partially antagonized the inhibitory effect of cytochalasin D; that is, cytochalasin D inhibited cell-to-cell fusion more effectively at lower calcium concentrations. Pretreatment of cells with theophylline partially prevented inhibition of cell fusion by cytochalasin D. The fact that cell-to-cell fusion is separable from cell-to-viral envelope fusion under the conditions described above should be considered in the mechanism of EAT cell fusion induced by the virus.

G. Perturbation of Cell Membrane Structure by HVJ Much information about clustering of intramembrane particles (IMPS) during the fusion of erythrocytes has been obtained by freeze-fracture electron microscopy. This clustering seems to be a general phenomenon induced by fusogens such as Sendai virus (Biichi r t al., 1973). Calcium ionophores (Vos et id., 1976),and Calcium phosphate (Zakai et al., 1977). When erythrocytes treated with these fusogens at 37°C are fixed after cooling to a low temperature, such as 4°C. and then freeze-fractured, many clusters of IMPs and smooth membrane areas depleted of IMPS are observed on the P-face. Erythrocytes that have not been treated with fusogens do not show this clustering of IMPs. Fixation at low temperature is necessary for demonstration of clustering in fusogen-treated erythrocytes; no clustering is observed when the cells are fixed at a temperature suitable for fusion, such as 37°C. Thus, it is believed that clustering is induced by phase separation of IMPS from lipids at low temperatures (Volsky and Loyter, 1978) and that the erythrocyte membrane structure available for fusion has the characteristic of allowing phase separation. However, it was difficult to decide whether this characteristic is critical for fusion or is a side effect of treatment with fusogens. Sekiguchi and Asano (1978) obtained the following interesting results concerning the correlation between IMP clustering and fusion of erythrocyte ghosts. Human erythrocyte ghosts, loaded with bovine serun albumin, fused on treatment with HVJ and showed IMP clustering. But erythrocyte ghosts loaded with anti spectrin antibody plus bovine serum albumin showed greatly reduced fusion and no clustering of IMPs. Fab fragments prepared from the antibody did not have this effect. These results suggest that if the spectrin meshwork is fixed by antispectrin anti-

324

YOSHIO OKADA

b

10

2‘0

30

4b rnin

Incubation time at 37°C

FIG.19. Phospholipid intermixing between HVJ envelopes and EAT cell membranes at 37°C under conditions with 0.5 M glucose (W), with 10 pg/ml cytochalasin D (O)?and HVJ containing spin-labeled phosphatidylcholine (PC*) in its without either reagent (0). envelope was mixed with EAT cells at 0°C for 5 min and then incubated at 37°C. The intermixing was assayed by the change of the ESR spectrum of PC*.(From A. Asano, T. Maeda, S . Ohnishi, and Y. Okada, unpublished data.)

body, the IMP pattern is also fixed and results in inhibition of erythrocyte fusion by the virus. Cold-induced clustering of IMPs has also been observed in mouse L cells treated with polyethylene glycol (PEG) (Robinson et al., 1979; Roos et al., 1983). L cells, a fibroblast strain, treated with PEG at a less than critical concentration for fusion, showed no clustering of IMPs, but, when treated with 50% PEG, more than 90% of the cells underwent fusion and displayed IMP clustering. Furthermore, mutant lines of L cells that were resistant to cell fusion showed no fusion or IMP clustering when treated with 50% PEG but, when treated with higher concentrations, showed extensive fusion and IMP clustering. These findings are similar to what happens when erythrocytes undergo fusion.

325

10. SENDAI VIRUS-MEDIATED CELL FUSION

The cold-induced clustering of IMPs was also observed in EAT cell fusion by HVJ, but the appearance was transient. At 37°C for 2 min, slight aggregation of IMPs was observed but disappeared within a few minutes. Fusion at 37°C is too rapid to allow identification of sequential morphologic changes such as IMP clustering. This differs from the above cases. In the erythrocyte system, there is no restoration stage of IMP clustering, and IMP clustering of L cells with PEG seems to appear slowly and to persist for over 90 min at 37°C (Roos c ~ ta / . , 1983). For the demonstration, it was necessary to make the fusion reaction slower. Distinct and marked clustering of IMPS was observed when the reaction temperature was lowered from 37°C to 28°C (Kim and Okada, 1981). At 28"C, clustering was seen most clearly after incubation for 10 min (Fig. 20). This stage corresponds to a very early stage of the cell fusion reaction compared with the stage incubated at 37°C for only 2 min, because transformation of organelles has been seen in the cytoplasm of all of the cells and fusion is initiated in only a few cells, but the rest are still at the stage of cell-to-cell attachment. This cold-induced clustering of IMPs is apparently a reversible step that occurs in the early stage of cell fusion, like the transformation of organelles. When the cells are incubated at 28°C for 20 rnin and then chilled, only S0-60% of them still have IMP clusters. On incubation at 28°C for 10 rnin and then at 37°C for 10 min, no cells showed cold-induced IMP clustering. Since cytochalasin D inhibited the clustering of lMPs induced by the treatment with HVJ at 28°C for 10 min, the inhibition of cell-to-cell fusion by cytochalsin D may be due to appearance of conditions that do not allow the clustering of IMPs. H. Evidence for Close Attachment of Cell Membranes Induced by HVJ

For cell fusion to occur, the lipid bilayers of adjacent cell membranes must bind closely, and mutual repulsion between them must be overcome. In virus-mediated cell aggregates at O'C, the plasma membranes of adjacent cells remain at least IS0 apart, and the distance between the viral envelopes and plasma membranes is also IS0 When these cell aggregates are shifted to 37"C, cell fusion occurs very rapidly, and it is difficult to identify the stage of close binding of two lipid bilayers. But on incubation at a lower temperature, this stage becomes detectable as in the case of the IMP clustering (Kim and Okada, 1981). On incubation at 28°C for 10 rnin, the initial huge cell aggregates dissociate into small aggregates, which are stable, and the cells in the aggregates fuse during further incubation. In the aggregates, it is possible to

A

A.

FIG.20. Freeze-fracture image showing the region of close contact between two agglutinated cells (A and B). Cells were incubated at 28°C for 10 min, cooled, and fixed at 4°C. Striking aggregation of intramembrane particles is clearly seen on the P-face. The fracture face is hollowed here and there, and the intramembrane particles appear clustered in the hollows. Leaflets of plasmalemma (arrowheads) and virus particles (arrows) are seen on the P-face of cell A. The asterisk shows ice present between a virion and an IMP cluster. ~ 4 5 , 0 0 0 .Bar, 0.1 pm. (From Kim and Okada, 1981.)

10. SENDAI VIRUS-MEDIATED CELL FUSION

327

identify many sites where two apposing membranes are protruding and external leaflets of the two plasma membranes are in direct contact (Fig. 21). This occurs at the same stage as cold-induced IMP clustering with transformation of organelles in the cytoplasm. Virus particles are not seen in these regions of membrane contact. but they are frequently detected slightly apart from these regions in hollows covered by two cell membranes. A few fusing sites of cell membranes are detectable immediately after the formation of narrow cytoplasmic bridges between two cells at this stage; this may represent a step that follows formation of the contact sites on the two membranes. The freeze-fracture image shows that the fractured P-face (Fig. 20) is not flat, but has many hollows and bulging areas. The bulging areas on the fracture plane are smooth, with no IMPs, and they seem to resemble the attachment sites of two plasma membranes (Fig. 21), which are also projecting areas. The IMPs are concentrated in the hollows. Several virus envelopes can be observed on the fracture face attached to the clustered IMPs. These appear to be bound to receptors on the cell membrane and not fused with the cell membrane, because ice is detectable between the two surfaces. Thus, they could correspond to the virus particles attached to two cell membranes by a distance of about I50 A (Fig. 21). The image of the fractured plane in Fig. 20 may be interpreted to mean that the structure of the area in which two plasma membranes are in close contact is so stable that IMPs are excluded from this area when the cell aggregate is cooled to 4°C. On the other hand, IMPs on which virus particles are adsorbed are fixed in their original position by formation of the virus-receptor complex. On cooling to 4"C, these IMPs may act as nuclei for clustering other IMPs that are excluded from the areas of direct attachment of two plasma membranes, thus forming large IMP clusters. This speculation seems to be supported by the fact that at the same stage, single cells that have been liberated from cell aggregates show a flat fracture face with much smaller randomly distributed IMP clusters. On every IMP cluster in hollows of the fracture plane (Fig. 20), a virus particle may not be adsorbed, because the hollows are very numerous and it is unlikely that so many virus particles would have been adsorbed under the conditions used. In hollows without virus particles, clustering of IMPs may occur because there is no pressure to exclude them from these areas; this is not true in areas of attachment of two bilayers. Areas of attachment of two lipid bilayers should be the sites of rapid fusion at 37°C. The areas of attachment of two plasma membranes may not be the sites at which viral envelopes have fused, because the fractured P- and E-faces provide no evidence for this. It is conceivable to consider

FIG.21. Direct attachment of two lipid bilayers of EAT cells treated with HVJ at 0°C for 5 min and then at 28°C for 10 min. (a) Two cells are very close together and are in direct

contact in small localized regions immediately before fusion. (b) Higher magnification of regions of contact of two apposing cell membranes. The membranes are very close together over a broad area and are in direct contact in several small regions. (c) Higher magnification of closely approximated membranes showing virus particles near the area. (d) Cytoplasmic bridges formed immediately after membrane fusion. Virus particles are seen near the bridges. Arrows mark virus particles. Bar, 0.1 pm. (From Kim and Okada, 1981.)

10. SENDAI VIRUS-MEDIATED CELL FUSION

329

that the rapid fusion of EAT cells may be carried out through the sites of direct attachment of two plasma membranes induced by the perturbation of cell membranes before the viral envelope-cell membrane fusion.

1. Discussion and Summary of Erlich Ascites Tumor Cell Fusion by HVJ The process of cell fusion may be summarized by the model shown in Fig. 2 2 . ( I ) On incubation at temperatures below 15°C no cell fusion occurs, but conditions that allow fusion develop (Fig. 2 2 , 1 and 11). Virus particles bind first to sialosaccharide groups that may be part of glycoproteins, then to sialolipids, and approach closely the cell membrane; the amino-terminal hydrophobic residue of F , can then be inserted into the lipid bilayer of the cell membrane. This process may involve stepwise binding to sialo residues that have stronger binding activities, and depends mainly on the activities of HANA spike proteins. The combination of HANA and F spikes on the envelope may facilitate the transport of virus particles. As result, a strong binding force is established between the virus and the cell membrane, two cell membranes become attached and fixed by the virus, and a suitable structure is established for fusion. ( 2 ) On incubation at 37"C, FI spikes inserted into cell membranes undergo conformational change. This results in the folding in cholesterol and the perturbation of the lipid bilayer, followed by marked damage to the cell membrane. This damage in turn induces disarrangement of the cytoskeletal system, caused by the sudden introduction of calcium ions into the cytoplasm, and allows the membrane proteins to become freely mobile. The two lipid bilayers can then come into direct contact without interference from membrane proteins (Fig. 2 2 , 111). In these areas of attachment, formed at low temperatures, the membranes promptly fuse at 37°C (Fig. 2 2 , IV). Lucy and Ahkong (1986) concluded from studies on a model of fusion of biological membranes that the protein-free region required for fusion is only 30 nm in diameter. When considered in relation to the whole cell fusion reaction, the fusion efficiency can be seen to depend on the balance between the rate of dissociation of cells from the initial cell aggregates and the rate of cell fusion in the aggregates. Dissociation of cells occurs rapidly when cell aggregates are incubated at 37°C. It is due to destruction of viral receptors by the neuraminidase activity of HVJ, followed by mutual repulsion of the cells. Quantitiative studies (Okada et al., 1966) showed that the rate of dissociation of EAT cells is greatest between 2.5 and 5 min after the beginning of incubation at 37°C. Within the first 5 min, about 7 5 4 0 % of the cells have dissociated from the initial cell aggregates. Thus, in such

YOSHIO OKADA

330

I

a t 31°C

m

FIG.22. Model of fusion of EAT cells by HVJ.

conditions, cell-to-cell fusion must be carried out before the cells have time to dissociate. In fact, on incubation of EAT cells at 37"C, fusion is detectable by electron microscopy within the first 2 min. This stage corresponds to that incubated for 10 min at 28°C. In Fig. 21d, cytoplasmic bridges form immediately after fusion of two cells, and virus particles still remain near the bridges. These findings suggest that the viral envelope fusion is slower than the rapid fusion observed (Fig. 22, V). This rapid fusion occurs where two lipid bilayers become attached directly and not through viral envelope-cell membrane fusion. As indicated in Sections IV,G and IV,H, on incubation at 37°C such an attachment region appears immediately. If the establishment of such regions is delayed, the apparent fusion efficiency decreases greatly as described in Section IV,F, yet the viral envelope still undergoes fusion with cell membranes. Cell-to-cell fusion is, of course, possible through the viral envelope fusion step, if one virus envelope fuses with two different cell membranes. In fact, Knutton (1977) has observed this phenomenon in an elec-

10. SENDAI VIRUS-MEDIATED CELL FUSION

33 1

tron microscopy study during human erythrocyte fusion. It must be considered different from the mechanism described here. Calcium ions and ATP are essential for cell-to-cell fusion of EAT cells but not for viral envelope-EAT cell membrane fusion. Volsky and Loyter (1978) reported that calcium ions are also essential for cell-to-cell fusion of chicken erythrocytes, but not for viral envelope-erythrocyte membrane fusion. Chicken erythrocytes are nucleated and contain cell microtubules. On the other hand, calcium ions and ATP are not needed for cell-to-cell fusion of human erythrocytes (Perez rf a l . , 1974). These observations suggest that viral envelope fusion could be involved to some degree in human erythrocyte fusion, but it does not appear to be involved to any significant degree in fusion of chicken erythrocytes and EAT cells. This may be due to differences between structure and fuction of cytoskeleton-cell membrane system of enucleated erythrocytes and nucleated cells. Two such mechanisms may be involved. One is through viral envelope fusion, and another is the fusion of two cell membranes at the sites of direct attachment mentioned here. The rate of the former is slower than the rate of the latter, and the latter becomes dominant when the dissociation rate from cell aggregates is high, as in the case of EAT cells and chicken erythrocytes. The mode of inhibition of EAT cell fusion by cytochalasin D can be explained by assuming that areas of close contact of the two lipid bilayers are either delayed or fail to be established. This in turn may be explained by the 2-min incubation at 37°C and disappearance for a few minutes in EAT cells. This may be the reason why toxicity of HVJ is lower than that of PEG. In the case of PEG, IMP clustering was observed when L cells were incubated for 90 min at 37°C (Roos et a / . , 1983). It is well known that surface antigens of two different cells are rapidly mixed after fusion induced by Sendai virus (Frye and Edidin, 1970). This may involve the perturbation of the cytoskeleton-membrane protein interaction induced by the virus at 37"C, as described here. ACKNOWLEDGMENTS The author wishes to thank Drs. J . Kim and T'. Uchida of the Institute for Molecular and Cellular Biology. Osaka University, and Dr. A . Asano of the Cancer Research Institute. Sapporo Medical College. for helpful discussions. The excellent electron micrographs were provided through the courtesy of Dr. J . Kim. REFERENCES Ahkong, Q. F.. and Lucy, J. A. (1986). Osmotic forces in artificially induced fusion. Biochim, Biophvs. Actu 858, 206-216. Asano. A , . and Asano. K. (1984). Molecular mechanism of virus entry to target cell>. 7Lnior Nes. 19. 1-20,

332

YOSHIO OKADA

Asano, A., and Okada, Y . (1977). Inhibition of virus-induced fusion of Ehrlish ascites tumor cells by cytochalasin B and D. Life Sci. 20, 117-122. Asano, A., and Sekiguchi, K. (1978). Redistribution of intramembrane particles of human erythrocytes induced by HVJ (Sendai virus): A prerequisite for the virus-induced cell fusion. J. Suprumol. Struct. 9, 441-452. Asano, K., and Asano, A. (1985). Why is a specific amino acid sequence of F glycoprotein required for the membrane fusion reaction between envelope of HVJ (Sendai virus) and target cell membranes? Biochem. Int. 10, 114-122. Bachi, T., Aguet, M., and Howe, C. (1973). Fusion of erythrocytes by Sendai virus studied by immuno-freeze-etching. J. Virol. 11, 1004-1012. Blumberg, B . , Giorgi. C., Roux, L., Raju, R., Dowling, P., Chollet. A., and Kolakofsky. D. (1985a). Sequence determination of the Sendai virus HN gene and its comparison to the influenza virus glycoproteins. Cell 41, 267-278. Blumberg, B. M., Giorgi, C . , Rose, K., and Kolakofsky, D. (1985b). Sequence determination of the Sendai virus fusion protein gene. J. C e n . Virol. 66, 317-33 I . Davidson, R. L., and Gerald, P. S. (1977). Improved techniques for the induction of mammalian cell hybridization by polyethylene glycol. Somatic Cell Genet. 2 , 165-176. Frye, L. D . , and Edidin, M. (1970). The rapid intermixing of cell surface antigens after fusion of mouse-human heterokaryons. J. Cell Sci. 7 , 319-335. Fukai, K., and Suzuki, T. (1955). On the characteristics of a newly isolated hemagglutinating virus from mice. M e d . J . Osaka Uniu. 6 , 1-15. Furusawa, M., Nishimura, T., Yamaizumi, M., and Okada, Y . (1974). Injection of foreign substances into single cells by cell fusion. Nature (London) 249, 449-450. Gething, M. J., White, J. M., and Waterfiled, M. D. (1978). Purification of the fusion protein of Sendai virus: Analysis of the NH?-teminal sequence generated during precursor activation. Proc. Natl. A c a d . Sci. U . S . A . 75, 2737-2740. Homma, M. (1971). Trypsin action on the growth of Sendai virus in tissue culture cells. 1. Restoration of the infectivity for L cells by direct action of trypsin on L cell-borne Sendai virus. J. Virol. 8, 619-629. Homma, M. (1972). Trypsin action on the growth of Sendai virus in tissue culture cells. I I . Restoration of the hemolytic activity of L cell-borne Sendai virus by trypsin. J . Virol. 9, 829-835. Homma, M., and Ohuchi, M. (1973). Trypsin action on the growth of Sendai virus in tissue culture cells. 111. Structural difference of Sendai viruses grown in eggs and tissue culture cells. J . Virol. 12, 1457-1465. Homma, M., Shimizu, Y. K., and Ishida, N . (1976). On the study of Sendai virus hemolysis. I. Complete Sendai virus lacking in hemolytic activity. Virology 71, 41-47. Hosaka, Y . (1958). On the hemoytic activity of HVJ. Biken J. 1, 70-89. Hosaka, Y . , and Shimizu, Y. K. (1972a). Artificial assembly of envelope particles of HVJ (Sendai virus). I. Assembly of hemolytic and fusion factors from envelopes solubilized by Nonidet P40. Virology 49, 627-639. Hosaka, Y . , and Shimizu, Y. K. (1972b). Artificial assembly of envelope particles of HVJ (Sendai virus). 11. Lipid components for formation of the active hemolysin. Virology 49, 640-646. Ishida, N., and Homma, M. (1961). Host-controlled variation observed with Sendai virus grown in mouse fibroblast (L) cells. Virology 14, 486-488. Ito, M. (1973). Studies on the behavior of meiotic protoplasts: Induction of a high fusion frequency in protoplasts from liliaceous plants. Plant Cell. Physiol 14, 865-872. Kao, K . N . , and Michayluk, M. R. (1974). A method for high-frequency intergenic fusion of plant protoplasts. Planta 115, 355-367.

