Current Topics in Membranes and Transport VOLUME 30
Cell Volume Control: Fundamental and Comparative Aspects in Animal...
4 downloads
469 Views
15MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Current Topics in Membranes and Transport VOLUME 30
Cell Volume Control: Fundamental and Comparative Aspects in Animal Cells
Advlsoy Board
G . Blobel E . Carufili J . S. Cook
D . Louvard
Current Topics in Membranes and Transport Edited by
Arnost Kleinzeller Department of Physiology Universiry of Pennsylvania School of Medicine Philadelphia, Pennsylvania
VOLUME 30
Cell Volume Control: Fundamental and Comparative Aspects in Animal Cells Guesr Editors
R. Gilles
Arnost Kleinzeller
Loboratory of Animal Physiology Institute of Zoology Universiry of LiPge LiPge, Belgium
Department of Physiology University of Pennsylvania School of Medicine Philadelphia, Pennsylvania
L. Bolls Department of General Biology University of Milan Milan, Italy
ACADEMIC PRESS, INC. Harcourt Brace Jovanovich, Publishers
San Diego New York Berkeley Boston London Sydney Tokyo Toronto
COPYRIGHT (B 1987 BY ACADEMIC PRESS. 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 Slxth Avenue, San Diego, California 92101
United Kingdom Edition published by
ACADEMIC PRESS INC. (LONDON) LTD. 24-28 Oval Road. London NWI 7DX
LIBRARY OF CONGRESS CATALOG C A R D N U M B E R :
ISBN 0-12-153330-1
(alk. paper)
PRINTED IN THE UNITED STATES OF AMERICA 87 88 89 90
9 8 7 6 5 4 3 2 I
70-1 17091
Preface, ix Yale Membrane Transport Processes Volumes, XI
PART 1.
VOLUME CONTROL IN ISOSMOTIC CONDITIONS
Volume Malntenance In lsosmotlc Condltlons ANTHONY D. C. MACKNIGHT 1. Introduction, 3 11. Experimental Approaches to Study Cellular Volume, 4 111. Questions of Cell Organization. 11 IV. The Magnitude of the Osmotic Forces Generated by Impermeant Cellular Solutes, 16 V. What Are the Cellular lmpermeant Solutes?, 20 VI. How Is the Cellular Osmotic Swelling Force Offset?, 21 VII. What Limits Cellular Swelling?, 31 VIII. The Role of Medium Anions in Cell Volume Regulation, 32 IX. Summary, 37 References, 38
Role of Cytoplarmlc Veslcles In Volume Malntenance G. D. V. VAN ROSSUM, M. A. RUSSO, AND J. C. SCHISSELBAUER I. Introduction, 45 11. 111. IV. V.
Examples of Water Movement and Compartmentalization, 47 Volume Regulation in Vertebrates. 51 Ion Transport by lntracellular Membranes, 68 Conclusion, 69 References. 7 I
The Cell Cytocrkeleton: Possible Role In Volume Control JOHN W. MILLS I. Introduction, 75 11. Cell Shape and the Cytoskeleton. 76
vi
CONTENTS
111. IV. V. VI. VI1.
Membrane-Cytoskeletal Interactions, 77 Relationship between Cell Volume and Cell Shape, 78 Possible Role of the Cytoskeleton in the Mechanism of Cell Volume Control, 80 Cytoskeleton and Cell Volume in MDCK Cells, 85 Summary, 96 References, 96
PART 11.
VOLUME CONTROL IN ANISOSMOTIC CONDITIONS
Volume Regulation In Epithelia MIKAEL LARSON AND KENNETH R. SPRING I.
Introduction, 105
11. Basic Principles for the Maintenance of Epithelial Cell Volume during Steady-State
111. IV . V. VI . VII . VIII .
Conditions, 106 Transepithelial NaCl Transport, 108 Response of Epithelial Cell Volume to Osmotic Perturbations, 110 Methods for Measuring Cell Volume Changes, 1 11 Volume Regulatory Increase: Necrurus Gallbladder Epithelium, 1 12 Volume Regulatory Decrease, 1 15 Physiological Significance of Volume Regulation, 120 References, I20
Volume Regulation In Cultured Cells ELSE K. HOFFMANN 1. Introduction, 125 11. Classification of the Ion Transport Mechanisms Involved in Volume Regulation, I28 Ill. Regulatory Volume Increase (RVI), 131 IV. Regulatory Volume Decrease (RVD), 144 V. Summary and Perspectives, 166
References, 172
Cell Volume Regulation In Lower Vertebrates LEON GOLDSTEIN AND ARNOST KLEINZELLER I. Evolutionary Considerations, 181 11. Environmental Considerations, 182
vii
CONTENTS
Ill. IV. V. VI. VII.
Cell Solute Composition, 183 Changes in Cell Composition during Osmotic Stress, 187 Extracellular Fluid Regulation and Cell Volume Control, 189 Role of Intracellular Osmolytes in the Mechanism of Cell Volume Regulation, 191 Cell Volume Maintenance and the Role of the Cytoskeleton, 197 References, 201
Volume Regulation In Cells of Euryhallne Invertebrates
R. GiLLES 1. Introduction, 205 11. Coping with Changes in Membrane Tension and Cell Volume: Cell Volume Control in Anisosmotic Media, 206 111. Coping with Changes in Ion Content: Modification in Structure and Activity of Macromolecular Components, 23 1 1V. Conclusions, 235 References, 236
PART 111.
PHYSICOCHEMICAL PROSPECTIVES
Non-Donnan Effects of Organic Osmolytes In Cell Volume Changes MARY E. CLARK I. Introduction, 251 Donnan Theory as a Model, 252 Ill. Perturbing and Stabilizing Solutes, 259 IV. Mechanisms of Action of Stabilizing Solutes, 262 V. Conclusions, 269 References, 270 11.
Index, 273 Contents of Recent Volumes, 283
This Page Intentionally Left Blank
Preface The ability for volume control must be considered a fundamental property of living cells. Basically, cells can indeed be viewed as complex chemical machineries in which localization and concentration of the various interacting molecular species must be carefully controlled in order to maintain optimal activity. A priori, cell volume control results from a balance between three major components (see figure): the amount of osmotically active water; the amount of osmotically active solutes; the elastic properties of cells and tissues and their structural resistance to changes in volume. Structural resistance to changes in volume is related to the architectural characteristics of the plasma membrane and of the extra- and intracellular matrix. Control of the amount of osmotically active water can be related to fluxes, binding, and/or sequestration. Control of the amount of intracellular solutes can be achieved by fluxes, binding, andlor metabolism, depending on the type of osmotic effector considered. The knowledge accumulated over the past twenty-five years on cell volume control deals chiefly with regulation of the fluxes and concentrations of a few major intracellular osmotic effectors: the inorganic ions K , Na , and C1- in vertebrate cells, the free amino acids, and some quaternary ammonium derivatives in invertebrate cells. This volume will, however, consider, whenever possible, some of the other mechanisms quoted above. It will also consider, to some extent, interactions between intracellular solutes of different concentrations and the structure of various membrane and intracellular macromolecules. Results of comparative studies using species that can normally withstand large changes in blood osmolality and ion concentration indicate that the processes at work in both +
+
SOLUTES FLUXES- BINDING SEQUESTRATION
FLUXES-BINDING METABOLISM
CELL STRUCTURAL RESISTANCE AND ELASTICITY* The cell's component of volume maintenance and regulation.
ix
X
PREFACE
volume regulation and in macromolecular structure stabilization are closely related and cannot be dissociated. Structure destabilization of membrane-related proteins by ions could indeed be among the trigger mechanisms for changes in solute fluxes. Changes in enzyme activity due to the structural modifications by ions are implicated in the control of the amount of different amino compounds important as osmotic effectors in volume regulation and as “compensatory solutes” acting against protein destabilization. R. GILLES
Yale Membrane Transport Processes Volumes Joseph F. Hoffman (ed.). (1978). “Membrane Transport Processes,” Vol. 1. Raven, New York. Daniel C. Tosteson, Yu. A. Ovchinnikov, and Ramon Latorre (eds.). (1978). “Membrane Transport Processes,” VoI. 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 Current Topics in Membranes and Transport (F. Bronner and A. Kleinzeller, eds.). Academic Press, New York. William H. Miller (ed.). (1981). “Molecular Mechanisms of Photoreceptor Transduction”: Volume 15 of Current 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 and Transport (A. Kleinzeller and F. Bronner, eds.). Academic Press, New York. Joseph F. Hoffman and Bliss Forbush I11 (eds.). (1983). “Structure, Mechanism, and Function of the Na/K Pump”: Volume 19 of Current 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 Transport (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. 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.). (1987). “Potassium Transport: Physiology and Pathophysiology”: Volume 28 of Current Topics in Membranes and Transport (F. Bronner and A. Kleinzeller, eds.). Academic Press, Orlando. xi
This Page Intentionally Left Blank
Part I
Volume Control in lsosmotic Conditions
This Page Intentionally Left Blank
CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 30
Volume Maintenance in lsosmotic Conditions ANTHONY D . C . MACKNIGHT Department of Physiology University of Oiago Medical School Dunedin. New Zealand
1.
INTRODUCTION
Cellular volume remains constant under steady state conditions. In vertebrates, this constancy, in part, reflects the maintenance of a stable extracellular environment. It also requires that cellular composition be held constant. Since water crosses most plasma membranes readily, we can write simply, cell osmoles/cell water content = extracellular osmolality. Discussion of the mechanisms by which extracellular fluid osmolality and volume are regulated is outside the scope of this article. Recent reviews provide an introduction to these topics (Anderson et af., 1984; Fitzsimons, 1985; Robertson, 1985). Broadly speaking, cellular osmoles may belong to one of two classes: impermeant solutes, which can not cross the plasma membranes, and diffusible, permeant solutes which can move by a variety of means between the extracellular and cellular compartments. The cellular content of impermeant solutes will depend upon the balance between anabolism and catabolism, and will thus reflect the metabolic status of the cells. The distribution of the diffusible solutes will be determined by requirements of electroneutrality, by the balance of the forces (principally chemical and electrical) which act on the solutes, and by the ease with which they can cross the membrane. Inasmuch as energy is required to move solutes across the plasma membrane against their electrochemical potential gradients, the distribution of some diffusible solutes will also reflect the metabolic status of the cells. The aim of this article is to examine aspects of the regulation of cellular volume only under isosmotic conditions. Fuller discussions of many of the issues 3
Copyright 0 1987 by Academic FTcrs. Inc. All nghls of repduction in any form reserved.
4
ANTHONY D. C. MACKNIGHT
raised here and discussion of volume regulation in anisosmotic media will be found in other articles in this volume as well as in reviews (Cala, 1983; Grinstein er al., 1984; Hoffmann, 1977; Kregenow, 1981; Macknight, 1983; Macknight and Leaf, 1985; Rink, 1984; Siebens, 1985) and in the symposium organized by Finn (1985).
II. EXPERIMENTAL APPROACHES TO STUDY CELLULAR VOLUME Cellular volume may be estimated directly by microscopy. Determination of absolute volumes in fixed specimens is complicated by the problem of possible alterations in volume during fixation and processing. Though modem techniques are designed to minimize volume changes, this approach is best suited to detecting changes in volume in a preparation induced by different experimental protocols rather than to determining absolute volumes. Direct observation of living tissues and cells during incubation, using conventional light microscopy or phase contrast or Nomarski optics, has provided important information about volume regulation, however, particularly in relation to the responses of cells to anisosmotic media (Grantham er al., 1981; Gonzilez et al., 1982; Kirk er al., 1984; MacRobbie and Ussing, 1961; Spring and Hope, 1978). However, the methods are not easily applied to all cells and provide no direct information about any changes in cell solutes which may be associated with the changes in volume. Volumes of single cells may be estimated simply and reliably by electronic sizing techniques such as Coulter counting. Unfortunately, this technique is not applicable to cells in organs and also provides no direct information about cell solutes. In many studies of cellular volume, water contents rather than volumes are measured. Since some 70 to 80% of the mass of most cells is water, provided that solid matter in the cells remains constant during the course of the experiment, changes in water content can be regarded as synonymous with changes in volume. In this article the two terms will be used interchangeably. The major problem of studies in which water contents are measured by chemical techniques, and one which remains to be resolved, is to decide upon the most suitable means for determining the size of the extracellular space. The problems in selecting a suitable marker are not trivial (Law, 1982). While inulin probably remains the marker of choice (McIver and Macknight, 1974), and noninulin space water appears to approximate cell water quite closely, there are unresolved problems in deciding what fractions of total sodium and chloride to assign to the cellular compartment (Macknight, 1980). The best indication of the suitability of an extracellular marker is provided by a comparison of the steady state cellular potassium content determined from chemical analysis (which is not affected by variations in measured extracellular space)
5
ISOSMOTIC VOLUME MAINTENANCE
with the derived cellular water content. For a variety of tissues from amphibia and mammals, we have found in our laboratory the relationships shown in Fig. 1. There is a direct correlation between potassium content and noninulin space water content such that, despite a range of potassium contents from some 250 to 450 mmol/kg dry wt, derived cell potassium concentrations are comparable and approximate 140 mmol/liter cell water. Derived cell sodium and chloride concentrations based upon chemical analysis are also relatively similar in the different tissues from a variety of species (Fig. 2), but are, in general, higher than those predicted from the results of alternative techniques, particularly microelectrode studies, as discussed later. To study cellular volume regulation under isosmotic conditions, preparations are required in which both water and ions are easily measured in a variety of experimental situations. There is no one ideal preparation. Experimentation in vivo is necessarily limited to conditions which animals can tolerate, though much important information (for example, about how cells respond to anisosmotic conditions) has been obtained from such studies. For many types of experiment, such as those in which drugs or metabolic inhibitors are used, or in which the effects of different medium solutes are examined, in vitro systems are required. One unresolved problem with in vitro incubation is that tissue water, sodium, and chloride contents may increase rapidly (within minutes) when the tissue is first incubated in balanced isosmotic extracellular-like ionic media. For example, in unpublished experiments (Tarbotton, Leader, and Macknight), we have found
zkol,’$ P-g 300 U
+
,,,,,,,
200 I ’
toad bladder rabbit colon A Necturus gallbladder rat renal cortex V mouse diaphragm
+
,
OO
1.0 2.0 3.0 noninulin space H,O content
4.0
FIG. I . Relationship between cell potassium content and noninulin space water content in a variety of tissues. Symbols represent mean values obtained in experiments in our laboratory.
6
ANTHONY D. C. MACKNIGHT
0toad urinary bladder
rabbit descending colon
0rat renal cortex
Necturus gallbladder
muse diaphragm
U
Na conc.
K
Cl conc.
COM.
FIG. 2. Noninulin space ion concentrations in a variety of tissues [derived by dividing noninulin space ion contents (mmollkg dry wt) by noninulin space water contents (kg HnOlkg dry wt)]. Values were obtained in experiments in our laboratory.
that skeletal muscle in mouse diaphragm analyzed immediately upon removal from the animal has a tissue water content of 2.81 0.05 kg/kg dry wt and contains (in mmol/kg dry wt) sodium, 150 & 5, potassium, 382 & 7, and chloride, 105 f 5 (mean f SEM,n = 24). After incubation at 37°C in balanced isosmotic medium in vitro tissue water content has increased to 3.14 k 0.08 kglkg dry wt and tissue now contains sodium, 210 2 13, potassium, 377 12, and chloride, 174 f 9 (n = 48). Such tissue swelling is often interpreted as indicating cellular damage. Though this may sometimes be true, an alternative explanation deserves consideration. Cellular damage should result not only in uptake of sodium, chloride, and water but also in loss of potassium. Yet, tissue potassium content, more than 98% of which must be cellular, can remain remarkably constant. This suggests that the initial uptake of fluid may be confined to the extracellular compartment. Indeed such an uptake would be expected, for the interstitial fluid in vivo exists as an unsaturated gel (Aukland and Nicolaysen, 1981). It is maintained in this condition by the colloid osmotic force exerted across the capillaries by the plasma proteins with their counterions and by the system of lymphatic drainage. Neither system operates in vitro. Nor can the presence of proteins or other large molecular weight solutes in the medium in v i m substitute satisfactorily for the in vivo situation as, without the restrictive capillary endothelial barrier, such solutes will simply tend to diffuse throughout the interstitial space. Thus, in v i m , one should expect an isosmotic uptake of extracellular medium by the tissues until the interstitial gel becomes saturated.
*
*
ISOSMOTIC VOLUME MAINTENANCE
7
Such diffuse uptake of fluid would not result in any readily detectable histological changes, and attempts to identify swollen compartments have proved, at best, equivocal (for example, Sheff and Zacks, 1982). Furthermore, it would account for the increased size in vitro of the space into which a variety of markers distribute, and does not necessitate their cellular penetration. As an illustration, we can take the data for mouse diaphragm cited above and assume, in each case, that cell potassium concentration is some 150 mmol/liter. Then, both in vivo and in v i m , cell water would approximate 2.50 kg/kg dry wt. Thus, before incubation extracellular water would be 0.31 kg/kg dry wt, which represents an extracellular space of 11% of tissue water. After incubation for 60 min at 37"C, extracellular water is now 0.64 kg/kg dry wt, which represents an extracellular space of 20% of tissue water. Derived cell sodium and chloride contents are very similar for both situations (sodium, 104 and 114 mmol/kg dry wt; chloride, 71 and 90 mmol/kg dry wt). Much information about volume regulation has come from in v i m studies using tissue slices, most commonly renal cortical slices, an approach developed by Robinson (1949, 1950) and by Mudge (1951). Like all techniques, it has important limitations which must be kept in mind in designing appropriate experiments. It is instructive to compare the advantages and disadvantages for studying cell volume regulation of the major renal preparations (Table 1). The major advantages of the slice are its ease of preparation, handling, and analysis. Its major disadvantages relate to its relative thickness. Though occasional attempts are made to work with renal slices at 37°C our experience, even with relatively thin slices (about 0.2 mm), is that at this temperature some degree of tissue hypoxia is found. For example, direct comparison of slices at 25, 30, and 37°C shows that the higher temperatures are associated with lower potassium content and, at 37°C with some swelling (Table 11). Certainly, thicker slices (0.5 mm hick) are clearly hypoxic at 37°C (Balaban et al., 1980). Even renal tubular suspensions may be difficult to oxygenate adequately at this temperature (Burg and Orloff, 1966). These problems reflect the high metabolic rate of renal tissue. Thin preparations of other tissues are much more easily incubated in vitro at 37°C (for example, liver slices, Macknight et al., 1974, and skeletal muscle, as illustrated above). That renal cortical slices of 0.2 to 0.3 mm thickness can be oxygenated adequately at 25°C is illustrated in Fig. 3. An important and unresolved question in in vitro preparations of kidney is the extent to which solute movements occur across the apical plasma membranes. In the kidney slice, and in many preparations of isolated tubule suspensions, the lumina are collapsed. This must limit solute and water fluxes from lumen to cell and, therefore, affect the rates of transepithelial transport compared with the normal situation. However, by combining collagenase perfusion of the kidney in sifu with subsequent dispersion and purification, suspensions of tubules with open lumens have been obtained (Soltoff and Mandel, 1984). If, as seems true
8
ANTHONY D. C. MACKNIGHT TABLE I ADVANTAGESAND DISADVANTAGES OF DIFFERENT RENALPREPARATIONS Preparation
Slices
Suspensions of isolated tubules
Isolated individual tubules
Advantages 1. Readily prepared 2. Not exposed to proteolytic enzymes 3 . Obtainable from any species 4. Easily handled and analyzed 5 . Data obtained following brief periods (minutes) of incubation
I . Minimal diffusional delay between cells and medium 2. Can be used for compartmental analysis 3 . Can be incubated at 37°C 4. New methods provide preparations with patent proximal tubule lumens 1. Minimal diffusional delay between cells and medium Can be incubated at 37°C Direct visualization of changes in cell volume possible No enzymatic treatment required Luminal perfusion possible
Disadvantages 1. Some cell damage during
preparation 2. Satisfactory only at 25°C and then only if relatively thin ( c 0 . 3 mm thick) 3. Cellular heterogeneity and structural complexity preclude compartmental analyses and studies which require rapid equilibration of medium with interstitial fluid 4. Lumens of proximal tubules collapsed, restricting exchanges across apical membranes 1. Some cell damage during preparation 2. Absence of tubular basement membrane 3. Cellular heterogeneity 4. Less easily obtained for analysis of ions and water
I . Minute quantities of tissue make analysis of cell composition difficult 2. Lumens collapsed in absence of perfusion 3. Difficult to obtain from species other than rabbit
for other epithelia, basolateral membrane sodium permeability is low relative to apical membrane permeability, the extent of ouabain inhibition of oxygen consumptions should allow an indirect comparison of the sodium fluxes across the apical plasma membranes in different preparations. In suspensions of isolated rabbit tubules with open lumens incubated at 37"C, Balaban et al. (1980) reported a decrease in oxygen consumption of almost 70% with maximal doses of ouabain, whereas in rabbit renal cortical slices incubated at 25"C, both Whittam
9
ISOSMOTIC VOLUME MAINTENANCE
TABLE II COMPOSITION OP RAT RENALCORTICAL SLICESINCLIBATED x r 25, 30,
25°C (8) 30°C (7) 37°C (5)
2.79 k 0.05 2.86 +- 0.04 3.34 f 0.09
277 2 10 339 ? I S 472 -+ I 1
AND
*
291 5 236 k 10 176 k 8
37"CU
229 ? 8 283 f 13 387 2 10
a Slices (0.2 to 0.3 mm thick) were incubated in isosmotic sodium chloride Ringers, at the temperatures shown, for 60 min. Stimng and oxygenation were achieved by bubbling oxygen through the medium throughout the experiment. Data. which give mean k SEM for the numbers of slices shown in parentheses, are from Hughes and Macknight (1976). In mmollkg dry wt.
and Willis (1963) and Cooke (1979) found an inhibition of 40%. These data suggest, therefore, that, in the absence of ouabain, there is an appreciable transepithelial transport of sodium in renal cortical slices despite the absence of a patent lumen histologically. One very sensitive indicator of cell viability (and one which is not affected by problems of extracellular space measurement) is the steady state potassium con-
y: 0.012~ 9381
%oxygen salurotion
%oxygen soturotion
RAT RENAL CORTICAL SLICES
Fa. 3. Tissue composition and oxygen consumption (QoJ in rat renal cortical slices (0.2 to 0.3 mm thick) incubated at 25°C. Note that Qo, remains relatively constant at Po, greater than 60% but declines slowly below that. Nevertheless, at PO, greater than 40% tissue composition remains constant. Note also that the Qo, of rat renal tissue is about double that of rabbit. This study, therefore, provides a more critical test of the adequacy of tissue oxygenation at 25°C than would a study of rabbit renal cortical slices. (Data redrawn from Cooke, 1979.)
10
ANTHONY D. C. MACKNIGHT TABLE 111 COMPARISON OF DERIVED CELL ION A N D WATER CONTENTS FROM CHEMICAL ANALYSIS I N PREPARATIONS FROM RABBIT RENALCORTEX Temperature (“C)
H20 ( k g h dry wt)
Naa
Ka
CIU
25 25 28
2.08 2.14 2.36 2.078 1.73 2.31
86 I47 I94 I26 89 165
347 285 253 300 238 26 1
108 121 153 -
Slicesb Suspensionsc Closed lumensd Open lumens*
25f
Single tubulesh
37 25
132
In mmol/kg dry wt. Data from Cooke and Macknight (1984). Data from Burg er al. (1964). Data from Palliard et al. (1979). Data from Soltoff and Mandel (1984). f Recalculated using protein/dry wt ratio of 0.71 given in reference. 8 H20content at 25°C calculated on the assumption that gains of Na and K compared with 37°C were accompanied by isosmotic water uptake. Data from Burg er al. (1966).
tent. A comparison of potassium contents in different preparations (Table 111) reveals that the renal cortical slice incubated at 25°C is certainly as good in this regard as the other preparations incubated at a comparable temperature. Water, sodium, and chloride contents also compare favorably with the values reported for tubule suspensions and for single tubules. Every preparation has both advantages and disadvantages and much can still be learned from applying the simple slice technique to problems of volume regulation. However, in terms of the conventional “pump-leak’’ hypothesis (discussed later), a complete description of cellular volume regulation requires knowledge of (1) the contribution of impermeant solutes to cellular osmolarity and electroneutrality, (2) plasma membrane potentials, (3) cellular ion activities, (4) kinetics of the active transport processes, ( 5 ) plasma membrane ion conductances, and (6) the extent of involvement of co- and/or countertransport processes in solute fluxes. We have some information about steady state volume and ion composition, and about the effects on cell volume of metabolic inhibitors, drugs which block membrane transport processes, and perturbations of extracellular osmolality in a variety of epithelial and other cells. However, there is no mammalian preparation, as yet, which is sufficiently well studied for a complete characterization of volume regulation under isosmotic conditions to be provided.
ISOSMOTIC VOLUME MAINTENANCE
111.
11
QUESTIONS OF CELL ORGANIZATION
A. The Propertles of Cell Water The concept that water in cells has the properties of bulk water appears overly simplistic and many working in this field feel uncomfortable with the assumption. Cells contain dense networks of cytoskeletal elements, together with many membranes of organelles. The depressant effect on water activity of close proximity to surfaces suggest that cellular water may be, in part at least, in some way “bound” or immobilized within cells and have solvent properties which differ from those of bulk water. None would deny that the properties of water in the immediate vicinity of surfaces are altered. But the argument relates to whether “immediate” means but one or two layers (and therefore perhaps 5% or so of cell water) or refers to water over much larger distances from surfaces, with the bulk of cellular water affected. An article by Clegg (1984) contains an interesting discussion of this and related matters. Attempts to obtain experimental evidence about the properties of cell water have provided controversial results. Nuclear magnetic resonance (NMR) spectroscopy has been applied to investigate the state of cellular water but with conflicting conclusions. In their comprehensive review, Shporer and Civan (1977) argue that the most likely interpretation of the data is that only a small fraction (some 5%) of cellular water is bound. The remainder has the kinetic properties of bulk water but exchanges rapidly with the bound water. More recently, as described by Clegg ( 1984), an alternative technique-quasielastic neutron scattering (QNS)-has been applied to the study of water in Arfemiu cysts. Though results with this and other techniques suggest that water within the cysts differs substantially in properties from those of bulk water, the unusual nature of these cysts raises questions about the applicability of these results to vertebrate cells. An alternative approach of great interest is the intracellular reference-phase technique, results from which are reviewed by Horowitz and Miller (1984). A gelatin solution is injected into amphibian oocytes. Time is allowed for this gel to reach diffusion equilibrium. The oocyte is then frozen, and the gel and other portions of the cell removed by cryomicrodissectionand analyzed for solutes and water. Results so far have been taken to indicate that nuclear water is different in its structure from bulk water. Though cell water may be modified in its properties, nevertheless the results of many experiments are consistent with much of the cell water behaving as bulk water and this is often implicitly assumed in discussions of cellular volume regulation.
12
ANTHONY D. C. MACKNIGHT
6. Cell Ions Clearly, divalent and multivalent ions form complexes with cellular constituents and may be both bound and compartmentalized. However, the state of the univalent ions in cells and tissues continues to be debated today, just as it was in the early years of this century. For example Moore et al. (1912) wrote, “The varying concentrations of sodium, potassium, chlorine and phosphatic ions within and without the cell are an expression of specific affinities of the definite colloids of each particular cell-type for these ions, and do not mean that there is a membrane acting as a closed gate to these ions” and “sodium ions can readily be washed out of cells because the cell proteins have no affinity for them, while potassium ions are retained to the last.” Today, perhaps the most widely promulgated interpretation of this type is that propounded by Ling over the past 20 or so years (see Ling, 1984, for a full discussion). In his association-induction hypothesis, Ling argues that potassium is preferred by cells over sodium, not because of the transport and permeability properties of the plasma membrane, but because of potassium-selective sites on cellular macromolecules. It is more conventional, however, to assume either explicitly or implicitly that the univalent ions are in free solution in all of the cellular water, and to attribute the distributions of potassium and of sodium to the properties of the plasma membranes. One approach to answer this question is to measure total cellular ions and water by chemical and/or isotopic techniques and to compare derived concentrations with direct measurements of ion activities obtained with ion-sensitive microelectrodes. Table IV illustrates results from our laboratory in two tissues, Necturus gallbladder and rat skeletal muscle. There are two major problems with this approach. First, the interpretation of results of ion analyses in tissues is
TABLE IV RELATIONSHIPS BETWEEN CELLION ACTIVITIES AND ESTIMATED CELLION CONCENTRATIONS Cell concentrations0
Ion activities0 Tissue ~~
Na ~
Necrurus gallbladderb Rat skeletal musclec
K
CI
~~
13
7
Required activity coefficients
Na
K
CI
Na
K
c1
62 26
136 168
52 26
0.21 0.27
0.68 0.59
0.50 0.18
~
92 99
In mmol/liter. Data from Leader er al. (1981). Data from Leader ef al. (1984).
26
5
ISOSMOTIC VOLUME MAINTENANCE
13
complicated not only by the problems of dividing the total water between the cellular and extracellular spaces (discussed above) but also by the possibilities of extracellular ion binding or compartmentalization, or both. In the calculation of cellular ions from standard chemical and isotopic analyses, it is usual to assume that binding of univalent ions to components of the extracellular matrix and compartmentalization within some restricted interstitial space are both negligible. However, there is little direct evidence in favor of this view and many studies of connective tissues elements which argue against it. For example, polyanionic glycosaminoglycans are known to bind sodium (for a review of ions in relation to connective tissue elements see Comper and Laurent, 1978). In a study by Law (1984a), the possibility of interstitial ion binding in mammalian renal cortical slices has been examined. These were subjected to prolonged metabolic inhibition after exposure to chondroitinase ABC, which degrades a variety of sulfated glycosaminoglycans. Certainly tissue sodium content was decreased but it is difficult to be certain that the enzyme's effects remained localized solely to the extracellular compartment. Perhaps the best studied mammalian tissue in this regard is arterial wall, in which there is good evidence for the extracellular binding of cations (Villamil et al., 1968; Law, 1984b) and the possibility that anions are also bound extracellularly. Second, what are the correct activity coefficients for the conversion of measured cell ion activities to concentrations'?For potassium, where correction for extracellular ion is negligible, the use of activity coefficients close to that for free solution at the same concentration (0.76) gives reasonable agreement between the values obtained by the two techniques in both cell types (Table IV). This suggests that much, if not all, of the cellular potassium is in free solution and also that the bulk of the cell water is available as solvent for it. For both sodium and chloride, however, there are considerable discrepancies between measured activities and derived cellular concentrations which require quite different activity coefficients from those in free solution for their resolution. In the absence of data from alternative techniques such as electron microprobe analysis, applied to the same tissues under identical experimental conditions, it is not possible to decide whether these discrepancies reflect difficulties in estimating the extracellular quantities of sodium and chloride, actual differences in activity coefficients between cellular ions, or cellular binding or compartmentalization. The possibilities of cellular binding and compartmentalization have been widely canvassed and many attempts using a variety of tissue and techniques have been made to clarify this issue. Simple arithmetic has suggested to some that cellular cations are bound, for the sum of potassium plus sodium in the cellular compartment always exceeds that of the measured univalent anions. All that this reveals, however, is the presence in the cell of polyvalent anions. It says nothing about the state of the cations in relation to these.
14
ANTHONY D. C. MACKNIGHT
C. The Nature of Cytoplasm Much discussion treats the cytoplasm as a solution containing ions and neutral molecules in which are suspended the organelles. As knowledge of the complex and highly organized cytoskeleton grows (see Mills, this volume, for discussion of this in relation to cell volume), such a view may need revision. An alternative is to describe the cytoplasm as a gel, a view put forward in the early years of cellular physiology. Observations, such as the apparent lack of Brownian motion in the cytoplasm of living cells viewed by phase-contrast or Nomarski optics, and the gellike nature of cytoplasm extruded from cells such as the squid giant axon (Spyropoulos, 1977), lend support to the concept. One experimental approach has been to examine the behavior of cells exposed to osmotic gradients. However, it is difficult to interpret the results of such studies as favoring the gel rather than the solution hypothesis. Formally, in both models, the steady state activities of water in cell and medium must be identical, and Ap,, the chemical potential gradient for water, will be zero. Since Apw = Vw (AP - AT) (where V , is the partial molal volume of water, AP is the hydrostatic pressure gradient, and An the osmotic pressure gradient), in the steady state hp = A n . A volume flow, J v , can result either from a hydrostatic pressure gradient in the absence of an osmotic gradient, J V ~ , , ==~ Lphp
or from an osmotic gradient in the absence of a hydrostatic gradient, JvW,=, = LpdAn
where Lp is the hydraulic conductivity of the membrane, and Lpd is the coefficient of osmotic flow. The ratio -LpdlLp is the reflection coefficient, u. Using this, JV
=
Lp(AP - uhn)
(3)
For a semipermeable membrane across which only solvent flows, u = 1. Measured differences between volume flows induced by comparable hydrostatic and osmotic gradients will result when u < 1, i.e., Lpd < Lp, and are usually explained in terms of finite membrane permeability to one or more of the solutes in solution. The size of mammalian cells and the fragility of their plasma membranes have precluded studies in which cellular hydrostatic pressure is raised and consequent volume flows examined. Only osmotic gradients, produced by changing medium osmolality, have been employed to produce volume flows between cell and medium to study and assess the behavior of cell water. However, in squid, Spyropoulos (1979) and Vargas (1968) have made measurements of water movements between the inside and outside of internally perfused giant
ISOSMOTIC VOLUME MAINTENANCE
15
axons as functions of both hydrostatic and osmotic pressure gradients. The ratio LpdlLp was much less than 1. Spyropoulos (1979) argued that this discrepancy could not be accounted for by membrane permeability to the solutes used to raise the osmotic pressure of the medium. Together with results from other experiments, he interpreted these findings in terms of a cytoplasmic gel whose swelling is restrained by the nerve sheath. The detailed treatment of the effects of a gel on water activity is dependent upon the model chosen. Water activity will be affected by such factors as the degree of cross-linking in the gel, the nature of fixed ionic groups, and the counterions in solution and their concentrations. In the steady state water activities of medium and gel must be equal. The depression by solutes of medium water activity must therefore equate with the effects on gel water activity of dilution by the gel molecules, of the elastic deformation of the gel network, and of interactions between fixed charges on the network, as well as with the depression associated with dissolved solutes within the gel. In such a steady state, with the gel not restrained by any boundary which resists deformation, there will be no hydrostatic pressure gradient between cell and medium and the equilibrium volume of the gel will be achieved. However, should the gel be bounded by a barrier which prevents its swelling to its equilibrium volume, then a hydrostatic pressure-the swelling pressure of the gel-will be generated to raise gel water activity to equal that in the medium. The effects of the gel on water activity are described in terms which depend upon the particular model chosen. In essence, an identical view of the forces involved in cellular swelling is taken in the more common approach which treats the cellular contents as a mixed solution containing two components-the nondiffusible solutes (together with their counterions) which cannot penetrate the plasma membrane (sometimes referred to as colloids to reflect the fact that some, at least, are macromolecules) and the small molecular weight solutes to which the membrane is permeable. Here, A n consists of two terms, R n c ' and R ~ ~ U A C where P, R n c i is the total contribution of impermeant cellular solutes and R n u A c P the difference in osmotic pressure due to the diffusible solutes in cell and medium. (Since for impermeant solutes u = 1, it is omitted from the appropriate term). Thus, inserting these terms into Eq. (3), Jv = Lp(AP - R n c ' - R ~ u A c P )
(4)
In assessing this formulation with regard to the properties of the gel, discussed above, we can consider the osmotic term for the impermeant solutes as equivalent, in terms of effects on water activity, to the combination of factors of dilution of water by gel molecules, elastic deformation, and interactions of fixed charges. The term R n a A c P simply expresses in both situations the difference between medium and cells in the depression of water activity by permeant solutes. AP is formally equivalent to the swelling pressure of the gel and, like it,
16
ANTHONY D. C. MACKNIGHT
will only be seen if cellular swelling is restricted by a relatively indistensible membrane or sheath which can withstand sufficient tension without rupture. The simple formulation [Eq. (4)], given above, which has often been applied in discussing the regulation of cellular volume and the forces involved in flows of water between cells and medium is, therefore, equally valid whether the cytoplasm is viewed as a gel or a solution. Studies of the effects on cellular water contents of osmotic or hydrostatic gradients (where possible) cannot, of themselves, distinguish between these two possibilities. There may be situations, such as in the squid giant axon (Spyropoulos, 1979), mammalian renal tubules (Linshaw, 1980), and mammalian heart (Pine et al., 1981), where an external investiture of connective tissue components limits swelling, and it is also possible that cytoskeletal components may modify the behavior of the plasma membrane (Geiger, 1983). However, it is widely held that plasma membranes of animal cells cannot withstand hydrostatic pressure gradients to any extent (Harvey, 1954). Even an immeasurable difference of + 1 mOsm/liter water between cells and medium would require a cellular hydrostatic pressure of 2.6 Pa (19.3 mm Hg) above atmospheric and would generate a tension in the plasma membrane of some 16 dynes * cm-I in a spherical cell of radius 12.5 pm if uniformly distributed. It is generally accepted, therefore, that both cells and medium are at the same osmolality (Maffly and Leaf, 1959), though it must be realized that, given the experimental error and the fact that some 20 to 30% of most tissue water is in extracellular fluid, it is not possible to exclude a small difference in osmolality between cell and extracellular fluids. This then raises several questions. What is the magnitude of the osmotic forces generated by the impermeant cellular solutes? What are the cellular solutes which contribute toward it? How is this osmotic force offset so that cellular volume can remain constant? What limits cell swelling?
IV. THE MAGNITUDE OF THE OSMOTIC FORCES GENERATED BY IMPERMEANT CELLULAR SOLUTES As has been discussed, whether the impermeant solutes are in free solution or aggregated into a gel, fluid movements between cell and medium can be analyzed in terms of hydrostatic and osmotic forces. In the steady state, and in the absence of a hydrostatic gradient between cells and medium, osmotic equilibrium must exist and there must be electroneutrality in each solution. Thus (Nq + K, + Cli - + A," -) = extracellular mOsm/kg dry wt, and (Na, + K i + ) = (Clip + nA"-), where Nai+, K i + , and Clip represent cellularcontents (mmol/kg dry wt), A,"- represents the sum of the individual osmotic contributions of all individual impermeant anions (mmol/kg dry wt) and nA"- the total +
+
+
17
ISOSMOTIC VOLUME MAINTENANCE TABLE V CALCULATIONS OF CELL-IMPERMEANT SOLUTESFROM CHEMICAL ANALYSIS" H20
Kidney cortex Chloride medium Gluconate medium Mouse diaphragm Chloride medium Gluconate medium 0
(kg/kg dry wt)
Nab
Kb
Clb
Ab
&An-
nAn-IA
2.09 1.52
114 19
316 262
92 4
126 126
338 331
2.68 2.61
2.46 1.86
104
115
333 244
85 0
241 219
352 359
1.46 1.64
Data from unpublished observations of Scott, Tarbotton, Leader, and Macknight. In mmol/kg dry wt.
content of impermeant solute of valence n - (mmol/kg dry wt). The ratio nAin- /A," - approximates the average number of excess negative charges per impermeant ion. Unfortunately, even such seemingly straightforward calculations are bedeviled by our inability to decide what values to assign to the cellular permeant ions and what osmotic and activity coefficients to use to correct for the differences between concentrations and osmotic activities and ion activities, respectively. To illustrate this approach and the information to be gained from it, Table V shows data obtained in our laboratory from recent in vitro studies of rabbit renal cortical slices and mouse diaphragm. From this analysis it appears that skeletal muscle fibers contain about twice as much nondiffusible Osm/kg dry wt as do renal cortical cells but with about half the average unit charge. Inasmuch as osmotic and ionic activities will be less than, rather than greater than, 1, these estimates of impermeant anion contents and of average charge will underestimate the true values. Using the cell water contents to derive concentrations, and applying the van't Hoff equation, it is a simple matter to calculate that, if no other mechanisms were available, then the hydrostatic pressure required in the cell to offset the swelling which would result from the impermeant cell solute would be some 1.5 atm for kidney slices and 2.5 atm for skeletal muscle. This is a minimum value for it ignores the fact that the diffusible cell cations (sodium and potassium) in excess of chloride are balancing charge on impermeant cell solutes. If these ions are in solution as opposed to "bound," they will also contribute to the effective osmotic force exerted by the impermeant solutes. An alternative approach to estimate the magnitude of the osmotic effect of cellular impermeant solutes has been to offset swelling by incubating tissues in media containing relatively large-molecular-weightsolutes which should not be able to penetrate the plasma membrane. Such studies carried out with metabolisrn inhibited to prevent energy-dependent solute transport suffer from the crit-
18
ANTHONY D. C. MACKNIGHT
icism that cellular autolysis will break down impermeant solutes. If these autolytic products are retained in the cell, a greater content of impermeant solute will result. If they are lost from the cell, impermeant solute content will be underestimated. In addition, as metabolic inhibition progresses, the permeability of the plasma membrane may increase and the initially nonpermeant extracellular solute will gradually diffuse into the cell. Results obtained with this approach in metabolically inhibited tissues (reviewed by Macknight and Leaf, 1977) yield an estimate of some 20 mOsm/kg water for cellular colloid osmolality. This compares with the estimates in Table V of 60 mOsm/kg water for cortical slices and 98 mOsm/kg water for skeletal muscle. Rather than inhibiting metabolism, ouabain can be employed in metabolizing tissue to inhibit the sodium pump specifically. It should then be possible to determine the concentration of impermeant solute required to prevent subsequent swelling. However, in many preparations cellular swelling is very slow following pump inhibition (Macknight and Leaf, 1977) and this approach is often not feasible. Using isolated renal tubules exposed to collagenase to remove the basement membrane, which may limit cellular swelling in this preparation, Linshaw et al. (1977) and Linshaw and Stapleton (1978) determined the medium albumin concentration required to offset swelling in ouabain-treated tubules. Additionally, since the basement membrane was thought to limit swelling, with the consequent generation of a hydrostatic pressure within the cells and tubule, the applied hydrostatic pressure necessary within the tubular lumen to give the same increase in outer tubular diameter as occurred after ouabain treatment was determined. The estimates of cellular colloid osmotic pressure obtained by these two approaches were in reasonable agreement, giving pressures of 40 to 100 cm H20 (Linshaw et al., 1977), which corresponds to an osmolality of some 1.3 to 3.8 mOsm/kg H20. Whether estimates of the contribution of impermeant cellular solutes to cellular osmolality based on the addition of large-molecular-weight medium solutes are made from studies with nonmetabolizing or metabolizing tissues, the values are clearly well below those obtained from simple calculations. An alternative approach, which we are studying, is to replace all the medium permeant anion (chloride in our bicarbonate-free media) with the relatively small-molecularweight impermeant univalent anion gluconate. The major advantages of this approach are as follows. First, the absence of a permeant anion means that, in the steady state, cellular osmolality can be contributed to only by the impermeant solutes and their associated counterions. Once the metabolizing cells have equilibrated with the medium and lost their chloride (with associated cation), which takes some 15 to 30 min, they can neither gain nor lose ions. Thus they must behave as perfect osmometers, swelling and shrinking as predicted with changes in medium osmolality. Such is indeed the case, at least over the range studied (Fig. 4). It is, therefore, a simple matter to estimate the medium osmolality at
19
ISOSMOTIC VOLUME MAINTENANCE
renal cortical slices mouse diaphragm
$1 2
2.0
+ c aJ + c
e
,,’,_‘ ,,,,‘,/
1.0
Tcontrol/
Vexpt~
FIG. 4. The relationship between cell water content and relative medium osmolality ( R controll experimental) for rabbit renal cortical slices and mouse diaphragm incubated in gluconate medium. The arrows indicate the relative osmolalities required for a water content equal to that obtained in isosmotic sodium chloride medium. (Data from unpublished studies of Scott, Tarbotton, Leader, and Macknight.)
R
which cell volume is normal and this must equate with the osmotic contribution of the impermeant cell solutes. Second, and of particular importance, under these experimental conditions the sodium pump can play no role in volume regulation, for, whatever its rate of activity, total cellular solute and therefore volume cannot change. Its only effect can be on the ratio of cell potassium to cell sodium. Thus the calculated osmotic contribution of impermeant solutes is not in error because of sodium pump activity. Table V shows the effects of incubating both renal slices and skeletal muscle in chloride-free gluconate media. Of particular importance is the excellent agreement between the values derived for impermeant anions in chloride and in gluconate media. (This agreement, of course, reflects the fact that both types of cell lose cations with chloride as an isosmotic neutral electrolyte solution when they are equilibrated in the gluconate medium. If cells were gaining gluconate, the loss of chloride would not be accompanied by an equivalent loss of cation and the loss of water would be less than that predicted from the loss of chloride.) Thus we can feel some confidence in the estimates of cellular impermeant anions provided in the table. We can also calculate the medium osmolality at which cells in gluconate media will have the same volume as do cells in isosmotic sodium chloride media (osmolality 275 mOsm/kg H,O). This is 208 mOsm/kg H,O for renal slices and 219 mOsm/kg H,O for muscle. The arrows on the graph (Fig. 4) indicate the osmolality ratios (isosmotic/hyposmotic) required for these hyposmotic media to be isotonic.
20
ANTHONY D. C. MACKNIGHT
The values obtained for cellular impermeant solutes and associated counterions from this approach are clearly greater than those estimated from the addition of small concentrations of large-molecular-weight solutes to isosmotic chloride-containing media. The simplest explanation for this discrepancy is that tissues studied by the latter technique were not in a true steady state. In particular, the full colloid osmotic effect of the cellular impermeant solutes was not expressed. Neither sodium nor chloride has a reflection coefficient of zero across the plasma membrane, even in the complete absence of pump activity. Thus these ions will continue to exert a greater osmotic effect in the medium until they have equilibrated completely between the two compartments. Even in the absence of metabolism this may take hours, during which time deterioration of the preparation may result in the plasma membrane becoming permeable to the medium solute as well as in autolysis of cellular solutes and the loss of the resulting smaller solutes to the medium. For all of these reasons, it seems more accurate to accept the larger values for impermeant cellular solutes estimated from the simple analysis of metabolizing tissues incubated in isosmotic chloride or gluconate media. Thus we conclude that under normal conditions some 75% of cellular osmolality results from impermeant solutes and their associated counterions in these mammalian tissues.
V. WHAT ARE THE CELLULAR IMPERMEANT SOLUTES? There are surprisingly few studies of mammalian tissues which have attempted to provide a complete description of the various impermeant cellular solutes which contribute to osmotic balance and cell electroneutrality. Perhaps because of the greater diversity in cellular osmolalities found in some lower vertebrates and invertebrates, rather more detailed information is available about cellular composition in these species. Brief discussion only is justified here, for Burton (1983) has reviewed this topic recently and aspects of it are also dealt with elsewhere in this volume. Unlike cells of lower species, mammalian cells in general, with the apparent exception of cells in the brain (Pollock and Arieff, 1980), maintain their total solute content relatively constant despite variations in extracellular osmolality over a pathophysiological range. They do not appear to contain the quantities of free amino acids and other nitrogenous solutes which are reported in many marine and fresh water species. Burton (1983) provides a table giving some representative values for rat skeletal muscle, drawing on data from a number of studies. Estimates of cell sodium and chloride provided there are lower than in Table V, as is the case when chloride space is used as an estimate of extracellular fluid volume. High energy phosphate compounds, chiefly phosphorylcreatine, contribute some 50 mmol/kg H,O as do low-molecular-weight
ISOSMOTIC VOLUME MAINTENANCE
21
nitrogenous solutes including free amino acids. Of these compounds, only phosphorylcreatine contributes significantly to the estimated net negative charge on the cellular impermeant solutes. Much of this charge must, therefore, be carried on proteins or exist on membranes of organelles. No estimates of cellular protein concentration are provided by Burton (1983). However, the estimated contributions of the small-molecular-weight impermeant solutes are large enough of themselves to more than account for all the impermeant solutes required for osmotic balance, suggesting that the osmotic, though not the charge contributions of those macromolecules, must be quite small.
VI.
HOW IS THE CELLULAR OSMOTIC SWELLING FORCE OFFSET?
In the absence of a positive cellular hydrostatic pressure, this can only be accomplished by the presence of an extracellular solute which is excluded from the cells, as was recognized independently by Wilson (1954) and by Leaf (1956). Sodium appears to be the solute excluded. Most workers favor the view that this reflects a combination of a low plasma membrane sodium permeability together with active extrusion of sodium from the cells by a sodium pump-the “pumpleak” hypothesis (Tosteson, 1964; Tosteson and Hoffman, 1960). However, as discussed earlier, alternative models which place emphasis on properties of the cytoplasm favoring potassium accumulation and sodium exclusion, rather than on the properties of the plasma membrane, continue to be advocated. It has proved difficult to obtain experimental evidence which allows the unequivocal rejection of one or other of these hypotheses. However, the continuing success in identifying specific membrane-localized ion transport processes argues in favor of a major role of the membrane in controlling cellular ions and volume. Also, the recent observation that gross cellular swelling induced by replacement of medium chloride by acetate is associated with a net accumulation of potassium in renal cortical cells (Cooke and Macknight, 1984) is not predicted by the association-induction hypothesis but is fully consistent with a dominant role of the plasma membrane. Since the weight of the evidence favors the membrane-based model, it will be examined in more detail. An adequate description of the regulation of cellular volume in such a model requires knowledge of the relationships between rates of passive and active ion movements, as well as of membrane permeabilities to ions and of their electrochemical gradients. The latter in turn requires knowledge of plasma membrane potentials and of cellular and medium ion and osmotic activities. Unfortunately, we lack the data required to provide a detailed description of cellular volume regulation for any mammalian cell type. The simplest cell to study would be one which possessed a plasma membrane with only conductive pathways for
22
ANTHONY D. C. MACKNIGHT
passive ion movements, and with only the Na+ ,K+-ATPase providing a carriermediated transport mechanism for the predominant univalent cations. In such a cell, chloride ions would be distributed at electrochemical equilibrium and the steady state cellular volume would be constrained by five simple relationships, as discussed and analyzed by Jakobsson (1980). Two, the necessities for osmotic balance and for cellular electroneutrality, have been discussed already and used to estimate impermeant cellular solute contents and average charge. The others involve equations for the equilibrium distribution of chloride, and for the relationships between passive and active sodium and potassium fluxes. Simplifying, but reasonable, assumptions that the electric field within the membrane is constant and that pump flux is directly proportional to cellular sodium concentration are required to derive these latter equations. In addition, an extension of this approach provides a set of equations with which one can examine the behavior of cells in conditions under which volume will alter. Mammalian skeletal muscle might be expected to be a good tissue in which to study factors influencing cellular volume. Since it is a symmetrical cell, one is not confronted with the problem of plasma membranes having different transport characteristics, as one is in epithelia. Also, it is a homogeneous tissue with cells that are relatively easily studied with microelectrode techniques. Furthermore, in the resting state, it is commonly regarded as having conductive pathways for potassium, chloride, and sodium while lacking the cotransport mechanisms found in epithelia. It appears, however, to have a sodium-hydrogen countertransport system (Aickin and Thomas, 1977) whose principle role may be in the regulation of cell pH. We have begun such a study but have yet to resolve problems in the interpretation of data obtained from microelectrode and from chemical analysis. Tables IV and V illustrate these problems. In particular, while the microelectrode data indicate that chloride is distributed at electrochemical equilibrium, the chemical analysis suggests that cells contain more chloride than predicted and that this is all lost, together with potassium (which must be cellular), when cells are incubated in a chloride-free medium. Until we have additional data, further speculation is unwarranted. Given our difficulties in resolving data obtained with these different techniques in skeletal muscle, it is not surprising that in the more complicated epithelial tissue no fully satisfactory description of volume regulation under isosmotic conditions exists. Following the description of volume regulation in terms of the “doubleDonnan,” or “pump-leak” model in the late 1950s, several groups examined the effects of cardiac glycosides on cellular volume. The expectation that cells would swell when their sodium pumps were inhibited was met in a variety of tissues. However, some workers reported little or no swelling in metabolizing tissue under their experimental conditions (see Macknight and Leaf, 1977, for
23
ISOSMOTIC VOLUME MAINTENANCE
detailed discussion with references). This contrasts with the swelling produced by metabolic inhibition (Fig. 5). More impressively, tissue swollen by metabolic inhibition was able to recover its volume, when metabolism was restored, just as rapidly and completely in the presence of ouabain as under control conditions, even though net potassium uptake was inhibited completely (Fig. 6). Such results were interpreted as indicating that cells possessed a mechanism (or mechanisms) other than the conventional ouabain-sensitive sodium pump by which they regulated their volume. Suggested mechanisms included a second sodium pump insensitive to ouabain (Whittembury, 1968) and some type of contractile mechanism which could extrude an isosmotic solution from the cell (Kleinzeller, 1965). The interpretation of these findings remains controversial. Many of the discrepancies between different studies in the extent to which cellular swelling occurs after cardiac glycosides seem to reflect differences in the rates at which such rabbit
A
L.0 /
ouabain
U
m
Y
- 40
F
,* guinea
I L
I
0
,/’
I
1
I
I
60 120 time (rnin)
I
J
180
FIG. 5. The effects on tissue water content of incubation, at 25°C. of renal cortical slices in sodium Ringers alone, or containing ouabain or iodoacetamide. Concentrations (mmollliter) of ouabain used were rabbit, 10, guinea pig, 10, and rat, 15. Media containing iodoacetamide ( I rnmol/liter) were gassed with nitrogen, not oxygen. Slices were all equilibrated for at least 15 min in oxygenated sodium Ringers prior to the start of the experiment. (Data from Hughes and Macknight, 1977.)
24 A
ANTHONY D. C. MACKNIGHT B
4 9
0
e
&
FIG. 6 . The behavior of rat renal cortical slices reincubated at 25°C at time zero in oxygenated sodium Ringers without (A) or with (B) ouabain (10 mmol/liter). Slices were first equilibrated in oxygenated sodium Ringers at 25°C for at least 15 min (values shown at arrow), then placed in nitrogenated sodium Ringers at 0.5"Cfor 150 min (leaching). Slices reincubated with ouabain were also exposed to the glycoside throughout the incubation at 0.5"C.Each point represents the mean f SD of 7 to 23 separate observations. (Figure reproduced from Macknight and Leaf, 1977.)
swelling can occur, together with variations in experimental protocols (e.g., differences in preliminary incubation and in temperature of incubation, phosphate-buffered as opposed to bicarbonate-buffered media) and different degrees of sensitivity of tissues from different species to these drugs. There seems little doubt now that cells incubated with ouabain in concentrations sufficient to produce maximal inhibition of the Na+ ,K+ -ATPase will swell eventually. The time that this takes, however, will reflect the balance between the rates of net loss of cell potassium and of uptake of extracellular sodium and chloride. In some epithelia, at least, increased cell sodium appears to decrease apical plasma membrane sodium permeability, either directly or indirectly (Chase, 1984), so that swelling becomes a very slow process (Macknight et al., 1980). In addition, swelling in epithelial cells may increase the availability of mechanisms for potassium loss across the basolateral plasma membrane either directly, by increasing the numbers of potassium-selective conductive pathways (Lewis et al., 1985), or indirectly, by activating chloride pathways (Ussing, 1982). Such an increase in potassium loss with chloride will also retard swelling. Though such
25
ISOSMOTIC VOLUME MAINTENANCE
mechanisms could alter the rate at which cells swelled, and even cause some cell shrinkage after an initial swelling phase, they could not prevent the inevitable swelling which must occur in the absence of energy-dependentnet sodium extrusion from cells. Within a conventional framework, several factors might contribute to the recovery of volume, illustrated in Fig. 6, when metabolism is restored to ouabain-inhibited tissue. First, though the absence of net uptake of potassium argues for a significant degree of pump inhibition when metabolism is first restored, the study of Mills et af. (1981), in a line of tissue cultured kidney cells (LLC-PK,), shows that ouabain binds only to pumps turning over in the sodium mode. During metabolic inhibition little or no binding occurs and, when metabolism is restored, some minutes elapse before binding is complete (Fig. 7). The possibility that some initial activity of the ouabain-sensitive sodium pump could contribute to volume regulation under these conditions is supported by the studies of Cooke (1978a,b, 1979, 1981a,b) who found an inverse correlation between the restoration of cellular volume and species sensitivity to ouabain in renal cortical slices from rat, guinea pig, and rabbit (Fig. 8). In addition, the
I
012 5
1
10 15
I
I
30 time iminl
60
FIG.7. The binding of [3H]ouabain to cultured pig kidney epithelial cells. The rate of binding ( B ) was examined at 37°C in cells maintained throughout at that temperature or first incubated with 13H]ouabain at 4°C before incubation at 37°C. Note that virtually no ouabain binding occurred at the low temperature. The faster rate of binding at 37°C by cells incubated first at 4°C may reflect the greater sodium concentration and lower potassium concentration resulting from the period of metabolic inhibition. At 37°C the initial activity of the Na+,K+-ATPase on which ouabain binding depends should, therefore, be greater in these cells than in the control cells maintained at 37°C throughout the experiment. (From Mills er a l . , 1981.)
26
ANTHONY D. C. MACKNIGHT C
A
6oo
I--”\
a
t
40
600
*,e-=-=-+
200
0 0 -
q; I m cm
0
30 Time (min)
60
0
30 Time (min)
60
0
30
60
Time ( m i d
FIG.8. A comparison of the rates of change of tissue water and ions when swollen renal cortical slices from rabbit (A), guinea pig (B). or rat (C)were incubated with or without ouabain at 25°C in oxygenated media. To maximize the initial water contents slices were transferred to medium at 02°C rather than being first equilibrated in oxygenated sodium Ringers for at least 15 min at 25°C. Ouabain concentrations (mmol/liter) were rabbit, 1 , guinea pig, 10, and rat, 10. Each point represents the mean k SD of 6 to 8 separate observations. (Data from Cooke, 1981b.)
failure to accumulate potassium in the early minutes of incubation in rat renal cortical slices could be contributed to, in part, by increased plasma membrane potassium permeability which, as discussed earlier, may be associated with cell swelling, in some epithelia at least. Second, it is possible that at least some of the initial recovery in volume when metabolism is restored might be associated with resynthesis of impermeant solutes from autolytic products retained within the cells. Such resynthesis, by reducing the numbers of osmotically active cellular solutes, would increase cellular water activity above that in the medium and the consequent loss of this water would be accompanied by loss of diffusible solutes. Such a mechanism could make some contribution to volume recovery both in the normal and in the ouabain-inhibited tissues. This recovery could only be temporary, however, and would be followed by further swelling unless sodium was excluded from the cells in some way. Any contribution of impermeant solute resynthesis to volume recovery is likely to be small. The available information suggests that the products of autolysis are lost to a large extent from metabolically inhibited cells (Swan and Miller, 1960; Deymp 1953a,b). Furthermore, if such a mechanism played an important role in the initial restoration of volume, cell volume should recover in the first minutes after metabolism is restored in media in which sodium has been replaced by other cations. However, there is no evidence of this, at least in renal cortical tissue. As shown in Fig. 9, the restoration of volume appears to be sodium dependent.
27
ISOSMOTIC VOLUME MAINTENANCE
Third, it is possible that some membrane transport pathway other than that through the sodium pump might contribute to recovery of cellular volume by producing a net extrusion of sodium chloride from the cells. Possibilities could only include transport processes for which favorable gradients between cell and extracellular fluid existed at the end of the period of metabolic inhibition. Furthermore, once metabolism was restored, the rate coefficients for such processes would have to increase to a greater extent than did the rate coefficients for processes involved in the passive uptakes of solutes from the medium. Such increases could be achieved, for example, by inserting previously unavailable pathways into the plasma membrane. In experiments in which metabolic inhibition had been caused by chilling, it could reflect the higher Q,, of carriermediated transport compared with simple diffusive movements when the temperature was increased. Assuming that derived concentrations adequately reflect ion activities under the conditions, examination of results from experiments with
equilibroted ( N a Ringer) 25OC. 02 leached 150 min. 0.5'C.
-
'
6.0C
0
N2
reincuboted
2 min. 25OC. 02
reincubated
60 min. 25OC. 02
P
D
m
Y \
-2
I
5.0C
c
0
E
i_
L.0C
Om
I 0
m
2
3.00
Sodium Ringer
Choli ne Ringer
Cesium Ringer
Rubidium Ringer
Potassium Ringer
FIG. 9. The effects on tissue water content of rat renal cortical slices of incubation in chloride media with different univalent cations. All slices were first equilibrated in oxygenated sodium Ringers for at least 15 min before transfer to the media shown. Note that only tissues incubated in sodium chloride medium showed significant recovery of volume when leached slices were reincubated in oxygenated medium at 25°C. The differences in the degrees of swelling in the other media appear to reflect the relative permeabilities of the plasma membranes to the different cations. (Reproduced from Macknight, 1981.)
28
ANTHONY D. C. MACKNIGHT
renal cortical slices (Macknight, 1968) allows us to exclude a one-to-one coupled sodium-chloride cotransporter as a possibility, for even at the end of metabolic inhibition the sodium chloride gradient would continue to drive sodium chloride into the cells. However, the energetic gradient for a potassium-chloride cotransporter and, also, for a sodium-potassium-two chloride cotransporter, would favor net solute loss from cell to medium. One ingenious variation on this possibility is that put forward by van Rossum and his collaborators (Russo et af., 1977, 1985; van Rossum and Russo, 1981, 1984) as discussed in van Rossum el al., this volume. In their hypothesis, a chloride-dependent cotransport mechanism drives ions and thereby water into cellular vesicles which then merge into the plasma membrane and discharge their contents by exocytosis. If the sodiumpotassium-two chloride cotransporter were involved in volume recovery, specific inhibitors like furosemide and bumetanide should block this recovery. There are conflicting results about their effects in swollen, ouabain-inhibited tissue which may reflect species differences and tissue sensitivities, as well as the concentrations of inhibitors used. At concentrations greater than 2 mmol/liter, furosemide causes gross cellular swelling in renal cortical slices, associated with disruption of metabolism (Macknight, 1969; van Rossum et af., 1981). However, at concentrations up to.2 mmol/liter furosemide alone causes no swelling in rat renal cortical slices (Macknight, 1971), nor does it prevent restoration of volume when swollen rat renal slices (Macknight, 1969; van Rossum et al., 1981) or rat liver slices (van Rossum and RUSSO,1984) are reincubated in oxygenated medium. In contrast, at a concentration of 1 mmol/liter, it prevents volume recovery in guinea pig renal cortical slices (P6rez-Gonziilezet al., 1980). Whereas in our study (Macknight, 1969) the combination of ouabain and furosemide together had no detectable inhibitory effect on volume recovery in swollen slices (Fig. lo), some inhibition of recovery in both rat renal slices (Russo et al., 1985) and rat liver slices (van Rossum and Russo, 1984) has been reported. In the absence of knowledge of the pH gradients, it is not possible to explore possible contributions of sodium-hydrogen and chloride-bicarbonate countertransport systems to volume recovery. Although amiloride does not prevent volume recovery or inhibit potassium reaccumulation in ouabain-inhibited renal cortical slices when metabolism is restored (Macknight, 1969), it is not possible to obtain the full effect of this diuretic on sodium-hydrogen exchange with normal medium sodium concentrations (Kinsella and Aronson, 1981). In our studies, though amiloride inhibited tissue oxygen consumption significantly, steady state sodium, potassium, chloride, and water contents were unchanged (Macknight, 1971). Thus there was no evidence that amiloride had any direct effect on cell Na+ ,K+-ATPase, as has been claimed from studies with suspensions of rabbit renal proximal tubules (Soltoff and Mandel, 1983). Again it must be emphasized that while the activation of transport systems
29
ISOSMOTIC VOLUME MAINTENANCE
ordinory
CI-
V
0
cI-content
+
furosemide
170
4
+
L60
-
380
-
300
-
220
- 0
+ + $ $
290
:-
+ O Q 0
210
# # @ @ ( + * @
+ @
130 ..
3.30
-
2.50
- A
before leaching
a @ Q
Q
Q
1
1
4
0
1
2
4
Q
ij
,
I
I
@
0
if 4 6)
2.90 -.
'issue H,O content tissue K+ content hssue ~d content tissue lmmol/ k g dry w t 1 kg H,O/kg dry w t )
(Y
8
Q
Q
pi 9(
1 6 3 0 6 0
time Imtn)
FIG. 10. The effects of furosemide on the restoration of volume in rat renal cortical slices. Slices were first equilibrated at 25°C in oxygenated sodium chloride medium (mow). They were then transferred to nitrogenated medium at 0.5"C containing furosemide or furosemide and ouabain where they remained for 150 min (leaching) before being reincubated at time zero with these drugs in oxygenated medium at 25°C. Comparison with Fig. 6 illustrates that furosemide alone had no appreciable effect on restoration of tissue water or ions. Though ouabain produced the expected inhibition of potassium reaccumulation, the combination of the two drugs had no effect on the rate or extent of volume recovery. Note that, for clarity, the time scale in this figure is logarithmic, in contrast to that in Fig. 6. Each point represents the mean 2 SD of six separate observations. (From Macknight. 1969.)
30
ANTHONY
D.C. MACKNIGHT
which move ions down a gradient could slow or reverse swelling temporarily, as the gradients were dissipated cells must continue to swell as water, sodium, and chloride are driven in as a consequence of the osmotic force of the impermeant solutes. In summary, though under some experimental conditions the volume of ouabain-treated cells does not vary as expected from the simple pump-leak model, it may still be possible to account for this apparently anomalous behavior in terms of the conventional sodium pump, without having to invoke an alternative primary energy-dependent transport process. Nevertheless, workers continue to explore the possibility of a second, ouabain-insensitivesodium pump and recently two groups have reported preparation of sodium-sensitive ATPases from epithelial plasma membranes which differ in their properties from the conventional Na+,K+-ATPase (Marin ef al., 1985; Del Castillo and Robinson, 1985a; Proverbio and Del Castillo, 1981; Proverbio et al., 1975, 1982). Furthermore, evidence has also been obtained that this ouabain-insensitive ATPase may support sodium transport across basolateral plasma membrane vesicles prepared from guinea pig small intestinal and rat renal cortical cells, respectively (Del Castillo and Robinson 1985b; Marin et al., 1985). Inasmuch as sufficient ouabain virtually abolishes transepithelial sodium transport, the role of this putative sodium transport mechanism in intact epithelial cells is unclear. It is conceivable, however, that the inhibition of transepithelial transport by ouabain is a consequence of decreased sodium entry to the cells across the apical plasma membrane associated with the disturbances of cellular ion concentrations which result from the inhibition of the basolateral membrane Na+ ,K+ -ATPase. In this case, continued activity of another basolateral membrane sodium pump might not result in transepithelial sodium transport. Further advances in our analysis of these problems require a more detailed knowledge of cell ion activities, of plasma membrane potentials, and of the nature and control of membrane transport processes than we currently possess. While it has not proved possible to establish unequivocally that a second primary energy-dependentprocess is not involved in volume regulation, we have been able to show that the conventional ouabain-sensitive sodium pump does play an important role in determining the extent of volume recovery by swollen tissue. Rabbit renal cortical cells, incubated in isosmotic media in which acetate completely replaces chloride, about double their water content (Cooke and Macknight, 1984). Ouabain does not affect this swelling, though cells lose potassium and gain sodium as expected. In the absence of ouabain, swollen slices transferred to chloride medium recover their volume and ionic composition completely. Ouabain largely prevents this (Fig. 11). The small recovery which does occur is readily explained by acetate leaving cells faster than chloride can enter to replace it. Thus some cation is also lost as is water to maintain osmotic balance.
31
ISOSMOTIC VOLUME MAINTENANCE
VII. WHAT LIMITS CELLULAR SWELLING? Red blood cells exposed to hypotonic media swell and, if the uptake of fluid is great enough, they hemolyze. Obvious rupture of plasma membranes with loss of cellular contents under an osmotic stress is, however, seen less often in other cell types. Some cells even appear to withstand exposure to distilled water for finite periods of time. A variety of factors may protect against cell lysis, their relative importance varying from tissue to tissue. Unlike single cells, cells in organs may have their swelling limited by extracellular matrix such as basement membrane (renal tubules, Linshaw, 1980) or by collagen fibers (cardiac muscle, Pine e t a l . , 1981). They may also exchange solutes through gap junctions, thus sharing osmotic loads between adjacent cells. Under hyposmotic conditions losses of solutes may limit swelling. In mammals, over a pathophysiological range of osmolalities, cells in the brain appear to limit their swelling in this way (Pollock and Arieff, 600 -
-
control
10.’ M ouabain
+
W u
O1
NaCl medium
No acetate medium 60 min
NaCl medium 60 min
Rabbit renal cortical slices incubated in sodium chloride and sodium acetate media with FIG. 1 I . or without ouabain. Slices were first equilibrated in oxygenated sodium chloride medium for at least 15 min at 25°C. They were then transferred to oxygenated media at 25°C in which chloride was replaced by acetate. Note that cell water increased markedly under these conditions. Cells lost chloride and accumulated both sodium and potassium under control conditions. Ouabain did not affect chloride loss or cell swelling but caused the expected inhibition of potassium uptake so that the swelling was associated with sodium accumulation alone. Control cells returned to chloride medium restored their original water and electrolyte compositions. Ouabain largely prevented this restoration of volume. (Figure drawn from data in Cooke and Macknight, 1984.)
32
ANTHONY D. C. MACKNIGHT
1980). There is less evidence for this in other mammalian cell types. However, with extreme hyposmolality in in vitro experiments, large losses of ions accompany, and must limit, the extent of the swelling which occurs. The rate of swelling may also be important. Rapid swelling may result in the activation of membrane transport processes with consequent compensatory losses of solutes (often referred to as regulatory volume decrease) as has been described in a variety of isolated cells (discussed by Hoffmann, this volume). I n vitro, in an organized tissue, cells may be exposed less abruptly to the osmotic gradient because of problems of diffusion and mixing within the tissue interstices, with a less dramatic initial effect on cellular volume. This more gradual swelling may limit the extent to which transport processes are activated in the membrane and account for the difficulty in demonstrating in tissue slices the dramatic alterations in cellular solutes with partial volume recovery after exposure to hyposmotic media which is seen in some isolated cells and in isolated renal tubules (Gonzalez et al., 1982). It is also possible that the cellular matrix itself, either through the cytoskeleton or because of the gellike nature of the cytoplasm, or through a combination of the two, limits swelling under extreme conditions. Nevertheless, despite the limitations of swelling which can occur under some circumstances, cells incubated in isosmotic media may double their water content with an associated accumulation of potassium, thereby maintaining a high potassium concentration. Such behavior is seen in renal cortical cells incubated in isosmotic media in which acetate replaces chloride (Fig. 11). Renal cortical cells also swell to a similar extent when incubated in media in which cesium, rubidium, or potassium completely replace sodium. Swelling in these isosmotic media is of much the same magnitude as that associated with very dilute incubation media, suggesting that this may represent the maximum swelling obtainable. Presumably, further swelling is limited by viscoelastic forces generated within the tissue.
VIII. THE ROLE OF MEDIUM ANIONS IN CELL VOLUME REGULATION Much emphasis has been placed on the effective exclusion of sodium from cells as the determinant of cellular volume under isosmotic conditions. Yet volume is determined not simply by this alone. In the steady state it reflects the distribution reached by all permeant species as well as the osmotic and charge effects of the nonpermeant cellular solutes. It is possible for isosmotic media in which chloride is replaced by other anions to be either hypertonic or hypotonic. Figure 12 illustrates a range of steady state cellular water contents found in rabbit renal cortical slices incubated in isosmotic media in which chloride has been replaced partially or completely either by the
33
ISOSMOTIC VOLUME MAINTENANCE
- 600 - 500
0 +
I
3
- 400$ m
c 2
P v -300 E
a
75 0
,
,
44
88
l,i,l,
200
100 100
I
Y
, , ,
132 100 67 40 10 medium [ C I I
200;
I
0
I
0
and concentration (A)in rabbit renal cortical Fic. 12. Cell water (0)and potassium content (0) slices incubated in isosmotic oxygenated media (275 mOsm/kg H20)at 25°C in which chloride is substituted for, partially or completely, by either gluconate or acetate. Note the relative constancy of derived cell potassium concentration despite the range of cell water contents. (Figure reproduced from Macknight, 1985.)
impermeant anion gluconate or by acetate. The latter ion appears to cause cellular swelling because the undissociated weak acid can cross the plasma membrane whereas the acetate ion cannot. So long as a favorable gradient for nonionic acid diffusion exists cells will swell. Figure 13 summarizes the steps believed to be involved. Water content is a little greater when renal cortical slices are incubated in a medium containing a physiological concentration of bicarbonate gassed with 95% 0,-5% CO, than when slices are incubated in bicarbonate-free phosphatebuffered chloride medium gassed with 100% 0, (Cooke, 1976). However, tissue behavior is otherwise comparable and bicarbonate-free medium has the advantage that changes in univalent cations (sodium and potassium) can be directly related to measured anion (chloride). Law (1984~)has published data on the effects of a wide range of anions on volume in rat kidney cortical slices. With different medium anions, therefore, it is possible to produce cells varying some threefold in water content yet having relatively constant potassium concentrations. Furthermore, the changes in water and ion contents are completely reversible, with restoration of a normal composition upon return to an isosmotic chloride medium.
34
ANTHONY D. C. MACKNIGHT
C
FIG.13. (A) The mechanism of cellular swelling in acetate media. Acetic acid (HAc) equilibrates across the plasma membrane which is relatively impermeable to the acetate ion (Ac-). The resulting cellular accumulation of solute causes a net entry of water to the cells with swelling. There is also a loss of cellular chloride ions to the medium in which the chloride ion concentration has been reduced. There are two further processes which will accompany the changes in (A). (B)The cellular hydrogen ions are exchanged for extracellular sodium ions. The mechanism illustrated involves sodium-hydrogen ion countertransport but there is no direct experimental evidence for this from our studies. (C) The sodium ions gained are actively extruded from the cells by the Na .K -ATPase. This step can be inhibited by ouabain or by metabolic inhibition. The final cellular composition reflects the extent of acetate accumulation and the relative activities of the Na+ , K + pump compared with the rates of passive diffusion of these ions across the membrane. (Reproduced from Cooke and Macknight, 1984.) +
+
The osmotic contribution of cellular chloride can be estimated from the difference between cellular water content in gluconate and in chloride media (Table V). For example, in rabbit renal cortical slices, this difference of 0.57 kg H,O dry wt is associated with a difference in cell chloride content of 88 mmol/kg dry wt [and of cation content (Na + K) of 89 mmol/kg dry wt, as required for
ISOSMOTIC VOLUME MAINTENANCE
35
electroneutrality]. Thus the fluid moving under these conditions is isosmotic (calculated osmolarity 3 10 compared with calculated medium osmolarity of 293 mOsm/liter water), and chloride with associated diffusible counterions contributed some 0.57L2.09 or 27% to total cellular osmolarity. Yet what determines cellular chloride content in isosmotic chloride media? It is now widely accepted that chloride is not distributed at electrochemical equilibrium in many epithelia, for both absorptive and secretory cells contain more chloride than would be predicted from measurements of membrane potentials. For example, were chloride in electrochemical equilibrium, the estimated concentration in the renal slices referred to, of 44 mmol/liter, would require a plasma membrane potential of some -30 mV (inside negative), about 20 to 30 mV less negative than that measured in guinea pig renal cortical slices (Proverbio and Whittembury, 1975). Whether or not chloride is distributed at equilibrium in other cell types is often controversial and unresolved. An example of the difficulties is provided by our data in mammalian skeletal muscle (Table IV) where measurements of membrane potentials and chloride activities indicate an equilibrium distribution, but where estimates of cellular chloride from chemical analysis require a very low activity coefficient for cell chloride for the results with the two techniques to be harmonized. There are several possible pathways through which chloride can cross plasma membranes. Emphasis has been placed in the past on conductance pathways through which chloride diffuses. More recently, carrier-mediated co- and countertransport pathways have been identified and characterized. Cotransport pathways include sodium-chloride, potassium-chloride, and sodium-potassiumtwo chloride carriers; countertransport pathways include the bicarbonate-chloride exchange carrier. These electroneutral carriers are not detectable through electrical measurements but can be identified through the use of a variety of more or less specific inhibitors. Thus cellular chloride may be influenced not only by plasma membrane potential but also by the relative activities of a variety of membrane-based carriers. Furthermore, the availabilities of both conductance channels and carriers within the plasma membranes for a variety of ions appear not to be constant. Instead, they may vary as a consequence of alterations in cell volume. For example, in a recent study examining properties of the basolateral plasma membrane in toad urinary bladder epithelial cells, Lewis el al. (1985) found that incubation in isosmotic media in which chloride was replaced by gluconate resulted in shrinkage of epithelial cells. This shrinkage was associated with loss from the basolateral membrane of barium-sensitive highly selective potassium conductance channels. These channels were reactivated when cell volume was increased again by incubation in hyposmotic gluconate medium. Associated with these changes in potassium conductance were changes in sodium pump activity, suggesting that both passive and active pathways through the basolateral membrane were affected in parallel by the changes in volume. Decreases in membrane
36
ANTHONY 0. C. MACKNIGHT
conductance and in pump activity with cell shrinkage would tend to minimize the decrease in volume by limiting losses of cellular ions. Conversely, increases in membrane conductance and in pump activity with cell swelling would tend to minimize the increase in volume by accelerating losses of cellular ions. Thus these epithelial cells may, to some extent, be protected against changes in volume. However, the mechanisms by which the conductance channels are affected remain to be established. Possibilities include inactivation or activation within the plasma membrane itself, or removal to or insertion from submembrane stores. The behavior of toad bladder epithelial cells in these experiments could be described adequately primarily in terms of ion conductances, with initial loss of cell chloride into gluconate medium via a chloride conductance pathway followed, as cell volume decreased, by loss of potassium conductance pathways and decreased sodium pump activity. However, in other epithelia, carrier-mediated chloride movements may be of greater importance. For example, Ussing (I 982) has discussed volume regulation in frog skin epithelial cells in terms of a balance between the availabilities of basolateral membrane cotransporters and of chloride conductive pathways. As originally presented, a sodium-chloride cotransporter was postulated, but a sodium-potassium-two chloride system may be involved (Ussing, 1985). He suggests that in the steady state, with normal cellular volume, the activity of the cotransporters is minimal and there is negligible chloride conductance. Under these conditions, cellular potassium is close to its equilibrium distribution but cells contain more chloride than predicted, for chloride had been driven into the cell by the cotransporter energized by the sodium gradient. It is argued that the swelling associated with a hyposmotic serosal medium is minimized by activation of chloride conductive pathways which allow chloride to move out of the cells down its electrochemical gradient, taking potassium with it, whereas, in a hyperosmotic serosal medium, shrinkage of cells is minimized by activation of the cotransporter. The mechanisms described above may play a role, not only in protecting epithelial cells from undue changes in volume under anisosmotic conditions, but may also serve to maintain cellular volume relatively constant in the face of wide variations in rates of transepithelial transport (Schultz, 198 1). In view of our uncertainties about so many aspects of cellular volume regulation in isosmotic media, it is perhaps fitting to end with yet another apparently simple but unanswered question. How is it that cells achieve similar steady state volume, and sodium and potassium contents, in isosmotic media in which chloride is replaced completely by nitrate? This fact, demonstrated by Mudge (195 1) in renal cortical slices, is illustrated in Fig. 14, which displays data for rabbit renal cortical slices and for mouse diaphragm obtained recently in our laboratory (Scott, Tarbotton, Leader, and Macknight, unpublished observations). This must imply that the pathways utilized by chloride in crossing the plasma membrane are
37
ISOSMOTIC VOLUME MAINTENANCE L.0
0
chloride medium [7 nitrate medium
r
I
c
3.0 u L 1 m
\
m Y
c
2.0
r
2 8 0 ,
= --
1.0
2
0 rabbit renal cortical slices
mouse diaphragm
FIG. 14. The effects on cell water and ions of incubation in sodium chloride or sodium nitrate media. Rabbit renal cortical slices were incubated for 60 min in oxygenated media at 25°C whereas mouse diaphragms were incubated for 60 min in oxygenated media at 37°C. Columns represent means of 12 to 16 separate observations. (Scott, Tarbotton, Leader, and Macknight, unpublished observations.)
also equally available to nitrate. To what extent can nitrate replace chloride on the various co- and countertransport mechanisms so far described? Available evidence indicates that nitrate substitutes poorly, if at all, for chloride in direct sodium cotransport mechanisms (Eveloff et al., 1978; Geck et al., 1980; Haas et al., 1982). However, it appears to substitute well for chloride in bicarbonatechloride countertransport (Aronson and Seifter, 1984). Can nitrate pass through the same conductance pathways as does chloride? So long as we believed that chloride was distributed at electrochemical equilibrium, we had simply to postulate a common conductance pathway. But, if specific carrier mechanisms for chloride are responsible for forcing this ion out of equilibrium, a more complex explanation is required.
IX. SUMMARY Despite our greater knowledge of membrane transport processes, many questions fundamental to our understanding of the regulation of cellular volume remain to be answered. However, in summary, given osmotic equilibrium between cells and extracellular fluids, it can be argued that cellular water content is determined by two factors. First, the amount of irnpermeant cellular solute
38
ANTHONY D. C. MACKNIGHT
determined by metabolism, together with the diffusible ions required for charge neutrality, sets a lower limit to cellular water content. It is argued here that this represents about 75% of the total water content. The remaining water is associated osmotically with the diffusible solutes, principally the univalent sodium, potassium, and chloride ions. The cellular content of these reflects a balance between ion fluxes through the sodium pump, co- and countertransport mechanisms, and conductance pathways. The magnitudes of these fluxes and the steady state cellular concentrations reached depend upon both driving forces for ion movements and upon the relative availabilities of the different pathways, which may vary in such a way as to minimize changes in cellular volume both in isosmotic and anisosmotic media. Thus, though the sodium pump may be primarily responsible for preventing cell swelling, in that it offsets the osmotic force of the impermeant cell solutes, the steady state volume of the cells also reflects the distribution of the permeant anions, principally chloride. ACKNOWLEDGMENTS Work in the author’s laboratory is supported by a program grant from the Medical Research Council of New Zealand.
REFERENCES Aickin, C. C., and Thomas, R . C. (1977). An investigation of the ionic mechanism of intracellular pH regulation in mouse soleus muscle fibres. J. Physiol. (London) 273, 295-316. Andersson, B., Leksell, L. G . , and Rundgren, M. (1984). Regulation of body fluids: Intake and output. In “Edema” (N. C. Staub and A. E. Taylor, eds.), pp. 299-318. Raven, New York. Aronson, P. S., and Seifter, J. (1984). CI- transport via anion exchange. Fed. Proc., Fed. Am. SOC. Exp. Eiol. 43, 2483-2484. Aukland, K.,and Nicolaysen, G. (1981). Interstitial fluid volume: Local regulatory mechanisms. Physiol. Rev. 61, 556-643. Balaban, R. S . , Soltoff, S. P., Storey, J. M., and Mandel. L. I. (1980). Improved renal cortical tubule suspensions: Spectrophatometric study of 0 2 delivery. Am. 1. Physiol. 238, F50-F59. Burg, M. B., and Orloff, J. (1966). Effect of temperature and medium K on Na and K fluxes in separated renal tubules. Am. J. Physiol. 211, 1005-1010. Burg, M. B., Grantham, J., Abramow, M., and Orloff, J. (1966). Preparation and study of fragments of single rabbit nephrons. Am. J . Physiol. 210, 1293-1298. Burg, M. B., Grollman, E. F., and Orloff, J. (1964). Sodium and potassium flux of separated renal tubules. Am. J. Physiol. 206, 483-491. Burton, R. F. (1983). The composition of animal cells: Solutes contributing to osmotic pressure and charge balance. Comp. Biochem. Physiol. 76B, 663-671. Cala, P. M. (1983). Volume regulation by red blood cells: Mechanisms of ion transport. Mol. Physiol. 4, 33-52. Chase, H . , Jr. (1984). Does calcium couple the apical and basolateral membrane permeabilities in epithelia? Am. J. Physiol. 247, F869-F876. Clegg, J. S. (1984). Properties and metabolism of the aqueous cytoplasm and its boundaries. Am. J. Physiol. 246, R133-RI51.
ISOSMOTIC VOLUME MAINTENANCE
39
Comper, W. D., and Laurent, T. C. (1978). Physiological function of connective tissue polysaccharides. Physiol. Rev. 58, 255-315. Cooke, K. R. (1976). Effect of the C02-bicarbonate buffer system on the water and ion contents of rat renal cortical slices. Biochim. Biophys. Acta 437, 280-288. Cooke, K . R. (1978a). Ouabain and regulation of cellular volume in freshly prepared slices of rabbit renal cortex. J . Physiol. (London) 279, 361-374. Cooke, K. R. (1978b). Effects of ouabain and potassium-free media on cellular volume regulation in rat renal cortical slices. J . Physid. (London) 279, 375-384. Cooke, K. R. (1979). Oxygen consumptions and potassium contents of slices of rat renal cortex. Q. J . Exp. Physiol. 64, 69-78. Cooke, K. R. (1981a). Species differences in the effect of ouabain on cell volume recovery in renal cortex. In “Epithelial Ion and Water Transport” (A. D. C. Macknight and I. P. Leader. eds.), pp. 329-338. Raven, New York. Cooke, K . R. (1981b). Ouabain and regulation of cellular volume in slices of mammalian renal cortex. J. Physiol. (London) 320, 319-332. Cooke, K. R., and Macknight. A . D. C. (1984). Effects of medium acetate on cellular volume in rabbit renal cortical slices. J . Physiol. (London) 349, 135- 156. Del Castillo, I. R., and Robinson, J . W. L. (1985a). MgZ+-ATP-dependentsodium transport in inside-out basolateral plasma membrane vesicles from guinea-pig small intestinal epithelial cells. Biochim. Biophys. Acta 812, 402-412. Del Castillo, I. R., and Robinson, J. W. L. (1985b). Na+-stimulated ATPase activities in basolateral plasma membranes from guinea-pig small intestinal epithelial cells. Biochim. Biophys. Acta 812, 413-422. Deyrup, I. (1953a). A study of the fluid uptake of rat kidney slices in virro. J . Gen. Physiol. 36,739749. Deyrup, 1. J. (1953b). Reversal of fluid uptake by rat kidney slices immersed in isosmotic solutions in vitro. Am. J . Physiol. 175, 349-352. Eveloff, I., Kinne, R., Kinne-Saffran, E., Murer, H., Silva, P., Epstein, F. H., Stoff, J., and Kinter, W. B. (1978). Coupled sodium and chloride transport into plasma membrane vesicles prepared from dogfish rectal gland. Pf2iiger.Y Arch. 378, 87-92. Finn, A. L. (1985). Symposium: Volume-dependent pathways in animal cells. Fed. Proc.. Fed. Am. SOC. Exp. Biol. 44, 2499-2529. Fitzsimons, I. T. (1985). Physiology and pathology of thirst and sodium appetite. In “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), Vol. 2, pp. 885-901. Raven, New York. Geck, P., Pietrzyk, C.. Burckhardt, B. D.. Pfeiffer, B., and Heinz. E. (1980). Electrically silent cotransport of N a + , K + , and CI- in Ehrlich cells. Biochim. Biophys. Acta 600,432-447. Geiger. B. (1983). Membrane-cytoskeleton interaction. Biochim. Biophys. Acta 737, 305-34 I . Gonzilez, E., Carpi-Medina, P., and Whittembury, G. (1982). Cell osmotic water permeability of isolated rabbit proximal straight tubules. Am. J . Physiol. 242, F32 I -F330. Grantham, J., Linshaw, M., and Welling, L. (1981). Volume regulation in isotonic and hypotonic media in isolated rabbit renal proximal tubule. In “Epithelial Ion and Water Transport” (A. D. C. Macknight and J. P. Leader, eds.), pp. 339-347. Raven, New York. Grinstein, S., Rothstein. A,, Sarkadi, B.. and Gelfand, E. W. (1984). Responses of lymphocytes to anisosmotic media: Volume-regulating behaviour. Am. J . Physiol. 246, C204-C2 15. Haas, M., Schmitt, W. F., 111, and McManus. T. J . (1982). Catecholamine-stimulated ion transport in duck red cells. Gradient effects in electrically neutral (Na + K + 2C1) cotransport. J . Gen. Physiol. 80, 125-147. Harvey, E. N. (1954). Tension at the cell surface. Protnp/asmato/ogia 2, E5. Hoffmann, E. K. (1977). Control of cell volume. In “Transport of Ions and Water in Animals”
40
ANTHONY D. C. MACKNIGHT
(B. L. Gupta, R. B. Moreton, J. L. Oschman, and B. J. Wall, eds.), pp. 285-332. Academic Press, London. Horowitz, S. B., and Miller, D. S. (1984). Solvent properties of ground substance studies by cryomicrodissection and intracellular reference-phase techniques. J. Cell Biol. 99, 172s- 179s. Hughes, P. M., and Macknight, A. D. C. (1976). The regulation of cellular volume in renal cortical slices incubated in hyposmotic medium. J. Physiol. (London) 257, 137-154. Hughes, P. M., and Macknight, A. D. C. (1977). Effects of replacing medium sodium by choline, ’ caesium or rubidium, on water and ion contents of renal cortical slices. J . Physiol. (London) 267, 113-136. Jakobsson, E. (1980). Interactions of cell volume, membrane potential, and membrane transport parameters. Am. J . Physiol. 238, C196-C206. Kinsella, J . L., and Aronson, P. S. (1981). Amiloride inhibition of the Na+ -H exchanger in renal microvillus membrane vesicles. Am. J. Physiol. 241, F374-F379. Kirk, K. L., DiBona, D. R., and Schafer, J. A. (1984). Morphologic response of the rabbit cortical collecting tubule to peritubular hypotonicity: Quantitative examination with differential interference contrast microscopy. J . Membr. Biol. 79, 53-64. Kleinzeller, A. (1965).The volume regulation in some animal cells. Arch. Biol. Liege 76, 217-232. Kregenow, F. M. (1981). Osmoregulatory salt transporting mechanisms: Control of cell volume in anisotonic media. Annu. Rev. Physiol. 43, 493-505. Law, R. 0. (1982). Techniques and applications of extracellular space determinations in mammalian tissues. Experienria 38, 41 1-421. Law, R . 0. (1984a). Characteristics of ionic binding by rat renal tissue in virro. J. Physiol. (London) 353, 67-80. Law, R. 0 . (1984b). The influence of age on fluid and sodium chloride distribution in rat aortic wall. Q. J . Exp. Physiol. 69, 737-751. Law, R. 0. (1984~).Some effects of monovalent anion replacement on the volume and composition of cells in incubated slices of rat renal cortex. Biochim. Biophys. Acra 773, 246-252. Leader, J . P., Macknight, A. D. C., Mason, D. R., and Armstrong, W. McD. (1981). Cellular composition of Necrurus gallbladder epithelial cells measured by chemical analysis and by ionspecific micwlectrodes. Proc. Clniv. Orugo Med. Sch. 59, 82-84. Leader, J. P., Bray, J. J., Macknight, A. D. C., Mason, D. R., McCaig, D., and Mills, R. G . (1984). Cellular ions in intact and denervated muscles of the rat. J. Membr. Biol. 81, 19-27. Leaf, A. (1956). On the mechanism of fluid exchange of tissues in virro. Biochem. J . 62, 241-248. Lewis, S. A., Butt, A. G., Bowler, J. M., Leader, I. P., and Macknight, A. D. C. (1985). Effects of anions on cellular volume and transepithelial Na+ transport across toad urinary bladder. J . Membr. Biol. 83, 119-137. Ling, G. N. (1984). “In Search of the Physical Basis of Life.” Plenum, New York. Linshaw, M. A. (1980). Effect of metabolic inhibitors on renal tubule cell volume. A m . J. Physiol. 239, F571-FS77. Linshaw, M. A., and Stapleton, F. B. (1978). Effect of ouabain and colloid osmotic pressure on renal tubule cell volume. A m . J. Physiol. 235, F480-F491. Linshaw, M. A., Stapleton, F. B., Cuppage, F. E., and Grantham, J. J. (1977). Effect of basement membrane and colloid osmotic pressure on renal tubule cell volume. Am. J. Physiol. 233, F325-F332. Mclver, D. J. L., and Macknight, A. D. C. (1974). Extracellular space in some isolated tissues. J. Physiol. (London) 239, 3 1-49. Macknight, A. D. C. (1968). Water and electrolyte contents of rat renal cortical slices incubated in potassium-free media and media containing ouabain. Biochim. Biophys. Acru 150, 263-270. Macknight, A. D. C. (1969). Effects of metabolic inhibition and of diuretic agents on the composition of rat renal cortical slices. M.D. Thesis, University of Otago. +
ISOSMOTIC VOLUME MAINTENANCE
41
Macknight, A. D. C. (1971). Effects of frusemide, hydrochlorothiazide and amiloride on rat renal cortical slices. Proc. Univ. Orago Med. Sch. 49, 51-53. Macknight, A. D. C. (1980). Comparison of analytic techniques: Chemical, isotopic and microprobe analysis. Fed. Proc., Fed. Am. Soc. Exp. Eiol. 39, 2881-2887. Macknight, A. D. C. (1981). Ouabain-insensitive volume regulation-a reappraisal. In “Epithelial Ion and Water Transport” (A. D. C. Macknight and J. P. Leader, eds.), pp. 357-362. Raven, New York. Macknight, A. D. C. (1983). Volume regulation in mammalian kidney cells. Mol. Physiol. 4, 1731. Macknight, A. D. C. (1985). The role of anions in cellular volume regulation. Pflugers Arch. Macknight, A. D. C., and Leaf. A. (1977). Regulation of cellular volume. Physiof. Rev. 57, 510573. Macknight, A. D. C., and Leaf, A. (1985). Cellular responses to extracellular osmolality. I n “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), pp. 117-131. Raven, New York. Macknight, A. D. C., Pilgrim, J . P., and Robinson, B. A. (1974). The regulation of cellular volume in liver slices. J. Physiol. (LondonJ238, 279-294. Macknight, A. D. C., DiBona, D. R., and Leaf, A. (1980). Sodium transport across toad urinary bladder: A model “tight” epithelium. Physiol. Rev. 60,615-715. MacRobbie, E. A. C., and Ussing, H. H. (1961). Osmotic behaviour of the epithelial cells of frog skin. Acra Physiol. Scand. 53, 348-365. Maffly, R. H., and Leaf, A. (1959). The potential of water in mammalian tissues. J. Gen. Physiol. 42, 125771275, Marin, R.. Proverbio, T., and Proverbio, F. (1985). Active sodium transport in basolateral plasma membrane vesicles from rat kidney proximal tubular cells. Eiochim. Eiophys. Acra 814, 363373. Mills, J. W., Macknight, A. D. C., Jarrell, J. A,, Dayer. J . M., and Ausiello, D. A. (1981). Interaction of ouabain with a Na+ pump in intact epithelial cells. J . Cell Eiol. 88, 637643. Moore, B., Roaf, H. E., and Webster, A. (1912). Direct measurement of the osmotic pressure of casein in alkaline solution. Experimental proof that apparent impermeability of a membrane to ions is not due to the properties of the membrane but to the colloid contained within the membrane. Eiochem. J. 6, 110-121. Mudge, G. H. (1951). Studies on potassium accumulation by rabbit kidney slices: Effect of metabolic activity. Am. J. Physiol. 165, 113-127. Paillard, M., Leviel, F., and Gardin, J. P. (1979). Regulation of cell volume in separated renal tubules incubated in hypotonic medium. Am. J . Physiol. 236, F226-F231. Perez-Gonzilez, M., Proverbio, F., and Whittembury, G. (1980). ATPases and salt transport in the kidney tubule. Curr. Top. Membr. Tramp. 13, 315-335. Pine, M. B., Brooks, W. W., Nosta, J. J., and Abelmann, W. H. (1981). Hydrostatic forces limit swelling of rat ventricular myocardium. Am. J. Physiol. 241, H740-H747. Pollock, A. S., and Arieff, A. 1. (1980). Abnormalities of cell volume regulation and their functional consequences. Am. J . Physiol. 239, F195-F205. Proverbio, F., and Del Castillo, J. R. (1981). Na+-stimulated ATPase activities in kidney basallateral plasma membranes. Eiochim. Eiophys. Acra 646,99-108. Roverbio, F., and Whittembury, G. (1975). Cell electrical potentials during enhanced sodium extrusion in guinea-pig kidney conex slices. J . Physiol. (London) 250, 559-578. Proverbio, F., Condrescu-Guidi, M., and Whittembury, G. (1975). Ouabain-insensitive Na+ stimulation of an MgZ+-dependent ATPase in kidney tissue. Eiochim. Eiophys. Acra 394,281-292. Roverbio, F., Proverbio, T., and Marin, R. (1982). Ouabain-insensitive Na+ -stimulated ATPase
42
ANTHONY D. C. MACKNIGHT
activity of basolateral plasma membranes from guinea-pig kidney cortex cells. Biochim. Biophys. Acra 688, 757-763. Rink, T. 1. (1984). Aspects of the regulation of cell volume. J. Physiol. (Paris) 79, 388-394. Robertson, G. L. (1985). Regulation of vasopressin secretion. In “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), Vol. 2, pp. 869-883. Raven, New York. Robinson, J . R. (1949). Some effects of glucose and calcium upon the metabolism of kidney slices from adult and newborn rats. Biochem. J. 45, 68-74. Robinson, J . R. (1950). Osmoregulation in surviving slices from the kidneys of adult rats. Proc. R . Soc. London Ser. B 137, 378-402. Russo, M. A,, Van Rossum, G. D. V., and Galeotti, T. (1977). Observations on the regulation of cell volume and metabolic control in virro; changes in the composition and ultrastructure of liver slices under conditions of varying metabolic and transporting activity. J . Membr. Biol. 31,267299, Russo, M. A,, Ernst, S . A,, Kapoor, S. C., and Van Rossum, G. D. V. (1985). Morphological and physiological studies of rat kidney-cortex slices undergoing isosmotic swelling and its reversal: A possible mechanism for ouabain-resistant control of cell volume. J. Membr. Biol. 85, 1-24. Schultz, S . G. (1981 ). Homocellular regulating mechanisms in sodium-transporting epithelia: Avoidance of extinction by “flush-through”. Am. J . Physiol. 241, F579-FS90. Sheff, M. F.. and Zacks, S. I. (1982). Interstitial space of mouse skeletal muscle. J. Physiol. (London) 328, 507-519. Shporer, M., and Civan, M. M. (1977). The state of water and alkali cations within the intracellular fluids: The contribution of NMR spectroscopy. Curr. Top. Membr. Trunsp. 9, 1-69. Siebens, A. W. (1985). Cellular volume control. In “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), pp. 91-115. Raven, New York. Soltoff, S. P., and Mandel, L. J. (1983). Amiloride directly inhibits the Na,K-ATPase activity of rabbit kidney proximal tubules. Science 220, 957-959. Soltoff, S. P., and Mandel, J. (1984). Active ion transport in the renal proximal tubule. 1. Transport and metabolic studies. J. Gen. Physiol. 84, 601-622. Spring. K. R., and Hope, A. (1978). Size and shape of the lateral intercellular spaces in a living epithelium. Science 200, 54-58. Spyropoulos, C. S. (1977). Water fluxes in nerve fiber. J . Membr. Biol. 32, 1-18. Spyropoulos, C. S. (1979). Cytoplasmic gel and water relations of axon. J. Membr. Biol. 47, 195238. Swan, A. G., and Miller, A. T. JM. (1960). Osmotic regulation in isolated liver and kidney slices. Am. J. Physiol. 199, 1227-1231. Tosteson, D. C. (1964). Regulation of cell volume by sodium and potassium transport. In “The Cellular Functions of Membrane Transport” (J. F. Hoffman, ed.), pp. 3-22. Prentice Hall, Englewood Cliffs, N.J. Tosteson, D. C., and Hoffman, J . F. (1960). Regulation of cell volume by active cation transport in high and low potassium sheep red cells. J. Gen. Physiol. 44, 169-194. Ussing, H. H. (1982). Volume regulation of frog skin epithelium. Acfa Physiol. S c a d . 114, 363371. Ussing, H. H. (1985). Volume regulation and basolateral co-transport of sodium, potassium and chloride in frog skin epithelium. P’ugers Arch. Van Rossum, G. D. V., and Russo, M. A. (1981). Ouabain-resistant mechanism of volume control and the ultrastructural organization of liver slices recovering from swelling in vitro. J. Membr. Eiol. 59, 191-209. Van Rossum, G . D. V., and Russo, M. A. (1984). Requirement of CI- and Na+ for the ouabainresistant control of cell volume in slices of rat liver. J. Membr. Biol. 77, 63-76.
ISOSMOTIC VOLUME MAINTENANCE
43
Van Rossum, G. D. V., Ernst, S. A , , and Russo, M. A. (1981). Relative effects of furosemide and ethacrynic acid on ion transport and energy metabolism in slices of rat kidney-cortex. NaunynSchmeideberg’s Arch. Pharmacol. 317, 90-96. Vargas, F. F. (1968). Filtration coefficients of the axon membrane as measured by hydrostatic and osmotic methods. J . Gen. Physiol. 51, 13-27. Villamil, M. F.. Rettori, V., Yeyati, N . , and Kleeman. C. R. (1968). Chloride exchange and distribution in the isolated arterial wall. Am. J. Physiol. 215, 833-839. Whittam, R., and Willis. J. S. (1963). Ion movements and oxygen consumption in kidney cortex slices. J . Physiol. (London) 168, 158-177. Whittembury, G . (1968). Sodium and water transport in kidney proximal tubular cells. J . Gen. Physiol. 51, 303s-314s. Wilson, T. H. (1954). Ionic permeability and osmotic swelling of cells. Science 120, 104-105.
This Page Intentionally Left Blank
CURRENT TOPICS IN MEMBRANES AND TRANSPORT, VOLUME 30
Role of Cytoplasmic Vesicles in Volume Maintenance G . D . V . VANROSSUM Department of Pharmacology Temple University School of Medicine Philadelphia, Pennsylvania 19140
M . A. RUSSO Dipartimento di Medicina Sperimentale Universitd La Sapienza 00161 Rome. Italy AND
J . C . SCHISSELBAUER Department of Pharmacology Yale University School of Medicine New Haven, Connecticut 06510
1.
INTRODUCTION
Living cells need to regulate their water content, even under apparently isosmotic conditions, because of the osmotic forces which result from high intracellular concentrations of large, charged molecules (Macknight and Leaf, 1977). Regulation of cellular water content is important not only for the maintenance of an appropriate physical size of the cell and its membrane-bound organelles, but also for maintaining appropriate concentrations of soluble materials such as ions, substrates, and soluble enzymes. Its importance is emphasized by the fact that swelling of cells and their organelles is the earliest histological alteration seen in many tissues subjected to a wide range of pathological insults in vivo, and that similar swelling is readily induced in vitro by the general reduction of metabolic activity produced by cooling or by more specific inhibition of various aspects of cell function, e.g., energy metabolism and ion transport. 45
Copyrighl Q 1987 by Academic Press. lnc. All nghls of npmluction in any form reserved.
G.D. V. VAN ROSSUM ET AL.
46
In the first article of this volume, Macknight has discussed the role of active transport of ions across the plasma membrane in preventing entry of water from an isosmotic medium, and hence in maintaining cell volume. As originally proposed by Wilson (1954) and Leaf (1956), the distribution of Na+ appears to be of prime importance, with a major contribution being made by the coupled transport of Na+ and K (Na-K transport) that is sensitive to ouabain. Yet there is a considerable body of evidence suggesting that inhibition of the Na-K transport with ouabain does not always totally abolish regulation of cellular volume in isosmotic conditions. Several proposals have been made to account for the ouabain-resistant fraction of volume regulation (see Section 111), but the attention of Garfield and Daniel (1977a,b) and of ourselves (Russo er al., 1976, 1977) was particularly attracted by the coincidence between extrusion of water continuing in the presence of ouabain and the occurrence of characteristic vesicles in the cytoplasm. Independently, both groups suggested that the vesicles might provide a vehicle for the expulsion of water, by exocytosis, when Na-K transport was inactive. The concept of volume and osmotic regulation by intracellular, membranebound vesicular organelles is well established in the case of the contractile vacuoles of certain unicellular organisms, while the tonoplast provides a means whereby the plant cell may have a sufficient water content to maintain turgor without dilution of its cytoplasm. In mammalian cells, segregation of excess water in cisternae of the endoplasmic reticulum is a well-recognized pathological response to toxic treatments, leading eventually to “vacuolar degeneration. Collection of the water in vesicles, in this case, could represent a protection of cytoplasmic components from dilution. More generally, membrane-bound organelles and vesicles are recognized to be of great importance in cellular physiology and biochemistry, with particular importance for intracellular distribution of materials and for their transport to and from the exterior by way of endo- and exocytosis. Some vesicles, such as the storage granules of secretory cells, have densely packed, almost solid contents. But many vesicles are clear when seen in the electron microscope and must contain a high proportion of water. This last clearly constitutes a compartmentalization of water and, if the vesicles undergo exocytosis, they must also expel some of this water together with any associated solutes from the cell. Conversely, endocytotic vesicles will carry water into the cell. While the concept of a role for vesicular structures in the overall compartmentalization and homeostasis of cellular water is, in general terms, incontrovertible, the clear demonstration of a specific role in the net expulsion of water in a particular instance is not simple. A general difficulty arises from the nature of the evidence available. While the role of the relatively small vesicles in situ can only be judged from static, morphological observations with the electron microscope, movement of water can only be judged by biochemical methods. It is impossible directly to correlate such different lines of evidence. One is therefore dependent +
”
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
47
on parallel observations of vesicle formation and water movements in tissues subjected to conditions which affect one or other of these functions as specifically as possible. A bridge between the two lines of evidence is, to some extent, offered by the finding that vesicles isolated from intracellular membranes possess mechanisms allowing them to accumulate ions, which must be followed by water. Agents able to alter transport activities of the vesicles may then be examined for their effect on the Occurrence of vesicles in situ, on the one hand, and cellular water movements on the other. In this article we start by considering some instances in which the role of intracellular membranes or vesicular systems in the compartmentalization or movement of water is either well established or strongly suspected. We then consider evidence that intracellular vesicles are implicated in regulation of the volume of vertebrate cells when Na-K transport is inhibited under isosmotic conditions.
II. EXAMPLES OF WATER MOVEMENT AND COMPARTMENTALIZATION
A. Contractile Vacuoles The contractile vacuoles of protozoa and algae are membrane-bound, cytoplasmic organelles which regularly fill with fluid derived from the cytoplasm and discharge it to the exterior of the cell by a contraction mechanism. Different organisms may have one or several vacuoles and the period of the filling and expulsion cycle varies with the organism and the ambient conditions. As an example, the alga, Vucuoluriu virescens, has a cycle lasting 49-62 sec at 23°C (Heywood, 1978). Contractile vacuoles are found predominantly in freshwater organisms (Bovee and Jahn, 1973), where there is a strong tendency for water to enter the cells, and they clearly have an important role in volume regulation. The mechanism effecting transfer of water from the cytoplasm into the contractile vacuole differs from one organism to another. Three examples may be taken. In the trypanosomatid, Leptomonus collosomu, an array of cytoplasmic tubules, the spongiome, attaches to the vacuole during filling and is apparently concerned in water segregation (Linder and Staehelin, 1979); the nature of the driving force for water segregation into the spongiome elements is not known. In many types of amoeba, numerous vesicles of small diameter (< I pm) as well as tubular structures are believed to empty their contents into the contractile vacuole by fusing with it (Hickinger, 1973); the small vesicles and tubules may themselves arise from the endoplasmic reticulum. In V. virescens the contractile vacuole is found in close association with the Golgi apparatus. “Subsidiary vesicles” derived from the Golgi system fill the region between it and the contractile vacuole, and are seen to fuse with the latter (Heywood, 1978). Pre-
48
G. D. V. VAN ROSSUM ET AL.
sumably, the vacuole receives at least some of its contents from the Golgi cisternae and, indeed, under appropriate conditions of fixation Heywood (1978) observed similar electron-dense matter in the Golgi, subsidiary, and contractile elements. He suggested that this substance may have a high affinity for water (as, for instance, mucopolysaccharides)and could lead to the entry and retention of water within the membranous structures. Direct studies of the ionic composition of the contents of contractile vacuoles have been made with Paramecium caudarum, using X-ray microanalysis. All the common inorganic elements of biological material were present (Schmitz et a f . , 1985) and it seems possible that ion transport across the membranes of the Golgi and other vesicles may contribute to water entry.
B. Tonoplast of Plant Cells Structural rigidity in nonwoody parts of plants is largely dependent on the turgor pressure developed by the intracellular water against the nondeformable cell wall. Most plant cells contain only a small amount of cytoplasm lining the cell wall and the bulk of the cellular water is accumulated within the large, central vacuole, or tonoplast . Transport properties of the tonoplast membrane lead to an accumulation of a wide range of solutes within the vacuole, the osmotic activity of which results in accumulation of water. This device permits a relatively large body of water to be present in the cells, allowing development of turgor, while avoiding dilution of cytoplasmic materials. The tonoplast is derived from the Golgi apparatus during growth, small vesicles from the apparatus fusing to form the large vacuole. Among the transport systems in the tonoplast membrane is a proton-translocating mechanism which is expressed as an H -stimulated adenosine triphosphatase (Marin and Gidrol, 1985). Proton uptake appears to be coupled to C1- movements, for the ATPase activity is stimulated by C1and inhibited by the anion-exchange inhibitor, 4,4’-diisothiocyano-2,2’stilbenedisulfonic acid (DIDS). It will be seen below that mammalian Golgi membranes also transport H + and C1- (Section IV). +
C. Vesicles in Vertebrate Tlssues 1. SECRETION
Here we consider vesicles formed from intracellular membranes which expel their contents, also derived solely from intracellular materials, by exocytosis at one pole of the cell. In several cases such vesicles are derived from the Golgi apparatus and their contents become highly concentrated and electron dense, e.g., adrenal medulla, exocrine and endocrine pancreas, anterior pituitary, mast
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
49
cells, and platelets (Farquhar and Palade, 1981; Orci and Perrelet, 1977; Lawson et al., 1975). Others contain more electron lucent, and so presumably aqueous, contents, e.g., casein-containing vesicles of mammary gland, bile acid-induced vesicles of liver, and vacuoles induced by intense stimulation of several exocrine glands (Patton and Keenan, 1975; Jones et af., 1979; Mills and Quinton, 1981; Tapp, 1969, 1970). Based on studies with isolated chromaffin granules, a mechanism has been proposed for osmotic lysis of granules (Pazoles and Pollard, 1978) which has been suggested to account for exocytosis from a variety of intact cells (Pollard et al., 1977; Brown et al., 1978; Orci and Malaise, 1980). The membranes of chromaffin granules bear a proton-translocating adenosine triphosphatase which accumulates H+. This ATPase is stimulated by anions able to penetrate the membrane, e.g., C1- , which also accumulate in the granule and eventually lead to osmotically driven lysis of the membranes and release of contents. In intact cells, the release of granule contents to the medium is stimulated by C1- and inhibited by increasing the osmotic activity of the medium, e.g., by sucrose or the impermeable anion, isethionate. It is suggested that lysis occurs when the granules are closely adjacent to the plasma membrane and that the external medium acts as the source of C1- and water. It would seem that the system could also accumulate intracellular CI- and water when presented with the former at higher than usual concentrations as a result of some derangement of the internal milieu; subsequent exocytosis could then contribute to expulsion of excess C1and water. A case in point may be seen in the serous cells of tracheal submucosal glands. Secretory granules apparently deep in the cytoplasm are seen to be swollen with clear material after intense hormonal stimulation (Mills and Quinton, 1981). Like Tapp (1969, 1970), Quinton (1981) suggested that this results from a disturbance of intracellular ionic contents arising from the stimulation. However, an ion transport system different from that of the chromaffin granules may be concerned, for Tapp (1969) showed that replacement of Na+ ,but not of C1- , inhibited vesicle swelling. The liver is of interest because it elaborates materials that must be secreted at opposite poles of the hepatocyte. Protein components of the blood are synthesized and secreted at the sinusoidal (basal) surface by exocytosis, while bile is secreted at the apical pole. Sinusoidal secretion is inhibited by agents acting on microtubules (Orci et al., 1973; Redman et al., 1975), while bile secretion is blocked by agents inhibiting microfilament function (Phillips et al., 1975). The water of the bile is probably derived from more than one source (Boyer, 1980) but it is relevant here that stimulation of bile production by infusion into rats of taurocholic acid results in a marked increase of Golgi-related membranes and that a number of clear vesicles appear near the canaliculi (Jones et al., 1979). We have also noted a marked incidence of pericanalicular vesicles in liver slices incubated in vitro with glyco- and taurocholic acids (DeFeo, Holowecky, and
50
G. D. V. VAN ROSSUM ET AL.
van Rossum, unpublished observations). The Golgi apparatus may thus have a role in bile secretion, possibly via vesicular exocytosis. In the light of the suggested importance of C1- movements in exocytosis, it may be significant that replacement of external C1- by NO,- also reduces bile formation (Graf and Peterlik, 1975). Absence of external Na+ also reduces bile salt-stimulated bile flow (Van Dyke et af., 1982), but this could arise from the requirement of Na+ for accumulation of bile salts in the cells. 2. TRANSCELLULAR WATERMOVEMENT Three ways have been described in which this can depend on intracellular vesicular structures. Most obviously, vesicles taking up material at one cellular surface may traverse the cell and release their contents by exocytosis at another surface. A clear example of this possibility is seen in the transfer across the hepatocyte of a number of serum proteins (e.g., polymeric immunoglobulin A, haptoglobulin) which appear in the bile in concentrations higher than in serum (Mullock and Hinton, 1981). Endocytotic uptake from the sinusoidal surface requires a glycoprotein, secretory component, which is synthesized in the liver and acts as a receptor. Vesicles bringing about the transfer do not appear to interact with other membranous components within the cell. The movement of vesicles from blood to bile, as indicated by appearance of secretory component in the latter, continues in the absence of the specifically transferred blood proteins. Such a system must inevitably transfer certain amounts of water and ions across the cell. Water and solutes also traverse endothelial cells of muscle capillaries by this first mechanism (Simionescu, 1979). Vesicles with clear contents line the plasma membranes and are also seen in the cytoplasm; their diameter is approximately 700 A. Time course studies with electron-dense markers, varying in molecular weight (diameter) from 1900 (20 A) to 400,000 (1 10 A), indicate the movement of material from the blood and across the endothelial cell in the vesicles, with electron-dense material appearing in the pericapillary space after 30 sec to 3 min, depending on the size of the marker molecule. The second possible mechanism for transcellular movement of materials by involvement of vesicles is also illustrated by these same cells. The vesicles can form channels which completely traverse the endothelium, either when a single vesicle fuses simultaneously with luminal and abluminal membranes of the cell, or when a chain of two to four fused vesicles is formed across the cell. Movement of material via separate vesicles and by channel formation can be seen in the same cell, with the relative incidence of each varying with the position of the endothelium in the capillary bed (Simionescu, 1979). Neither the vesicles nor the channels appear to communicate with other membranous elements of the cells, but presumably represent a means whereby movement of water and solutes across a cell can be compartmen-
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
51
talized and so occur without directly affecting cytoplasmic volume and composition. A third mechanism for transcellular movement is illustrated by the increased transport of water and ions across epithelia, such as amphibian bladder and renal collecting ducts, which are sensitive to neurohypophyseal hormones. In this case, the hormones stimulate passage of the material through the body of the cell by an increase in permeability specifically of the apical plasma membrane. The permeability increase is mediated by cyclic AMP and is associated with an increase of both endo- and exocytosis at the apical border (Masur et al., 1972). Morphological and freeze-fracture studies suggest that hormones stimulate the association with the apical membrane of cylindrical structures of the cytoplasm, with transfer of membrane-associated protein particles to the apical membrane (Wade el al., 1981). It is believed that these particles are channels for water movement and that their insertion into the apical membrane is responsible for the increased permeability. While the intracellular structures are thus carriers of water channels rather than of water itself, their association with the apical membrane, like exocytosis of vesicles, is prevented by increased osmotic activity of the medium. The cylinders then swell markedly to form vesicles with clear contents (Kachadorian et af.,1981). This suggests the presence in the cytoplasmic membranes of transport mechanisms which have an analogous function, with a similar sequence of events, to those proposed by Pollard and co-workers for exocytosis (Section II,C, 1). This section has shown several cases in which vesicles can be important in the segregation and transfer of water through a cytoplasmic environment. In many instances there appears to be an association with, or origin from, the Golgi apparatus as well as an important role for H and/or C1- transporting mechanisms of the vesicular membranes leading to accumulation of water in the vesicles. +
111.
VOLUME REGULATION IN VERTEBRATES
Under isosmotic conditions, cell volume regulation in vertebrates is to a large extent explicable by passage of water directly through the plasma membrane in response to the ionic balance maintained by the Na-K transport system (see Macknight, in the preceding article). But observations on a variety of tissues indicate that at least part of the ability to regulate cell volume is resistant to ouabain or the absence of K + from the medium (Kleinzeller and Knotkova, 1964; Garfield and Daniel, 1977a,b; Macknight, 1968; Russo e t a l . , 1976, 1977, 1985). This is prima facie evidence that the persisting fraction of volume-regulating capacity is not dependent on the Na-K transport system and it is for this fraction that there is evidence for a role of cytoplasmic vesicles in the segregation and eventual expulsion of cellular water.
G. D. V. VAN ROSSUM ET AL.
52
Several proposals have been made to account for the ouabain-resistant extrusion of water under isosmotic conditions, based mainly on experiments with renal cortex (see Macknight and Leaf, 1977). An early suggestion of a “cryptic pump” (Willis, 1968) proposed that the Na-K transport system is incompletely inhibited because some of the transporter molecules are at sites of restricted access to ouabain or restricted diffusion of K + away from the cell membrane. Furthermore, a much used animal model, the rat, is rather insensitive to ouabain and Cooke (1981a,b) showed that water extrusion from renal cortical slices of rabbit, unlike rat, was slowed by ouabain. However, even in rabbit slices a significant extrusion persisted. Moreover, results we describe below suggest that the properties of the water extrusion occurring in the presence of ouabain are different from those seen without ouabain, indicating that different mechanisms may be concerned. Of the other proposals to account for ouabain-resistant water extrusion, two will be mentioned briefly. Whittembury and Proverbio (1970) proposed that a “second Na+ pump,” insensitive to ouabain and uncoupled from K + movement, is responsible for a net extrusion of Na+ ,and hence volume regulation, in renal cortex. Recently, Mm’n et al. (1985) have detected an Na-dependent adenosine triphosphatase in basolateral plasma membranes of rat proximal tubules. The system is inhibited by furosemide and they suggest that it may form the basis for a second pump. Completely different is the proposal of Kleinzeller (1972) and Rorive er al. (1972), that a contractile mechanism at the cell periphery, which requires Ca2 and ATP, may act to restrict cell volume mechanically and, by contraction, to extrude water and associated ions that have entered during a period of induced swelling. The above work was all done with little or no published morphological control. Garfield and Daniel (1977a), using uterine smooth muscle, and we, using liver and kidney slices, have shown that the presence of ouabain is associated with the Occurrence of cytoplasmic vesicles as well as with continued extrusion of water. Evidence for the proposal that the vesicles provide a vehicle for water extrusion in a number of tissues will now be described. +
A. Ouabain-Resistant Volume Reguiatlon In Liver 1. EXPERIMENTAL MODEL The work has been done mainly with tissue slices of rat liver. Slices can be prepared rapidly and are convenient to handle and analyze for ions and water. Criticism is sometimes made of their metabolic adequacy, especially that they are prone to hypoxia in their central regions. Such comments have usually referred to kidney slices (e.g., Balaban et al., 1980; Mandel, 1982), but Cooke (1979) has shown that under appropriate conditions (thin slices, appropriate
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
53
temperature, thorough oxygenation of the medium) there is no indication of hypoxia as judged by physiological criteria, and our morphological findings argue similarly (Russo et al., 1985). Slices of brain tissue are also a good model (Nicholson and Hounsgaard, 1983). Liver slices are, in our experience, more robust than renal cortical slices. They maintain good and constant levels of ATP, ions, and water as well as an excellent ultrastructural appearance after 60-180 min at 38°C (Russo et al., 1977; van Rossum, 1972; van Rossum and Russo, 1981, 1984). Some damaged cells appear, but they are scattered through the thickness of the slice and are not concentrated in the central region. The experimental procedure was that of Mudge (1951). Slices are first cooled to 0- 1°C in an isosmotic Ringer solution; this inhibits general metabolic and iontransporting activity and thus allows an equilibration of ionic gradients accompanied by entry of water. Upon restoration of activity by warming to 38"C, the ionic composition and water content are restored to values close to those of fresh, unincubated tissue (Elshove and van Rossum, 1963; van Rossum, 1970a). Perhaps the most remarkable aspect of this system is the rapid and complete recovery of cellular ultrastructure, from the swollen and apparently disorganized state seen at 1°C (Fig. la) to an appearance at 38°C which is nearly indistinguishable from liver tissue in vivo (Fig. lb). 2. RECOVERYWITH AND
WITHOUT
OUABAIN
The structural recovery in the absence of ouabain is already well advanced after 5 min at 38"C, at which time a significant extrusion of water, Na+, and C1- has occurred (Fig. 2). A minimum water content is attained by about 15 min. A peculiarity of the liver is that net reaccumulation of K + is delayed until after 15 min (Elshove and van Rossum, 1963; Judah and McLean, 1962; van Rossum, 1970b), a feature which allows it to be seen that the extrusion of Na+ can be considered to have two components. The first is equimolar with the extrusion of C1- and occurs together with the loss of water; the continued extrusion of Na+ after water and C1- have attained a steady state is approximately equimolar with the reaccumulation of K . The restoration of composition and ultrastructure at 38°C is prevented by agents blocking mitochondria1 energy conservation (Elshove and van Rossum, 1963; Russo et al., 1977). although 0, consumption and ATP levels must be reduced by more than 50% before any effect on ions and water is seen (van Rossum, 1972). By contrast, incorporation of amino acids into slice protein is very sensitive to smaller degrees of respiratory inhibition (van Rossum et al., 1976). Experiments with ouabain used concentrations (2-5 mM) which completely prevented net reaccumulation of K . Maximal extrusion of water was reduced by about 50%, but the time course of recovery was rather similar to that of controls without ouabain (Fig. 2). Omission of K + from the medium as an +
+
54
G.D. V. VAN ROSSUM ET AL.
55
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE b 40C
1
41a
-
3 >
300
t m
X
fE
200
0
0
100
0 t
3b
0
81
t at 30% (mh)
FIG. 2. Time course of (a) water and (b) ionic content of rat liver slices during incubation at 38"C, after 90 min incubation at 1°C. General conditions were as for Table I. Points at zero time Controls; --0--, represent the tissue composition at the end of the incubation at 1°C. (a) -O-, incubated with 2 mM ouabain. The uppermost pair of lines shows the total water content, the middle pair shows intracellular water, and the lowest pair shows extracellular water. (b) The intracellular contents of the ions are shown for control slices only. -. K; -0-. , Na ; --0--,CI - . +
+
alternative means for inhibiting Na-K transport caused a similar, partial inhibition of water extrusion both in the presence and absence of ouabain, but significant extrusion of water, N a + , and C1- persisted (Russo er al., 1977). Slices incubated with ouabain for 60 min at 38"C, like controls, showed excellent structural recovery, but differed in displaying groups of rounded vesicles in the region of the Golgi apparatus, between nuclei and canaliculi (Fig. Ic and d). Similar vesicles appear in the absence of K + , while control slices show FIG. I . Micrographs of rat liver slices incubated under various conditions. (a) Electron micrograph of portion of cell after incubation for 90 min at 1°C. Cisternae of the endoplasmic reticulum are dilated and mitochondria swollen. Ground substance of cytoplasm and nucleoplasm are rather electron clear. (b) Electron micrograph after incubation for 90 min at 1°C followed by 60 min at 38°C. Detail of lateral canalicular region of two cells. The ultrastructure of all organelles, especially the endoplasmic reticulum and mitochondria, has recovered from the swelling at 1°C. (c) Optical micrograph of slice after incubation for 90 min at 1°C and 60 min at 38°C in the presence of ouabain (2 mM). This illustrates the large number of vesicles formed in the presence of ouabain. (From van Rossum and Russo, 1984: with permission of Springer-Verlag. Berlin and New York.) (d) Electron micrograph of slice incubated as in (c). This detail of the vesicles shows that they are found near the bile canaliculi (BC) and in close association with the Golgi apparatus ( G ) .
56
G.D.V. VAN ROSSUM ET AL.
only a few vesicles of smaller dimensions. These observations are not peculiar to slices subjected to preincubation at O-lOC,for slices incubated with ouabain at 38°C as soon as possible after preparation (1-2min) maintain cell volume well and contain similar vesicles to those of Fig. lc. These and other results (Russo et al., 1977), as well as the increased bile flow induced by ouabain in perfused liver (Graf et al., 1972), led us to propose that, by accumulating water from the cytoplasm and expelling their contents into the canaliculi, the vesicles could contribute to the regulation of cell volume in the presence of ouabain. In the absence of water extrusion driven by Na-K transport at the basolateral membrane (Blitzer and Boyer, 1978), water could be accumulated in the vesicles more rapidly than they dispose of it by exocytosis. A new steady state would be established in which the vesicles are more obvious than usual (Fig. 3). The following sections describe experiments to test aspects of this suggestion by parallel studies of water extrusion and ultrastructure.
FIG. 3. P r o p a l for two mechanisms of volume regulation in rat liver cells. In the absence of ouabain, most water extrusion follows Na-K transport at the basolateral membrane. When Na-K transport is inhibited, the accumulation of CI- and Na+ in vesicles closely associated with the Golgi apparatus leads to an entry of water and vesicular swelling. The vesicles pass to the canaliculi, by a microfilament dependent mechanism, and there expel their contents by exocytosis. The suggested site of action of a number of treatments is indicated.
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
57
3. ORIGIN OF VESICLES IN THE PRESENCE OF OUABAIN The vesicles seen after 60 min at 38°C with ouabain or in the absence of K are of a wide range of sizes and, depending on whether they have clear or granular contents, seem to have different origins. The clear vesicles, which include most of the smaller range and about half of the larger, arise from the vesiculations of the endoplasrnic reticulum that are very evident during the incubation at 1°C (Fig. la). After 5 min at 38"C, the reticulum is extensively recovered, except in certain regions of the cytoplasm where terminal cisternae show increased vesiculation. At 15 min, larger, rounded vesicles are apparent; the Golgi system is also distinguishable and its elements are sometimes seen in communication either with swollen terminal cisternae of the reticulum or with the larger, rounded vesicles (Fig. 4a). The majority of these elements have clear contents. In experiments in which electron-dense particulates that only enter the cells by endocytosis (e.g., thorium dioxide, ferritin) were included in the medium, no particles were seen in the vesicles at 15 min (Russo et al., 1977, and unpublished observations), so that the contents were derived only from intracellular sources. After longer incubation times (30-60 min), the proportion of larger vesicles with granular contents increases and these also contain the electron-dense particles. Since the particles are initially found only in the sinusoidal spaces, the vesicles containing them must be formed by endocytosis at this border of the cells. Vesicles without granular contents remain free of the electron-dense particles. While the two types of vesicle thus appear to arise differently, they are nevertheless frequently seen to fuse with each other and thus form a single system which becomes expanded upon inhibition of Na-K transport. +
4. RELATION OF VESICLES TO VOLUME CONTROL
The model in Fig. 3 predicts that extrusion of water in the presence of ouabain should be prevented either if segregation of water from the cytoplasm into the vesicles is inhibited or if exocytosis is blocked. In the former case, formation of vesicles should be reduced while in the latter an accumulation of vesicles in the cytoplasm might be expected. Since ouabain-resistant water extnrsion is accompanied by a net loss of Na+ and C1- (Russo et al., 1977), transfer of water from the swollen cytosol across the vesicular membranes may follow a movement of one or both of these ions. Replacement of C1- in the medium with NO, - or SO,2 - completely prevented water extrusion in the presence of ouabain (Table I) and vesicles were almost completely absent (Fig. 4b). Restoration of C1- after 15 min at 38°C led rapidly to onset of water extrusion (Fig. 5 ) and to appearance of vesicles, with clear contents, in the vicinity of the canaliculi (Fig. 4c and d). Replacement of mediurn Na+ by Li , in the presence of ouabain, also reduced the incidence of +
50
G. D. V. VAN ROSSUM ET AL.
FIG.4. Electron micrographs of liver slices incubated for 90 min at 1°C followed by incubation at 38°C under the conditions indicated. (a) Incubation for 60 min at 38"C, with 2 mM ouabain present throughout. This shows the close association of the vesicles with Golgi apparatus (G), with regions of proposed origin.from the Golgi and of vesicle fusion indicated with arrows. (b) Incubation at 38°C for
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
59
Time at 38OC (min) *p< .05 FIG.5. Restoration of water extrusion from liver slices upon introduction of CI- to the incubation medium. Ouabain (2 mM) was present throughout and the incubation for 90 rnin at 1°C and the first 15 min at 38°C was in Cl--free medium with either NO3- or S042- (as indicated) being the major anion. At “Transfer,” some of the slices were restored to CI- medium, still with ouabain. Micrographs of these slices are illustrated in Fig. 4c and d. The diagrams show the intracellular water contents. Note the complete absence of water extrusion in the CI--free media. *, Significantly different from water content at 15 min.
vesicles, although less markedly (van Rossum and Russo, 1984). The diuretic, furosemide, which inhibits cotranspon of Naf and C1- at the plasma membrane (see later articles), was found to cause effects similar to the absence of C1- . At 1 mM, it reduced the rate of water extrusion in the presence of ouabain and much reduced the number of cytoplasmic vesicles (van Rossum and Russo, 1984). 60 min, with the slices being in CI--free medium containing 2 mM ouabain throughout. Nitrate is substituted for CI - in the medium. Note that the pericanalicular regions are completely devoid of the vesicles normally seen in the presence of ouabain. A similar picture is obtained if 1- or S042- is substituted for CI - . (c) Incubation at 1°C and for 15 min at 38°C was in CI--free (NO3-1 medium containing 2 mM ouabain; the slices were then transferred to CI- medium, also containing ouabain, for a further 30 rnin at 38°C. Vesicles are reformed around the canaliculi (BC) together with the onset of water extrusion (see Fig. 5). (d) As (c) but the CI--free medium contained S042- as the major anion. The reappearance of vesicles near the canaliculi is evident. [Parts (c) and (d) are from van Rossum and Russo. 1984; with permission of Springer-Verlag. Berlin and New York.]
OF
VARIOUSTREANENTS
ON
TABLE I INTRACELLULAR WATER CONTENT OF LIVERSLICES IN THE PFESENCEAND ABSENCE OF OUABAIN Intrawllular water content (kg waterlkg dry wt)"
Then at 38°C for 60 min ouabain plus
9ominat
Treatment
1°C
Control
Treatment
ouabaia
treatment
Cytochalasin B (100 pglml) Cytochalasin E (33 pglml) Colchicine (1 m M ) CI--free medium, with NO3- as substitute
2.42 -t 0.33 2.77 -t 0.07 2.73 2 0.10 2.60 2 0.09
1.48 2 0.10 1.75 -t 0.13 1.68 2 0.11 1.52 2 0.08
1 . 6 0 f 0.17 1.92 f 0.08 1.61 f 0.08 1.70 f 0.08
1.72 f 0.10 1.96 f 0.06 1.98 f 0.17 2.00 2 0.06
2.19 2 O.llc 2.38 2 0.13C.d 2.15 2 0.25 2.47 -t 0 . 1 3 ~
(4 ( 8)
(18) (10) (15)
Intriicellular water was determined from the total tissue water, measured gravirnetrically, and the volume of distribution of inuli (see Russo et al.. 1976). Values are mean f standard error of the mean. b Slices were incubated in a medium containing 146 mM Na+, 5 mM K+, 1 mM Mg2+, 1.3 mM Ca2+, I61 mMC1- ,2 mM phosphate, 1 mM SO&, Tris (10 m M ,pH 7.4). It was gassed with 0 2 . C1-free medium contained NO3- instead of CI- . Ouabain was at 2 m M .Cytochalasins were added from stock solutions in dimethyl sulfoxide; final concentration in medium was 0.1% (vlv) DMSO. C Significantly greater than value with ouabain alone, by t test; p < 0.01. Significantly less than value after incubation at 1°C. by t test; p < 0.01.
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
61
Entry of materials into bile canaliculi can be inhibited by cytochalasins, but not by colchicine (see Section II,C,l). Cytochalasin B and, somewhat less effectively, D and E, markedly inhibited water extrusion in the presence of ouabain (Table I) and resulted in the presence of large numbers of vesicles (Fig. 6a). The latter were distributed throughout the cytoplasm, rather than being restricted to regions close to the canaliculi, and tended to be larger than with ouabain alone. In addition, the canaliculi became completely smooth in outline and devoid of microvilli. The impression given is that the vesicles are able to form, by accumulation of water, but are unable to pass to the canaliculi or to expel their contents. By contrast, colchicine had no effect on water extrusion or vesicle formation (Table I). The above results indicate an important role for ion movements in the formation of vesicles in the presence of ouabain and suggest that microfilaments are important in their exocytosis. There is a clear correlation between the Occurrence of water extrusion resistant to ouabain and vesicular formation. In the absence of ouabain, neither replacement of C1- by NO,- nor the presence of furosemide or cytochalasins caused a significant inhibition of water extrusion or K+ accumulation (Table I), indicating that the Na-K transport is able to account fully for volume regulation when it is active. Nevertheless, cytochalasin B produced an accumulation of vesicles similar to that described above, so that the mechanisms responsible for vesicle formation (i.e., transfer of ions and water) are functional even in the absence of ouabain, the vesicles being Seen because they are unable to undergo exocytosis. 5.
ROLE OF EXOCYTOSIS
Exocytosis of vesicular contents into the canaliculi is an important postulate of the proposed mechanism but the evidence for it is not yet fully convincing. The effects of cytochalasin €3 cited above are consistent with the suggestion. More direct is the finding that the electron-dense particulates discussed in Section III,A,3 are first confined to the sinusoidal spaces of the slices but appear later in intracellular vesicles and canaliculi (Russo et af.,1977 and unpublished observations). Moreover, the particles in the canaliculi are frequently associated with rounded collections of granular material, not surrounded by membranes, that resemble the vesicular contents (Russo et al., 1977) and probably represent the products of exocytosis. Instances of small, clear vesicles communicating with the canalicular lumina at the base of microvilli are not infrequently seen in the presence of ouabain (Russo et al., 1977), but could as well be endocytotic as exocytotic figures. There is an occasional instance in which we have observed a larger vesicle to fuse with a canaliculus (see also Section 111,A,6), but this is rare. However, it is likely that the fusion and stretching of the membranes is a very rapid event which produces figures too unstable to survive fixation of the tissue.
FIG.6. (a) Rat liver slice incubated for 90 min at 1°C and 60 min at 38°C in the presence of cytochalasin B (100 pg/mI) and ouabain (2 mM). Many vesicles, of various sizes, are scattered throughout the cytoplasm. Canaliculi (BC) are altered in shape and devoid of microvilli. (From van Rossum and Russo, 1981, with permission from Springer-Verlag, Berlin and New York.) (b) Slice of hepatoma 3924A after 120 min at 1°C and 60 min at 38°C without ouabain. A region between two cells is shown;vesicles are absent. (c) Slice of hepatoma 3924A, as in (b), except that I M o u a b a i n was present throughout. Vesicles are now visible.
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
63
6. ISOLATEDHEPATOCYTES For comparison with liver slices, we have performed experiments with rat hepatocytes in primary culture for 24 hr, in collaboration with J. L. Farber. Ouabain (2 mM) induces a marked loss of K during incubation for 60- 180 min at 37°C and induces formation of vesicles very similar to those seen in slices treated with ouabain (Fig. 7a). During culture, numbers of cells unite with formation of canaliculi between them and Fig. 7b shows an example of a large vesicle fusing with a canaliculus, in the presence of ouabain. Furthermore, phalloidin, which interacts with microfilaments, induces vesicle formation which resembles that caused by cytochalasins in liver slices (Russo er a l . , 1982). These results are so far consistent with the slice experiments. The above considerations indicate that the vesicles seen in liver cells are a consequence of reduced activity of Na-K transport and that they are not a peculiarity of the slice preparation. There is a close correspondence, as far as it has been tested, between the observed behavior of vesicles and that predicted from the proposal of Fig. 3. +
8. Other Tissues If our arguments above are correct, analogous mechanisms might play a more general role in cellular volume regulation. Evidence to test this has been sought in a number of tissues. I . HEPATOMA3924A
This transplantable rat tumor of the Morris series (Moms and Wagner, 1968) is poorly differentiated but retains some structural features of the hepatwyte, including the distinction between sinusoidal and canalicular regions. Experiments with slices show a capacity to regulate cellular volume and ions that is very similar to that of rat liver (van Rossum et al., 1971). With endogenous substrates only, ouabain (1 mM) completely inhibited net accumulation of K but reduced water extrusion by only 50%; this was associated with a more active appearance of the Golgi apparatus than in controls, and of large vesicles in the Golgi regions near the canaliculi (Fig. 6b and c), which led to the proposal of a role for the vesicles in volume regulation (Russo et al., 1976). The cells have a high glycolytic capacity and water extrusion, both with and without ouabain, was increased when glucose was added to the medium (Table 11). Moreover, a greater fraction of the water extrusion was resistant to ouabain in the presence of glucose, even though inhibition of K + accumulation by ouabain was undiminished. This suggests that ATP synthesized in the cytosol is more effective than mitochondria1 ATP as a source of energy for volume regulation in this tissue, especially for that fraction of water extrusion which is resistant to oua+
FIG.7. Electron micrographs of rat hepatocytes in primary culture for 24 hr and then incubated at 37°C for 240 min in the presence of 2 mM ouabain. Incubation was in Williams E medium. (a) Many clear, rounded vesicles are seen in the cytoplasm. Many fewer are seen if incubation is without ouabain. (b) Detail of reformed bile canaliculus (BC)recognized by the junctional complexes which limit it (arrows). Note apparent fusion of some vesicles with the canaliculus. N, Nucleus. Experiment conducted in collaboration with J. L. Farber.