10. SENDAI VIRUS-MEDIATED CELL FUSION

333

Klenk. H. D.. and Choppin, P. W. (1970). Plasma membrane lipids and parainfluenza virus assembly. Virology 40, 939-947. Kim. J . , and Okada, Y. (1980). Morphological changes in Ehrlich ascites tumor cells during the cell fusion reaction with HVJ (Sendai virus). I.Alteration of cytoplasmic organelles and their reversion. Exp. Cell Rex. 130, 191-202. Kim. J.. and Okada. Y . (1981).Morphological changes in Ehrlich ascites tumopr cells during the cell fusion reaction with HVJ (Sendai virus). 11. Cluster formation of intramembrane particles in the early stage of cell fusion. Exp. Cell Res. 132. 125-136. Kim. J . . and Okada. Y . (1982). Morphological changes in Ehrlich ascites tumor cells during the cell fusion reaction with HVJ (Sendai virus). 111. Morphological characterization of HVJ glycoproteins integrated into the plasma membrane and their internalization by coated vesicles. Exp. Cell Res. 140, 127-136. Kim. J . , Miyake. K.. and Okada. Y. (1979). Transformation of intramembrane particles of HVJ (Sendai virus) envelopes from an invisible to visible form on aging of virions. Virology 95, 523-535. Kimura. H.. Uchida, T.. Futami. Y . . Yoshida. K.. Shinomiya, T.. Tarui, S . , and Okada. Y . ( 19x1). Evidence for guanylate cyclase activity associated with hemagglutinating virus of Japan (Sendai virus). J . B i d . Chem. 256, 2.508-2513. Kingsbury, D. W . (1974). The molecular biology of paramyxoviruses. M e d . Microbiol. Immrrnol 160, 73-83. Knutton. S . (1977). Studies of membrane fusion. 11. Fusion of human erythrocytes by Sendai virus. J . Cell Sci 28, 189-210. Lamb. R. A,. Mahy. R. W . J., and Choppin. P. W. (1976). The synthesis of Sendai virus polypeptides in infected cells. Virology 69, 116-131. Loyter. A , , and Volsky, D. J. (19x2). Reconstituted Sendai virus envelopes as carrier for the introduction of biological material into animal cells. Cell Surf: Reu. 8, 215-266. Lucy, J . A , . and Ahkong. Q. F. (1986). An osmotic model of the fusion of biological membranes. FEBS L e f f . 199, I - I I , McCrath. J . . and Solter, D. (1983). Nuclear transplantation in the mouse by microsurgery and cell fusion. Science 220, 1300-1307-. Maeda, T., Asano. A,. Ohki, K., Okada. Y.. and Ohnishi, S . (1975). A spin-label study on fusion of red blood cells induced by hemagglutinating virus of Japan. Biochemisfry 14, 3736-3741. Maeda. Y.. Kim. J.. Koseki, E.. Mekada, E., Shiokawa. Y.. and Okada. Y. (1977). Modification of cell membranes with viral envelopes during fusion of cells with HVJ (Sendai virus). Ill. Effects of mono- and disaccharides on cell fusion and membrane movement of fused cells. Exp. Cell Rex. 108, 95-106. Markwell, M . A . K.. Svennerholm, L.. and Paulson, J . C. (1981). Specific gangliosides function as host cell receptor for Sendai virus. Proc.. N u / / .A c d . Sci. U . S . A .78, 540654 10. Matsumoto, T.. and Maeno, K. (1962). A host-induced modification of hemagglutinating virus ofJapan (HVJ. Sendai virus) in its hemolytic and cytopathic activity. Virology 17, 563-570. Miura, N . , Uchida. T.. and Okada, Y . (1982). HVJ (Sendai virus)-induced envelope fusion and cell fusion are blocked by monoclonal anti-HN protein antibody that does not inhibit hemagglutination and neurarninidase activity of HVJ. Exp. Cell Res. 141, 409420. Miura, N., Ohtsuka, E.. Yamaberi, N., Ikehara. M.. Uchida, T.. and Okada. Y. (198%). Use of the deoxyinosine-containing probe to isolate and sequence cDNA encoding the fusion (F) glycoprotein of Sendai virus (HVJ). Gene 38, 271-274.

334

YOSHIO OKADA

Miura, N., Nakatani, Y., Ishiura, M., Uchida, T., and Okada. Y. (1985b). Molecular cloning of a full-length cDNA encoding the hemagglutinin neuraminidase glycoprotein of Sendai virus. FEES L e f t . 188, 112-1 16. Miyake, Y., Kim, J., and Okada, Y. (1978). Effects of cytochalasin D on fusion of cells by HVJ (Sendai virus). Cell-cell fusion is separable from cell-virus fusion. Exp. Cell R e s . 116, 167-178. Morgan, C., and Howe, C. J. (1968). Structure and development of viruses as observed in the electron microscope. IX. Entry of parainfluenza I (Sendai) virus. J . Virol. 2, 11221132. Nagai, Y . , Klenk, H. D., and Rott. R. (1976). Proteolytic cleavage of the viral glycoproteins and its significance for the virulence of Newcastle disease virus. Virology 72, 492-508. Nakanishi, M., Uchida, T., and Okada, Y. (1982). Glycoproteins of HVJ (Sendai virus) have critical ratio for fusion between virus envelopes and cell membranes. Exp. Cell R e s . 142, 95-101. Ohki, K., Nakama, S. , Asano, A., and Okada, Y. (1975). Stimulation of virus-induced fusion of Ehrlich ascites tumor cells by 3’,5’-cyclic AMP. Biochem. Biophys. R e s . Commun. 67, 331-337. Ohki, K., Asano, A.. and Okada. Y. (1978). Effects of cyclic AMP and calcium ions on virus-induced fusion of Ehrlich ascites tumor cells. Biochim. Biophys. A c w 507, 507516. Ohno-Shosaku, T., and Okada, Y. (1985). Electric pulse-induced fusion of mouse lymphoma cells: Roles of divalent cations and membrane lipid domain. J . Membr. B i d . 85, 269280. Okada, Y . (1958). The fusion of Ehrlich’s tumor cells caused by HVJ virus In uitro. Biken’.c J . 1, 103-110. Okada. Y. (1962a). Analysis of giant polynuclear cell formation caused by HVJ virus from Ehrlich’s acsites tumor cells: Microscopic observation of giant polynuclear cell formation. Exp. Cell R e s . 26, 98-107. Okada, Y. (1962b). Analysis of giant polynuclear cell formation caused by HVJ virus from Ehrlich’s ascites tumor cells: 111. Relationship between cell conditions and fusion reaction or cell degeneration reaction. Exp. Cell Res. 26, 119-128. Okada, Y. (1972). Fusion of cells by HVJ (Sendai virus). In “Membrane Research” ( F . Fox, ed.), pp. 371-382. Academic Press, New York. Okada, Y., and Murayama, F. (1966). Requirement of calcium ions for cell fusion reaction of animal cells by HVJ. Exp. CeIl Res. 44, 527-551. Okada, Y ., and Murayama, F. (1968). Fusion of cells by HVJ: Requirement of concentration of virus particles at the site of contact of two cells for fusion. Exp. Cell R e s . 52, 34-42. Okada, Y., and Tadokoro, J . (1962). Analysis of giant polynuclear cell formation caused by HVJ virus from Ehrlich’s ascites tumor cells: 11. Quantitative analysis of giant polynuclear cell formation. Exp. Cell R e s . 26, 108-118. Okada, Y., Suzuki, T., and Hosaka, Y . (1957). Interaction between influenza virus and Ehrlich’s tumor cells. 111. Fusion phenomenon of Ehrlich’s tumor cells by the action of , HVJ Z strain. M e d . J . Osaka Uniu. 7, 709-717. Okada, Y., Murayama, F., and Yamada, K. (1966). Requirement of energy for the cell fusion reaction of animal cells by HVJ. Virology 27, 115-130. Okada, Y., Koseki, I., Kim, J . , Maeda, Y., Hashimoto, T., Kanno, Y . , and Matsui, Y . (1975). Modification of cell membranes with viral envelopes during fusion of cells with HVJ. I. Interaction between cell membranes and virus in the early stage. Exp. Cell R e s . 93, 368-378. Ozawa, M., Asano, A., and Okada, Y . (1976). Importance of interpeptide disulfide bond in a

10. SENDAI VIRUS-MEDIATED CELL FUSION

335

viral glycoprotein with hemagglutination and neuraminidase activites. FEBS L e i / . 70, 14.5- 149. Ozawa. M.. Asano. A.. and Okada. Y . (1979a). The presence and cleavage of interpeptide disulfide bond in viral glycoprotein. J . l3iochc.m. 86, 1361-1369. Orawa. M . , Asano, A , , and Okada, Y . (1979b). Biological activities ofglycoproteins of HVJ (Sendai virus) studied by reconstitution of hybrid envelope and by concanavalin Amediated binding: A new function of HANA protein and structural requirement for F protein in hemolysis. Virology 99, 197-202. Paaternak, C. A.. and Micklem, J . J . (1973). Permeability changes during cell fusion. J . Mcmhr. B i d . 14, 293-303. Paterson. R. G.. Harris, T. J . R., and Lamb, R . A . (1984). Fusion protein of the parainyxovirus simian virus 5: Nucleotide sequence of mRNA predicts a highly hydrophobic glycoprotein. Proc.. Nail. Accrd. S c i . U . S . A . 81, 6706-6710. Perez. H.. Toister, Z., Laster. Y . , kind Loyter. A. (1974). Fusion of intact human erythrocytes and erythrocyte ghosts. J . Cell B i d . 63, 1-1 I. Pollard. H. B., Pazoles. C. J . . Creutz, C. E.. Scott, J . H.. Zinder, O., and Hotchkiss. A. ( 1984). An osmotic mechanism for exocytosi5 from dissociated chromaffin cells. J . B i d . Chern. 259, 1114-1121. Pontecorvo, G. (1975). Production of mammalian somatic cell hyhrids by means of polyethylene glycol treatment. Somutic. Cell G e n r t . 1, 397-400. Richardson. C. D.. Scheid, A,, and Choppin. A. (1980). Specific inhibition of paramyxovirus and myxovirus replication of oligopeptide with amino acid sequences similar to those at the N-terminal of the F, or HA? viral polypeptides. Virologv 105, 205-222. Robinson. J . M.. Roos. D. S . . Davidson, R. L.. and Karnovsky. M. J . (1979). Membrane alterations and other morphological features associated with polyethylene glycol-induced cell fusion. J . Cell Sc,i. 40, 63-75. Roos, D. S . . Robinson, J . M.. and Davidson. R. L. (1983). Cell fusion and intramembrane particle distribution in polyethylene glycol-resistant cells. J . Cell B i d . 97, 909-917. Scheid. A,, and Choppin, P. S . (1974). Identification of biological activities of paramyxovirus glycoproteins. Activation of cell fusion. hemolysis and infectivity by proteolytic cleavage of an inactive precursor protein of Sendai virus. Virology 57, 470-490. Scheid. A,, Caliguiri, L. A,, Compans, R. W.. and Choppin, P. W. (1972). Isolation of paramyxovirus glycoproteins. Association of both hemagglutinating and neuraminidase activities with the larger SV5 glycoprotein. Virology 50, 640-652. Sekiguchi. K.. and Asano. A. (1978). Participation of spectrin in Sendai virus-induced fusion of human erythrocyte ghosts. Proc. Nurl. Acud. Sci. U . S . A . 75, 1740-1744. Sekiguchi. K., Kuroda. K., Ohnishi, S . . and Asano, A. (1981). Virus-induced fusion of human erythrocyte ghosts. I. Effects of macromolecules on the final stages of the fusion reaction. Biochim. Biophys. Aciu 645, 21 1-225. Shioda. T., Iwasaki. K., and Shibuta. H. (1986). Determination of the complete nucleotide sequence of the Sendai virus genome RNA and the predicted amino acid sequence of the F, HN and L proteins. N w l e i c Acids Re.7. 14, 1545-1.563. Sugawa, H., Immamoto, N., Wataya-Kaneda, M., and Uchida. Y . (1985). Foreign protein can be carried into the nucleus of mammalian cells by conjugation with nucleoplasmin. Exp. Cell R e s . 159, 419-429. Tanaka. K., Sekiguchi, M.. and Okada, Y . (197.5). Restoration of ultraviolet-induced unscheduled DNA synthesis of xeroderma pigmentosun cells by the concomitant treatment with bacteriophage T4 endocnuclease V and HVJ (Sendai virus). Proc. N u t / . Acud. Sci. U . S . A . 72, 407 1-4075. Tozawa, H.. Watanabe. M.. and Ishida, N. (1973). Structural components of Sendai virus.

336

YOSHIO OKADA

Serological and physiochemical characterization of hemagglutinin subunit associated with neuraminidase activity. Virology 55, 242-253. Uchida, T., Kim, J., Yamaizumi, M., Miyake, Y., and Okada, Y. (1979). Reconstitution of lipid vesicles associated with HVJ (Sedai virus) spikes. Purification and some properties of vesicles containing nontoxic fragment A of diphtheria toxin. J . Cell B i d . 80, 10-20. Veomett, G., Prescott, D. M., Shay, J., and Porter, K. R. (1974). Reconstitution of mammalian cells from nuclear and cytoplasmic components separated by treatment with cytochalasin B. Proc. Nail. Acud. Sci. U . S . A . 71, 1999-2002. Volsky, D. J., and Loyter, A. J. (1978). Role of Ca?' in virus-induced membrane fusion. Ca2+ accumulation and ultrastructural changes induced by Sendai virus in chicken erythrocytes. J . Cell B i d . 78, 465-476. Vos, J., Ahkong, Q. F., Botham. G. M., Quirk, S . J.. and Lucy, J. A. (1976). Changes in the distribution of intramembrane particles in hen erythrocytes during cell fusion induced by the bivalent-cation ionophore A23 187. Biochem. J . 158, 651-653. Yamaizumi, M., Mekada, E . , Uchida, T., and Okada, Y. (1978). One molecule ofdiphtheria toxin fragment A introduced into a cell can kill the cell. Cell 15, 245-250. Yamaizumi, M., Uchida, T., and Okada, Y. (1979). Macromolecules can penetrate the host cell membrane during the early period of incubation with HVJ (Sendai virus). Virology 95, 218-221. Yanovsky, A., and Loyter, A. (1972). The mechanism of cell fusion. 1. Energy requirement for virus-induced fusion of Ehrlich ascites tumor cells. J . B i d . Chem. 247, 4021-4028. Zakai, N., Kulka, R. G., and Loyter, A. (1977). Membrane ultrastructural changes during calcium phosphate-induced fusion of human erythrocyte ghosts. Proc. Nuil. Acud. Sci. U . S . A . 74, 2417-2421. Zimmerberg, J., and Whitaker, M. (1985). Irreversible swelling of secretory granules during exocytosis caused by calcium. Nuiure (London) 315, 581-584.

CURRENT TOPICS I N MEMBRANES AND TRANSPORT, VOLUME 32

Chapter I I

Fusion Activity of the Hemagglutinin of Influenza Virus MARY-JANE GETHING,**tJEAN HENNEBERRY," A N D JOE SAMBROOK* * Drpartnzent of Biochemistry and f Hmwrd Hughes Medical Institute University of Texas Soirthwestern Medical Center Drillas, Texas 75235

I. 11

t

111. IV. V.

v1. VI1. VIlI.

Introduction Influenza Virus-Mediated Fusion: Role of the Hemagglutinin Assays for the Fusion Activity of HA Expression of HA in Cultured Cells from Cloned HA cDNAs Genetic Approaches to Studies of HA-Mediated Membrane Fusion A. Studies on Variant Influenza Viruses That Induce Fusion at Elevated pH B. Site-Directed Mutagenesis of the Fusion Peptide of HA C. Possible Involvement of Other Regions of HA in the Process of Fusion Characterization of the Low pH-Induced Conformational Change in HA Studies on the Cleavage Activation of HA Conclusion References

1.

INTRODUCTION

The ability of influenza virus to change its antigenic properties presents a major obstacle to controlling the disease by herd immunity or vaccination, so that influenza continues to be one of the major unconquered pathogens of man (Kendal and Patriarca, 1986). The major antigen of the 337 Copyright 0 I988 by Academic Press. Inc. All rights of reproduction in any form reserved.

338

MARY-JANE GETHING ET AL.

virus particle is the hemagglutinin (HA) glycoprotein which, like the less abundant neuraminidase (NA), is inserted into the lipid membrane that envelops the virion (Laver and Kilbourne, 1966; Drzenick et al., 1966). Influenza viruses with the potential to cause new pandemics or epidemics in an immune population carry antigenically novel hemagglutinins (Laver and Webster, 1979). Apart from its importance to man as a mutable antigen, HA plays the major role in the penetration of the host cell by the virus (Lazarowitz and Choppin, 1975; Klenk et al., 1975). HA is responsible not only for the initial attachment of the virus to receptors on the surface of the cell (Hirst, 1941) but also for the fusion of viral and cellular membranes that marks the onset of infection (White et ul., 1981; Huang et al., 1981). Because of its importance in all these processes, much work has focused on elucidating the molecular details underlying the structure, function, and biological activities of HA. Initially this work depended almost exclusively on the techniques of classical protein chemistry (Laver, 1963, 1964, 1973; Laver and Webster, 1968; Brand and Skehel, 1972; Skehel and Waterfield, 1975; Waterfield et al., 1979, 1980; Ward and Dopheide, 1980; Ward, 1981) and on the study of temperature-sensitive mutants of HA (Ueda and Kilbourne, 1976; Scholtissek and Bowles, 1975; Klenk et al., 1981). Since the advent of recombinant DNA technology, the genes encoding the HAS of many strains of influenza have been cloned and sequenced (reviewed in Lamb, 1983). Knowledge of the amino acid sequence of the HA from the X-31 influenza strain, together with X-ray diffraction studies of crystals of the protein led to elucidation of the three-dimensional structure of the ectodomain of the HA molecule (Wilson et al., 1981). Monoclonal antibodies raised against influenza viruses have allowed a detailed analysis of the antigenic structure of the protein (Wiley et al., 1981), and from studies of the recognition of HA and other viral polypeptides by cytotoxic T lymphocytes is emerging a complex picture of the cellular immune response to influenza virus (reviewed by Braciale et al., 1986; Mills, 1986). Recently, assays have been developed to monitor the state of folding and oligomerization of nascent and mature forms of HA (Bachi et al., 1985; Gething et ul., 1986b; Copeland et ul., 1986). The culmination of work from these and many other sources has been a detailed description of the physical domains of the molecule, the locations of its antigenic sites, the points at which it is glycosylated, its organization into trimeric structures, and its orientation with respect to the membrane. HA is therefore the best characterized of all eukaryotic membrane proteins, providing a superb model system for studies of the mechanism of low pH-mediated membrane fusion.

11. HEMAGGLUTININ OF INFLUENZA VIRUS

339

II. INFLUENZA VIRUS-MEDIATED FUSION: ROLE OF THE HEMAGGLUTI N IN

Influenza viruses, like many other enveloped viruses, enter and infect cells by a process involving binding to receptors on the cell surface, endocytosis of the virions, and low pHmediated fusion of the viral membrane with the membranes of acidic intracellular vesicles or endosomes (Matlin et al., 1981, 1982: Miller and Lenard, 1981: Marsh r t u l . , 1982; reviewed by White er d . , 1983). This pH-dependent fusion activity has been extensively studied in uitro using as targets cultured cells (White r t e l l . , 1980, 1981; Huang et cil., 1981), erythrocytes (Vaanfinen and Kfiariainen, 1980; Huang et [ I / . , 1981), and liposomes (White et d . , 1980, 1982a; Maeda er al., 1981). It has been shown that it is the viral surface glycoproteins that mediate the lipid bilayer fusion with specific pH dependences that are characteristic of each virus species and strain (White et a/., 1981, 1983). Early studies with influenza virus suggested that H A plays the key role in both the infectivity and the fusion activity of the virus (Lazarowitz and Choppin, 1975: Klenk et ( I / . , 1975: Maeda and Onishi, 1980; Huang et a/., 1981; Maeda c’t N / . . 1981). By expressing the protein in simian CV-1 cells infected with SV40-HA vectors, we have demonstrated that the HA molecule displays low pH-mediated fusion activity in the absence of any other influenza virus-encoded components (White rf al., 1982b). The HA molecule in its neutral form is a trimer that projects from the viral envelope as a rod-shaped structure 135 A in length (Wilson et [ I / . , 1981). To be active in fusion. the HA precursor, which is synthesized as a single polypeptide chain, must be processed by a posttranslational proteolytic cleavage into disulfide-bonded HA I and HA2 subunits (Laver, 1971: Klenk et a / . , 1975; Lazarowitz and Choppin, 1975; White et a / . , 1981, 1982b). A new hydrophobic amino terminus. the fusion peptide, is generated on the HA2 subunit. This peptide is highly conserved in HAS from different virus strains (reviewed by Lamb, 1983) and in many studies has been implicated as being intimately involved in the fusion process (Gething ef a/., 1978, 1986a; Richardson e f ul., 1980; Garten er al., 1981: White et a/., 1982b). In the native structure of HA, the hydrophobic fusion peptide in each monomer is tucked into the interface between the subunits of the trimer, approximately 30 A from the site of insertion of the polypeptide chain into the lipid bilayer of the virus envelope or the plasma membrane (Wilson et al., 1981). It has been proposed (Skehel et ul., 1982; White er al., 1982a; Daniels er al., 1983, 1985; Doms et al., 1985) that protonation of one or more amino acid side chains results in partial disso-

340

MARY-JANE GETHING ET AL.

ciation of the HA trimer and exposure of the fusion peptide which then inserts into the lipid bilayer. HA would then become an integral component of both the viral and target membranes, presumably bringing them close enough together to fuse. 111.

ASSAYS

FOR THE FUSION ACTIVITY OF HA

The fusion activity of HA can be manifested experimentally as cell-cell fusion (i.e., polykaryon formation) when monolayers of cells displaying cleavage-activated HA on their plasma membranes are transiently exposed to low pH. This activity can be measured following infection of cultured cells with influenza viruses (Maeda and Ohnishi, 1980; Huang et al., 1981; White et al., 1981) or following expression of recombinant HA from eukaryotic vector systems (see Figs. I and 3, below, and White et al., 1982b; Sambrook et al., 1985; Gething et al., 1986a). Such experiments have shown that HA-induced cell-cell fusion displays a characteristic pH profile, with the threshold pH varying between 5 and 6 depending on the viral origin of the HA molecule (Huang et al., 1981; White et al., 1983). In a second manifestation of the fusion activity of HA, molecules such as enzymes, antibodies, or oligonucleotides can be loaded into erythrocytes and delivered into cells that express cell surface HA by a process that involves binding of the loaded red cells via the hemagglutinating activity of HA, followed by HA-mediated fusion of the red cell and host cell membranes. Fusion can be measured quantitatively by delivery of horseradish peroxidase (HRP)followed by staining of the cell cytoplasm with diaminobenzidine (Figs. 1 and 3, below; Doxsey et al., 1985; Sambrook et al., 1985; Gething et al., 1986a). Finally, it is often desirable to follow the low pH-induced conformational change in HA that is necessary for initiation of the fusion reaction. This conformational change, which in wild-type HA closely parallels that of the fusion activity, has been monitored by following the acquisition of protease sensitivity by an ectodomain fragment of HA (BHA), released from the membrane by treatment with bromelain (Brand and Skehel, 1972), or by assessing the ability of this fragment to bind to lipid vesicles or detergents or to aggregate in lipid- or detergent-free solutions (Skehel et al., 1982; Doms et al.. 1985). As noted above, to undergo the low pH-induced confoFmational change or to be active in fusion, the HA precursor must have been cleaved into HA1 and HA2 subunits. During influenza virus infections, this cleavage is

11. HEMAGGLUTININ OF INFLUENZA VIRUS

341

performed by an endogenous protease in the epithelial cells of the respiratory tract. However, the HAS from human virus strains are not cleaved when expressed in the great majority of cultured cell lines. Thus, to perform fusion experiments it is necessary to effect an exogenous proteolytic cleavage of the cell surface HA molecules. This can be readily achieved by treating intact monolayers of HA-expressing cells with low levels of trypsin (White et al., 1982b; see Fig. 2).

IV.

EXPRESSION OF HA IN CULTURED CELLS FROM CLONED HA cDNAs

Key to our studies on HA-mediated fusion has been the ability to express wild-type and mutant forms of HA from a variety of eukaryotic expression vectors. We have developed a range of vector systems to express cDNAs encoding the HAS from the A/Japan/305/57 (H2 subtype) and A/Aichi/68 (H3 subtype) influenza virus strains. These vectors allow control of expression levels in a variety of cell types as well as choice between short-term, high level production in transient or lytic viral systems and constitutive expression in continuous cell lines. The expression systems that we have used most frequently for production and analysis of wild-type and mutant HAS are based on the doublestranded DNA virus, SV40. The characteristics of these transient or lytic viral vectors, from which HA cDNA is expressed in simian cells under the control of either the SV40 early or late promoters, have been reviewed in detail previously (Gething and Sambrook, 1981, 1983). Levels of expression of approximately lo8 molecules/cell/24 hr can routinely be achieved following infection of CV-I cells with high titer SV40-HA recombinant virus stocks. Figures I and 2 illustrate the techniques that are available to characterize the synthesis, cellular location, and functional activities of the wild-type Japan HA protein expressed in CV-1 cells. The protein, which can be shown using indirect immunofluorescence to be located at the cell surface as well as in the endoplasmic reticulum and Golgi regions of the cell, is active in hemagglutination, cell-cell fusion and red blood cell-mediated delivery. Although the high levels of HA expression obtained using the lytic SV40 viral vectors are very useful for the initial characterization of HA mutants, the ability to generate cell lines that continuously express wildtype or mutant forms of HA provides significant advantages for longer term studies and for experiments designed to investigate the function of

342

MARYJANE GETHING ET AL.

FIG.1 . Analysis of HA expressed from SV40-HA recombinant vectors. Forty-eight hours after infection of CV-I cells with the SVEHA3 recombinant vector (Gething and Sambrook, 198I), the following assays were performed: immunofluorescence on permeabilized cells (A), hemagglutination (B), cell-cell fusion (C), and erythrocyte-cell fusion meaThe details of the experisured by delivery of HRP and staining with diaminobenzidine (D). mental protocols have been described previously (Gething and Sambrook 1981, 1982, 1986a; Doyle et a/., 1985).

cellular components involved in the transport and maturation of HA. We have therefore constructed vectors based on bovine papilloma virus (BPV) and used them to develop lines of cells that constitutively express HA at levels suitable for biochemical, immunological, and functional analysis. In initial experiments, BPV-transformed cell lines expressing HA were identified by the labor-intensive method of screening

343

11. HEMAGGLUTININ OF INFLUENZA VIRUS

v .C

c-‘ +

~

68K-

d

z

+

-a

f f a cn m

92K-

3 0)

I=

-a Ln In 4

-k

L

v ._ C

a

0

Ln

L

a a I I > a

2

m

L

cu

0

2

0 L

L

(u

cu

a r

5

> a m

I

d h 0 L

1=

cu

h

L

$

Q,

+

I=

+

d

-k

0 L

L

cu

0 L

r

cu

3

2

+

0 L -c

cu

a I > a m

r

7HA0 -HA

non-glyc

46K -

~ H A I

3 O K - w

3H A 2

Fic. 2. Comparison of HA synthesized from SV40-HA and BPV-HA vectors. The figure shows an autoradiograph of an SDS-PAGE separation of HA polypeptides immunoprecipitated from [’sS]methionine-labeled extracts of cells producing HA from SVEHA3 or pBVI-MTHA vectors (from Sambrook P I u / . , 1985). Cells were pulsed for 15 min with 1”SImethionhe in either the presence or absence of tunicamycin ( 5 pg/ml). Following a 2-hr chase period in an excess of nonradioactive methionhe. the cell monolayers were mocktreated or treated with trypsin ( 5 pg/ml) or neuraminidase (0.5 unitsiml) for IS rnin before cell extracts were prepared. immunoprecipitated with a high titer rabbit anti-HA serum. and separated by SDS-PAGE.

large numbers of individual clones by radioimmune assay (Sambrook et d.,1985). This was feasible for cell lines such as C127 and NlH-3T3 that were efficiently transfected by the BPV-HA vector but did not prove successful with other cell types. However, use of a fluorescence-activated cell sorter to identify cells that bind FITC-conjugated anti-HA antibodies (or hemagglutinate FITC-labeled red blood cells) has facilitated the rapid

344

MARY-JANE GETHING ET AL.

selection of rare HA-expressing cells and has also allowed us to clone out those cells that display the greatest amounts of HA at the cell surface. The Japan or Aichi (X-31) HA genes have now been introduced into cell lines derived from a number of different species including mouse (C127, NIH-3T3, and MME cells), dog (MDCK cells), pig (PK cells), and hamster (CHO cells). The level of expression of HA varies between cell types: up to 10’ molecules of HA per cell can be obtained in C127, NIH-3T3, or CHO cell lines, while somewhat lower levels (104-106 molecules per cell) are obtained in the other, more differentiated cell types. Obviously, the development of an increasing variety of HA-transformed cell lines of different types and species coupled with the ability to deliver efficiently into the cytoplasm probes such as antibodies or oligonucleotides provides great opportunities for studies of intracellular events and architecture in living cells. Figures 2 and 3 characterize the synthesis, cellular location, and functional activities of the wild-type Japan HA protein expressed in a continuous line of NIH-3T3 cells transformed with the BPV-HA vector, pBVIMTHA (Sambrook et al., 1985). Comparison of HAS produced from SV40-HA and BPV-HA vectors in CV-1 cells and NIH-3T3 cells, respectively (Fig. 21, indicates that the nonglycosylated forms of HA synthesized in each cell type in the presence of the drug tunicamycin are identical in size, as are the core-glycosylated forms of the proteins synthesized during a 15-min pulse. However, variation between these cell types in the trimming and modification of the oligosaccharide side chains results in differences in the terminal glycosylation patterns of the HA molecules, and thus in differences in their mobilities on SDS-polyacrylamide gels. Figure 2 also illustrates the quantitative cleavage by exogenous trypsin of terminally glycosylated precursor HA0 to HA1 and HA2 subunits. Figure 3 indicates the cell surface location of the HA synthesized in NIH-3T3 cells and demonstrates that the protein is active in hemagglutination, cellcell fusion, and red blood cell-mediated delivery assays. The availability of cell lines that express HA at different levels has allowed us to estimate the level of expression required on a per cell basis for cell-cell fusion to occur. Table I compares, for six different transformed NIH-3T3 cell lines, the number of HA molecules per cell and the cell-cell fusion and hemagglutination activities. The results indicate that a minimum expression level of approximately 3 x 106molecules per cell is necessary for the manifestation of fusion activity. A similar result was obtained when red blood cell-mediated delivery experiments were performed (results not shown).

11. HEMAGGLUTININ OF INFLUENZA VIRUS

345

FIG.3. Analysis of HA expressed from BPV-HA recombinant vectors. A continuous NIH-3T3 cell line transformed with the pBVI-MTHA vector (Sambrook ef n l . , 1985) was characterized using the following assays: immunofluorescence on nonpermeabilized cells (A), hemagglutination (B), cell-cell fusion (C), and erythrocyte-cell fusion measured by delivery of HRP and staining with diaminobenzidine (D). The details of the experimental protocols have been described previously (Gething and Sambrook, 1981. 1982: Sambrook et al., 1985; Doxsey et al., 1985).

V.

GENETIC APPROACHES TO STUDIES OF HA-MEDIATED MEMBRANE FUSION

Two genetic approaches have been employed to analyze the mechanism of the fusion reaction mediated by HA. The first involves studies of variant influenza viruses that induce fusion with raised pH thresholds. Sequence analyses of HAs from the variant viruses have identified altered

346

MARY-JANE GETHING ET AL.

TABLE 1 CORRELATION OF THE LEVELOF EXPRESSION OF HA IN MURINE BVi-MTHA CELLLINESWITH THE CAPACITY TO UNDERGO LOW PH-MEDIATED CELL-CELLFUSION Cell line

TRI-I TRI-4 TR 1-5 NTRI-4 NTRI-5 NTRI-I I

HA rnolecules/cell" 6 I 1 3 I 8

x x x x x

lo0 I06 106

I@ 106

x 106

Fusion activityh

Hemagglutination'

+++

++t ++t

+ + +++ -

+++

+++ +++ +++ +++

" Cell extacts were prepared from a known number of cells and assayed by solid-phase radioimmunoassay (Cething and Sambrook. 1981), Ir Fusion activity was measured by polykaryon formation following treatment of cell monolayers with trypsin and neurdminidase (Sambrook P I a / . . 1985) and repeated ( 2 x ) brief exposure to buffer at pH 5.0 (PBS containing 10 mM HEPES). Symbols: -, no polykaryons observed; +. small numbers of polykaryons containing 4-5 nuclei: + + +. massive formation of polykaryons containing large numbers of nuclei (see. for example. Fig. 3 0 . Hemagglutination assays were performed after treatment of cell monolayers with trypsin and neuraminidase (Sambrook el d.,1985) using a IW solution of guinea pig erythrocytes. ++ +. Erythrocytes bound to SO-IOO%, of cells in the monolayer (see, for example. Fig. 3 B ) . 1

amino acids that play a role in the pH dependence of fusion. The second approach ulitizes oligonucleotide-directed, site-specific mutagenesis of a cloned HA gene to alter the nucleotide sequence encoding selected amino acids in the HA molecule. Expression of the mutant genes in simian cells has confirmed the central role of the fusion peptide and provided insights into the mechanism of the fusion reaction. A. Studies on Variant Influenza Vlruses That Induce Fusion at Elevated pH

To gain insight into the molecular mechanism underlying the pH dependence of HA-mediated fusion, influenza virus variants have been isolated that induce fusion at threshold pH values higher than those of their parent viruses. Rott et al. (1984) showed that variants of the X-31 strain, selected for their ability to undergo activation cleavage and grow in MDCK cells, also displayed an elevated threshold pH for fusion that was approximately 0.7 pH units higher than the wild type. Variants of the X-31 and Weybridge virus strains have been selected for by growth in the presence of amantadine, a compound that raises endosomal pH (Daniels et af., 1985). Variant viruses were obtained that mediated fusion at pH values 0.1-0.7

1 1 . HEMAGGLUTININ OF INFLUENZA VIRUS

347

unit higher than the parental strain. Finally, we have isolated and analyzed a naturally occurring variant ofthe X-31 strain whose pH threshold for fusion was elevated by 0.2-0.3 units (Doms P t al., 1985). Sequence analysis of the HAS from these variant viruses has identified individual amino acids that appear to play a role in the low pHmediated conformational change in the molecule. Some of these residues are located along the interface between the subunits of the trimer. while others stabilize the unexposed location of the fusion peptide at the amino terminus of HA2 (see Fig. 4A). It has been proposed that these amino acids participate in interactions that stabilize the structure of the wild-type molecule at neutral pH (Daniels et d., 1983. Substitution of these residues would lower the energy barrier necessary for the low pH-induced conformational transition to the fusion-active state. All the results are consistent with the widely held theory that dissociation of the HA trimer at low pH is a necessary and early step in the fusion mechanism. 6. Site-Directed Mutagenesis of the Fusion Peptide of HA

Although the majority of the amino acids that were altered in the fusion variants were located along the trimer interface (Fig. 4A), three altered residues were located within the hydrophobic fusion peptide at the amino terminus of the HA2 subunit. The fact that these three residues have undergone conservative substitutions probably reflects the fact that mutant viruses unable to mediate the fusion reaction at any pH could not enter and infect cells and therefore could not be propagated. To obtain mutant HAS that might be inactive or disabled for fusion and to probe the consequences of altering the length and hydrophobicity of the fusion peptide. we have used oligonucleotide-directed mutagenesis of the Japan HA cDNA to introduce single base changes into the sequence that encodes this peptide (Gething ef ul., 1986a). Three mutants were constructed that introduce single, nonconservative amino acid changes in the fusion peptide (Fig. 4B). When the mutants were assayed for fusion activities and for the low pHmediated conformational change and acquisition of lipid binding capacity (Table 11), three fusion phenotypes were observed: 1 . Substitution of glutamic acid for the glycine residue at the amino terminus of HA2 abolished all fusion activity although the mutant HA could still undergo a conformational change (at lower pH than the wild type) that resulted in protease sensitivity and lipid binding capability. Study of this mutant has provided the first indication that the conformational change can be temporally separated from lipid binding, and also

FIG.4. Schematic diagrams of the ectodomain of the HA monomer showing the location of amino acids that have been altered in the fusion variants (A) or the site-directed mutants (B). The schematic diagram of the three-dimensional structure of the HA monomer is taken from Wilson et a / . (1981). The data summarized in A are drawn from Rott er al. (1984),

348

- GLYl

-

GLY4

GLU

-

GLU

1- GLY

Daniels et a / . (1989, and Doms et u / . (1986), while that in B is from Gething ct u / . (1986a). In A the residues marked with an asterisk are from the HA1 subunit; all other residues are from the HA2 subunit.