65
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
EFFECT OF
OUABAlN ON
TABLE I1 CONTENTS OF SLICES
INTRACELLULAR WATER
OF VARIOUS
TISSUES
lntracellular water content (kg/kg dry wt) Then 60-min warm incubation" Tissue
I "Cb
Control
Rat hepatoma 3924A Rat renal cortex Rat lung 21-day fetus 2-4 hr postnatal Adult (>2 months) Duck salt gland
4.08 2 0.16 2.85 2 0.09
3.36 2 0.08 1.63 2 0.10
4.20 3.30 2.78 2.10 ~~~~~
t 0.15 t 0.16 2 0. I 1 2 0.09 ~
2.68 2.58 2.05 1.98 _
_
_
OuabainC
2 0.1 I 2 0.06 2 0.1 I 2 0.08 _
_
_
~
(n)
3.56 2 0.08d 1.63 2 O.Md
(32) (22)
3.63 3.02 2.43 2.41
(
~
2 0.16d,C 2 0.16. 2 0.07d.e 2 O.lOd.e
9)
( 9) ( 9)
(29)
~
Warm incubation was at 38°C for all tissues except renal cortex, which was at 25°C. Incubation medium in each case was that described in Table I, except that medium for the hepatoma also contained 20 mM glucose. Incubation at 1°C was for 120 min in the case of hepatoma and salt gland and 90 min for all others. Ouabain concentration was 0.5 m M for salt gland, 1 m M for hepatoma, and 2 m M for all others. Significantly different from value at I T , by t test; p < 0.01. Significantly different from control value, by f test; p < 0.01.
bain. Less directly, the finding suggests that the ouabain-resistant mechanism is able to work even in the absence of ouabain.
2. RENALCORTEX This has been the most frequently used tissue in studies of volume regulation, but suffers from the disadvantage of being rather sensitive to high temperatures (i.e., 38°C) and of being markedly heterogeneous with respect to cell types. As others have found (e.g., Macknight. 1968; Whittembury and Proverbio, 1970). volume recovery of cortical slices at 25"C, after swelling at 1°C was little affected by ouabain, despite complete inhibition of K reaccumulation (Table 11). That there may be two separate mechanisms for Na-K transport and water extrusion is suggested by our observation that the latter is less sensitive to inhibition of respiration and lowering of cellular ATP (Russo et al., 1985). Cells of proximal, distal, and collecting tubules in the slices showed a few, rounded membrane-bound vesicles in their apical regions at 25°C. The number of vesicles was very much increased in the presence of ouabain, when basal vesicles and dilatation of the intercellular spaces of the basolateral infoldings also occurred (Fig. 8a; also Russo et al., 1985). Fragments of cortical tubules, isolated by the method of Rasmussen (1975). also contained many vesicles in the presence of ouabain (Garcia-Caiiero, Rabinowitz, and van Rossum, unpublished observations). +
66
G. D. V. VAN ROSSUM ET AL.
FIG. 8. Electron micrographs of slices of rat renal cortex after incubation for 90 min at 1°C and 60 min at 25°C in the presence of 2 mM ouabain. (a) Proximal tubular cells contain many vesicles of various sizes. BM, Basement membrane; M, microvilli of brush border. (b) Detail of vesicles, showing fusion between vesicles themselves (single m w s ) and of vesicles with plasma membrane of basal infoldings (double arrows). (From Russo et al., 1985, with permission of Springer-Verlag, Berlin and New York.)
The extrusion of water from slices in the presence of ouabain (2 mM) was inhibited 40-60% when medium C1- was replaced by NO,- or I- and 30% by furosemide, while the number of cytoplasmic vesicles was greatly reduced (Russo et af., 1985). Cytochalasin B reduced water loss by 25% and somewhat increased the number and intracellular dispersal of vesicles (van Rossum, Emst, and Russo, unpublished observations). These findings are very similar to those with liver and two extra features of the vesicles in cortical slices emphasize their possible role in exocytosis. First, nearly all vesicles had clear contents, suggesting mainly aqueous materials and, second, many instances are seen of vesicles fusing with the membranes of the basolateral infoldings (Fig. 8b). We suggest that a vesicular mechanism may contribute to water expulsion from renal tubular cells, differing from that proposed for liver in that the water is expelled at the basolateral rather than the apical border. 3. MYOMETRIUM
Rangachari et af. (1973) found myometrial muscle to be resistant to ouabain and Garfield and Daniel (1977a) noted two types of vesicle in the cells-small
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
67
ones arranged along the plasma membranes in all conditions, and larger ones in the cytoplasm that appeared in the presence of ouabain; the latter bear some resemblance to those we have seen in liver cells. Garfield and Daniel proposed that particularly the smaller vesicles were involved in exocytotic extrusion of water. The effects on the vesicles of ion substitution were not examined, but water extrusion itself was not inhibited if C1- was replaced by NO,- and Na+ by Li (Rangachari et a / ., 1973). It will be important to determine the effects of these ions on the two types of vesicles in order to decide whether there is a role for either in volume control. +
4. LUNG
Slices of whole lung from late fetal (20 days gestation), postnatal (2-4 hr), and adult rats swell at 1°C and subsequently extrude water at 38°C by a mechanism which is only partially inhibited by ouabain (Mariani and van Rossum, 1985; and Table 11). Further studies with the fetal tissue indicate little or no inhibition when CI- is replaced by NO,- or when furosemide is added, in the presence of ouabain. With ouabain alone, several large vesicles appear. Ouabain also induces release of surfactant material from storage vesicles into the alveolar lumen, as shown by electron microscopy. The remaining empty vesicles stay open to the alveolar space and it is possible that these may be the origin of at least some of the clear vesicles seen with ouabain. The induction of secretion of surfactant material by ouabain could provide a connection between the hypothesis of Pollard and co-workers for exocytosis and a role in volume regulation, for the partial inhibition of water extrusion caused by ouabain would leave a greater availability of water and C1- in the cytoplasm. There could thus be an intracellular, rather than an extracellular, source of water and CI- to stimulate exocytosis. The system bears some resemblance to the stimulation of watery vesicles in serous cells of tracheal submucosal glands, but differs from it in that the swelling of serous cell vesicles is inhibited, not induced, by ouabain (Mills and Quinton, 1981).
5. AVIANSALTGLAND The hypothesis that ouabain-resistant volume regulation proceeds by a vesicular mechanism further requires that a tissue showing no ouabain resistance should not form the characteristic vesicles in the presence of ouabain. Slices of salt gland from Pekin ducks have been studied by a similar protocol to that described for other tissues. The intracellular water content after incubation for 120 min at 1°C is lower than that of other tissues and subsequently shows only a small decrease at 38°C (Table 11). Apparently these cells swell less at 1°C. The presence of ouabain, however, causes a considerable swelling beyond that seen at 1°C suggesting a complete loss of volume regulation when the Na-K transport
68
G. D. V. VAN ROSSUM ET AL.
system is inhibited. These slices show no evidence of formation of vesicles in the presence of ouabain (van Rossum and Russo, 1987).
IV. ION TRANSPORT BY INTRACELLULAR MEMBRANES Any role vesicles may play in controlling the dis$bution of cellular water is likely to follow from a movement of osmotically active materials across their membranes. In several instances referred to above, there is direct evidence that isolated organelles can actively transport ions, with the ATP-driven movement of H being frequently the primary mechanism which subsequently drives the movement of other substances, e.g., bioactive amines or C1- in chromaffin granules (Pazoles and Pollard, 1978; Wilkins and Salganicoff, 1981; Cidon et al., 1983). The requirement of C1- and Na+ for the formation of vesicles in the cells of liver and kidney incubated with ouabain (Sections III,A,4 and II,B,2) is indirect evidence for the importance of ion movements in expansion of the membranous elements of the cell into water-containing vesicles. The apparent association of the vesicles, especially in liver (Fig. 4a) and hepatoma 3924A (Section III,B,l), with the Golgi system suggests that isolated Golgi membranes are an appropriate source for the study of the transport properties of the vesicular membranes. Zhang and Schneider (1983) and Glickman et al. (1983) have shown the accumulation of H + , by a proton-dependent ATPase, in preparations of Golgienriched vesicles and the latter provide a preliminary indication of C1- uptake. We have studied the C1- uptake system in more detail, using vesicles derived from rat liver. The preparation is enriched at least 10-fold in marker enzymes for the Golgi apparatus and has endoplasmic reticulum as its main contaminant; electron microscopy indicates the presence of vesicles derived from both these sources. Our results confirm that C1- can be accumulated by a mechanism that specifically requires ATP and that is saturable by increasing external C1- (Table 111; see also Schisselbauer and van Rossum, 1985). An external C1- concentration of about 20 mM gives half-maximal accumulation. The accumulation is prevented by the anion ionophore, tributyl tin, suggesting that uptake occurs across a membrane and against an electrochemical gradient, while its dependence on proton movements is indicated by inhibition with the proton ATPase inhibitor, N,N’-dicyclohexylcarbrbodiimide(DCCD) and the proton ionophore, carbonyl cyanide rn-chlorophenylhydrazone (CCCP). In contrast to Glickman er al. (1983), we find little response to a K + diffusion potential, imposed by external K in the presence of valinomycin. Nevertheless, either K or Na stimulates the uptake and the ionophore, monensin, which induces exchange of H+ for other monovalent cations, also stimulates C1- uptake. +
+
+
+
69
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE TABLE 111
UPTAKE OF
c1- B Y GOLGI-ENRICHED VESICLES ISOLATED FROM RAT LIVER
Conditions tested A.
B.
C.
Concentration
-
No nucleotide ATP GTP CTP ATP + DCCD ATP Valinornycin (no ATP) ATP + CCCP ATP + tributyl tin ATP + monensin ATP ( I mM) in all Tris Na+ + Tris K + + Tris Na+ + K +
1 mM 1 mM
I mM 1 mM
+ 250 pM
1 mM
1 CLM 10 pM
10 pM 10 pM
150 mM
100mM 100 mM 50mM
+ 50mM + 50 mM + 100mM
CI- uptake in 15 min (nmol/mg protein)o 0.004
(n)
0.032 0.055 ? 0.006 0.057 2 0.009 0.124 & 0.032
(4) (4) (4) (4) (3)
0.60 ? 0.13 0.05 2 0.01 0.02 0.02 1.30 ? 0.27
(9) (7) (1) (1) (4 )
0.02
(10) (3) (4) (7)
0.055
?
0.214
_t
0.18
?
0.40 2 0.06
0.43 0.40
? ?
0.08 0.05
a Vesicles were separated by density gradient centrifugation and enrichment of galactosyl transferase was approximately 10-fold. Incubation was at 25°C in medium which contained, mainly, I50 mM K + , 40 mM N a + , 150 mM gluconate, 2 mM CI- (labeled with '6c1) and citrate-maleate buffer, in experiments A and B. In experiment C, the major cation concentrations were as shown. Values are means 2 standard error of the mean.
We conclude that an ATP-dependent accumulation of H is accompanied by C1- and that the subsequent exchange of H for external Na+ or K results in a net uptake of NaCl or KCl, which could induce water entry. Such ionic movements could occur in intact cells if the cytoplasmic concentration of C1- were high enough, inducing entry of water. The system could utilize whichever of the two cations predominated in its vicinity, and in the case of cells treated with ouabain this would be Na+ . Such a mechanism fulfills several of the requirements for a role in ouabain-resistant volume regulation which are imposed by the observations on vesicle behavior in the intact liver slices. +
+
V.
+
CONCLUSION
Vacuoles and vesicles are widespread in living cells. Those with watery contents necessarily contribute to the partition and distribution of cell water and, if they undergo exocytosis, will contribute in some measure to the net expulsion of water from the cell. Such factors play a major role in volume regulation in
70
G. D. V. VAN ROSSUM ET AL.
organisms possessing contractile vacuoles, but in vertebrates it is the movement of water through plasma membranes, driven by Na-K transport in isosmotic conditions and also by furosemide-sensitive cotransport systems in anisosmotic experiments, that appears to be most important. However, net extrusion of water from several vertebrate tissues continues to a considerable extent under isosmotic conditions even when the Na-K transport system is inhibited to the degree that no net accumulation of K + takes place. These conditions coincide with those under which cytoplasmic vesicles are often observed. At least in liver and some cells of renal cortex, there is a close correspondencebetween the expectations of a model for water extrusion by exocytosis and the appearance of the vesicles when water extrusion is altered, while the ion-transporting properties of Golgiderived membranes are consistent with the proposal. In this regard, it is noteworthy that the vesicles responsible for transferring water into contractile vacuoles (Section II,A), the tonoplast membrane (Section II,B), and many secretory granules are all derived from the Golgi apparatus and have in many cases been shown to display proton-transporting activity. These considerations suggest that, at the least, reduced activity of Na-K transport leads to a redistribution of cell water into membrane-bound vesicles which depends on ionic movements across the vesicular membranes. It remains possible that the association with the observed net extrusion of water is coincidental and that, as argued by Willis (1968) and Cooke (1981), a residual activity of the Na-K transport is actually responsible for the loss of cellular water. While it is difficult to exclude totally the persistence of a small activity of Na-K transport, we suggest that the properties of the water extrusion from liver and renal cortex continuing in the presence of ouabain or absence of medium K + , especially its specificity for C1- and inhibition by furosemide and cytochalasin, are so different from those of water extrusion in the presence of fully active NaK transport, that different systems are likely to be responsible. Direct evidence that the vesicles are involved in water extrusion is hard to obtain because of the difficulty of correlating morphological with functional evidence. However, the choleretic effect of ouabain (Graf et al., 1972) is consistent with the proposed mechanism in liver. Further evidence can be sought by electron microprobe analysis of the vesicle contents and, possibly, by antibody labeling of proteins in the vesicular membranes. Studies of volume regulation in cells genetically resistant to ouabain and in the avian salt gland, in which water extrusion shows no resistance to ouabain (Table II), will also be important (English et al., 1985). That different mechanisms for volume regulation may predominate when NaK transport is active or inhibited, as we suggest for liver and kidney, is not to say that some contribution from each mechanism may not play a part in each case. For example, some Golgi-associated vesicles are present in control liver slices, although not very evident, and are presumably concerned in the intracellular
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
71
transport and/or secretion of materials synthesized in the Golgi apparatus. These functions will involve some segregation of ions and water. But only when the cellular water levels are raised, as a consequence of some inhibition of the water extrusion dependent on Na-K transport, will the availability of cell CI- and water become sufficient to render the vesicles large and numerous. A possible physiological role for an ouabain-resistant mechanism is suggested by the finding that at least some of the humordl natriuretic substances occurring in mammals appear to act as inhibitors of the Na-K transport mechanism (Haddy and Pamnani, 1985; Martin et al., 1985). The amounts secreted are particularly marked in neonatal animals, during pregnancy and in some pathological conditions, such as hypertension (Hamlyn et al., 1985; Valdes, 1985). Modification of Na-K transport activity under such circumstances could compromise volume regulation unless a compensating mechanism, independent of Na-K transport, can come into play. REFERENCES Balaban, R. S., Soltoff, S. P., Storey, J. M., and Mandel, L. J . (1980). Improved renal cortical tubule suspension: Spectrophotometric study of 02 delivery. Am. J. Physiol. 238, F50-F59. Blitzer, B. L., and Boyer, J. L. (1978). Cytochemical localization of Na+,K+-ATPase in the rat hepatocyte. J. Clin. Invest. 62, 1104-1 108. Bovee, E. C., and Jahn, T. L. (1973). Taxonomy and phylogeny. In “The Biology of Amoeba” (K. W. Jeon. ed.). pp. 37-70. Academic Press, New York. Boyer, J. L. (1980). New concepts of mechanisms of hepatocyte bile formation. Physiol. Rev. 60, 303-326. 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. Proc. Nail. Acad. Sci. U.S.A. 75, 876-880. Cidon. S.,Ben-David, H., and Nelson. N. (1983). ATP-driven proton fluxes across membranes of secretory organelles. J . Biol. Chem. 258, 11684-1 1688. Caoke, K. R. (1979). Oxygen consumptions and potassium contents of slices of rat renal cortex. Q. J . EXP. Physiol. 64, 69-78. Cooke, K. R. (1981a). Ouabain and regulation of cellular volume in slices of mammalian renal cortex. J. Physiol. (London) 320, 319-332. Cooke. K. R. (1981b). Species difference in the effect of ouabain on cell volume recovery in renal cortex. In “Epithelial Ion and Water Transport” (A. D. C. Macknight and J. P. Leader, eds.), pp. 329-338. Raven, New York. Elshove, A., and van Rossum, G. D. V. (1963). Net movements of sodium and potassium, and their relation to respiration. in slices of rdt liver incubated in virro. J. Physiol. (London) 168, 531553.
English, L. H., Epstein, J . , Cantley, L., Hausman, D., and Levenson, R. (1985). Expression of an ouabain resistance gene in transfected cells; ouabain treatment induces a K + -transport system. J. B i d . Chem. 260, I 1 14-1 119. Farquhar, M.G., and Palade, G. E. (1981). The Golgi apparatus (complex)-(1954-1981)-from artifact to center stage. J. Cell Biol. 91, 77s- 103s. Flickinger, C. J. (1973). Cellular membranes of Amoebae. In “The Biology of Amoeba” (K. W. Jeon, ed.), pp. 171-199. Academic Press, New York. Garfield, R. E., and Daniel, E. E. (1977a). Relation of membrane vesicles to volume control and
72
G. D. V. VAN ROSSUM ET AL.
Na+ transport in smooth muscle: Effect of metabolic and transport inhibition on fresh tissues. J. Meckanochem. Cell Motil. 4, 113-155. Garfield, R. E., and Daniel, E. E. (1977b). Relation of membrane vesicles to volume control and Naf transport in smooth muscle: Studies on Na+-rich tissues. J. Mechanochern. Cell Moril. 4, 157- 176. Glickman, J . , Croen, K., Kelly, S., and Al-Awqati, Q. (1983). Golgi membranes contain an electrogenic H + pump in parallel to a chloride conductance. J. Cell Biol. 97, 1303-1308. Graf. J., and kterlik, M. (1975). Mechanism of transport of inorganic ions into bile. I n “The Hepatobiliary System-Fundamental and Pathological Mechanisms” (W. Taylor, ed.), pp. 4358. Plenum, New York. Graf, J., Korn, P., and Peterlik, M. (1972). Choleretic effects of ouabain and ethacrynic acid in the isolated perfused rat liver. Naunyn-Schmiedeberg’s Arch. Pharmacol. 272, 230-233. Haddy, F. J., and Pamnani, M. B. (1985). Evidence for a circulating endogenous Na+-K+ pump inhibitor in low-renin hypertension. Fed. Proc., Fed. Am. SOC. Exp. B i d . 44, 2789-2794. Hamlyn, J . M., Levinson, P. D., Ringel, R., Levin, P. A., HamiIton, B. P.. Blaustein, M.P..and Kowarski, A. A. (1985). Relationships among endogenous digitalis-like factors in essential hypertension. Fed. Proc., Fed. Am. SOC.Exp. Biol. 44,2782-2788. Heywood, P. (1978). Osmoregulation in the alga Vacuofaria virescens. Structure of the contractile vacuole and the nature of its association with the Golgi apparatus. J. Cell Sci. 31, 213-224. Jones, A. L., Schmucker, D. L., Mooney, J. S., Ockner, R. K., and Adler, R. D. (1979). Alterations in hepatic pericanalicular cytoplasm during enhanced bile secretory activity. Lab. Invesr. 40,512-517. Judah, J. D., and McLean. A. E. M. (1962). Action of antihistamine drugs in vifro. 11. Ion movements and phosphoproteins in whole cells. Biochem. Pharmacoi. 11, 593-602. Kachadorian, W. A., Muller, J., and Finkelstein, A. (1981). Role of osmotic forces in exocytosis: Studies of ADH-induced fusion in toad urinary bladder. J. Cell Biol. 91, 584-588. Kleinzeller, A. (1972). Cellular transport of water. I n “Metabolic Pathways” (L. Hokin, ed.), 3rd Ed., Vol. VI., pp. 91-131. Academic Press, New York. Kleinzeller, A., and Knotkovh, A. (1964). The effect of ouabain on the electrolyte and water transport in kidney cortex and liver slices. J. Physiof. (London) 175, 172-192. Lawson, D., Fewtrell, C., Gomperts, B., and Raff, M. C. (1975). Anti-immunoglobulin-induced histamine secretion by rat peritoneal mast cells studied by immunofenitin electron microscopy. J. Exp. Med. 142, 391-402. Leaf, A. (1956). On the mechanism of fluid exchange of tissues in vitro. Biochent. J . 62,241-248. Linder, J. C., and Staehelin, L. A. (1979). A novel model for fluid secretion by the trypanosomatid contractile vacuole apparatus. J . Cell Biol. 83, 371-382. Macknight, A. D. C. (1968). Water and electrolyte contents of rat renal cortical slices incubated in K-freemedia and in media containing ouabain. Biochim. Biophys. Acra 150, 263-270. Macknight, A. D. C., and Leaf, A. (1977). Regulation of cellular volume. Physiol. Rev. 57, 510573. Mandel, L. J. (1982). Use of noninvasive fluorometry and spectrophotometry to study epithelial metabolism and transport. Fed. Proc.. Fed. Am. SOC. Exp. Biol. 41,36-41. Mariani, M.F., and van Rossum, G.D. V. (1985). Regulation of cell volume and ionic content in slices of fetal rat lung. Fed. Proc.. Fed. Am. SOC. Exp. Biol. 44,642. Marin, B. P., and Gidrol, X. (1985). Chloride-ion stimulation of the tonoplast H+-translocating ATPase from Hevea brasiliensis (rubber tree). A dual mechanism. Biochem. J. 226, 85-94. Marln, R., Proverbio, T., and h v e r b i o , F. (1985). Active sodium transport in basolateral plasma membrane vesicles from rat kidney proximal tubular cells. Biochim. Biophys. Acra 814, 363373. Martin, B. C., Faure, H., and Siegenthaler, G. (1985). Indirect estimation of natriuretic factor: Its concentration and affinity constant for Na-K-ATPase. Biochem. SOC. Trans. 13, 256.
CYTOPLASMIC VESICLES IN VOLUME MAINTENANCE
73
Masur, S. K., Holtzman, E., and Walter. R. (1972). Hormone-stimulated exocytosis in the toad urinary bladder. Some possible implications for turnover of surface membranes. J. Cell Biol. 52, 211-219. Mills, J. W., and Quinton, P. M. (1981). Formation of stimulus-induced vacuoles in serous cells of tracheal submucosal glands. Am. J. Physiol. 241, C18-C24. Moms, H. P.. and Wagner, B. P. (1968). Induction and transplantation of rat hepatomas with different growth rate (including “minimal deviation” hepatomas). Methods Cancer Res. 4, 125-152. Mudge, G. H. (1951). Studies on potassium accumulation by rabbit kidney slices: Effect of metabolic activity. Am. J . Physiof. 165, 113-127. Mullock, 9. M., and Hinton, R. H. (1981). Transport of proteins from blood to bile. Trends Biochem. Sci. 6, 188-191. Nicholson, C., and Hounsgaard, J . (1983). Diffusion in the slice microenvironment and implications for physiological studies. Fed. Pror.. , Fed. Am. SOC. Exp. Biol. 42, 2865-2868. Orci, L., and Malaisse., W. (1980). Single and chain release of insulin secretory granules is related to anionic transport at exocytotic sites. Diabetes 29, 943-944. Orci, L., and Perrelet, A. (1977). Morphology of membrane systems in pancreatic islets. I n “The Diabetic Pancreas” (9.W. Volk and K. F. Wellmann, eds.), pp. 171-210. Plenum, New York. Orci, L., Le Marchand, Y , , Singh, A , , Assimacopoulos-Jeannet, F.,Rouiller, C., and Jeanrenaud, B.(1973). Role of microtubules in lipoprotein secretion by the liver. Nature (London) 244, 3032. Patton, S., and Keenan. T. W. (1975). The milk fat globule membrane. Biochim. Biophys. Acta 415, 273-309. Pazoles, C. J . , and Pollard, H. 9. (1978). Evidence for stimulation of anion transport in ATP-evoked transmitter release from isolated secretory vesicles. J. Biol. Chem. 253, 3962-3969. Phillips, M. J., Oda, M., Mak, E., Fischer, M. M.. and Jeejeebhoy, K. N. (1975). Microfilament dysfunctions as a possible cause of intrahepatic cholestasis. Gastroenterology 69, 48-58. Pollard, H. B., Tack-Goldman, K., P a l e s , 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. Narl. Acad. Sci. U.S.A. 74, 5295-5299. Quinton, P. M. (1981). Possible mechanisms of stimulus-induced vacuolation in serous cells of tracheal secretory glands. Am. J. Physiol. 241, C25SC32. Rangachari, P. K., Daniel, E. E., and Paton, D. M. (1973). Regulation of cellular volume in rat myometrium. Biochim. Biophys. Acta 323, 297-308. Rasmussen, H. (1975). Isolated mammalian renal tubules. I n “Methods in Enzymology” (B. W. O’Malley and S. G. Hardman, eds.), Vol. 39, pp. 11-20. Academic Press, New York. Redman, C. M., Banerjee. D., Howell, K., and Palade, G . E. (1975). Colchicine inhibition of plasma protein release from rat hepatocytes. J . Cell B i d . 66, 42-59. Rorive, G.,Nielsen, R., and Kleinzeller, A. (1972). Effect of pH on the water and electrolyte content of renal cells. Biochim. Biophys. Acta 266, 376-396. Russo, M. A., Ernst, S. A,, Kapoor, S. C . , and van Rossum, G. D. V. (1985). Morphological and physiological studies of rat kidney cortex slices undergoing isosmotic swelling and its reversal: A possible mechanism for ouabain-resistant control of cell volume. J. Membr. Biol. 85, 1-24. Russo, M. A , , van Rossum, G. D. V.,and Galeotti, T. (1977). Observations on the regulation of cell volume and metabolic control in virro; changes in composition and ultrastructure of liver slices under conditions of varying metabolic and transporting activity. J . Membr. Biol. 31, 267299.
Russo, M. A., Kane, A. B., and Farber. J. L. (1982). Ultrastiuctural pathology of phalloidinintoxicated hepatocytes in the presence and absence of extracellular calcium. Am. J . Parhol. 109, 133-144. Russo, M. A,, Galeotti, T . , and van Rossum, G.D. V. (1976). The metabolism-dependent mainte-
74
G. D. V. VAN ROSSUM ET AL.
nance of cell volume and ultrastructure in slices of Moms hepatoma 3924A. Cancer Res. 36, 4 160-4 174. Schisselbauer, I. C., and van Rossum, G. D. V. (1985). An ATP-dependent mechanism for CIuptake by Golgi-derived vesicles. Fed. Proc., Fed. Am. SOC. Exp. B i d . 44, 1044. Schmitz, M., Meyer, R., and Zierold, K. (1985). X-ray microanalysis in cryosections of natively frozen Paramecium caudarum with regard to ion distribution in ciliates. Scan. Electron Microsc. i, 433-445. Simionescu, M. (1979). Transendothelial movement of large molecules in the microvasculature. In "Pulmonary Edema" (A. P. Fishman and E. M.Renkin, eds.), pp. 39-52. American Physiological Society, Bethesda, Maryland. Tapp, R. L. (1969). The mechanism of watery vacuolation in the acinar cells of the submandibular gland. J . Cell Sci. 4, 55-70. Tapp, R. L. (1970). Anoxic and secretory vacuolation in the acinar cells of the pancreas. Q. J. Exp. Physiol. 55, 1-15. Valdes, R. (1985). Endogenous digoxin-immunoactive factor in human subjects. Fed. Proc., Fed. Am. SOC.Exp. Biol. 44, 2800-2805. Van Dyke, R. W., Stephens, J . E., and Scharschmidt, B. F. (1982). Effects of ion distribution on bile acid-dependent and -independent bile formation by rat liver. J . Clin. Invest. 70, 505-517. van Rossum, G. D. V. (1970a). Net movements of calcium and magnesium in slices of rat liver. J. Gen. Physiol. 55, 18-32. van Rossum, G. D. V. (1970b). Relation of intracellular Caz+ to retention of K+ by liver slices. Nulure (London) 225, 638-639. van Rossum, G. D. V. (1972). The relation of sodium and potassium ion transport to the respiration and adenine nucleotide content of liver slices treated with inhibitors of respiration. Eiochem. J. 129, 427-438. van Rossum, G. D. V., and Russo, M. A. (1981). Ouabain-resistant mechanism of volume control and the ultrastructural organization of liver slices recovering from swelling in virro. J . Membr. Biol. 59, 191-209. van Rossum, G. D. V., and Russo, M. A. (1984). Requirement of Cl- and Na+ for the ouabainresistant control of cell volume in slices of rat liver. J. Membr. Biol. 77, 63-76. van Rossum, G. D. V., and RUSSO,M. A. (1987). Volume regulation in slices of avian salt gland. In preparation. van Rossurn, G. D. V., GosBlvez, M., Galeotti, T., and Moms, H.P. (1971). Net movements of monovalent and bivalent cations. and their relation to energy metabolism, in slices of hepatoma 3924A and of a mammary tumour. Biochim. Biophys. Acra 245, 263-276. van Rossum, G. D. V., Holowecky, 0. 0.. and Moms, H. P. (1976). Relationship between energyrequiring processes and energy levels in slices of liver and hepatomata. Acra Med. Rom. 14, 262-28 1. Wade, J. B., Stetson, D. L., and Lewis, S. A. (1981). ADH action: Evidence for a membrane shuttle mechanism. Ann. N.Y.Acad. Sci. 372, 106-1 16. Whitternbury. G., and Proverbio, F. (1970). Two modes of Na extrusion in cells from Guinea pig kidney cortex slices. Pfliigers Arch. 316, 1-25. Wilkins, 3. A,, and Salganicoff, L. (1981). Participation of a transmembrane proton gradient in 5hydroxytryptamine transport by platelet dense granules and dense-granule ghosts. Eiochern. J. 198, 113-123. Willis, 1. S. (1968). The interaction of K ,ouabain and Na+ on the cation transport and respiration of renal cortical cells of hamsters and ground squirrels. Biochim. Biophys. Actu 163,516-530. Wilson, T. H. (1954). Ionic permeability and osmotic swelling of cells. Science 124, 104-105. Zhang, F.,and Schneider, D. L. (1983). The bioenergetics of Golgi apparatus function: Evidence for an ATP-dependent proton pump. Biochem. Biophys. Res. Commun. 114, 620-625. +
CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOI.UME 30
The Cell Cytoskeleton: Possible Role in Volume Control JOHN W.MILLS Department of Anatomy Dartmouth Medical School Hamver, New Hampshire 03756
1.
INTRODUCTION
A substantial body of information has accumulated to support the pump-andleak hypothesis for steady state maintenance of cellular ions and volume. The essence of this model is that operation of an energy-dependent Na+:K+ exchange pump located in plasma membranes counteracts the dissipation, via “leak” pathways, of the respective electrochemical gradients for bulk ionic species (i.e., Na+, K + , and Cl-). The site for these ion exchanges is presumably the membrane surrounding the cell, the transporters dispersed throughout the membrane in the case of nonepithelial cells, or arranged uniquely in either the apical or basolateral domain in epithelial cells. Thus the maintenance of cell volume (and the ion fluxes underlying this process) can be aptly described as an inherent property of the cell membrane. With such a paradigm it is clear that investigations into the mechanism of ion fluxes across membranes can involve studies on whole cells, membrane vesicles derived from cells, and membrane fragments. A possible role of cytosolic components in the mechanism of ion fluxes across the cell has only recently received attention. Most of the interest stems from the rapid increase in our knowledge of the diversity and specific location of cytosolic proteins. Proteins that are especially of interest in membrane transport studies are those grouped in the category of “cytoskeleton.” This stems from what may be two important properties of the cytoskeleton. One is that the proteins could be linked directly to the cell membrane, and thus able to act on it or the integral membrane proteins (Geiger, 1983; Dentler, 1981). Another is that certain parts of the cytoskeleton may be able to “contract” and 75
Copyright Q 1987 by Academic Press, Inc. All righls of rcpmducuon in any form reserved.
76
JOHN W. MILLS
generate force and, while clearly affecting cell shape, may also alter volume (Kleinzeller, 1972). The following article is meant to give an overview of the research related to the possible role of the cytoskeleton in regulating ion transport events or fluid flux across cell membranes. Only selected examples have been chosen and no attempt has been made to be exhaustive. Research on the cytoskeleton, cytoskeleton-associated proteins, and cytoskeleton-membrane interactions is progressing at a rapid rate (for reviews see Geiger, 1983; Bennett, 1985; Dentler, 1981; Jacobsen, 1983; Dustin, 1984). Hopefully many of the questions raised here will soon be answered.
II.
CELL SHAPE AND THE CYTOSKELETON
It is quite clear that changes in cell shape are associated with changes in organization of different components of the cytoskeleton. One example can be seen in cultured epithelial cells growing to confluency. When Madin-Darby canine kidney (MDCK) cells are seeded at a low density such that they exist individually or have an extensive free surface, the arrangement of F-actin and microtubules is quite different than when the epithelial monolayer is fully confluent (Mesa et af., 1980). Other cell types, both epithelial and nonepithelial, show changes in cell shape that are correlated with changes in the cytoskeleton including cells undergoing mitosis, cells migrating on a substrate, and cells involved in phagocytosis (Albertini and Herman, 1984). Support for the role of the cytoskeleton in determining and maintaining cell shape comes from studies using drugs which directly alter the cytoskeletal proteins. The two most commonly used are the cytochalasins and colchicine. The cytochalasins bind to actin and alter polymerization and cross-linking (MacleanFletcher and Pollard, 1980). Cells treated with cytochalasin show characteristic morphological alterations (Godman and Miranda, 1978). These include loss of microvilli, formation of membrane blebs, and cell retraction (rounding) with the appearance of long, thin retraction fibers. Associated with these changes is an extensive alteration of the microfilament (actin containing) network. Upon removal of the cytochalasin, the cells resume a normal morphology and actin distribution. A change in cell shape can also be induced in fibroblasts by exposure to colchicine (Goldman et af., 1973). This drug causes depolymerization of microtubules. When fibroblasts are exposed to taxol, a drug which increases the rate and extent of microtubule polymerization (Schiff et al., 1979), cell shape is also altered (Green and Goldman, 1983). Thus microtubules may play a role in maintaining and altering cell shape. However, this role may not be universal
THE CELL CYTOSKELETON
77
since treatment of epithelial cell lines with colchicine or taxol did not alter cell shape (Goldman er al., 1973; Green and Goldman, 1983).
111.
MEMBRANE-CYTOSKELETAL INTERACTIONS
The key for assigning a role to the cytoskeleton in mediating changes in cell shape is that the cytoskeletal proteins be linked to the membrane. One of the best examples of such an interaction comes from studies on the erythrocyte cytoskeleton. An extensive research effort has produced a detailed description of how actin, existing as oligomers in the red cell cytoplasm, is linked to the cell membrane. The linkage involves spectrin and band 4.1 with spectrin making the link to the membrane by association with ankyrin (Bennett, 1985). Interestingly, ankyrin binds to an integral membrane glycoprotein termed band 3 which contains an anion transport pathway. In nonerythroid cells the link between the cell membrane (and/or integral membrane proteins) is not so well defined. However, much indirect evidence suggests that a similar actin-cell membrane linkage, mediated by spectrin-like proteins, exists (Geiger, 1983). One area involving an actin-membrane interaction that has been extensively studied in nonerythroid cells (Bennett, 1985; Geiger, 1983) is the structural basis for the existence of clustered receptors and the phenomenon known as patching and capping that occurs after ligands bind to uniformly distributed receptors on cells. In both attached and nonattached cells patches become closely associated with actin filaments or stress fibers (Albertini and Anderson, 1977; Bourguignon and Singer, 1977; Ash et al., 1977). Capping can be prevented by treatment with cytochalasins. In addition, after patching and capping, the surface receptors become more resistent to detergent extraction, indicating they are bound to the underlying cytoskeletal proteins. The identity of the proteins making the link between actin and the membrane is not known. Recently, however, evidence has been produced that a spectrin-like molecule is involved. In the T-lymphoma cell line, BW B147, patching and capping is induced by exposure to a monoclonal antibody to the integral cell membrane glycoprotein gp 180 (Bourguignon et af., 1985). As mentioned for other instances of patching and capping, the amount of gp 180 in the membrane-associated cytoskeleton (detergent extracted) increases in direct proportion to the number of patched/capped cells. In addition there is a similar increase in the amount of fodrin in the membrane-associated cytoskeletal fraction. Fodrin is a spectrin-like molecule that binds calmodulin and actin. The key point in this study was that, by the use of two nonionic detergents with different extraction properties, a gp 180-fodrin complex free from actin could be isolated from uncapped cells but during capping the complex became associated
78
JOHN W. MILLS
with the actin-containing cytoskeleton. Thus the integral membrane protein gp 180 may exist in close association with fodrin and this molecule play a key role in anchoring the complex to the actin cytoskeleton. The fact that the amount of membrane glycoprotein associated with actin changes is important since it indicates that, under steady-state or stimulated conditions, the amount of membrane protein interacting with actin could also change and this could alter the position or activity of the bound molecule.
IV. RELATIONSHIP BETWEEN CELL VOLUME AND CELL SHAPE In spite of the large body of work on the control of cell shape, the relationship between cell volume control processes and cell shape changes has received little attention. It is reasonable to assume that at least in the cases where shape changes are rapid and extensive (Hsie and Puck, 1971; Westermark and Porter, 1982; Willingham and Pastan, 1975), cell volume may be altered as well. As suggested by Melmed et al. (1981), it is possible that alteration of cell shape is coordinated with volume control processes. They pursued this question by analyzing shape and volume changes in the 5774.2 macrophage. In this cell line a shape change, marked by formation of protuberances, occurs after exposure to colchicine. There is also a reduction in volume of approximately 20%. A similar reduction in volume occurs when the 5744.2 macrophage is exposed to dibutyryl-CAMP but in this case no protuberances form. With colchicine, the volume reduction was shown to be sensitive to SITS, an inhibitor of C1 transport. It was proposed, therefore, that microtubule disassembly may activate a SITS-sensitive transport system. The ionic basis of the CAMP-stimulated volume reduction was not analyzed, so no mechanism was suggested. What was demonstrated was that the two agents had at least partially separate pathways since colchicine elicited a response in a protein kinase-deficient mutant line whereas cAMP did not. Melmed et al. (1981) made two important suggestions based on this study. The first was that, since many peptide hormones stimulate cAMP and alter shape, volume changes may accompany this hormone stimulation. The second was that the protuberance formed by colchicine treatment was similar, morphologically, to the uropod formed by motile cells. Thus, volume changes may occur regularly in motile macrophages. Another example of a correlation between changes in shape and volume comes from studies on human erythrocyte ghosts (Johnson et al., 1980). As the salt concentration of the medium bathing the ghosts increases, the ghosts change shape and eventually assume an echinocyte form. The shape change is accompanied by a volume reduction. A similar correlation was also demonstrated for the shape and volume change induced by an alteration in pH. Interestingly,
79
THE CELL CYTOSKELETON
contraction of the erythrocyte spectrin-actin lattice, obtained after extraction of the ghosts with Triton, could be demonstrated under the same conditions that induced the volume change. In contrast, a dissociation of volume and shape was reported in goose erythrocytes (Nikinmaa and Huestis, 1984). In these cells exposure to the ionophore A23 187 in the presence of a Ca2 concentration greater than 10 - M resulted in a clear morphological alteration to the echinocyte form. Under this condition the cellular water content was identical to untreated cells. A well-studied example of the correlation between changes in cell shape, the cytoskeleton, and volume is the acrosome reaction of echinoderm sperm. When the sperm comes in contact with egg jelly or is exposed to the calcium ionophore A23187 a long thin process rapidly (< 10 sec) forms which is several times the length of the nonactivated sperm (Inoue and Tilney, 1982). Associated with this process formation is the rapid assembly of actin monomers into actin filaments within the core of the process. Also associated with this elongation is an increase in volume of the periacrosomal region (Dan and Hagiwara, 1964; Tilney and Inoue, 1982). In addition, depolarization of the membrane, gain of Na+ and CI , loss of H + , and an alkalinization of the cytoplasm (Schackmann et af., 1978) occurs during activation. Based on these results, Tilney and Inoue (1985) proposed that the force required to drive the extension of the acrosomal process resulted from a combination of water entry and growth of actin filaments. The scheme for this process is as follows: Entry of Na and CI - after depolarization is accompanied by osmotic water movement; as protons are eliminated (perhaps mediated by Na /H exchange) actin monomers are released from the profilactin storage pool (Tilney el af., 1978); these monomers may exert an osmotic effect and also assemble at the growing tip of the filaments; the filaments then act, in combination with the increased water in the acrosomal region, on the acrosomal membrane to contribute to extension. This model clearly does not involve the cytoskeletal changes directly with the alterations in membrane ion fluxes that must occur. In fact, the change in the actin is thought to occur as a secondary effect of cytosolic alkalinization brought about by a stimulation of H efflux (Tilney et al., 1978). The postulated role for osmotically driven water entry, combined with actin assembly mediated by a pH change, as the driving force for elongation of the acrosomal process is intriguing since stimulation of the Na+ / H + exchanger can lead to volume increases and pH changes (Siebens, 1985). Thus activation of the echinoderm sperm may include, as a key event, activation of a membrane Na+ /H exchanger. One possible problem with the above interpretation is that the volume measurements were conducted on fixed specimens (Dan et d..1964) or included only the periacrosomal region (Inoue and Tilney, 1982; Tilney and Inoue, 1982). Thus a quantitative analysis of volume change in the whole sperm was not performed. At least some of the size increase seen in this region could be related to +
+
+
+
+
+
+
80
JOHN W. MILLS
the movement of water from one part of the sperm cytoplasm to another rather than an actual entry of water from the external medium. Since the formation of an actin gel from a sol in v i m effectively extrudes water (Pollard, 1976), this could occur in the sperm and provide the force for moving the water into the acrosomal region.
V. POSSIBLE ROLE OF THE CYTOSKELETON IN THE MECHANISM OF CELL VOLUME CONTROL There are at least two possible roles for the cytoskeletal proteins in the mechanism of cell volume control. One involves the possibility that the actin and actinassociated proteins are potentially contractile. Through a connection with the cell membrane, the contractile apparatus could exert enough tension to act as a resistive force that prevents swelling (Heubusch et al., 1985). Alternatively, the cytoskeletal proteins might interact with membrane proteins and this interaction may involve control of the activity or the insertion/retrieval of these proteins (Siman et al., 1985; Insel and Koachman, 1982; Hays, 1983). The membrane proteins could include ion channels, pumps, or exchangers such as the band 3 anion transport pathway already mentioned.
A. Contractile Mechanism In Nonmuscle Cells The concept that a contractile mechanism could be involved in volume control has its origins in the original mechanicochemical hypothesis proposed by Kleinzeller and co-workers (1972). This model envisioned that the cell membrane was involved in the extrusion of water and electrolytes via an integral contractile mechanism. A major criticism of this model was that estimates of cellular hydrostatic pressure and membrane tension were well below that required either for the cell to balance the colloid osmotic pressures of the impermeant ions, or for the membrane to withstand the pressure if it existed (Macknight and Leaf, 1978). However, the key feature of the analysis of membrane properties, as pointed out by MacKnight and Leaf (1978), was that any mechanism that acted to link the membrane to the cytoplasm would have the effect of reducing the radius of the cell and thus reducing the tension in the membrane. A recent review of membrane cytoskeletal interaction indicates that at least in the erythrocyte, an adequate number of cytoskeleton-membrane linking proteins exist that, if uniformly spaced, would act to reduce effectively the radius of individual membrane elements (Jacobson, 1983) For example, erythrocyte spechin binds to the membrane via ankyrin which is bound to the integral membrane
THE CELL CYTOSKELETON
’
81
protein band 3. There are lo5 copies of ankyrin and if these were all in contact with band 3, there would be 10s; attachment sites. Using the value of Westerman et al. (1961) for the surface area of the erythrocyte and assuming that the bonds to the membrane are uniformly distributed a 40-nm spacing can be estimated. This would have an impact on the membrane tension that must exist in order to withstand the colloid osmotic pressure. This is due to the fact that the membrane-cytoskeleton binding pattern results in an effective reduction in the diameter of the membrane that could form a vesicle between bonds to approximately 20 nm. Another important point obtained from these calculations, as pointed out by Jacobson (1983), is that the spacing of 40 nm results in another constraint. The minimum size of a phospholipid vesicle is approximately 50 nm in diameter. The fact the spacing results in a radius of less than that indicates that the bonding of the cytoskeleton to the membrane may aid in preventing bulging or membrane blebbing caused by a build-up of hydrostatic pressure generated either by the influx of water or the action of a contractile mechanism. This implies that any “strength” of the membrane may lie in the bond between the integral membrane protein and the linking protein, thereby imparting stiffness to the membrane, and also in the force generation that occurs between contractile proteins. The above discussion deals exclusively with the erythrocyte. There is evidence that a spacing of membrane-cytoskeleton interactions similar to that in the erythrocyte exists in the microvilli of the brush border of intestinal epithelium (Geiger, 1983) although the evidence of actual binding (as compared to close association) is not as well developed in this model and the spacing has not been worked out. These results indicate that the cytoskeleton, via a linkage to the cell membrane, could add stability to the membrane. The cytoskeletal proteins, actin, spectrin, ankyrin, etc., could thus exist as an internal scaffolding which, due to the many links throughout the cell, possibly aid in counteracting the swelling pressure of the intracellular colloid. A similar mechanism was proposed by Heubusch et al. (1985) to explain nonideal osmotic behavior of human erythrocytes exposed to anisotonic media. They reported that the cells did not behave as predicted for an osmometer over a range of 200-700 mOsm. Ideal osmometric behavior was obtained after treatments known to disrupt the spectrin-actin lattice. The second possibility, that an actual contraction (i .e., isotonic contraction) aids in expulsion of ions and water, has not received much experimental support. It is very difficult in nonmuscle cells to see movement analogous to the contraction or shortening seen in muscle cells. One place where contraction may occur is in the cytoplasm of macrophages. These cells move and change shape. A model of this movement (Stossel et al., 1981) involves a coordinated gel-sol transition of the cytoplasm. The sol portion of the cytoplasm is forced to the leading edge by a contraction in the gel. The gel portion of the cytoplasm contains large
82
JOHN W. MILLS
amounts of cross-linked actin filaments and myosin. Thus the force necessary to produce streaming of the sol could be achieved by a contraction of the actin bundles mediated by the myosin. This contraction, which would produce a pressure change, results in deformation of the membrane. Whether or not a volume change occurs is not known but was hypothesized by Melmed et al. (1981).
A well-known example of nonmuscle contraction occurs during cytokinesis. During anaphase a ring of actin filaments assembles just beneath the cell membrane. This ring also contains myosin and a-actinin (Fujiwara and Pollard, 1976; Fujiwara e?al., 1978). Thus the elements necessary for force generation (actin and myosin) pulling in the membrane (via a link with a-actinin) are present. Contraction is also thought to play a role in neural tube formation (Karfunkel, 1974). The presumptive neural ectoderm begins to change shape and then invaginate, curl into a tube, and separate from the ectoderm to form the neural tube. Cell elongation appears to be dependent on microtubules. It is thought that the force necessary to produce the rolling up of the epithelial sheet is generated by a contraction of the microfilaments present in the apical region of the epithelial cells since treatment with cytochalasin prevents invagination and can reverse the tube formation once initiated. Another part of the cell in which contraction may take place is at the sites where actin is organized into bundles known as stress fibers or actin cables. The stress fibers are known to contain actin, myosin, and a-actinin (Lazarides and Burridge, 1975; Lazarides, 1976). In addition, the myosin and a-actinin have an alternating periodic arrangement along the fiber. Evidence for contraction comes primarily from experiments on glycerinated cells. Using such a preparation, combined with laser beam microsurgery, Isenberg et al. (1976) demonstrated that the now-isolated fibril would shorten when an ATP-containing contraction medium was added. Kreis and Birchmeier (1980) also employed the glycerinated cell model but used a different approach. Prior to glycerination rhodaminelabeled or-actinin was injected into the fibroblasts. The or-actinin organized into periodic banding along the stress fiber and withstood the glycerin treatment. Upon addition of ATP-containing contraction medium the stress fibers contracted as judged by a shortening of the distance between a-actinin bands. Although the data from these two experiments indicate that stress fibers have the capability to contract, whether they do so in intact cells is still not known (Bemdge, 1981). As pointed out by Kreis and Birchmeier (1980) they could not elicit contraction in nonglycerinated cells. Thus the contraction of the fiber may be an artifact of the glycerinated cell model. On the other hand, the contraction phenomenon elicited in these studies might actually represent an isometric contraction that occurs in the intact cell where tension development for maintenance of cell shape is the primary function. Harris et af. (1980) demonstrated that fibroblasts grown on a flexible substrate made from silicone rubber produced
THE CELL CYTOSKELETON
83
wrinkles in the sheet, indicating tension development. Thus the shortening of stress fibers seen in the glycerinated models may not occur in vivo. In summary, the possibility that the cytoskeleton could contribute to cell volume maintenance and ionic homeostasis by counterbalancing swelling or exerting a contractile force is an attractive hypothesis. However, some key points need to be established. These include data on whether in nonerythroid cells the units of the cytoskeleton are linked to the membrane at intervals close enough to reduce the tension that individual membrane “units” must withstand and whether isotonic contraction occurs in nonmuscle cells.
B. Cytoskeletal Control of the Activity or Number of Membrane Proteins Rather than playing a direct role in fluid extrusion via a contractile mechanism or resisting swelling pressure by tension, the cytoskeleton may play a role in controlling the number or activity of ion transport units in the cell membrane. An obvious example of where this could take place is in the erythrocyte where the anion exchanger (band 3 protein) is linked via ankyrin to the spectnn-actin lattice underlying the cell membrane (Bennett, 1985). However, little is known about the activity of this exchanger in relationship to changes in organization of the cytoskeleton (Branton et al., 1981). An immunologically similar band 3 protein has been identified in nonerythroid cells (Cox et al., 1985; Drenckhahn and Zinke, 1984; Kay et al.. 1983). Thus the framework for analyzing a direct role of the cytoskeleton in controlling the activity of this transporter in many cell types has been established. This is especially important since the anion exchanger appears to play a critical role in volume regulatory responses and regulation of intracellular pH (Siebens, 1985). Control of the activity (other than ion transport) of a membrane event by the cytoskeleton has been proposed in S49 lymphoma cells (Insel and Koachman, 1982). In this cell, treatment with cytochalasin B enhances P-adrenergic and cholera toxin-induced increases in CAMP. In addition, cytochalasin B decreases the affinity of the receptors for P-adrenergic agonists. Cytochalasin treatment does not alter the number of P-adrenergic receptors nor does it affect the response to forskolin, which activates the catalytic component of the adenylate cyclase. Since both the binding to P-adrenergic receptors and receptor-mediated stimulation of adenylate cyclase require the nucleotide coupling protein of the adenylate cyclase complex, Insel and Koachman (1982) proposed that microfilaments may regulate this protein. Another postulated regulatory role for a cytoskeletal protein has been proposed for the actin-binding protein fodrin (brain spectrin). In rat brain synaptosomes, Ca2+ stimulates an increase in [3H]glutamate binding and this is associated with
84
JOHN W. MILLS
a change in maximum binding rather than affinity (Siman et af., 1985). The increase can be inhibited by incubation with EGTA or leupeptin, a thiol proteinase inhibitor (Baudry et d . , 1983). When synaptic membranes are incubated with 100 pM calcium, proteolysis of a previously distinct band on SDS-PAGE occurs. This proteolysis is blocked by leupeptin (Baudry et al., 1981). The Ca2+-sensitive protein has a molecular weight similar to fodrin. In addition, anti-fodrin antibodies block both the Ca2 -induced proteolysis and increase in [3H]glutamatebinding (Siman et al., 1985). Based on these results, Siman eta!. (1985) proposed that degradation of fodrin leads to an uncovering of membrane receptors and thus plays a direct role in regulating the number of glutamate receptors, Microtubules could also be involved in control of the activity or number of transport units. As noted earlier, Melmed ef al. (1981) proposed that depolymerization of microtubules activated an anion pathway in the 5744.2 macrophage. There is extensive indirect evidence that microtubules or tubulin interact with cell membranes (Dentler, 1981; Bernier-Valentin et al., 1983; Bhattacharyya and Wolff, 1985). This could be direct, i.e., tubulin being an integral part of the membrane, or via some linking protein. In this regard it is interesting to note that brain ankyrin is also a tubulin-binding protein (Bennett, 1985). Further evidence that microtubules may be involved in membrane transport events comes from studies of the sodium current in squid giant axons (Matsumoto et al., 1984). When axons were perfused with colchicine, the Na+ current was decreased. When substances that support microtubule assembly (tax01, D,O, DMSO) were perfused, the Na current increased. Finally, partial restoration of Na currents in tubulin-depleted axons could be accomplished by perfusion with tubulin and complete restoration was achieved by perfusion of tubulin in combination with cAMP and a 260K protein that acts as a microtubule crosslinker (Murofushi et af., 1983), all in a medium that favors microtubule assembly. Interestingly, in light of the present discussion, was the proposal that the axoplasmic microtubules play a role in generating the Na current via a crosslinking to the axolemma (Matsumoto er al., 1984). Another possible mechanism that might involve the cytoskeleton is a vesicle insertion/retrieval process. The vesicles would exist in the cytoplasm in close proximity to the cell membrane and contain the appropriate pathways for ions or water. Insertion of vesicles containing water channels is a major component of the “shuttle” hypothesis (Wade, 1980) that attempts to explain the change in osmotic water permeability that occurs after exposure of the toad urinary bladder to vasopressin or other agents that raise cellular cAMP levels (Hays, 1983). A cytoskeleton-mediated insertion of vesicles containing transport pathways into plasma membranes has been proposed as the mechanism responsible for the regulatory volume decrease seen in cells of Necturus gallbladder after the initial swelling upon exposure to dilute media (Foskett and Spring, 1985; see also +
THE CELL CYTOSKELETON
85
Larson and Spring, this volume). A microfilament-mediated vesicle insertion mechanism that would deliver preformed Na channels to the apical membrane has also been recently proposed for rabbit urinary bladder (Loo et al., 1983). Microtubules also appear to play a key role in vesicle-mediated secretion. The site where microtubules are involved is thought to be distal to the Golgi apparatus, where secretory material is packaged into granules (Dustin, 1984). The actual exocytosis of the vesicle contents may not be controlled by microtubules but be a function of the actin-containing terminal web. However, Mueller et af. (1980) have proposed that, since colchicine inhibits vesicle-apical cell membrane fusion events observed after exposure of the toad urinary bladder to ADH, microtubules may play a role in this process. Thus. if vesicle insertion and retrieval is a mechanism by which ion transport is regulated, then microtubules may play a critical role in the delivery of the transport units. Another type of vesicle insertion mechanism mediated by the cytoskeleton has been studied in detail by van Rossum and colleagues (1981). In this case rat liver cells swell during incubation at 1"C. Upon rewarming, fluid-filled vesicles appear and are expelled via exocytosis (see van Rossom et al., this volume). The recovery of volume, but not the appearance of fluid-filled vesicles, was inhibited by exposure to cytochalasin B. Colchicine had no effect. Thus, the role of actin filaments in this case may be related to "shuttling" the vesicle, which contains fluid destined to be expelled, to the cell membrane rather than controlling the number or activity of transport units.
VI.
CYTOSKELETON AND CELL VOLUME IN MDCK CELLS
With the above considerations in mind we initiated a study of the role that the cytoskeleton might play in volume and ion transport processes. For these studies we have utilized the Madin-Darby canine kidney (MDCK) cell line. This cell line was chosen because it has many well-defined characteristics in culture (Cereijido et al., 1978), appears to be of distal tubule origin (Valentich, 1981), and some information regarding the cytoskeleton is available (Mesa et af., 1980; Fey et al., 1984). In addition, many ionic transport characteristics of the cell membrane have been described (Stefani and Cereijido, 1983; Simmons, 1982; McRoberts et al., 1982), and the response to many agents that affect intracellular CAMP levels has been documented (Rindler et al., 1979). We have focused on the distribution of F-actin and microtubules. There is an extensive literature on the assembly, sensitivities, and possible functions of these filament systems. The MDCK cells have an extensive array of intermediate size filaments (Fey ef al., 1984). However, since little is known of the function that this filament system serves, we have not pursued the question as to how its
THE CELL CYTOSKELETON
87
organization and distribution changes during alterations in cell volume and ion content. F-actin, as revealed by staining with the F-actin-specific compound phallacidin linked to a fluorescent marker (Barak et al., 1980), is primarily distributed in three distinct areas in the confluent MDCK cell (Fig. 1). The base of the cell attached to the substrate is marked by the presence of stress fibers, a characteristic of cultured cells, but a cytoskeletal structure that is also present in cells in situ (Wong et al., 1983; Byers et al., 1984). The lateral margin of the cell, presumably that in association with the lateral cell membrane, demonstrates a continuous fluorescence from apex to base. At the apical cell surface there are small spots of fluorescence distributed across the entire cell. The stress fibers can be resolved as dark bands in phase-contrast microscopy and can be seen as bundles with the electron microscope. We have not been able to identify clearly the structures that account for the actin along the lateral cell membrane, but we have occasionally observed filamentous arrays in this area and similar membrane-associated F-actin has been identified in other cells and termed MAG for membrane-membrane F-actin aggregates at cell-cell contacts (Carley et ul., 1983) The F-actin seen as spots on the cell surface most probably represents actin arrayed in the core of the many microvilli that cover the surface of the cell (Burgess and Prum, 1982). All of these fluorescent patterns are altered by exposure to cytochalasins (Mills and Lubin, 1986), and, therefore, according to this criterion can be assumed to reflect the distribution of F-actin. Microtubules are distributed in a relatively even three-dimensional pattern throughout the cytoplasm (Fig. 2) marked by a decreasing density from nuclear region to cell periphery. This pattern is similar to that of other epithelial cells in culture. The one exception to this was the occasional presence of a cell or group of cells devoid of microtubules. We have not yet determined whether this indicates a cell that does not contain tubulin assembled into microtubules or is an artifact due to the inability of either the primary or secondary antibody to penetrate the cell. Since MDCK cells have an adenylate cyclase and elevate intracellular levels of cAMP in response to several agents, we investigated whether treatment with dibutyryl cAMP (db CAMP) altered cell volume, ion content, shape or the cell cytoskeleton. We found that 1 mM db cAMP caused a reproducible reduction in the volume of MDCK cells (Table I). Associated with the volume reduction, in FIG. I, Fluorescent micrographs of nitrobenzoxadiazole-phallacidin (NBD-PH) staining of Factin in MDCK cells. (A) Control cultun: showing NBD-PH staining at base of cell. Notice presence of stress fibers (arrows) and staining of lateral membrane (arrowheads). (B) Focus on cell surface at region of microvilli and junctional complex. Fluorescence is seen as continuous band around periphery and as mottling of cell surface. (C) NBD-PH localization after 90-min treatment with 5 pg/ml cytochalasin B. Stress fibers are not present. Level of focus at base of cells. Bar equals 10 pm. (From Mills and Lubin, 1986.)
88
JOHN W. MILLS
FIG. 2. Localization of tubulin in MDCK cells. Procedure as described in Mills and Lubin (1986). Microtubules can be seen throughout the cytoplasm (arrows).Bar equals 5 )~m.
both a time- and dose-dependent manner, is a large reduction in Naf and a smaller but consistent loss of Kf (Table I). This effect on volume and ion content could be mimicked by exposure of the cells to various agents that elevate intracellular levels of cAMP such as isobutylmethylxanthine, forskolin, and epinephrine (Mills and Skiest, 1985). After exposure to db cAMP or any of the agents that elevate cAMP in the cells, the organization of F-actin is altered. The most obvious change is seen at the base of the cells where the array of stress fibers disappears and the F-actin appears as dense bundles around the periphery of the cells (Fig. 3). This change in F-actin can be detected as early as fifteen minutes after exposure to db cAMP
89
THE CELL CYTOSKELETON TABLE I EFFECTOF db CAMP AND AMILORIDE ON CELL ~~
Treatment Control d b cAMP Amiloride dbcAMP Amiloride
+
Volume (pl HzOlcell)
Na (nmol/106 cells)
2.16 2 0.16
108 2 22 55 2 18 83 22 51 ? 15
1.71 2 0.14
I .92 ? 0.17 1.72 2 0.13
______~
~~
VOLUME AND 10NSa.b
~
_f
K (nrnol/lOh cells)
230 218 223 217
t 12 2 10 2 12 2 10
“a] ( m M )
[K]( m M )
48.8 2 6.0 30.0 ? 6.7 41.6 2 7.5
107.7 t 5.4
28.2 t 5.7
128.3 2 6.9 118.1 6.0 128.2 2 6.6 _f
~
Values are means and standard errors of five separate experiments From Allen er al. (1986).
and is mimicked and usually enhanced by exposure to isobutylmethylxanthine (Fig. 3) or forskolin. This time course correlates with the time course of the change in volume (Fig. 4) and the reduction in Na+ content after db cAMP treatment. As we previously pointed out (Mills and Lubin, 1986; Mills and Skiest, 1985), the fluorescent image at the apical cell surface was not identical to that seen in the control. However, analysis with the scanning electron microscope did not reveal any systematic differences in the organization of the apical membrane. Thus until a more detailed analysis of the cytoskeleton in this area is achieved, the main and possibly only effect of increased levels of cAMP on the cytoskeleton is an alteration in the F-actin associated with the microfilament system of the stress fibers. This implies differential sensitivity of the various Factin containing structures. Such a sensitivity has been proposed in the case of metabolic inhibition (Sanger er al., 1983). A general relaxation of the actin filament network within the cell might also occur but could not be detected by the phallacidin method. Initially, we felt that the volume change and alteration in F-actin occurred without any detectable change in cell shape. However, we have found that in several preparations, there is a change in morphology that can be seen with phase optics (Fig. 5). This change is not present in every preparation. The ability to detect the morphological alteration may relate to the fact that the volume reduction seen with db cAMP has a range of 10-30% in individual preparations. Since the cytoskeletal preparations are done on cells grown on coverslips and the volume change is not determined under these conditions, the lack of a detectable change in cellular morphology, separate from that seen with actin, could be related to the magnitude of the volume response. We could detect no change in microtubule array under any of the conditions that affect intracellular levels of cAMP (Mills and Lubin, 1986). In addition, if the cells were treated with colchicine at concentrations and times that result in a
91
THE CELL CYTOSKELETON
I
0
30
90
I I00
TIME (minl FIG. 4. Effect of db cAMP (CAMP)and cytochalasin B (cyto) on MDCK cell volume. Cells were incubated with [ 14CIurea for 90 min and then exposed to I mM db CAMP, 5 pglml cytochalasin B, or the two together. Volume was determined at indicated time points for the next 180 min. Although the rate of reduction of volume was more rapid with cytochalasin than with db cAMP the eventual end points were similar. (From Mills and Lubin, 1986 )
near complete loss of microtubules, there was no significant change in volume M (Table 11). However, at this concentration, there was a clear except at increase in the number of Trypan Blue-positive cells, indicating a toxic effect. Finally, treatment with colchicine did not inhibit the db cAMP effect on volume (Table 111). Thus, at least in confluent cultures of MDCK cells, disruption of microtubules does not alter steady state volume nor does it inhibit the db CAMPinduced volume reduction (and alteration of F-actin). Since actin appeared so markedly altered after treatment with db cAMP and this change was correlated with the volume and ion shift, we attempted to see if a direct alteration of F-actin had an effect on cell volume, ion content, and shape. When confluent cultures of MDCK cells were exposed to 5 pg of cytochalasin B, FIG. 3. NBD-PH staining of F-actin. (A) Base of cells after exposure to I mM db cAMP for 90 min. F-actin appears as bundles of bright fibers. (B) Base of cells after exposure to 0.1 mM isobutylmethylxanthine for 90 min. Large bundles of F-actin can be seen. (C) Apical surface of culture treated as in (A). Fluorescence appears brighter on the surface of some cells. Notice that cell surfaces are at various focal planes. Bar equals 10 p m . (From Mills and Lubin, 1986.)
92
JOHN W. MILLS
FIG.5. Phasecontrast micrograph of MDCK cells. Paired culture was exposed to normal Ringer’s medium (A) or medium containing 1 mM db cAMP (B). Notice that outlines of cells and intercellular spaces are easily discernible in (A) but that after exposure to db cAMP the spaces are not visible (arrows). Bar equals 10 pm.
93
THE CELL CYTOSKELETON TABLE I1 TO VARlOlJS CONCENTRATIONS EFFECTOF 90-min EXFOSURE ON CELLVOLUME o b MDCK CELLS
OF
COLCHICINE
pl H20/cell ( N = 9)' ~
Control
10-7 M
2.26 t 0.14
2.25 2 0.11
M
IO-'M
2.27 t 0.1 I
10-4
2.16 t 0.08
M
2.64 t 0.46
Mean 2 SEM.
there was a reduction in volume (Fig. 4) and ions (Allen et a l . , 1986) that reaches a steady state in 90 min. At this time, the normal array of F-actin was completely altered and the F-actin was now seen as bright fluorescent spots dispersed throughout the cytoplasm and along the lateral cell membrane (Fig. 1). When the cells were treated with cytochalasin B and db CAMP, the change in volume and ions was identical, both in the time course and amount, to cells treated with cytochalasin alone. Thus cytochalasin B mimicked the db cAMP response. It is possible that cytochalasin could be affecting the adenylate cyclase and thus altering the intracellular levels of cAMP (Insel and Koachman, 1982). However, using a radioimmunoassay for cAMP we have found that cytochalasin treatment for up to 90 min did not alter intracellular cAMP (Mills et al., unpublished observations). Alternatively, cytochalasin could be altering intracellular ion content and thus volume by its well-known inhibitory effect on the Na-dependent hexose transport (Kletzien and Perdue, 1973). There are two reasons why this is probably not the case. Since the medium we used does not contain any glucose, this carrier is not operating in the steady state; and we can duplicate the cytochalasin B effect with cytochalasin D which does not inhibit the carrier (Tannenbaum et al., 1977). In both guinea pig and Necrurus gallbladder, cAMP inhibits a Na /H exchanger in the apical membrane (Reuss and Petersen, 1985; Petersen et a l . , 1985). There is evidence to indicate that MDCK cells have a N a + / H + ex+
TABLE I11 VOLUMEOF MDCK CELLSAFTER EXFOSURE TO db cAMP OR db cAMP PLUS 1 0 - 5 M COLCHICINE
pl H20/cell Control
db cAMP
2.09 2 0.03
1.77 t 0.08
?