349

350

MARY-JANE GETHING ET AL.

TABLE I1 FUSION PHENOTYPES OF T H E WILD-TYPE A N D MUTANT HA

SUMMARY OF THE ~~

~

~

Threshold pH Efficiency at pH 5 of erythrocyteof cellcell fusion cell fusion (%)

Threshold pH of erythrocytecell fusion

WT

PROTEINS"

~

Efficiency at pH 5 of cell-cell fusion (%)

pH at which 50% of BHA is converted to protease sensitivity and lipid binding

5.3

5.3

90

90

5.3

MI

-

-

M4 MI1

5.7 5.3

5.6 4.6

0 50

0 50

5.0 5.5

90

0

5.3

~~

~

This table is adapted from and summarizes results described in Cething

PI

ul. (1986a).

allows distinction of the stages of lipid binding and bilayer fusion, indicating that HA does more than simply bring the two membranes close together. 2. Substitution of glutamic acid for the glycine residue at position 4 in HA2, which decreased the length of the apolar stretch to 6 amino acids, raised the threshold pH both for the conformational change and for fusion and also reduced the efficiency of fusion. It appears that the mutation has destabilized the neutral conformation of the HA trimer in a similar fashion to the amino acid alterations identified in the variant viruses described above. The results suggest that the amino acid at position 4 in HA2 may play two roles: a structural role in maintaining the fusion peptide in its neutral conformation and another role in the stage of bilayer destabilization. 3 . Extension of the hydrophobic stretch by replacement of the glutamic acid at position 11 with glycine yielded a mutant protein that underwent the conformational change and induced fusion of erythrocytes with cells with the same efficiency and pH profile as the wild-type protein. However, the ability of this mutant to induce polykaryon formation was greatly impaired. This phenotype provides the evidence for a distinction between cell-cell fusion and erythrocyte-cell fusion. The mutant HA is competent to mediate fusion of the cell and erythrocyte membranes over the small area necessary for injection of HRP into the cytoplasm, but, except at very low pH, the mutant HA is unable to induce bilayer destabilization over areas sufficient to cause polykaryon formation. Analysis of these mutants has allowed us to delineate several stages of the mechanism of HA-mediated membrane fusion which were not separated in previous studies. These include (1) the low pH-induced conformational change in HA that exposes the fusion peptide; (2) the interaction

11. HEMAGGLUTININ OF INFLUENZA VIRUS

351

of the fusion peptide and/or other regions of the HA molecule with the target lipid bilayer; ( 3 ) destabilization of the lipid bilayer and membrane coalescence over small areas; and (4) membrane coalescence and separation over large areas, resulting in polykaryon formation. The amino terminus of the HA2 subunit is the most highly conserved region of HA (Gething P I a / . . 1980; Kawaoka rt a / . , 19841, indicating that this region of the molecule must be of great importance for the structure and function of HA and the life cycle of the virus. Nevertheless, although the HAS of field strains never show amino acid changes in the fusion peptide, it is possible, either by selection of variant viruses by growth in amantadine or by in vitro mutagenesis, to introduce alterations into this peptide that do not inhibit the assembly, intracellular transport, or fusion activity of this protein. Presumably variant viruses carrying these altered HA molecules would be at a disadvantage in a competition with field strains, either because occurrence of the conformational change or the fusion reaction at a higher pH is not desirable or because the mutant HA proteins are more unstable or thermolabile. Whatever their defects would be in the real world, HA mutants generated in uitro provide the opportunity to study the role of this hydrophobic peptide in the fusion reaction. C. Possible Involvement of Other Regions of HA in the Process of Fusion

Although these studies have confirmed a function for the amino terminus of HA2 in the fusion mechanism, they have not ruled out a role for other regions of the molecule in the interaction with the target lipid bilayer. Very promising candidates for other structures that may be involved in fusion are the two amphipathic helices of the stalk domain, which are made up of sequences from the HA2 subunit (Wilson el ul., 1981). In the neutral trimer, hydrophobic amino acids which lie along one face of the top half of the long helix form extensive contacts between the individual subunits. Following the low pH-induced conformational change when the subunits ofthe trimer separate, the helices will no longer interact with each other and may well be available to interact instead with the lipid bilayer. The short helix is also amphipathic. Physiologically important interactions of amphipathic surfaces (Q helices or p sheets) with membranes have been described (Kaiser and Kezdy, 1984) with the amphilicity of the structure being more conserved than specific amino acid sequence. Design and analysis of mutations in this region of HA will be quite complex, since it is clear that the long helix in HA2 is intimately involved in the folding and final structure of the trimeric molecule (Wilson et ul.,

352

MARY-JANE GETHING ET AL.

1981). It will be necessary to design mutations that do not affect the folding and transport of the protein but that might interfere with the interaction with the target bilayer following the low pH-induced conformational change. Candidates for oligonucleotide-directed mutagenesis would include residues with aliphatic or aromatic side chains on the hydrophobic face of the helix, but which do not appear to be so intimately involved in the interactions that connect the subunits in the trimer. Conservative substitutions would test the role of individual amino acids in the fusion mechanism, while nonconservative substitutions would test the importance of amphilicity rather than specific sequence. Alteration of the pH threshold for fusion would indicate that the substituted amino acid plays a structural role in the conformational change, while alteration of the efficiency of fusion would suggest that the amphipathic helices might indeed interact with the target lipid bilayer during the fusion reaction. VI.

CHARACTERIZATION OF THE LOW-pH INDUCED CONFORMATIONAL CHANGE IN HA

A current view of the low pH-induced conformational change in HA is that protonation of one or more amino acid side chains in the subunit interface causes the dissociation of the HA trimer. Because the quaternary interactions between the three globular domains are few in number in the neutral conformation of the trimer (Wilson et al., 1981), dissociation of the subunits may simply occur by separation of the globular domains at the top of the molecule. This would then be followed by breaking of the interactions between the amphipathic helices in the stalk which leads to exposure of the previously buried fusion peptide. Recent experiments, however, have suggested that a more complicated series of conformational changes may occur. White and Wilson (1987) have used antipeptide antibodies, directed against individual sequences located near the fusion peptide and along the subunit interface in both the globular and stalk domains, to probe the temporal sequence of separate conformational alterations that take place following treatment at low pH. Surprisingly, their results indicated that the conformational change begins with exposure of a loop (HA1 residues 14-52) which in the neutral structure is embedded in the trimer interface, followed by exposure of the fusion peptide and then the carboxy terminus of HAl. The separation of the subunits, and in particular of the globular domains, appears to be a relatively late event in the process. Thus, although these results must be interpreted with the caveat that the experiments were carried out using HA solubilized in

11. HEMAGGLUTININ OF INFLUENZA VIRUS

353

detergent, it appears that conformational changes around the stem of the molecule may precede dissociation of the subunits. Neither scenario is contradicted by the data from the fusion variant studies which identified amino acids important in stabilizing the neutral positions of both the HA1 carboxy terminus and the fusion peptide, as well as the intersubunit connections, but could not define any order of events in the conformational change. Whichever model is the more correct, we still need to understand better the events that lead to disruption and reestablishment of the lipid bilayer. It is to be hoped that elucidation of the role of individual amino acids of HA in the fusion reaction might lead to a physiochemical model for the rearrangements that occur within the fusing membranes.

VII. STUDIES ON THE CLEAVAGE ACTIVATION OF HA

As described above, the fusion activity of HA and thus the infectivity of influenza virus requires that the HA0 precursor be processed by a posttranslational proteolytic cleavage to an active form of the molecule. The HAS from human type A influenza viruses contain a single arginine residue at the processing site, and activation involves cleavage on the carboxyl side of this amino acid by a trypsinlike protease, followed by remove1 of the arginine by a carboxypeptidase activity (reviewed by Rott and Klenk, 1986). Cellular proteases that carry out these cleavages are expressed only in primary epithelial cells, a situation which normally restricts viral replication to the respiratory epithelium during influenza virus infections of humans. Virulent avian influenza viruses encode HAS that have several arginine and lysine residues at the processing site, so that the HA is cleaved in most cell types (Rott and Klenk, 1986), leading to systemic infections. The pathogenicity of these viruses tends to correlate with the numbers of basic amino acids at the cleavage site of HA, although other structural features of HA such as the positioning of carbohydrate side chains can modulate the virulence of the infection (Kawaoka et al.. 1984). In addition, pathogenicity has proven to be a multigenic trait (Schulman, 1983). In cells in culture, HAS derived from human viruses are not cleaved so that it is the precursor HA molecule that is displayed at the cell surface; treatment with exogenously added trypsin is required to activate the protein for fusion. HAS derived from pathogenic avian viruses are activated by cellular proteases in cultured cells; it is as yet uncertain whether cleavage takes place before or after arrival of the protein at the cell surface.

354

MARY-JANE GETHING ET AL.

To study the structural requirements for activation cleavage of HA we have used oligonucleotide-directedmutagenesis to alter the amino acid sequence at the processing site of X-31 HA. A single nucleotide change, AGA- ACA, wasintroduced togenerate the cleavage-minus mutant, C-T, in which the arginine residue at the cleavage site is substituted by a threonine residue. Loop-in mutagenesis of 3 or 15 nucleotides was performed to generate the other two mutants. The first, C+RR, contains an additional arginine residue at the cleavage site. The second, C+RRKKR, contains two extra arginines and two extra lysine residues at the cleavage site (reproducing the cleavage site in the HA from the A/Turkey/Ontario/ 66 strain of influenza virus). Figure 5 shows the results of DNA sequence analysis to confirm the nucleotide sequences encoding the region around the cleavage sites in the wild-type and mutant HA cDNAs. The nucleic acid and corresponding amino acid sequences of the relevant segment of the wild-type and mutant proteins are shown in Fig. 5B.Following confirmation of the desired mutant sequences, the mutated X-31 cDNAs were used to replace the corresponding wild-type fragments in the SVEXHA recombinant viral genome (Doyle ef d.,1986).The recombinant genomes were transfected into CV-I cells, and high titer virus stocks were developed (Doyle et al., 1985) and used to infect fresh monolayers of CV-I cells for analysis of the biosynthesis, intracellular transport, and functional activities of the wild-type and mutant HA proteins. An analysis of the transport and proteolytic cleavage of these proteins is shown in Fig. 6. Fifteen minutes after synthesis, the wild-type HA is present predominately as the core-glycosylated HA0 species which appears as a sharper, faster migrating band. Following a 2-hr chase period, most of the protein has been converted to the slightly slower migrating, terminally glycosylated form of the precursor and has been transported to the cell surface where it is available for cleavage by exogenously added trypsin. The C-T protein is also quantitatively converted to the terminally glycosylated form during the 2-hr chase period. However, although this protein can be demonstrated to be at the cell surface by immunofluorescence and hemagglutination assays (results not shown), the mutant protein cannot be cleaved by exogenously added trypsin. The C+RR mutant is indistinguishable from the wild-type protein in this experiment, demonstrating that the presence of an additional arginine residue at the cleavage site does not result in endogenous cleavage of HA in CV-I cells. Finally, the C+RRKKR mutant is cleaved into HA1 and HA2 subunits in the absence of any exogenous protease. The addition of trypsin to the medium above the cells does not result in any further cleavage of the remaining (core-glycosylated) precursor protein. This result demonstrates that the presence of multiple arginine and lysine residues (five total) at the

11. HEMAGGLUTININ OF INFLUENZA VIRUS

355

cleavage site facilitates the cleavage of the HA precursor by a protease endogenous to CV-I cells. The data shown in Fig. 6 do not provide evidence that this mutant protein reaches the cell surface; however, immunofluorescence and erythrocyte binding experiments reveal the presence of this protein at the plasma membrane. Each mutant protein was then tested for a requirement for in uirro proteolytic activation for mediation of low pH-induced cell-cell fusion. Figure 7 shows that (as previously reported; Gething c't a / . , 1986a) cells expressing the wild-type HA protein can be fused by treatment at low pH only after the HA0 precursor at the cell surface has been treated with exogenous trypsin, thereby cleaving the molecule into disulfide-bonded HA1 and HA2 subunits. However, the C T mutant cannot mediate low pH-induced fusion even after treatment with trypsin, presumably because alteration of the cleavage site prevents cleavage activation of the molecule. The phenotype of the C'RR mutant is identical to that of the wildtype protein, requiring pretreatment with trypsin to activate the molecule for fusion. The C+RRKKR mutant is active in low pH-mediated cell fusion without prior treatment with exogenous trypsin because the HA molecules at the cell surface have already been activated for fusion via cleavage by an endogenous protease. In summary, a mutant in which the single arginine residue at the cleavage site was altered to threonine was not cleaved under any circumstances, even after treatment with exogenous protease. This cleavageminus mutant was transported normally to the cell surface and displayed hemagglutinating activity although it could not mediate low pH-induced fusion. The second mutant contained two arginine residues at the processing site; this protein was indistinguishable from the wild-type HA in its transport and biological activities and was cleaved only after treatment with external trypsin. The final mutant reproduced the processing site of the HA from a pathogenic avian virus, i.e., . . . Arg.Arg.Lys.Lys.Arg. . . . This mutant HA was cleaved by a cellular protease in CV- I cells and mediated low pH-induced fusion without prior treatment with trypsin. Further experiments are planned to probe the precise structural requirements for cleavage activation. In addition to testing whether endogenous processing occurs when the site contains three or four basic amino acids, it will be of interest to probe any differences betwen arginine and lysine as protease substrates and to study the role of the conserved nonpolar residues that both precede and follow the processing site. The processing site mutants that are already available will be useful for a number of purposes. First, the cleavage-minus mutant will provide an opportunity to produce precursor HA0 for crystallographic studies. A significant, although possibly localized, conformational change must take

WT

C'T

WT

C+RR

C+RRKKR

FIG. 5. Nucleic acid and corresponding amino acid sequences around the cleavage activation sites of wild-type and mutant HA molecules. The Clal-BarnHI restriction fragment that encompasses the entire coding sequence of X-31 HA was cloned into an M13 mp18 vector. Single-stranded phage DNA was used as the template for oligonucleotide-directed mutagenesis to generate mutants that encode altered sequences at the activation cleavage site of HA. The autoradiographs (A) show the results of DNA sequence analysis to confirm the nucleotide sequences encoding the region around the cleavage sites in the wild-type and mutant HA cDNAs. The arrowheads and dashed lines show the positions and extent of nucleotide changes or insertions. The amino acids corresponding to each codon are shown in single letter code: R, Arg; K, Lys. (B) The nucleic acid and corresponding amino acid sequences of the relevant segment of the wild-type and mutant proteins. The vertical arrows show potential cleavage sites.

357

11. HEMAGGLUTININ OF INFLUENZA VIRUS

B

1HA2

XHA-WT

..Asn.Val.Pro.Glu.Lys.Gln.Thr Arg.Gly.Leu.Phe.Gly. AAT GTA CCA GAG AAA CAA ACT AGA GGC CTA TTC GGC

XHA-C-T

..Asn.Val.Pro.Glu.Lys.Gln.Thr

XHA-C'RR

,

XHA-C'RRKKR

. . A m .Val .Pro.Glu.Lys.Gln.Thr.Arg.Arg. 1 1Lys. Lys.Arg.Gly.Leu.Phe.Gly.. I IHA2

ihr.Gly.Leu.Phe.Gly. ACA

1

.Asn.Val .Pro.Glu. Lys.Gln.Thr.Arg.Arg. /HA* .Gly.Leu.Phe.Gly.. AGG AGA

I

AGG AGA AAG AAG AGA

FIG.5 .

(Con~in~ed)

place on cleavage activation since the carboxy terminus of the HA1 chain and the amino terminus of the HA2 chain, which are separated by only one amino acid in the precursor polypeptide, end up separated by a gap corresponding to approximately 10 residues in the cleaved molecule (Wilson PI a/.. 1981). The only satisfactory way to analyze this conformational change is by comparison of the known three-dimensional structure of the cleaved molecule with that of the precursor form. Uncleaved HA produced in tissue culture systems has not been a suitable source of precursor because it undergoes activation cleavage during harvesting from the cell surface with bromelain. Furthermore, the anchor-minus precursor HA that is secreted in large amounts from CV-1 cells that have been infected with the SVEHA20-A- recombinant virus (Gething and Sambrook, 1982) also is not useful for structural studies. Although the anchor-minus HA is initially assembled into trimeric structures, the oligomers are not stable and fall apart into monomers over time. By contrast, the solublized trimers of wild-type HA (BHA) that can be released from the virion or cell surface with bromelain (Brand and Skehel, 1972) are very stable. Preliminary studies indicate that soluble BHAO trimers can be prepared from the cleavage-minus variant of X-3 1 HA. A second application involves the C+RRKKR mutant which undergoes endogenous activation cleavage in cultured cells. We have constructed a composite mutant X-31 HA cDNA that encodes a protein containing both the cleavage site insertion and the substitution of asparagine for aspartic acid at residue 132 (Dams et al., 1985), which results in a higher pH threshold for fusion activity. When this cDNA was expressed in CV-I

358

MARY-JANE GETHING ET AL.