SEM (N = 3) Colchicine
+ db cAMP
1.75 2 0.07
+
94
JOHN W. MILLS
changer (Rindler and Saier, 1981). If the exchanger is inhibited and the Na+ pump continues to turn over, then this could result in a net reduction of Na+ which could then be followed by C1- and water. To test this we exposed the MDCK cells to amiloride, an inhibitor of Na+ /H+exchange (Benos, 1982), and monitored changes in volume and ions (Table I). Volume was reduced by exposure to amiloride but not to the same extent as with db cAMP alone. If amiloride and db cAMP were added together, the change was equal to that of db cAMP alone. Thus inhibition of the Na+/H+ exchanger partially mimicks the db cAMP effect. What is even more important is whether this effect of amiloride on volume also resulted in a change of F-actin. If actin is altered by amiloride, then this would imply that the change in actin is related to a change in the ionic composition or pH of the cytoplasm. This is especially pertinent since polymerization and stability of actin may be sensitive to changes in pH (Tilney et al., 1978; Condeelis and Vahey, 1982). Our results, however, show that the organization of F-actin within the cell was not altered by treatment with amiloride alone but did change when db cAMP and amiloride were added together (Allen et al., 1986). Thus direct inhibition of Na+ /H+ exchange did not lead to an alteration in actin and indicates that the db CAMP-induced change in actin is not a secondary effect of a change in ionic content or pH. In rabbit colonic mucosa and dog tracheal epithelium, cAMP stimulates CIsecretion (Frizzell et al., 1976; Smith et al., 1982) and in Necturus and guinea pig gallbladder cAMP induces an increase in apical membrane C1- permeability in addition to inhibiting Na+ /H+ exchange (Petersen and Reuss, 1983; Petersen et al., 1985). In MDCK cells exposure to agents that elevate intracellular levels of cAMP leads to an increase in short-circuit current that appears to be the result of a stimulation of C1- secretion (Brown and Simmons, 1982; Simmons et al., 1984). Our preliminary results show that C1- content drops significantly in MDCK cells after exposure to db cAMP (Allen and Mills, unpublished observations). Thus a combination of inhibition of Na+/H+ exchange and increased C1- permeability of the apical membrane could be the ionic mechanism underlying the volume shifts. One possible explanation for the results described above is that the exposure to db cAMP results in a contraction of the cytoskeleton which leads to a volume change. There are several reasons why this may not be the case. First raising intracellular levels of cAMP leads to relaxation in smooth muscle cells (Adelstein, 1982) via a phosphorylation of myosin light chain kinase. Phosphorylation of the kinase leads to a decrease in activity which then leads to a decrease in phosphorylation of myosin. Phosphorylation of the myosin is necessary to activate actomyosin ATPase activity. We have not analyzed this in MDCK cells but assume that any contractile mechanism that exists would follow similar biochemical pathways as for smooth muscle. Second, treatment with cytochalasins
95
THE CELL CYTOSKELETON
produces a volume and ion shift identical to that obtained with db CAMP. This agent alters the microfilament system in the cell by inhibiting actin monomer addition to the filaments and by disrupting actin network formation, and not via stimulation of contraction. Finally, if the decrease in intracellular ions and water were due to pressure developed during contraction, one would predict that the loss of ions would be in the same ratio as they exist in the intracellular water. This is clearly not the case since Na loss far exceeds K loss. An alternative mechanism that would involve actin in the ion shifts associated with the volume change is via a control of either the activity or number of transporting units in the cell membrane. Since actin has been shown to be linked to the anion transport protein of the red blood cell and may also be linked to a similar transporter in the kidney tubule (Drenckhahn ef al., 1985) it may be that at least some of the actin filaments in the MDCK cell are linked to an ion transporter in the cell membrane. When intracellular levels of cAMP rise, this alters the state of the cytoskeletal proteins (possibly reducing the number of membrane attachments) such that the activity of the transporter is altered. This scheme implies a regulatory role for cytoskeletal proteins as already proposed for control of glutamate receptors and the N unit of the adenyl cyclase. We do not know what type of pathways anions follow in the MDCK cell exposed to db CAMP. However, it is of interest to note that Greger et al. (1985) recently reported that in shark rectal gland exposure to cAMP results in the appearance of a C1- channel in the apical cell membrane. In addition, Kolb et al. (1985), using patch clamp methods, demonstrated that there was an anion-selective channel in the apical membrane of MDCK cells which could be detected more readily in the patch-excised condition than in the cell-attached mode as well as when the cell was exposed to epinephrine. Rather than having a regulatory role in the activity of transporters already present in the cell membrane, the MDCK cell cytoskeleton may be involved in a vesicle insertion mechanism. In this scheme the ion pathways exist in vesicles located in the cytoplasm. Upon alteration of the cytoskeleton, either by db cAMP or cytochalasins, the vesicles come in contact with the cell membrane and fuse with it, thus inserting the transporter. A vesicle insertion mechanism has been proposed for insertion of water channels into the apical membrane of toad bladder epithelial cells (see Hays, 1983, for review), sodium channels in rabbit urinary bladder (Loo et al., 1983), proton pumps in turtle urinary bladder (Gluck et al., 1982), and KCI pathways in Necfurus gallbladder during volume regulation (Foskett and Spring, 1985). Whether this is an active process, involving actin in a contractile or cross-linking mechanism, which would move the vesicles to the membrane, or a process involving a “relaxation” of the subplasmalemmal cytoplasmic gel (DiBona, 1983), which would facilitate contact between the vesicle membrane and cell membrane (Hirokawa et al., 1983), is not known. Our +
+
96
JOHN W. MILLS
results with cytochalasin and the known effects of CAMPon actin and actomyosin force generation would indicate that, if vesicle insertion is part of the mechanism in MJXK cells, then a relaxation or gel to sol transformation may be involved.
VII. SUMMARY There is a significant body of indirect evidence that indicates that the cytoskeleton may play a role in regulating the number and/or activity of membrane transport pathways. Thus, modifications of ion transport processes which lead to alterations in cell volume may be mediated by changes in the cytoskeleton. One possible mechanism could involve the delivery to or retrieval from the cell membrane of specific ion transporters. Another could be that a change in the number or position of membrane-cytoskeletal attachment sites results in the stimulation or inhibition of transport proteins already present in the cell membrane. There is also evidence to indicate that certain components of the cytoskeleton can develop tension and contract. This action might be able to provide a stabilizing or resistive force to counteract hydrostatic pressure changes. Resolution of the actual role of the cytoskeleton in regulating ion transport and cell volume will require further detailed studies on the types of cytoskeletal proteins. their distribution in cells, and whether or not these proteins interact with membrane transport proteins. ACKNOWLEDGMENTS The author wishes to thank Drs. Peter Friedman and Roger Sloboda for critical review of earlier drafts of the manuscript. The research reported here was supported by the American Heart Association.
REFERENCES Adelstein, R. S. (1982). Calmodulin and the regulation of the actin-myosin interaction in smooth muscle and nonmuscle cells. Cell 30,349-350. Albertini, D. F., and Andersen. E. (1977). Microtubule and microfilament rearrangements during capping of concanavalin A receptors on cultured ovarian granulosa cells. J . Cell Biol. 73, 11 1127.
Albertini, D. F., and Herman B. (1984). Cell shape and membrane receptor dynamics: Modulation by the cytoskeleton. I n “Cell and Muscle Motility” (J. W. Shay, ed), Vol. 5, Ch. 6. Plenum, New York. Allen, J. C., Coffey, A. K., and Mills, J. W.(1986). Involvement of cytoskeletal organization and NalH exchange in dibutyryl cyclic AMP induced alterations of MDCK cell volume. I n “Ion Gradient-Coupled Transport, INSERM Symp. No. 26” (F. Alvarado and C. H. van Os, eds.), pp. 415-419. Elsevier, Amsterdam. Ash, J. F.. Louvard, D., and Singer, S. J. (1977). Antibody-induced linkages of plasma membrane
THE CELL CYTOSKELETON
97
proteins to intracellular actomyosin-containing filaments in cultured fibroblasts. P roc. Nail. Acad. Sci. V . S . A . 74, 5584-5588. Barak, L. S., Yocum. R. R., Nothnagel, E. A., and Webb, W. W. (1980). Fluorescence staining of
the actin cytoskeleton in living cells with 7-nitrobenz-2-oxa-l,3-diazole-phallacidin. Proc. Nail. Acad. Sci. V.S.A. 77, 980-984.
Baudry, M..Bundman, M. C., Smith, E. K.. and Lynch, G. S. (1981). Micromolar calcium stimulates proteolysis and glutamate binding in rat brain synaptic membranes. Science 212, 937-938. Baudry, M.,Kramer, K., and Lynch, G. (1983). Irreversibility and time course of calcium stimulated [3H]glutamatebinding to rat hippocampal membranes. Brain Res. 270, 142-145. Bennett, V. (1985). The membrane skeleton of human erythrocytes and its implications for more complex cells. Annu. Rev. Biochem. 54, 273-304. Bennett, V.,Davis, J., and Fowler, W. E. (1982). lmmunoreactive forms of erythrocyte spectrin and ankyrin in brain. Philos. Trans. R. SOC. London Ser. B 299, 301-312. Benos, D. J. (1982). Amiloride: A molecular probe of sodium transport in tissues and cells. Am. J. Physiol. 242, C131-C145. Bernier-Valentin, F., Aunis, D., and Rousset, B. (1983). Evidence for tubulin binding sites on cellular membranes: Plasma membranes, mitochondria1 membranes, and secretory granule membranes. J. Cell Biol. 97, 209-216. Benidge, K . (1981). Are stress fibers contractile‘?Nature (London) 294, 691-692. Bhattacharyya, B., and Wolff, J. (1975). Membrane-bound tubulin in brain and thyroid tissue. J. Biol. Chern. 250, 7639-7646. Bourguignon, L. Y.W., and Singer, S. J. (1977). Transmembrane interactions and the mechanism of capping of surface receptors by their specific ligands. Proc. Nafl. Acud. Sci. U.S.A. 74, 5031-5035. Bourguignon, L. Y. W., Suchard, S. J., Nagpal, M. L., and Glenney, J. R., Jr. (1985). A Tlymphoma transmembrane glycoprotein (GP 180) is linked to the cytoskeletal protein, fodrin. J . Cell Biol. 101, 477-487. Boyles, J., Fox, J. E. B., Phillips, D. R., and Stenberg. P. E. (1985). Organization of the cytoskeleton in resting, discoid platelets: Preservation of actin filaments by a modified fixation that prevents osmium damage. J. Cell Biol. 101, 1463-1472. Branton, D., Cohen, C. M.,and Tyler, J. (1981). Interaction of cytoskeletal proteins on the human erythrocyte membrane. Cell 24, 24-32. Brown, C. D. A., and Simmons, N. L. (1982). Catecholamine stimulation of C1- secretion in MDCK cell epithelium. Biochim. Biophys. Acfa 649, 427-435. Burgess, D. R., and Rum, B. E. (1982). A reevaluation of brush border motility: Calcium induces core filament solation and microvillar vesiculation. J . Cell Biol. 94, 97-107. Byes, H. R.. White, G. E., and Fujiwara, K. (1984). Organization and function of stress fiber in cells in viiro and in situ: A review. I n “Cell and Muscle Motility” (J. W. Shay, ed.), Vol. 5 , Chap. 2, pp, 83-137. Plenum, New York. Carley, W. W., Lipsky, M.G., and Webb, W. W. (1983). Regulation and drug insensitivity of Factin association with adhesion areas of transformed cells. J. Cell Physiol. 117, 257-265. Cereijido, M., Robbins, E. S., Dolan, W. I.,Rotunno, C. A., and Sabatini, D. D. (1978). Polarized monolayers formed by epithelial cells on a permeable and translucent support. J . Cell Biol. 77, 853-880. Condeelis, J., and Vahey, M. (1982). A calcium- and pH-regulated protein from Dictyostelium discoideum that cross-links actin filaments. J. Cell B i d . 94, 466-471. Cox, J. V,, Moon, R, T., and Lazarides. E. (1985). Anion transporter. Highly cell-type specific expressions of distinct polypeptides and transcripts in erythroid and non-erythroid cells. J . Cell Biol 100, 1546-1557.
9a
JOHN W. MILLS
Dan, J. C., and Hagiwara, Y.(1967).Studies on the acrosome. IX. Course of acrosome reaction in the stafiish. J. Ulrrastrucr. Res. 18, 562-579. Dan, J. C., Ohori, Y.,and Koshida, N. (1964). Studies on the acrosome. VII. Formation of the acrosomal process in sea urchin spermatozoa. J. Ulrrasrrucr. Res. 11, 508-524. Dentler, W.L. (1981).Microtubule-membrane interactions in cilia and flagella. Int. Rev. Cyrol. 72,
1-48. DiBona, D. R. (1983).Cytoplasmic involvement in ADH-mediated osmosis across toad urinary bladder. Am. J. Physiol. 245, C297-C307. Drenckhahn, D.,and Zinke, K. (1984).Identification of immunoreactive forms of human erythrocyte band 3 in nonerythroid cells. Eur. J. Cell Biol. 34, 144-150. Drenckhahn. D.,Schluter, K.. Allen, D. P., and Bennett, V. (1985). Colocalization of band 3 with ankyrin and spectrin at the basal membrane of intercalated cells in the rat kidney. Science 230,
1287- 1289. Dustin, P. (1984). “Microtubules,” 2nd Ed. Springer-Verlag, New Yo&. Fey, E. G . , Wan, K. M.,and Penman, S. (1984). Epithelial cytoskeletal framework and nuclear matrix-intermediate filament scaffold: Three-dimensional organization and protein composition. J. Cell Biol. 98, 1973-1984. Foskett. J. K.,and Spring, K. R. (1985). Involvement of calcium and cytoskeleton in gallbladder epithelial cell volume regulation. Am. J. Physiol. 17, C27-C36. Fox, J. E. B., Boyles. J. K., Reynolds, C. C., and Phillips, D. R. (1984).Actin filament content and organization in unstimulated platelets. J. Cell Biol. 98, 1985- 1991. Frizzell, R. A,, Koch, M. J., and Schultz, S. G.(1976).Ion transport by rabbit colon. I. Active and passive components. J. Membr. Biol. 27, 297-316. Fujiwara, K.,and Pollard, T. D. (1976). Fluorescent antibody localization of myosin in the cytoplasm, cleavage furrow, and mitotic spindle of human cells. J. Cell Biol. 71, 848-875. Fujiwara, K., Porter, M. E., and Pollard, T. D. (1978).Alpha-actinin localization in the cleavage furrowduring cytokinesis. J. Cell Biol. 79, 268-275. Geiger, B. (1983).Membrane-cytoskeleton interaction. Biochim. Biophys. Acra 737, 305-341. Cluck, S., Cannon, C., and Al-Awqati, Q. (1982).Exocytosis regulates urinary acidification in turtle bladder by rapid insertion of H + pumps into the luminal membrane. Proc. Narl. Acad. Sci. U.S.A. 79, 4327-4331. Godman, G.C., and Miranda, A. F. (1978).Cellular contractility and the visible effects of cytochalasin. In “Cytochalasins: Biochemical and Cell Biological Aspects” (S. W.Tanenbaum, ed.), pp. 277-429. Elsevier, Amsterdam. Goldman, R. D., Berg, G . , Bushnell, A,, Chang, C.-M.. Dickerman, L., Hopkins, N., Miller, M.L., Pollack, R., and Wang. E. (1973).Fibrillarsystems in cell motility. CibOFoumf. Symp. 14, 83-102. Green, K. J., and Goldman, R. D. (1983).The effects of taxol on cytoskeletal components in cultured fibroblasts and epithelial cells. Cell Moril. 3, 283-305. Greger, R.,Schtatter, E., and Gogelein, H. (1985).C1--channels in the apical cell membrane of the rectal gland “induced” by CAMP. Pflugers Arch. 403, 446-448. Harris,A. K., Wild, P., and Stopak, D. (1980). Silicone rubber substrata: A new wrinkle in the study of cell locomotion. Science 208, 177-179. Hays, R. M. (1983).Alteration of luminal membrane structure by antidiuretic hormone. Am. J. Physiol. 245, C289-C296. Heubusch, P.,lung, C. Y.,and Green, F. A. (1985).The osmotic response of human erythrocytes and the membrane cytoskeleton. J. Cell. Physiol. 122, 266-272. Hirokawa, N., Cheney, R. E., and Willard, M. (1983).Location of a protein of the fodrin spectrinTW260/240family in the mouse intestinal brush border. Cell 32, 953-965. Hsie, A. W.,and Puck, T. T. (1971).Morphological transformation of Chinese hamster cells by
THE CELL CMOSKELETON
99
dibutyryl adenosine cyclic 3‘,5’-monophosphate and testosterone. Proc. Nutl. Acad. Sci. U.S.A. 68, 358-361. Inoue, S., and Tilney, L. G. (1982). The acrosomal reaction of Thyone sperm. 1. Changes in the sperm head visualized by high resolution video microscopy. J. Cell Biol. 93, 812-820. Insel, P. A , , and Koachman, A. M. (1982). Cytochalasin B enhances hormone and cholera toxinstimulated cyclic AMP accumulation in S49 lymphoma cells. J . Biol. Chem. 257, 9717-9723. Isenberg, G., Rathke, P. C., Hulsmann, N., Franke, W. W., and Wohlfarth-Bottermann, K. E. (1976). Cytoplasmic actomysin fibrils in tissue culture cells. Cell Tissue Res. 166, 427-443. Jacobson, B. S. (1983). interaction of the plasma membrane with the cytoskeleton: An overview. Tissue Cell 15, 829-852. Johnson, R. M., Taylor, G., and Meyer, D. B. (1980). Shape and volume changes in erythrocyte ghosts and spectrin-actin networks. J. Cell Biol. 86, 371-376. Karfunkel, P. (1974). The mechanisms of neural tube formation. fnr. Rev. Cyrol. 38, 245-271. Kay, M. M. B., Tracey, C. M., Goodman, J. R.. Cone, 1. C.. and Bassel, P.S. (1983). Polypeptides immunologicallyrelated to band 3 are present in nucleated somatic cells. Proc. Narl. Acad. Sci. U.S.A. 80, 6882-6886. Kleinzeller, A. (1972). Cellular transport of water. I n “Metabolic Pathways: Metabolic Transport” (L. E. Hokin, ed.), Vol. 6, pp. 91-131. Academic Press, New York. Kletzien. R. F., and Perdue, J. F. (1973). The inhibition of sugar transport in chick embryo fibroblasts by cytochalasin 8. Evidence for a membrane specific effect. J . &of. Chem. 243, 711-719. Kolb, H. A., Brown, C. D. A., and Murer, H.(1985). Identification of a voltage-dependent anion channel in the apical membrane of a CI--secretory epithelium (MDCK). Pflugers Arch. 403, 262-265. Kreis, T. E., and Birchmeier, W . (1980). Stress fiber sarcomeres of fibroblasts are contractile. Cell 22, 555-56 I . Lazarides, E. (1976). Actin, a-actinin and tmpomyosin interaction in the structural organization of actin filaments in nonmuscle cells. J . Cell Biol. 68, 202-219. Lazarides, E., and Bunidge, K. (1975). a-actinin: Immunofluorescent localization of a muscle structural protein in nonmuscle cells. Cell 6 , 289-298. Loo. D. D. F., Lewis. S. A., Ifshin, M. S., and Diamond, J. M. (1983). Turnover, membrane insertion, and degradation of sodium channels in rabbit urinary bladder. Science 221, 12881290. MacKnight, A. D. C., and Leaf, A. (1977). Regulation of cellular volume. Physiol. Rev. 57, 510573. MacLean-Fletcher. S., and Pollard, T. D. (1980). Mechanism of action of cytochalasin B on actin. Cell 20, 329-341. McRoberts, J. A., Erlinger, S . , Rindler, M. J., and Saier, M. H., Jr. (1982). Furosemide sensitive salt transport in the Madin-Darby canine kidney cell line. J . Biol. Chem. 257, 2260-2266. Matsumoto, G., Ichikawa, M., Tasaki, A., Murofushi, H., and Sakai, H. (1984). Axonal microtubules necessary for generation of sodium current in squid giant axons: I. Pharmacological study on sodium current and restoration of sodium current by microtubule proteins and 260K protein. J. Membr. Biol. 77, 77-91. Melmed, R. N., Karanian, P. I., and Berlin, R. D. (1981). Control of cell volume in the 5774 macrophage by microtubule disassembly and cyclic AMP. J . Cell Biol. 90, 761-768. Mesa, I., Ibarra, G., Sabanero, M., Martinez-Palomo, A.. and Cereijido, M. (1980). Occluding junctions and cytoskeletal components in a cultured transporting epithelium. J . Cell Biol. 87, 746-754. Mills, J. W . , and Lubin. M. (1986). Effect of adenosine 3’.5’-cyclic monophosphate on volume and cytoskeleton of MDCK cells. Am. J . Physiol. (Cell Phvsiol.) 250(19), C319-C324.
100
JOHN W. MILLS
Mills, J . W., and Skiest, D. J. (1985). Role of cyclic AMP and the cytoskeleton in volume control in MDCK cells. Mol. Physiol. 8, 247-262. Mueller, J . , Kachadorian, W. A., and DiScala, V. A. (1980). Evidence that ADH-stimulated intramembrane particle aggregates are transferred from cytoplasmic to luminal membranes in toad bladder epithelial cells. J. Cell Eiol. 85, 83-95. Murofushi, H., Minami, Y., Matsumoto, G., and Sakai, H. (1983). Bundling of microtubules in virro by a high molecular weight protein prepared from the squid axon. J. Biochem. 93, 639650. Nikinmaa, M., and Huestis, W. H. (1984). Shape changes in goose erythrocytes. Eiochirn. Eiophys. Acra 773, 317-320. Osborn, M., and Weber, K. (1984). Actin paracrystal induction by forskolin and by db cAMP in CHO cells. Exp. Cell Res. 150, 408-418. Petersen, K.-U., and Reuss, L. (1983). Cyclic AMP-induced chloride permeability in the apical membrane of Necfurus gallbladder epithelium. J . Gen. Physiol. 81, 705-729. Petersen, K.-U., Wehner, F., and Winterhager, J. M. (1985). NalH exchange at the apical membrane of guinea-pig gallbladder epithelium: Properties and inhibition by cyclic AMP. Pflugers Arch. 405, S115-SI20. Pollard, T. D. (1976). The role of actin in the temperature-dependent gelation and contraction of extracts of Acanrhomoeba. J. Cell Biol. 68, 579-601. Reuss, L., and Petersen, K.-U. (1985). Cyclic AMP inhibits N a + / H + exchange at the apical membrane of Necfurus gallbladder epithelium. J . Gen. Physiol. 85, 409-430. Rindler, M. J . , and Saier, M. H . , Jr. (1981). Evidence for N a + / H + antiport in cultured dog kidney cells (MDCK). J. Eiol. Chem. 256, 10820-10825. Rindler. M. J., Chuman, L. M., Shaffer, L., and Saier, M. H., Jr. (1979). Retention of differentiated properties in an established dog kidney epithelial cell line (MDCK). J . Cell Biol. 81,635648. Sanger, J. W., Sanger, J. M., and Jockusch, B. M. (1983). Differential response of three types of actin filament bundles to depletion of cellular ATP levels. Eur. J . Cell Eiol. 31, 197-204. Schackmann, R. W., Eddy, E. M., and Shapiro, B. M. (1978). The acrosome reaction of Srrongylocentrotus purpurarus sperm: Ion movements and requirements. Dev. Eiol. 65, 483-485. Schiff, P. B., Fant, J., and Horwitz, S. B. (1979). Promotion of microtubule assembly in virro by taxol. Nature (London) 277, 665-667. Siebens, A. W. (1985). Cellular Volume Control. In “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), Ch. 5, pp. 91-1 15. Raven, New York. Siman, R., Baudry, M., and Lynch, G. (1985). Regulation of glutamate receptor binding by the cytoskeletal protein fodrin. Nature (London) 313, 225-228. Simmons, N. L. (1982). Cultured monolayers of MDCK cells: A novel model system for the study of epithelial development and function. Gen. Pharmacol. 13, 287-291. Simmons, N. L . , Brown, C. D. A,, and Rugg, E. L. (1984). The action of epinephrine on MadinDarby canine kidney cells. Fed. Proc., Fed. Am. SOC. Exp. Biol. 43, 2225-2229. Smith, P. L., Welsh, M. J., Stoff, J. S . , and Frizzell, R. A. (1982). Chloride secretion by canine tracheal epithelium: I. Role of intracellular cAMP levels. J . Membr. Biol. 70, 217-226. Stefani, E., and Cereijido, M. (1983). Electrical properties of cultured epitheloid cells (MDCK). J . Membr. Biol. 73, 177-184. Stossel, T. P., Hartwig. J. H., and Yin, H. L. (1981). Actin gelation and the structure and movement of cortical cytoplasm. Cell Surf. Rev. 7, 140-169. Tannenbaum, I., Tannenbaum, S. W., and Godman, G. C. (1977). The binding sites of cytochalasin D, 11. Their relationship to hexose transport and to cytochalasin B. J . Cell. Physiol. 91, 225238.
THE CELL CYTOSKELETON
101
Tilney, L.G . , and Inoue, S . (1982). The acrosomal reaction of Thyone sperm. 11. The kinetics and possible mechanism of acrosomal process elongation. J . Cell B i d . 93, 820-827. Tilney, L. G.. and Inoue, S. (1985). Acrosomal reaction of Thyone sperm. 111. The relationship between actin assembly and water influx during the extension of the acrosomal process. J . Cell Biol. 100, 1273-1283. Tilney, L. G.,Kiehart, D., Sardet, C., and Tilney, M. (1978). The polymerization of actin IV. The role of Ca+ and H + in the assembly of actin and in membrane fusion in the acrosomal reaction of echinoderm sperm. J. Cell Biol. 77, 536-550. Valentich, J. D. (1981). Morphological similarities between the dog kidney cell line MDCK and the Mammalian cortical collecting tubule. Ann. N.Y . Acud. Sci. 372, 384-405. van Rossum, G. D. V., and Russo, M.A. (1981). Ouabain-resistant mechanism of volume control and the ultrastructural organization of liver slices recovering from swelling in virro. J . Mernbr. Biol. 59, 191-209. Wade, J. B. (1980). Hormonal modulation of epithelial structure. Curr. Top. Membr. Transp. 13, 123- 147. Westetman, M. P., Pierce, L. E., and Jensen, W. N. (]%I). A direct method for quantitative measurement of red cell dimensions. J . Lab. Clin. Med. 57, 819-825. Westermark, B., and Porter, K. R. (1982). Hormonally induced changes in the cytoskeleton of human thyroid cells in culture. J. Cell Biol. 94, 42-50. Willingham, M.C., and Pastan, I. (1975). Cyclic AMP and cell morphology in cultured fibroblasts: Effects on cell shape, microfilament and microtubule distribution, and orientation to substratum. J . Cell Biol67, 146- 159. Wong, A. J., Pollard, T. D., and Herman, I. M . (1983). Actin filament stress fibers in vascular endothelial cells in vivo. Science 219, 867-869. +
This Page Intentionally Left Blank
Part II
Volume Control in Anisosmotic Conditions
This Page Intentionally Left Blank
CURRF;NT
TOPICS IN MEMBRANES A N D TRANSPORT. VOLUME 30
Volume Regulation in Epithelia MIKAEL LARSON' Department of Physiology and Medical Biophysics Biomedical Center Uppsala University S-751 23 Uppsala. Sweden AND
KENNETH R . SPRING Laboratory of Kidney and Electrolyte Metabolism National Heart, Lung, and Blood Institute National Institutes of Health Bethesda, Maryland 20205
1.
INTRODUCTION
Intracellular water constitutes about 40% of the total body weight and regulation of cellular volume is essential for homeostasis. Cellular membranes of leaky epithelia are freely permeable to water and epithelial cell volume would be sensitive to changes in the osmolality of the surrounding extracellular fluid compartments. A reduction in interstitial fluid osmolality would result in water entering the cell, diluting the intracellular solutes, and cell swelling would occur. Conversely, if the osmolality of the surrounding interstitial spaces is suddenly increased, water would be osmotically extracted, resulting in shrinkage of the epithelial cells. Although the regulation of cellular volume has been studied for many years, the mechanisms behind volume regulation are still controversial and under intensive debate. Beause of the high water permeability of the epithelial cell membranes regulation of cellular volume cannot be determined by the water permeability of the membranes. Physiologists have instead confined their studies of the regulatory processes to mechanisms, leading primarily to changes in solute 'Present address: Depanment of Physical Education, University of Orebro, S-701 30 Orebro, Sweden. 105
Copynght 0 1987 by Academic Press. Inc All nghls of nprcduclion In any form reserved
106
MIKAEL LARSON AND KENNETH R. SPRING
content of the cell. Increased fluxes of ions into or out of the cell, as the consequence of osmotically induced shrinkage or swelling, will result in return of cell volume back to the original value, a phenomenon observed in virtually all cell types studied over the years. To assess the specific mechanisms involved in volume regulation, a large number of different tissue preparations have been investigated in detail (i.e., liver and kidney slices, isolated cells such as red blood cells, lymphocytes, and tumor cells, isolated renal tubules, isolated gallbladder epithelium, brain cells, skeletal muscle). Many different techniques have also been developed over the years and today we have access to methods that allow the study of regulatory processes in situations that we can consider to be more or less normal physiological conditions of the living cell. The mechanisms described to be responsible for volume regulation show discrepancies between cell types, but some solute pathways appear to be in common.
II. BASIC PRINCIPLES FOR THE MAINTENANCE OF EPITHELIAL CELL VOLUME DURING STEADY-STATE CONDITIONS Epithelial cells are confronted by a steady influx of solutes across one cell membrane and an efflux across the opposite membrane. The work done by these cells in the process of transepithelial transport can be substantial. A rat proximal tubule cell is continuously subjected to a transcellular flow of fluid which is of such a magnitude the cell volume is replaced every 20 sec. Proximal tubule cells and those of other fluid-transporting epithelia must possess regulatory systems capable of balancing the influx and efflux of fluid. The consequences of failing to maintain this balance are obvious: profound cell swelling or shrinkage could occur within seconds of derangement of the regulatory process. The balance between fluid entry and exit from epithelial cells is a reflection of the balance of solute entry and exit mechanisms. As will be described below, the rate of transcellular fluid movement is, in the absence of an osmotic gradient, solely determined by the rate of transcellular solute transport.
A. Epithelial Cell Water Permeability The water permeability of the cell membranes of only a few epithelia has been directly determined. The subject was recently reviewed for renal tubules by Berry (1983). Direct measurements of water permeability of rabbit proximal tubule cell membranes have been reported (Welling et al., 1983; Gonzales et al.,
VOLUME REGULATION IN EPITHELIA
107
1984). The water permeability of both principal and intercalated cells of rabbit cortical collecting duct has also been measured by quantitative light microscopic techniques (Strange and Spring, 1985). These studies, and those on Necturus gallbladder (Persson and Spring, 1982), show that the water permeability of the basolateral membrane is approximately the same per unit area in all epithelial cells studied (Berry, 1983). The basolateral membrane area is sufficiently high in epithelial cells that the osmolality of the cell is predominantly determined by the osmolality of the solution bathing the serosal (basolateral) surface (Strange and Spring, 1985). This means that, regardless of the magnitude of the water permeability of the apical (luminal) membrane of the cells, water will enter or leave the cell in response to alterations in solute content and therefore osmolality of the cytoplasm. Even in cells subjected to large osmotic gradients, the intracellular osmolality will be very close to that of the serosal bath unless significant barriers to diffusion exist beneath the basolateral membrane (Spring, 1983). Diffusion barriers, introduced because of experimental design constraints, can alter the access of the basolateral membrane to the serosal bathing solution as seen in studies OfNecturus gallbladder (Spring, 1983). It must be pointed out, however, that in vivu, the serosal capillaries probably prevent significant unstirred layerrelated osmolality differences from developing. The high water permeability of one or both cell membranes effectively means that the rate of water flow across the cell membranes is not a limiting factor which determines cell volume.
6. Solute Flow and Steady-State Cell Volume Processes which result in an increase in cell solute content will cause cell swelling. An increase in cell solute content causes a rise in intracellular osmolality and water influx. Once the osmotic pressure difference across the cell membrane is dissipated, cell volume ceases to change. Blocking the exit of solute from a cell or increasing the entrance rate of solute both result in cell swelling. Blocking solute exit causes swelling because the normal entry process continues for some time (Ericson and Spring, 1982a). Increasing solute entry also may lead to swelling; a good example is given by the recent reports on the effects of glucose addition on the short-circuit current (a measure of the active transport rate) and cell volume of intestinal epithelia (Lau et al., 1984). Adding glucose to the mucosal perfusate stimulates Na+ entry via Na+-glucose cotransport, cell solute content increases, and swelling results. Subsequently the rate of Na+ exit increases, as reflected by the short-circuit current, cell Na+ returns to normal values, and presumably cell volume also returns to control values. Other techniques for increasing the rate of solute entry include the use of channel formers (e.g., nystatin), alteration of the gradients for ions across the cell membranes, or stimulation of solute entry by hormone addition (e.g., al-
108
MIKAEL LARSON AND KENNETH R. SPRING
dosterone). Unless solute exit increases in parallel with entry during such manipulations, cell volume will increase. This volume increase may be transient or sustained depending on the compensatory capabilities of the epithelium. Thus there is the expectation that any impediment to solute exit or enhancement of solute entry should lead to cell volume increase. In a similar way, any procedures which lead to decreased solute content will also lead to a diminution in cell volume. Solute depletion lowers intracellular osmolality and leads to water efflux. Inhibition of solute entry by ion replacement or drug addition should cause cell shrinkage as the normal solute exit processes continue. An example of this sequence of events is given by the effects of the inhibitors bumetanide (Larson and Spring, 1984) or hydrochlorthiazide (Larson and Persson, 1985) on steady state cell volume of Necfurus gallbladder epithelial cells. Similar observations have been made in Amphiuma diluting segment (Guggino et af., 1985), where inhibition of Na+ ,K+ ,C1- cotransport into the cell across the apical membrane leads to cell shrinkage and hyperpolarization of the membrane potential as the cell NaCl content falls. Another way to reduce cell volume is to stimulate solute exit by acceleration of normal exit processes, by activation of quiescent exit steps, or by introduction of ionophores or channel formers which cause the cell to leak solute. For example, cell swelling appears to activate solute in many systems by the activation of quiescent channels which stimulate solute exit. C. Polarlty of Solute Movements and Cell Volume Changes Solute movements across all epithelial cells are polarized. In general, absorptive epithelia take up solutes across the apical cell membrane and transport them out across the basolateral membrane. In the steady state these two processes must be in balance and the transepithelial flux of the solute is a direct measure of the net flux across the individual cell membranes. Volume regulatory flows are similarly polarized in the few epithelial preparations which have been investigated. Volume regulatory increase in Necturus gallbladder is accomplished by transport events which occur only at the apical membrane of the cells (Marsh and Spring, 1985). Volume regulatory decrease occurs because of transport across the basolateral membrane of the cells (Larson and Spring, 1983). This polarity of transport is a major feature of epithelia and constitutes an essential characteristic of the directed work done by such tissues.
111. TRANSEPITHELIAL NaCl TRANSPORT The mechanism of NaCl entry into the Necfurus gallbladder cells is a question in dispute. It has been demonstrated that transepithelial transport of salt and
109
VOLUME REGULATION IN EPITHELIA
water is associated with carrier-mediated NaCl cotransport across the apical membrane. Other investigators disagree with this conclusion as they have evidence that the mechanism for the NaCl uptake during salt transport is the action of the parallel Na /H and CI- IHCO, - exchangers operating across the apical gallbladder membrane. Ericson and Spring (1982b) suggested that these parallel exchangers are normally inactive and therefore not participating in transepithelial transport of salt. They proposed that the exchangers were activated by cell shrinkage and were associated with the volume regulatory increase. The diuretic bumetanide, a very potent inhibitor of coupled NaCl transport in the Necturus gallbladder epithelium, has been shown to significantly affect steady state cell volume (Fig. 1). Bumetanide, which will not interfere with the parallel ion exchangers, reduces cell volume rapidly by about 17%, due to blockage of coupled NaCl entry at the apical membrane, while the Na+/K+ pump, at the basolateral membrane, continues to extrude Na from the intracellular space at a normal rate (Larson and Spring, 1983). The intracellular Na+ pool is virtually emptied by the pump and the efflux of C1- parallels that of Na . The quantity of C1- in the cell is, however, greater than the Na+ quantity and the calculated loss of C1- is in excess of the Na loss during shrinkage. Some of the C1- is probably lost together with K + , possibly by a cotransport process at the basolateral membrane. This conclusion is supported by measurements of the intracellular K activity which together with the volume measurements enabled the calculation of the quantity of intracellular K . The results were interpreted to show some loss of K + during the bumetanide-induced shrinkage (Larson and Spring, 1983). It is not clear, however, if the parallel exchangers are normally totally inactive and just transiently activated, or if their activity increases above normal basal activity by extracellular hypertonicity. If the parallel exchangers are in part +
+
+
+
+
+
+
-e c
8
0
1
2
3
4
5
Time (minutes)
FIG. I . The effect of the diuretic bumetanide on steady state cell volume when applied to the mucosal side of Necrurus gallbladder epithelium. Note the rapid shrinkage as an immediate effect of the drug.
110
MIKAEL LARSON AND KENNETH R. SPRING
responsible for the transepithelial fluid transfer, inhibition of the C1- /HCO, antiport should reduce steady state cell volume similarly to the effect of bumetanide. A recent investigation (Larson and Persson, 1985) demonstrated an effect of the carbonic anhydrase inhibitor, acetazolamide, on steady state cell volume. A decrease by 7% was reached after bilateral perfusion of the blocking agent. These authors suggested that the reduction in volume was due to reduced CO, turnover with subsequent decrease in substrates (H+ and HC0,-) available for the two ion exchangers. In addition, it was shown that transepithelial fluid transport, measured across gallbladder sacs, was reduced by 69% in the presence of acetazolamide. It was not possible, however, to totally block the fluid transport with the drug, suggesting that cellular uptake of Na+ and C1- across the apical membrane is accomplished, to about the same extent, by both neutrally coupled NaCl entry and Na and C1- entry via the two parallel ion exchangers. A recent report by Davis and Finn (1985) gave evidence for the activity of both cotransport and parallel exchangers in the apical membrane of Necturus gallbladder. NaCl uptake has also been shown by Jensen and co-workers (1984) to be regulated by a feedback mechanism. When Necturus gallbladder cells are exposed to the Na+ ,K+-ATPase inhibitor ouabain, they start to swell because of apical NaCl entry. The entry is automatically shut off when the cells reach peak volume and instead the cells start to shrink, probably as a result of basolateral KCl exit. Bumetanide in the apical bath will block the swelling phase, but does not further increase the rate of shrinkage during the normally activated shrinking phase, suggesting that the uptake of Na+ and C1- is regulated by a feedback mechanism. This type of feedback regulation was earlier observed in tight epithelia by Windhager and Taylor (1983) and also in renal proximal tubules (Friedman et al., 1981), and the proposed mechanism for regulation of Na+ entry in epithelial cells involves alterations in intracellular Ca2 level. +
+
IV. RESPONSE OF EPITHELIAL CELL VOLUME TO OSMOTIC PERTURBATIONS
When epithelial cells were challenged by anisotonic media, many showed an osmometric response during the initial swelling or shrinking phase. In most leaky epithelia this change in volume is transient, immediately followed by a very rapid regulatory readjustment accomplished by solute fluxes across either the epithelial apical (mucosal) or basolateral (serosal) cell membrane. This polarized nature of the regulatory process will be discussed primarily with regard to the accumulated knowledge of the sequence of events in the Necturus gallbladder epithelium. A schematic representation of ion movements proposed to be involved in the volume regulatory responses in Necturus gallbladder after perfusion with anisotonic perfusion solutions is shown in Fig. 2.
111
VOLUME REGULATION IN EPITHELIA Mucoso
Serosa
Hypertonic No'
1 n I I
Volume Regulotory Increase
Isotonic
Hypotonic
Fro. 2 . Ion movements involved in the regulatory responses observed in Necrurus gallbladder. In the hypertonic state a parallel exchange mechanism is activated, leading to uptake of Na , CI - , and osmotically obligated water. When perfused with hypotonic solution KCI is exiting the cell across the basolaterdl membrane (potassium is both coupled to CI- and conductively leaving the cell). +
V. METHODS FOR MEASURING CELL VOLUME CHANGES Morphometric methods have been used to quantitatively determine epithelial cell volume. From electron micrographs of serial sections of fixed tissues Blom and Helander (1 977) and Welling and Welling (1979) made estimates of cellular volume for tissues fixed in a particular steady state condition. When using this technique it is not possible, however, to get any information of the dynamic behavior of the epithelium. Optical, light microscopic methods for volume studies on living epithelium were first used by MacRobbie and Ussing (1961), who mounted frog skin in a perfusion chamber. The thickness of the epithelium was repeatedly measured during continuous perfusion with solutions having different osmolarities. When Necturus gallbladder epithelium is mounted in a thin Ussing-type charnber the outlines of each cell can be visualized using the light microscope (Persson and Spring, 1982). The cells are large, regularly shaped and have few lateral infoldings and are thus very suitable for optical measurements. With differential interference contrast microscopy, as developed by Nomarski (1955), an image with high contrast, shallow depth of field, and good resolution is produced. A great advantage with this optical technique is that only a thin cross section of the cell is in focus at a particular time, leaving the remainder of the cell out of focus and thus invisible, which permits optical sectioning of the cell. The differential interference contrast optics, with high magnification ( 1oOX ), reduces the thick-
112
MIKAEL LARSON AND KENNETH R. SPRING
ness of the optical section to 0.5 pm or less. An image at the focal plane is stored on video disk using a low light level television camera. A complete sectioning of a cell, with 5 to 10 recordings from apex to base, is achieved in less than 2 sec and can be repeated every 5 to 6 sec. The achieved time resolution is quite significant and recordings of rapid dynamic events are possible. The stored video images can be replayed later at any time, and with image processing techniques the cell outlines of each cross section can be traced. Cellular volume is determined from the calculated cell cross-sectional areas and known displacements of the focus. This technique provides the possibility of frequent determinations of epithelial cell volume during controlled changes in the composition of the bathing solutions on either side of the epithelium.
VI. VOLUME REGULATORY INCREASE: Necturus GALLBLADDER EPITHELIUM When Necturus gallbladder epithelial cells are exposed to hypertonic solution in either mucosal or serosal bath they undergo an initial osmotically induced shrinkage followed by recovery to their original volume (Persson and Spring, 1982; Ericson and Spring, 1982b). The recovery occurs despite the fact that the bathing solution osmolality remains increased (Fig. 3). The initial volume decrease, measured in Necturus gallbladder, is of the magnitude predicted from the change in the composition of osmotically active solutes, if the cell behaves as an osmometer. Persson and Spring (1982) reported that an 18% change in osmolality causes a maximal decrease in cellular volume of about 17% within 30 sec after the increase in mucosal osmolality. They also calculated the rate of
5
Hypertonicity -
u 3 0
1 2 Time (minutes)
FIG.3. Cell volume as a function of time in response to an 18% increase in mucosal (or serosal) bath osmolality.
113
VOLUME REGULATION IN EPITHELIA
volume flow out of the cell during the osmotically induced shrinkage and found that it corresponds to an apical membrane hydraulic conductivity (15,) of 3.5 x l o t 4 cm sec- * Osm- I . Subsequent to the osmotically induced shrinkage a very rapid regulatory increase in cell volume was observed. Cell volume returned to control values in about 60 to 90 sec. The rate of the fluid transfer into the cell during the restoration of cell volume has been demonstrated to be at least five times greater than the normal fluid transport across the epithelium. This regulatory process requires Na+ and CI- as well as HC0,- in the mucosal bathing solution (Fisher et al., 1981). Similar to what has been described in Amphiuma red blood cells (Cala, 1980), volume recovery in Necrurus gallbladder after a hypertonic shock was shown by Ericson and Spring (1982b) to be dependent on the inwardly directed transport of Na and C1- . The route for the Na is via the amilonde-inhibitable Na+ /H+ antiport, located in the apical membrane. By blocking this transport mechanism with amiloride, or by removal of mucosal Na+ , the volume regulatory increase was completely abolished. Regulatory swelling is also dependent of the electroneutral C1- /HCO, - exchanger, running in parallel with the Na+/H+ transporter, a process similar to that described previously by Cala (1980) in the Amphiuma red cell. CI- uptake, and thereby also volume regulation, is inhibited by the stilbenes (SITS, DIDS) indicating that the regulatory process requires the activation of the C1- /HCO, - exchanger. Fisher and Spring (1984) found that the cellular gain of Na+ and C1- accounted for not more than 20% of the increase in intracellular solute content. The major solute increase was K + (70%). The remainder of the solutes are so far unidentified, but may be attributed to changes in intracellular protein concentrations (or possibly to changes in their osmotic coefficients), release of amino acids or organic substances from intracellular stores, or a combination thereof. The question of the sidedness of the transport processes involved in volume regulatory increase has recently been investigated. Marsh and Spring (1985) determined the polarity of the parallel exchangers responsible for volume regulatory increase. They showed that volume regulatory increase in gallbladder cellular volume was the consequence of apical C1- /HCO, - exchange regardless of whether the osmotic change is made in mucosal or serosal bath perfusing the Necturus gallbladder cells. These authors showed that the potent inhibitor of the C1- /HCO, - exchange mechanism, DIDS, is without effect on volume regulatory increase when added to the basolateral side of the epithelium. DIDS was only effective in the apical bath, showing that volume regulatory increase is the result of activation of transporters located only in the apical membrane. Other epithelia, such as frog urinary bladder (Davis and Finn, 1981) and frog skin (Ussing, 1982), do not exhibit volume regulatory increase. If apical C1entry is required for the recovery in cell volume, not only in Necrurus gallbladder, but also in other epithelia, the lack of an apical CI - transport mechanism in these tissues may be the reason for the nonregulatory behavior. A similar lack of +
+
114
MIKAEL LARSON AND KENNETH R. SPRING
response to hypertonic serosal perfusates has been demonstrated in single collapsed renal tubules (Gagnon et al., 1982) and kidney cortical slices (Gyory et al., 1981; Gilles et al., 1983), preparations which do not permit perfusion of the luminal side of the epithelium. If volume recovery is dependent on apically bound membrane transporters, the occluded lumen could prevent the response during these experimental conditions. It is also interesting to mention that when Necturus gallbladder cells are swollen in hypotonic mucosal solution and thereafter returned into control osmolality the cells respond by shrinking and regulating cell volume back to the original control volume, a response identical to that following hypertonic perfusion (Marsh and Spring, 1985). DIDS blocks the volume regulatory increase which occurs, indicating the presence of the same transporters as those activated after perfusion with hypertonic solutions. Other cells show the same kind of behavior. Lymphocytes, Ehrlich cells, and frog skin do not normally regulate their volume after a hypertonic challenge, but when previously swelled in hypotonic solution they activate volume regulatory increase mechanisms. Regulation in Ehrlich cells (Geck et al., 1980; Hoffmann et af., 1983) and frog skin (Ussing, 1982) is dependent on transient activation of Na+, K + , and C1- cotransport. According to Grinstein and co-workers (1983), volume regulation in lymphocytes involves the amiloride-sensitive Na /H transporter and is dependent on C1- and HCO,- in the medium. Volume regulatory increase in Necturus gallbladder has been shown by Foskett and Spring (1985) to be insensitive to agents that interfere with cell Ca2+ or calmodulin-mediated events and is not blocked by substances that cause changes in the cytoskeleton. The recovery increase is therefore not dependent on intact microfilaments or microtubules. A problem that needs further to be investigated is the question of the mechanism that triggers the volume regulatory increase. The reduction in cell volume itself does not seem to be the direct cause of the transient activation of the membrane-bound transporters (Spring and Hope, 1979; Ericson and Spring, 1982b). A possibility, suggested by Cala (1980), was that the trigger for volume regulatory increase in the Amphiuma red cell was changes in intracellular pH. In support of a similar conclusion for the Necturus gallbladder are the studies made by Reuss and Costantin (1984), in which intracellularpH increases in response to C1- removal and that this pH change is sensitive to the CI-/HCO,- blocker SITS. The effect of stilbenes on the CI-/HCO,- exchange will thus result in alkalinization of the cell and, subsequent to this effect, the Na+/H+ exchange may be inactivated in order to restore the pH level of the cell interior, which obviously will result in an abolished volume regulation after a hypertonic exposure. Larson and Persson (1985) suggested that interfering with carbonic anhydrase and the hydration of CO, by acetazolamide should slow down the C1- /HCO, - and Na /H exchangers. Volume regulatory increase is signifi+
+
+
+
115
VOLUME REGULATION IN EPITHELIA
cantly reduced by acetazolamide after a hypertonic challenge, possibly due to a reduced amount of substrates available for the two ion exchangers.
VII. VOLUME REGULATORY DECREASE A. Review of Different Mechanisms Cell volume regulation in most cells occurs readily in response to a reduction in extracellular osmolality. Virtually all cells respond to hypotonicity, first by swelling and then by a regulatory phase during which the cells reduce their volume actively back to, or toward, their original volume. This volume regulatory decrease occurs despite the fact that the cells still remain in a solution with reduced osmolality (Persson and Spring, 1982). The mechanism behind the decrease in cell volume in most cells involves cellular loss of KCI and osmotically obligated water. The frog skin epithelium, a tight epithelium, was shown by Ussing (1982) to respond to a decrease in the extracellular concentration of osmotically active solutes by increasing the basolateral permeability of CI - . This increase in CI permeability possibly occurs in combination with an increase in the permeability to K + of the basolateral membrane. C1- will thus leave the cell together with K+. A similar transport of K + , independent from that of CI- out of the cell, induced by cell swelling in hypotonic medium, has also been demonstrated in lymphocytes, where conductive exit processes of both ions were described by Grinstein and collaborators ( 1982). Several lines of evidence support the view of independent conductive movements of K + and C1- (Grinstein et al., 1982, 1984) and findings are observed suggesting that the C1- conductance is the dominant determinant of the membrane potential during the volume regulatory decrease response. A possible role for Ca2 in lymphocyte regulatory shrinkage has been postulated, because volume-induced increase in K + conductance is Ca2 dependent, possibly the Ca2 -dependent K conductance observed in red cells (Gardos, 1958; Gardos er al., 1976; Schwartz and Passow, 1983). It has also been suggested that the C1- conductance possibly is increased by Ca2 . In Ehrlich ascites tumor cells the volume regulatory decrease response has been shown to have many similarities with the mechanisms described above for lymphocytes. Ehrlich cells lose K + and C1- in combination with amino acids when enlarged in hypotonic media. Taurine constitutes the major part of the amino acids lost (Hendil and Hoffmann, 1974; Hoffmann and Hendil, 1976). According to Kregenow ( 1 97 1 ), the swelling-induced regulatory response observed in the duck red cell also involves net loss of K and C 1 ,~but the loss of +
+
+
+
+
+
116
MIKAEL LARSON AND KENNETH R. SPRING
K is dependent on the C1- , as demonstrated by Kregenow and Caryk (1979) and MacManus and Haas (1981). Replacement of C1- by inorganic anions blocks volume regulatory decrease; the probable mechanism for this exit process is KCl cotransport. Part of the C1- leaving the cell during volume recovery is regained by the cell in exchange for HCO,- (or OH-) via the Cl-/HCO,exchanger. The net result of this parallel transport will be an alkalinization of the medium and cell acidification (Kregenow, 1981). This type of C1--dependent K loss during the regulatory shrinkage has also been detected in other red cells. A furosemide-blockable K and C1- exit response was observed in the toadfish by Lauf (1982), and in low K sheep red cells after osmotic enlargement (Ellory and Dunham, 1980; Dunham and Ellory, 1981). It is worthy to note that a Cl-dependent K+ exit, induced by swelling, could, according to Parker (1983a), be activated in dog red cells after loading them with K . These cells have normally an inverted relationship between intracellular Na+ and K + and lack measurable quantities of Na+ ,K+-ATPase (Parker, 1977). They are thus high Na+ and low K + cells and therefore lose Na+ rather than K+ during the volume recovery. The Na+ transport occurs against its electrochemical gradient and may be accomplished by an Na+ /Ca2+ exchange mechanism (Parker et al., 1975; Parker and Harper, 1980; Parker, 1983b). The mechanism for volume regulatory decrease in the Amphiuma red cell has been shown to have different characteristics from other cell types. The process still involves loss of K + and C1- , but it has been hypothesized by Cala (1983, 1985) that K+ is exchanged for H+ via a K + / H + antiport mechanism. The basis for this conclusion is measurements of the membrane potential during both isotonic and anisotonic conditions, as well as after drug-induced changes of the K + transport. The interdependency of K + and H + is quite evident, but the direct coupling of the two ions is complicated by findings indicating that the fluxes of both ions might be conductive. Although no clear distinction can be made, the proposed mechanism for the regulatory response involves K + / H + exchange. Low intracellular Ca2+ concentrations are observed to block the volume regulatory decrease and Ca2 was concluded to be an important modulator of the electroneutral K + flux accomplished by the K + / H + exchanger. Renal cortical cells have been studied in different preparations. Grantham and co-workers (1977) demonstrated a loss of K + and Na+ in combination with anions from isolated collapsed rabbit proximal straight tubules, subsequent to immersion of the tubules in hypotonic medium. The magnitude of the loss was sufficient to account for the regulation of cell volume. Linshaw and Grantham (1980) further studied the proximal tubule of the rabbit and were able to conclude that volume regulation was dependent on the normal function of the ouabainsensitive Na /K pump at the basolateral membrane. Furthermore, they found evidence for a regulatory function of the tubular basement membrane, generating counteracting hydrostatic forces when these cells were swelled in hypotonic +
+
+
+
+
+
+
+
117
VOLUME REGULATION IN EPITHELIA
medium. Paillard and collaborators (1979) observed in their experiments on renal tubules separated from the rabbit kidney that cell volume regulation after hypotonic incubation was immediate, without initial swelling, was accomplished by the loss of NaCl with no changes in intracellular K + , and that the regulation could be eliminated with ouabain. They suggested that the loss of NaCl was due to blocking of passive NaCl entry into the cells, probably occurring across the luminal membrane. This finding of an NaCl loss, resulting in volume regulation when the tissue was transferred into hypotonic medium, supports the results previously presented by Hughes and Macknight (1976). These authors were not able to find any evidence for loss of K + in renal cortical slices and attributed the volume regulation to the ouabain-sensitive Na /K pump. Gilles and co-workers (1983) questioned the importance of the Na+/K+ pump in the regulatory readjustment of cell volume, based on observations in rabbit renal cortical slices, but supported the previous conclusion of an Na -dependent volume regulatory mechanism. It must be emphasized that in the described preparations on renal tissues the lumen of the tubules is collapsed and consequently not perfused when volume changes occur. Any process acting over the apical membrane of the tubular cells might therefore be totally, or at least partly, diminished because of the experimental conditions. On the basis of this, we cannot exclude a role of apical transport mechanisms, participating in volume regulation of renal tubular cells following extracellular hyposmolality . These are preparations in need of further investigations in order to establish the ionic basis for the regulatory compensation. Luminal perfusion of isolated renal tubular segments might result in different characteristics of the volume decrease process. A very different time course of the regulatory process after hyposmotic swelling is observed in brain and skeletal muscle. In the brain, swelling is slow compared to single cells and volume regulatory decrease involves cellular loss of K , Na , and probably amino acids over several hours or days (Arieff et al., 1976; Pollock and Arieff, 1980). Skeletal muscle shows less compensation for the induced swelling and the loss of K + appears to be insignificant (Dila and Pappius, 1972). +
+
+
+
+
B. Ngcturus Gallbladder Epithelium The response of Necrurus gallbladder epithelial cells to an imposed extracellular osmolality reduction was first described by Persson and Spring (1982). They observed an initial enlargement of the cells of about 22% above the control volume when the cells were exposed to an 18% reduction in osmolality and peak volume was reached within about 30 sec. The swelling phase was directly followed by a reduction in cell volume, despite the fact that mucosal bath os-
118
MIKAEL LARSON AND KENNETH R. SPRING
molality remained hypotonic, resulting in cell volumes close to control values in less than 3 min after the challenge. Similar regulatory decrease was also observed when the cells were returned to control Ringer solution, after having been previously shrunken in hypertonic solution. In most of the cell preparations discussed above the compensatory decrease in cell volume is accomplished by the exit of K , C1- , and osmotically obligated water. Although Necfurus gallbladder epithelial cells were shown to volume regulate after decrease in luminal osmolality (Persson and Spring, 1982) the ionic basis for the change in cell volume was only recently determined (Larson and Spring, 1984). +
1. IONIC DEPENDENCE OF VOLUME REGULATORY DECREASE
To analyze the ionic basis of volume regulation, Larson and Spring (1984) determined the effect of a hypotonic mucosal perfusate on gallbladder cell volume. The composition of the serosal bath remained constant and its osmolality was not changed. The typically observed response to hypotonicity is shown in Fig. 4. Cells swelled due to osmotic influx of water, to a maximum volume about 16% above control within 30 sec and regulation was completed about 60 sec later. Volume regulation appeared not to require HC0,-, but removal of C1from both bathing solutions completely blocked the regulatory response. Increasing the K + concentration 10-fold in the serosal perfusate also resulted in complete inhibition, due to the change of the electrochemical gradient for K + . Larson and Spring (1984) concluded that both K + and C1- are involved in the volume regulatory decrease process in the Necrurus gallbladder.
2. EFFECTSOF INHIBITORS ON VOLUMEREGULATORYDECREASE Larson and Spring (1984) further excluded any possible activation of the apical Na /H and C1- IHCO, - exchange mechanisms during volume reg+
+
-- 120 2 + C
8
c
- 100 U
0
1
2
3
lime (minutes)
FIG. 4. Cell volume as a function in time in response to an 18% reduction in mucosal bath osmolality .
119
VOLUME REGULATION IN EPITHELIA
ulatory decrease because neither amiloride nor SITS affected the rate of volume reduction. A possible role of the basolateral Na ,K -ATPase was also ruled out, as ouabain was without effect in this tissue, but the classical pump may be of importance in renal tubules (Linshaw and Grantham, 1980). The very potent diuretic and blocker of coupled NaCl entry in Necturus gallbladder (Ericson and Spring, 1982a; Larson and Spring, 1984). bumetanide, when added to the serosal perfusion solution, significantly slowed the regulatory process. The bumetanide experiments made it possible to localize the regulatory decrease mechanism to the serosal side of the gallbladder epithelium. Thus the transport processes in this tissue are polarized, the regulatory mechanisms occurring across the apical or basolateral membrane, in response to hyper- or hypotonicity , respectively. +
+
3. MOVEMENT OF KCI ACROSS THE BASOLATERAL MEMBRANE DURING VOLUMEREGULATORY DECREASE The possibility of a directly coupled exit mechanism for K + and C1- under normal conditions was first proposed by Reuss (1979). He later described this cotransport mechanism as Na independent and suggested that the symport was an important pathway for CI- transport out of the cell across the basolateral membrane of Necrurus gallbladder cells, as well as in other NaCl absorbing epithelia (Reuss, 1983). On the basis of their volume measurements, Larson and Spring (1984) suggested that volume regulatory decrease involves K and Cl - , and that these ions are transported across the basolateral membrane by a cotransport mechanism. The volume data also indicated that the rate of volume regulatory decrease was a function of the electrochemical gradient for both of these ions across the serosal membrane. Further studies with ion-sensitive microelectrodes showed that, under control conditions, the electrochemical gradients for both ions favor exit from the cell. A 10-fold increase of the K concentration in the serosal bath was one of the procedures found to block volume regulation and this increase led to an opposing gradient for the K + exit, but did not change the CI- gradient. When the electrochemical Cl- gradient in turn was increased, favoring exit, by a reduction of serosal CI- concentration in combination with high serosal K + , volume regulatory decrease proceeded at rates about 65% of control. Reducing the driving force for C1- across the basolateral membrane by a decrease in the intracellular C1- activity slowed down the rate of volume regulatory decrease, but did not completely block regulation. Larson and Spring ( 1984) concluded from the electrophysiological and volume data, that volume regulatory decrease required movement of KCI out of the cell across the basolateral membrane. An estimation of the stoichiometric relationship between K + and C1- failed to show a fixed ratio, indicating electrogenic coupling of the two ions. Potassium exited the cell in excess of CI - . This question needs to be further investigated, to more precisely evaluate the stoichiometry of the coupling and characterize other organic anions involved in the regulatory response. A +
+
+
120
MIKAEL LARSON AND KENNETH R. SPRING
transient hyperpolarization has previously been observed during the regulatory decrease phase (Fisher and Spring, 1984) consistent with an increase of the K+ permeability and loss of this cation, due to an activated KCl transport. It is clear, from the studies by Larson and Spring (1984),that KCl cotransport plays a significant role in the readjustment of cell volume that follows an osmotic swelling. It is also evident that an increase of the K+ conductance is partly responsible for the regulatory adjustment, but the relative magnitude of these two mechanisms is not known today. Contrary to what was observed during volume regulatory increase, Foskett and Spring (1985) reported that activation of the transport processes involved in volume regulatory decrease is dependent on intact microfilaments and Ca2 . +
VIII.
PHYSIOLOGICAL SIGNIFICANCE OF VOLUME REGULATION
Hyposmolality with hyponatremia is probably the most commonly observed electrolyte disturbance in clinical practice and it is crucial for normal cell function that the cells are capable of regulating their solute and water content (MacKnight and Leaf, 1985). Whether the very powerful transport systems described in this article are responsible for the regulation of cell volume in vivo is, however, not clear today. In red cells, the ion content and volume probably are maintained by the volume regulatory mechanisms, but the final composition of electrolytes is determined by the interdependent action of volume regulatory mechanisms and, for instance, the Na+/K+ pump. The question concerning whether or not the activation of the regulatory processes also involves transepithelial transport of solutes and water and, thus, may be responsible for normal epithelial solute and water composition, has to be further evaluated. Ericson and Spring (1982b) and Jensen and co-workers (1984) found no evidence for any interaction of the regulatory mechanisms and transepithelial transport. Their view is that the regulatory responses are effected by an activation of mechanisms that are normally not operating and transepithelial transport is the result of continuous exchange of solutes and water via different pathways. Whether the very rapid responses observed in the gallbladder epithelium are of significance in vivo is not yet clear. The critical, common factor for NaCl transport mechanisms appears to be pH (Reuss and Costantin, 1984). It still remains to be established whether the volume regulatory mechanisms are present in order to keep the solute and water composition of epithelial cells constant during the transepithelial absorptive process. REFERENCES Arieff, A . I., Llach. F., and Massry, S. G . (1976). Neurological manifestations and morbidity of hyponatremia: Correlation of brain water and electrolytes. Medicine 55, 121- 129.
VOLUME REGULATION IN EPITHELIA
121
Berry, C. A. (1983). Water permeability and pathways in the proximal tubule. Am. J. Physiol. 245, F279-F294. Blom, H., and Helander. H. D. (1977). Quantitative electron microscopical studies on in virro incubated rabbit gallbladder epithelium. J. Membr. Biol. 37, 45-61, Cala. P. M. (1980). Volume regulation by Amphiitmu red blood cells. J . Gen. fhysiol. 76, 683-708. Cala, P. M. (1983). Cell volume regulation by Amphiumu red blood cells. The role of Ca2+ as a modulator of alkali metal/H+ exchange. J. Gen. Physiol. 82, 761-784. Cala, P. M. (1985). Volume regulation by Amphiumu red blood cells: Strategies for identifying alkali metal/H+ transport. Fed. Proc.. Fed. Am. Soc. Exp. Biol. 44,2500-2507. Davis, C. W., and Finn. A. L. (1981). Regulation of cell volume of frog urinary bladder. I n “Membrane Biophysics” (M.A. Dinno, A. B. Callahan, and T. C. Rozzel, eds.), pp. 25-36. Liss, New York. Davis, C. W . , and Finn. A. L. (1985). Effects of mucosal sodium removal on cell volume in Necturirs gallbladder epithelium. Am. J. Physiol. 249, C304-C3 12. Dila, C. J., and Pappius, H. M. (1972). Cerebral water and electrolytes: An experimental model of inappropriate secretion of antidiuretic hormone. Arch. Neurol. 29, 85-90. Dunham, P. B., and Ellory, J. C. (1981). Passive potassium transport in low potassium sheep red cells: Dependence upon cell volume and chloride. 1.Physiol. (London) 318, 51 1-530. Ellory, J . C., and Dunham, P. B. (1980). Volume-dependent passive potassium transport in LK sheep red cells. AIfred Benson Symp. 14, 409-427. Ericson, A.-C., and Spring, K. R. (1982a). Coupled NaCl entry into Necturus gallbladder epithelial cells. Am. J. Physiol. 243, C140-Cl45. Ericson, A.-C., and Spring, K. R. (1982b). Volume regulation by Necturus gallbladder: Apical Na+-H+ and CI--HCO?- exchange. Am. J. Physiol. 243, C146-CI50. Fisher, R. S., and Spring, K. R. (1984). Intracellularactivities during volume regulation by Necrurus gallbladder. J. Membr. Biol. 78, 187-199. Fisher, R. S., Persson, B.-E., and Spring, K. R. (1981). Epithelial cell volume regulation: Bicarbonate dependence. Science 214, 1357- 1359. Foskett, I. K.,and Spring, K. R. (1985). Involvement of calcium and cytoskeleton in gallbladder epithelial cell volume regulation. Am. J. Physiol. 248, C27-C36. Friedman, P. A,, Figueiredo, J. F., Maack. T., and Windhager, E. E. (1981). Sodium-calcium interactions in the renal proximal convoluted tubule of the rabbit. Am. J . fhysiol. 240, F558F.568. Gagnon, J., Ouimet, D., Nguyen, H., Lahdde, R., Le Grimellec, C., Carri&re,S.,and Cardinal, J. (1982). Cell volume regulation in the proximal convoluted tubule. Am. J. Physiol. 243, F408F415. Gardos, G. (1958). The function of calcium in the potassium permeability of human erythrocytes. Biochim. Biophys. Actu 30, 653-654. Gardos, G., Lassen, U. V., and Page, L. (1976). Effect of antihistaminesand chlorpromazineon the calcium-induced hyperpolarization of the Amphiumu red cell membrane. Biochim. Biophys. ACtU 448,599-606. Geck, P., Pietrzyk, C., Burckhardt, B.-C.. Pfeiffer, B., and Heinz. E. (1980). Electrically silent cotransport of Na, K and CI in Ehrlich cells. Biochim. Biophys. Actu 600, 432-447, Gilles, R., Duchene, C., and Lambert, I. (1983). Effect of osmotic shocks in rabbit kidney cortex slices. Am. J. Physiol. 244, F696-F705. Gonzales, E., Carpi-Medina, P., Linares, H., and Whittembury, G . (1984). Osmotic water permeability of the apical membrane of proximal straight tubular (PST) cells. Pfliigers Arch. 402, 337-339. Grantham, J. J., Lowe, C. M.,Dellasega. M.,and Cole. B. (1977). Effect of hypotonic medium on K and Na content of proximal renal tubules. Am. J. Physiol. 232, F42-F49. Grinstein, S., Clarke, C. A., Dupre, A.. and Rothstein, A. (1982). Volume-induced increase of anion permeability in human lymphocytes. J . Gen. Physiol. 80, 801-823.
122
MIKAEL LARSON AND KENNETH R. SPRING
Grinstein, S., Clarke, C. A,, and Rothstein, A. (1983). Activation of N a + / H + exchange in lymphocytes by osmotically induced volume changes and by cytoplasmic acidification. J . Gen. Physiof. 82, 619-638. Grinstein, S . , Rothstein, A,. Sarkadi, B., and Gelfand, E. W. (1984). Responses of lymphocytes to anisotonic media: Volume-regulating behavior. Am. J. Physiol. 246, C204-C215. Guggino, S. E., Guggino, W. B., Suarez-Isla, B. A , , Green, N., and Sacktor, B. (1985). The influence of barium on apical membrane potentials and potassium channel activity in cultured rabbit medullary thick ascending limb cells (MTAL). Fed. Proc.. Fed. Am. SOC. Exp. Biol. 44, 443. Gyory, A. Z., Kweifio-Okai, G., and Ng, J. (1981). Hypo- and hyperosmolal saline and raffinose on kidney cortical cell volume at 37°C. Am. J. Physiol. 240, F180-FI84. Hendil, K. B., and Hoffmann, E. K. (1974). Cell volume regulation in Ehrlich ascites tumor cells. J. Cell. Physiol. 84, 115-126. Hoffmann. E. K., and Hendil, K. B. (1976). The role of amino acids and taurine in isosmotic intracellular regulation in Ehrlich ascites mouse tumor cells. J. Comp. Physiol. 108, 279-286. Hoffmann, E. K., Sjoholm, C., and Simonsen, L. 0. (1983). Na+,CI- cotransport in Ehrlich ascites tumor cells activated during volume regulation (volume regulatory increase). J. Membr. Biol. 76. 269-280. Hughes, P. M., and Macknight, A. D. C. (1976). The regulation of cellular volume in renal cortical slices incubated in hyposmotic medium. J . Physiol. (London) 257, 137-154. Jensen, P. K.,Fisher, R. S., and Spring, K. R. (1984). Feedback inhibition of NaCl entry in Necturus gallbladder epithelial cells. J. Membr. Biol. 82, 95- 104. Kregenow, F. M. (1971). The response of duck erythrocytes to nonhemolytic hypotonic media: Evidence for a volume controlling mechanism. J. Gen. Physiol. 58, 372-395. Kregenow, F. M. (1981). Osmoregulatory salt transporting mechanisms: Control of cell volume in anisotonic media. Annu. Rev. Physiol. 43, 493-505. Kregenow, F. M., and Caryk, T. (1979). Co-transport of cations and CI during the volume regulatory response of duck erythrocytes. Physiologist 22, 73. Larson, M., and Persson, B.-E. (1985). Carbonic anhydrase inhibition slows down volume regulatory increase in Necturus gallbladder epithelial cells. Acra Physiol. Scand. 124, 118. Larson, M., and Spring, K. R. (1983). Bumetanide inhibition of NaCl transport by Necrurus gallbladder. J . Membr. B i d . 74, 123-129. Larson, M., and Spring, K. R. (1984). Volume regulation by Necrurus gallbladder: Basolateral KCI exit. J. Membr. B i d . 81, 219-232. Lau, K. R., Hudson, R. L., and Schultz, S. G. (1984). Cell swelling increases a barium inhibitable potassium conductance in the basolateral membrane of Necrurus small intestine. Proc. Natl. Acad. Sci. U.S.A. 81, 3591-3594. Lauf, P. K. (1982). Evidence for chloride dependent potassium and water transport induced by hyposmotic stress in erythrocytes of the marine teleost, Opsanus tau. J. Comp. Physiol. 146,916. Linshaw, M. A., and Grantham, J. J. (1980). Effect of collagenase and ouabain on renal cell volume in hypotonic media. Am. J. Physiol. 238, F491-F498. Macknight, A. D. C.. and Leaf, A. (1985). Cellular responses to extracellular osmolality. In “ m e kidney: Physiology and Pathophysiology” (D. W. Seldin and G. Giebisch, eds.), pp. 117-132. Raven, New York. MacManus, T. J., and Haas, M. (1981). Catecholamine stimulation of K/K (K/Rb) exchange in duck red cells. Fed. Proc., Fed. Am. Soc. Exp. Biol. 40, 484. MacRobbie, E. A. C., and Ussing, H. H. (1961). Osmotic behaviour of the epithelial cells of frog skin. Acra Physiol. Scand. 53, 348-365. Marsh, D. J., and Spring, K. R. (1985). The polarity of volume regulatory increase by Necrurus gallbladder epithelium. Am. J . Physiol. 249, C471-475.