FIG.6. Analysis of the transport and proteolytic cleavage of wild-type and mutant HA proteins. CV-1 cells infected for 36 hr with SV40-HA viruses encoding the wild-type or mutant HA proteins were labeled for I5 min with [Y5]methionine. Cell extracts were prepared immediately (A) or following further incubation for 2 hr in medium containing an excess of nonradioactive methionine (B,C). In C trypsin at a concentration of 10 j&ml was included in the medium above the cells for the final I5 min of the 2-hr period. Proteins were precipitated from the cell extracts with anti-HA serum, separated by SDS-PAGE, and autoradiographed. The positions of the precursor H A 0 molecules and of the HA1 and HA2 cleavage products are shown to the right.

cells from an SV40-HA vector, the expected phenotype was obtained of fusion activity at pH 6.0 in the absence of prior treatment with trypsin. The double-mutant cDNA has also been inserted into a BPV-based vector, and continous cell lines (NIH-3T3and CHO) that express the variant HA are being selected by fluorescence-activated cell sorting. Because these cell lines will undergo cell-cell or erythrocyte-cell fusion under very mild and simple conditions (only moderately low pH and no protease treatment), they should provide excellent experimental systems for studies of fusion or as universal recipients for any materials of biological interest (antibodies, enzymes, oligonucleotides, etc.) that can be loaded into erythrocyte ghosts (Doxsey et al., 1985).

FIG.7. Low-pH-induced cell-cell fusion of CV-I cells expressing wild-type or mutant HA proteins: requirement for proteolytic cleavage activation. CV-I cells were infected with SV40-HA vectors containing the wild-type or mutant HA genes. Fifty hours after infection. the medium was aspirated from the monolayers which were then washed 3 times with medium lacking serum. Following incubation at 37°C for IS min with medium alone (top row) or with medium containing trypsin (10 &ml) (bottom row). the cells were briefly treated with buffer at pH 5.3 and then incubated further in medium containing serum for 8 hr to permit visualization of the formation of polykaryons with clustered nuclei. The cell monolayers were then fixed with formaldehyde, stained with Giemsa. and photographed.

360

MARY-JANE GETHING ET AL.

VIII.

CONCLUSION

The studies described in this chapter illustrate how the use of recombinant DNA techniques can advance our knowledge of the mechanism of membrane fusion. Cloning of cDNA copies of genes encoding wild-type and variant fusogenic proteins provides information on their primary amino acid sequences. Site-directed mutagenesis of the cloned genes can be used to probe the importance of chosen domains of a protein in the fusion mechanism. Expression of wild-type and mutant genes in mammalian cells using various types of eukaryotic vectors facilitates the analysis of their fusion phenotypes and in addition provides cell lines that can be used for cell fusion experiments or for delivery of macromolecules into the cell cytoplasm. Future experiments that combine the power of recombinant DNA techniques with the elegance of the model system provided by influenza hemagglutinin should reveal further details of the molecular mechanism of the fusion reaction induced by this particular membrane protein. The availability of cloned copies of a number of other viral fusion proteins (for references, see Gething et al., 1986a) will now permit similar analyses in other systems. Such studies should reveal any common or unique features of the mechanisms by which these different fusogenic proteins mediate membrane fusion. ACKNOWLEDGMENTS This work was supported by grants from the National Institutes of Health to M.J.G. and J.S. REFERENCES Bachi, T.,Gerhard, W., and Yewdell, J. W. (1985). Monoclonal antibodies detect different forms of influenza virus during viral penetration and biosynthesis. J. Virol. 55,307-313. Braciale, T. J., Lukacher, A. E., Morrison, L., Braciale, V. J., Smith, G., Moss, B., Gething, M. J., and Sambrook, J. (1986). Influenza viral antigen recognition by Class I and Class 11 MHC restricted cytolytic T lymphocytes. In "Options for the Control of Influenza Virus" (P. A. Kendal and P. A. Patriarca, eds.), pp. 407-421. Liss, New York. Brand, C. M., and Skehel, J. 3. (1972). Crystalline antigen from the influenza virus envelope. Nature (London) New Biol. 238, 145-147. Copeland, C., Doms, R. W., Bolzau. E. M., Webster, R. G., and Helenius, A, (1986). Assembly of influenza hemagglutinin trimers and its role in intracellular transport. J. Cell Biol. 103, 1179-1 191. Daniels, R. S., Downie, J. C., Hay, A. J., Knossow, M., Skehel, J. I., Wang, M. L., and Wiiey. D. C. (1985). Fusion mutants of the influenza virus hemagglutinin glycoprotein. Cell 40,431-439. Doms, R. W., Helenius, A., and White, J. M. (1985). Membrane fusion activity of the

11. HEMAGGLUTININ OF INFLUENZA VIRUS

361

influenza virus hemagglutinin. The low pH-induced conformational change. J . Biol. Chem. 260, 2913-298 I . Doms. R. W., Gething, M.-J.. Henneberry, J . H., White, J . . and Helenius, A. (1986). A variant influenza hemagglutin that induces fusion at elevated pH. J. Virol. 57,603-613. Doyle, C.. Roth. M. G., Sambrook, J . , and Gething, M-J. (1985). Mutations in the cytoplasmic domain of the influenza virus hemagglutin affect different stages of intracellular transport. J . Cell B i d . 100, 704-714. Doyle, C., Sambrook, J . , and Gething, M.-J. (1986). Progressive deletions of the transmembrane and cytoplasmic domains of influenza hemagglutinin. J. Cell B i d . 103, 11931204. Doxsey. S. J., Sambrook, J . . Helenius, A.. and White. J. M. (1985). An efficient method for introducing macromolecules into living cells. J. Cell Biol. 101, 19-27. Drzenick, R., Seto, J. T., and Rott. R. (1966). Characterization of neuraminidases from myxoviruses. Biwhim. Biiplrys. Acru 128, 547-558. Garten, W., Bosch, F.-X., Linder, D., Rott, R., and Klenk. H.-D. (1981). Proteolytic activation of the influenza virus hemagglutinin: The structure of the cleavage site and the enzymes involved in cleavage. Virok>gy115, 361-374. Gething. M . 4 . . and Sambrook, J . (19811. Cell-surface expression of influenza haemag glutinin from a cloned DNA copy of the RNA gene. Nuture (London)293, 620-625. Gething. M.-J.. and Sambrook, J. (1982). Construction of influenza haemagglutinin genes that code for intracellular and secreted forms of the protein. Nuture (London)300,598603. Gething. M.-J.. and Sambrook, J. (1983). Expression of proteins on the cell surface using mammalian vectors. In “Genetics of Influenza Viruses” (P. Palese and D. W. Kingsbury, eds.), pp. 169-191. Springer-Verlag. Berlin and New York. White, J. M., and Waterfield, M. D. (1978). Purification of the fusion protein Gething, M.-J.. of Sendai virus: Analysis of the amino terminal sequence generated during precursor activation. Proc. Natl. Acad. Sci. U.S.A. 75, 2137-2140. Gething, M.-J., Bye, J., Skehel, J. J., and Waterfield. M. D. (1980). Cloning and DNA sequence of double stranded copies of haemagglutinin genes from H2 and H3 strains elucidates antigenic shift and drift in human influenza virus. Nuture (London)287,301306. Gething, M.-J., Doms, R. W., York, D.. and White, J . (1986a). Studies ofthe mechanism of membrane fusion: Site-specific mutagenesis of the hemagglutinin of influenza virus. J . Cell Biol. 102, 11-23. Gething, M.-J., McCammon, K., and Sambrook. J . (1986b). Expression of wild-type and mutant forms of influenza hemagglutinin: The role of folding in intracellular transport. Cell 46, 939-950. Gething. M. J . . Doms, R. W.. White, J. M., and Helenius, A. (1986~).Studies on the mechanism of membrane fusion. I n ”Protein Engineering: Applications in Science, Medicine and Engineering” (M. Inouye and R. Sarma, eds.). pp. 193-215. Academic Press, New York. Hirst. G. K. (1941). The agglutination of red cells by allantoic fluid of chick embryos infected with influenza virus. Science 94, 22-23. Huang, R. T., Rott, R., and Klenk, H.-D. (1981). Influenza viruses cause hemolysis and fusion of cells. Virology 110, 243-247. Kaiser, E. T., and Kezdy, F. J . (1984). Amphiphilic secondary structure: Design of peptide hormones. Science 223, 249-255. Kawaoka. Y., Naeve, C. W., and Webster, R. G . (1984). Is virulence of H5N2 influenza viruses in chickens associated with loss of carbohydrate from the hemagglutinin? Virology 139, 303-316.

362

MARY-JANE GETHING ET AL.

Kendal, A. P., and Patriarca, P. A., eds. (1986). ”Options for the Control of lnfluenza Virus.” Liss, New York. Klenk, H.-D.,Rott, R.,Orlich, M., and Blodorn, J . (1975). Activationof influenzaA viruses by trypsin treatment. Virology 68, 426-430. Klenk, H.-D., Garten, W., Keil, W., Niemann. H., Bosch, F. X.,Schwartz, R. T.. Scholtissek, C., and Rott, R. (1981). Processing of the hemagglutinin. I n “Genetic Variation among Influenza Viruses (D. P. Nayak, ed.), pp. 193-21 1. Academic Press, New York. Klenk, H.-D., Garten, W., and Rott, R. (1984). Inhibition of proteolytic cleavage of the hemagglutinin of influenza virus by the calcium-specific ionophore A23 187. EMBO J . 3, 2911-2915. Lamb, R. A. (1983). The influenza virus RNA segments and their encoded proteins. In “Genetics of Influenza Viruses” (P. Palese and D. W. Kingsbury. eds.), pp. 21-69. Springer-Verlag, Berlin and New York. Laver, W. G. (1963). The structure of influenza viruses. 3. Disruption of the virus particle and separation of neuraminidase activity. Virology 20, 25 1-262. Laver, W. C. (1964). Structural studies on the protein subunits from three strains of influenza virus. J . Mid. Biol. 9, 109-124. Laver, W. G. (1971). Separation of two polypeptide chains from the hemagglutinin subunit of influenza virus. Virology 45, 275-288. Laver, W. C. (1973). The polypeptides of influenza virus. Adu. Virus Res. 18, 57-103. Laver, W. G., and Kilbourne, E. D. (1966). Identification in a recombinant influenza virus of structural proteins derived from both parents. Virology 30, 493-501. Laver, W. G., and Webster, R. G. (1968). Selection of antigenic mutants of influenza viruses: Isolation and peptide mapping of their hemagglutinating proteins. Virology 34, 193-202. Laver, W. G., and Webster, R. G. (1979). Ecology of influenza viruses in lower mammals and birds. Br. Med. Bull. 35, 29-33. Lazarowitz, S. G., and Choppin, P. W. (1975). Enhancement of the infectivity ofinfluenza A and B viruses by proteolytic cleavage of the hemagglutinin polypeptide. Virology 68, 440-454. Lenard, J., and Miller, D. (1982). Uncoating of enveloped viruses. CeII 28, 5-6. Maeda, T., and Ohnishi. S. (1980). Activation of influenza virus by acidic media causes hemolysis and fusion of erythrocytes. FEBS Lett. U2,283-287. Maeda, T., Kawasaki, J., and Ohnishi, S. (1981). Interaction of influenza virus hemagglutinin with target membrane lipids is a key step in virus-induced hemolysis and fusion at pH 5.2. Proc. Natl. Acad. Sci. U . S . A . 78, 4133-4137. Marsh, M., Wellsteed, J., Kern, H . . Harms, E., and Helenius, A. (1982). Monensin inhibits Semliki Forest virus penetration into culture cells. froc. Natl. Acad. Sci. U.S.A. 79, 5297-5301. Matlin. K., and Simons, K. (1983). Reduced temperature prevents transfer of a membrane glycoprotein to the cell surface but does not prevent terminal glycosylation. Cell 34, 233-243. Matlin, K., and Simons, K . (1984). Sorting of an apical plasma membrane glycoprotein occurs before it reaches the cell surface in cultured epithelial cells. J . Cell Biol. 99, 2131-2 139. Matlin, K., Reggio, H., Helenius, A,, and Simons, K. (1981). The effective entry of influenza virus into MDCK cells. J. Cell Biol. 91, 601-613. M a t h K.,Reggio, H., Simons, K., and Helenius, A. (1982). Pathway of vesicular stomatitis virus entry leading to infection. J . Mol. Biol. 156, 609-631. Miller, D. K., and Lenard, J. (1981). Antihistaminics, local anesthetics, and other amines as antiviral agents. f r o c . Natl. Acad. Sci. U.S.A. 78, 3605-3609.

11. HEMAGGLUTININ OF INFLUENZA VIRUS

363

Mills. K. H . G . (1986). Processing of viral antigens and presentation to class 11-restricted T cells. Immrrnol. 7 i h y 7, 260-263. Morrison. L. A , . Lukacher. A . E.. Braciale. V . L . . Fan. D. P . . and Braciale. T. J . (1986). Differences in antigen presentation to MHC Class I-and Class Il-restricted influenza virus-specific cytolytic T lymphocyte clones. J . E.1-p.Merl. 163, 903-92 I , Richardson, C.. Scheid, A , . and Choppin, P. (1980). Specific inhibition of paramyxovirus and myxovirus replication by oligopeptides with amino acid sequences similar to those at the N-termini of the F, o r HA: polypeptides. Virology 105, 205-222. Rott. R.. and Klenk, H.-D. (1986). Pathogenicity of influenza virus in model systems. I n “Options for the Control of Influenza Virus” ( P . A . Kendal and P. A . Patriarca. eds.). pp. 53-62. Liss. New York. Rott. R.. Orlich. M., Klenk. H.-D.. Wang. M. L . , Skehel, J . J., and Wiley. D. C. (1984). Studies on the adaptation of influenra viruses to MDCK cells. EMBO J . 3, 33293332. Sambrook. J.. Rodgers. L.. White, J . . and Gething, M.-J. (1985). Lines of BPV-transformed murine cells that constitutively express influenza virus hemagglutinin. EMBO J . 4, 91103. Schiffer. M.. and Edmundson. A . B. (1967). Use of helical wheel\ to represent the structures of proteins and to identify segments with helical potential. Biophvs. J . 7. 121-135. Schlotissek, C.. and Bowles. A. L. (1975). Isolation and characterization of temperaturesensitive mutants of fowl plague virus. Virology 67, 576-581. Schulman, J . L. ( 1983). Virus-determined differences in the pathogenesis of influenza virus infections. In “Genetics of Influenza Viruses” ( P . Palese and D. W. Kingsbury. eds.). pp. 305-320. Springer-Verlag Berlin and New York. Skehel. J . J.. and Waterfield. M. D. (1975). Studies on the primary structure o f t h e influenza virus haemagglutinin. Pro(.. Nurl. Ac.od. .Sc,i. U . S . A . 72, 93. Skehel. J . J . . Bayley, P. M.. Brown. E. B . , Martin. S . R.. Waterfield. M. D . , White, J . M.. Wilson. I. A , , and Wiley. D. C. (1982). Changes in the conformation of influenza virus haemagglutinin at the pH optimum of virus-mediated membrane fusion. /‘roc.. Nor/. Ai.ud. Sci. U . S . A . 79, 968-972. Ueda. M.. and Kilbourne. E. D. (1976). Temperature-sensitive mutants of influenza virus: A mutation in the hemagglutinin gene. Virolog,v 70, 425-431. VBAninen. P., and Kaariainen, L. (1980). Fusion and haemolysis of erythrocytes caused by three togaviruses: Semliki Forest, Sindbis and rubella. J . Gcn. Virol. 46,467-475. Ward. C . W. (1981). Structure of the influenra virus hemagglutinin. Crtrr.. 7 o p . M i c r ~ h i c ~ l . Imtnrtnol. 94, 1-74. Ward, C . W . . and Dopheide, T. A . A . (1980). The Hong K o n g ( H 3 ) hemagglutinin. Complete amino acid sequence and oligosaccharide distribution for the heavy chain of AiMemphisi102172. In “Structure and Variation in Influenza Virus” ( G . Laver and G . Air, eds.), pp. 27-37. Elsevier. New York. Waterfield. M. D.. Espelie, K., Elder. K . . and Skehel, J . J . (1979). Structure of the haemaglutinin of influenza virus. Br. M c d . Brill. 35, 57-63. Waterfield, M. D.. Gething. M.-J., Scrace, G . . and Skehel, J. J . (1980). The carbohydrate side chains and disulphide bonds of the hemagglutinin of the influenza virus AiJapan 305/57 ( H 2 N I ) . In “Structure and Variation in Influenza Virus“ (C. Laver and G . Air. eds.). pp. 11-20. Elsevier. New York. White, J. M., and Wilson, 1. A. (1987). Anti-peptide antibodies detect steps in a protein conformational change: Low-pH activation of the influenza virus hemagglutinin. J . Cell B i ~ l 105, . 2887-2896. White, J . . Kartenbeck. J . , and Helenius, A . (1980). Fusion of Semliki Forest virus with the plasma membrane can be induced by low pH. J . Cell B i d . 87, 264-272.

364

MARYJANE GETHING ET AL.

White, J. M., M a t h , K., and Helenius, A. (1981). Cell fusion by Semliki Forest. influenza, and vesicular stomatitis viruses. J . Cell B i d . 89, 674-679. White, 3.. Kartenbeck, J., and Helenius, A. (1982a). Membrane fusion activity of influenza virus. EMEO J . 1, 217-222. White, J . M . , Helenius, A. A., and Gething, M.4. (1982b). Haernagglutinin of influenza expressed from a cloned gene promotes membrane fusion. Nature (London)300, 658659. White, J., Kielian, M., and Helenius, A. (1983). Membrane fusion proteins of enveloped animal viruses. Q. Rev. Eiophys. 16, 151-195. Wiley, D. C., Skehel, J . J., and Waterfield, M. D. (1977). Evidence from studies with a cross-linking reagent that the hemagglutinin of influenza virus is a trimer. Virdogy 79, 446-448. Wiley, D. C., Wilson, 1. A . , and Skehel, J . J . (1981).Structural identification of the antibody binding sites of the Hong Kong influenza virus haemagglutinin and their involvement in antigenic variation. Nuture (London) 289, 373-378. Wilson, 1. A.. Skehel, J. J., and Wiley, D. C. (1981). The haemagglutinin membrane glycoprotein of influenza virus: Structure at 3 A resolution. Nurure (London) 289, 366-373.