123
VOLUME REGULATION IN EPITHELIA
Nomarski. G. (1955). Microinterferometri differentielle a ondas polarisees. J. Phys. Radium 16,9S13s. Paillard, M.. Leviel, F., and Gardin, J.-P. (1979). Regulation of cell volume in separated renal tubules incubated in hypotonic medium. Am. J. Physiol. 236, F226-F231. Parker, J . C. (1977). Solute and water transport in dog and cat red blood cells. In “Membrane Transport in Red Cells” (J. Ellory and V. Lew, eds.), pp. 427-466. Academic Press, New York. Parker, J . C. (1983a). Hemolytic action of potassium salts on dog red blood cells. Am. J. Physiol. 244, c313-c317. Parker, J . C . (1983b). Passive calcium movements in dog red blood cells: Anion Effects. Am. J. Physiol. 244, C318-C323. Parker, I . C., and Harper, J. R. (1980). Calcium movements in dog red blood cells. AlfredBenson Symp. 14, 274-282. Parker, 1. C., Gitelman, H. G.. Glosson, P. S . , and Leonard, D. L. (1975). Role of calcium in volume regulation by dog red blood cells. J. Gen. Physiol. 65, X4-96. Persson, B.-E., and Spring, K. R. (1982). Gallbladder epithelial cell hydraulic water permeability and volume regulation. J. Gen. Physiol. 79, 481-505. Pollock, A. S . , and Arieff, A. 1. (1980). Abnormalities of cell volume regulation and their functional consequences. Am. J. Physiol. 239, F195-F205. Reuss, L. (1979). Electrical properties of the cellular transepithelial pathway in Necrurus gallbladder: 111. Ionic permeability of the basolateral cell membrane. J. Membr. Biol. 47, 239-259. Reuss, L. (1983). Basolateral KCI co-transport in a NaCI-absorbing epithelium. Nature (London) 3Q5, 723-726. Reuss, L., and Costantin, 1. L. (1984). CI-/HC03- exchange at the apical membrane of Neciurus gallbladder. J. Gen. Physiol. 83, 801-818. Schwartz. W., and Passow, H. ( I 983). Ca2 + -activated K channels in erythrocytes and excitable cells. Annu. Rev. Physiol. 45, 359-374. Spring, K. R. (1983). Fluid transport by gallbladder epithelium. J. Exp. Biol. 106, 181-194. Spring, K . R., and Hope, A. (1979). Fluid transport and the dimensions of cells and interspaces of living Necturus gallbladder. J. Gem Physiol. 73, 287-305. Strange, K., and Spring, K. R. (1987). Cell membrane water permeability of rabbit cortical collecting duct. J. Membr. Biol. %(1), in press. Ussing, H. H. (1982). Volume regulation of frog skin epithelium. Acta Physiol. Scand. 114, 363369. Welling, D. J . , and Welling, L. W. (1979). Cell shape as an indicator of volume reabsorption in proximal nephron. Fed. P r o c . , Fed. Am. SOC. Exp. Biol. 38, 121-127. Welling, L. W., Welling, D. J., and Ochs, T. J. (1983). Video measurement of basolateral membrane hydraulic conductivity in the proximal tubule. Am. J. Physiol. 245, F123-FI29. Windhager, E. E., and Taylor, A. (1983). Regulatory role of intracellular calcium ions in epithelial Na transport. Annu. Rev. Physiol. 45, 51 1-532. +
This Page Intentionally Left Blank
CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 30
Volume Regulation in Cultured Cells ELSE K . H O F F M A ” Institute of Biological Chemistry A August Krogh Institute University of Copenhagen DK-2100 Copenhagen 0 , Denmark
1.
INTRODUCTION
The ability of animal cells to control their volume is among the most fundamental properties of living cells. Because cells contain charged macromolecules in their cytoplasm, some mechanism must be operating to prevent colloid osmotic swelling and lysis. Thus, their very existence depends on their ability to control their volume. Four factors are important in determining steady-state cell volume in animal cells: (1) Water is in thermodynamic equilibrium across the membrane. Because of this osmotic equilibrium, cell solute content determines cell volume. (2) The permeable cations and anions tend to move until the products of activities on both sides of the membrane are equal, i.e., until Donnan equilibrium is attained. ( 3 ) Electroneutraliry is maintained on both sides of the membrane. (4) Hydrostatic pressure differences must be negligible, because animal cells are easily distensible and cannot maintain hydrostatic pressure gradients. Cells contain impermeant, anionic macromolecules, ‘‘fixed anions,” that affect cell volume. Since cells are permeable to H,O and small ions such as the major extracellular ions (Na , K , and CI -), passive distribution of these ions according to Donnan equilibrium will result in colloid osmotic cell swelling and eventual lysis. A simple example is given in Fig. 1 (for further description see, e.g., Hoffmann, 1977, 1983); a more complete treatment of the Donnan equilibrium is found in Sten-Knudsen (1978). From Fig. 1 it is seen that a positive osmotic pressure difference will develop between the cell and its surroundings. Water enters the cell and tends to level out the difference in osmotic pressure, but at the same time the ions will redistribute according to the Gibbs-Donnan rule. This will go on until all the diffusible ions are inside or until the cells burst +
+
125
Copyrighi 0 1987 by Academic Rear. Inc. All righis of reproduction in any form rrxcrved.
126
ELSE K. HOFFMANN (1)
li)
(0)
lo)
(1)
Ck. x c - = Ck. x c , -
(2)
lo) Ck. =
I31
Ck.
I
11)
I
lo)
c,- = co (il
C1
+
nCpon-
Thus an osmotic pressure difference
II > RT [Cpon-]
develops
FIG. 1. A simple cell model. The cation k+ and the anion j - can penetrate the cell membrane.
Po"- is an impermeable anion. Equation (1) is the Gibbs-Donnan rule; C is the concentration and i and o indicate the inner and outer compartment. Equations (2) and (3) state that electroneutrality is maintained on both sides of the membrane. Equation (4) gives the osmotic pressure difference (n). Equation ( 5 ) follows from Eqs. (1). (2). and (3) since, mathematically, if two positive quantities are related such as to have a constant product [see Eq. (I)], their sum is minimal when they are equal [compare Eq. (2) and Eq. (3)l.
(colloid osmotic lysis). This means that animal cells must, even in isotonic solutions, somehow prevent passive distribution of permeant ions across the cell membrane. Animal cells also continue to regulate their volume when exposed to anisosmotic media, which means that some of the ion transport mechanisms must be volume controlled. There is general agreement that control of cellular volume reflects a balance between passive and active ion movements across the cellular membrane, with the colloid pressure of intracellular macromolecules being offset by the extrusion of sodium ions from the cells. This pump-and-leak concept was developed about 25 years ago by Leaf (1959), Ussing (1960), and Tosteson and Hoffman (1960). In the course of time this concept still holds valid but has proved to be somewhat simplistic. In recent years evidence has accumulated to demonstrate that volume regulation in mammalian cells is achieved also via dynamic and controlled changes of the leak pathways (for references, see Hoffmann, 1978, 1985a-c; Hoffmann et al., 1983, 1984a; Grinstein et al., 1984c, 1985a,b), involving transient stimulation of normally dormant leak pathways (Hoffmann, 1985~;Hoffmann el al., 1986b). Furthermore, the ''leaks'' turn out to be a composite of a number of specific transport pathways involving also cotransport and ion exchange systems, as described below.
VOLUME REGULATION IN CULTURED CELLS
127
Two basic approaches are typically used in studies of cell volume regulation. One approach is to use drugs or alter conditions so that normal pump-leak relationships are perturbed under isotonic conditions (see the first article in this volume). The second approach is to experimentally change the cell volume by varying the medium osmolality , and monitor volume recovery and associated solute movements. This approach addresses the question of how the cells respond when they are not at their normal volume. This is the approach used in the studies to be described in this article. Permeabilities to the predominant ions, K + , Na+, and C1-, are generally much lower than the water permeability. In a typical mammalian cell, the Ehrlich ascites tumor cell, Na+, K + , and CI- permeabilities are all five orders of magnitude lower than the water permeability (Hoffmann et al., 1979; see Table IV). This means that the membrane of animal cells can be regarded as semipermeable and that the cell must behave as a perfect osmometer. In hypotonic media animal cells swell initially as nearly perfect osmometers. Subsequently, however, the cells reduce their volume (regulatory volume decrease, RVD) to a new steady state at a volume only slightly above the original. The volume recovery is accompanied by a parallel net loss of KCI. Restoration of 300 mOsm tonicity after hypotonic pretreatment induces an initial osmotic shrinkage to a cell water content close to that predicted for a perfect osmometer, followed by a net uptake of water, K + , and C1- . This net uptake results in a recovery of cell volume (regulatory volume increase, RVI) up to a level near the original cell volume in the 300 mOsm saline solution. These phenomena have been reported for a variety of cell types, although detailed information has been developed only in relatively few. In this article we shall discuss the basic issues involved in studies of cell volume regulation in cells from vertebrates, focusing mainly on unattached cells such as erythrocytes, lymphocytes, and Ehrlich ascites tumor cells. These cells are especially well suited for studies of cell volume regulation, because the relevant parameters (cell volume, cell water, ion and amino acid contents, fluxes of tracer ions, etc.) can be measured far more simply and accurately than in attached cells. These symmetrical cells are particularly useful as possible model systems for investigating the ionic mechanisms of transporters found in certain epithelia. For epithelial volume regulation see the fourth article in this volume and papers by Spring and Ericson (1982), fistensen and Ussing (1983, and Ussing (1985, 1986). Erythrocytes are only discussed when they are useful as reference background. The interested reader is also referred to several recent reviews (Siebens, 1985; Hoffmann, 1983, 1985a,b; Grinstein et al., 1984c, 1985a,b; Cala, 1983a, 1985; MacKnight, 1983; Ellory er al., 1985a,b). Cell volume regulation in invertebrates has been reviewed by Rorive and Gilles ( 1979) and Gilles (1979); see also the seventh article in this volume).
128
ELSE K. HOFFMANN
II. CLASSIFICATION OF THE ION TRANSPORT MECHANISMS INVOLVED IN VOLUME REGULATION Figure 2 illustrates four general mechanisms by which ions cross membranes. The first of these is electrodifision which must respond to the electrochemical potential of the ion at the two sides of the membrane. Ion diffusion will follow the Goldman-Hodgkin-Katz equation (Hodgkin and Katz, 1949), unidirectional fluxes of ions measured by isotopes should follow the Ussing flux ratio equation (Ussing, 1949), and the permeability as measured at steady state by unidirectional isotope flux should equal that measured under conditions of net ion flow. Such ion diffusion, illustrated by (a), (b), and (c) in Fig. 2, will tend to dissipate the ion gradients created by secondary and primary active transport processes, and hence may be unfavorable for the economy of cell metabolism. In many cells it has been found that the ion leak by electrodiffusion is in fact very small in the normal steady state situation. The second mode of transport involves a one-for-one electroneutral exchange of ions, illustrated by (d), (e), and (0 in Fig. 2. In this case, ion influx and efflux
,Na*
%
?A-
t
FIG.2. Schematic diagram of ion transport pathways through the cell membrane of an isolated vertebrate cell. Electrodiffusion of ions: (a), (b), and (c). Exchange diffusion pathways: (d), (e), and (f). Cotransport pathways: (g) sodium chloride cotransport; (h) sodium, potassium, chloride cotransport; (i) potassium chloride cotransport. The influx of chloride via mechanisms (g) and (h) depends on the extrusion of intracellular sodium by primary active trunsporr mediated by (i) the sodiumpotassium pump. The efflux of chloride via mechanism (i) could also drive chloride out against a gradient, depending on the uptake of potassium by the sodium-potassium pump, (j).
VOLUME REGULATION IN CULTURED CELLS
129
are tightly coupled, so that net ion permeability is much lower than that measured by isotope self exchange or by experiments in which one ion is exchanged for another across the cell membrane. The ion transport causes no transfer of electrical charge across the cell membrane, and it is therefore characterized as an electrically silent process. Because it canies no current, it does not influence the membrane potential of the cell. Another characteristic is that flow of an ion from one side of the membrane (cis) to the other (trans) is strongly dependent on the presence of transportable ions at the trans side. In such a system the isotope exchange rates are typically independent of the membrane potential, but this is not always true, since potential changes may affect the membrane structure and alter the rates of ion translocation. Changes in potential may also affect ion distribution caused by ion flow through electrodiffusional transport pathways in parallel with the exchange system, and this may secondarily affect the rate of ion exchange. A third mode of transport which has recently attracted considerable attention is the coupled flow (corrunsporr) of anions and cations, illustrated by (g), (h), and (i) in Fig. 2. Cotransport is often identified by the dependence of cation flow on anion concentration, and vice versa. Such experiments, however, only provide evidence for anion-activated cation flow or vice versa cation-activated anion flow. The demonstration of true, tightly coupled cotransport requires evidence that the flows of anions and cations are mutually dependent and have a specified stoichiometry. Such transport systems have a high degree of chemical specificity toward the ions transported, thus most cotransport systems transport chloride but not nitrate, thiocyanate, or sulfate. Proof of true cotransport also requires evidence that the flows of anions and cations are not coupled indirectly through, e.g., changes in pH or membrane potential. Anion diffusion can set up a potential which will drive cations, and anion exchange for OH- can set up pH gradients, which will drive cation/H+ exchange. Thus, there must be evidence that cotransport occurs in the absence of or against such gradients. Most cotransport systems so far described involve coupled transport of anions and cations, the transported complex having a zero net charge, and hence they are electrically silent and do not affect the membrane potential nor are they influenced by it. It is also possible, however, for cotransport systems to transport net charge. The transport systems (g) and (h) shown in Fig. 2 are able to accumulate chloride ions in the cell against an electrochemical gradient. The uphill ion transport is energized by the downhill movement of sodium ions, and is possible only in the presence of an inwardly directed electrochemical gradient for sodium. The cotransport thus depends on subsequent removal of sodium ions from the cell by the action of the Na+,K+ pump (i) in Fig. 2. This active transport of sodium and potassium is fueled by the catalyzed hydrolysis of adenosine tri-
130
ELSE K. HOFFMANN
phosphate. The Na+ ,K+ transport is a “primary active” transport process because it is directly coupled to the degradation of chemical energy stores. In contrast, the cotransport of chloride against an electrochemical gradient is classified as a “secondary active” transport process because the energy for the uphill chloride movement is derived from the sodium movement and depends on the maintenance of a sodium gradient by the primary active transport process.
3.5-
-E;
3.0.
c
0
H2O Perfect osrnorneter
t CI-
2501
Time after change in osrnolarity (rnin)
FIG.3. Regulatory volume increase in Ehrlich ascites cells after increase in external osmolarity. The cells were pretreated by exposure to low external osmolarity (225 rnOsm) for 20 to 40 min. At zero time a tonicity of 300 mOsm was restored by addition of one-fourth volume of a double-strength saline solution, and cell water, K +,and CI - content were followed with time. The cell water content calculated for a perfect osmorneter is indicated by the broken line. The original cell water content in the 300 mOsm saline solution before the hypotonic pretreatment is indicated on the figure for reference. (Reproduced from Hoffmann er al.. 1983.)
131
VOLUME REGULATION IN CULTURED CELLS
111.
REGULATORY VOLUME INCREASE (RVI)
Following shrinkage in a hyperosmotic saline a regulatory volume increase (RVI) can be observed in some tissues and cell types. Other cells simply shrink as osmometers with no sign of volume recovery (see, e.g., Hendil and Hoffmann, 1974). Nearly all cells shows a volume recovery, however, when hypotonic pretreatment, resulting in a net loss of KCI, is followed by restoration of tonicity. After the initial shrinkage the cells recover their volume with an associated KCI uptake (see Fig. 3). Figure 4 gives a summary of the mechanisms shown to be involved in RVI in different cell types. Among the vertebrate cells that have been more intensively studied are the cells shown in Table I, which summarizes the proposed mechanisms of various RVI responses. The evidence for the mechanisms suggested is given below.
( n ) Normal
Cell volume volume
Na+
3 NO+ Na'
0
Yo
Cotransport system
(2)
Cotransport system
(3)
CI-
Exchange systems
(1) FIG. 4. Types of ion fluxes activated in various cell types during regulatory volume increase (RVI). ( I ) Electroneutral Na+ /H+ exchange functionally coupled to CI-/HC03 - exchange, giving NaCl uptake without change in pHi. Both RVI (2) Na , K + , 2C1- cotransport, and RVI (3) Na+ , CI - cotransport, are examples of secondary active transport driven by the sodium gradient. Note that in all the RVI mechanisms the cell takes up NaCl with subsequent replacement of Na by K via the N a + / K + pump, which is the only primary active transport involved. See Table I for examples. +
+
+
132
ELSE K. HOFFMANN TABLE I VOLUME-REGULATORY ION TRANSPORT MECHANISMS IN VARIOUS VERTEBRATE VOLUME INCREASE (RVI) CELL TYPESACTIVATEDDURING REGULATORY
hOWSED
Mechanism Amiloride-sensitive Na +-H exchange functionally coupled to CI - , H C 0 3 exchange (see Fig. 4)
Cell type
Referenceso
Amphiumu RBC
+
Furosemide- and bumetanidesensitive Na+ , K , 2CI cotransport, or Na , CI cotransport (see Fig. 4)
Lymphocytes Dog RBC Necrurus gallbladder apical membrane Avian RBC
+
14,15,16,17 18,19,20,2 1.22.23
+
Ehrlich ascites tumor cells
24,25,26,27
Frog skin basolateral membrane
28.29.30
Key to references: (1) Kregenow (1981), (2) Cala (1980), (3) Cala (1983a), (4) Cala (1986). (5) Siebens (1985). (6) Siebens and Kregenow (1985), (7) Kregenow er al. (1985), (8) Grinstein et al. (1983a), (9) Grinstein er al. (1984c), (10) Grinstein er al. (1985a), ( I I ) Parker (1983~).(12) Parker (1986), (13) Spring and Ericson (1982), (14) Kregenow (1971b). (15) Kregenow (1981), (16) Kregenow er al. (1976), (17) Schmidt and McManus (1977b), (18) Schmidt and McManus (1977a). (19) Palfrey er al. (1980), (20) Haas er al. (1982), (21) Kregenow and Caryk (1979), (22) BakkerGmndwald (1981). (23) Palfrey (1983), (24) Hoffmann (1982). (25) Hoffmann er al. (1983), (26) Hoffmann (1985a), (27) Geck and Pfeiffer (1985), (28) Ussing (1982). (29) Ussing (1985), (30) Ussing (1986).
A. Furosemide- and Bumetanide-lnhibitabie Catlon-Anion Cotransport involved In RVI
Since the involvement of cation-anion cotransport in RVI was first demonstrated in avian red blood cells, I shall here give a very brief overview of RVI in avian red cells as a reference background for the description of RVI in other cells. For a recent new review on Na+ , K , C1--cotransport systems see Chipperfield (1986). +
1. AVIANRED CELLS When bird red cells [e.g., pigeon red cells (0rskov, 1954), duck and turkey red cells] are suspended in hypertonic media, the cells at first shrink osmotically, but then swell back toward their initial volume (Kregenow, 1971b). The data showed that cell volume was regulated at an external [ K + ] of 15 mM but not at 2.8 mM. This RVI seems to involve influx of Na+ and K + , accompanied by C1- . Careful kinetic studies confirmed the coupled influx of Na and K after cell shrinkage (Schmidt and McManus, 1977a), and net flux measurements under a variety of conditions indicate a Na+ :K stoichiometry of 1:1 (Schmidt +
+
+
133
VOLUME REGULATION IN CULTURED CELLS
and McManus, 1977b; Haas er al., 1982). A similar coupled influx of Na+ and K + is activated by catecholamines such as norepinephrine (Schmidt and McManus, 1977b; Riddick et a l . , 197 1) and the kinetics suggest that the same flux mechanism is activated in each case (Schmidt and McManus, 1977b,c; Kregenow, 1973). The activation by norepinephrine seems to be mediated by changes in intracellular cAMP (Riddick et al., 1971) and may involve changes in the phosphorylation of specific sites on a 180,000-Da protein known as goblin (Alper er al., 1980a.b). The mechanism of stimulation by cell shrinkage is different, since formation of cAMP is apparently not involved (Kregenow er al., 1976). In both cases, the movements of Na and K are insensitive to ouabain (Kregenow, 1971b, 1973), but are inhibited by “loop” diuretics such as furosemide (Schmidt and McManus, 1977a.b) and bumetanide (Palfrey et al., 1980). The involvement of C1- in Na plus K cotransport was first suggested from kinetic considerations (Kregenow and Caryk, 1979; Bakker-Grunwald, 1981). The kinetic approach of Kregenow and Caryk (Kregenow and Caryk, 1979) showed that when SITS was present to inhibit the anion exchange system, cationcoupled C1- fluxes were unmasked. Definitive evidence for cotransport of two C1- ions with Na and K was provided based upon thermodynamic considerations by Haas et al. (1982), who demonstrated that when valinomycin is used to change the membrane potential (V,), the Na+ flux is not affected. Experiments with valinomycin (to keep V , equal to the Nernst potential for K + ) showed that the Na+ flux is dependent on the chemical chloride gradient, but not on the membrane potential, arguing strongly for electroneutral C1- cation cotransport rather than an effect of V , on an electrogenic Na+ + K + cotransport. Finally, it was shown that a C1- gradient can drive Na+ uphill, even if the membrane potential is kept constant with valinomycin (Haas er al., 1982). These data strongly suggest that cell shrinkage or catecholamines activate a Na+ K+ + 2C1- cotransport mechanism. It is interesting to note, as discussed by Cala (1983a), that the equation for the force driving net ion flow across the membrane assuming electrogenic Na K cotransport (Schmidt and McManus, 1977~) is exactly the same as the equation obtained on the basis of electroneutral Na + K + 2C1- cotransport (Haas et a l . , 1982). This is the case because in the earlier experiments CI - was always at electrochemical equilibrium, thus Schmidt and McManus (Schmidt and McManus, 1977c) misinterpreted the significance of CI involvement by assuming it was important only as it related to the membrane potential. This example demonstrates that conclusions about the number of cotransported species from thermodynamic driving force experiments are only valid if all of the possibly cotransported ions have been displaced from electrochemical equilibrium. If some ions are at electrochemical equilibrium, such experiments do not reveal whether or not these ions are involved in the cotransport system (see Cala, 1983a). +
+
+
+
+
+
+
+
+
+
+
+
134
ELSE K. HOFFMANN
It is suggested from kinetic studies (Palfrey et al., 1980; Haas and McManus, 1982, 1983) that bumetanide binds to a chloride site which is affected by cations. This may prove useful in defining a bumetanide-binding component involved in the transport process in avian red cells, as has recently been done (using [3H]bumetanide) for similar transport processes in the kidney (Forbush and Palfrey, 1983) and in the Ehrlich cells (Hoffmann et al., 1986a). The mechanism which triggers the flux increase during RVI is not known. Experiments with ATP-depleted cells suggest, however, that both volume-dependent and catecholamine-stimulated fluxes are dependent on cell metabolism (Palfrey, 1983). 2. EHRLICH ASCITESTUMORCELLS Evidence for electrically silent cotransport of anions and cations in single mammalian cells was first presented by Geck et al. (1980), who reported a furosemide-sensitiveinflux of Na+ , K , and C1- with a stoichiometry of 1:1:2 into Ehrlich cells which have been previously depleted of K + and loaded with Na+. The flux of each ion is dependent on the presence of the others, and activation of K + flux is hyperbolic with respect to C1- concentration. Activation of this flux does not affect the membrane potential, and changes in membrane potential have no influence on the flux, indicating an electrically neutral cotransport process. The flux is linearly related to the combined chemical potential gradients for the ions, but is nonzero when the combined chemical potential is zero. This suggests possible coupling to metabolic energy (or inaccuracies in estimating chemical potentials in cells), but no coupling to ATP hydrolysis was observed (Geck el al., 1980). In Erhlich cells at steady state, unidirectional K flux is linearly dependent on the chloride concentration (Bakker-Grunwald, 1978), and the stoichiometry of C1- to K + flux is near one (but sometimes less than one) (Aull, 1981), in fair agreement with the concept of 1:l KCl cotransport, but not with Na+, K + , 2C1- cotransport. Levinson (1985) has recently examined the role of Na+ in cotransport under steady-state conditions. The results are consistent with the proposal that the C1-Qdependent cation cotransport system, when operating at steady state, mediates an exchange of KCl for KCl or NaCl for NaCl, with the relative proportions of each being determined by the extracellular "a+]. At low extracellular Na+ concentrations (1 to 55 mM) the cotransport system mediates the flux of KCl but not of NaCl across the membrane. However, with increasing Na+ concentrations (60to 150 mM), progressively less KCl is transported. The decrease in KC1 transport is offset by an increase in NaCl transport such that the sum (KCl + NaCl) is maintained relatively constant. +
+
a. Activation of Cotransport during RVI. When Ehrlich cells are preincubated in hypotonic medium, in order to reduce their intracellular ionic content (see Section IV,A), and then resuspended in isotonic medium, the initial osmotic
135
VOLUME REGULATION IN CULTURED CELLS
cell shrinkage is followed by a rapid volume increase back to the original cell volume (Hoffmann et al., 1983). This involves a net KCl uptake followed by a parallel water uptake (see Fig. 3). It was proposed that the primary process is an activation of an otherwise quiescent bumetanide-sensitive Na /C1- cotransport system, with subsequent replacement of Na+ by K + via the N a + / K + pump, stimulated by the Na+ influx (Hoffmann et al., 1983). The evidence for the activation of a cotransport system during RVI can be summarized as follows. +
1. The net K and water uptake was found to be C1- dependent, as seen from the inhibition of the uptake following substitution of NO,- for CI- (Fig. 5). Moreover, the net uptake of KCl and water is Na+ dependent: the uptake was abolished in media with Na+ replaced by choline (at 5 mM external K + ) (Fig. 5 ) , and inhibited in Na -free media with K substituted for Na during restoration of toxicity (not illustrated). The RVI is inhibited by furosemide or bumetanide, but is not inhibited by DIDS or amiloride arguing against the involvement of functionally coupled Na /H and C1- /HCO, - exchanges. 2. The observed net C1- flux during RVI was more than 10-fold larger than the conductive net C1- flux (Hoffmann et al., 1983). Under steady state conditions the bumetanide-sensitive component of the unidirectional 36Cl flux was negligible (Table II), and the unidirectional 36C1flux was unaffected by substitu+
+
+
+
+
+
cell w a t e r ( m l / g d r y w t = m i n )
1
0.20
control
NO,
Choline
Furosemide
Bujnetanide
1mM
10pM
I FIG. 5 . Initial rate of water uptake during RVI in Ehrlich ascites tumor cells. The cells were pretreated at low external osmolarity (225 mOsm) for 20-40 min; 300 mOsm tonicity was restored at zero time. and cell water, K + ,and CI- content were followed with time during the RVI (only the data for cell water are shown). The choline medium contained the same K concentration ( 5 mM) as +
the standard medium. Furosemide and bumetanide were added with the double-strength saline solution used for restoration of tonicity a1 zero time. The rate of uptake was determined as the slope of the first part of the uptake curves. (From Hoffmann, 1985b.)
136
ELSE K. HOFFMANN
tion of choline for Na,+ . In contrast, during RVI a large bumetanide-sensitive 36Cl influx was observed which amounted to about half the total unidirectional W l flux (Table 11). Geck and Pfeiffer (1985) have also demonstrated that the cotransport system is silent at a cell volume above a threshold volume which is in the range 3.0-3.5 ml cell water/g dry wt. It should be noted that the cells under steady state conditions in our laboratory have a slightly higher cell volume of about 3.8 ml water/g dry wt (Hoffmann et al., 1984). 3. The flux ratio for the bumetanide-sensitive component of unidirectional C1- influx and efflux during RVI was 1.9 (Hoffmann et al., 1983). Flux ratio analysis of the bumetanide-sensitiveCl - flux under these conditions shows that the flux can only be accounted for by NaCl cotransport, since the driving force for Na , K , 2C1- cotransport or other processes would not be sufficient to account for the magnitude of the ratio influx/efflux (Hoffmann et al., 1983). 4. Under conditions where the Na+ / K + pump was inhibited with ouabain a bumetanide-sensitive uptake of Na+ and C1- in about equimolar amounts could be demonstrated (Fig. 6), whereas the uptake of K + was negligible. This provides direct experimental evidence for the involvement of the Na+ ,K+ pump, and supports the conclusion that an Na , C1- cotransport is involved rather than +
+
+
TABLE I1 ACrlVATlON OF BUMETANIDE-SENSITIVEUNIDIRECTIONAL c1- FLUXES DURING RVI IN EHRLICH ASCITESTUMORCELLS" Unidirectional 36cI influx (pmollg dry wt x min)b Experimental condition
Total
Physiological steady state RVI (initial phase)
43.6 2.0 (7) 67.5 f 3.1 (8)
*
Bumetanide-sensitive component
-0.8 f 2.3 ( 3 ) 38.9 2 3.4 (3)
The cells at steady state were incubated in 300 mOsm saline solution, and the unidirectional CI - influx was determined as the tracer exchange flux under steady-state conditions. Bumetanide (25 p M ) was added 20 min before the W1 addition for the flux measurements. For the measurements during RVI the cells were pretreated at low external osmolarity, with restoration of 300 mOsm tonicity at zero time. 36CI was added with the double-strength saline solution used for restoration of tonicity, and the cell 36CI activity was followed with time during the RVI. The initial rate of W I influx was determined as the slope of the first part of the influx curve. Bumetanide (25 p M ) was added at zero time with the double-strength saline solution (values are from Hoffmann et al. (1983). b The values are given as mean t SEM with the number of independent experiments in parentheses.
137
VOLUME REGULATION IN CULTURED CELLS
I
Na'
CI-
250-
Ouabain + bumetanide
200. Ouabain+ burnetanide
150.
0
0
5
10
1
t
0
0
5
10
1
Time after change in osmolarity ( m i n I FIG.6 . Effect of bumetanide on Na and CI - uptake during regulatory volume increase (RVI) +
in the presence of ouabain. Experimental protocol as in Fig. 3. The volume recovery and the ion uptake were monitored in the presence of ouabain with and without addition of bumetanide. The data presented are from two independent experiments, both showing inhibition of volume recovery (data not shown) and of Na+ uptake (left frame) and CI- uptake (right frame).Ouabain (1.5 mM) was added at zero time in the experiment shown in the right panel and 2 rnin before the addition of the double-strength saline solution used for restoration of tonicity in the experiment shown in the left panel. In both experiments bumetanide (50 pM) was added with the double-strength saline solution at zero time. No significant net movements of K + were found in the presence of ouabain nor in the presence of ouabain plus bumetanide (data not shown). (Reproduced from Hoffmann er 01.. 1983.)
a K + , Na+, 2C1- cotransport system. It is unclear how the Na+/K+ pump is activated. Only a small increase in the Na+ concentration in cell water was observed about 0.5 min after restoration of tonicity, and the further increase in cell Na+ during RVI was insignificant. This process thus seems to involve coupled NaCl influx, possibly triggered in part by low intracellular C1- (Ussing, 1982) acting together with the Na+ .K+ pump to mediate a net KCI influx. It is still unclear whether this mechanism is distinct from the Na , K , 2C1- transport (see above) observed in K -depleted Ehrlich cells (Geck et al., 1980; Geck and Pfeiffer, 1985) and from the KCl and NaCl cotransport seen under certain conditions in steady state (BakkerGrunwald, 1978; Aull, 1981; Levinson, 1985). Maybe the coupled ion transports can all be mediated by the same transport system operating under different conditions and thus functioning in different modes. This question is unsolved for Ehrlich cells as well as for avian red cells (see also Section IV,B,l). +
+
+
138
ELSE K. HOFFMANN
b. The Number of Chloride-Cation Cotransport Sites on Ehrlich Ascites Tumor Cells Measured with rHlBumetanide. Simultaneous measuresments were made of net C1 influx and [3H]bumetanidebinding to Ehrlich ascites tumor cells in which the chloride cation cotransport pathway had been activated by hypertonic challenge following hypotonic pretreatment (Hoffmann et al., 1986a). Bumetanide inhibits net chloride influx during RVI with a K , / 2 of about 6 pM. Since the number of bumetanide molecules bound per cell is a saturable function of bumetanide concentration with a K , of 8.7 f 3.1 pM, which is not significantly different from K L I 2for the inhibition of net C1- influx, it is likely that this predominantly represents specific binding to the transport sites. This is supported by the finding that up to 35 pM bumetanide there is a good linear correlation between inhibition of C1- influx during RVI and numbers of bumetanide molecules bound per cell (Hoffmann et al., 1986a). The extrapolation to the number of bumetanide binding sites per cell at maximal inhibition of C1 transport thus gives the number of cotransport sites per cell: 2.0 X lo6. From this number and the bumetanide-sensitivefluxes measured (not necessarily maximum fluxes), the turnover number is calculated at 50 C1- ions per site per second. The bumetanide-sensitive cotransport system is quiescent under isotonic steady state conditions (Table 11). Nevertheless, these resting cells have the same number of bumetanide-binding sites as the cells where the cotransport system is activated, indicating that recruitment of new transport site is not involved in the activation of anion-cation cotransport. 3. OTHERCULTURED CELLS There are several other cultured cells in which furosemide- and bumetanideinhibitable cotransport is demonstrated but where an eventual role in RVI remains to be established.
a. MDCK Cells. The canine kidney cell line MDCK shows both a K + stimulated Na+ uptake (Rindler et al., 1979) and a furosemide-sensitive K + influx (Aiton et a f . , 1981). Experiments with oriented, high-resistance monolayers of MDCK cells on Millipore filters suggest that there is a K+ transport mechanism at the basolateral cell membrane which is Na activated, furosemide sensitive, and SITS insensitive (Aiton et al., 1982). The fluxes of Na+ and K + are mutually dependent (Aiton et al., 1982; Rindler et al., 1982), with each ion increasing the V,, and decreasing the K , for the other (Rindler et al., 1982). Br- can be substituted for C1-, but other anions greatly inhibit the K + (Rb+) (Aiton et al., 1982; McRoberts et al., 1982) and Na+ (McRoberts et al., 1982) fluxes. There is a hyperbolic dependence of cation flux on C1- concentration (Aiton et al., 1982; McRoberts et al., 1982), which can be accounted for by assuming a dependence on the square of the C1- concentration (McRoberts et +
139
VOLUME REGULATION IN CULTURED CELLS
a l . , 1982). There is no effect of changes in membrane potential or transmembrane pH gradient on the cation uptake, suggesting an electroneutral Na +, K , 2C1- cotransport, which also fits the observed stoichiometry of the fluxes (McRoberts et a l . , 1982). The transport system shows some metabolic dependence, since it is inhibited when ATP is decreased to very low levels, but it does not seem to be directly coupled to ATP hydrolysis (Rindler et a l . , 1982). +
b. Glial Cells. Chloride is not at equilibrium with the membrane potential in astroglial cells, the internal Cl-- being four to five times larger than predicted from the Nernst equation (Kimelberg, 1981). Furosemide or an ethacrynic acid derivative can decrease the steady state internal C1- concentration, suggesting that a specific transport process (possibly anion-cation cotransport) is responsible for the increased internal C1- (Kimelberg and Bourke, 1982). Consistent with this idea, a CI-- or Br- -dependent, Na+ -activated, furosemide- or bumetanide-inhibited, ouabain-insensitive 86Rb uptake has been described in rat glioma cells (Johnson et a l . , 1982). +
c . Other Cultured Cells. Many other cultured cells, including primary chick heart cells, BC,Hl smooth muscle cells, HeLa cells (Aiton et al., 1981), LM(TK-) mouse fibroblasts (Gargus and Slayman, 1980), and BALB/c 3T3 preadipose cells (O’Brian and Krzeminski, 1983), exhibit furosemide-sensitive K and Rb fluxes. In HeLa cells, the K + flux is a saturable function of Na and K + , while the dependence on C1- is sigmoid (Aiton et a l . , 1981). Brsubstitutes for C1-, other anions do not activate the flux, and acetate is strongly inhibitory in the presence of C1- . The data seem compatible with Na+ , K , 2C1- cotransport. In BALB/c 3T3 cells, the Na+ and K + fluxes are mutually dependent, and are stimulated by C1- , suggesting Na+ , K , 2C1- cotransport (O’Brian and Krzeminski, 1983), but the electroneutrality of the system has not yet been established. The furosemide-sensitive Rb+ or N 2 + uptake is rapidly inhibited by the tumor promoter, phorbol 12-myristate 13-acetate (PMA). PMA thus changes the intracellular cation content and the cell volume and these changes may be involved in the growth-promoting effect of phorbol esters (O’Brian and Krzeminski, 1983). +
+
+
+
+
B. Amlloride-Sensitive Na /H Exchange Functionally +
Coupled to CI - /HCO,-
+
Exchange
1. Amphiuma RED CELLS Figure 4 also gives a model of the RVI response of red cells from the salamander Amphiuma means. The volume-regulatory mechanism in these cells is an amiloride-sensitive Na+ / H + exchange (Cala, 1980, 1983a,b, 1985; Kregenow,
140
ELSE K. HOFFMANN
1981; Kregenow et al., 1985; Siebens and Kregenow, 1985). For a recent review on volume-sensitive alkali metal-H transport in Amphiuma red blood cells, see Cala (1986). As in the duck red cell and the Ehrlich cell RVI response, secondary movements of ions occur through the Na+ / K + pump and the anion exchanger, and so these transporters are included in the model. Considering Na and K movements first, Na enters in exchange for H , resulting in an increase in Na+ content and cell volume (Cala, 1980; Siebens and Kregenow, 1985). Importantly, the increase in cell volume results from the exchange of osmotically active Na+ for H , almost all of which comes from the intracellular buffers such as hemoglobin and is therefore not osmotically active. Millimolar quantities of H can be released by these buffers with only minor intracellular pH changes. As intracellular Na content increases, [Na+Ii increases, resulting in a stimulation of the Na /K pump so that part of the Na that has entered is exchanged for K + , and net increases in both Na+ and K + are observed during RVI (Siebens and Kregenow, 1985). Like in Ehrlich cells this net K+ uptake is entirely blocked by ouabain (Siebens and Kregenow, 1985). The Na+ /H+ entry step is completely blocked by 1 mM amiloride (Cala, 1980; Kregenow, 1981; Kregenow et al., 1985; Siebens and Kregenow, 1985), and is in contrast to the duck and the Ehrlich cell RVI response, insensitive to furosemide and bumetanige (Siebens and Kregenow, 1980, 1985). With regard to anion and H + movements, the exchange of Na+ for H + will result in a disequilibrium of H + and HCO, despite the effect of intracellular buffers to minimize pH changes. The anion exchanger will reequilibrate anions so that +
+
+
+
+
+
+
+
+
+
[Cl -Ii/ [Cl - 1, = [HCO,
+
-Ii/ [HCO, - Ii
= [H + I,/ [H
+
Ii
Thus, in the Amphiuma red cell, chloride moves in through the anion exchanger, whereas in the duck red cell and the Ehrlich cell, chloride moves back out if HC0,- is present in the medium. In the Amphiuma RVI response, all net chloride uptake is blocked by inhibition of the anion exchanger by SITS or DIDS (Cala, 1980; Kregenow, 1981). In addition, by blocking reequilibration of pH, SITS also greatly increases the alkalinization of the cell and the acidification of the medium (Kregenow, 1981; Kregenow et al., 1985). The evidence for an Na+/H+ exchange mechanism with a stoichiometry of 1:l is that when the Na+ /K+ pump is blocked with ouabain and the anion exchanger is simultaneously blocked with SITS or DIDS, the amount of H released to the medium during volume regulation (measured by tritration of medium pH changes) is equal to the amount of Na+ entering the cell (Cala, 1980; Kregenow, 1981; Kregenow et al., 1985; Siebens and Kregenow, 1980). Some other findings consistent with the model shown in Fig. 4 are as follows: (1) SITS-treated cells gain less water for a given Na+ uptake than untreated cells, because the SITS-treated cells gain no C1- (Kregenow, 1981; Kregenow +
141
VOLUME REGULATION IN CULTURED CELLS
et al., 1985). (2) Volume recovery is blocked by 1 mM amiloride or replacement of medium Na+ with choline (Sieben and Kregenow, 1985). (3) Cells in hyper-
tonic Na-free choline medium lose Na+ and alkalinize the medium, implying that the Na+ /H exchange process is reversible (Kregenow et a l . , 1985; Siebens, 1985). (4) There is no major change in the membrane potential measured with microelectrodes [these are giant red cells, -70 pm in diameter (Cala, 1980)] or estimated from chloride concentration ratios (Sieben and Kregenow, 1985). (5) The net Na+ flux proceeds according to (ApNa+ - Ap,+) where ApNa+and ApH+are the chemical potentials for Na+ and H + , respectively (Cala, 1985). (6) Substitution of Br-, I - , SCN-, or NO,- for medium CIdoes not substantially affect volume-regulatory Na+ uptake (Kregenow er a l., 1985). Some of these anions are, however, partially inhibitor (Cala, 1983a,b; Kregenow er al., 1985), even when the anion exchanger is blocked by inhibitors. (Cala, 1983a,b). The mechanism of inhibition by these anions is unknown. The Na+ /H exchanger of the Amphiuma red cell RVI response is qualitatively similar to that in other systems (Benos, 1982), such as microvillus vesicles from the rabbit renal cortex (Kinsella and Aronson, 1981, 1982). In the Amphiuma red cell RVI response, ( I ) Na+ uptake is a saturable function of “a+], (K, 39 mM), (2) Li+ substitutes for Na+, whereas K + and choline do not, and (3) amiloride acts as a reversible competitive inhibitor of Na+ uptake, with a Ki of -2 X lop6 M (Siebens and Kregenow, 1980, 1985). +
+
-
2. DOG RED CELLS The dog red cell RVI response has long been known to involve passive movements of sodium (Parker and Hoffman, 1965) which now also appears to be an amiloride-sensitive Na /H exchange process (Parker, 1983~).Shrinkagestimulated Na+ fluxes are almost completely blocked by 0.2 mM amiloride (Parker, 1983c), and evidence consistent with direct coupling between Na+ and H + movements has been presented (Parker, 1983d). Given appropriate ion gradients, Na -driven H movements and H -driven Na movements can be demonstrated (Parker, 1983d). Li+ appears to substitute for Na+, whereas K + and choline do not (Parker, 1983~).As in renal microvillus vesicles (Kinsella and Aronsen, 1981, 1982), Li appears to have a higher affinity for the transport mechanism than does Na+ (Parker, 1983~).A finding in the dog red cell that remains to be explained is that replacement of medium C1- with SCN- is completely inhibitory, and NO, - is partially inhibitory, although both of these anions are transported by the anion exchanger (Parker, 1981). Some insight into the role of anions in this transport system was achieved from experiments with a membrane fixative (glutaraldehyde). The ability of glutaraldehyde to fix the transporter in a functional or nonfunctional state made it possible to distinguish whether the anion effects are exerted on the activation of the transporter or on its function once activated. The conclusion was (see Parker 1984, 1986), that the +
+
+
+
+
+
+
142
ELSE K. HOFFMANN
chloride requirement of the Na+ ,H+ exchanger involves the activation or triggering of the mechanism rather than the transport function. A model for the Na+ ,H+ exchanger in dog red blood cells and its interaction with chloride and lithium ions has been presented (see Parker, 1986). The model supposes that protons, Li+ ions, and chloride have "allosteric" effects on the activation of the exchanger that are separate from the role of these ions as transporter substrates. Chloride is necessary for the triggering of the Na+ ,H+ exchanger by cell shrinkage and lithium is regarded by this transporter as being more like a proton than like Na+ . Raising the concentration of internal Li activates the exchanger. +
3. LYMPHOCYTES The lymphocyte preparation which is used contains 5% (Bui and Wiley, 1981) to 20% (Cheung et a f . , 1982) nonlymphocytes (monocytes, platelets, polymorphonuclear leukocytes), and is therefore often termed a peripheral blood monocyte preparation. Like Ehrlich cells, lymphocytes do not regulate their volume when osmotically shrunken in hypertonic media (Grinstein et a f . , 1983a, 1984c; Hempling et a f . , 1971; Roti-Roti and Rothstein, 1973), but they do regulate their volume if the RVI-after-RVD protocol is used (Grinstein et al., 1983a, 1984~).The reason for this behavior is not known. As in Ehrlich cells, [C1-Ii may play an important role in the activation of the RVI response in lymphocytes (Grinstein et al., 1983a, 1984~). In contrast to the response of Ehrlich cells, the RVI response of lymphocytes appears to involve amiloride-sensitive Na+ /H+ exchange (Grinstein er al., 1983a, 1984c), like that reported for Amphiuma red blood cells. For a recent review, see Grinstein et al., (1986). The response is ouabain insensitive, Na+ dependent, and electroneutral (measured with a membrane potential-sensitive dye). In isotonic cells, stimulation of the Na+ /H+ antiport can be accomplished by lowering the cytoplasmic pH or by addition of P-phorbol diesters. These procedures are also found to induce cellular swelling in lymphocytes, thus resembling RVI (see Grinstein et al., 1985b). Analysis of the cellular ionic content following RVI indicates a net gain of K with only a moderate increase in Na+ . However, cells osmotically shrunken in the presence of ouabain can still volume regulate, but then only the Na content is elevated, with a small decrease in K + content (Grinstein et al., 1983a). It was therefore concluded that net Na uptake is the primary event in RVI and that the increase in K content occurs secondarily, because the Na /K pump is accelerated by the elevated intracellular Na+ concentration (Grinstein et al., 1983a), similar to what has been shown for Ehrlich cells (Hoffmann et al., 1983). Like in Amphiuma red blood cells, amiloride substantially inhibited RVI. The +
+
+
+
+
+
143
VOLUME REGULATION IN CULTURED CELLS
associated increases in total Na * or K content, and the elevated rate of 22Na uptake, are also blocked by amiloride. Since no significant change in membrane potential was recorded during volume regulation (Grinstein et al., 1983a), this finding is consistent with an electroneutral process, such as N a + / H + countertransport. To test the involvement of Na /H exchange in RVI more directly, Grinstein and co-workers undertook measurements of transmembrane H transport during RVI, where the H fluxes were estimated from changes either in the cytoplamic pH (pH,), or of the extracellular pH (pH,) measured in lightly buffered media. Hypertonic shrinking of thymic lymphocytes results in a marked cytoplasmic alkalinization. Several observations suggest that the alkalinization is a consequence of countertransport of extracellular Na+ (Na;) for internal H (Hi+). (1) The ApH, is strictly dependent on the presence of Na,+:(2) The alkalinization is completely prevented by amiloride (Grinstein et al., 1985a,b). Extracellular pH determinations demonstrate that the internal alkalinization is due to the transmembrane displacement of H equivalents. (3) The cytoplasmic alkalinization is found to be accompanied by a commensurate acidification of the extracellular solution. (4) The external acidification is dependent on Nazand is susceptible to inhibition by amiloride (Grinstein et al., 1985a,b). In summary, the available data support an activation of the electroneutral Na /H antiport as the primary event underlying RVI. The mechanism of activation of Na / H exchange is analyzed by comparing the kinetic parameters of transport in resting (isotonic) and hyperosmotically stressed cells. It is concluded that changes in the extracellular cation affinity of the antiport are not involved in the osmotically induced activation (Grinstein et al., 1985a,b). pH, is one of the most important determinants of the rate of Na /H exchange in isotonic cells (Grinstein et al., 1984a,b) and is therefore a likely target for regulation. For this reason Grinstein and co-workers have compared the pH, dependence of the rate of H extrusion in normal and osmotically shrunken cells. Although the form of the relationship between pHi and the rate of Na+ /H exchange is similar in both conditions, osmotic shrinking appears to cause a 0.2-0.3 pH unit alkaline shift of the pH dependence curve. This shift, which renders the quiescent transporter active at physiological pHi (7.0-7. I), can account for the osmotic activation of Na+ /H exchange during RVI. The shift in the pHi dependence of the Na /H exchange probably reflects an altered pH, sensitivity of an allosteric “modifier” site different from the transport site (see Aronson et al., 1982) According to this model, protonation of internal groups at the modifier site increases the rate of N a + / H + exchange. After cell shrinking the “set point” of the modifier, which normally prevents transport at pH, 7. I or higher, is adjusted upward (Grinstein et al., 1985a,b). As a result, the nearly quiescent exchanger is activated, but the activation persists only until pH, attains the new set point (Grinstein et al., 1985a,b). +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
144
ELSE K. HOFFMANN
Can the activation of cation countertransport account for the volume recovery observed in osmotically shrunken cells? Although in principle Na+ /H exchange is an osmotically neutral event, reswelling of the shrunken cells occurs because (1) the cytoplasmic buffering power will replenish most of the extruded H and (2) the elevated cytoplasmic pH will increase the HCO, - content of the cells. This in turns drives C1- into the cell through the Cl-/HCO,- exchange system (see Cala 1983a; Grinstein ef a / . , 1984c, for detailed discussion). With the data presently available, the contribution of C1- to the reswelling can only be approximated, with variations depending on the assumptions made. Maximal effects are expected if the distribution of C1- is mainly determined by at an electroneutral anion exchanger, so that Cl; /Cl; equals HCO,,/HCO,, all times. Smaller effects are expected if the C1- conductance contributes significantly to the distribution of C1- (assuming that both membrane conductance and potential remain constant during RVI). Assuming predominance of the exchange pathway an anion distribution (consistent with the low C1- conductance found in human lymphocytes; Grinstein et al., 1984c), a change of up to 50 mmol Cl-/liter cells can be expected (see Grinstein et al., 1985a,b). When combined, the net gains of Na+ , HCO,-, and C1- and the osmotically obliged water should result in a significant volume increase. In conclusion, an activation of the Na+/H+ antiport and the attendant changes in HC0,- and C1- distribution can largely account for the reported observations of RVI in lymphoid cells. A tentative model of the activation process is given in a review by Grinstein et al. (1986). +
+
IV. REGULATORY VOLUME DECREASE (RVD) In hypotonic media vertebrate and invertebrate cells initially swell by osmotic water equilibration but subsequently regulate their volume (RVD) by a net loss of KC1 and associated loss of cell water [for references see reviews by Rorive and Gilles (1979), Kregenow (1981), Spring and Ericson (1982), Hoffmann (1983, 1985a,b), Cala (1983a, 1985), and Grinstein et al. (1984c)l. Several transport mechanisms have been proposed to be activated during RVD and the principal types are shown in Fig. 7. Table I11 summarizes the proposed mechanisms in the various cell types. In most cell types the KCl loss appears to be mediated via electroneutral ion transport mechanisms. An electroneutral K+ /C1- cotransport involved in RVD has been reported in red cells from duck (see McManus and Schmidt, 1978; Kregenow, 1981; McManus et al., 1985), dog (Parker, 1983a), LK-sheep (Dunham and Ellory, 1981; Ellory et al., 1985a), and fish (Lauf, 1982). In human red cells a volume-sensitive K+/Clcotransport can be stimulated by high hydrostatic pressure (Ellory and Hall, 1985; Ellory et al., 1985a,b; Hall and Ellory, 1986). A C1--dependent K + transport can also be induced by N-ethylmaleimide (NEM) in LK-sheep and goat
145
VOLUME REGULATION IN CULTURED CELLS
(3) Conductlve systems
FIG. 7. Pnncipal types of ion fluxes activated during regulatory volume decrease (RVD). (1) Electroneutral K /H exchange functionally coupled to CI - /HC03 - exchange, giving KCI loss without change in pH,. (2) Electroneutral K CI - cotransport. (3) Conductive K flux functionally coupled to conductive C1- flux. +
+
+
+
red cells (Lauf and Theg, 1980a,b), human red cells (Ellory et af., 1982, 1985b; Wiater and Dunham, 1983; Lauf et a l., 1984; Lauf, 1985; and Ehrlich ascites tumor cells (Kramheft et al., 1986), and moreover by Ca2 depletion in LKsheep red cells (Lauf and Mangor-Jensen, 1984) and Ehrlich cells (Kramheft et al., 1986). In Amphiuma red cells the RVD involves electroneutral, functionally coupled exchange of K + / H + and CI-/HCO,- (Cala, 1980, 1985; Kregenow, 1981; Siebens, 1985). In contrast, separate conductive K and C1- transport pathways have been reported to be activated during RVD in human lymphocytes (Grinstein ef al., 1982a,b, 1984c; Sarkadi et al., 1984a. 1985) and in Ehrlich ascites tumor cells (Hoffmann, 1978; Hoffmann et a l . , 1984, 1986b). A volume-sensitive conductive C1- transport pathway has been reported in the basolateral membrane of frog skin epithelial cells (MacRobbie and Ussing, 1961; Ussing, 1982; Kristensen and Ussing, 1985). A conductive K + transport pathway activated by cell swelling has been reported in the basolateral membrane of frog and toad urinary bladder epithelial cells (Davis and Finn, 1982), in liver cells (Kristensen and Folke, 1984), and in enterocytes (Lau et al., 1984; Schultz et al., 1985). In Necturus gallbladder epithelial cells activation of both K /C1- cotransport and K + conductance in the basolateral membrane during RVD has recently been suggested (Larson and Spring, 1984). In Ehrlich cells at low pH and after Ca2 +
+
+
+
146
ELSE K. HOFFMANN TABLE 111 VOLUME-REGULATORYION TRANSPORT MECHANISMS IN VARIOUS VERTEBRATE CELLTYPESACTIVATED DURING REGULATORY VOLUME DECREASE (RVD)
PROPOSED
Mechanism K , H exchange functionally coupled to C1- , HCOj exchange (see Fig. 8) K , CI - cotransport (see Fig. 7) +
+
+
Conductive CI - flux functionally coupled to conductive K + flux (see Fig. 7)
Cell type
Referenceso
Amphiuma RBC
Avian RBC Dog RBC LK sheep RBC Fish RBC Necturus gallbladder apical membrane Ehrlich ascites tumor cells
6J.8.9 10
Lymphocytes Frog skin basolateral membrane
20,21,22,23,24,25 26.27.28
11
12 13 14,15,16,17,18,19
( I ) Kregenow (1981). (2) Cala (1980). (3) Cala (1983a), (4) Cala (1983b), (5) Cala (1985). (6) Kregenow (1974), (7) Kregenow and Caryk (1979), (8) McManus (1982), (9) McManus et al. (1985). (10) Parker (1983a), (11) Dunham and Ellory (1982). (12) Lauf (1982), (13) Spring and Ericson (1982), (14) Hoffrnann (1978), (15) Hoffmann (1982). (16 and 17) Hoffmann (1985b,c), (18) Hoffmann et al. (1984). (19) Hoffmann ef al. (1986b), (20) Grinstein et al. (1982a), (21) Sarkadi et al. (1984a). (22) Sarkadi ef al. (1984b). (23) Grinstein ef al. (1982b), (24) Grinstein et al. (1984). (25) Sarkadi et al. (1985), (26) MacRobbie and Ussing (1961), (27) Ussing (1982), (28) Ussing (1986).
depletion the combined activation during RVD of K /C1- cotransport and of conductive K + and C1- transport has also been reported (Kramhoft et al., 1986). +
A. Volume-Activated CI- and K + Conductance Pathways Volume-induced increase in K + permeability (which was presumed to be conductive) has been demonstrated in lymphocytes (Roti-Roti and Rothstein, 1973; Ben Sasson et al., 1975; Bui and Wiley, 1981), in Ehrlich cells (Hendil and Hoffmann, 1974; Hoffmann, 1978), in the basolateral membrane of frog and toad urinary bladder epithelial cells (Davis and Finn, 1982), in liver cells (Knstensen and Folke, 1984), and in enterocytes (Lau et al., 1984; Schultz et al., 1985). In the earlier papers on isolated cells the anion conductance was simply assumed to be initially high, although this assumption was not tested. Activation by cell swelling of uncoupled, conductive fluxes of K and CI - in single cells was first proposed for Ehrlich cells (Hoffmann, 1978) based on C1+
147
VOLUME REGULATION IN CULTURED CELLS
flux measurements and the demonstration of an initially low C1- conductance (Heinz et al., 1975; Simonsen et al., 1976), and later described in detail in lymphocytes (Grinstein et al., 1982a,b, 1984c; Sarkadi et al., 1984a, 1985; see Section IV,A,2) and in Ehrlich cells (Hoffmann, 1978; Hoffmann et al., 1984, 1986b; see Section IV,A, 1). A volume-sensitive conductive C 1 transport ~ pathway was reported in the basolateral membrane of frog skin epithelial cells already in 1961 (MacRobbie and Ussing, 1961; see Ussing, 1982; Kristensen and Ussing, 1985). 1. EHRLICH ASCITESTUMORCELLS The conductive C1 permeability accounts for only about 5% of the value deduced from tracer flux measurements. This was shown by Heinz et al. (1973, who used the valinomycin technique of Hunter (1977). The same conclusion was reached using a different electrophysiological approach (Simonsen et al., 1976; Hoffmann et al., 1979). The membrane potential (V,) was measured as a function of the concentration of external K + , substituting K + for Na+ . The transference number for K (IK+) was estimated from the slope of V , against log,, [K,+], and tc,- and tNa+ were calculated, neglecting current carried by ions other than C1- , K , and Na . The diffusional net flux of K was calculated from the steady state flux of 42K+, assuming the flux ratio equation to be valid for the passive K + fluxes. From this value the K + conductance and the Na+ and CIconductances were calculated. They were all measured at about 5 p S (siemens) c m P 2 [data from Hoffmann et ul. (1979), corrected for previous underestimate of the area/volume ratio (see Hoffmann et al. 1986a)l. The conductive permeabilities derived from the conductances are given in Table IV. The conductive C1+
+
+
+
TABLE IV THEMEMBRANF. C O N D l l C T l V t PLRMEABILIIY TO K , Na , CI - A N D T H OSMOTIC ~ PERMLABILITY TO W A r E R IN EHRLICH ASCITESTUMOR CELl S +
+
Permeability constant (10-8 cm sec-1) K+ Na C1H,Oa +
3.6 I .5 1.4
8.9 x 105
0 From Henipling ( 1967); the other values are from Hoffmann et a / . ( 1979) corrected, for previous underestimate of the area/volume ratio (see Hoffmann er a / . . 1986a).
148
ELSE K. HOFFMANN
permeability at 1.4 X cm sec- * in Ehrlich cells at 38°C is similar to that reported in the human red blood cell (see Knauf, 1979) and in red blood cells from sheep (Tosteson er al., 1973) and dog (Parker ef al., 1977). It should be noted that the Na+ and K+ conductive permeabilities of human red cells are about two orders of magnitude lower than the conductive permeability to C1- , whereas they are all of the same order of magnitude in the Ehrlich cell (see Table IV). When Ehrlich cells are suspended in hypotonic media, however, a RVD response occurs which decreases the Na+ permeability but greatly increases both the K and C1- permeability of the cells and causes a net efflux of K and C1to decrease the cell volume toward normal volume (Hendil and Hoffmann, 1974; Hoffmann, 1978). The chloride permeability (Pel) is increased to a greater extent that that of K + , and it is concluded that the K + permeability limits the rate of volume change during RVD since volume changes are more rapid when val+
+
2.01
Gramicidin
I Quinine
1.5-
1.0--
I. 0
2
&
6
Time (minl
FIG.8. Effect of quinine and gramicidin on regulatory volume decrease in Ehrlich ascites tumor cells in Na+-free choline medium. Ehrlich ascites cells were preincubated at 3 to 8% cytocrit in standard incubation medium containing Ca2+ ( 1 mM). At zero time a sample of the cell suspension was diluted 500- to 1500-fold (final cell density about 70,000cells/ml) with hypotonic (150 mOsm) choline medium containing Ca2+ (0.5 mM), and the cell volume was followed with time using a Coulter counter. The initial cell volume was measured by dilution of a parallel sample of the cell suspension in standard incubation medium (mean value of two experiments). In one group quinine (1 mM) was present in the hypotonic choline medium (0). In one group gramicidin (5 ILM) was added at the time indicated by the m o w (M). The experiment illustrated was performed at pH 8.2 but similar results were obtained in experiments at pH 7.2. (From Hoffmann er nl.. 1986b.)
149
VOLUME REGULATION IN CULTURED CELLS
inomycin (Hoffmann et al., 1984) or gramicidin (Hoffmann et al., 1986b) is used to provide a parallel pathway for K + . The volume change in the presence of gramicidin has been used to monitor the changes in Pel. P,, increases abruptly when the cells are swollen but the change in P,, is transient, with inactivation within about 10 min. The initial increase in P,, is about 60-fold. As discussed in detail in the following sections, several points of evidence demonstrate that the CI- conductance pathway is separate from the K + permeation mechanism. The chloride transport pathway has a different anion selectivity and sensitivity to inhibitors than the anion exchange system and the cation-anion cotransport system in Ehrlich cells (Hoffmann et al., 1986b). The volume-sensitive pathways can also be activated in isotonic cells; thus, addition of the Ca2+ ionophore A23187 in isotonic medium induces a fast net loss of KCI both in Ca2 -containing and Ca2 -free media (Hoffmann et al., 1984). The net loss of KCI induced by A23187 shows several parallels to the KCl loss induced during RVD, and it is demonstrated that A23187 induces a substantial increase of both the conductive K + and the conductive C1-permeability. The A23187-induced increase in P,, in Ca2+-free media (which is probably mediated by release of Ca2 from internal stores) is transient like the activation during the volume response, whereas the activation of P,, is persistent in Ca2 containing media (Hoffmann et al., 1986b). These findings suggest that a transient increase in free cytosolic Ca2 may account for the transient activation of the C1- transport pathway during RVD. The increase in P,, is 12-fold after addition of A23187. The potassium permeability increases to a somewhat greater extent than that of chloride. The ionophore A23 187 plus Ca2 -induced increase in P , is estimated at 21-fold. The number of Ca2+-activated K + channels is estimated at about 100 per cell assuming a single channel conductance at 20 pS (Hoffmann et af., 1986b). A comparison of the A23187-induced K + conductance estimated from tracer flux measurements at high external K conductance estimated from tracer flux measurements at high external K and from net flux measurements suggests single file behavior of the Ca2 -activated K channel involved in RVD. Anti-calmodulin drugs, e.g., pimozide block the volume- or A23187-induced C1- transport pathway as well as the volume- or A23187induced K transport pathway. It is proposed that the Ca2 activation of both the K + and the C1- channel during RVD is mediated by calmodulin (Hoffmann et ul., 1984, 1986b). Preliminary results (B. Aabin and B. Kristensen, 1987) have demonstrated that calmodulin is found in the cytoplasma of Ehrlich cells and that the cell membranes bind calmodulin in a Ca2+-dependent manner. These findings support the notion that calmodulin plays a role in the Ca2+ activation of the channels. The main evidence for these findings is given in Section IV,A,l,a through Section IV,A,l,f. +
+
+
+
+
+
+
+
+
+
+
+
150
ELSE K. HOFFMANN
a . Activation of Separate K+ and C1- Transport Pathways by Cell Swelling and by Addition of Ionophore A23187. During RVD the apparent K + permeability is increased (Hendil and Hoffmann, 1974; Hoffmann, 1978; Hoffmann et al., 1984). A concomitant volume-induced increase in C1- net permeability is demonstrated in Figs. 8 and 9. Under conditions where the Ca2+-sensitive K + channel is blocked by quinine, the addition of gramicidin, which imposes a high K+ permeability, induces a fast net efflux of KCI, demonstrating a high C1- net permeability which greatly exceeds the low C1- net permeability found in unperturbed cells (Hoffmann et al., 1984). The ionophore A23 187 when added to a suspension of Ehrlich ascites cells in steady state induces a net loss of KCI with associated cell shrinkage (Hoffmann et al., 1984, 1986b). A substantial concomitant increase in net K + and C1permeabilities is demonstrated, and addition of ionophore A23187 also induces a substantial activation of the unidirectional 42K fluxes and 36Cl fluxes (Hoffmann et al., 1986b). The KCI loss induced by ionophore A23187 shows in several respects parallels to the KC1 loss observed during RVD: (1) the volume response is in both cases unaffected by substitution of nitrate or thiocyanate for C1- (Hoffmann e f a l . , 1984, 1986b), which provides evidence against KCl loss via a volume-sensitive, C1--dependent cotransport system, similar to that reported in erythrocytes of a variety of species (see Section IV,B,l); (2) the KCI loss is in both cases inhibited by quinine (see Figs. 8 and lo), suggesting the involvement of the Ca2+dependent K+ transport pathway found in several cell types (see reviews by Lew and Ferreira, 1978; Schwartz and Passow, 1983) and also demonstrated in Ehrlich cells (Valdeolmillos et al., 1982); (3) the anti-calmodulin drug pimozide inhibits the KCI loss in both cases (see Fig. 11; Hoffmann, 1985c), suggesting that calmodulin may play a role in the activation of the K + and C1- transport pathways; (4) the KCI loss can in both cases be observed in Ca2+-containing as well as in Ca2+-free media (containing excess EGTA) (see Hoffmann et al., 1984), suggesting that release of Ca2+ from internal stores or perhaps modulation of Ca2+ sensitivity may play a role in the activation of the transport pathways. Quinine blocks the net loss of KCl induced by cell swelling or by ionophore A23 187, but the quinine inhibition is overcome when the blocked K conductance is bypassed by addition of valinomycin or gramicidin (Figs. 9 and 10). These findings demonstrate that quinine blocks a separate K transport pathway activated in parallel with the C1- transport pathway. An increased 36Cl- flux induced by ionophore A23 187 in the presence of quinine is directly demonstrated (Hoffmann et al., 1986b). The activation of the C1- flux was also observed in the presence of DIDS and bumetanide, which would inhibit Cl- flux via the anion exchange system (Hoffmann et al., 1979; Sjaholm et al., 1981) and via the cotransport system (Hoffmann et al., 19831, respectively, consistent with the +
+
151
VOLUME REGULATION IN CULTURED CELLS
interpretation that the activated C1- flux is effected via a conductive C 1 trans~ port pathway. b. Activation and Inactivation of the C1- Transport Pathway. Time Dependence and Role of CaZ+. The C1- transport pathway is strongly activated during RVD (see Figs. 8 and 9). The increase in CI- net flux is estimated at about 60fold, comparing the rate of cell shrinkage in the presence of gramicidin 1 min after hypotonic exposure (Figs. 8 and 9) with that observed for isotonic cells. The activation of the C1- net permeability during RVD exceeds that of the K + permeability as seen by the acceleration of the volume response by gramicidin (see Fig. 8). After addition of ionophore A23187 the increase in net C1- permeability is estimated at 12-fold calculated from net C1- flux measurements (Hoffmann et al., 1986b). The A23187-induced increase in DIDS- and bumetanide-insensitive 36C1influx is estimated at about 20-fold. These findings demonstrate a substantial A23 187-induced activation of the C1- transport pathway which, however, is smaller than that induced by cell swelling. During ionophore A23187-induced cell shrinkage the CI - permeability is rate limiting, as gramicidin fails to accelerate the volume response, and the calculated K + conductance exceeds the Cl- conductance (Hoffmann et al., 1986b). Figure 9 shows a time-dependent inactivation within 5 to 10 min of the volume-induced C1- -transport pathway. This inactivation may, rather than a time dependence per se, reflect a transient increase in cytosolic free Ca2 . This notion is supported by the finding in Fig. 10 that the ionophore A23 187-induced activation of the C1- transport pathway is transient in the absence of external Ca2 but persistent in the presence of 1 mM external Ca2 . In the latter case the Ca2+ pump is probably swamped by the A23187-induced Ca2+ leak across the cell membrane, although the Ca2 pump in Ehrlich cells has been reported to be extremely powerful (Cittadini et al., 1982; Klaven et a l . , 1983). In Ca2+-free media A23187-induced release of Ca2+ from internal stores (see Arslan et al., 1985) and a subsequent hysteretic activation of the Ca2 pump (see Scharff et al., 1983) is likely to produce a transient increase in cytosolic Ca2 . A similar transient increase in cytosolic Ca2 could conceivably be produced by volumeinduced release of Ca2+ from internal stores or by Ca2+ entry across the cell membrane. An increase in cytosolic free Ca2+ induced by cell swelling has recently been demonstrated in toad bladder epithelial cells using the intracellular Ca2 indicator quin 2 (Chase and Wong, 1985). Moreover, the KCl loss during RVD was reduced following cellular Ca2 depletion in Ehrlich cells (Hoffmann et al., 1984). The effect of cell swelling could, however, also be accounted for by modulation of the Ca2+ sensitivity of the transport pathways, as previously suggested in relation to the pH dependence of the volume recovery in hypotonic media which was found to be similar in the presence and absence of external Ca2+ (Hoffmann et a l . , 1984). +
+
+
+
+
+
+
+
+
152
ELSE K. HOFFMANN Grarnicidin I
I
0
c
5
10
1'5
Time (rnin) FIG. 9. Transient increase of chloride permeability of Ehrlich ascites tumor cells after hypotonic cell swelling in choline medium containing quinine. Experimental protocol as in Fig. 8. pH in the present experiment was 7.2. Ehrlich cells were preincubated at 3 to 8% cytocrit in standard incubation medium containing Ca2+ ( I mM). At zero time a sample of the cell suspension was diluted 500to 1500-fold (final cell density about 70,000 cells/ml) with hypotonic (150 mOsm) choline medium containing CaZ+ (0.5 mM), and the cell volume was followed with time using a Coulter counter. The initial cell volume was measured by dilution of a parallel sample of the cell suspension in standard incubation medium (mean value of two experiments). Quinine (1 mM) was added to the hypotonic choline medium in-order to block the Ca2+-dependent K + permeability (control, open symbols). At the times indicated Gramicidin, 0.5 pA4 (O), was added to impose a high cation permeability. The figure is representative for two independent experiments. (Reproduced from Hoffmann et al., 1986b.)
The finding that Ca2+ is involved in the activation of the C1- transport pathway appears to be at variance with the findings in human lymphocytes where the increase in C1- net permeability induced by A23 187 plus Ca2 is significant (Grinstein er al., 1982b) but not appreciable compared to that induced by cell swelling (Sarkadi er al., 1984a,b). A Ca2+-activated C1- transport pathway has recently been reported in the luminal membrane of salivary gland epithelial cells (Nauntofte and Poulsen, 1984, 1986). +
c. Selecrivity of the Anion Transport Pathway. In the absence of activation of the anion transport pathway the net permeability of the cell membrane is substantially higher to nitrate and thiocyanate than to C1- ,as seen by Comparing the rate of cell shrinkage induced by valinomycin in these media (Hoffmann er al., 1986b). Following ionophore A23 187-induced activation of the anion transport pathway the rate of cell shrinkage in the presence of valinomycin is substantial
,
153
VOLUME REGULATION IN CULTURED CELLS A23[7
Gramicidin
I
1
i
Co’*-f ree
0.8
0
A23187
2
8
6
L
Gramicidin I
\
1mM Ca2‘
cI 0
12
8
4
Time ( m i n )
FIG. 10. Time dependence of chloride net permeability of Ehrlich ascites tumor cells following addition of ionophore A23187 (2 p%f)in Ca2+-free and Ca2+-containing choline medium. The media contained quinine ( I mM) to block the Ca2 -dependent K channel. Gramicidin (0.5 +M) +
+
was added to impose a high cation permeability. In the experiment shown in the upper frame the choline medium was nominally Ca2+ free and contained 0.5 mM EGTA. Choline (0.8 mM) was replaced by Na+ in order to give an external Na+ concentration near electrochemical equilibrium with cell N a + . The experiment is representative for two and three experiments in Ca2+-free and Caz+-containing medium, respectively. The cell volume is given relative to the value measured before addition of A23187 and gramicidin (open symbols). The curves shown in the upper frame are compiled from two experiments marked individually. (Reproduced from Hoffmann et nl., 1986b.)
154
ELSE K. HOFFMANN
both in C1- , Br - , NO, - , and SCN - medium and the absolute increase in the rate of cell shrinkage is of similar magnitude in the four media, indicating the activation by Ca2 of a rather unselective anion transport pathway (Hoffmann et a l . , 1986b). The volume-induced anion transport pathway in human lymphocytes was also found to be rather unselective (Grinstein et al., 1982a). +
d . Role of Calmodulin in the Activation of K+ and C1- Transport Pathways. A number of drugs reported to inactivate the Ca2 -binding protein calmodulin (Weiss et al., 1980), among others the diphenylbutyl piperidine neuroleptic pimozide, have previously been demonstrated to inhibit the KCl loss induced by cell swelling or by A23187 in lymphocytes (Grinstein et al., 1982c) and in Ehrlich cells (Hoffmann et al., 1984; see also Fig. 11). It may be noted that the inhibition by pimozide of the A23187-induced KCI loss is strongly reduced at high external Ca2 concentration indicating that Ca2 is involved in the process which is inhibited by pimozide. The Ca2+-activated K + channel in red cells is sensitive to anti-calmodulin drugs (Lackington and Orrego, 1981; Yingst and Hoffman, 1984). It should be noted, however, that the involvement of calmodulin in the activation of the Ca2 -dependent K channel in red cells is still an open question. Pape and Kristensen (1984) have presented direct evidence supporting the involvement of calmodulin, whereas negative results have been reported by Lew et al. (1982) and by Plishker (1984). In Ehrlich cells pimozide is found to block the volume- or A23 187 plus Ca2 induced KCI loss also in the presence of anions with a high conductive permeability (nitrate or thiocyanate, see Fig. 11; Hoffmann et al., 1986b), and also when a high K + net permeability is ensured by addition of gramicidin (see Fig. 11; Hoffmann et al., 1986b). These findings demonstrate that the volume- or A23 187 plus Ca2 -activated K C1- transport pathways are both inhibited by pimozide. The inhibition by pimozide of the A23 187-activated K+ transport pathway is moreover directly demonstrated by 42K flux experiments (Hoffmann et al., 1986b). Although pimozide is not selective as an anti-calmodulin drug, the findings suggest that calmodulin is involved in the activation of both the K and the C1- transport pathway in Ehrlich cells. +
+
+
+
+
+
+
+
+
e . Conductance of the Ca2+-Activated K+ Channel. Single-file diffusion has been demonstrated in K + channels in giant axons (Hodgkin and Keynes, 1955; Bergenisich and Smith, 1984) and in frog striated muscle (Horowicz et al., 1968), and recently for the Ca2+-activated K+ channel in human red cells (Vestergaard-Bogind et al., 1985). A comparison of the A23187-induced K + conductance estimated from 42K flux measurements at high external K + , and from net K flux measurements, suggests single-file behavior of the Ca2+activated K channel involved in RVD in Ehrlich ascites cells (Hoffmann et al., 1986b). The single-channel conductance of Ca2 -activated K channels ranges +
+
+
+
155
VOLUME REGULATION IN CULTURED CELLS
Gromicidin
1 Pimozide
Control
Control
Chloride medium
2
0
L
Choline medium
0
Thlocyanate medium
2
0
2
L
Time I min 1
FIG. 1 1 . Inhibition by pimozide of the increase in CI- and K + net permeabilities in Ehrlich ascites tumor cells induced by cell swelling. Parallel groups of cells were preincubated at 4% cytocrit for 25 min in CI- medium (standard incubation medium), choline medium, or thiocyanate medium (left, middle, and right panels, respectively) with shift of the medium once after 15 min preincubation. The media were buffered at pH 8.2 and contained 0.15 mM Ca2 and 0.15 mM Mg2 . At zero time the cell suspensions were diluted 1000-fold in the corresponding media with the tonicity reduced to 150 mOsm, and the cell volume followed with time. Pimozide was added to the experimental groups (closed symbols) in the concentrations indicated. In the experiment in choline medium was added I ) at I = I min to control and experimental groups in (middle frame) gramicidin (0.5 @ order to impose a high K permeability. The cell volume is given relative to the initial cell volume, measured by dilution of a parallel sample of the cell suspension in the corresponding isotonic incubation medium. Similar results were obtained in nitrate medium (data not illustrated). (From Hoffmann el a / ., 1986b.) +
+
+
from 20 to about 200 pS (Latorre and Miller, 1983). In the case of the Ehrlich ascites cell the findings suggesting single filing would argue against a “maxichannel. Assuming, therefore, a single-channel conductance of 20 pS similar to that reported for human red cells (Grygorczyk et a l ., 1984) the number of activated K + channels in the Ehrlich ascites cell can be estimated at 7 X lo6 cmP2, or about 100 per cell (Hoffmann er al., 1986b). ”
f. Conclusions. The present findings demonstrate that separate transport pathways for K and C1- can be activated in Ehrlich cells by cell swelling in hypotonic media or by addition of the Ca2+-ionophore A23187. The anion transport pathway seems to be rather unselective, in sharp contrast to the C1-dependent cotransport systems. The activation of the CI- transport pathway in hypotonic media is transient, which may reflect a transient increase in cytosolic free Ca2+ concentration, as suggested by the marked difference in the time dependence of the ionophore A23 187-induced response in the presence and absence of external Ca2 . Calmodulin appears to be involved in the activation of both the K + and the C1- transport pathway. The volume- or ionophore A23187+
+
156
ELSE K. HOFFMANN
induced K transport seems to be mediated via the Ca2 -activated K channels reported in red cells and in several other cell types (see review by Schwartz and Passow, 1983), including Ehrlich cells (Valdeolmillos et al., 1982). There is evidence of single-file behavior of the K + channels in Ehrlich cells similar to that recently demonstrated in red cells (Vestergaard-Bogind et al., 1985). The nature of the volume- or A23 187 plus Ca2 -induced anion transport system and its possible relation to other specific anion transport systems, e.g., the volumedependent, potential-gated anion channels in epithelial cell basolateral membranes (see Ussing, 1986), remains to be explored. +
+
+
+
2. LYMPHOCYTES Anion conductance is small in lymphocytes and, similar to the findings in Ehrlich cells, limits the rate of salt flow induced by ionophores such as gramicidin and valinomycin (Grinstein et al., 1982a). When lymphocytes are suspended in hypotonic media, however, an RVD response occurs which greatly increases both the K and C1- permeability of the cells and causes a net efflux of K + and C1- to decrease the cell volume toward normal values. The chloride permeability (P,,) is increased to a greater extent than that of K + , so that the membrane becomes depolarized (Grinstein et al., 1982a), and it is concluded that the K + permeability limits the rate of volume change during RVD since volume changes are more rapid when gramicidin is used to provide a parallel pathway for K + . The findings discussed above have recently been reviewed (Grinstein et al., 1982b, 1984~). The volume change in the presence of gramicidin has been used to monitor the changes in P,,. P,, increases abruptly when the cells are swollen more than 12% above their initial volume. The response is all or none, and two populations of responding and nonresponding cells can be distinguished near the critical volume threshhold (Grinstein er al., 1984~).The change in P,, is transient and decays with a half-time of 5-8 min (Sarkadi er al., 1984a,b). If the cells earlier than this time are shrunken below the threshold volume the chloride conductance increase promptly disappears. Several points of evidence demonstrate that the C1- conductance pathway is separate from the K permeation mechanism: (1) Substitution of K (by Na +) does not affect the anion flux (Grinstein et al., 1982a). (2) Quinine inhibits the K + pathway at lower concentrations than those which affect the C1- flux (Grinstein et al., 1984c; Sarkadi et al., 1985). (3) DIDS and NAP-taurine partially inhibit, and dipyridamole completely inhibits, the C1- flux with no effect on K flux (Grinstein et al., 1984c; Sarkadi et al., 1985). (4) RVD is present in T lymphocytes, but absent in B cells (Cheung et al., 1982), due to the absence of the K + permeability increase in B cells. Despite this, B cells exhibit a normal anion permeability increase in response to cell swelling (Grinstein et al., 1983b). (5) Depletion of intracellular Ca2+ by A23187 and EGTA inhibits the K + flux +
+
+
157
VOLUME REGULATION IN CULTURED CELLS
increase, but does not affect the anion permeability increase (Grinstein et al., 1984c; Sarkadi et al., 1985). All of these data suggest that increased cell volume opens a new anion conductance pathway in lymphocytes as well as in Ehrlich cells which is different from the K + flux pathway, and which has a different anion selectivity and sensitivity to inhibitors than does the red cell anion exchange system (Grinstein et al., 1984c; Sarkadi et af., 1985). The mechanism whereby changes in cell volume trigger this permeability increase is unknown at present, although it may involve effects of mechanical stress on the membrane. An increased 4sCa efflux during RVD has been demonstrated, suggesting a redistribution of Ca2 inside the cell with a transient increase in cytosolic Ca2 concentration (Grinstein et al., 1982~).Moreover, the KCI loss during RVD was reduced following cellular Ca2+ depletion (see review by Grinstein et al., 1984~).On the other hand, in experiments where the cytoplasmic free Ca2+ concentration was directly monitored in these cells using quin 2, no changes could be demonstrated during RVD (Rink et al., 1983). As discussed by Grinstein et al. (1984c), this finding can be explained by assuming either a ‘‘local’’ increase in cytoplasmic free Ca2+, or a Ca2+-independent activation of K + channels by cell swelling. The absence of a detectable increase in cytosolic free Ca2 during RVD and the inhibitory effect of cellular Ca2 depletion could, however, also be accounted for by modulation of the Ca2+ sensitivity of the transport pathways, as previously suggested in relation to the pH dependence of the RVD in Ehrlich cells which was found to be similar in the presence and absence of external Ca2+ (Hoffmann et al., 1984). A number of anti-calmodulin drugs have been demonstrated to inhibit both K and C1- transport induced by cell swelling, although with a weaker effect on the C1- transport pathway in the case of pimozide and R24571 (Grinstein er al., 1982c, 1983b; Sarkadi et al., 1985). The findings are taken to indicate the involvement of Ca2+ and calmodulin in the volume-induced activation of the K transport pathway. The involvement of Ca2 and calmodulin in the activation of the C1- transport pathway appears to be unlikely, however, In these cells, at variance with the findings in Ehrlich cells, because Ca2+ depletion or repletion had no effect on volume-induced CI- transport, and because the increase in CI- conductance induced by A23187 plus Ca2+ was not substantial (Sarkadi et al., 1984a,b). +
+
+
+
+
+
+
B. K + X I - Cotransport 1. DUCKRED CELLS The first report on volume regulation by duck red blood cells was by Kregenow (Kregenow, 1971a). The study showed that following osmotic swelling the duck red cell regulates volume back to control levels (RVD) in about 90 min.
158
ELSE K. HOFFMANN
This RVD results from activation of a transport system which mediates the efflux of K - and C1- from the cells (Kregenow, 1971a) in approximately stoichiometric amounts (Kregenow, 1981), driven by the K concentration gradient. This K + flux is dependent on the presence of C1- or Br- in the medium and is insensitive to SITS (Kregenow and Caryk, 1979). It has been reported that there is a component of the C1- flux which parallels the increased K+ flux (Kregenow and Caryk, 1979), suggesting cotransport of one C1- with one K + , and electroneutrality of this transport process has recently been established (McManus et al., 1985). The K+ flux increases linearly with increasing C1-, with no sign of saturation as in the case of RVI, and is inhibited only by high concentrations of bumetanide. Bumetanide inhibition is not affected by the C1concentration (McManus, 1982). Despite the differences between RVD and RVI, some data suggest that the two may be different manifestations of a common transport system [for discussion of this hypothesis see McManus (1982)l. +
2. OTHERRED BLOODCELLS When dog red blood cells are swollen in hypotonic media, there is an increase in a C1--dependent K + flux which is inhibited by furosemide, and may involve KCl cotransport (Parker, 1983a). In human red cells a volume-sensitive K + /CI cotransport can be stimulated by high hydrostatic pressure (Ellory et al., 1985a,b; Hall et al., 1982) or by treatment with N-ethylmaleimide (see Section IV,B,4). Another case of possible KCl cotransport is seen when red cells of the oyster toadfish (Opsanus tau) are placed in hypotonic media. K + flux occurs onlywithC1- orBr- present, andnot withN0,-,I-,orSCN-. K + andC1are lost in approximately equal amounts, and the flux is inhibited by furosemide (Lauf, 1982). The fact that the flux is also inhibited by SITS casts some doubt, however, on the concept that KCl cotransport is involved. The process might instead involve K /H countertransport coupled to SITS-sensitive C1- /OH exchange similar to that reported in red cells of the salamander, Amphiuma (Cala, 1983a) (see Section IV,C). The K + permeability of LK (low potassium) sheep red cells also increases with cell volume, while Na+ permeability is unaffected (Dunham and Ellory, 1981). This response is absent or very weak in HK (high potassium) sheep red cells (Ellory et al., 1982). The K + flux in LK red cells is stimulated about twofold when C1- is replaced by Br- , but is greatly inhibited in SCN- , I- , NO,-, isethionate, MeSO,-, or acetate media (Dunham and Ellory, 1981). In the absence of C1- (with MeSO, substitution), there is no volume-sensitive K + flux (Dunham and Ellory, 1981). A very high furosemide or bumetanide concentration (about 2 mM) is required for 50% inhibition (Ellory et al., 1982). The K flux in LK sheep red cells is inhibited by anti-L, ,a specific component of an +
+
+
159
VOLUME REGULATION IN CULTURED CELLS
antiserum against LK sheep red cells (Smalley et al., 1982). Since there are only about 850 anti-L binding sites per cell (Smalley et al., 1982), the flux must involve a small number of sites, or else the effect of anti-L, must be highly cooperative. It is even likely that the number of antigen sites may exceed by an order of magnitude the number of cotransport sites (Smalley ef al., 1982). Although this system is K + selective, it is distinct from a Gardos-type Ca2+stimulated K + channel (Dunham and Ellory, 1981; Ellory and Dunham, 1980). Compression with high hydrostatic pressure also induces a large C1 --dependent K + flux in LK sheep red cells which shows many of the characteristics of the hypotonically induced flux (see Ellory et al., 1985a,b). The N-ethylmaleimide (NEM)-stimulated C1- -dependent K + flux in LK sheep red cells is described in Section IV,B,4a. 3. EHRLICH ASCITESTUMORCELLSAT Low pH OR ARER Ca2 DEPLETION +
In Ehrlich cells Thornhill and Laris (1984) have reported a C1 --dependent and quinine-insensitive net loss of KCI induced by cell swelling or by addition of ionophore A23187 plus Ca2+, which was concluded to occur via an electroneutral K /C1- cotransport mechanism. The results reported by Thornhill and Laris (1984) deviate in several respects from those found in our laboratory (see Section IV,A,l). In their study the rate of KCI loss during RVD (Thornhill and Laris, 1984, Fig. 2) is about 20-fold slower than that found in our laboratory (Hoffmann et al., 1984; Figs. 8 and 11). Ca2+ depletion has been demonstrated to inhibit the RVD in Ehrlich cells (Hoffmann et al., 1984) and in lymphocytes (Grinstein et al., 1982c), and the prolonged preincubation in Ca2 -free media in the protocol of Laris and co-workers may perhaps account for the slow volume response. K+/Cl- cotransport has recently been found to be stimulated by Ca2+ depletion in LK sheep red cells (Lauf and Mangor-Jensen, 1984) and by swelling in Ca2+-depleted Ehrlich cells (Kramhgft et al., 19861, and such an activation would explain the C1- -dependent KCl transport observed by Laris and co-workers. Figure 12 summarizes a number of experiments designed to establish the role of Ca2+ and/or pH in a possible chloride-dependent component of RVD. The figure shows the chloride-dependent component (hatched) and the total volume recovery under the various experimental conditions. In each case experiments were carried out with parallel groups in C1- and NO, - media, and the initial rates of volume recovery were estimated. The chloride-dependent component was calculated by subtraction of the rate of volume recovery in NO,- medium from that in C1- medium in each set of parallel groups. It is seen that at pH 7.4 in the presence of Ca2+, i.e., normal control conditions, only a small (about 8%) fraction of chloride-dependent RVD could be demonstrated. This is in agreement +
+
160
ELSE K. HOFFMANN
with the previously established observation that RVD under standard conditions in these cells is dominated by the opening of independent anion channels and K channels (see Section IV,A,l). In contrast to this, it is also clear from Fig. 12 that if the cells have been preincubated in Ca2 -free medium prior to the hypoosmotic treatment, about 50% of the volume recovery at normal pH becomes anion dependent. Figure 12 also shows that the fraction of initial RVD, which is chloride dependent, is different at different pH values. At high pH volume regulation seems to occur almost exclusively by unselective anion channels whereas at low pH more than 50% of the observed volume regulation is dependent on the presence of C1-. The measurements of K+ ion movements following hypoosmotic shocks at low pH shows a C1--dependent net efflux of K + at 56 ? 6 pEq/g dry wt/5 min (+- SEM in three experiments), which is equal to an average 52% of the total K + loss measured in C1- cells (at pH 6.6). This indicates that at low pH about 50% of the volume regulation occurs by a K /C1- cotransport rather than by independent K and C1- conductive channels (Kramhgft et al., 1986). The above results probably elucidate the difference between our previous results and the results by Thornhill and Laris (1984). Thus, in Ehrlich ascites tumor cells alternative possibilities seem to exist, so that if cell swelling occurs under conditions where the usual Ca2 -dependent K and C1- channels cannot function (e.g., at low pH or after Ca2+ depletion) the cells instead turn on a K+/Cl- cotransport system and thus reduce their volume (Kramhgft et al., 1986) +
+
+
+
+
+
4. N-ETHYLMALEIMIDE (NEM)-STIMULATED
C1- -DEPENDENT K
+
FLUX
a. Red Blood Cells. The sulfhydryl-reactive reagent N-ethylmaleimide (NEM) greatly stimulates a C1- -dependent selective K flux in mature LK sheep red cells (Ellory et al., 1982; Bauer and Lauf, 1983; Lauf, 1985; Lauf and Theg, 1980; Logue et al., 1983) and in all sheep reticulocytes (Lauf, 1983b). This stimulation is absent in mature HK red cells (Lauf and Theg, 1980), indicating that the system disappearsfrom the HK cell during maturation. K efflux occurs in the absence of external K (and Rb ), and is actually inhibited by external K or Rb+, whichrulesout a 1:l K+-K+ exchange (Lauf, 1983a;Logueetal., 1983). NEM-activated K + flux is dependent upon the presence of either C1- or Br- , (Ellory er al., 1982; Logue et al., 1983), and the net K + flux stops when KJK, = Cl,/Cl, (Lauf, 1983a), where c and o refer to intra- and extracellular concentrations, respectively. The NEM-stimulated KCl cotransport in LK sheep red cells is sensitive to furosemide and the sensitivity is greatly enhanced by addition of Rb+ to the +
+
+
+
+
161
VOLUME REGULATION IN CULTURED CELLS 03
I 0 21
I 010
T
0 pH Ca
74
66
+
-
+
82
-
+
-
n 4 6 11 6 3 3 FIG. 12. The chloride-dependent component of regulatory volume decrease (RVD) in Ehrlich cells at pH 6.6, 7.4, and 8.2 with and without external Ca*+. The ordinate gives the initial rate of RVD calculated from the slope of volume response vs time curves (fractional reduction of cell volume per minute). The cells without external Ca2+ were preincuabted in Ca2+-free medium containing EGTA (0.5 mM).Experimental protocol as in Fig. 11. The chloride-dependent part of volume regulation (hatched part of columns) was calculated as the difference between the initial rate of volume regulation in CI --medium (whole column) and in N03-medium in n paired experiments. Chloride-dependent K + loss at pH 6.6 (measured in three separate experiments) was 56 ? 6 peq/g dry wt15 min, equal to an average of 52% of the total K + loss. [Reproduced from Kramh~ftet al. ( 1986).]
162
ELSE K. HOFFMANN
external medium, whereas CI- seems to have no effect on the sensitivity to furosemide (Lauf, 1984). Furosemide seems to bind to a site distinct from the sites for Rb+ (K+) or C1- (Lauf, 1984). There are several similarities between the NEM-stimulated CI - -dependent K+ flux and the C1--dependent K + flux stimulated after volume increase (RVD) in LK sheep red cells (see Section IV,B,2). The affinities for K+ and C1- are similar (Dunham and Ellory, 1981; Ellory et al., 1982; Lauf, 1983a, 1985); both occur in LK but not HK cells (Ellory et al., 1982; Lauf and Theg, 1980), and are inhibited by anti-L antibody (Dunham and Ellory, 1981; Ellory et al., 1982; Logue et al., 1983; Lauf, 1985). Iodoacetamide inhibits the effect of both NEM (Bauer and Lauf, 1983) and cell swelling (Dunham and Ellory, 1981) and Ca2+ is inhibitory to both NEM- and volume-induced KCl transport (Lauf and Mangor-Jensen, 1984). On the other hand, neither response blocks the other: (Ellory er al., 1982) and cells with different volumes show about the same response to NEM (Lauf, 1983a). Furthermore, metabolic depletion inhibits the NEM response, but not the basal or volume-stimulated KCl flux (Lauf, 1983c, 1985), suggesting that the mechanisms which trigger the increase in K+ flux differ in their dependence on cell metabolism. The most simple and likely hypothesis is that the K + transport system itself is the same in both cases, but the triggering mechanism is different. NEM-treated cells are still volume sensitive This, however, would not explain the observation that the volume-stimulated K + flux is sensitive to SITS (Ellory and Dunham, 1980), while the NEM system is not (Lauf and Theg, 1980), or why the volume-stimulated K + flux is better activated by Br- (Dunham and Ellory, 1981) than the NEM-stimulated flux (Lauf and Theg, 1980). For further discussion on the relationship between volume- and NEM-stimulated K+/Cl- flux in LK sheep red cells see Lauf (1985). In addition to sheep red cells, NEM stimulation of K flux is seen in red cells of the LK goat (Lauf and Theg, 1980), cow, rabbit, hamster, hedgehog, rat (Ellory er al., 1982), and in human red cells (Ellory et al., 1982, 1985a,b; Duhm, 1981; Lauf et al., 1982, 1984; Lauf, 1985; Wiater and Dunham, 1983). In human red cells, the NEM-induced flux is K + specific (Falke et al., 1984; Wiater and Dunham, 1983) and is inhibited by Na+ (Ellory et al., 1982). The flux is C1- or Br- dependent (Ellory er al., 1982; Lauf er al., 1982) but with a reduced flux in Br- compared to C1- . The flux is only weakly inhibited by furosemide or bumetanide (Ellory et al., 1982), in contrast to the CI - -stimulated Na+ /K+ cotransport described in Section III,A. It should be noted that both in LK sheep red cells (Ellory et al., 1982; Bauer and Lauf, 1983; Logue et al., 1983) and in human red blood cells (Wiater and Dunham, 1983) NEM at low concentration inhibits the K + flux. +
b. Ehrlich Ascites Tumor Cells. NEM treatment of Ehrlich ascites tumor cells under steady state conditions induces a substantial C1- -dependent net loss of
163
VOLUME REGULATION IN CULTURED CELLS
cellular K + and cell shrinkage (Kramh~ftet a f . , 1986). On the basis of anion substitution experiments it can be calculated that about 60% of the K + loss is chloride dependent. Therefore a C1 --dependent K flux is inducible by NEM in Ehrlich cells. The anion preference of the K + loss is C1- > Br- % SCN- = NO,-. From estimates of membrane potential {using the fluorescent dye, 1 , l dipropyloxadicarbocyanine [DiOC,-(5)]} after treatment with NEM it is concluded that the K loss is a result of electroneutral ion fluxes. Measurements of changes in external pH in unbuffered, HC0,--free media in the presence of the anion exchange inhibitor DIDS showed that only a minor part of the K loss can be attributed to K + /H+ exchange. Thus NEM seems to activate a K + /C1cotransport in Ehrlich cells. In addition, NEM seems to inhibit the previously described C1--dependent Na+ uptake which is activated during RVI (see Section III,A,2). Addition of NEM to cells undergoing RVD after a hypoosmotic shock results in a C1--dependent acceleration of RVD, suggesting that a K /C1- cotransport can be induced also under hypotonic conditions. A similar C1--dependent K efflux can be demonstrated during RVD in calcium-depleted cells or at decreased pH (see Section IV,B,3). The combined results show that Ehrlich cells possess a quiescent K + /CI - cotransport system, which becomes active after interference with SH groups, or during RVD after Ca2+ depletion or pH decrease. +
+
+
+
+
C. K + /H+ Exchange Functionally Coupled to CI - /HCO,- Exchange Since the involvement of K IH exchange was first demonstrated in Amphiuma red blood cells, I shall here give a very brief overview of RVD in this cell type. For a more detailed review on RVD in Amphiuma red blood cells, see Cala +
+
(1986). Amphiuma
The Amphiuma red cell RVD response involves a net loss of K + and C1(Cala, 1980; Kregenow, 1981; Siebens and Kregenow, 1985). It has been hypothesized that the K loss is indicated through a K /H exchange mechanism (Cala, 1980, 1983a,b, 1985). This suggestion is based on the findings that the membrane potential (measured with microelectrodes) does not change during RVD, that volume-regulatory K + loss is approximately the same despite valinomycin-induced changes in membrane potential, and that the medium becomes more alkaline during the RVD response (Cala, 1980). However, direct evidence for coupling between K + and H + movements has not yet been presented. In Ehrlich cells, where separate channels are activated during RVD (see +
+
+
164
ELSE K. HOFFMANN
Hoffmann et al., 1986b) a K + loss in excess of C1- loss can also occur (Hendil and Hoffmann, 1974). To account for the very big difference between the K + and C1- loss during RVD in Amphiuma cells on the basis of separate channels for K and C1- an extremely large proton buffering capacity of the cells would however be needed. The strongest argument in support of a K+ /H+ exchange is, however, that the membrane conductance does seem to remain low during RVD (Cala, 1983a,b). As in the Ehrlich cell and the lymphocyte system, a role for intracellular Ca2 in controlling the RVD response has been proposed in Amphiuma red cells (Cala, 1983a,b). In the Amphiuma system, the effect of Ca2+ is reported to be mainly on electroneutral K + movements (Cala, 1983a,b) as opposed to the Ca2+activated conductive K + movements thought to occur in the Ehrlich cell RVD (see Section IV,A,l) and the lymphocyte RVD (see Section IV,A,2). The arguments presented for a role for Ca2+ as a modulator of K + / H + exchange are based upon inferences drawn from ion flux and electrical measurements and from thermodynamic considerations (Cala, 1983a,b). The flux studies (1) demonstrate similarities between volume- and A23 187-stimulated cell K loss, (2) illustrate a large disparity between net cell K + and C1- loss, an observation consistent with the notion that a substantial fraction of Ca2+-induced K + loss is electroneutral, and (3) establish that K /H exchange during RVD is sensitive to external Ca2+ in the absence of A23187. The electrical studies provide information regarding membrane voltage and, when evaluated in terms of evidence demonstrating that G, is volume independent, they permit an upper limit estimate of the Ca2 -induced conductive K loss. The thermodynamic analysis performed under conditions where the driving forces for A23 187-induced K flux via conductive and electroneutral pathways differ in both magnitude and direction, permit a clear distinction between the two modes of transport. These latter studies establish not only that the A23187-induced K + flux is electroneutral but that it is electroneutral by virtue of coupling with H+ (OH-). Finally, evidence has been presented supporting the hypothesis that K /H exchange and Na+ /H+ exchange are different transport modes mediated by the same membrane component(s) (Cala, 1983a,b, 1985) . +
+
+
+
+
+
+
+
+
+
D. The Role of Amino Acids and Taurine A great relative decrease in the concentration of ninhydrin-positive substances occurs during RVD in Ehrlich cells. This decrease accounts for -30% of the total decrease in osmotically active substances (Hoffmann and Hendil, 1976). The majority of the free amino acid pool is made up of nonessential amino acids such as alanine, glycine, glutamic acid, proline, and aspartic acid. These amino
VOLUME REGULATION IN CULTURED CELLS
165
acids made a major contribution to the intracellular adjustment during RVD (Hoffmann and Lambert, 1983), with the most pronounced effect seen on the cellular concentrations of glycine and of taurine. Similar results have been found for all invertebrate phyla studied so far, as well as for some vertebrates [for a review see Gilles (1979) and also the article by Gilles in this issue]. A priori, such regulation of the amino acid pool may be controlled either by a change in the rates of synthesis and degradation of these compounds, by a modification of protein metabolism, or by a change in the rate of transport through the cell membrane. In Ehrlich cells we have found an increase in the oxidative catabolism of both alanine and glycine under hypoosmotic conditions. In the case of alanine, this increased degradation accounts for 33% of the decrease in cellular alanine content, while the degradation of glycine and taurine plays no significant role. Changes in the rate of protein turnover do not seem to be involved (Lambert and Hoffmann, 1982). Of the three possibilities proposed above, changes in taurine and amino acid membrane transport in hypotonic media are most important in Ehrlich cells. It has been demonstrated that a 30% increase in cell volume causes a 7-fold increase in the diffusional taurine permeability and a 1.5-fold increase in the diffusional glycine permeability (Hoffmann and Lambert, 1983). The same increase in volume doubles the K + permeability (Hoffmann, 1977). The maximal flux for the Na -coupled glycine transport is decreased under the same conditions (Hoffmann and Lambert, 1983). With respect to taurine, the following is demonstrated. The amount of taurine lost from the cells is practically equivalent to the amount gained in the medium. This indicates that modification in the taurine transport parameters is the main mechanism in the regulation of the cellular taurine pool during RVD. Dilution of the medium results in a reduction in the taurine uptake and an increase in the taurine release (Hoffmann and Lambert, 1983). Reducing the extracellular sodium concentration by substitution of NaCl with choline C1 reduces the initial taurine uptake. The unidirectional taurine influx is also reduced when NO,- is substituted for C1-. The reduction in taurine influx seen after dilution of the external medium therefore seems to result from the reduction in external Na+ and C1- (Lambert, 1984, 1985). Reduction in osmolarity, on the other hand, increases cellular permeability to taurine, leading to the increase in passive taurine efflux seen under hypotonic conditions. Thus, an increse in the passive permeability to certain amino acids and a decrease in the Na -dependent amino acid influx results in a leakage of amino acids from the cellular pool to the external medium during RVD in Ehrlich cells. The leakage of taurine is inhibited by pimozide, which is known to inactivate the Ca2 -binding protein calmodulin, suggesting the involvement of Ca2 and calmodulin in the regulation of the taurine leak pathway (Lambert, 1985). +
+
+
+
166
ELSE K. HOFFMANN
V. SUMMARY AND PERSPECTIVES In this article we have discussed the mechanisms whereby Ehrlich cells, lymphocytes, and red cells regulate their volume back toward control values following osmotic swelling or shrinkage. This volume regulation appears to be a consequence of the activation by the osmotic perturbation of previously quiescent ion transport pathways. Among the volume-regulatory mechanisms proposed are bumetanide- and furosemide-sensitive cotransport, amiloride-sensitive Na /H exchange, pimozide-sensitive changes in K + and C1- conductance, and K /H exchange. Many of these transporters are similar to those observed in more complex systems, such as epithelia (see Tables I and I11 for a few examples). Ehrlich cells, lymphocytes, and red cells may thus prove useful as model systems for the investigation of mechanisms of cell volume regulation and ion transport in more complex systems. The future purpose of these studies will be to explore the mechanisms by which volume perturbation activates these regulatory mechanisms. +
+
+
+
A. Possible Activation Mechanisms Although the mechanisms of activation of volume-regulatory responses are poorly understood some factors seem to have an obvious role. In the following we shall briefly deal with the possible regulatory function of intracellular Ca2 , calmodulin, the microfilament network, prostaglandins, leukotrienes, and CAMP. Ca2+ appears to play a key role in the RVD in Amphiuma red cells, human lymphocytes, and in Ehrlich cells (for references see above). In the Amphiuma red cell Ca2+ seems to act as a modulator of K + /H+ exchange (Cala, 1983b). In lymphocytes and in Ehrlich cells there is strong evidence that Ca2+ and calmodulin are involved in the activation of the K transport pathway (Grinstein et al., 1982c, 1984c; Sarkadi et al., 1984a, 1985; Hoffmann et a f . , 1984). In Ehrlich cells also the C1- transport pathway is activated by Ca2 and inhibited by the anti-calmodulin drug pimozide (Hoffmann et al., 1986b). This is in contrast to the C1- transport pathway in lymphocytes which seems to be essentially Ca2 independent (Grinstein et a f . , 1982c; Sarkadi et al., 1984a). The precise role of Ca2+ in RVD is at present unclear. An influx of Ca2+ from the external medium is not necessary to activate RVD (Hendil and Hoffmann, 1974); instead release of Ca2 from internal stores seems to play a role both in lymphocytes (Grinstein et al., 1982c, 1984c) and in Ehrlich cells (Hoffmann et al., 1984). No change in cytosolic free Ca2+ concentration could, however, be detected in human lymphocytes during RVD (Rink er al., 1983). An increase in cytosolic free Ca2 induced by cell swelling has recently been demonstrated in toad bladder epithelial cells using the intracellular Ca2 indicator quin 2 (Chase +
+
+
+
+
+
+
167
VOLUME REGULATION IN CULTURED CELLS
and Wong, 1985). Modulation of the Ca2+ sensitivity of the transport pathways has also been suggested in relation to the pH dependence of the volume recovery during RVD in Ehrlich ascites cells (Hoffmann er al., 1984). Also in epithelial cells there is evidence that volume regulation (RVD) depends on calcium and calmodulin. This has been shown in Necrurus gallbladder cells by Foskett and Spring (1985). The link between cell swelling and the putative transient increase in the intracellular level of Ca2 by Ca2 entry across the cell membrane, or by Ca2+ release from intracellular stores, is unknown. Phosphoinositide metabolism and release of inositol trisphosphate have been demonstrated to play a key role in receptor-mediated Ca2+ mobilization (see Bemdge, 1984; Sekar and Hokin, 1986). but to the best of the authors’ knowledge, inositol trisphosphate has not been implicated in RVD. Calcium and calmodulin play an essential role in the release of arachidonic acid from membrane phospholipids (see Wong and Cheung, 1979; Bemdge, 1982) and in the production of leukotrienes via activation of the lipoxygenase pathway (Feinstein and Sha’afi, 1983). It has recently been demonstrated (Lambert er al., 1987) that Ehrlich ascites cells possess the ability to synthesize and release the arachidonic acid metabolites, prostaglandins, and leukotrienes, and that addition of arachidonic acid stimulates the production of both compounds. During RVD the synthesis of leukotrienes is stimulated while, concomitantly, prostaglandin synthesis is reduced (see Fig. 13). Addition of LTD, accelerates RVD (which is rate limited by the K + permeability), while addition of inhibitors of leukotriene synthesis blocks the volume response, even when a high K conductance has been ensured by the presence of gramicidin. It is proposed that the activation of the K + and C1- transport pathways after hypotonic swelling (Hoffmann er al., 1986) involves an increase in leukotriene synthesis. Addition of PGE, inhibits RVD in hypotonic medium containing sodium but not in sodium-free hypotonic media, indicating that PGE, increases the passive sodium permeability in Ehrlich cells. The reduced prostaglandin synthesis during RVD could thus be responsible for the concomitant reduction in the passive permeability for sodium previously reported by Hoffmann (1978). Moreover, the increased prostaglandin synthesis following addition of arachidonic acid and the resulting increase in sodium permeability could explain the observed inhibition of RVD by arachidonic acid. In Necturus gall bladder cells, cytochalasin B inhibits RVD, suggesting that an intact microfilament network is also a prerequisite for normal RVD. It is proposed that actin filament-mediated fusion of cytoplasmic vesicles with the cell membrane could result in insertion of new ion channels (Foskett and Spring, 1985). A possible role of microtubules and the microfilament network during volume regulation has also been suggested in cultured rat pheochromocytoma cells (Delpire et al., 1985). Mills and Skiest (1 985) have analyzed the role of the cytoskeleton in cell volume regulation using MDCK kidney cells. They propose +
+
+
168
ELSE K. HOFFMANN
-
0
5
10
15
C
I
L
-m3 -
" Q
m
isotonic
100
c L
X
W
0
K> 0
5
hypotonic
10
15
Time ( m i n )
FIG. 13. LTC4 and PGE2 synthesis in Ehrlich cells suspended in standard incubation medium and in hypotonic incubation medium. Ehrlich cells preincubated in standard incubation medium were gently spun down and resuspended in either standard or hypotonic (150 mOsm) incubation of medium and the cytocrit adjusted to 6%. Release of LTC4 and PGEl was followed with time by serially isolating cell-free medium by centrifugation and measuring the LTC4 and PGEl concentrations by radioimmunoassay. The extracellular content (nglg cell dry wt) was calculated from the concentration in the medium (nglml) and the cell density of the cell suspension (g dry wt/ml). The figure is representative of four experiments (Lambert et al., 1987).
that the state of organization of F-actin, which can be influenced by changes in the levels of CAMP, could have a direct effect on the activity, stability, or presence of the membrane elements that play a role in volume control processes. The actual mechanism of activation of the RVI response in osmotically shrunken cells is in most cases unknown. Ca2+, calmodulin, and the cytoskeleton do not seem to be involved in RVI, neither in Ehrlich cells (Geck and Pfeiffer, 1985), red blood cells (Cala, 1983b), or gallbladder epithelial cells (Foskett and Spring, 1985). It is suggested that the cytoplasmic C1- concentration may be involved in activation of both the Ehrlich cell (Hoffmann et al., 1983) and lymphocyte (Grinstein et al., 1983b) RVI responses. However, C1- is thought to play only a permissive role in these cells, modulating the response to
VOLUME REGULATION IN CULTURED CELLS
169
cell shrinkage as first suggested for frog skin epithelial cells (Ussing, 1982). The actual mechanism of activation is unknown. In avian red cells, cAMP may exert a regulatory role on the operation of the Na + , K + , 2C1-- cotransport by inducing a conformational change normally induced by shrinkage (see Siebens, 1985; Palfrey and Rao, 1983). In contrast, the cAMP level seems to play no role in the volume-sensitive cotransport in Ehrlich cells (Geck and Pfeiffer, 1985) and cAMP has been reported to even inhibit cotransport in human red blood cells (Garay, 1982). An Na+ /H exchange is activated by shrinking cells in hypertonic media (see Section III,B) or by activation of protein kinase C in lymphocytes with phorbol esters (see Grinstein et al., 1985b). In both instances the possible mechanism is suggested to be a shift in the pH sensitivity of the modifier site (see Section III,B,3). This could be brought about by phosphorylation of the exchanger itself or of an ancillary protein. The mechanism whereby cell shrinking could stimulate a protein kinase is at present not clear. However, it is conceivable that, as described for red blood cells (Dise et al., 1980), shnnking promotes phospholipase activity in the plasma membrane. This could result in liberation of diacylglycerol from inositol phospholipids which would, in turn, stimulate protein kinase C (see Grinstein et d.,1985b). +
6. Identification of the Membrane Constituents A discussion of differences and similarities of the transport systems at the molecular level must await the development of suitable reagents for identification of the membrane constituents involved in the transport. A step in this direction has already been taken with labeling with specific inhibitors of the different transport system. I shall only mention three examples: (1) attempts to identify the anion exchange protein, (2) attempts to identify some cotransport systems, and (3) attempts to isolate K + and C1- channel proteins. 1. The original identification of the anion exchange protein in red blood cells was based upon labeling with specific inhibitors like DIDS (first reviewed by Cabantchik et al., 1978) and the anion transporter from a red cells has now been sequenced (Kopito and Lodish, 1985). DIDS has also been used to identify the anion exchange protein in Ehrlich cells (Jessen er al., 1986). 2. Labeled loop diuretics have recently been introduced as reversible probes for labeling the C1- -dependent cotransporter in canine kidney membranes (Forbush and Palfrey, 1983), plasma membrane vesicles of shark rectal gland (Hannafin e f al., 1983; Kinne et a l . , 1984, 1983, in cultured epithelial cells (MDCK) (Rugg, Simmons and Tivey, 1986), and in Ehrlich ascites tumor cells (Hoffmann et al., 1986a). The density and turnover number for C1 --dependent cotransport systems
170
ELSE K. HOFFMANN
have recently been determined using r3H]bumetanide in MDCK cells, and in Ehrlich cells in which the cotransport system was activated by hypotonic pretreatment followed by hypertonic challenge. In MDCK cells the density was measured at about 0.5 X lo6 sites per cell with a turnover number about 100 K + ions/site/sec (Rugg et al., 1986). These values are in broad agreement with those reported for Ehrlich cells stimulated by cell shrinkage: about 2 X lo6 sites/cell with a turnover number (not necessarily a maximal value) at 50 C1- ions/site/sec (Hoffmann et al., 1986a). In both cell types, the K , for the saturable high-affinity bumetanide binding was similar to the K,,, for bumetanide inhibition of cotransport, suggesting that the bumetanide binding predominantly represents a specific binding to cotransport sites. In Ehrlich cells the bumetanide-sensitive cotransport system is quiescent under isotonic steady-state conditions (see Table 11). Nevertheless, resting cells showed the same number of bumetanide-binding sites as the cells where the cotransport system was activated by cell shrinkage, indicating that recruitment of new transport site is not involved in the activation of the cotransport system during the RVI response in these cells. In recent experiments attempts have been made to isolate membrane constituents related to cotransport systems. Covalent binding of [3H]bumetanide induced by near-UV light was used by Jargensen et al. (1984) to partially purify a 34-kDa bumetanidebinding protein from pig kidney outer medullary luminal membranes. This protein was assumed to represent (a portion of) the Na+ , K , 2C1- cotransport system probably including the C1- -binding site, although the specificity of bumetanide binding was not tested. UV activation of bumetanide has been demonstrated to produce nonspecific perturbations of other membrane transport systems than the cotransport system in cultured renal epithelial cells (Amsler and Kinne, 1986; Rugg et al., 1986), although the interaction with other cellular components was minimized at low bumetanide concentrations (Amsler and Kinne, 1986). The interaction of bumetanide with multiple membrane components was also reported by Jargensen et al. (see above) and has also been found in the case of Ehrlich ascites tumor cells (Feit, Hoffmann, Kristensen, and Dunham, unpublished). Moreover, in both studies the bumetanide binding to several membrane components was found to exhibit apparent saturation-bindingcharacteristics. The difficulties arising from the possible inappropriate labeling produced by photoactivation of bumetanide has been circumvented in a recent study on membrane proteins from Ehrlich ascites tumor cells, applying affinity chromatography using bumetanide as ligand (Feit, Hoffmann, Kristensen, and Dunham, unpublished). SDS-PAGE electrophoresis of total membrane proteins and membrane polypeptides nonretained and retained on the bumetanide affinity column identified three major membrane polypeptides which are extracted by the bumetanide gel. The apparent molecular weight of the three components was 35, 40, and 78 kDa, respectively. Reconstitution experiments are needed to clarify +
VOLUME REGULATION IN CULTURED CELLS
171
Molecular weight I k D o J
FIG. 14. Scanning of gels from SDS-PAGE of membrane proteins from Ehrlich ascites tumor cells purified on a furosemide affinity gel. The cells were disintegrated by high-pressure decompression and the membranes purified by differential centrifugation. The cell membranes were solubilized in I 0 0 mM sodium citratei31 sodium choleate and were incubated with furosemide affinity resin. The hound membrane proteins were eluted with furosemide (3 mM), and finally analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the presence of dithiotreitol (3%) (a) and in its absence (h). (From Cherksey er al., 1987.)
the relation between these membrane polypeptides and (parts of) the cotransport system. 3. In recent experiments, an effort has been made to isolate K + and C1channel proteins from several cells and tissues by affinity chromatography using quinine or furosemide as ligand (Cherksey and Zeuthen, 1987; Cherksey er al., 1987). From apical cell membranes from bovine choroid plexus both quinine and furosemide affinity gels extracted a single, 300-kDa protein which seemed to have two channels: a K + channel, which was blocked by quinine, and a C1channel, which was blocked by furosemide (Cherksey and Zeuthen, 1987a,b). A similar protein was extracted from isolated membranes from Ehrlich cells (Cherksey et af.,1987). Under reducing conditions, the 300-kDa protein apparently breaks easily into 95- and 43-kDa subunits. This is illustrated in Fig. 14, which shows scanning of gels from SDS-polyacrylamide gel electrophoresis processed in the presence of a reducing agent (a) and in its absence (b). The 300kDa protein can be reformed by Cu2 /phenanthroline oxidation, suggesting that the subunits easily connect by disulfide bonds. +
The final goal of these studies is the isolation, reconstitution, and characterization of the molecular structures involved in the volume regulatory movements of ions across cell membranes. ACKNOWLEDGMENTS The author would like to thank L. 0. Simonsen for critical reading of the manuscript. Work in the author’s laboratory has been supported by the Danish National Science Research Council.
172
ELSE K. HOFFMANN REFERENCES
Aabin, B., and Hoffmann, E. K. (1986). Inhibitors of the volume- andCa2+-activated CI- transport pathway in Ehrlich ascites tumor cells. Acra Physiol. Scand. 128, P54. Aabin, B., and Kristensen, B. J. (1987). Identification of calmodulin in Ehrlich ascites tumor cells and its CaZ+-dependent binding to the cell membrane. Acra Physiol. Scand. (In press). Adragna, N., Canessa, M., Bize, I., Garay, R., and Tosteson, D. C. (1980). (Na+K) co-transport and cell volume in human red blood cells. Fed. Proc., Fed. Am. SOC. Exp. Biol. 39, 1842. Aiton, J. F., Chipperfield, A. R., Lamb, I. F., Ogden, P., and Simmons, N. L. (1981). Occurrence of p p i v e furosemide-sensitive transmembrane potassium transport in cultured cells. Biochim. Biophys. Acra 646,389-398. Aiton, J. F., Brown, C. D. A., Ogden, P., and Simmons, N. L. (1982). K + transport in “tight” epithelial monolayers of MDCK cells. J. Membr. Biol. 65, 99-109. Alper, S. L., Beam, K. G., and Greengard, P. (1980a). Hormonal control of Na+-K+ co-transport in turkey erythrocytes: Multiple site phosphorylation of goblin, a high molecular weight protein of the plasma membrane. J. Biol. Chem. 255, 4864-4871. Alper, S. L., Palfrey, H. C., DeRiemer, S. A,, and Greengard, P. (1980b). Hormonal control of protein phsophorylation in turkey erythrocytes: Phosphorylation by CAMP-dependent and Ca2+ -dependent protein kinases of distinct sites in goblin, a high molecular weight protein of the plasma membrane. J. Biol. Chem. 255, 11029-11039. Amsler, K., and Kinne, R. (1986). Photoinactivation of sodium-chloride cotransport in LLC-PKI/C14 cells by bumetanide. Am. J. Physiol. 250, C799-C807. Aronson, P. S . , Nee, J., and Suhm, M. A. (1982). Modifier role of internal Na+ in activating the Na+ /H exchanger in renal microvillus membrane vesicles. Narure (London) 299, 161- 163. Arslan, P., Virgilio, F. D., Beltrame, M., Tsien, R., and Pozzan, T. (1985). Cytosolic Ca2+ homeostasis in Ehrlich and Yoshida carcinomas. A new, membrane-permeant chelator of heavy metals reveals that these ascites tumor cell lines have normal cytosolic free Ca*+ . J. Biol. Chem. 260, 2719-2721. Aull, F. (1981). Potassium chloride cotransport in steady-state ascites tumor cells. Does bumetanide inhibit? Biochim. Biophys. Acia 643, 339-345. Bakker-Gmnwald. T. (1978). Effect of anions on potassium self-exchange in ascites tumor cells. Biochim. Biophys. Acra 513, 292-295. Bakker-Gmnwald, T. (198 1). Hormone-induced diuretic-sensitive potassium transport in turkey erythrocytes is anion dependent. Biochim. Biophys. Acra 641, 427-431. Bauer, J., and Lauf, P. K. (1983). Thiol-dependent passive K/CI transport in sheep red cells: 111. Differential reactivity of membrane SH groups with N-ethylmaleimide and iodoacetamide. J. Membr. Biol. 73, 257-261. Benjamin, M. A., and Dunham, P. B. (1983). Asymmetry of Na/K co-transport in human erythrocytes. J. Gen. Physiol. 82, 27a. Benos, D. J. (1982). Amiloride: A molecular probe of sodium transport in tissues and cells. Am. J. Physiol. 242, C131-CI45. Ben-Sasson, S., Shaviv, R., Bentwich, Z., Slavins, S., and Doljanski, F. (1975). Osmotic behaviour of normal and leukemic lymphocytes. Blood 46, 891-899. Bergenisich, T., and Smith, C. (1984). Multi-ion nature of potassium channels in squid axons. Curr. Top. Membr. Transp. 22, 353-369. Bui. A. H.. and Wiley, J. S. (1981). Cation fluxes and volume regulation by human lymphocytes. J . Cell. Physiol. 108, 47-54. Cabantchik, Z. I., Knauf, P. A., and Rothstein, A. (1978). The anion transport system of the red blood cell. The role of membrane protein evaluated by the use of “probes.” Biochim. Biophys. Acra 515, 239-302. +
VOLUME REGULATION IN CULTURED CELLS
173
Cala, P. M. (1980). Volume regulation by Amphiuma red blood cells. The membrane potential and its implications regarding the nature of the ion-flux pathways. J. Gen. Physiol. 76, 683-708. Cala. P. M. (1983a). Volume regulation by red blood cells: Mechanisms of ion transport. Mol. Ph.vsiol. 4, 33-52. Cala, P. M. (1983b). Cell Volume regulation by Amphiumo red blood cells. The role of Ca2+ as a modulator of alkali metal/H+ exchange. J. Gen. Phvsiol. 82, 761-784. Cala, P. M. (1985). Volume regulation by Amphiurna red blood cells: Characteristics of volumesensitive KIH and Na/H exchange. Mol. Phvsiol. 8, 199-214. Cala, P. M. ( I 986). Volume-sensitive alkali metal-H transport in Amphiuma red blood cells. Curr. Top. Membr. Tramp. 26, 79-99. Chase, H.. and Wong, S . (1985). Cell swelling increases intracellular calcium. a requirement for the increase in K + permeability which underlies volume regulation in toad bladder. Kidney Int. 27, 305. Cherksey, B. D., and Zeuthen, T. (1987). A membrane protein with a K + and a CI- channel. Acta Phvsiol. Srand. (In press). Cherksey, B. D., Hoffmann, E. K., lessen, F.. and Zeuthen, T. (1987). The K + and CI- channel protein from Ehrlich ascites turnout cells under reducing and oxidizing conditions. J . Physiol. (London) (In press). Cheung. R. K.,Grinstein, S., Dosch, H. H., and Gelfand, E. W. (1982). Volume regulation by human lymphocytes. Characterization of the ionic basis for regulatory volume decrease. J . Cell Ph?/siol. 112, 189-196. Chipperfield, A. R. (1986). The (Na*-K+-CI-) co-transport system. Clin. Sci. 71, 465-476. Cittadini, A,, Dani, A. M., Wolf, F., Bussi, D., and Calviello, G. (1982). Calcium permeability of Ehrlich ascites tumour cell plasma membrane in vivo. Biochim. Biophys. Acta 686, 27-35. Davis, C. W., and Finn, A. L. (1982). Sodium transport inhibition by amiloride reduces basolateral membrane potassium conductance in tight epithelia. Science 216, 525-527. Delpire, E., Gilles. R.. Duchene, C., and Goessens, G . (1985). Effects of osmotic shocks on the ion content and ultrastructure of rat pheochromocytoma cells of line PC 12. Mol. Physiol. 8, 293307. Dise, C. A , , Goodman, D. B. P., and Rasmussen, H. (1980). Selective stimulation of erythrocyte membrane phospholipid fatty acid turnover associated with decreased cell volume. J. Biol. Chem. 255, 5201-5207. Duhrn, J. (1981). Lithium transport pathways in erythrocytes. In "Basic Mechanisms in the Action of Lithium" (H. M. Emrich, J. B. Aldenhoff. and H. D. Lux, ed.). pp. 1-20. Excerpta Medica. Amsterdam. Dunham, P. B., and Ellory, J. C. (1981). Passive potassium transport in low potassium sheep red cells: Dependence upon cell volume and chloride. J. Physiol. (London) 318, 51 1-530. Ellory. I. C., and Dunham, P. B. (1980). Volume-dependent passive potassium transport in LK sheep red cells. Alfred Benzon S y n p . 14, 409-423. Ellory. J. C.. and Hall, A. C. (1985). The effects of high hydrostatic pressure on volume-sensitive cation transport in human red cells. J. Physiol. (London) 362, 15P. Ellory, J. C., Dunham, P. B., Logue, P. J., and Stewart, G. W. (1982). Anion-dependent cation transport in erythrocytes. Philos. Trans. R . Soc. London Ser. B 299, 483-495. Ellory, I . C., Hall, A. C.. and Stewart, G.W. (1985a). Volume-sensitive passive potassium fluxes in red cells. I n "Transport Processes, lono- and Osmoregulation" (R. Gilles and M . GillesBaillien, eds.), pp. 401-410. Springer Verlag, Berlin and New York. Ellory, J . C . , Hall, A. C., and Stewart, G. W. (1985b). Volume-sensitive cation fluxes in mammalian red cells. Mol. Physiol. 8, 235-246. Falke, J. J., Pace, R. J. and Chan, S . 1. (1984). Direct observation of the trans-membrane recruitment of band 3 transport sites by competitive inhibitors. A chlorine-35 NMR study. J. Biol. Chem. 259, 6481-6491.
174
ELSE K. HOFFMANN
Forbush, B., 111, and Palfrey, H. C. (1983). 3H-bumetanide binding to membranes isolated from dog kidney outer medulla. J . Biol. Chem. 258, 11787-1 1792. Foskett, J. K., and Spring, K. R. (1985). Involvement of calcium and cytoskeleton in gallbladder epithelial cell volume regulation. Am. J . Physiol. 248, C27-C36. Garay, R. (1982). Cation transport in essential hypertension. Lancer 1, 501-502. Gargus, J. I . , and Slayman, C. W. (1980). Mechanism and role of furosemide-sensitive K + transport in L cells: A genetic approach. J . Membr. Biol. 52, 245-256. Geck, P., and Pfeiffer, B. (1985). Na+ + K + + 2CI- cotransport in animal cells-Its role in volume regulation. Ann. N.Y. Acad. Sci. 456, 166-182. Geck, P., Pietrzyk, C., Burckhardt, B.-C., Pfeiffer, B., and Heinz, E. (1980). Electrically silent cotransport of N a + , K + and CI- in Ehrlich cells. Biochim. Biophys. Acra 600,432-447. Gilles, R. (1979). Intracellular organic osmotic effectors. In “Mechanisms of Osmoregulation in Animals. Maintenance of Cell Volume” (R. Gilles, ed.), pp. 111-154. Wiley, New York. Grinstein, S., Clarke, C. A,, DuPre, A,, and Rothstein, A. (1982a). Volume-induced increase of anion permeability in human lymphocytes. J . Gen. Physiol. 80, 801-823. Grinstein, S., Clarke, C. A., and Rothstein, A. (1982b). Increased anion permeability during volume regulation in human lymphocytes. Philos. Trans. R . SOC. London Ser. B 299, 509-518. Grinstein, S., DuPre, A., and Rothstein, A. (1982~).Volume regulation by human lymphocytes. Role of calcium. J . Gen. Physiol. 79, 849-868. Grinstein, S., Clarke, C. A , , and Rothstein, A. (1983a). Activation of N a + / H + exchange in lymphocytes by osmotically induced volume changes and by cytoplasmic acidification, J. Gen. Physiol. 82, 619-638. Grinstein, S . , Clarke, C. A., Rothstein, A., and Gelfand, E. W. (1983b). Volume induced anion conductance in human B lymphocytes is cation independent. Am. J . Physiol. 245, C160-CI63. Grinstein, S . , Cohen, S., and Rothstein, A. (1984a). Cytoplasmic pH regulation in the thymic lymphocytes by an amiloride-sensitive NalH antiport. J . Gen. Physiol. 83, 341-369. Grinstein, S., Goetz, J. D.. and Rothstein, A. (1984b). Na fluxes in thymic lymphocytes 11. Amiloride sensitive Na/H exchange pathway: Reversibility of transport and asymmetry of the modifier site. J . Gen. Physiol. 84, 585-600. Grinstein, S., Rothstein, A., Sarkadi, B., and Gelfand. E. W. (1984~).Responses of lymphocytes to anisotonic media: Volume-regulating behavior. Am. J. Physiol. 246, C204-C2 15. Grinstein, S., Goetz, J. D.. Cohen, S., Furuya, W., Rothstein, A , , and Gelfand, E. W. (1985a). Mechanism of regulatory volume increase in osmotically shrunken lymphocytes. Mol. Physiol. 8, 185-198. Grinstein, S . , Goetz, J. D., Cohen, S., Rothstein, A,, and Gelfond, E. W. (1985b). Regulation of N a + / H + exchange in lymphocytes. Ann. N.Y. Acad. Sci. 456, 207-219. Grinstein, S . . Cohen, S., Goetz, J. D., Rothstein, A., Mellors, A., and Gelfand, E. W. (1986). Activation of the Na+ /H antiport by changes in cell volume and by phorbol esters; Passible role of protein kinase. Curr. Top. Membr. Tramp. 26, 115-134. Grygorczyk, R., Schwartz, W., and Passow, H. (1984). Ca2+-activated K + channels in human red cells. Comparison of single-channel currents with ion fluxes. Biophys. J. 45, 693-698. Haas, M., and McManus, T. J. (1982). Bumetanide inhibition of (Na+K+2C1) co-transport and KlRb exchange at a chloride site in duck red cells: Modulation by external cations. Biophys. J. 37, 214a. Haas, M., and McManus, T. J. (1983). Bumetanide inhibits (Na+K+2CI) co-transport at a chloride site. Am. J . Physiol. 245, C235-C240. Haas, M., Schmidt, W. F., 111, and McManus, T. J. (1982). Catecholamine-stimulated ion transport in duck red cells. Gradient effects in electrically neutral (Na+K+2C1) co-transport. J . Gen. Physiol. 80, 125-147. Hall, A. C., and Ellory, J. C. (1986). Effects of high hydrostatic pressure on passive monovalent cation transport in human red cells. J . Membr. Biol. 94, 1-17. +
175
VOLUME REGULATION IN CULTURED CELLS
Hall, A. C.. Ellory. J. C., and Klein. R. A. (1982). Pressure and temperature effects on human red cell cation transport. J. Membr. Biol. 68, 47-56. Hannafin. J . , Kinne-Saffran. E., Friedman, D., and Kinne, R. (1983). Presence of a sodiumpotassium chloride cotransport system in the rectal gland of Sqwlus ucunrhias. J. Membr. Biol. 75, 73-84. Heinz, E., Geck, P., and Pietrzyk, C. (1975). Driving forces of amino acid transport in animal cells. Ann. N.Y. Acad. Sci. 264, 428-441. Hempling, H. G . (1967). Application of irreversible thermodynamics to a functional description of the tumor cell membrane. J. Cell Physiol. 70, 237-256. Hempling, H. G . , Thomas, S., and Dupre, A. (1977). Osmotic properties of human lymphocytes. J . Cell. Physiol. 93, 293-302. Hendil, K. B., and Hoffmann, E. K. (1974). Cell volume regulation in Ehrlich ascites tumor cells. J. Cell. Physiol. 84, 115-125. Hodgkin, A. L . , and Katz, B. (1949). The effect of sodium ions on the electrical activity of the giant axon of the squid. J. Physiol. (London) 108, 37-77. Hodgkin. A. L., and Keynes, R. D. (1955). The potassium permeability of a giant nerve fibre. J . Physiol. (London) 128, 61-88. Hoffmann. E. K. (1977). Control of cell volume. I n “Transport of Ions and Water in Animals” (B. J . Gupta, R. B. Moreton, J . L. Oschman, and B. J. Wall. eds.), pp. 285-332. Academic Press, London Hoffmann, E. K. (1978). Regulation of cell volume by selective changes in the leak permeabilities of Ehrlich ascites tumor cells. Alfred Benzon Symp. 11, 397-417. Hoffmann, E. K. (1980).Cell volume regulation in mammalian cells. I n “Animals and Environmental Fitness” (R. Gilles, ed.), pp. 43-59. Pergamon, Oxford. Hoffmann, E. K . (1982). Anion exchange and anion-cation co-transport systems in mammalian cells. Philos. Trans. R. Soc. London Ser. B 299, 519-535. Hoffmann, E. K. (1983). Volume regulation by animal cells. SOC.Exp. Biol. Semin. Ser. 18. 55-80. Hoffmann, E. K. (1985a). Cell volume control and ion transport in a mammalian cell. I n “Comparative Physiology and Biochemistry Current Topics and Trends. B. Transport ProcessesIon0 and Osmoregulation” (R. Gilles, and M .Gilles-Baillien, eds.), pp. 389-401. SpringerVerlag, Berlin and New York. Hoffmann, E. K. (1985b). Role of separate K and CI - channels and of Na+ ICI - cotransport in volume regulation in Ehrlich cells Fed. Proc.. Fed. Am. SOC. Exp. Biol. 44, 2513-2519. Hoffmann, E. K. (1985~).Regulatory volume decrease in Ehrlich ascites tumor cells: Role of inorganic ions and amino compounds. Mol. Physiol. 8, 167-185. Hoffmann, E. K., and Hendil, K. B. (1976). The role of amino acids and taurine in isosmotic intracellular regulation in Ehrlich &$citesmouse tumour cells. J . Comp. Physiol. 108, 279-286. Hoffmann. E. K . , and Lambert. I. H. (1983). Amino acid transport and cell volume regulation in Ehrlich ascites tumour cells. J . Physiol. (London) 338, 613-625. Hoffmann, E. K., Simonsen, L. O., and Sjsholm, C. (1979). Membrane potential, chloride exchange, and chloride conductance in Ehrlich mouse ascites tumour cells. J. Physiol. (London) 296, 61-84. Hoffmann, E. K., Sjsholm. C., and Simonsen. L. 0. (1983). N a + , CI- co-transport in Ehrlich ascites tumor cells activated during volume regulation (regulatory volume increase). J. Membr. Biol. 76, 269-280. Hoffmann, E. K., Simonsen, L. O., andLambert, I. H. (1984). Volume-induced increaseof K + and CI - permeabilities in Ehrlich ascites tumor cells. Role of internal Ca2 . J. Membr. Biol. 78, 21 1-222. Hoffmann, E. K.. Schipdt, M., and Dunham, P. B. (1986a). The number of chloride-cation cotransport sites on Ehrlich ascites cells measured with 3H-bumetanide. Am. J. Physiol. 250, C688-C693. +
+
176
ELSE K. HOFFMANN
Hoffmann, E. K., Lambert, I. H., and Simonsen, L. 0. (1986b). CaZ+-activated, separate K + and CI- transport pathways in Ehrlich ascites tumor cells. J. Membr. Biol. 91, 227-244. Horowicz, P., Gage, P. W., and Eisenberg, R. S. (1968). The role of the electrochemical gradient in determining potassium fluxes in frog striated muscle. J . Gen. Physiol. 51, 193s-203s. Hunter, M. J. (1977). Human erythrocyte anion permeabilities measured under conditions of net charge transfer. J . Physiol. (London) 268, 35-49. Jessen, F., Sjuiholm, C., and Hoffmann, E. K. (1986). Identification of the anion exchange protein of Ehrlich cells: A kinetic analysis of the inhibitory effects of 4,4‘-diisothiocyano-2,2’-stilbenedisulphonic acid (DIDS) and labeling of membrane proteins with 3H-DIDS. J. Membr. Biol. 92, 195-205. Johnson, J. H., Dunn, D. P., and Rosenberg, R. N. (1982). Furosemide-sensitive K + channel in glioma cells but not neuroblastoma cells in culture. Biochem. Biophys. Res. Commun. 109, 100-105.
Jergensen, P. L., Petersen, I . , and Rees, W. D. (1984). Identification of a N a + , K + , CI-cotransport protein of M, 34,000 from kidney by photolabeling with [3H]bumetanide. Biochim. Biophys. Acru 775, 105- 110. Kimelberg, H. K. (1981). Active accumulation and exchange transport of chloride in astroglial cells in culture. Biochim. Biophys. Acra 646, 179-184. Kimelberg, H. K., and Bourke, R. S. (1982). Anion transport in the nervous system. I n “Handbook of Neurochemistry” (A. Lajtha, ed.), 2nd Ed., Vol. 1, pp. 31-67. Plenum, New York. Kinne, R., Hannafin, J., Koenig, B., and Kinne-Saffran, E. (1984). Mode of action of loop diuretics in chloride transporting epithelia. I n “Diuretics” (J. B. Puschett, ed.), pp. 447-449. Elsevier, New York. Kinne, R., Koenig, B., Hannafin, J., Kinne-Saffran, E., Scott, D. M., and Zierold, K. (1985). The use of membrane vesicles to study the sodium chloride-potassium chloride cotransporter involved in active transepithelial chloride transport. P’ugers Arch. 405, SlOl-SIO5. Kinsella, J. L.. and Aronson. P. S. (1981). Amiloride inhibition of the Na+-H+ exchange in renal microvillus membrane vesicles. Am. J. Physiol. 241, 7374-7379. Kinsella, J. L. and Aronson, P. S. (1982). Determination of the coupling ratio for Na+-H+ exchange in renal microvillus membrane vesicles. Biochim. Biophys. Acra 689, 161- 164. Klaven, N. B., Pershadsingh, H. A., Henius, G. V., Laris, P. C., Long, J. W., Jr., and McDonald, J. M. (1983). A high-affinity, calmodulin-sensitive (Ca2+ + Mg* +)-ATPase and associated calcium-transport pump in the Ehrlich ascites tumor cell plasma membrane. Arch. Biochem. Biophys. 226, 618-628. Knauf, P. A. (1979). Erythrocyte anion exchange and the band 3 protein: Transport kinetics and molecular structure. Curr. Top. Membr. Transp. 12, 249-263. Kramheft, B., Lambert, I. H., Hoffmann, E. K., and Juirgensen, F. (1986). Activation of K, CI cotransport in Ehrlich ascites tumor cells. Am. J. Physiol. 251, C369-C379. Kregenow, F. M. (19’71a). The response of duck erythrocytes to nonhemolytic hypotonic media. Evidence for a volume-controlling mechanism. J. Gen. Physiol. 58, 372-395. Kregenow, F. M. (1971b). The response of duck erythrocytes to hypertonic media. Further evidence for a volume-controlling mechanism. J . Gen. Physiol. 58, 396-412. Kregenow, F. M. (1973). The response of duck erythrocytes to norepinephrine and an elevated extracellular potassium: Volume regulation in isotonic media. J . Gen. Physiol. 61, 509-527. Kregenow, F. M. (1974). Functional separation of the Na-K exchange pump from the volume controlling mechanism in enlarged duck red cells. J. Gen. Physiol. 64, 393-412. Kregenow, F. M. (1981). Osmoregulatory salt transporting mechanisms: Control of cell volume in anisotonic media. Annu. Rev. Physiol. 43, 493-505. Kregenow, F. M., and Caryk, T. (1979). Co-transport of cations and C1 during the volume regulatory responses of duck erythrocytes. Physiologist 22, 73. Kregenow, F. M., Robbie, D. E., and Orloff, J. (1976). Effect of norepinephrine and hypertonicity of K influx and cyclic AMP in duck erythrocytes. Am. J. Physiol. 231, 306-312.