Index

A

4-Acetamido-4’-isothiocyanostilbene 2.2’-disulfonic acid inhibition of anion transport in erythrocyte membranes, 208 chromaffin granule ATPase, 209 5-hydroxytryptamine secretion. 2 14 parathyroid hormone secretion, 2 14-2 IS Acetate. inhibition of Ca2+-dependent catecholamine release in permeabilized chromaffin cells, 2 16-2 18 Acetylcholine release from nerve terminals, 143 at neuromuscular junction. 116-1 17. 145 N-Acetylglucosamine, effect on myoblast fusion. 95 Acidic residues. protonation in hydrophobic segment, 281 Acridine orange, 229 complexing by anionic substances, 2 32 -2 33 deficiency of, 235 marker trapping, 235 mobility of forms, 233-234 permeation of membranes, 232 suramin pretreatment effect, 245 transfer to phagosomes. 229, 245 uptake of sulfonated fluors and. 237 Acrosomal membranes intramembranous particles. 16- 17 sperm-egg fusion site, 13 Acrosome reaction freeze-fracture images of acrosomal membrane, 16

membrane phospholipid changes and, 17 sea urchins, 6 sturgeon spermatozoa and, 9-10 Acrylamide, fluorescence quenching, 272 Adenosine triphosphate analogs, chromaffin granule lysis and. 206 and Ca2+,requirements in chromaffin cells and platelets. 149 lysis induced in isolated chromafin granules, 208 removal, secretory system sensitivity, 126 Adenovirus, cell entry route. 288 Adenylate cyclase system, G proteins, 54-55 Adrenal chromaffin cells, botulinum toxin effect, 75 Adrenal medullary cells, permeabilized, exocytosis inhibition, 132-134 Aequorin Ca” role in stimulus-secretion coupling and, 120 microinjection, in sea urchin eggs. 51 Aldehyde fixation blebs, 175 exocytosis and, 170 vesicular structures, 175 Alphavirus E l glycoproteins, hydrophobic segments, 260 Amiloride, protein kinase C inhibition, I30

Amino acid residues, hydrophobic. HA2, FI. and gp36 glycoproteins, 259 Aminopeptidase M,effect on FI amino terminus, 267 Amino terminals, HA2, FI. and gp36 glycoproteins. 259

365

366 Ammonia chromaffin granule lysis mediated by, 208 effect on methylamine accumulation in permeabilized chromaffin cells, 216-2 I7 veratridine-induced catecholamine secretion in chromaffin cells, 216-219 Amphibians eggs, vitelline envelope, 10 sperm-egg fusion morphology, 10 Amphipathic helix, in hemagglutinin, 351-352 Amphomycin, effect on sperm-egg fusion, 33 Anion permeability, pH in isolated chromaffin cells and, 205 Anion permeability series, 207, 212 Anion transport blockers anion transport mechanisms in chromaffin granules and, 208 inhibition of 5-hydroxytryptamine secretion, 214 Antibodies anticalmodulin cortical exocytosis inhibition, 66 exocytosis and, 152 antihemagglutinin, 342-344 antimembrane, sperm-egg fusion and, 32 delivery into cytoplasm, 344 Antigenic properties, influenza virus, 337-338 Antigenic structure, hemagglutinin, 338 Artifacts with fluorescent lysosomal probes, 234-235 in rapid freezing, 173-175 ATP, see Adenosine triphosphate ATPase chromaffin granules, potentiation and inhibition, 209 clathrin-dependent, clathrin coat removal and, 194 F , - , mitochondrial, in chromaffin membrane preparations, 204-205 proton-translocating, see Proton-translocating ATPase

INDEX

ATPase I, chromaffin granule membranes, 204-205 ATPase 11, chromaffin granule membranes, 204-205 AtT20 cells, a-latrotoxin effect. 155 Azurophilic granules, polymorphonuclear leukocytes, 142-143

B Band 3 proteins, mobilization by influenza virus, 275 Basophils, human, antigen-induced histamine release, permeant anion role, 215 BAY-K 8644, calcium agonist, 123 BHA, see Bromelain hemagglutinin BHK-21 cells, Semliki Forest virus envelope fusion, 263 Bilayers, membrane, role in membrane fusion, 101-102 Bindin acrosomal process and, 6 in sperm-egg fusion, 34 Birds, sperm-egg fusion morphology, I I Blebs in glutaraldehyde-fixed cells, formation, 181-182 production by aldehyde fixation, 175 Botulinum toxin effect on adrenal chromaffin cells, 75 capacitance, 125 exocytosis inhibition, 124-125, 132, 134, 153-1 54 type D, effect on exocytosis, 125 Bovine papilloma virus vectors, 341-345 Bromelain hemagglutinin (BHA) circular dichroism spectrum, 274 three-dimensional structure, 272 trypsin action on, 274 Bromide anion Class 1, 208 inhibition of Ca*+-dependent catecholamine release in permeabilized chromaffin cells, 216-218 permeability, 207 use of blocker-sensitive anion transport site, 209

367

INDEX Er& juponica. sperm-egg fusion, 10 BV I-MTHA cell lines. hemagglutinin expression and low pHmediated cell-cell fusion, correlation. 346 C

C127 cells bovine papilloma virus-transformed, hemagglutinin expression, 342 X-3 I gene introduction. hemagglutinin expression, 344 Calcium ion agonists and antagonists, 123 and ATP. requirements in chromaffin cells and platelets, 149 channels agonist-gated, 123 in nervous tissue. 123 voltage-sensitive, 123 clamping. 150 in cortical exocytosis (sea urchin), 51-52 G proteins and, 54-55 phosphoinositide cycle, 52-54 phospholipase C, phosphatidylinositol-specific, 52. 74 signal transduction. 52-54 effect on hemagglutinating virus of Japan-mediated cell fusion, 310-31 I elevation. transiency of release responses induced by. 149 endocytosis regulation and, 159-160 exocytosis control and, 120-121 extracellular, in sperm-egg fusion, 27-28 fusogenic process and, 126 granule swelling and, 220 influx in spermatozoa. I5 inhibition of phagosome-lysosome fusion by polyanionics and, 247-248 pore formation in virus-induced hemolysis, 265 myoblast fusion dependence on, 89, 91-93, 99-101, 105 permeability in secretory cells, 120-121 phase separation of phospholipids and, 35 as primary messenger in exocytosis, 122- 123

protein kinase C activators and, I50 in regulated exocytosis. 147-151 in regulated secretion, 47-48 release from endoplasmic reticulum, 124 requirement for protein kinase C. 129- I30 requirement in secretory systems, 122 stimulation of proteolysis, membrane fusion and, 103-104 Calcium ionophore A23187 lysosomal enzyme release induced by, inhibition by 4-acetamido4’-isothiocyanostilbene 2.2‘disulfonic acid, 215 stimulation of 5-hydroxytryptamine secretion, inhibition by anion transport blockers, 2 14 Carboxyl groups, protonation, 282-284 Catecholamine secretion (chromaftin cells) digitonin-permeabilized.2 18-219 osmotic effects, 216-219 permeabilized cells. 216-219 proton gradient and, 213 Cation-exchange resins acridine orange complexing, 232-233 phagosome-lysosome fusion and. 248 Cations, impermeability of chromaffin granules, 206 Cell-free secretory systems, 149 Cell lines. myogenic. 89 Cell lysis. hemagglutinating virus of Japan-mediated in Erlich ascites tumor cells. 309-3 1 1 Cell specificity. of myoblast fusion. 89-91 Cell surface complex preparations, 59 Chumu mucerphyllu, sperm-egg fusion morphology, 6-8 Chemically fixed cells, exocytosis. 170175 Chemiosmotic hypothesis, 203-204 alternative hypotheses. 219-220 Chloride anion Class 1. 208 CI-.ATP-induced lysis of isolated chromaffin granules, 207-208 inhibition of Ca”-dependent catecholamine release in permeabilized chromdffin cells, 2 16-21 8 permeability, 207

INDEX

use of blocker-sensitive anion transport site, 209 Chloroquine, phagosome-lysosome fusion and, 245-246 CHO cells, 31 gene introduction, hemagglutinin expression, 344 Cholera toxin, cortical exocytosis stimulation, 74 Cholesterol effect on fusion of hemagglutinating virus of Japan, 264-265 membrane distribution during exocytosis, 185 in target membranes, binding of FI amino-terminal segment, 267 Chromaffin cells digitonin-permeabilized, catecholamine secretion, 218-2 19 exocytotic fusion-fission, GTP-binding protein role, 152 permeabilized membrane recycling rate and Ca2+, 160 osmotic effects on secretion, 216-219 rapidly frozen, exocytosis, 177-179 regulated exocytosis, ATP and Ca2' requirements, I49 secretion, osmotic effects, 21 1-214 catecholamine secretion, 213-214 osmotic strength of media effects, 212 permeant anion effects, 212 probenecid effects. 212 proton electrochemical gradient and, 213-214 pyridoxal phosphate effects, 212 p-trifluoromethoxyphenylh ydrazone effects, 2 12-2 I4 Chromaffin granules ghosts, 204-205, 210-21 I ionic permeabilities, 206-209 isolated electrochemical proton gradient, 205 lysis, anion groups and, 208 pH. intragranular. 205 proton-translocating ATPase, 204-205 valinomycin,K+-induced lysis, 208 osmotic fragility in vitro, 213 osmotic lysis, 206-209 Chromobindin 9, membrane fusion-fission in exocytosis and, 152

Chymotrypsin, effect on sperm-egg fusion, 3 1-32 Circular dichroism bromelain hemagglutinin (BHA) spectrum, 274 measurement of influenza virus hemagglutinin, 279 Clathrin basket, 191. 194 curvature, enzymatic activities and, 194 Cleavage, proteolytic, hemagglutinin precursor, 339, 358 Clostridium toxins, exocytosis inhibition, 153-155

Clustering, intramembrane particle arrays, 323-32.5 "C-NMR, protonation of carboxyl groups, 282-284 Coated vesicles in endocytosis, 157, 159 virus uptake, 262, 286 Complementary DNA, encoding influenza virus hernagglutinin, 341-345 Concanavalin A, effect on myoblast fusion, 94-95 Conformational change, hemagglutinins, 352-353 Constitutive secretion, in exocytosis. 46-47 Cortical exocytosis (sea urchin egg), 48-51 calcium signal, 51-55 cell surface complex preparations, 59 cortical lawn preparations, 59 cortical vesicle discharge, 59-62 cortical vesicle-plasma membrane junction, 70 G proteins and, 54-55 inhibitors, 64-66 metabolic energy and, 71-72 osmotic forces and, 70-71 phosphoinositide cycle, 52-54 phospholipases and, 73 vectorial nature of, 61 viral model, 72-73 Cortical granules, proteolytic component, 25 Cortical lawn preparations, 59 Cortical vesicles (sea urchin egg), 48-51 discharge assays, 62-64 morphology, 59-62

369

INDEX

junction with plasma membrane, 70 purification. 66-67 purified, reconstitution of active cortex from. 67-69 Cross fertilization species specificity of sperm-egg fusion. 20-24

in various species. 24 vitelline envelope and, 22 CV-I cells anchor-minus precursor hemagglutinin, 357

hemagglutinins produced from SV40-HA and BPV-HA vectors in,

macrophage lysosome\ and, 232-233

hehavior of nonionic water-dispersihle polymers and. 239-240 cells labeled with Thorotrast and sulfonated fluors. dye transfer to phagosomes. 240 fusion inhibition. lysosomal pH and. 247 phagosome-lysosome fusion and. 229 Diac ylgl ycerol analogs. effect on exocytosis. 128 in cortical exocytosis, 73 myoblast fusion and. I02 as primary messenger in exocytosis.

344

122-123

low pH-induced fusion. 3S9 mutant hemagglutinin protein activity. 354

Cyclic AMP effect on exocytosis in parotid cells. I50 exocytosis regulation in parotid acinar cells, 153 hemagglutinating virus of Japanmediated fusion and, 317 levels after myoblast fusion, 106 in salivary and pancreatic exocrine secretion, 120 Cytochalasin B . inhibition of hemagglutinating virus of Japan-mediated fusion, 320-323 Cytochalasin D inhibition of Ehrlich ascites tumor cell fusion, 331 hemagglutinating virus of Japan-mediated fusion. 320-323 Cytochalasins effect on sperm-egg fusion, 33 inhibition of hemagglutinating virus of Japan-induced cell fusion, 266 Cytoplasmic organelles, alteration, hemagglutinating virus of Japan-mediated fusion and, 317-320

D 2-Deoxyglucose. effect on sperm-egg fusion, 33 Dextrans, exocytosis inhibition. 186 Dextran sulfate acridine orange accumulation in

signal transduction and, 52-54 Diaphragms, freeze-fracture imaging. 17 I Dicyclohex ylcarhodiimide effect on secretion from intact chromaffin cells, 213 reactive peptide in ATPase I. 205 Differential delivery, lysosomal constituents. 240-241 Digestion, intracellular. polyanionics and. 242-244

Digitonin-permeahilized chromaffin cells, catecholamine secretion, 218-219 Dihydropyridines, calcium antagonists. 123 Diphtheria toxin cell entry route, 288 fusion. hemagglutinating virus of Japan-mediated, 303-305 Disaccharides, inhibition of hemagglutinating virus of Japan-induced cell fusion. 266 S.S’-Dithiohis(2-nitrohenzoicacid]. inhibition of cortical exocytosis. 64 Dithiothreitol. effect on sperm-egg fusion, 33

Dowex-SO. affinity for acridine orange, 233 Dynein arms, sperm motility and, 18

E Ectodomain fragment. hemagglutinin, 340 Efficiency of fusion, influenza virus mutants, 350 Egg cortex (sea urchin) cortical vesicle purification, 66-67

370 plasma membrane purification, 67 reconstitution, 67-69 Egg plasma membrane electric potential, 30-3 1 intermingling with sperm plasma membrane, 19-20 membrane potential. 25 sea urchin, 48 junction with cortical vesicles, 70 purification, 67 purified, reconstitution of active cortex from, 67-69 sperm-egg fusion sites, 17-18 E glycoproteins, Semliki Forest virus, 276 EGTA, microinjection, in sea urchin eggs, 51

Ehrlich ascites tumor cells hemagglutinating virus of Japanmediated fusion cell lysis and, 309-3 I I close attachment of cell membranes, 325-329 clustering of intramembrane particle arrays, 323-325 concentration of virus required for. 308-309 cyclic AMP levels and, 317 cytoplasmic organelles, alteration, 3 17-320 effect of cations, 310-31 1 freeze-fracture imaging, 326-327 incubation at 37°C. 314-317 inhibition by cytochalasins B and D, 33 I , 320-323 membrane potential and, 315 model, 329-331 perturbation of cell membrane structure, 323-325 pH range required for, 308-309 viral envelope fusion and cell fusion. 320-323 virus-cell interaction at low temperature, 312-314 Elaidic acid, effect on myoblast fusion, 98 Electronectin, synthesis before myoblast fusion, 95-% Electron microscopy Ca2'-mediated fusion, 59-62 exocytosis monitoring by, 117-1 18 exocytosis rates and, 140

INDEX

illusion of inhibited fusion of secondary lysosomes, 236 markers, hydrosol trapping of, 240 transfer of acridine orange and Thorotrast to phagosomes, 229-230 viral envelope fusion, 262 Electropermeabilizdtion,122 Endocytic vesicles, prelysosomal, rapid acidification, 287 Endocytosis, 189- 194 clathrin basket, 191, 194 exocytosis and, 196 mechanisms, fast and slow, 158-159 membrane sorting, 156-159 regulation, 159-160 rotary-shadow method, 189 synaptic vesicle depletion, 160 triggered, 156- 160 Endoplasmic reticulum CaZ+release, 124 hemagglutinin localization, 341 Endosomes, Semliki Forest virus and influenza virus genome transfer, 262 Eosine. transient dichroism, 264 Epinephrine secretion chromaffin cells, 212 isolated chromaffin granules, 208 Erythrocyte membranes anion transport, inhibition, 208 band 3 proteins, mobilization by influenza virus, 275 fusion, viral envelope fusion and. 331 hemagglutinating virus of Japan fusion. 263-265 influenza virus effects, 274-275 influenza virus fusion, 269, 339 trypsinized, vesicular stomatitis virus envelope fusion and hemolysis, 263 Erythrosine B, effect on sperm-egg fusion, 33 Escherichiu coli

affinity for acridine orange, 233 microbicidal activity, phagosomelysosome fusion and, 244-246 N-Ethylmaleimide, inhibition of cortical exocytosis, 64 Exocrine gland cells, membrane recycling in, 159 Exocytosis biochemical study, limitations, 140

371

INDEX

CaZ+permeability in secretory cells. 120-12 I in chemically fixed cells, 170-17.5 constitutive secretion, 46-47, 142 continuous, I16 cortical. see Cortical exocytosis electron microscopy, 117-1 18 electrophysiological approach, 140-141 endocytosis and, 196 guanosine 5'4riphosphate binding-protein role in fusion-fission. I52 inhibition by botulinum toxin. 132, 134 Closrridium toxins, 153-155 dextrans. 186 GTPyS. 132 hyperosmotic forces, 186 inhibition in permeable adrenal medullary cells. 132-134 in v i m models, 56-73 Caz'-triggered, egg cortex and, 59-70 cell-free systems, 58 permeabilized cell systems, 56-57 membrane capacitance and, 117-1 18 membrane fusion-fission in, 151-1.53 Mg-ATP role. 126-129 permeabilized-cell studies, 121- 123 primary message generation. 123-124 protein kinase C and, 129-130 in rapidly frozen cells, 175-189 rates, activation of protein kinases and Ca", 152 regulated ATP and CaZ+requirements, 149 release, 142-143 second messenger control, 147-15 I secretion, 46-47 stimulation by a-latrotoxin, 155-156 stimulus-secretion coupling. 47-48 toxins and, 124-125 triggered, 116, 118, 120-121 types, 128-129 viral fusion model, 1 18- I20 Exo-endocytotic coupling, 141-142

F FI-ATPase, mitochondrial, in chromaffin membrane preparations, 204-205

Fatty acids, see ulso specific w i d effect on myoblast fusion. 98-99 in sperm-egg fusion, 34 cis-unsaturated. requirement for vesicular stomatitis virus. 277 Ferririn, 229 Fertilization envelope, sea urchin egg, SO Fibroblasts, in myoblast cultures. 88-89 Field strains, influenza virus, 351 Fishes, sperm-egg fusion morphology, 8-10 Fission. in exocytotic process, 206. 220 Fluoresceinated dextran, and mobile impermeant Huors, differential and sequential delivery, 241-242 Fluorescein isothiocyanate. in pH measurements, 287 Fluorescence, relief of quenching, 264 Fluorescence microscopy, illusion of inhibited fusion of secondary lysosomes, 236 Fluoride anion Class 111, 208 permeability, 207 Fluors, sulfonated, see Sulfonated Huors Free energy of transfer, virus hydrophobic segment peptide from aqueous to lipid bilayer phases, 279-280 Freeze-fracture electron microscopy myoblast fusion, 101 particle-free patches in sperm membrane, 34 particle-free regions in myoblast membrane at fusion sites, 99 Freeze-fracture imaging cell membranes induced by hemagglutinating virus of Japan, 327-328 diaphragms, 171 intramembrane particle arrays. 171-173 sperm membrane before and after acrosome reaction, 16 Freeze-thawing, hemagglutinating virus of Japan, 265 Fura2 Ca" measurement with. 149 microinjection, in sea urchin eggs, 5152 Fusers, microorganisms as, 246-247 Fusion-fission. membrane, in exocytosis, IS 1-153

372

INDEX

G Gullus gullus, sperm-egg fusion, 1 I Gangliosides, toxin binding, 154 Gap junctions, myoblast fusion and, 100 Gelatinous structure gelatinous trap model. 239-240 residual bodies, 246-249 testing, 241-242 in lysosomes, 239 @-I ,3-Glucanase, cortical vesicle exocytosis assay, 63-64 in sea urchin cortical vesicles, 50 Glutamate, effect on Ca”-dependent catecholamine release in permeabilized chromaffin cells, 2 16-218 Glutaraldehyde-fixed cells, bleb formation, 181 -182 Glycerol, dehydration, induction of intramembrane particle arrays, 173-175 GI ycoproteins myoblast fusion and. 94-96 viral envelope, see speciJic envelope glvcoprotein

Golgi regions, hemagglutinin localization, 34 I Granule membranes cholesterol distribution during exocytosis, 185 intermixing with plasma membrane constituents during fusion. 183 mobility before and during fusion, 183-185 GTP-binding protein, role in exocytosis, 75 Guanosine 5‘-diphosphate, role in exocytosis. 54-55 Guanosine 5’-triphosphate analogs exocytosis inhibition, 132 role in exocytosis, 54-55 GTPyS, 124 phospholipase C stimulation, 124 role in exocytosis, 54-55, 123

H Hemagglutinating virus of Japan (HVJ) Ca2+ inhibition of pore formation, 265

cholesterol effect on fusion, 264-265 clustering of intramembrane particle arrays, 266 effect on target membranes, 266 envelope structure, 301-303 FI glycoprotein amino-terminal hydrophobic segment, 267 amino-terminal segment, 259 binding to cholesterol in target membrane, 267 free energies for transfer, 280 molecular weight and function, 258 posttranslational cleavage, 301 precursor forms, 258 F2 glycoprotein molecular weight and function, 258 posttranslational cleavage, 301 precursor forms, 258 freeze-thawing, 265 fusion of Ehrlich ascites tumor cells, 306-308 cell lysis and, 309-3 I 1 close attachment of cell membranes, 325-329 clustering of intramembrane particle arrays, 323-325 concentration required for, 308-309 cyclic AMP levels, 317 cytoplasmic organelles, alteration, 3 17-320 effect of cations, 310-311 freeze-fracture imaging, 326-327 incubation at 37”C, 314-317 inhibition, 320-323 model, 329-33 I membrane potential and, 315 perturbation of cell membrane structure, 323-325 pH range required for, 308-309 viral envelope and cell fusion separation, 320-323 virus-cell interaction at low temperatures, 3 12-3 14 fusion cell, 266, 300-301 envelope, 263-265 with erythrocyte membranes, 263-265 factors required for, 303-306 lipid intermixing, 264

INDEX

with liposomes. 264-265 rate constant. 264 harvests, early and late, 265 hemolysis, 265 hemolytic activity. 300-301 HN glycoprotein, molecular weight and function. 258 infectivity, 300-301 intramembrane particle arrays, 265-266 nucleocapsid strands. 265 proteins, 299-300 sequence homology, 259 sonication, 265 structure. 299-300 Hemagglutinin amphipathic helix role in fusion. 351-352 anchor-minus precursor secreted from CV-I cells, 357 antigenic structure, 338 assays for fusion activity, 340-341 circular dichroism measurements. 279 cleavage activation, 353-359 cloned genes, site-specific mutagenesis. 346 conformational change, low pH-induced. 352-353 ectodomain fragment, 340 expression in cells from cloned hemagglutinin cDNAs, 341-345 fusion peptide, site-directed mutagenesis. 347-351 HA I . 258, 339-340 interaction with liposomes. 271-272 molecular weight and function. 258 HA2. 258, 339-340 amino-terminal segment, 259 hydrophobic segment, 272 secondary structure formation and. 284 free energies for transfer. 280 hydrophobic segments. 260 interaction with liposomes. 271-272 molecular weight and function, 258 primary sequence, as a factor in interaction with lipid bilayers, 284 interaction with liposomes, 271-272 low pH-induced conformational change, 272-274 mutants C'RR, 354

373 C'RRKKR, 354-355, 357-358 C T, 354-355 precursor forms, 258 posttranslational cleavage, 339, 358 reconstituted vesicles, 272 rosettes. 270 synthetic peptide from. 271 synthesized from SV40-HA and BPV-HA vectors, comparison, 343 three-dimensional structure. 338 transport, 358 vectors based on bovine papilloma virus, 34 1-345 Hemolysis, hemagglutinating virus of Japan-induced. 265, 300-301 Histamine release, from human basophils, permeant anion role, 215 Hoechst 33342, sea urchin egg staining, 26 Horseradish peroxidase, endocytosed, in dextran sulfate-inhibited cells, 242 Host-controlled modification, hemagglutinating virus of Japan, 300-301 HVJ virus, see Hemagglutinating virus of Japan Hyaline layer barriers to excess spermatozoa and. 30 resistance to fusion with spermatozoa and, 25 in sea urchin cortical vesicles, 50 Hydration repulsion, myoblast fusion and, 102-103 Hvdroides he.rugonits, sperm-egg fusion morphology, 6-8 Hydrophobicity, HA2, FI. and gp36 amino-terminal segments, 259 Hydrophobic segment entrance into lipid bilayer, 279-280 lipid bilayer as target, 279 protonation of acidic residues, 281-282 viral fusion glycoproteins, 260-261 Hydrophobic stretches, internal. Semliki Forest virus El and vesicular stomatitis virus G glycoprotein, 259 Hydrosols accumulation of polyanionic substances in lysosomes and, 23 I formation in lysosomes. 239-240 movement in lysosomes, 239-240

INDEX

structure in lysosomes, 239-240 trapping of colloidal electron microscopic markers, 240 5-Hydroxytryptamine, transport in porcine anterior pituitary, 210-21 1 Hyperosmotic buffers, inhibition of cortical vesicle discharge, 7 I Hyperosmotic forces, exocytosis inhibition, 186 I

Immune response, cellular, to influenza virus, 338 Immunocytochemistry, secretogranin 11 identification, 143 lmmunofluorescence indirect, hemagglutinin localization, 341 surface changes during myoblast fusion, 96-97 Infectivity, hemagglutinating virus of Japan, 300-301 Influenza C virus glycoproteins, hydrophobic segments, 260 sequence homology, 259 Influenza virus antigenic properties, 337-338 binding to liposomes lacking receptors, 279 brornelain hemagglutinin (BHA) three-dimensional structure, 272 trypsin action on, 274 cellular immune response to, 338 in coated pits and coated vesicles, 268 conformational change in viral glycoprotein, 278-279 effect on erythrocyte membranes, 274-275 endocytosis in MDCK cells, 268 in endosomes, 268 entry mechanism, 262 envelope fusion in acidic media, 262 and cell fusion with MDCK cells, 263 fusion with cultured cells, 339 erythrocytes, 269, 339 liposomes, 272, 339 genome transfer by fusion with endosomes, 262

interaction with erythrocytes and MDCK cells, 268 mutants pH modification of fusion activity, 285 threshold pH and efficiency of fusion, 350 NA glycoprotein, molecular weight and function, 258 pH dependence of viral fusion, 270 variants inducing fusion at elevated pH, 346-347 X-31 strain, 269 Inositol phospholipids, breakdown, myoblast fusion and, 102, 104 Inositol 1,4,5-triphosphate, signal transduction and, 52-54 Inositol trisphosphate, Ca2' release from endoplasmic reticulum and, 124 Insulinoma cells, exocytotic fusion-fission, GTP-binding protein role, 152 Insulin secretion, pencreatic islets, chemiosmotic hypothesis-based predictions, 215-2 16 Intermediates, in virus membrane fusion, 285 lntramembrane particle arrays, 171-173 acrosomal membranes, 16-17 clustering, in hemagglutinating virus of Japan-mediated fusion, 266, 323-325 formation, 182 glycerol dehydration and, 173-175 hemagglutinating virus of Japanmediated fusion, 265 influenza virus-induced, in erythrocyte membranes , 274-275 Iodide anion Class I, 208 use of blocker-sensitive anion transport site, 209 Ionic permeability, isolated chromaffin granules, 206-209 Isethionate, Class I1 anion, 208

K Kargagener's syndrome, spermatozoa-in, 18

L La Crosse virus cell entry route, 288

INDEX

GI glycoprotein, molecular weight and function. 258 G2 glycoprotein, molecular weight and function, 258 low pH-induced fusion activity, 278 pH dependence of viral fusion, 270 Lanthanum ion, effect on sperm-egg fusion, 33 a-Latrotoxin Ca'+-free effect, 156 exocytosis stimulation, 155-156 Leishmania mexicana. survival, phagosome-lysosome fusion effect. 245-246 Limulus amebocytes exocytosis, 177 plasma and granule membrane boundary during exocytosis, 183-185 Linoleic acids. effect on myoblast fusion, 98. 106 Lipid bilayers entrance of hydrophobic segment into, 279-284 interaction with HA2 primary sequence, 284 mobility, 180- 182 target for hydrophobic segment, 279 Lipidic particles, point defects and, 179-180 Lipids binding capacity of hemagglutinin mutants, 347-3.50 exchange, virus-cell fusion and, 264 intermixing, in virus-cell fusion, 264 role in myoblast fusion, 97-99, 101 targets for vesicular stomatitis virus, 277 Lipid-soluble ions, chromaffin granule pH and. 204-205 Liposomes ganglioside-containing, as hemagglutinin receptor, 271 hemagglutinating virus of Japan fusion, 264-265 influenza virus binding, 279 influenza virus fusion, 272, 339 interaction with influenza virus hemagglutinin, 271-272 vesicular stomatitis virus binding and fusion, 277 Lissamine rhodamine, label for secondary lysosomes, 236-237

375 Lucifer yellow, label for secondary lysosomes, 236-237 Lysin, in sperm-egg fusion, 34 Lysis CI ,ATP-induced in isolated chromaffin granules, 207-208 osmotic, chrornaffin granules. 206-209 Lysophosphatidylserine. effect on sperm-egg fusion, 33 Lysophospholipids. in sperm-egg fusion, 34 Lysosomal probes, fluorescent, set> speriJjc probe

Lysosomes constituents. differential delivery. 240-24 1 differential and sequential delivery of mobile impermeant fluors and tluoresceinated dextran, 241-242 free dye accumulation, 236 gelatinous structure in. 239 hydrosol formation, structure, and movement, 239-240 in macrophage antimicrobial activity. 245-246 nonionic hydrocolloids in, 238-242 pH. fusion-inhibiting qualities of polyanionics and, 247 primary accumulation of polyanionic substances and. 242 contents, intrusion into yeast phagosomes, 236 fusion inhibition and digestion of S. cereuisiue. 242 in yeast phagosornes, 236 secondary, 229 accumulation of polyanionics, 248 digestion, polyanionic substances in, 244 effects on nonionic water-dispersible polymers, 239 impermeant fluorescent labels. 236-238 intactness after phagosome-lysosome fusion, 236 Thorotrast transfer from, 230 in viral infection, 288 Lysosomotropic reagent, effect on viral replication, 287

376

INDEX

M Macrophages, nuclei, affinity for acridine orange, 233 Magnesium ion extracellular, in sperm-egg fusion, 29 phase separation of phospholipids and, 35 Mammals, sperm-egg fusion morphology, 11-13 Marsupialia, sperm-egg fusion, 11-13 Mast cells exocytosis, 122 intermixing of granule and plasma membrane constituents during fusion, 183 membrane recycling i n , 159 rapidly frozen, exocytosis. 175-177 small vesicles fused with granule membranes, 183 MDCK cells influenza virus envelope fusion and cell fusion, 263 growth in, 346-347 interaction, 268 X-3 I gene introduction, hemagglutinin expression, 344 Melittin, succinylated, pH-dependent fusion activity, 282-284 Membrane capacitance, 149 botulinum toxin effects, 125 electrophysiological approach, 141 endocytosis and, 158-159 exocytosis monitoring by, 117-1 18, 122 tetanus toxin effects, 125 Membrane fluidity inhibition of phagosome-lysosome fusion by polyanionics and, 247 membrane fusion and, 100-101 Membrane fusion in chemically fixed cells, 170-175 initiation at point defects, 179-180 microdomains and, 17 1- I73 Membrane perturbations, polyanioninduced fusion dysfunction and, 248 Membrane potential egg plasma membrane, 25 hemagglutinating virus of Japanmediated, 3 I5

Membrane retrieval, see Exo-endocytotic coupling Membranes pentalaminar and trilaminar structures, 170- 175 permeation by acridine orange, 232 Metabolic energy hypothesis, cortical exocytosis and, 71-72 Metal ions, role in myoblast fusion, 92-93 Metalloendoproteases, inhibition of myoblast fusion, 104 Me thy lamine accumulation in permeabilized chromaffin cells, 216-218 effect on veratridine-induced catecholamine secretion in chromaffin cells, 216-219 Mg-ATP, role in exocytosis, 126-129 Microbicidal activity, role of phagosome-lysosome fusion, 244-246 Microdomains, membrane fusion and, 171 - 173 Microscopic assay, for cortical vesicle exocytosis, 64 MME cells, X-31 gene introduction, hemagglutinin expression, 344 MMTV virus. see Mouse mammary tumor virus Mobility acridine orange, 233-234 sulfonated fluors, 238 Monensin chromaffin cell exposure in Na+- or K+-containing media, granule lysis. 2 19-220 effect on methylamine accumulation in permeabilized chromaffin cells, 216-2 I7 Monoclonal antibodies, surface changes during myoblast fusion, 96 Monosaccharides, inhibition of hemagglutinating virus of Japan-induced cell fusion, 266 Monotremata, sperm-egg fusion. 11-13 Motility, spermatozoa, sperm-egg fusion and, 18-19 Mouse hepatitis virus 90A glycoprotein, molecular weight and function, 258 E2 glycoprotein, molecular weight and function, 258

377

INDEX Mouse mammary tumor virus cell entry uncoating route, 288 gp36 glycoprotein amino-terminal segment. 259 conformational change, 28 1 free energies for transfer. 280 molecular weight and function. 258 gp52 glycoprotein, molecular weight and function. 258 low pH-induced fusion activity, 278 pH dependence of viral fusion, 270 Mucopolysaccharides, in sea urchin cortical vesicles, 50-51 Muscarinic receptors, activation. Ca?' and, 150 Mutagenesis site-directed, of hemagglutinin fusion peptide, 347-35 1 site-specific, of cloned hemagglutinin genes, 346 Mycobucr~~riitrn ruhercitkosis. sulfatides 228 Myconectin, synthesis before myoblast fusion. 95-96 Myoball cell cultures, 89 Myoblast fusion aggregation and adhesion stages, 92 Ca2+dependence, 89. 91-93 cell specificity, 89-91 lineage and, 89-91 lipids and, 97-99 mechanisms bilayer structure changes, 101-102 Ca'+-stimulated proteolysis, 103I04 I 2-diacylglycerol effects, 102-103 hydration repulsion and, 102- 103 protein phosphorylation role, 104 membrane changes, 105 membrane fluidity and, 100-101 precursor cell heterogeneity, 89-91 primary cultures. 88-89 CaZ+dependence, 89 cell lines, 89 fibroblast reduction, 88 myoball cultures. 89 receptor-stimulated, 106 recognition stage. 91-92 role of proteins, 94-96 stimulation, 106-017 temperature effects, I00 I

time dependence. 94 ultrastructural changes, 99-100 Myoblast plasma membranes antibody studies, 98-97 glycoproteins. 94-95 lipids, 97-99 protein phosphorylation and, 96 Myristic acid, effect on myoblast fusion, 98

N Neiinrhes j u p o n i w . sperm-egg fusion morphology. 6-8 Neuromuscular junction exocytosis acetylcholine release, 116-1 17, 145 Ca2' role, 147 effect of protein kinase C activators. I50 electrophysiological approach, 141 inhibition by botulinum toxin. 154 a-latrotoxin effect, 155-156 endocytosis Ca2+ and, I60 models, 157 Neurosecretory vesicles, intragranular pH, 210 Neurotransmitters acetylcholine release, 143 conventional. I16 peptide. 116. 143 Neutrophils, lysosomal enzyme release. inhibition by 4-acetamido-4'isothiocyanostilbene 2.2'-disulfonic acid, 215 Nigericin chromaffin granule lysis mediated by. 208 effect on veratridine-induced catecholamine secretion in chromaffin cells. 216-219 NlH-3T3 cells bovine papilloma virus-transformed, hemagglutinin expression, 342 cell surface location of hemagglutinin synthesized in, 344 hemagglutinins produced from SV40-HA and BPV-HA vectors in, 344 X-3 I gene introduction, hemagglutinin expression, 344

378

INDEX

Nuclear magnetic resonance spectroscopy, I3C-, protonation of carboxyl groups, 282-284 Nucleocapsid strands, hemagglutinating virus of Japan, 265 0