VOLUME REGULATION IN CULTURED CELLS
177
Kregenow, F. M., Caryk. T., and Siebens, A. W. (1985). Further studies of the volume-regulatory response of Amphiuma red cells in hypertonic media. Evidence for amiloride-sensitive Na/H exchange. J. Gen. Physiol. 86, 565-584. Kristensen. L. 0., and Folke, M. (1984). Volume-regulatory K + efflux during concentrative uptake of alanine in isolated rat hepatocytes. Biochem. J. 221, 265-268. Kirstensen, P.. and Ussing, H. H. (1985). Epithelial organization. In “The Kidney: Physiology and Pathophysiology” (D.W. Seldin and G. Giebisch, eds.), pp. 173-188. Raven, New York. Lackington, I . , and Orrego, F. (1981). Inhibition of calcium-activated potassium conductance of human erythrocytes by calmodulin inhibitory drugs. FEES Lett. 133, 103-106. Lambert, I. H. (1984). Na+-dependent taurine uptake in Ehrlich ascites tumour cells. Mol. Physiol. 6, 233-246. Lambert, 1. H . (1985). Taurine transport in Ehrlich ascites tumour cells. Specificity and chloride dependence. Mol. Physiol. 7, 323-332. Lambert, I. H., and Hoffmann, E. K. (1982). Amino acid metabolism and protein turnover under different osmotic conditions in Ehrlich ascites tumor cells. Mol. Physiol. 2, 274-286. Lambert, I. H . , Hoffmann, E. K., and Christensen, P. (1987). Role of prostaglandins and leukotrienes in volume regulation by Ehrlich ascites tumor cells. J. Membr. Biol. (In press). Larson, M., and Spring, K. R. (1984). Volume regulation by Necrurus gallbladder: Basolateral KCI exit. J . Membr. Biol. 81, 219-232; Letter to the Editor, J. Membr. Biol. 84, 191 ( 1985). Latorre, R., and Miller, C. (1983). Conduction and selectivity in potassium channels. J. Membr. Biol. 71, 11-30. Lau, K. R., Hudson, R. L., and Schultz, S. G. (1984). Cell swelling increases a barium-inhibitable potassium conductance in the basolateral membrane of Necrurus small intestine. Proc. Natl. Acad. Sci. U.S.A. 81, 3591-3594. Lauf, P. K. (1982). Evidence for chloride dependent potassium and water transport induced by hyposmotic stress in erythrocytes of the marine teleost, Opsanus tau. J. Comp. Physiol 146, 916. Lauf, P. K. (1983a). Thiol-dependent passive K/CI transport in sheep red cells: I . Dependence on chloride and external K + ( R b + ) ions. J. Membr. Biol. 73, 237-246. Lauf, P. K. (l983b). Thiol-dependent passive K/CI transport in sheep red cells: 11. Loss of C1- and N-ethylmaleimide sensitivity in maturing high K + cells. J. Membr. Biol. 73, 247-256. Lauf, P. K. (1983~).Thiol-dependent passive K -CI - transport in sheep red cells: V. Dependence on metabolism. Am. J . Physiol. 245, C445-C448. Lauf, P. K. (1984). Thiol-dependent passive K+-CI- transport in sheep red cells: IV. Furosemide inhibition and the role of external R b + , Na+ and CI-. J . Membr. Biol. 77, 57-62. Lauf. P. K. (1985a). On the relationship between volume- and thiol-stimulated K + , C1- fluxes in red cell membranes. Mol. Physiol. 8, 215-234. Lauf, P. K. (1985b). K+:CI- cotransport: Sulfhydryl, divalent cations, and the mechanism of volume activation in a red cell. J. Membr. Biol. 88, 1-13. Lauf, P. K., and Mangor-Jensen, A. (1984). Effects of A23187 and Ca2+ on volume- and thiolstimulated, ouabain-resistant K +Cl - fluxes in low K sheep erythrocytes. Biochem. Biophys. Res. Commun. 125, 790-797. Lauf, P. K., and Theg, B. E. (1980). A chloride dependent K + flux induced by N-ethylmaleimide in genetically low K + sheep and goat erythrocytes. Biochem. Biophys. Res. Commun. 92, 14221428 Lauf, P. K., Garay. R., and Adragna, N . C. (1982). N-Ethylmaleimide stimulates chloride-dependent K + but not Na+ fluxes in human red cells. J. Gen. Physiol. 80, 19a. Lauf. P. K., Adragna, N. C., and Garay, R. P. (1984). Activation by N-ethylmaleimide of a latent K. CI flux in human red blood cells. Am. J. Physiol. (London) 246, C385-C390. Leaf, A. (1959). Maintenance of concentration gradients and regulation of cell volume. Ann. N . Y . Acad. Sci. 72. 396-404. +
178
ELSE K. HOFFMANN
Levinson, C. (1985). Sodium-dependent ion cotransport in steady-state Ehrlich ascites tumor cells. J . Membr. Biol. 87, 121-130. Lew, V. L., and Ferreira, H. G. (1978). Calcium transport and the properties of a calcium-activated potassium channel in red cell membranes. Curr. Top. Membr. Tramp. 10, 217-277. Lew, V. L., Muallem, S., and Seymour, C. A. (1982). Properties of theCa2+-activated K + channel in one-step inside-out vesicles from human red cell membranes. Nature (London) 296, 742-
744. Logue, P., Anderson, C . , Kanik, C., Farquharson, B., and Dunham, P. (1983). Passive potassium transport in LK sheep red cells: Modification by N-ethylmaleimide. J. Gen. Physiol. 81, 861885. MacKnight, A. D. C. (1983). Cellular response to injury. I n “Edema: Basic Science and Clinical Manifestations” (N. C. Staub and A. E. Taylor, eds.), pp. 489-520. Raven, New York. McManus, T. J. (1982). Catecholamine-stimulated and volume-sensitive ion movements in avian red cells: Alternate modes of chloride-dependent cation transport. Jacobs-Parpart-Ponder Mem. Leer., Red Cell Club Apr. 19. McManus, T. J., and Schmidt, W. F., 111 (1978). Ion and co-ion transport in avian red cells. I n “Membrane Transport Processes” (J. F. Hoffman, ed.), Vol. 1, pp. 79-106. Raven, New York. McManus, T. J., Haas, M., Starke, L. C., and Lytle, C. Y. (1985). The duck red cell model of volume-sensitive chloride-dependent cation transport. Ann. N . Y . Acad. Sci. 456, 183- 186. MacRobbie, E. A. C., and Ussing, H. H. (1961). Osmotic behaviour of the epithelial cells of frog skin. Acra Physiol. Scand. 53, 348-365. McRoberts, J. A., Erlinger, S., Rindler, M. J., and Saier, M.H.,Jr. (1982). Furosemide-sensitive salt transport in the Madin-Darby canine kidney cell line: Evidence for the transport of Na , K + , and CI-. J. Biol. Chem. 257, 2260-2266. Mills, J. W., and Skiest, D. J. (1985). Role of cyclic AMP and the cytoskeleton in volume control in MDCK cells. Mol. Physiol. 8, 247-262. Nauntofte, B., and Poulsen, J. H. (1984). Chloride transport in rat parotid acini: Furosemidesensitive uptake and calcium-dependent release. J . Physiol. (London) 357, 61P. Nauntofte, B., and Poulsen, J. H. (1986). Effects of Ca2+ and furosemide on C1- transport and O2 uptake in rat parotid acini. Am. J . Physiol. 251, C175-Cl85. O’Brien, T. G . , and Krzeminski, K. (1983). Phorbol ester inhibits furosemide-sensitive potassium transport in BALB/c 3T3 preadipose cells. Proc. Nail. Acad. Sci. U.S.A. 80, 4334-4338. Brskov, S. L. (1954). The potassium absorption by pigeon blood cells. A considerable potassium absorption by pigeon and hen blood cells is observed when a hypertonic sodium chloride solution is added. Acta Physiol. Scand. 31, 221-229. Palfrey, H. C. (1983). Na/K/CI cotransport in avian red cells. Reversible inhibition by ATP depletion. J. Gen. Physiol. 82, 10a. Palfrey, H. C., and Rao, M. C. (1983). Na/K/CI co-transport and its regulation. J. Exp. Biol. 106, 43-54. Palfrey, H. C., Feit, P. W., and Greengard, P. (1980). CAMP-stimulated cation cotransport in avian erythrocytes: Inhibition by “loop” diuretics. Am. J . Physiol. (London) 238, C139-CI48. Pape, L., and Kristensen, B. I. (1984). A calmodulin activated CaZ+-dependent K + channel in human erythrocyte membrane inside-out vesicles. Biochim. Biophys. Acra 770, 1-6. Parker, J. C. (1981). Heterogeneity among dog red blood cells. J. Gen. Physiol. 78, 141-150. Parker, J. C. (1983a). Hemolytic action of potassium salts on dog red blood cells. Am. J. Physiol. 244, C313-C317. Parker, J . C. (1983b). Passive calcium movements in dog red blood cells: Anion effects. Am. J. Physiol. 244, C318-C323. Parker, I. C. (1983~).Volume-responsive sodium movements in dog red cells. Am. J . Physiol. 244, C324-C330. +
179
VOLUME REGULATION IN CULTURED CELLS
Parker, J . C. (39834). Further characterization of sodium-proton countertransport in dog red blood cells. J. Gen. Physiol. 82, lla-12a. Parker, J. C. (1984). Glutaraldehyde fixation of sodium transport in dog red blood cells. J. Gen. Phvsiol. 84, 789-803. Parker, J. C. (1986). Na-proton exchange in dog red blood cells. Curr. Top. Membr. Transp. 26, 101-1 15. Parker, I . C., and Hoffman, J. F. (1965). Interdependence of cation permeability, cell volume, and metabolism in dog red blood cells. Fed. Proc., Fed. Am. Soc. Exp. B i d . 24,589. Parker, J . C., Castvanova, V., and Goldinger, J. M. (1977). Dog red blood cells: Na and K diffusion potentials with extracellular ATP. J. Gen. Physiol. 69, 417-430. Plishker, G. A. (1984). Phenothiazine inhibition of calmodulin stimulates calcium-dependent potassium efflux in human red blood cells. Cell Calcium 5, 177-185. Riddick, D. H., Kregenow, F. M., and Orloff, J. (1971). The effect of norepinephrine and dibutyryl cyclic-AMP on cation transport in duck erythrocytes. J. Gen. Physiol. 57, 752-766. Rindler, M. J., Taub, M., and Saier, M. H., Jr. (1979). Uptake of 22Na+ by cultured dog kidney cells (MDCK). J. Eiol. Chem. 254, 11431-1 1439. Rindler, M. J., McRoberts, J. A , , and Saier, M. H., Jr. (1982). ( N a + , K+)-cotransport in the Madin-Darby canine kidney cell line: Kinetic characterization of the interaction between Na and K . J. Eiol. Chem. 257, 2254-2259. Rink, T. J., Sanchez, A., Grinstein, S . , and Rothstein, A. (1983). Volume restoration in osmotically swollen lymphocytes does not involve changes in free CaZ concentration. Eiochim. Eiophys. Acra 762, 593-596. Rorive. G., and Gilles, R. (1979). Intracellular inorganic osmotic effectors. In “Mechanisms of Osmoregulation in Animals” (R. Gilles, ed.), pp. 83-109. Wiley, New York. Roti-Roti, W . L., and Rothstein, A. (1973). Adaptation of mouse leukemic cells (L5178Y) to anisotonic media. Exp. Cell Res. 79, 295-310. Rugg, E. L., Simmons, N. L., and Tivey, D. R. (1986). Specificity of labelling Na + K + C1 cotransport by [3H]bumetanide. J. Exp. Phvsiol. 71, 165-182. Sarkadi, B., Mack, E., and Rothstein, A. (1984a). Ionic events during the volume response of human peripheral blood lymphocytes to hypotonic media. I. Distinctions between volumeactivated CI- and K + conductance pathways. J. Gen. Physiol. 83, 497-512. Sarkadi. B.. Mack, E., and Rothstein, A. (1984b). Ionic events during the volume response of human peripheral blood lymphocytes to hypotonic media. 11. Volume- and time-dependent activation and inactivation of ion transport pathways. J. Gen. Physiol. 83, 513-527. Sarkadi, B.,Cheung, R., Mack, E., Grinstein, S . . Gelfand, E. W . , and Rothstein, A. (1985). Cation and anion transport pathways in volume regulatory response of human lymphocytes to hypoosmotic media. Am. J. Physiol. 248, C480-C487. Scharff, 0 . .Foder, B., and Skibsted, U. (1983). Hysteretic activation of the Ca2+ pump revealed by calcium transients in human red cells. Eiochim. Biophvs. Acra 730, 295-305. Schmidt, W. F., 111, and McManus, T. J. (1977a). Ouabain-insensitive salt and water movements in duck red cells. 1. Kinetics of cation transport under hypertonic conditions. J. Gen. Physiol. 70, 59-79. Schmidt, W. F . . 111, and McManus. T. J. (1977b). Ouabain-insensitive salt and water movements in duck red cells. 11. Norepinephrine-stimulation of sodium plus potassium cotransport. J. Gen. P hvsiol. 70, 8 1-97, . salt and water movements in Schmidt. W. F., 111, and McManus, T. J . ( 1 9 7 7 ~ )Ouabain-insensitive duck red cells. 111. The role of chloride in the volume response. J. Gen. Physiol. 70, 99-121. Schultz, S. G . , Hudson, R. L., and Lapointe, J.-Y. (1985). Electrophysiological studies of sodium cotransport in epithelia: Towards a cellular model. Ann. N.Y. Acad. Sci. 456, 127-135. Schwartz, W.. and Passow, H. (1983). CaZ+-activated K channels in erythrocytes and excitable cells. Annu. Rev. Physiol. 45, 359-374. +
+
+
+
180
ELSE K. HOFFMANN
Siebens, A. W. (1985). Cellular volume control. In “The Kidney: Physiology and Pathophysiology” (D. W. Seldin and G.Giebisch, eds.), pp. 91-115. Raven, New York. Siebens, A. W., and Kregenow, F. M. (1980). Analysis of amiloride-sensitive volume regulation in Amphiuma red cells. Fed. Proc., Fed. Am. SOC. Exp. Biol. 39, 379. Siebens, A. W., and Kregenow, F. M. (1985). Volume-regulatory responses of Amphiuma red cells in anisotonic media. The effect of amiloride. J. Gen. Physiol. 86, 527-564. Simonsen, L. O., Hoffmann, E. K.,and Sjsholm, C. (1976). Chloride conductance of the Ehrlich ascites tumor cell membrane. FEES Symp. Biochen. Membr. Tramp. p. 225. Sjsholm, C.. Hoffmann, E. K., and Simonsen, L. 0. (1981). Anion-cation co-transport and anion exchange in Ehrlich ascites tumour cells. Acfa Physiol. Scand. 112, 24A. Smalley, C. E., Tucker, E. M., Dunham, P. B., and Ellory, J. C. (1982). Interaction of L antibody with low potassium type sheep red cells: Resolution of two separate functional antibodies. J. Membr. Biol. 64, 167-174. Spring, K. R . , and Ericson A,-C. (1982). Epithelial cell volume modulation and regulation. J . Membr. Biol. 69, 167-176. Sten-Knudsen, 0. (1978). Passive transport processes. In “Membrane Transport in Biology. Vol. 1. Concepts and Models” (G.Giebisch, D. C. Tosteson, and H. H. Ussing, eds.), pp. 5-113. Springer-Verlag. Berlin and New York. Thornhill, W. B., and Laris, P. C. (1984). KCI loss and cell shrinkage in the Ehrlich ascites tumor cell induced by hypotonic media, 2-deoxyglucose and propranolol. Biochim. Biophys. Acra 773, 207-2 18. Tosteson, D. C., and Hoffman, J. F. (1960). Regulation of cell volume by active cation transport in high and low potassium sheep red cells. J. Gen. Physiol. 44, 169-194. Tosteson, D. C., Gunn, R. B., and Wieth, I. 0. (1973). In “Erythrocytes, Thrombocytes, Leucocytes” (E. Gerlach, K. Mosev, E. Deutsch, and W. Wilmanns, eds.), pp. 62-66. Theme, Stuttgart. Ussing. H. H. (1949). The distinction by means of tracers between active transport and diffusion. Acfa Physiol. Scand. 19, 43-56. Ussing, H. H. (1960). Active and passive transport of the alkali metal ions. In “The Alkali Metal Ions in Biology” (H. H. Ussing, P. Kruhsffer, J. Hess Thaysen, and N. A. Thorn, eds.), pp. 45- 143. Springer-VerIag, Berlin. Ussing, H. H. (1982). Volume regulation of frog skin epithelium. Acra Physiol. Scand. 114, 363369. Ussing, H. H. (1985). Volume regulation and basolateral cotransport of sodium, potassium, and chloride ions in frog skin epithelium. Pfliigers Arch. 405, S2-S7. Ussing, H. H. (1986). Epithelial cell volume regulation illustrated by experiments in frog skin. Renal Physiol. (Easel) 9, 38-46. Valdeolmillos, M., Garcia-Sancho, J., and Herreros, B . (1982). Ca2+-dependent K + transport in the Ehrlich ascites tumor cell. Biochim. Biophys. Acfa 685, 273-278. Vestergaard-Bogind, B., Stampe, P., and Christophersen, P. (1985). Single-file diffusion through the Ca2+-activated K%‘-channelof human red cells. J. Membr. Biol. 88, 67-75. Weiss, B., Prozialeck, W., Cimino, M., Bamette, M. S . , and Wallace, T. L. (1980). Pharmacological regulation of calmodulin. Ann. N.Y.Acad. Sci. 356, 319-345. Wiater, L. A., and Dunham, P. B. (1983). Passive transport of potassium and sodium in human erythrocytes: Effects of sulfhydryl binding agents and furosemide. Am. J. Physiol. 245, C348C356. Yingst, D. R., and Hoffman, I. F. (1984). Ca-induced K transport in human red blood cell ghosts containing menazo 111. Transmembrane interactions of Na, K, and Ca and the relationship to the functioning Na-K pump. J . Gen. Physiol 83, 19-45.
CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 30
Cell Volume Regulation in Lower Vertebrates LEON G O D S T E I N Division of Biology and Medicine Brown University Providence, Rhode Island 02912 AND
ARNOST KLEINZELLER Department of Physiology University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104
1.
EVOLUTIONARY CONSIDERATIONS
The ancestors of modem-day fishes lived in a marine environment throughout most of their history, but at the end probably inhabited brackish areas. Most of the modem-day fishes (e.g., teleosts, elasmobranchs) live in a marine environment, but they probably had a period of freshwater in their history (Lockwood, 1966). The only modem fish that seems to have had no freshwater period in its existence is the jawless (agnatha) hagfish, Myxine. Thus, most of the present-day fishes (marine and freshwater) had a period of existence in their evolutionary history that required them to adapt to an environment that placed a severe osmotic stress on them and necessitated the development of an osmoregulatory mechanism that would enable them to cope with this stress (Lockwood, 1966). These mechanisms had to deal with both the osmotic relationships between the internal and external environment of the fish and between the intracellular and extracellular fluids (cell volume regulation). Although the numbers of species involved are not as great as with the fishes, there are examples of other lower vertebrates (e.g., amphibia) that migrated from a semiterrestrial or terrestrial existence to a totally aquatic environment during their evolution. The early amphibia, which are thought to have evolved from 181
Copyrighi 8 1987 by Academic &ss. lnc. All rights uf reproduction in any fmI C S ~ N
~ ~ .
182
LEON GOLDSTEIN AND ARNOST KLEINZELLER
freshwater fishes, had to cope with a relatively desiccating environment. As with the fishes they were faced with the dual problems of maintaining the volume and composition of both the extracellular and cellular fluids constant. Some amphibia (e.g., Xenopus laevis) returned to a total freshwater existence, and a few invaded brackish environments. Similarly, while most reptiles live on land several have become semiaquatic and even totally marine. Again, the osmotic problems associated with migration from a terrestrial to semiaquatic or marine habitat had to be accompanied (or more likely preceded) by physiological adaptations that would allow the cells to cope with the new osmotic environment.
II. ENVIRONMENTAL CONSIDERATIONS The osmotic challenges that lower vertebrates faced during evolution are recapitulated in the periodic excursions that some of today’s lower vertebrates make into environments of different salinities and/or water availabilities. Euryhaline fishes face osmotic challenges as they go from marine to brackish to freshwater and vice versa. Euryhaline fishes are able to live in both a freshwater and marine environment. Although these fishes have mechanisms for regulating the osmotic pressure of their extracellular fluids, these mechanisms are not perfect and the extracellular fluid osmolality does change, more or less, with environmental salinity. This means that the cells of these fishes must possess mechanisms for regulating their volumes in the face of changes in extracellular fluid osmolalities. Euryhaline fishes are not the only ones that have to cope with changes in extracellular fluid osmolality. For example, there are fishes that leave the water periodically (e.g., lungfishes; Smith, 1930) and have to cope with the desiccating effects of a terrestrial environment as well as the increase in extracellular fluid osmolality caused by accumulation of metabolic products that are concentrated in the extracellular fluids (e.g., sulfate). The cells of these fishes must have mechanisms to prevent or decrease a reduction in cell volume caused by a hypersomotic extracellular environment. There are a few amphibia that venture into brackish water. They are the crabeating frog (Rana cancrivora) of Southeast Asia (Gordon et al., 1961) and some toads (Bufo marinus and Rufo viridis) (Tercafs and Schoffeniels, 1962; Weissberg and Katz, 1975) that inhabit areas near brackish water. In contrast to euryhaline fishes these arnphibia are limited in their ability to regulate the osmolality of their extracellular fluid which rises significantly when the animals are in a hyperosmotic environment. Thus, the cells of these amphibia face severe osmotic challenges and must possess mechanisms for coping with significant rises in extracellular fluid osmolality to avoid shrinking. Semiaquatic amphibia (e.g., frogs) leave the water for short periods of time
CELL VOLUME REGULATION IN LOWER VERTEBRATES
183
and the terrestrial amphibia (e.g., toads) return to water to breed. The terrestrial amphibia have thickened skins which prevent water loss while on land and minimize water uptake when they return to an aquatic environment. Nevertheless, in both types of amphibia, some water loss or uptake must occur in the extracellular fluid, forcing the cells to cope with changes in the osmolality of the extracellular fluid. Although there are reptiles that are euryhaline they have thick integuments that must minimize both the flux of water across their body surface as well as changes in extracellular fluid osmolality . Still, some extracellular changes may take place with which the cells have to cope. The thick integument also minimizes evaporative water loss that takes place when semiaquatic reptiles leave water for short periods. However, little work has been done on the changes in extracellular fluid osmolality under these conditions.
111.
CELL SOLUTE COMPOSITION
Since most cell membranes are highly permeable to water, cell volume ultimately depends on total solute concentrations inside and outside of cells. The major solutes in the extracellular fluids of vertebrates are Na+ , C1- , HCO, - , and in lower vertebrates urea and trimethylamine oxide (TMAO). Inside cells the predominant solutes are K , Na , organic anions (e.g., organic phosphates), and in lower vertebrates urea, amino acids, and quarternary ammonium compounds such as TMAO and betaine (Fig. 1). Some tissues of higher vertebrates, +
-.-i
+
r""l Na.K.CI
1000
0UREA
\
= 0
AMINO ACIDS TMAO UNDETERMINED
900
E UJ
000
Z
700
2
600
0
E w
u z
s p
500
FLOUNDER
400
300
200
3
s:
J
100 n "
SW W 2
SW FW
SW/Z F W
FIG. 1. Major cellular solutes in muscles from three representative lower vertebrates. Data for skate (Raju erimrea) from Forster and Goldstein ( 1 976), for flounder (Pleuronems flesus) from Lange and Fugelli (1965), and for toad (Bufo viridis) from Gordon (1965).
184
LEON GOLDSTEIN AND ARNOST KLEINZELLER
such as the renal medulla, also contain high concentrations of urea and trimethylamines. An inverse relationship is found, in some lower vertebrate species at least, between the concentrations of TMAO and free amino acids. In both hagfish (Myxine glufinosa) (Cholette and Gagnon, 1973) and skates (Raja erinacea) (Forster and Goldstein, 1976) those tissues that have high concentrations of TMAO have low concentrations of free amino acids and vice versa. This suggests that the two types of compounds are interchangeable from an osmotic point of view, but it is unclear at present what determines which compounds will accumulate in a particular tissue and how the differences are brought about. It is possible that both types of compounds compete for the same transport sites in the cell membrane and that tissue to tissue differences are determined by differences in the affinity of the carriers for TMAO and amino acids. Figure 1 shows the intracellular composition in three lower vertebrates: the osmotically tolerant skate, Raja erinucea (Forster and Goldstein, 1976), the euryhaline flounder, Pleuronecfesflesus (Lange and Fugelli, 1965), and the euryhaline toad Bufo viridis (Gordon, 1965). As in all marine elasmobranchs the intracellular and extracellular fluids of the skate are in osmotic equilibrium with the surrounding seawater. As shown in Fig. 1 electrolytes compose only about 20% of the total osmolarity of the intracellular fluids. This relatively low contribution of electrolytes to tissue fluid osmolality is characteristic of all elasmobranchs and reflects the limit of these compounds that are compatible with normal cell function. Krogh (1965) stated in reference to the low salt concentration in elasmobranchs: “It would appear, although we are unable to see why, that a total salt concentration slightly below 1%, is essential for the high development reached within the vertebrate phylum, irrespective of their habitat in the sea, in fresh water or on land.” It is likely that higher salt concentrations would be disruptive to electrostatic bonds in proteins and, therefore, interfere with enzyme action and all of the cellular processes (e.g., metabolism) that are dependent on these crucial agents (Yancey ef al., 1982). By substituting organic solutes such as urea, TMAO, and free amino acids for electrolytes elasmobranchs have developed the ability to be in osmotic equilibrium with the surrounding seawater while obviating the negative effects of high electrolyte concentrations. The major organic solute used by elasmobranchs for osmoregulatory purposes is urea. Since the body surface of these fishes has a low permeability to urea (Boylan, 1967), this solute is quite effective in raising the osmotic pressure of the body fluids and aiding in maintaining osmotic equilibrium between the inside and outside of the fish. However, the cell membranes not exposed to the outside medium are considered to be highly permeable to urea so that urea is thought to exert little, if any, osmotic force between the extracellular and intracellular fluids (Forster and Goldstein, 1979). Intracellular and extracellular urea concentrations are nearly the same in elasmobranchs.
185
CELL VOLUME REGULATION IN LOWER VERTEBRATES
It should be noted that not all elasmobranch cell membranes are freely permeable to urea. As shown in Table I, urea equilibrates across the cell membranes of cardiac muscle and rectal gland within 1 hr after intravenous injection of the labeled compound into the dogfish, Squalus acanrhias. However, urlta equilibrates much more slowly across the cell membrane of skeletal muscle; after I hr there is only 60% equilibration of injected [ ''C]urea between the intracellular and extracellular fluids of skeletal muscle. Cell membranes, in the dogfish at least, have a low permeability to TMAO. As shown in Table I, even after 20 days injected [14C]TMA0 has still not equilibrated across the muscle cell membrane. The erythrocyte membrane is somewhat more permeable to TMAO, but even here it takes days for the compound to equilibrate across the erythrocyte cell membrane. The passive permeability of cell membranes to the other major organic solutes such as amino acids is probably closer to that of TMAO than urea. Thus, of the intracellular solutes shown in Fig. 1, inorganic ions, TMAO, and amino acids exert the major osmotic effect across the cell membranes. In most cells urea exerts little osmotic effect. However, in certain cells such as skeletal muscle urea may make a significant contribution to the internal osmotic pressure. Urea concentrations can reach 0.4-0.5 M in some elasmobranchs (Bernard er al., 1966). This concentration of urea can disrupt hydrogen bonds and destabilize proteins. It has been suggested that the destabilizing effects of high concentrations of urea are counteracted by other organic solutes such as TMAO and certain free amino acids (e.g., taurine) that are accumulated in the body fluids of these fishes (Yancey ef al., 1982). As shown in Fig. 1, the wing muscle of the little TABLE I TISSUE~PLASMA RATIOSOF ['%]UREA AND ['4C]TRlMETHYLAMINE OXIDE ([I'TITMAO) AT DIFFERENT TIMESAFTER INTRAVENOUS INJECTION INTO THE DOGFISH, Squufus ucanrhius Tissue/plasrna ~~
~
TMAOh
Urea" Tissue
0.5 hr
1.0 hr
2 days
20 days
Skeletal muscle' Cardiac muscle Rectal gland Erythrocytes
0.29 0.76 0.80
0.44 0.83
0.05
0.26
0. I6
-
0.85 -
-
0.87
Data from Fenstermacher ef al. (1972). From Goldstein er 01. (1967). c Tissue water (g water/g tissue g water/g plasma): skeletal muscle 0.72; cardiac muscle = 0.81; rectal gland = 0.78. a
7
=
186
LEON GOLDSTEIN AND ARNOST KLEINZELLER
skate contains about 200 mM free amino acids and 50 mM TMAO, compared to about 350 mM urea. The ratio of urea to TMAO + free amino acids (I .4) is close to that reported to have little or no effect on enzyme activity of lower vertebrates. The accumulation of free amino acids in cells has been examined extensively in the little skate (Boyd er af., 1977). In general three amino acids are concentrated to high levels inside the cells of this fish: taurine, p-alanine, and sarcosine. The different tissues vary with regard to which of the three amino acids that they accumulate. Muscle has high concentrations of p-alanine and sarcosine; erythrocytes have p-alanine and taurine and heart has taurine. It is not clear why these three amino acids are accumulated except that in addition to having proteinstabilizing qualities, they are not in the mainstream of metabolism and, therefore, do not interfere with metabolic processes such as protein synthesis. Why certain tissues concentrate one or the other of the three amino acids is also not obvious. At least in the case of taurine and p-alanine cellular accumulation takes place by a Na+ -dependent uptake which can concentrate these amino acids to a very high degree (500- lOOOX their concentration in extracellular fluid) (Goldstein and Boyd, 1978; Forster and Hannafin, 1980). The cells accumulating the amino acids lack the metabolic machinery necessary for metabolizing them, so that enzymatic degradation is not a problem in maintaining high concentration of these amino acids. As noted above, there seems to be a reciprocal relationship between amino acid and TMAO accumulation. Thus, in comparison to the little skate in which cellular amino acid concentrations are high and TMAO relatively low, in the spiny dogfish the opposite holds. TMAO and other quarternary amines are found in high concentrations whereas free amino acid concentrations are relatively low compared to the little skate (Robertson, 1975). In marine teleosts, body fluid osmolality is much lower than the surrounding seawater. Most of the osmolality of the extracellular fluid is contributed by Na+ and C1- with other inorganic ions such as K + , Mg2+, Ca2+, and SO,2contributing to the remainder of the osmolality (Lange and Fugelli, 1965). As shown in Fig. 1 the muscle cells of the euryhaline teleost, PleuronectesjZesus, contain levels of inorganic ions similar to those found in the muscle cells of elasmobranchs, but the muscle cells of the former have much lower concentrations of organic solutes. Thus, inorganic ions contribute a greater percentage to the total osmolarity of teleost cells than in the cells of elasmobranchs. While the concentration of TMAO in teleost muscle is somewhat lower than in elasmobranchs, it still is substantial. As shown in Fig. 1, TMAO concentration in flounder muscle is approximately 30 mM compared to about 65 mM in the little skate. Similarly, the free amino acid concentration in flounder muscle is approximately 70 mM (compared to 220 mM in the skate), which, taken together with TMAO, comprises 25-30% of the total osmolarity of the intracellular fluid. The solute composition of intracellular fluids of amphibia and reptiles has not
CELL VOLUME REGULATION IN LOWER VERTEBRATES
187
been as well studied as in fishes. However, the studies that have been done indicate that similarities in composition do exist. Figure 1 shows the cell composition of the euryhaline toad, Bufo viridis. In freshwater, the muscle cells of this toad contain about the same concentration of inorganic ions as do the cells of skate and flounder. In addition, free amino acids are accumulated intracellularly just as in the fishes. There does not seem to be a significant concentration of TMAO, however. Nevertheless organic solutes (amino acids and urea) constitute about 40% of the intracellular osmolarity of this amphibian.
IV. CHANGES IN CELL COMPOSITION DURING OSMOTIC STRESS Most animal cells swell or shrink when the medium bathing them becomes hypo- or hypertonic. Return to original volume is achieved by a loss or gain of osmotically active solutes so that the internal and external concentrations of solute are in osmotic equilibrium. In invertebrates and lower vertebrates changes in both inorganic electrolytes and organic solutes participate in the osmotic adjustments taking place inside the cells during osmotic stress (Gilles, 1975). As shown in Fig. 1, environmental dilution of the little skate, with accompanying dilution of the extracellular fluids, produces a small drop in concentration of intracellular inorganic ions in muscle (wing) cells. Urea concentration falls significantly. Since the passive permeability of skate wing muscle to urea is unknown it is not possible to predict how much of an osmotic change the drop in urea produces. The concentration of total free amino acids also falls markedly. Since the passive permeability of cell membranes to amino acids is low the fall in amino acid concentration makes a significant contribution to the drop in cell osmolyte concentration. The decrease in TMAO concentration also aids in restoring cell volume but the exact contribution is difficult to assess since the permeability of the skate wing muscle cell to this compound is not well defined. Studies on the effects of environmental dilution in the hagfish have shown that both inorganic ions as well as organic solutes participate in volume regulation of muscle cells (Cholette et al., 1970; Cholette and Gagnon. 1973). In contrast to the skate, in which muscle K + concentration does not change significantly during environmental dilution, K concentration falls proportionally with the drop in osmotic pressure of the surrounding seawater in the hagfish. As in the skate, a major fraction of the change in cell solute concentration is due to a decrease in free amino acid concentration. TMAO concentration also drops significantly in diluted hagfish. Environmental dilution of flounder produces a response in muscle cell composition qualitatively similar to that observed in the skate, although the magnitude of the changes in the flounder are much less than those in the skate, +
188
LEON GOLDSTEIN AND ARNOST KLEINZELLER
commensurate with both the lower total osmotic pressure and its change during environmental dilution in thc: flounder. As shown in Fig. 1, the concentration of muscle inorganic ions charges very little when flounder are transferred from seawater to freshwater. TMAO concentration falls by 50% under these conditions but muscle osmolality is changed by only 15 mOsm/liter by this drop. On the other hand, total free amino acid concentration falls by 40% producing a 35 mOsmiliter drop in muscle osmolality. The fall in amino acid and TMAO concentrations in muscle accounts for about 80% of the decrease in muscle osmolality of flounder adapted to freshwater. The amino acid profile was not done in these experiments so ttat it is not known which amino acids fell during environmental dilution. However, in a similar study (Lasserre and Gilles, 1971) done on another flounder, Paralichrhys lethostigma, in which muscle amino acid concentration falls significantly in fish acclimated to freshwater, muscle taurine, glycine, and alanine, the three major amino acids in this tissue, were markedly lower in freshwater fish than in their saltwater counterparts. Similar studies done on other teleosts have yie1d:d results consistent with those found in the flounder. Compared to the numerous studies done on the effects of environmental salinity changes on cell composition in fish, relatively few such studies have been done on amphibia. One study done on the euryhaline toad Bufo viridis is shown in Fig. 1. In contrast to the skate and flounder, where only small changes in muscle inorganic ion contmt are observed during alterations in environmental salinity, toads adapted to 53% seawater show marked increases in the concentration of inorganic ions in the muscle cells. K + , Na+ , and C1- all rise in the muscle cells under these conditions. Urea concentration doubles in muscle, directly proportional to the increase in plasma urea concentration. Total free amino acid concentration, which constitutes about one-third of the tissue osmolarity, increases by about 90% in toads adapted to 50% seawater. Thus, increases in inorganic ions, urea, and free amino acids all contribute to the rise in muscle osmolarity during exposure of toads to a hyperosmatic environment. Degani (1985) has recently reported that K ,Na+ , and, to a minor extent, urea are elevated in erythrocytes of B. viridis acclimated in virro to hypertonic NaCl. The major amino acid participating in volume regulation by muscle cells of B. viridis adapted to a hyperosmotic environment is taurine (Gordon, 1965). Its concentration increases attout 50% (from 19 to 28 mM) in toads acclimated to 40% seawater. Alanine and glycine increase significantly, as well. This pattern of amino acid change re!+emblesthat described above for the flounder P. lerhosrigma during changes in environmental salinity, but differs from that observed in the skate where sarcosine and p-alanine are the major contributors to changes in muscle osmo1:uity. The mechanism(s) by which cellular amino acid concentrations are modulated during changes in enviror mental salinity and extracellular fluid osmolality have not been well studied in lower vertebrates. Extensive studies of this phenomenon +
CELL VOLUME REGULATION IN LOWER VERTEBRATES
189
have been done in invertebrates where experiments have shown that alterations in both cellular metabolism and membrane transport can play a role in bringing about changes in amino acid concentrations in the tissues of invertebrates acclimated to different salinities or exposed to hypoosmotic stress in v i m . Goldstein and Boyd (1978) did a detailed study on the mechanism of p-alanine regulation in the erythrocytes of skates (Raja erinacea) during hypoosmotic stress. When skates were acclimated to 50% seawater the erythrocyte p-alanine concentration fell to approximately one-half the control level. Since skate erythrocytes do not metabolize p-alanine a change in the rate of synthesis or degradation could not have played a role in lowering the amino acid level in the cell. Possible changes in membrane transport mechanisms that would lead to alterations in erythrocyte p-alanine accumulation were investigated in v i m . Lowering the Na+ concentration in the erythrocyte incubation medium to two-thirds the control level (a decrease similar to that which occurs in the plasma of skates acclimated to 50% seawater) while maintaining medium osmolarity constant decreased the rate of p-alanine influx by about 30%. However, palanine efflux rates were also decreased by a similar percentage, so that there was no net effect of medium sodium reduction (in the range of 300 to 200 mM) on the bidirectional flux of p-alanine in skate erythrocytes. When medium osmolality was lowered by one-third and extracellular Na held constant, p-alanine influx was unchanged while efflux increased about 70%. It appears, therefore, that changes in extracellular Na' concentration play little or no role in lowering erythrocyte@-alanineconcentration during environmental dilution of skates but that a reduction in extracellular osmolality has a major effect. The mechanism by which hypoosmolality leads to an increased efflux of (3-alanine is unknown. In this same study it was observed that acclimation of skates to 50% seawater for 7 days (but not 2 days) produced a 50% reduction in the rate of influx of p-alanine into erythrocytes incubated in v i m . +
V.
EXTRACELLULAR FLUID REGULATION AND CELL VOLUME CONTROL
The extracellular fluid concentration of a solute is important in the dynamics of its distribution between the intracellular and extracellular compartments. For example, the cellular concentrations of most amino acids are much higher than their concentrations in the extracellular fluid. Thus, there is a favorable diffusion gradient for efflux of amino acids from the cell. During environmental dilution when there is an increased efflux of amino acids into the extracellular fluid, the ensuing rise in concentration of released amino acids in the extracellular fluid could decrease the favorable gradient necessary for the continued efflux from the cells and hinder cell volume regulation. In addition, the rise in plasma amino
190
LEON GOLDSTEIN AND ARNOST KLEINZELLER
acid concentration could stimulate reuptake of the amino acid into the cells from which it is being released since the K , values of most amino acid transport systems are well above plasma concentrations. Therefore, mechanisms are needed to remove amino acids released into the extracellular fluid from the body. The removal could be achieved by metabolism or excretion. In the skate (R. erinacea) the mechanism for removing p-alanine and sarcosine released from the tissues during hypoosmotic stress is metabolism (mainly in the liver) while taurine is excreted intact via the kidneys. Leech and Goldstein (1983) have shown that p-alanine is oxidized by skate liver and Shuttleworth and Goldstein (1984) have done a detailed kinetic analysis of the uptake of p-alanine by hepatocytes isolated from the skate liver. The latter authors showed that increased oxidation of p-alanine during hypoosmotic stress could be achieved, theoretically at least, by changes in the rate of uptake of p-alanine into the liver cells. Although no significant changes were found in the activities of the enzymes involved in p-alanine oxidation or the uptake rate of p-alanine into isolated hepatocytes, the kinetic constants (K,) of both uptake and oxidation are such that both processes are sensitive, self-regulating systems. The K , for the uptake of f3-alanineby skate hepatocytes is about 0.2 mM, which is in the normal range of plasma concentrations (0.2-0.4 mM). Thus, fluctuations in extracellular fluid concentrations of p-alanine resulting from altered release of the amino acids from cells during osmotic stress could produce significant changes in the hepatic uptake and, therefore, oxidation of p-alanine. Taurine is not oxidized by fish tissues, so it has to be eliminated from the extracellular fluid by excretion. Schrock et al. (1982) discovered that taurine is excreted by the kidneys of marine fishes by both filtration and secretion. The secretion of taurine by the fish kidney is the first example of the secretion of a naturally occuring amino acid by the kidneys. In vitro experiments with kidney slices, teased tubules, and membrane vesicles have shown that taurine secretion involves accumulation of the amino acid across the peritubular membrane by a sodium, chloride-dependent process followed by downhill movement of taurine across the luminal membrane, probably occurring by cotransport with sodium (King and Goldstein, 1985). The taurine transport systems in the peritubular membrane of both the dogfish and flounder have been found to have K, values above the normal plasma levels of the amino acid. Therefore, increases in plasma taurine concentrations occurring during environmental dilution will stimulate renal taurine secretion (as well as filtration) and lead to increased taurine excretion. It appears that both renal excretory as well as hepatic metabolic processes are geared to compensate for increases and decreases in the levels of amino acids released into the extracellular fluid during volume regulation so as to maintain stable concentration gradients of these solutes between the intracellular and extracellular fluids.
191
CELL VOLUME REGULATION IN LOWER VERTEBRATES
VI. ROLE OF INTRACELLULAR OSMOLYTES IN THE MECHANISM OF CELL VOLUME REGULATION The phenomena of cell volume maintenance (at isotonic conditions) and regulation (in anisotonic media) were first studied on models of mammalian cells (Leaf, 1956; Kleinzeller, 1972) where the role of osmolytes additional to the nondiffusable anions did not appear to deserve major consideration. The analysis above raised questions as to the possible involvement of osmolytes in volume control of cells displaying osmoregulation. Studies designed to define the determinants of volume regulation in elasmobranch cells (rectal gland of the spiny dogfish, Squalus acanthius) readily established a role of the pump-and-leak mechanism on the basis of several criteria (Kleinzeller and Goldstein, 1984; Kleinzeller et al., 1985a; Kleinzeller, 1985). The model of ionic fluxes in these cells (Greger et al., 1984) proved to be valuable for some points of the present analysis: Under steady state conditions in vitro (tissue bathed aerobically at 15°C in standard elasmobranch saline): 1. The cell membrane is permeable to the bulk ionic species, Na , K , and H,O and C1- , the passive fluxes of Na and K across the (basolateral) membrane are relatively slow, and may become rate limiting under some experimental conditions 2. The presence of a Gibbs-Donnan system as the basis for the establishment of an osmotic gradient across the cell membrane, and hence for cellular swelling, follows from considerations of the ionic distribution, particularly of [K ] and from the massive swelling seen in isotonic high-K media (cf. Boyle and Conway, 1941) 3. The role of the Na+ ,K+-ATPase in the extrusion of Na+ (and H,O) has been established on the basis of observed ouabain effects on the cellular ionic distribution 4. No osmotic pressure gradient across the cell membrane could be detected 5. The (basolateral) membrane of the rectal gland cells appears to be readily permeable to urea; however, the imposition of a diffusion gradient for urea (ureafree, hypotonic media) did demonstrate a transient osmotic effect of this solute, reflected by changes of cell volume 6. As opposed to the stimulated rectal gland cells (Silva et af., 1977), no effects of furosemide could be detected in the resting cells (cf. also Shuttleworth and Thompson, 1980; Greger et af., 1984); hence, in contrast to many other cells +
+
C1- ; as compared with the permeability of the membrane to +
+
+
,'
+
'The apparent intracellular concentrations of ionic species are indicated by square brackets, with subscript i; extracellular concentrations: subscript a.
192
LEON GOLDSTEIN AND ARNOST KLEINZELLER
(cf. Hoffmann, 1985; Geck and Pfeiffer, 1985; Spring and Ericson, 1982), the Na+-K+ -2C1- cotransport system may play only a minor role in the volume regulation of the resting rectal gland cells2 A closer analysis of the steady state cellular distribution of bulk ionic species drew attention to some additional points to be considered. It will be seen from Table I1 that the sum of concentrations of bulk cations in rectal gland cells ([Na+Ii + [K+Ii), was 215 mM, i.e., considerably lower than in the incubation medium ("a+], + [K+], = 290 mM). The above data are not an analytical error, being in essential agreement with those of other authors for this tissue in vitro and in vivo (Silva et al., 1977; Shuttleworth and Thompson, 1980; Greger et af.,19843), as well as with available data on Squalus acunrhias muscle in vivo (here, the absence of reliable data on the extracellular fluid compartment prevents an appraisal of the apparent intracellular solute concentrations). Hence, a diffusion gradient of bulk electrolytes exists across these cell membranes. This diffusion gradient R is defined here as the ratio ([Na+Ii + [K+],)/([Na+], + [K+I,). Two (not mutually exclusive) alternative mechanisms may account for the generation of such diffusion gradient of bulk cations across the cell membrane: (1) the presence of some additional cellular cationic solute which would participate in the Gibbs-Donnan distribution; in view of the relatively low values for cell Ca2+ and Mg2+ such cation could be presumed to be organic; and (2) an unsuspected effect of uncharged intracellular solutes (osmolytes, cf. Yancey er al., 1982) on the ionic distribution. An inspection of Table I11 demonstrates the presence of considerable amounts of organic solutes in dogfish muscle and rectal gland cells. As to organic cations (methylamines, betaine, creatine, and creatinine and also little dissociated organic compounds such as TMAO), the data provide clear evidence that the cells are capable of maintaining (or accumulating) these solutes against major concentration gradients by hitherto poorly understood mechanisms. However, given the known dissociation constants of many of these solutes (pK, 8- 10) the intracellular concentration of total (dissociated and undissociated) organic compounds would have to be at least one order of magnitude higher than found in order to satisfy the requirements of a Gibbs-Donnan distribution. The relatively high concentration of free amino acids is noteworthy. That of taurine (54 & 8.3 2It would appear that the operation of this system can be triggered by lowering the intracellular activity of CI- (Greger et al., 1984). 3From the data of Greger et al. (1984), the derived apparent intracellular concentrations were (d): [Na+Ii, 16; [K+Ii, 175; i.e.. "a+] was considerably lower than that given in Table 11. Several possible explanations for this discrepancy may be considered: ( I ) an underestimate of the true extracellular space would tend to increase the chemically assessed "a+ 1; (2) cellular compartmentation of Na+ , including possible "binding," would decrease the activity of intracellular Na measured with ion-selective microelectrodes, and hence would decrease the derived [Na +],
TABLE II SOLUTE DISTRIBUTION IN MUSCLEAND RECTAL GLAND CELLSOF THE DOCFISH(Squnlus oranthias) Steady state intracellular ionic composition in dogfish rectal gland [mmolikg cell H,O (mM)]“
Fresh tissue (mmol/kg H,O) Solute
Plasmab
Na
250
+
K+ Caz + Mg‘
+
CI Nitrogeneous compounds “NH4“ Trimethylamine Dimethy lamine Trirnelhylarnine oxide Urea Bulk amino acids Tau GlY Ala Ser Glu ASP B-Ala Ethanolarnine Betaine Creatinine Creatine Myoinositol
7 3 I .3 240 0.2 0 F . e 2.64 I .7” 72r 3306.‘
Muscleb
Rectal glandc
18 I30
88 130 I .o 5.0 I26
2 7 13
0.0Y 0.05’
3 1
216
-
10.8“
3.1“ 68i 338c
I 80 333
350
-
45 55
-
-
0.95
2.w 48 0.39 2.3 8.5
-
0. Ih
28s 5
9d.c
4.7 -
0 53 0. I 0.I7 0.058 0.05 0.02 0.05 9h
Incubation medium
looh
-
-
686 Ih
-
12.5’ ~~
“Tissue incubated aerobically 2 hr in standard elasmobranch saline. bRobertson (1975).
‘Kleinzeller and Goldstein (1984) dC. Kahn and Kleinzeller (unpublished data obtained by mass spectrography with monomethylamine as standard).
.
<Erecinska and Kleinzeller (unpublished data obtained by enzyme analysis). ‘Kleinzeller (1985). K k c h er a/. (1979). hCha and Kleinzeller (unpublished data, obtained by an amino received from analyzer). ‘Kleinzeller. er of. (1985).
CellsC
64 151 0.7 7.4 97 10.4d.c 10.w I .7d
66f 3* 54 ‘4 0.9 3h 4.6
0.6 2.3 10.5
-
I 6‘
TABLE Ill STEADY-STATE CATIONDISTRIBUTIONIN VARIOUSCELLS” Plasma or medium Tissue
“a+],
[K+],
(~Na+],+[K+],)
Cells “a+],
[K+li
([Na+li+[K+li)
Rb
124 262 170
128 262 216
1.21 0.90 0.45
Conway (1957) Foster and Goldstein (1976) Clark and Hinke (1981)
173 218 230 184 239
1.15 1.48 0.79 1.12 1.43
Solomon (1952) Tosteson and Robertson (1956) Forster and Goldstein (1976) Christensen and Riggs (1952) Van Rossum and Russo (1981)
References
Muscle
Frog Skate Barnacle Red blood cells Human Duck Skate Ascites tumor cells (mouse) Liver slices (rat) Kidney Cortical slices (rabbit) Proximal tubule cytoplasm (rat) Distal tubule cytoplasm (rat) Papillary coll. tubule cytoplasm (rat)
b
104 290 476c
2 2 8c
106 292 484c
145 141 290
5
161
6 2 4 6
150 147 292 I64 I67
127
159 189 172 134 112
151
6
157
77
172
249
1.58
29
177
206
1.31
Kleinzeller and Knotkova (19w Beck et al. (1980)
17 34
194 169
21 1 203
1.34 0.37
Beck et al. (1980) Beck et al. (1984)
160
506
36
542
4 -
46 14 29 58 50
Apparent intracellular and extracellular cationic concentrations are given in mEqlkg H20 (mM). R : ([Na+],+[K+],)/([Na+],+[K 1,). Coelomic fluid. +
CELL VOLUME REGULATION IN LOWER VERTEBRATES
195
mM, n = 4) is in keeping with high values seen in other osmoregulatory tissues (Forster and Goldstein, 1976). However, this acidic amino acid (under physiological conditions, some 10% of the intracellular solute may be dissociated) would actually be expected to enhance (rather than decrease) the expected Donnan distribution of permeable cations. Hence, the above considerations tend to shift emphasis to alternative (2) as an explanation of the observed diffusion gradient of bulk cations. Table 111 demonstrates for some randomly chosen examples that the observed diffusion gradients for bulk cell cations (evidenced by R < 1) characterizes predominantly species and cells involved in osmoregulatory processes (elasmobranchs, e.g., dogfish and skate, barnacle muscle, renal medulla) whereas in other cells and tissues the conventional Gibbs-Donnan distribution (with R > 1) is found. The mechanism by which cellular osmolytes may affect cell volume and ionic distribution follows from an extension of the formal analysis of the pump-andleak system, as presented by Post and Jolly (1957). Here, the aqueous intracellular space contains impermeable, osmotically active components, A. A leak L allows the permeable solute S to pass through the membrane with a rate constant 1. In addition, S is extruded from the cells by a metabolically operated pump P (rate constant p ) . The concentration of S in the extracellular compartment, [S],, is constant. The steady state cell volume V is then given by
V = A/(S],(I - Ilp)
(1)
Hence, three parameters determine the cell volume: the amount of intracellular A, and the ratio of the rate constants for the leak and the pump. The formal analysis of the pump-and-leak system has hitherto emphasized particularly the role of the nondiffusible anions as the crucial determinant of the Gibbs-Donnan (and double Donnan) systems (cf. Leaf, 1956). It can be readily shown (Kleinzeller, 1985) that in the absence of an osmotic pressure gradient across the cell membrane, an increase of the intracellular osmotic components will decrease the intracellular concentration of bulk cations, R falling to values < 1. Equations derived as a first approximation for the steady state (Kleinzeller, 1985) demonstrate this point. For simplification, (“a+] + [K+]) is denoted [ B + ] , the subscripts i and o indicating the intra- and extracellular compartments, respectively. While A here is the nondiffusible anion of a Gibbs-Donnan system, with n charges,fis a constant corresponding to the reflection coefficients for the bulk electrolytes:
f = n,, -
+ uc,)
(where simplistically it is assumed that uNa= uK).Then, at steady state, [B+],/([B+]+ , A”--f)
= 1
(2)
196
LEON GOLDSTEIN AND ARNOST KLEINZELLER
clearly demonstrating that [B the cells, Eiq. (3) pertains: {[B+Ii
+
Ii > [B '1,.
When osmolyte 0 is accumulated in
+ O/(~U,, - l)}/{[B+], + A"-f}
= 1
(3)
The greater O/(no,, - l), the lower ([Na+Ii + [K+Ii). In effect, the cellular accumulation of an (effectively) impermeant osmolyte thus affects the intracellular ionic distribution indirectly, the osmotic equilibrium imposing a redistribution of permeant cell electrolytes. The present view is actually an extension of an analysis (Wilson, 1954; Freedman and Hoffman, 1979) showing that the oncotic pressure of cell proteins also plays an important role as a determinant of volume and ionic distribution in blood cells, in contrast to analyses emphasizing particularly the role of the charge of the nondiffusible anion. The above view permits some predictions which can be experimentally tested. Thus, the loss of intracellular osmolytes would be expected to produce an increase in R. It will be shown below that this prediction has been verified. The reverse test, i.e., a detailed analysis of cell volume and ionic distribution as a function of cellular accumulation of an uncharged osmolyte, has not yet been carried out. Experiments on volume regulation of rectal gland cells in anisotonic media (Kleinzeller and Goldstein, 1984) actually suggest two possibilities by which the accumulation of an osmolyte might affect cell volume, ionic distribution (and membrane potential): ( I ) The accumulation of an osmolyte is associated with an osmotic influx of H,O, thus diluting the intracellular electrolytes; and (2) an efflux of cell cations may be triggered, e.g., by a transient increase in the passive K + flux (cf. Bakker-Grunwald, 1983). In order to serve as a cellular osmolyte, the particular solute has to be impermeant, or effectively impermeant (being actively accumulated by the cells). Amino acids are an obvious major osmolyte in a variety of cells, e.g., in the barnacle muscle (intrawllular total concentration 500 mM, with a gradient of lo3) (Clark and Hinke, 1981), but also in elasmobranchs (particularly taurine, see Table 11). In dogfish muscle, TMAO is clearly accumulated against a major gradient, and in some shark species the concentration gradient exceeds 10 (Norr i s and Benoit, 1945); on the other hand, in the rectal gland the TMAO distribution between the cells and the plasma appears to be at equilibrium, and the role of this solute as an impermeant osmolyte becomes evident only when incubating the tissue in TMAO-free elasmobranch saline. Urea,which is quantitatively the most important osmolyte in the plasma of elasmobranchs (Forster and Goldstein, 1976), does not appear to qualify as a cellular osmolyte because (1) at steady state the concentration of this solute in the plasma and in the cells (Table 11) is practically identical and (2) because of the relatively high permeability of the cell membranes such as that of rectal gland to this solute (Kleinzeller and Goldstein, 1984). The significant effect of higher concentrations of urea (up to 2 M) on the ionic distribution in renal cortical tissue at 0°C (when the operation of the pump
197
CELL VOLUME REGULATION IN LOWER VERTEBRATES
is believed to be inhibited, and hence the Donnan system is chiefly responsible for the ionic distribution) (Robinson, 1962) requires additional study. Finally, attention has been paid recently to the role of polyols as osmolytes. Myoinsitol has been found to be a significant osmolyte in the renal medulla (Cohen et al., 1982; Bagnasco er al. (1986). Rather high concentrations of this polyol have now been found in the rectal gland (Table 111). D-Sorbitol another polyol present as a cellular osmolyte in a variety of cells (see Yancey et al., (1982). including the renal medulla (Bagnasco et al., 1986) has not been found in measurable amounts in the rectal gland (McGregor and Kleinzeller, 1986).
VII. CELL VOLUME MAINTENANCE AND THE ROLE OF THE CYTOSKELETON High-K+ incubation media produce a massive cell swelling (in muscle, but also in other cells) as a consequence of membrane depolarization and subsequent influx of K ,C1- , and H,O in accordance with the respective electrochemical gradients (Boyle and Conway, 1941). Dogfish rectal gland cells display the same phenomenon (Kleinzeller et a / ., I985a): In high-K + media (all Na replaced by K ), the cells lose Na ,and a massive influx of K + , C1- ,and H,O is seen; the membrane potential (assessed from the distribution of a lipophilic cation), E , and E,, fall to low values. The above effects are clearly reversible. On transfer of the swollen tissue to standard (high-Na+ , low-K+) medium, the cells extrude H,O, K ,and C1- ,and gain some Na ;also, the apparent membrane potential correspondingly increases. Considerations of this reversal process raised questions as to the possible involvement of mechanisms other than the pump-and-leak system in the observed cell volume regulation. A reversal of the K +-induced swelling has been previously seen in frog skin (McRobbie and Ussing, 1961), and the following mechanism was suggested: The permeability of the cell membrane to K is greater than that of Na . The driving forces being initially practically identical for both ionic species, the net efflux of K would exceed that of Na with a concominant net loss of diffusible anion (mostly C1 -) and water. Such mechanism cannot fully explain the reversibility of the swelling in view of the repolarization of the membrane and the apparent water extrusion against the prevailing osmotic gradient. The search for an additional mechanism involved in the reversal of the K induced swelling in rectal gland cells drew attention to the possible participation of the cytoskeleton (cf. Kleinzeller, 1972). An attractive possible mechanism for the observed phenomenon is based on studies of the fine cell structure under various experimental conditions (Masur et al., 1981). Figures 2, 3, and 4 show the light micrographs of portions of rectal gland tubules incubated in standard saline ( l ) , in high-K+ medium (2), and the reversal of the K + -induced swelling +
+
+
+
+
+
+
+
+
+
+
198
LEON GOLDSTEIN AND ARNOST KLEINZELLER
FIG.2. Light micrograph of portions of several tubules from a rectal gland slice after 120 min in control medium. The lumen is seen at a central triangular space (L), the nuclei are at N. and the “striped” cytoplasmic stain probably represents mitochondrial p u p s Oriented within the basal cytoplasm. X 1320. FIG. 3. Light micrograph of tubules from a rectal gland slice exposed for 120 min to high K + medium. Most cells are swollen, lacking discernible nuclei. Denser outlining may represent cytoplasm of cells (connective tissue?) which can regulate their volume under extreme conditions. A central lumen is not seen. x 1320. FIG.4. Light micrograph of tubules after exposure to swelling conditions followed by 60 min of normal saline. The nuclei (N)are distinguishable from the cytoplasm, the denser staining regions absent in Fig. 2 (mitochondria?)are discernible. Vacuoles are obvious in regions apical to the nucleus and adjacent to where a lumen would be. X 1300.
(3). It will be readily seen that high-K+ medium produces a massive swelling which obliterates the tubular lumen. This swelling reverses on transfer of the tissue to Na+ medium with a clearly marked vacuolization of the cells. Electron micrographs (Figs. 5 , 6, and 7) provide additional information on the above points. The swelling is associated with a loss of cytoplasmic and nuclear electron density and an enlargement of the osmotic space of the mitochondria. The reversal of the K+-induced swelling (Fig. 7) is detectable in a restoration of electron density to that of the control (Fig. 5 ) with multiple vesicles and vacuoles particularly in the apical cell region. The observed vesiculation (Figs. 4 and 7) may be a contributing element to the net extrusion of H,O and electrolytes, as postulated by van Rossum and Russo (1981). More recent studies revealed additional information on these points. First, actin was found to be a constituent component of the cytoskeleton of the dogfish rectal gland (J. W.Mills, personal communication). Thus, the necessary mechanism is present in the cells for a participation of the cytoskeleton in the regulation of cell volume. Second, the
FIG.5 . Electron micrograph of cells in a slice of rectal gland incubated for I hr in “standard” shark Ringer’s. Microvilli (arrowheads) border the open lumen (L). The cells have a relatively uniform cytoplasmic stippling and vesicles (V)are seen in the apical regions (A). Multivesicular bodies (endosornes, E), also seen in the apical region, are often associated with endocytosis. The basolateral membranes are seen as stacks of interdigitations (I). The ultrastructural appearance of the cell is comparable to that seen in sifu, containing “normal” mitochondria (M) and nuclei (N) as well. X
14000.
200
LEON GOLDSTEIN AND ARNOST KLEINZELLER
FIG. 6. Electron micrograph of apical regions of several cells after 60 min of incubation in high external K + medium. The lumen and microvilli are absent as are vesicles which normally sit below the microvilli. The apical (A) cytoplasm has a stippled appearance, suggestive of unoriented microfilamentousmaterial. The membrane-bounded profiles (P) containing dense cytoplasm may be extensions of connective tissue cells (C) found at the base of tubules. Stacks of Golgi lammellae are seen at (G).Basolateral membranes form interdigitating oval contours (I) with no apparent extracellular space (compare with accordian-like stacks in Fig. 5). The mitochondria (M)have enlarged “empty” compartments which appear to represent the cristae. Thickenings suggestive of desmosomes appear at (D).x7000. FIG. 7. Electron micrograph of apical portions of cells surrounding the tubule lumen in a rectal gland slice which has been incubated for 10 min in control medium following 60 min in high K + . The lumen is hook shaped, extending diagonally across the middle of the photograph, and is indicated by an L and mowheads at either end. Elongated finger-likecytoplasmic projections contain microfilamentsas does the apical cytoplasm (A) (compare with Fig. 4). Small vesicles are also seen in this region and larger vesicles and vacuoles (V) predominate toward the central cytoplasmic regions. The mitochondria (M) appear normally electron dense with narrow cristae and the contribution of two adjacent cell membranes is distinguishable within the basolateral folds. X7000.
microscopically observed K -induced swelling and impairment of the cell fine structure can be correlated with a significant increase in the size of the cellular pool of exchangeable 86Rb (as a measure of the K flux) corresponding to the slow efflux component (Kleinzeller et af., 1985b). This point is of particular interest when considering that a disruption of fine cellular structure and changes +
+
+
CELL VOLUME REGULATION IN LOWER VERTEBRATES
201
in the cellular pools of 86Rb are also found when cellular swelling in high-K media is prevented using relatively impermeant anions (such as isethionate or gluconate). Hence, the swelling effect of K may have a component additional to the conventional Boyle and Conway depolarization, i.e., a specific effect of K + (and Rb+) on the cytoskeleton. An additional set of observations is pertinent for this appraisal. An examination of the reversibility of the K -induced swelling noted two points: (1) the cell volume did not quite return to the original steady state value in controls; and (2) in the course of the swelling, the concentration of cell bulk cations rose to approximate that in the medium; while such a result might be taken to reflect the massive swelling (with an isotonic solution, mainly KCl) such explanation cannot be extended to the finding that on reversing the swelling, at the steady state ([Na+Ii + [K+Ii)was significantly higher than in the control. Subsequent studies (Kleinzeller, 1985; Kleinzeller et al., 1985) showed that K+-induced swelling produces a net loss of two major osmolytes, TMAO and myoinositol. These observations are therefore consistent with the model presented above. A loss of cell osmolyte would indeed produce an increase in R (Fig. 3). +
+
+
+
REFERENCES Bagnasco, S . , Balaban, R., Fales, H. M., Yang, Y.-M., and Burg, M. (1986). Predominant osmotically active organic solutes in rat and rabbit renal medulla. J . Biol. Chem. 261, 58725877.
Bakker-Grunwald, T. (1983). Potassium permeability and volume control in isolated hepatocytes. Biochim. Biophys. Acra 731, 239-242. Beck, F., Bauer, R., Bauer, R., Mason, J., Dorge, A . , Rick, R., and Thurau, K. (1980). Electron microprobe analysis of intracellular elements in the rat kidney. Kidney Inr. 17, 756-763. Beck, F., Dorge, A., Rick, R., and Thurau. K. (1984). Intra- and extracellular element concentrations of rat renal papilla in antidiuresis. Kidney Int. 25, 397-403. Bernard, G. R., Wynn, R. A., and Wynn, G. G. (1966). Chemical anatomy of the pencardial and penvisceral fluids of the stingray, Dusyuris umericunu. Biol. Bull. 130, 18-27. Boyd, T. A , , Cha, C.-J., Forster, R. P.,and Goldstein, L. (1977). Free amino acids in tissues of the skate Raju erinucea and the stingray Dasyatis sabinur Effects of environmental dilution. J . Exp. 2001. 199,435-442.
Boylan, J. W.(1967). Gill permeability inSquulus ucanrhius. In “Sharks, Skates and Rays” (P.W. Gilbert, R. F. Mathewson. and D. P. Rall, eds.), pp. 197-206. Johns Hopkins Press, Baltimore, Maryland. Boyle, P. J.. and Conway, E. J. (1941). Potassium accumulation in muscle and associated changes. J . Physiol. (London) 100, 1-63. Cholette, C., and Gagnon, A. (1973). Isosmotic adaptation in Myxine glurinosa L . 4 1 . Variations of the free amino acids, tnmethylamine oxide and potassium of the blood and muscle cells. Comp. Biochem. Physiol. &A, 1009-1021. Cholette, C., Gagnon, A., and Germain, P. (1970). Isosmotic adaptation in Myxine glurinosu L. I. Variations of some parameters and role of the amino acid pool of the muscle cells. Comp. Biochem. Physiol. 33, 333-346. Christensen, H.N., and Riggs, T. R. (1952). Concentrative uptake of amino acids by the Ehrlich ascites carcinoma cells. J . Biol. Chem. 194, 57-68.
202
LEON GOLDSTEIN AND ARNOST KLEINZELLER
Clark, M. E., and Hinke, J. A. M. (1981). Studies on water in barnacle muscle fibres. I. The dry weight components of fresh fish. J . Exp. Biol. 90, 33-41. Cohen, M. A. H., Hruska, K. A., and Daughaday, W. H. (1982). Free myo-inositol in canine kidneys: Selective concentration in the renal medulla. Proc. SOC.Erp. B i d . Med. 169, 380385. Conway, E. J. G. (1957). The nature and significance of concentration relations of potassium and sodium ions in skeletal muscle. Physiol. Rev. 37, 84-132. Degani, G. (1985). Osmoregulation in red blood cells of Bufo viridis. Comp. Biochem. Physiol. 81A, 451-453. Fenstermacher. J., Sheldon, F., Ratner, J.. and Roomet, A. (1972). The blood to tissue distribution of various polar materials in the dogfish, Squalus ucunrhius. Comp. Biochem. Physiol. 42A, 195-204. Forster, R. P., and Goldstein, L. (1976). Intracellular osmoregulatory role of amino acids and urea in marine elasmobranchs. Am. J. Physiol. 230, 925-93 I. Forster, R. P., and Goldstein, L. (1979). Amino acids and cell volume regulation. Yule J. Biol. Med. 52, 497-515. Forster, R. P., and Hannafin, J. A. (1980). Taurine uptake in atrial myocardium by a sodiumdependent p-amino acid system in the elasmobranch,Raju erinucea. Comp. Biochem. Physiol. 67A, 107-113. Freedman, J. C., and Hoffman, J. F. (1979). Ionic and osmotic equilibria of human red blood cells treated with mystatis. J. Gen. Physiol. 74, 157-185. Geck, P., and Pfeiffer, B. (1985). (Na+ + K + + 2Cl-)-cotransport in animal cells-its role in volume regulation. Ann. N . Y . Acud. Sci. 456, 166-182. Gilles, R. (1975). Mechanisms of ion and osmoregulation. In “Marine Ecology” (0.Kinne, ed.), Vol. 11, pp. 259-347. Wiley. London. Goldstein, L., and Boyd, T. A. (1978). Regulation of P-alanine transport in the skate (Raja erinucea) erythrocytes. Comp. Biochem. Physiol. 6OA, 25-30. Goldstein, L.. Hartman, S. C., and Forster, R. P. (1967). On the origin of trimethylamine oxide in the spiny dogfish, Squulus ocanrhias. Comp. Biochem. Physiol. 21, 719-722. Gordon, M. S., Nielsen. K., and Kelly, H. M. (l%l). Osmotic regulation in the crab-eating frog (Ranu canerivoru). 1. Exp. B i d . 38, 659-678. Gordon, S. (1965). Intracellular osmoregulation in skeletal muscle during salinity adaptation in two species of toads. B i d . Bull. 128, 218-229. Greger, R., Schlatter, E., Wang, F., and Forrest, J. N. (1984). Mechanism of NaCl secretion in rectal gland tubules of spiny dogfish (Squulus ucanthias) 111. Effects of stimulation of secretion by cyclic AMP. PJlugers Arch. 402, 376-384. Hoffmann, E.K. (1985). Cell volume control and ion transport in a mammalian cell. In “Transport processes, Ono- and Osmoregulation. Current Comparative Approaches” (R. Gilles and M. Gilles-Baillien, eds.), pp. 389-400. Springer-Verlag,Heidelberg. King, P. A,, and Goldstein, L. (1985). Renal excretion of nitrogenous compounds in vertebrates. Renal Physiol. 8, 261-278. Kleinzeller, A. (1972). Cellular transport of water. In “Metabolic Pathways” ( L . H. Hokin, ed.), Vol VI, pp. 92-131. Academic Press, New York. Kleinzeller, A. (1985). Trimethylamine oxide. and the maintenance of volume of dogfish shark rectal gland cells. J. Eq. h l . 236, 11-17. Kleinzeller, A., and Goldstein. J. (1984). Effect of anisotonic media on cell volume and electrolyte fluxes in slices of the dogfish Squulus acanthius rectal gland. J . Comp. Physiol. B 154,565571. Kleinzeller, A., and Knotkova, A. (1964). The effect of ouabain on the electrolyte and water transport in kidney cortex and liver slices. J . Physiol. (London)175, 172-192.