Octadecyl rhodamine B chloride, 264 Oleic acid, effect on myoblast fusion, 98 Oligonucleotides, delivery into cytoplasm, 344 Omega toxin, effect on exocytosis, 124 Osmotic effects in secretion intact cells chromaffin cells, 21 1-214 neutrophils, human, 2 I5 pancreatic islets, 215-216 parathyroid cells, dissociated, 214-215 platelets, human, 214 permeabilized chromaffin cells, 216-2 19 Osmotic forces cortical exocytosis and, 70-71, 75 role in fusion process, 185-186 Osmotic properties isolated chromaffin granules electrochemical proton gradient, 205 ionic permeabilities, 206-209 osmotic lysis, 206-209 proton-translocating ATPase, 204-205 Ouabain, effect on sperm-egg fusion, 33 Ovoperoxidase assay for cortical vesicle exocytosis, 63 in sea urchin cortical vesicles. 50

P Pancreatic acinar cells, exocytosis, 150 Pancreatic islets insulin secretion, chemiosmotic hypothesis-based predictions, 21 5-216 insulin secretory granules, intragranular pH, 209-210 P arumecium, trichocyst discharge, 186- 187 Paramyxovirus, fusion with cell surface plasma membrane, 286 Paramyxovirus FI, glycoproteins, hydrophobic segments, 260

Parathyroid cells, parathyroid hormone secretion, inhibition by 4-acetamido-4’-isothiocyanostilbene 2,2’-disulfonic acid and probenecid, 214-2 I5 Parotid cells acinar cells, regulation by CAMP, 147, 153 CAMPeffect on exocytosis, 150 Parotid gland, intragranular pH, 210 Particle-free patches in acrosome-reacted spermatozoa, 17 in myoblast membranes at fusion sites, 99 in sperm membrane, 34 Patch clamp technique, 149 Pathogenicity, influenza virus, 353 pBVI-MTHA vector, 344 PC12 cells ATP requirement, 150 a-latrotoxin effect, 155 regulated exocytosis requirements, 150 Pentalaminar structure, of membranes, 170- 175 Permeability, ionic, isolated chromaffin granules, 206-209 Permeabilized cells detergent, 122 electropermeabilization, 122 techniques in secretion studies, 149 Pertussis toxin, substrate protein in sea urchin eggs, 74 PH acidic, protonation at, 281 conformational change in hemagglutinin, 272-274 effects on sperm-egg fusion, 27 elevated, influenza virus variant fusion induction at, 346-347 endosomal, lysosomotropic reagents and, 287-288 extracelluar medium, virus replication and, 287 fluorescein isothiocyanate in measurement of, 287 fusion activity dependent on HA2 N-terminal peptide, 282 HA2 peptide, 282 influenza virus, 340

INDEX succinylated melittin, 282-284 viral envelope, 262 virus membrane, 285-286 gradient. catecholamine release in permeabilized chromaffin cells and. 2 17-218 hemolytic activity of Semliki Forest virus and, 275-276 vesicular stomatitis virus, 276-277 intragranular insulin secretory granules, pancreatic islets, 209-210 isolated chromaffin cells. ADP effects. 206 isolated chromaffin granules, 205 neurosecretory vesicles, 210 parotid gland, 210 porcine anterior pituitary granules, 2 10 low conformational change induced in hemagglutinin. 352-353 induced fusion in CV-I cells, 359 La Crosse virus fusion activity and. 27N mouse mammary tumor virus fusion activity and. 278 lysosomal. polyanionics and, 247 range required for hemagglutinating virus of Japan-mediated cell-to-cell fusion, 308-309 threshold, influenza virus mutants. 350 Phagosome-lysosome fusion fluorescent lysosomal probes acridine orange. 23 1-236 artifacts, 234-235 sulfonated fluors, 236-238 illusion of inhibited fusion of secondary lysosomes, 236 inhibition, Ca" and. 247-248 intracellular digestion and, 242-244 microbicidal activities and. 244-246 promotion by weak bases, 246 Phagosomes acridine orange transfer to, 229-230, 245 dye transfer to. 240 sulfonated Ruor uptake, 238 Thorotrast transfer to, 229-230 yeast digestion without marker, 242-243 intrusion of lysosomal contents in, 236

379 Phase separation, phospholipids. Ca" and M g " roles. 35 Phenothiazines, inhibition of cortical exocytosis. 65 Phenylalklamines. calcium antagonists, I23 Phorbol esters protein kinase C substrate, 12X sen3itivity of exocytotic systems to. 132 Phosphate anion Class 11, 208 permeability, 207 Phosphatidylinositol. intracellular signaling and, 52 Phosphatidylinositol 4.5-bisphosphate, intracellular signaling and, 52-54 Phosphatidylinositol 4-phosphate. intracellular signaling and, 52 Phospholipase. in sperm-egg fusion. 34 Phospholipase A. effect on myoblast fusion in culture. 98 Phospholipase A*. role in cortical exocytosis. 73 Phospholipase C effect on myoblast fusion in culture. 98 GTP-stimulated, 124 phosphatidylinositol-specific,52-53, 74 role in cortical exocytosis, 73 Phospholipids, phase separation. Ca" and Mg'+ roles. 35 Pinosome fusion, polyanionics and. 248-249 PK cells, X-31 gene introduction. hemagglutinin expression, 344 Plasma membranes cholesterol distribution during exocytosis. 185 and granule membrane constituents, intermixing during fusion, 183 intramembrane particle-free areas, 171 mobility before and during fusion. 183-185 Platelets granule membrane. proton-translocating ATPase, 2 I 1 5-hydroxytryptamine secretion, 214 regulated exocytosis, ATP and CaLf requirements, 149 Pneumovirus FI glycoproteins, hydrophobic segments. 260

INDEX

Point defects, membrane fusion initiation, 179- 180 Poliovirus, cell entry route, 288 Polyanionic substances accumulation as hydrosols in lysosomes, 23 I by primary lysosomes, 242 antagonism with weak bases, 246-247 within lysosomes. affinity for acridine orange, 233 polyanionics hypothesis, history, 22 8-230 in secondary lysosomes, digestion and, 242-244 Polycations, inhibition of cortical exocytosis, 65 Polyerythrocytes, spherical, formation, 266 Polyglutamic acid, acridine orange accumulation in macrophage lysosomes and, 232-233 Pol ykaryons influenza virus mutants, 350 spherical, formation, 266 Polylysine, inhibition of cortical exocytosis, 65 Polymers hydrosol formation in water, 239-240 nonionic water-dispersible, behavior, 239-240 Polymorphonuclear leukocytes, azurophilic granules and secretory vesicles, 142-143 Polymyxin 8 , protein kinase C inhibition, 130 Polyphosphoinositide phosphodiesterase, see Phospholipase C, phosphatidylinositol-specific Polyphosphoinositides, hydrolysis, 147- I48 Polyspermy block saccharide residue changes and, 31 sialic acid changes and, 31 Posttranslational cleavage hemagglutinin precursor, 339, 353-359 virus fusion proteins, 258 Potassium ion, extracellular, sperm-egg fusion and, 29-30 Precursor cells fusion proteins, 258-259

hemagglutinin myogenic, 89-90 Prefusion events, sperm-egg fusion, 26 Probenecid inhibition of anion transport in erythrocyte membranes, 208 5-hydroxytryptamine secretion, 214 parathyroid hormone secret ion, 2 14-2 I 5 veratridine-induced secretion in chromaffin cells, 212 Procrustean bed interpretation, 23 I , 236, 239 Pronase effect on sperm-egg fusion, 3 1-32 inhibition of cortical exocytosis, 65 Protease sensitivity, hemagglutinin mutants, 347-350 Proteinase inhibitors, sperm-egg fusion and, 32 Protein kinase C activators, effect on exocytosis, 150 Ca2+ requirement, 129-130 exocytosis and, 129-130 hydrolysis of polyphosphoinositides and, I49 inhibitors, 130- I32 myoblast fusion and, 104 plasma membrane association. 133I34 substrate for phorbol esters, 128 Protein phosphorylation in exocytosis, 126-127 exocytosis modulation and, 153 in membrane fusion, 96, 104 Proteoliasin, in sea urchin cortical vesicles, 50 Proteol ysis in membrane fusion, 103 in myoblast fusion, 104-105 Protonation acidic residues in hydrophobic segments, 28 1 carboxyl groups, 282-284 Proton gradient across chromaffin granule membrane, 205 cultured chromaffin cells, 213 Proton pump, activation, 127-128

INDEX

381

Proton-translocating AI'PdSe bovine anterior pituitary gland. 210 insulin secretory granules, pancreatic islets. 209-210 isolated chromdffin granules, characterization. 204-205 neurosecretory vesicles, 2 10 in platelet granule membrane, 21 1 Protozoa. rapidly frozen, exocytosis, 177-179 Pyridoxal phosphate inhibition of anion transport in erythrocyte membranes. 208 chromaffin granule ATPase, 209 5-hydroxytryptamine secretion. 214 veratridine-induced secretion in chromaffin cells, 212

Q Quick freezing, see Rapid freezing Quin2 Ca" measurement. 149 microinjection, in sea urchin eggs, 51-52 role of Ca?' in stimulus-secretion coupling and, 120

R Rabies virus, pH dependence of viral fusion, 270 Radioimmunoassay s bovine papilloma virus-transformed cells expressing hemagglutinin, 342 surface changes during myoblast fusion. 97 Rapid freezing artifacts, 173-175 cells. exocytosis in, 175-189 Recognition reactions, in myoblast fusion, 91-92.94-95 Recombinant DNA genes encoding influenza virus hemagglutinin. 338 mechanism of membrane fusion and, 360 Regulated secretion, in exocytosis, 46-47 Retroviridae, oncovirus glycoprotein Types B and C, hydrophobic segment, 26 1 Rotary-shadow method, 189

S Saccharide residues, in egg membrane, 31 Suc,c.ho,omyces cerrvisicie. 229 behavior, phagosome-lysosome fusion and, 245 digestion, fusion inhibition and. 242 Sea urchin eggs cell surface complex preparations, 59 cortical lawn preparations. S9 cortical vesicles, 48-51 constituents, 50-51 discharge morphology. 59-62 fertilization envelope, 50 fertilization steps, 48 granule and plasma membrane constituents, intermixing during fusion, 183 hyperosmotic inhibition of exocytosis, 186 plasma membrane glycocalyx. extracellular, 50 vitelline layer, SO rapidly frozen, exocytosis. 177-179 secretory granule swelling. 220 sperm-egg fusion morphology. 4-6 swelling of cortical granules. 185 Secondary structure, influenza virus HA2 amino-terminal segment and, 284 Secretogranin 11, immunocytochemical identification, 143 Secretory vesicles Ca"-activated K' channels, opening, 220 polymorphonuclear leukocytes, 142-143 .sea urchin eggs. 48 triggered exocytosis and, I16 Semliki Forest virus E 1 glycoprotein, 276 free energies for transfer, 280 internal hydrophobic stretch, 259 internal segment, 281-282 molecular weight and function, 258 E2 glycoprotein. 276 molecular weight and function, 258 E3 glycoprotein, 276 molecular weight and function, 258 entry mechanism, 262 envelope fusion in acidic media, 262 with BHK-21 cells, 263

382 genome transfer by fusion with endosomes, 262 low pH-induced hemolytic activity, 275-276 pH dependence of viral fusion, 270 Sendai virus, see Hemagglutinating virus of Japan Serine protease, in sea urchin eggs, 50 SFV virus, see Semliki Forest virus Sialic acid, in egg membrane, 31 Signaling, intracellular G proteins and, 54-55 phosphoinositide cycle and. 52-54 Sindbis virus low pH-induced hemolytic activity, 276 pH dependence of viral fusion, 270 Smooth vesicles, virus uptake, 262, 286 Sonication, hemagglutinating virus of Japan. 265 Species specificity, sperm-egg fusion, 20-24 Spermatozoa acrosome-reacted, particle-free patches, 17 Ca2+ influx, 15 fish sperm head plasma membrane, 9 sturgeon, 9 fusion with egg, see Sperm-egg fusion in Kargagener's syndrome, 18 motility, sperm-egg fusion and, 18-19 plasma membrane freeze-fracture imaging before and after acrosome reaction, 16 intermingling with egg plasma membrane, 19-20 particle-free patches, 34 sperm-egg fusion sites and, 13-17 Sperm-egg fusion antimembrane antibodies and, 32 cytochalasin effects, 33 dithiothreitol effects, 33 effects of extracellular Ca2+,27-28 K ' , 29-30 Mg*+, 29 erythrosine B effects, 33 glycoprotein biosynthesis inhibitors and, 33-34 mechanism, 34-35

INDEX morphology amphibians, 10 birds, I 1 fishes, 8-10 mammals, 11-13 marine invertebrates, 6-8 sea urchins, 4-6 motility of spermatozoa and, 18-19 pH effects, 27 prefusion events, 26 proteinase inhibitors and, 32 proteolytic enzymes and, 31-32 serine proteinase inhibitors and, 32 specific sites egg plasma membrane, 17-18 sperm plasma membrane, 13-17 temperature effects, 26-27 Stearic acids, effect on myoblast fusion. 98 Stimulus-secretion coupling, 47-48 Sulfate anion Class 111, 208 permeability, 207 Sulfolipids, mycobacterial acridine orange accumulation in macrophage lysosomes and, 232-233 acridine orange complexing, 232-233 affinity for acridine orange, 233 phagosome-lysosome fusion and. 229 Sulfonated fluors differential transfer from lysosomes, 24 1-242 and fluoresceinated dextran, differential and sequential delivery, 241-242 labeling of dextran sulfate cells, transfer to phagosomes, 240 labels for macrophage secondary lysosomes, 236-238 Sulforhodamine, 236-237 Suramin acridine orange accumulation in macrophage lysosomes and, 232-233 inhibition of 5-hydroxytryptamine secretion, 214 phagosome-lysosome fusion and, 229 pretreatment effect on acridine orange transfer to phagosomes, 245 SV40-HA vectors, 343

INDEX

Swelling Ca”-induced. 220 cortical granules. IUS sea urchin egg secretory granules. 220 Synapsin I exocytosis and, 117, 143 membrane fusion-fission in exocytosis and. 152 Synaptic vesicles, depletion in endocytosis. I 6 0

T Temperature low. virus-cell interaction at. 312-314 dependence inHuenza virus fusion activity, 269 viral envelope fusion, 264 effects o n myoblast fusion, 100 Temporal dependence, myoblast fusion, 94 Tetanus toxin effect on capacitance. 125 exocytosis, 124 exocytosis inhibition. 153-514 in chromaffin cells, IS4 mechanism. 154 structure and binding. 153-154 Thiocyanate anion permeability, 207. 207 inhibition of Ca”-dependent catecholamine release, 2 16-218 Thorium oxide marker. cross-linking with adsorbed polyanionics, 239 Thorotrast labeling of dextran sulfate cells, nontransfer to phagosomes, 240 polyanion cells, marker traces. 242 transfer to phagosomes. 229-230 Thrombin. stimulation of 5-hydroxytryptarnine secretion. inhibition by anion transport blockers. 214 Transport, of hemagglutinin, 358 Trapping. physical, of colloid,ri I e I eclron microscopic markers. 240 Trichocyst discharge, in P~ircimc~c,irrm, 186- I87 TriHuoperazine inhibition of cortical exocytosis. 65 protein kinase C inhibition, 130

p-TriHuoromet hox yphen ylhydrazone effect on secretion from intact chromaffin cells, 212-213 from permeabilized chromafin cells, 216-217 inhibition of platelet secretion, 214 Trilaminar structure, of membranes, 170- I75 Tripnrrrsres grarlrillti. sperm-egg fusion morphology, 4-6 Trypsin effect on sperm-egg fusion, 32 inhibition of cortical exocytosis. 65 Tunicamycin, inhibition of myoblast fusion, 95

U

UDP-glucose. effect on sperm-egg fusion. 33 Ultrastructure, myoblast, fusion-induced changes, 99-100

V Valinomycin, K -induced lysis, isolated chromaffin granules, 208 Vectorial transport assay. immunofluorescence-based. 62 Vesicular stomatitis virus entry mechanism, 262 envelope fusion and hemolysis with trypsinized erythrocytes, 263 envelope fusion in acidic media, 262 fusion with liposomes. 277 G glycoprotein, 258 free energies for transfer. 280 internal hydrophobic stretch. 259 inlernal segment. 28 1-282 niolecular weight and function. 258 lipids as targets for, 277 low pH-induced hemolysis and cell fusion. 276-277 Vesicular structures, production by aldehyde fixation. 17.5 Vesiculovirus G . glycoproteins, hydrophobic segments, 261 Viral envelope fusion, pH dependence. 262 +

384

INDEX

gl ycoproteins fusion, 262 molecular weight and function, 258 Viral fusion, exocytosis mechanism and, 118-120

Viral glycoproteins, conformation change, 278-279

cross fertilization and, 22 domestic fowl, I 1 fish eggs, 8 mammals, I I in sea urchin egg, 4 Vitelline layer, sea urchin eggs, 50 VSV, see Vesicular stomatitis virus

Viral model, cortical exocytosis and, 72-73

Virus glycoproteins envelope fusion and, 258 posttranslational cleavage of fusion proteins, 258 Virus membrane fusion hydrophobic segment entrance into lipid bilayer, 279-284 free energy of transfer of peptide from aqueous to lipid bilayer phases,

W Western equine encephalitis virus, E glycoproteins, 276 West Nile virus cell entry route, 288 low pH-induced hemolytic activity, 276 pH dependence of viral fusion. 270

Y

279-280

lipid bilayer domain as target, 279 primary sequence and hydrophobicity, 284

hydrophobic segment pH-dependent activity H A 2 N-terminal peptide, pH-dependent activity, 282 H A 2 peptide, 282 succinylated melittin, 282-284 intermediates, 285 mechanism membrane perturbation, 284-285 pH dependence, 285-286 Vitelline envelope amphibian egg, 10 barriers to excess spermatozoa and, 30 cortical granule material effect, 25

Yeasts affinity for acridine orange, 233 digestion by macrophages, 244 in phagosornes without marker, 242-243

marker delivery, phagosome-lysosome fusion and, 248 phagosomes, intrusion of lysosomal contents in, 236 Yellow fever virus low pH-induced hemolytic activity, 276 pH dependence of viral fusion, 270 Z

Zinc ion, effect on sperm-egg fusion, 33