CELL VOLUME REGULATION IN LOWER VERTEBRATES
203
Kleinzeller, A , , Forrest, J. N., Cha, C.-J.. Goldstein, J., and Booz, G. (1985a). Cell solute composition and potassium effects in slices of the rectal gland of the dogfish shark Squalus acanrhias. J . Comp. Physiol. B 155, 145-153. Kleinzeller, A., Mills, J. W., McGregor, L. J.. and Carre, D. A. (1985b). High external K + and cell volume in the dogfish (Squalus acanrhias) rectal gland. Bull. Mt. Deserr Isl. Biol. Lab. 25,64. Krogh. A. (1965). “Osmotic Regulation in Aquatic Animals.” Dover, New York. Lange, R., and Fugelli, K . (1965). The osmotic adjustment in the euryhaline teleosts, the flounder, P leuronecres Jesus L. and the three-spined stickleback Gasrerosreus aculeatus L. Comp. Biochem. Physiol. 15, 283-292. Lasserre, P., and Gilles, R. (1971). Modification of the amino acid pool in the parietal muscle of two euryhaline teleosts during osmotic adjustment. Experientia 27, 1434- 1435. Leaf, A. (1956). On the mechanism of fluid exchange of tissues in virro. Biochem. J . 62, 241-248. Leech, A. R., and Goldstein, L. (1983). p-alanine oxidation in the liver of the little skate, Raja erinacea. J . Exp. 2001.225, 9-14. Leech, A. R., Goldstein, L., Cha, C.-J., and Goldstein, I . M. (1979). Alanine biosynthesis during starvation in skeletal muscle of the spiny dogfish, Squalus acanrhias. J . Exp. Zool. 207, 73-80. Lockwood, A. P. M. (1966). “Animal Body Fluids and Their Regulation.” Harvard Univ. Press, Cambridge, Massachusetts. McGregor, L. C., and Kleinzeller, A. (1986). Osmolyte function of myo-inositol in dogfish (Squalus acanrhias) rectal gland. Bull. Mt. Desert I d . Biol. Lab. 26, 168. MacRobbie, E. A. C., and Ussing, H. H. (1961). Osmotic behaviour of the epithelial cells of frog skin. Acra Physiol. Scand. 53, 348-365. Masur, S. K., Goldstein, J., and Kleinzeller, A. (1981). Fine structural changes associated with cell volume regulation in slices of dogfish (Squalus acanrhias) rectal gland. Bull. Mr. Desert Is/. Biol. Lab. 21, 48-50. Noms, E. C., and Benoit, G . I. (1945). Studies on trimethylamine oxide. I. Occurrence of trimethylamine oxide in marine organisms. J . Biol. Chem. 158, 433-438. Parker, J. C. (1973). Dog red blood cells. Adjustment of density in vivo. J . Gen. Physiol. 61, 146157. Pierce. S. K. (1982). Invertebrate cell volume control mechanisms: A coordinated use of intracellular amino acids and inorganic ions as osmotic solute. Biol. Bull. 163, 405-419. Post, R. L., and Jolly, P. C. (1957). The linkage of sodium, potassium and ammonium active transport across the human erythrocyte membrane. Biochim. Biophys. Acra 25, 118-128. Robertson, J. D. (1975). Osmotic constituents of the blood plasma and parietal muscle of Squalus acanrhias. Biol. Bull. 148, 303-319. Robinson, I . R. (1962). Effects of urea upon the water and ionic content of kidney slices. J . Physiol. (London) 164, 552-562. Schrock, H., Forster, R. P., and Goldstein, L. (1982). Renal handling of taurine in marine fish. Am. J . Physiol. 242, R64-69. Shuttleworth, T. J., and Goldstein, L. (1984). P-Alanine transport in the isolated hepatocytes of the elasmobranch, Raja erinucea. J. Exp. Zoo/.231, 39-44. Shuttleworth, T. J., and Thompson, J. L. (1980). The mechanism of cyclic AMP stimulation of secretion in the dogfish rectal gland. J . Comp. Physiol. B 140, 209-216. Silva, P., Stoff, J., Field, M., Forrest, J. N., and Epstein, F. H. (1977). Mechanism of active chloride transport in the rectal gland: Role of Na-K-ATPase in chloride transport. Am. J. Physiol. 233, F298-F306. Smith, H. W. (1930). Metabolism of the lung-fish. Protopferus aethiopleus. J . Biol. Chem. 88,97130. Solomon, A. K. (1952). The permeability of the human erythrocyte to sodium and potassium. J. Cen. Physiol. 36, 57-1 LO.
204
LEON GOLDSTEIN AND ARNOST KLEINZELLER
Spring, K. R., and Ericson, A.-C. (1982). Epithelial cell volume modulation and regulation. J . Membr. Biol. 69, 167-176. Tercafs, R. R., and Schoffeniels, E. (1962). Adaptation of amphibians to salt water. Life Sci. 1, 1923.
Tosteson, D. C., and Robertson, J. S . (1956). Potassium transport in duck red cells. J. Cell. Comp. Physiol. 47, 147-166. Van Rossum, G. V. D., and Russo, M. (1981). Ouabain-resistant mechanism of volume control and the ultrashuctural organization of liver slices recovering from swelling in vifro.J. Membr. Biol. 59, 191-209. Weissberg, J . , and Katz, U. (1975). Effect of osmolality and salinity adaptation on cellular composition and on potassium uptake of erythrocytes from the euryhaline toad, Bufo viridis. Comp. Biochem. Physiol. 52A, 165- 169. Wilson, T. H. (1954). Ionic permeability and osmotic swelling of cells. Science 120, 104-105. Yancey, H.,Clark, M. E., Hand, S. C., Bowlus, R. D., and Somero, G. N. (1982). Living with water stress: Evolution of osmolyte systems. Science 217, 1214-1222.
CURRENT TOPICS IN MEMBRANES AND TRANSPORT. VOLUME 30
Volume Regulation in Cells of Euryhaline Invertebrates R . GILLES Laboratory of Animnl Physiology Institute of Zoology University of Liege 8 4 0 2 0 Liege, Belgium
1.
INTRODUCTION
So-called euryhaline, aquatic invertebrates have long been considered most useful models for the study of mechanisms of volume control in cells submitted to anisosmotic conditions as well as of the effects of changes in ion concentration on cell structure and activity. These species can indeed sometimes withstand very large changes in the salinity of their environmental medium, though they have only limited or no power of blood osmoregulation. During such changes, their cells and tissues have to sometimes cope with very important modifications in blood osmolality and ion concentration. For instance, in a species such as Eriocheir sinensis, a very euryhaline decapod crustacean, adaptation from sea water to fresh water results in a decrease in blood osmolality (mostly due to a drop in NaCl content) from some 1 100 mOsm/liter to about 550 mOsm/liter. Similarly, rather large changes in blood osmolality can also be found in many other euryhaline crustaceans as well as in a variety of mollusks and worms during acclimation to media of different salinities (for reviews, see for instance Potts and Parry, 1964; Gilles, 1975; Spaargaren, 1979). In one of the most extreme euryhaline blood osmoconformers known, the serpulid annelid Mercierella enigmatica (= Ficopatomus enigmaticus), which can tolerate natural waters of salinities between 1 and 55%0 S , blood osmolality will vary between 84 and 2304 mOsm/liter upon acclimation to these extreme media (Skaer, 1974; Benson and Treherne, 1978a,b). This is at variance with the situation found in most vertebrates in which, with the notable exception for some groups of fish, blood osmolality control is usually very effective. Cells, therefore, never experience important 205
Copynght 0 1987 by Academc Rcss. Inc All nghu or repducnon in any form reserved
206
R. GILLES
osmotic and ionic challenges in normal conditions. It is classically held that changes of only 25% in blood osmolality induced in vivo in mammalian species in the course of the development of a pathology or as part of an experimental procedure may cause large functional damages (see, for instance, Andreoli et al., 1977). Such changes in osmolality represent only some 60 mOsm/liter; nothing to compare with what can be seen in euryhaline invertebrates in which they are currently 10 times larger and occur in perfectly physiological conditions! The problems cells cope with during osmotic challenges are of two types: (1) the development of swelling pressures and thus membrane tension which will affect all membrane activities; (2) the adverse (structure disrupting) effects that large changes in inorganic ion concentration can induce at the level of a variety of membrane and intracellular proteins. This leads to a series of problems, some of which will be considered further in this review: (1) What are the mechanisms available to cope with the development of membrane tension and changes in intracellular volume? (2) How do the mechanisms found in invertebrates compare with those described in mammalian cells and do they have any specificity? (3) What are the eventual changes in structure and activity of cells that can induce changes in cell volume and/or in blood ion concentration? Do euryhaline invertebrates have specific features to deal with these problems? Some of these points have been considered in different reviews in the past few years. The reader is referred to the literature for complementary information: Gilles (1979, 1983), Gilles and PCqueux (1981, 1983), Treherne (1980, 1985), Yancey et al. (1982), and Clark (1985).
II. COPING WITH CHANGES IN MEMBRANE TENSION AND CELL VOLUME: CELL VOLUME CONTROL IN ANISOSMOTIC MEDIA A. The Volume Readjustment Responses Tissue responses to osmotic shocks have been studied in euryhaline invertebrates in vivo as well as in v i m . Many studies over the past 25-30 years have shown that the cells are not simply behaving as osmometers on variation of the osmolality of their environmental medium. They rather resume hydration values close to control levels in a slow phase of volume readjustment. Volume readjustment can be shown to occur in vivo, over long acclimation periods, both after swelling in hypoosmotic media or shrinkage in hyperosmotic ones. The time evolution of this process in vivo and in vitro is, however, somewhat different depending on the type of osmotic shock applied: hypo- or hyperosmotic. We will consider both types of response separately.
207
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
1. THE VOLUME RESPONSE TO HYPOOSMOTIC MEDIA(VRD) The in virro and in vivo responses of invertebrate tissue to hypoosmotic media have been studied in some detail, particularly in crustaceans. We will therefore concentrate on these species. Some work has also been done on other groups. The reader interested in this research is referred to the following literature: mollusks, Lange and Mostad (1967), Pierce (1971), Willmer (1978), Strange and Crowe (1979a), and Amende and Pierce ( 1980a); coelenterates, BensonRodenbough and Ross-Ellington ( 1982); polychaetes, Machin and O’Donnell (1977), and Costa et al. (1980). As shown in Fig. 1, application of hypoosmotic shock induces a biphasic response in the two types of crustacean tissue studied to date: muscle and nerve. There is first a rapid swelling which is followed by a much slower phase of volume regulatory decrease (VRD) during which cell volume may eventually resume values very close to control levels. This type of response can now be considered classical and of general occurrence since it has been demonstrated in all tissue and cell types studied, not only from euryhaline invertebrates but also from vertebrates (for reviews see Hoffmann, 1977, 1980, 1985a,b; this volume, article 5; Gilles, 1979, 1980a,b, 1983; and Cala, 1983).
B
:: 2
A
1
100
I
0
I
60 TIME (min)
I
120
FIG.I . Effect of hypoosmotic shock on the volume of isolated muscle fibers (A) and axons (B) of different crustaceans. The amplitude (71,/m2) of the shock applied is approximately equal to 1.5 for muscle fibers and 2 for axons. ( I ) Callinerres sapidus (Lang and Gainer, 1969a,b); (2) Homarus gammarus (Gainer and Grundfest, 1968); (3) Eriochrir sinensis (Gilles, 1973);(4)Callinecres sapidus (Gerard and Gilles, 1972a); (5) Carcinus maenas (KCvers ef a / . , 1979a); (6) H o m r u s gammrus (Gilles, 1973).
208
R. GILLES
It is worth noting in the experiments reported in Fig. 1 that both the maximum volume achieved and the importance of the regulation observed for the same osmotic shock are different, depending on the tissue and the species considered. In the decapod crustaceans studied, these differences in ability for cell volume regulation seem to be correlated with the ability of the species to tolerate media of different salinities. For instance, no really significant volume regulation was observed in the isolated nerve cord of the European lobster Homarus gammarus (= vulgaris), at least in the time course of the reported experiments. H . gammarus is a stenohaline species, strictly restricted to sea water. In the isolated axons of the crab Carcinus maenas, volume readjustment is far from being completed in 2 hr, while it is almost complete in the same tissue of another crab, Eriocheir sinensis. The green crab C. maenas is a fair to good euryhaline species which is rarely found in waters of salinity lower than lo%, S, while the Chinese crab E . sinensis is extremely euryhaline and can easily acclimate to fresh water or to sea water. Isolated muscle fibers of the good euryhaline blue crab Callinectes sapidus resume a volume close to control levels within a few hours, while no significant regulation can be observed in the same period of time in the muscle fibers of the stenohaline lobster H . gammarus (Fig. 1) or crayfish Astacus astacus (not shown; Reuben ef al., 1964). Similarly, no volume regulatory response is observed in the time course of acute application of hypoosmotic media to isolated muscle fibers of the poor euryhaline, blood-osmoconforming decapod Callianassa crassipes (Freel, 1982). Such differences in volume response efficiency in species showing different degrees of euryhalinity are probably of ecological importance. They could indeed account, at least partly, for differences in the ecological extension in brackish waters or in media with rapid salinity fluctuations such as estuaries. Rather comparable behavior is observed in tissues of crustaceans acclimating in vivo to media of different salinities. Decrease in blood osmolality indeed causes in vivo a biphasic response in muscle tissue quite similar to the one seen in vitro (Fig. 4). Also, differences in regulatory abilities with the ecological capabilities of the species can be demonstrated. For the same change in blood osmolality, complete regulation of muscle fiber volume is indeed achieved in about 24 hr in the very euryhaline E. sinensis and in about 70 hr in the less euryhaline C. muenus (Harris, 1976). In the stenohaline Cancer pagurus, cell volume readjustment is only partial after 15 days of acclimation to a dilute medium (Wanson et al., 1983). Another interesting feature of the volume response to hypoosmotic media in isolated tissues of crustaceans is that the maximum swelling achieved, at least for shocks of relatively large amplitude (n,/n2= l S ) , is always lower than what can be expected if the cells were to behave as perfect osmometers following the van’t Hoff equation. This is illustrated in Fig. 2, which gives the evolution of the maximum swelling achieved by isolated axons of E. sinensis and C. maenas
209
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
4
1507
100-
50
-
010
0.5 7
1.0 T
15 .
2.0
n2 l
FIG. 2. Effect of osmotic shocks of different amplitudes on the maximum volume achieved by isolated axons of Homurus gammurus (A),Corcinus muenus (O), and Eriocheir sinensis acclimated to sea water (8) or to fresh water (0). Compiled from Gilles (1973) and unpublished materials.
under different conditions of anisosmotic media. Usually, the maximum swelling T 2 ~ never exceeds 40achieved for a hypoosmotic shock of amplitude T ~ / = 50%, whereas it should be 75-80% considering the osmometric behavior and volume of osmotically inactive water of about 23%. As proposed previously (Gilles, 1980a,b), this indicates that the early response of the tissue to an important hypoosmotic shock is a swelling limitation process that should avoid the bursting out of the cells due to a too-large distension of the plasma membrane. This rapid, early swelling limitation phase is followed by a much slower one of volume readjustment during which cells may eventually resume volumes close to control levels. As is the case for volume readjustment and as already pointed out, the swelling limitation possibilities seem to be related to the ecological capabilities of the species to withstand salinity changes. Swelling limitation is indeed barely significant in the stenohaline lobster, while it is quite large in the euryhaline crabs E. sinensis, C. maenas, and C . sapidus (Fig. 1).
2. THE VOLUMERESPONSETO HYPEROSMOTIC MEDIA(VRI) At variance with what can be seen in hypoosmotic conditions, it has been impossible to date to demonstrate any significant regulatory response to hyperosmotic shocks of isolated crustacean tissue incubated in vitro. As a rule, the cells rapidly shrink and remain shrunk at least in the time course of the experiments performed, which usually does not exceed 3 to 4 hr [Fig. 3: muscle fibers, Callinectes sapidus (Lang and Gainer, 1969a,b) and Callianassa californiensis (Freel, 1982); axons, Callinectes sapidus (GCrard and Gilles, 1972a), Eriocheir sinensis (Gilles, 1973, 1974a), and Carcinus maenas (KCvers et a l . , 1979a)j. On a longer time scale (6-8 hr), Leader and Bedford (1978) reported only a slight,
210
R. GILLES
v2 XlOO v1
T I 0 0
60 120 180 TIME (min)
240
1 300
FIG. 3. Effect of a hyperosmotic shock on the cell volume of isolated axons of Callinecres sapidus ( x ) (Gerard and Gilles, 1972a). Eriocheir sinensis (0)(Gilles, 1973). Carcinus maenas (D) (Kkvers er ul., 1979a); and of isolated muscle fibers of Callinecres sapidus (0) (Lang and Gainer, 1969a.b). The amplitude of the shocks applied are 0.72 and 0.75, respectively, for the muscle and axon of C. supidus and 0.5 for the axons of the two other species. At the times marked by the arrows, the preparations are placed back in their control saline.
partial recovery of isolated muscle fibers in the crab Hemigrapsus edwardsi. This lack of a significant, rapid response of crustacean tissues to hyperosmotic shocks is in agreement with the results of most studies done to date, whatever the tissue or cell type considered. To my knowledge, a volume regulatory increase (VRI) in the usual salines has so far only been demonstrated in two vertebrate cell types: the gallbladder of Necfurus maculosus (Fisher et al., 1981; Persson and Spring, 1982) and the red blood cells of Amphiuma means (Cala, 1980). In both cases, cell response appears to be directly dependent on the presence of bicarbonate ions in the saline. The eventual dependency of VRI on HC0,- has never been really studied in invertebrate cells to date. No bicarbonate was present in the saline used in the experiments reported above, which showed partial recovery of H. edwardsi muscle fibers. Experiments done on isolated axons of C. maenas in the presence of bicarbonate ions failed to show any volume regulatory increase (Ktvers e f al., 1979b). That a VRI process is present in invertebrate tissues is, however, clearly seen in in vivo experiments. In such studies, indeed, tissues always resume hydration values close to control levels after acclimation of the animals to concentrated media. Though many references exist on the topic (Gilles, 1975, for review), the complete time course of tissue hydration changes occurring in vivo on changes in blood osmolality in both directions has only been studied in three cases: the coelenterate Bunodosoma cavernata (Benson-Rodenbough and Ross Ellington, 1982), the crustacean Eriocheir sinensis (Gilles, 1977; Fig. 4), and the mollusk Rangia cuneafa (Otto and Pierce, 1981). In these cases, the VRD process is much faster than the VRI one, which takes at least several days to be completed.
21 1
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
BLOOD : 59L '- 18 59L'-18 MUSCLE 586tlZ
L
T BLOOD.; 1017 BLOOD 1017:t Z L MUSCLE.1022: 36
70
'
I
1
I
I
I
0 1 L 9 15 ACCLIMATION TIME (DAYS) FIG.4 . Evolution of the water content of the leg muscle of Eriocheir sinensis during acclimation from sea water to fresh water (0) and from fresh water to sea water (0).After Gilles (1977), modified.
Failure to demonstrate any VRI process in in vitro preparations could thus be simply related to the slowness of the process in invertebrate tissue. It should, however, be noted that while VRI cannot be demonstrated in a variety of tissues and cell types shrunk in hyperosmotic salines, whatever the species (vertebrate or invertebrate) or the presence of bicaronate ions in the saline, such a regulatory response is always seen in cells osmotically shrunk by resuspension in control saline after VRD in a hypoosmotic one (see for instance Cala, 1977; Ussing, 1982; Hoffmann er al., 1983). This striking difference in volume response to two different types of hyperosmotic exposures has so far remained unexplained. It indicates that the problem is probably more complex than previously thought. More data will be needed to shed some light on this interesting point. We now come to the question of what mechanisms are implicated in volume readjustment responses.
B. Mechanisms of the Volume Readjustment Responses 1 . THEDIFFERENT COMFQNENTS INVOLVED
A priori, cell volume control must result from a balance between three major components (Fig. 5): (1) the amount of osmotically active water, (2) the amount of osmotically active solute, and (3) the elastic properties of cells and tissues and their mechanical resistance to changes in volume.
a. Water Physical Stare. There is in the literature an increasing amount of data showing that part of the intracellular water is in a physical state different from that of the bulk water inside and outside the cell. Unfortunately, nothing is
212
R. GILLES
BINDING FLUXES
METABOLISM BINDING FLUXES
aSTRUCTURAL RESISTANCE AND ELASTICITYFIG. 5. The components and mechanisms that could a priori be concerned with cell volume maintenance.
known to date on the eventual role of changes in the physical state of water on the volume regulation response. Information in this area would be welcome. b. Mechanical Resistance and Elasticity. To date, this problem has been the subject of only a few investigations, essentially on mammalian tissue. In kidney tubules treated with ouabain, removal of the basement membrane by collagenase disturbs the volume response to hypoosmotic media (Linshaw and Grantham, 1980). In Necturus maculosus gallbladder, RVD is inhibited by cytochalasin B, an agent causing depolymerization of the microfilaments intracellular network. These experiments point to the role of both extracellular and intracellular structures in the response of tissues to hypoosmotic media. Unfortunately, not much is known on the eventual role of such structures in invertebrates. Specific structures that might exert constraint forces restricting the development of membrane tension have been described by Treherne and colleagues at the level of the axons of the serpulid worm Ficoparamus (= Mercierella) enigmatica (Treherne, 1980, 1985, for reviews). As already stated, this species, one of the most extreme osmoconformers known, can live in media of largely fluctuating salinities and can withstand fairly abrupt changes in blood osmolality between 80 and 2300 mOsm/liter (Skaer, 1974; Benson and Treherne, 1978a,b). Such changes induce the development of enormous swelling pressures and tensions that will affect all membrane activities, particularly spike generation which is of importance for the immediate survival of the animal. Skaer et al. (1978) have described in the axons of that species, close-spaced, hemidesmosome-like structures that could provide some sort of support to the plasma membrane. These structures, associated with the axon membrane and connected with a network of intracellular neurofilaments, could indeed constitute rather rigid architectures opposing membrane distension. In addition, as proposed by Skaer er al. (1978), their close-spaced distribution should, in case of swelling, reduce the radius of curvature of the unrestrained portions of the axonal membranes, thus reducing tension at their level. This is a rather unique example of specific structure directly implicated in restricting the development of membrane tension.
213
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
A few years ago, we tested the possible role of the microtubules in RVD of isolated axons of Carcinus maenas by using vinblastin sulfate (Ktvers et a l . , 1979b). This compound largely impairs volume readjustment, but remains without effect on swelling limitation. These effects are, however, accompanied by large modifications in the intracellular level of Na , K , and C1- . They could, thus, be mediated through changes in the concentration of osmotically active solutes rather than by disruption of a macromolecular architecture showing elastic properties. More recently Gilles et al. (1986) have shown that cytochalasin B completely blocks the RVD seen in isolated axons of Carcinus maenas withstanding an hypoosmotic shock. This block concerns not only the volume readjustment phase but also the early swelling limitation phase (Fig. 6). Interestingly, the compound is without significant effect on the axonal volume when the tissue is incubated in control conditions. It is also without significant effect on the large decrease in K + level concomitant to volume regulation (see the next section). This shows that the decrease in the level of intracellular osmotic effectors normally accompanying the volume response (refer to the next section) cannot account for volume regulation unless elastic properties of the cell remain unaffected. In view of this, the decrease in intracellular osmotic effectors would induce a decrease in swelling pressure due to the application of the hypoosmotic medium. This in turn would allow a volume regulation essentially related to the elasticity of cellular structures. Obviously, much work remains to be done before a complete picture of the nature and relative importance of different elastic components in volume regula+
0 '
+
I I
0
I
I
30 60 TIME (min)
I
1
90
120
FIG. 6 . Effect of cytochalasin B on the volume of isolated axons of Carcinus maenas submitted to a hypoosmotic shock (wllm2 = 2). (0. A): Control hypoosmotic saline. ( 0 ,0): Hypoosmotic saline with cytochalasin B . The given data (181.07 and 141.42) refer to the maximum swelling achieved with and without cytochalasin B . [Original. from data of Gilles ef a / . (1986).]
214
R. GILLES
tion can be produced. The too few results available to date nevertheless point to an essential role of at least one such structure in the volume response following osmotically induced swelling. The microfilament network indeed seems to play a major role in volume readjustment at least once the swelling pressure induced by the application of an osmotic imbalance has been reduced by adjustment of the concentration of the different intracellular solutes. We will now consider this last problem in greater detail. c. Control ofsolute Level. Early experiments on isolated muscle fibers of the blue crab Callinectes sapidus (Lang and Gainer, 1969a,b) and on isolated axons of the Chinese crab Eriocheir sinensis (Gilles, 1973) showed that the cells shrink to a volume smaller than control level when the tissues are placed back in a control saline after volume readjustment in a hypoosmotic medium. This kind of experiment, shown in Fig. 7, leads to some interesting conclusions: (1) RVD implicates a decrease in the amount of intracellular osmotic effectors since the volume shrinks to values lower than control levels when the tissues are placed back in their control salines. (2) This control of intracellular solutes level is not primarily related to interactions between the cells and some organic constituent of the external medium as, for instance, could be the case in a double-Donnan equilibrium system. As a matter of fact, the salines used contain only inorganic ions. (3) For the same reason, neither is a hormonal control primarily implicated. This brings us to the question of what osmotic effectors are implicated and what the processes at work are in the control of their concentration.
-xvv12
100
0
60
120
180 240 TIME (min)
300
360
FIG. 7. Evolution of the volume of isolated muscle fibers of Callinecres sapidus ( 1 ) and of isolated axons of Eriocheir sinensis (2) submitted to hypoosmotic shocks and then returned to control saline (at the time marked by the arrow).The osmotic shock applied (a,/az)is approximately equal to I .5 for the muscle fibers (Lang and Gainer, 1969a.b) and to 2 for the isolated axons (Gilles, 1973).
21 5
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
2. THEINTRACELLULAR OSMOTICEFFECTORS A variety of studies have been devoted over the past 25 years to the nature of the different effectors involved in the volume control of invertebrate cells. Both monovalent inorganic ions and a large variety of amino compounds appear to be implicated. The respective parts played by these two groups of osmotic effectors as well as the nature of the amino compounds involved appear to vary from species to species and, in many cases, from tissue to tissue in the same species. We will now attempt to focus on the mechanisms implicated in the control of both groups of substances. The literature dealing with the nature of the organic osmotic effectors has been reviewed from time to time during the past decade (Gilles, 1975, 1979; Yancey et al., 1982). The reader is referred to these papers as well as to those listed on p. 221-222 for detailed accounts.
a . The Inorganic Ions. When isolated axons of the green crab Carcinus maenas are submitted to hypoosmotic conditions, there is a rapid decrease to new
steady state values of the intracellular content of K (Fig. 8). Transient changes in Na+ and C1- levels can also be observed. Most of the modifications in Na+ and CI- can, however, be accounted for by changes in cellular hydration. This is not the case for the large decrease in K , which appears to be directly related to +
+
OI,
I
,
0515 30
01, I
,
0515 30
1
I
1
60
120
180
,
60
I
1
120
180
80
60 40
C
20
'
0
0515 30
60 120 TIME (min)
180
FIG.8 . Effect of a hypoosmotic shock on the intracellular content of K (A), N a + (B), and CI(C) of Curcinus menus isolated axons. (A): Control conditions; (0, 0): hypoosmotic conditions; (0): results corrected for the measured changes in intracellular hydration. [Compiled from results of Kkvers er 01. (1979b).] +
R. GILLES
216
the volume control process. A rapid, though much smaller (only lo%), decrease in K has also been shown to occur in isolated coelomocytes of the blood worm Glycera dibranchiata subjected in vitro to hypoosmotic shock (Costa and Pierce, 1982). i. Control of K+ level. Thus, K+ appears to be an important osmotic effector at least in some invertebrates tissues. It is, on the other hand, the major solute implicated in volume regulation in most osmotically swollen vertebrates cells. Its role as well the mechanisms controlling its intracellular level in vertebrates cells have been the subject of a variety of studies in recent years (see the fifth article in this volume and also, for example, Kregenow, 1981; Cala, 1983, 1985; Hoffmann, 1985a,b; Grinstein et al., 1985; Lauf, 1985; Law, 1985; Ellory et al., 1985a,b, for reviews). The studies on inorganic ions in invertebrate cells unfortunately remain extremely sparse, and not much is known about the mechanism that could account for the control of the K + decrease. First, it should be stated that the relative importance of the role of K + , with respect to other osmotic effectors, appears to be different depending on the phase of volume regulation we are dealing with (swelling limitation or volume readjustment). As will be seen in the next section, the concentration of amino compounds decreases considerably with the decreasing level of K + in C. maenas axons withstanding hypoosmotic media. The importance of these changes can be directly related to the evolution of the volume regulation process. The decrease in these osmotic effectors is indeed important and rapid during the early, large swelling limitation phase. It is much slower during volume readjustment. Further, the changes in K+ concentration occurring during that last phase are barely significant. It thus seems that K+ loss is essentially important during the initial rapid swelling limitation. Subsequent slow volume readjustment would be more directly related to the regulation of the concentration of different organic osmotic effectors. This difference in the role played by K and the amino compounds we originally described in the course of in v i m experiments on isolated axons (Ktvers et al., 1979a,b) has been recently confirmed in in vivo studies. Moran and Pierce (1984) indeed showed that in the rock crab Cancer irroratus subjected to hypoosmotic media there is an early loss of K from the muscle tissue which seems to limit its initial rate of swelling. The slow volume recovery which occurs later seems to be mainly associated with control of amino acid levels. In invertebrates, the K + loss thus seems to be essentially associated with the swelling limitation process. It further seems related to transient changes in plasma membrane permeability. Application of hypoosmotic conditions indeed appears to induce a large increase in K + efflux from the isolated axons of C . maenus (KCvers et al., 1981). This increase is, however, only transient, taking place essentially during the rapid swelling limitation phase. It is followed by a phase during which the permeability to K + seems to be decreased as indicated +
+
+
217
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
by the increase in the half-renewal time calculated from 43K efflux kinetics data. To our knowledge, this is the only study done to date following this type of approach. Changes in membrane potential that are also indicative of changes in permeability have been reported by different authors. Usually, rapid hyperpolarization followed by slower depolarization and reduction in excitability occurs after application of hypoosmotic conditions to invertebrate tissues. This has been shown in axons of Maia squinado (Pichon and Treherne, 1976), Sabella penicillus (Treherne and Pichon, 1978), and Mercierella enigrnata (Benson and Treherne, 1978a,b; Skaer et a f . , 1978). This has also been shown in the cerebrovisceral connectives of Mytilus edulis (Willmer, 1978), in the cell bodies of neurons from the visceral ganglion of Mya arenaria (Beres and Pierce, 1981), or in the follower cells of the cardiac ganglion of the horseshoe crab Lirnulus polyphernus (Prior and Pierce, 1981). Many of these studies have been conducted over short periods of time. Recordings made over longer periods in L . polyphernus follower cells, M . arenaria cell bodies, or L i m a species salivary burstor neuron (Prior, 1981) indicated that the depolarization is only transient, as are the efflux changes observed in C . maenas axons. Most probably these variations in both permeability and membrane potential are related to the volume regulation process and end when the tissues have resumed volumes close to control levels. In this respect, it is interesting to consider that none of the electrical changes occurs if only the ion concentration is reduced, while the osmolality is maintained with sucrose (Beres and Pierce, 1981; Prior, 1981; Prior and Pierce, 1981). The changes in K efflux and in membrane potential are indicative of changes in one or several leak pathways for K . In this context, it is worth noting that the increase in efflux during hypoosmotic shock is not specific for K . A similar result has been obtained with amino acids in crustacean, molluscan, and worm tissues (see the following section). It has also been reported for other nonelectrolytes (urea and erythntol) in isolated rat portal veins preliminarily loaded with these compounds (Jonsson, 1971). As previously proposed (Gilles, 1974a, 1975, 1978), such results are indicative of nonselective changes in permeability that could be related to the osmotic swelling per se. In view of this, swelling would induce modifications in plasma membrane conformation that would in turn induce changes in the permeability to a variety of low-molecular-weight intracellular compounds. These compounds would then leave the cells more or less rapidly, depending on the concentration gradients, new values of permeability coefficients, and so on. This type of general unspecific response is of course not exclusive of others, more specific, implicating, for instance, control of specific channels opening. As a matter of fact, and at least so far as K + is concerned, the situation appears to be far more complex than what could be expected on the basis of a simple increase in general plasma membrane permeability. Experiments done on isolated axons of C. rnaenas incubated in salines +
+
+
218
R. GILLES
in which the osmolality and/or the content of Na and/or K + was varied lead indeed to the conclusion that the reduced level of intracellular K + achieved represents a balance between the necessity to contribute an osmotic equilibrium and to maintain a K + gradient across the plasma membrane compatible with bioelectric phenomena (KCvers et al., 1981). In this context, studies on Ca2+dependent K + effluxes and channels would be of interest. Several studies point to a possible important role of such channels in volume control of different types of vertebrate cultured cells (for recent reviews see Hoffmann, 1985a,b; Grinstein et al., 1985). The fact that Ca2+-free media impair cell volume control in isolated axons of the crab Callinectes sapidus (GCrard, 1975) and in coelomocytes of the polychaete worm Glycera dibranchiata (Costa and Pierce, 1982) could be taken as indicative of a similar mechanism. Other mechanisms could also be at work. A variety of them have been described in vertebrate cells (thiol sensitive, Lauf, 1985; pressure sensitive, Ptqueux et al., 1980; Zimmermann et al., 1980; Ellory et al., 1985b). They could also play a role in the control of K + levels in invertebrate tissues. The decrease in K + seen in vertebrate and invertebrate cells withstanding hypoosmotic conditions leaves us with a problem of cation-anion balance. In most studies, indeed, the eventual C1- changes have not been measured. The only study dealing with CI- in invertebrate cells is the one by KCvers et al. (1979a,b, 1981) reported above. In this case, it is clear that C1- changes do not correspond to those of K . A similar lack of correspondence between K and C1- changes has also been reported for different vertebrate cells such as mouse leukemic cells (Roti-Roti and Rothstein, 1973) and flounder red blood cells (Cala, 1977). At variance with these results, C1- appears to be extruded concomitantly with K in other vertebrate cell types (duck or Amphiuma red cells) (Kregenow, 1971; Siebens and Kregenow, 1978; Cala, 1980), frog skin (Ussing, 1982), and Ehrlich ascites mouse tumor cells (Hoffmann and Hendil, 1976). Further studies are needed to shed more light on this problem. As we will see in the next section, the concentration of different amino compounds decreases considerably in invertebrate cells in hypoosmotic conditions. Some of them, such as glutamic acid and aspartic acid, are most probably implicated in the cation-anion balance, the decrease in their intracellular level accounting, at least partly, for the discrepancy observed between the changes in K + and CI- concentrations. HCO, - , and possibly also OH - exchanging for C1- , could also play some part in the process. In this context, the relation between intracellular pH and volume regulation would be worth studying in invertebrate cells. ii. Control of Nu+ and C1- levels. As shown in Fig. 8, Na+ and C1- do not take part in the volume control process in C . maenas axons. There is even a slight increase in the amount of these ions during the volume readjustment phase. Such an increase is contradictory to the expected change in intracellular fluid osmolality and might account for the fact that volume regulation in C. maenas +
+
+
VOLUME REGULATION IN EURYHALJNE INVERTEBRATES
219
axons is slower than that reported in axons of C . sapidus or E . sinensis (Fig. 1). As already stated, this difference in volume control efficiency in tissues of euryhaline invertebrates appears to have interesting ecological correlates. A comparative study of the relative changes in permeability to Na+ and K + in tissues of different euryhaline species should be of interest in this context. Not much is known about the mechanisms implicated in the control of the Na+ level. Increase in Na+ concentration could be associated, at least partly, with an increase in permeability, leading to an increase in Na+ passive influx. This is consistent with the results of KCvers et al. (1979b) that show that application of hypoosmotic conditions induces a decrease in the 22Nahalf-renewal time in the intracellular compartment of C. maenas axons, the Na+ pumping mechanism of which has been blocked with ouabain. Increases in Na+ permeability have also been reported to occur in some vertebrate cells on application of hypoosmotic conditions (Doljanski et a / ., 1974; Schmidt-Nielson, 1975; for review see Hoffmann, 1977; Rorive and Gilles, 1979). The possible role of active transport system(s) that might be implicated in Na+ level maintenance during volume regulation has been little studied to date. It has been repeatedly shown during the past decade that ouabain remains without significant effects on volume regulation. This is true for crustacean tissues (Lang and Gainer, 1969a,b; GCrard, 1975) as well as for vertebrate tissues and cells (for reviews see Hoffmann, 1977; Rorive and Gilles. 1979). This can be taken as evidence that the Na /K exchange pump is not implicated in the process. If, in most cases, K + is the major inorganic ion lost during volume regulation in osmotically swollen cells, it seems that in some cases Na+ and C1- could be used as major inorganic osmotic effectors. Clear examples of this are found in mammalian kidney cortex cells (Hughes and MacKnight, 1976; Paillard et a l . , 1979; See1 et al.. 1980; Gilles et al.. 1983) and dog red blood cells (Parker and Hoffman, 1976; Parker, 1977). In invertebrates, it seems that Na+ and C1rather than K + are involved, with some amino compounds, in volume regulation in the isolated heart of the horseshoe crab Limulus polyphemus (Warren and Pierce, 1982). To my knowledge, this is the only example of the use of Na+ in volume control in invertebrate cells. In mammalian kidney cells, Na+ and C1are essentially implicated in a swelling limitation process. The large changes in Na observed during that phase can be directly related to the changes in the Na electrochemical gradient occurring when the hypoosmotic medium is applied. As a matter of fact, the calculated ratios of Na,/Na, remain in the same range in isosmotic conditions and after swelling limitation (Gilles et a l . , 1983). Unfortunately, nothing is known about the mechanisms at work in the control of NaCl concentration in L . polyphemus heart cells. The changes in inorganic ion concentration described to date are related to the response of cells to acute hypoosmotic stress conditions. It should be pointed out that further readjustment, probably involving further modifications in transport, +
+
+
+
220
R. GILLES
must occur in vivo and possibly also in vitro on a longer time scale. For instance, though K + is lost from coelomocytes of the osmoconforming worm Glycera dibranchiuta during acute exposure to hypoosmotic saline, there is no significant change in the intracellular concentration of this ion in the coelomocytes of animals acclimated to different salinities (Costa and Pierce, 1982). Similarly, the intracellular content of K + in both axons and muscle fibers of the blue crab Callinectes sapidus is not significantly different in animals acclimated to sea water or to 50% sea water (GCrard and Gilles, 1972b); nor is it different in the axons of the Chinese crab Eriocheir sinensis acclimated to sea water or to fresh water (Gilles, unpublished). From an ecophysiological standpoint, it is of interest to note that such readjustments in inorganic ions are far from being a general rule. In a variety of osmoconforming, rather stenohaline, crustaceans (Callianassa californianus, Cancer antennarius, and Emerita analoga, Freel, 1978a,b, 1982; Cancer pagurus, Wanson et al., 1983), changes in the muscle content of Na+ , K ,and C1- can still be recorded after long periods of acclimation to a dilute medium. These changes, however, can be accounted for by changes in tissue hydration and are not to be considered an integral part of the volume control process. As stated previously, it seems that these stenohaline species have extremely limited volume regulation capabilities and their tissues remain largely swollen even after several days at low salinity. Briefly, the results discussed above point to a role for inorganic ions, essentially K , in the short-term response of cells of euryhaline invertebrates to osmotic swelling. A decrease in the concentration of this ion is implicated chiefly in an early, rapid swelling limitation phase, while the subsequent, slower volume readjustment process would be more directly related to changes in the concentration of organic osmotic effectors. The changes in K + level can be associated with transient changes in passive fluxes through leak pathway(s). This type of response seems to be absent in tissues of osmoconforming stenohaline species in which a volume regulation process is either nonexistent or far too slow to be demonstrated in the course of the short-term experiment performed. We will now consider what organic osmotic effectors are implicated and what the mechanisms could be in the control of their concentration during the volume regulation responses. +
+
b. The Organic Osmotic Effectors. One hundred years ago, in a study of osmoregulatory mechanisms, U o n Fredericq, at the time head of the newly built laboratory of physiology of the Liege University, considered that “. . . In the invertebrate species, the independency towards the external medium (confered by these mechanisms) is only relative. It is interesting in this framework to study the effect that a sojourn in a more or less salty medium can exert on the salts composition of blood and tissues.” (Translated from French, Frederic, 1885). Starting such a study on species he was collecting in the marine stations of
22 1
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
Roscoff in France and of Napoli in Italy, he observed that the intracellular concentrations of ions were inadequate, especially in invertebrates and low vertebrates such as elasmobranchs, to explain the osmotic equilibrium existing between cells and their environmental medium, a result he summarized in sketch form in a paper in 1901 (Fig. 9). Some 65 years later, in the same institute, Florkin, following in the steps of his old master, initiated a study on the organic osmotic effectors, comparing concentrations of different amino compounds in euryhaline species acclimated to sea water or to dilute media. The first report of this study was published in 1955 by Duchateau and Florkin. It showed that in the Chinese crab Eriocheir sinensis, the total concentration of free amino acids in the muscle tissue was approximately doubled on acclimation from fresh water to sea water. Since there were only slight modifications of tissue hydration during the acclimation, Florkin (1956, 1962) proposed “to consider the variation in the amino acid component resulting from a change in the medium concentration, as exerting an intracellular osmotic regulation. i . Nature and importance of organic solutes in volume regulation. Since the original discovery of Florkin and co-workers, much work has been devoted to the chemical nature and to the relative importance of the amino compounds in cell volume maintenance in invertebrate tissues. These studies, largely developed during the past two decades, have been surveyed from time to time (Schoffeniels and Gilles, 1970, 1972; Gilles, 1974a, 1975, 1979; Gilles and PCqueux, 1983). Thus I will only briefly summarize the main conclusions drawn, referring the reader interested in more detail to these reviews and to the original papers listed below. In short, all the studies done to date have shown the following: ”
1. Amino compounds play a part in cell volume regulation in all the invertebrate species studied so far: Coelenterates: Webb et a l . , 1972; Pierce and Minasian, 1974; Shick, 1976;
Eau
CE nicr
A
Tissus
B
C
B B C FIG 9 The dismbution of osmotic effectors (salts and organic substances) In sea water, blood, and tissues of invertebrates. elasmobranchs, and bony fish [After Fredencq (1901). modified ]
222
R. GILLES
Howard and Kasschau, 1980; Benson-Rodenbough and Ross-Ellington, 1982. Annelids: Jeuniaux et al., 1961a; Duchateau et al., 1961; Clark, 1968; Freel et al., 1973; Costa et al., 1980. Siponcala: Virkar, 1966. Mollusks: Potts, 1958; Allen, 1961; Lange, 1963; Bricteux-Grkgoire et al., 1964a-c; Lynch and Wood, 1966; Peterson and Duerr, 1969; Virkar and Webb, 1970; Du Paul and Webb, 1970, 1974; Pierce, 1971; Gilles, 1972a; Kasschau, 1975; Hoyaux et al., 1976; Fyhn, 1976; Roesijadi et al., 1976; Baginski and Pierce, 1977; Strange and Crowe, 1979b; Henry et al., 1980; Amende and Pierce, 1980b; Powell et al., 1982; Pierce er al., 1983. Crustaceans: Duchateau and Florkin, 1955; Shaw, 1958; Duchateau et al., 1959; Duchateau-Bosson and Florkin, 1961; Jeuniaux et al., 1961b; Bricteux-Grkgoire et al., 1961, 1962; Vincent-Marique and Gilles, 1970a,b; Gilles, 1970; Gkrard and Gilles, 1972b; Weber and Van Marrewijk, 1972; Siebers et al., 1972; Bedford and Leader, 1977; Freel, 1978a,b; Farmer and Reeve, 1978; Burton and Feldman, 1982; Wanson et al., 1983; Moran and Pierce, 1984. Arachnornorphs: Bricteux-Grkgoire et al., 1963, 1966; Warren and Pierce, 1982. 2. Participation of amino compounds in volume regulation is not restricted to invertebrate tissues. Changes in the concentration of some of them (particularly amino acids) have been reported to occur in tissues of a variety of other organisms submitted in vivo or in vitro to osmotic challenges (vertebrates, plants, unicellulars): Bacteria-protozoa: Kaneshiro et al., 1969; Stoner and Dunham, 1970; Measures, 1975; MacKay et al., 1984. Algae and plants: Schobert, 1974, 1980; Liu and Hellebust, 1976: Ming and Hellebust, 1976; Gilles and Pkqueux, 1977; Storey and Wyn Jones, 1977; Tal et al., 1979; Reynoso and de Gamboa, 1982; Ahmad and Helleburt, 1985. Fish: Lange and Fugelli, 1965; Cholette et al., 1970; Huggins and Colley, 1971; Lasserre and Gilles, 1971; Cholette and Gagnon, 1973; Vislie and Fugelli, 1975; Ahokas and Duerr, 1975; Forster and Goldstein, 1976; Boyd et al., 1977; Pang et al., 1977; Ahokas and Sorg, 1977; Kaushik et al., 1977; Forster et al., 1978; Forster and Hannafin, 1980; Jurss, 1980; Vislie, 1980; see also Goldstein and Kleinzeller in this volume. Batracians: Gordon 1965; Baxter and Ortiz, 1966; Gordon and Tucker, 1968; Colley er al., 1972, 1974; Balinsky et al., 1972; Licht et al., 1975. Reptiles: Gilles-Baillien, 1973. Mammals: Hoffmann and Hendil, 1976; Thurston et al., 1980, 1981; Gilles er al., 1983.
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
223
As already pointed out in previous reviews (Gilles, 1974a, 1975, 1979), participation of amino compounds in the overall volume regulation process is a general phenomenon, occumng in all cell types studied so far. 3. In most cases, amino compounds obviously play a prominent role as osmotic effectors. In some systems, however, particularly mammalian cells, their participation to the osmolality of the internal medium is rather limited, indicating that their role may not be restricted to “broad osmotic effects.” In this context, they most probably have a part to play in the modulation of the effects that the changes in the level of the inorganic ions Na , K , and C1- occumng during volume readjustment exert on the structure and activity of macromolecular architectures inside the cells. This problem will be considered in more detail in the last section of this article. 4. The part played, as well as the nature of the major amino compounds involved in cell volume control, varies from species to species and often from tissue to tissue. 5. The prominent amino compounds determined so far are amino acids, the amine taurine, and a few quarternary ammonium derivatives such as trimethylamine oxide, glycine betaine, or isethionic acid. The occurrence of proline betaine has been reported once (Pierce et a l . , 1984). It also seems that small peptides can play a part in the process (Kevers et al., 1979a). Most of the studies done so far have dealt with the amino acids and the amine taurine. With the exception of arginine, amino acids showing the largest variations are always nonessential ones: glycine, proline, alanine, serine, aspartic acid, and glutamic acid. Only slight changes can be recorded at the level of the others. In most marine invertebrates studied, the amino acid concentration ranges between 150 and 350 mMlkg tissue wet wt. It can be much larger; it can also be much smaller (Gilles, 1975). This indicates that other organic compounds, yet undetermined, also participate in the control of the intracellular fluid osmolality . +
+
We will now consider the mechanisms implicated in the control of the amino acid levels. ii. Mechanisms of control of the amino acid intracellular level. Early experiments by Lang and Gainer (1969a,b) and Gilles and Schoffeniels (1969) showed modifications of the NPS or amino acid content that could not be accounted for by volume changes in isolated muscle fibers of the blue crab C. sapidus and in isolated axons of the Chinese crab E. sinensis. The time course evolution of such changes was studied later by Kevers et al. (1979a,b) in isolated axons of the green crab C. maenas. These results (Fig. 10) show that modification in NPS level are associated with the two different phases of the volume control process. A rapid, early decrease in amount is concomitant with the rapid, early swelling limitation, while a slow, subsequent modification relates to the slow volume readjustment process. Increased release of NPS was also reported in an isolated mollusk tissue by Pierce and Greenberg (1972, 1973).
224
R. GILLES
K
=
Ail\ o;
$0
I
I
120
180
TIME (min)
FIG.10. Effect of a hypoosmotic shock on the intracellular content of ninhydrin-positive sub-
0): hypoosmotic condistances of Carcinus m e n u s isolated axons. (A): Control conditions; (0, results corrected from the measured changes in intracellular hydration. [After Ktvers et tions; (0): al. (1979a), modified.] These early experiments, done on isolated tissues incubated in artificial salines containing only inorganic ions, indicated that the mechanism at work in the adjustment of the amino acid pool is intracellular and not primarily dependent on some interactions between tissues (hormonal processes, etc.). In 1969, we showed that in isolated axons of E. sinensis, the increase in amino acids recorded in the incubation saline after application of a hypoosmotic shock far from accounts for the total changes measured in the tissue (Gilles and Schoffeniels, 1969). These experiments were to our knowledge the first to indicate that though changes in fluxes play an important part in the control of the amino acid pool, they cannot explain the overall concentration variation; therefore, other mechanisms also had to be considered. In order to assess more accurately what could be the major mechanisms implicated, we initiated a study of the incorporation of I4C from labeled glucose in isolated axons of the blue crab C. supidus submitted to hypo- or to hyperosmotic conditions (Gilles and Gtrard, 1974). In these axons, application of a hypoosmotic shock leads to a decrease in the concentration of all the measured amino acids. At the same time, there is an increase in the specific activity of the amino acids that are labeled from [U-14C]glucose(Fig. 11). This increase in specific activity may be due to an increased synthesis from glucose. It may also be due to a rapid removal of amino acids from the metabolic pool at the beginning of the experiment. In this case, synthesis from glucose would occur in a pool of smaller size, thus leading to the increase in specific activity. Such a mechanism seems more likely than an increased synthesis. First, it can account for the large drop in concentration occumng during the early swelling phase, immediately after the application of the osmotic shock (see Fig. 1 l), much better than any metabolic process implicating increased synthesis. Further, an increase in synthesis would appear rather unlikely in a situation of decreasing concentration of amino acids. During hyperosmotic shock, the concentration of the amino acids increases while there is no significant change in the specific activity of those labeled from
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
225
FIG. 1 1 , Effect of (A) hypo- and (B) hyperosmotic shocks on the concentration (open bars) and specific activity (shadowed bars) of amino acids labeled from [U-14C]glucose in isolated axons of Callinecres sapidus. ASP*, Aspartic acid plus an undetermined compound. First bar: control conditions; second bar: experimental conditions. [After data from Gilles and Gtrard (1974).]
radioactive glucose. Such results point once more to a change in the output of amino acids from the pool rather than to a change in input. Any change in input should indeed have induced an increase in specific activity. In this framework, it is also interesting to note that no labeling can be found on proline and glycine, although they are nonessential amino acids, and that their concentration largely increases. This is one more argument to consider, that de novo synthesis is not primarily implicated in the control of the amino acid pool. Some years ago, Baginski and Pierce (1978) speculated on a possible increase in amino acid synthesis in isolated ventricles of the mollusk Modiofus demissus submitted to a hyperosmotic shock. Their results indeed showed an increased incorporation of 14C from glucose in alanine. There is, however, at the same time a decrease in radioactivity on glutamate and aspartate. Further, the results only give radioactivity measurements and not specific activity values. Since the amino acid pool size is changing, it is impossible to make any clear statement about an eventual participation of a de novo synthesis from the given data. Control of the intracellular concentration of the amino acids thus appears to be primarily achieved by modulation of the output from the metabolic pool. Release from that pool can be achieved either by degradation, efflux from the cell, synthesis of proteins or of other macromolecular compounds from the hydrocarbon skeletons left after deamination (Fig. 12). As discussed elsewhere (Gilles, 1978, 1979, 1980a,b), modifications in protein metabolism are unlikely to be an important process of control of the amino
226
R. GILLES
T I
OX IDATION
SYNTHESIS
FIG. 12. The mechanisms that could a priori be implicated in the control of the pool of free amino acids.
acids level in short-term experiments on isolated tissues. On the other hand, data obtained from experiments done in vivo on longer time scales remain far too scarce, not demonstrative, and, in some respects, contradictory. A clear-cut demonstration of an eventual specific role of protein metabolism is therefore still awaited. Similarly, there is actually no clear information as to an eventual participation of an interaction between amino acids and “storage” macromolecular compounds utilizing their hydrocarbon skeleton. Changes in fluxes of amino acids in and/or out of the cells have been indicated for a long time by experiments showing increased release of NPS or specific amino acids from isolated .tissues submitted to hypoosmotic conditions (Gilles and Schoffeniels, 1969; Lang and Gainer, 1969a,b; Gtrard and Gilles, 1972a; Pierce and Greenberg, 1972, 1973; GCrard, 1975; Costa et al., 1980; Amende and Pierce, 1980b). To our knowledge, however, flux kinetics studies have only been achieved on isolated axons of the blue crab Callinectes sapidus in the case of one amino acid: alanine (Gerard and Gilles, 1972a; GCrard, 1975; Gilles, 1978). Some of the results obtained are summarized in Fig. 13. Hypoosmotic shock appears to induce a large increase in alanine efflux and a decrease in the active component of the influx. In hyperosmotic conditions, on the other hand, there is no significant change in efflux while an increase in the active uptake of alanine can be recorded. As previously discussed, the changes in efflux could be related to changes in membrane permeability characteristics due to a swellinginduced distension. The modifications induced in the active transport are probably related to the changes in the Na+ concentration of the external medium imposed by the application of osmotic shocks. As shown by GCrard (1975), the uphill transport of alanine indeed appears to be Na dependent. In such conditions, the decrease in Na+ external content caused by application of the hypo-
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
227
HYPO HYPER FIG. 13. Effect of hypo- and hyperosmotic shocks on the fluxes of alanine from isolated axons of Callinecfes sapidus. Open bars: control; shadowed bars: experimental conditions. [After data from Gerard (1975) and Gilles (1978).]
osmotic medium will induce a decrease in alanine influx. On the other hand, the increased Na+ content of the hyperosmotic saline will induce an increase in alanine transport activity. That such changes in amino acids fluxes also occur in vivo in intact animals acclimating to salinity changes is demonstrated by the transient increase in blood amino acids shown to occur in a variety of species during acclimation to diluted media (crustaceans, Vincent-Marique and Gilles, 1970a,b; Gilles, 1977; Moran and Pierce, 1984; molluscs, Bartberger and Pierce, 1976; Dillon and Anderson, 1979; Strange and Crowe, 1979b; Henry and Mangum, 1980). The last mechanism to be considered involves control of the degradation of the amino acids. Isolated axons of the blue crab Callinectes sapidus show an increase in I4CO, production from different labeled amino acids when they are submitted to a hypoosmotic shock (Gilles, 1972b; Fig. 14). Concomitantly, there is an increase in 0, consumption and in total CO, production. Increase in NH, production has also been demonstrated in isolated axons of the Chinese crab Eriocheir sinensis upon application of hypoosmotic conditions (Gilles, 1974a). There is also an increase in I4CO, production from labelled glucose and pyruvate in these axonal preparations (Gilles, 1974b). Increased 14C02production from different labeled amino acids and from labeled glucose in hypoosmotic stress has been recently confirmed by Pressley and Graves (1983) at the level of isolated gill slices of the blue crab C . sapidus submitted to anisosmotic media. These different results demonstrate the participation of a deamination-oxidation sequence in the adjustment of the amino acid pool that occurs during volume regulation. Such changes in oxidative activity are further associated with redox changes occurring in the respiratory chain components (Gilles and Jobsis, 1972) as well as with changes in the general pattern of intermediary metabolism
228
R. GILLES
rn
::a
.I
“ C O ~ P R & ~ ~ T I O from N AMINO ACIDS
NH3 PRODUCTION nrnollhrlmgdry w i
A ALANINE
FIG. 14. Effect of a hypoosmotic shock on the 14C02 production from different ‘%-labeled amino acids and on NH, production of isolated axons of Eriocheir sinensis. Open bars: control; shadowed bars: experimental conditions. [After data from Gilles (1972b, 1974a).]
(Gilles, 1974b). This makes it clear that the changes in amino acid metabolism observed are not related to modifications in some specific pathways directly involved in their catabolism but are rather part of a more general readjustment of the intermediary and energy metabolism. Changes in deamination-oxidation appear also to occur in intact animals during salinity acclimation. Such acclimation leads indeed in mollusks as well as in crustaceans to changes in both blood level and excretion of NH, (NH, blood level: crustaceans, GCrard and Gilles, 1972b; Mangum et al., 1976; mollusks, Strange and Crowe, 1979b; Henry and Mangum, 1980. NH, excretion: crustaceans, Needham, 1957; Jeuniaux and Florkin, 1961; Mangum et al., 1976; Pressley et al. 1981; mollusks, Emerson, 1969; Allen and Garrett, 1971; Bartberger and Pierce, 1976; Henry and Mangum, 1980). On the other hand, acclimation of Carcinus maenas from sea water to 40% sea water increases by 50% the conversion of radiocarbon from glutamic acid to CO, in tissues of the green crab C. maenus (Chaplin et al., 1970). The question now arises of what mechanisms are implicated in these metabolic adjustments. This problem has been studied essentially by Gilles and colleagues in crustaceans and by Bishop and colleagues in mollusks. It has been discussed in the framework of different reviews during the past decade (Bishop, 1976; Bishop et al., 1981; Gilles, 1974a, 1975, 1979; Gilles and PCqueux, 1983). Thus we shall summarize only very briefly the main conclusions drawn. The studies by
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
229
Bishop concentrated on the possible metabolic origin(s) of the amino acids. As stated in his most recent reviews (Bishop er al., 1981), these origins seem to be multiple and complex, including protein turnover, the phosphoarginine pool, and possibly endogenous carbohydrates. The studies by Gilles and co-workers, on the other hand, did not primarily involve looking for an eventual major source of amino acids, considering that all the normal routes of entry into the pool participate. Further, since they showed that control is essentially exerted at the level of the exit pathways from the pool (see above), they concentrated on possible mechanism(s) of modulation of the transamination-amination-deamination sequence which control the amino acid pool level (Fig. 15). A priori, two different systems can be considered to play a part in the control of the transaminationamination-deamination sequence. A first process relates to genetically controlled modifications in the level and/or in the kinetics characteristics of one or several key enzymes. In another mechanism, the enzyme(s) activity is directly controlled by one or several modulators. As for the first possibility: Wickes and Morgan (1976) showed an important increase in the specific activity of glutamate dehydrogenase and aspartate aminotransferase in the oyster Crussosrrea virginica acclimating to concentrated media. Chaplin et al. (1965), however, were unable to detect any significant difference in glutamate dehydrogenase (GDH) activity in tissues of the green crab Carcinus maenas acclimated to 50% sea water or to 100% sea water. In the same way, we found no change in the specific activity or in the K, for aketoglutarate of glutamate dehydrogenase or aspartate aminotransferase from tissues of the Chinese crab Eriocheir sinensis acclimated to fresh water or to sea water (Gilles, 1974a, 1979). As for the possibility of a direct control of enzyme activity, we first showed
O D I KETOPRECURSORS SERINE AMINO ACIDS
NADH NADPH
HYDROXYPYRUVATE
ALANINE
&-KE TOGLUTARATE
OXALOACETATE
ASPAR TAT€
FIG. 15. The transamination-amination-deamination pathway controlling the free amino acids pool. ( 1 ) Glutamate dehydrogenase; (2) amino acid transaminases; (3) serine hydrolyase. +, 0. and - refer to activating effects, no effects, or inhibiting effects of NaCl on the enzymes’ activity.
230
R. GILLES
(Gilles, 1961; Gilles and Schoffeniels, 1964) that factors promoting changes in the intracellular inorganic ion equilibrium, such as veratrine or cocaine, induce changes in the incorporation of 14C from glucose and pyruvate. This led us to consider, as originally stated in Gilles (1961), “that the composition of the intracellular pool of free amino acids depends on the cationic composition of the cell by a variation of the degradation-synthesis equilibrium of the amino acids.” In 1963, we presented an initial, preliminary account on the effect of NaCl on the reductive amination activity of GDH from crayfish muscle (Schoffeniels and Gilles, 1963). This work has been confirmed and later extended with GDH extracted from a variety of sources (Chaplin er al., 1965; Corman and Kaplan, 1967; Gilles, 1974c; Measures, 1975). Interestingly, the activating effect of NaCl can be modulated by different amino acids as well as by the level of coenzyme (NADH) and substrates (a-ketoglutarate-NH, ) (see Fig. 16). It is, on the other hand, worth noting that in the same conditions of increasing NaCl concentration the activity of serine hydrolyase is decreased (Gilles, 1969), while the activity of alanine and aspartate aminotransferases remains unchanged (Turano er af., 1962; Chaplin er af., 1967; Huggins and Munday, 1968; Gilles, 1969). Also, the oxidative deamination activity of GDH is inhibited (Chaplin er al., 1965) or unaffected (Measures, 1975). Considering these effects, we have proposed (original paper, Gilles, 1969, see also Gilles, 1974a, 1975, 1978, 1979, for reviews) that in conditions of intracellular increase in ion concentration such as the one occumng in hyperosmotic conditions the equilibrium between amino acids and their keto-precursors is displaced, leading to a lower rate of metabolic output of amino acids from the pool through the transamination-amination-deamination sequence (Fig. 15). +
0.05 0:l 0.2 NADH (mmol liter-’)
FIG. 16. Effects of NaCl on the activity of glutamate dehydrogenase of Eriocheir sinensis leg muscle at different concentration of substrate (a-ketoglutarate) and coenzyme (NADH) in the presence or absence of serine. [After Gilles (1974c).]
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
231
A variety of other enzymes have also been studied by our group with respect to their sensitivity to salts (Gilles, 1969; Bolette-Dugaillay and Schoffeniels, 1970; Gilles et af., 1971). At the same time, the fate of the carbon units from glucose and pyruvate were also studied in isolated axons withstanding osmotic shocks (Gilles, 1974b). The results obtained, reviewed by Gilles (1979) and Gilles and PCqueux (1983), led to the conclusion that control of the fate of reducing equivalents as well as metabolic readjustment at the level of the Krebs oxidative cycle and of the general use of C-3 carbon units in glycolysis probably play some part in the metabolic adjustment of the amino acid pool occurring in cells withstanding osmotic challenges.
111. COPING WITH CHANGES IN ION CONTENT: MODIFICATION IN STRUCTURE AND ACTIVITY OF MACROMOLECULAR COMPONENTS The effects of monovalent inorganic ions on structure and activity of protein macromolecules have been studied for more than 20 years, essentially by the Von Hippel group and also by Warren and colleagues (Von Hippel and Wong, 1963, 1965; Von Hippel and Schleich, 1969a,b; Schleich and Von Hippel, 1970; Hamabata and Von Hippel, 1973; Warren and Cheatum, 1966; Warren et al., 1966; see Von Hippel, 1976, for review). In short, the net free energy stabilizing the folded form of a protein is in most cases very small, and environmental changes, as those induced for instance by the addition of monovalent ions to the medium, can “push” the protein across a phase transition boundary with the resultant formation of a different equilibrium structure. These “disturbing” effects can of course induce modifications in the activity of the proteins considered. These studies have been achieved, using physical methods, on a few isolated macromolecules used as models (collagen-DNA-myosin). They lead to the question of whether such “disrupting” effects of monovalent ions can be found in invertebrate cells and preparations of macromolecular components. Research on invertebrate material has been conducted to date using three different, though complementary, modes of approach: ( I ) Can disrupting effects of monovalent ions account for the modulation of enzyme activity leading to the metabolic adjustments in the amino acids level seen during cell volume control? (2) Can disrupting effects of monovalent ions be seen by electron microscopy in macromolecular components of cells submitted to osmotic challenges? (3) Can the disrupting effects of monovalent ions be “compensated” for by the high concentrations of organic solutes specifically found in cells of marine invertebrates? The last line of approach and the notion of “compensatory” solutes have been essentially studied in recent years by Clark and Somero and co-workers at San
232
R. GILLES
Diego, California, and have been the subject of different papers and reviews (Clark and Zounes, 1977; Bowlus and Somero, 1979; Clark et af., 1981; Yancey et at., 1982; see Clark, 1985, for review). Since this topic is considered at length by Clark, p. 25 1, this volume, we will not deal with it here. The two other modes of approach have been essentially developed in our laboratory.
A. Effects of Ions on Enzyme Actlvlty and Structure During the past two decades, we have shown that the effects of monovalent ions on the enzymes activity is directly related to a “disturbing” effect, chiefly of the anion on the structure of the enzyme protein. These effects have been integrated into a physiological model of control of the amino acid level by metabolic adjustment discussed in a previous section of this article (see Fig. 15). Briefly, our results showed that the series obtained when placing various anions in order of increasing effectiveness (i.e., Ac- < C1- C Br < NO,-;cf. Gilles, 1969; Gilles et al., 1971; Fig. 17) is similar to the one found by Von Hippel and colleagues in which the anions disturb the structure of different macromolecules 201 -
V
1510.
50100 200 300 LOO SALT ImM 1
..
.Nam
/’
,
?n
1 ““I
A-
.s.--;;;.1 -->-/ ., --I
-
,NaBf
1
1
1
I
05i)lOO 200 300 600 SALT (mM1
FIG.17. Effects of different monovalent salts of CI- (A) and of Na+ (B) on the activity of the succinate dehydrogenase of Myrilus californianus mantle tissue particulate preparations. [After data from Gilles et al. (1971).]
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
233
vi2 -
v; 1
000
‘
h 1 1
--
1;
0 lb 12 i6 1b i0 TIME (min) FIG. 18. Effect of NaCl on the thermosensitivity at 52°C of the glutamate dehydrogenase of Homarus gammarus muscle. Values are expressed as ratio of reaction velocity of the heated enzyme (Vi2) to reaction velocity at room temperature (V,l). (Original.)
as determined by physical methods. We have also shown that the addition of NaCl to the incubation medium largely modifies the thermosensitivity curve of glutamate dehydrogenase (Fig. 18). These different results indicate that the anion effect results from a direct “disturbing” effect on the enzyme protein configuration rather than from some indirect effect on association ion substrate, substrate pH, or electrostatic shielding. How monovalent ions can affect the structure and therefore the activity of enzymes and what kind of structural modifications are induced remain to date unanswered questions. The studies of the Von Hippel group led to the conclusion that ion effects are probably mediated, at least partly, through modifications in the structure of water in the vicinity of the protein (see von Hippel, 1976; Low, 1985, for discussions). A clear understanding of these effects unfortunately remains dependent on knowledge still-lacking on the equilibrium structure of that vicinal water and of the thermodynamical changes induced during the interactions of ions-water-proteins.
B. Effects of Ions on Other lntracellular Structures In the course of a recent study, we have shown that application of osmotic shocks induces important changes in the ultrastructure of mammalian cells (Delpire ef al., 1985a,b; Fig. 19). These modifications essentially concern (1) the electron density of the cytosol compartment, the general cytoplasmic background appearing much darker in the hyperosmotic conditions than in the control or hypoosmotic ones; we interpreted these changes as indicating modification in the
234
R. GILLES
FIG. 19. Effects of hypo- and hyperosmotic shocks on preparations of rat kidney slices and of Carcinus maenus leg muscle. [Original, other figures about rat kidney slices given in Delpire er al. (1985a).] Rat kidney (9520X): (A) hyperosmotic; (B) isosmotic; (C) hypoosmotic. Crab muscle (29300x): (D) hyperosmotic; (E) isosmotic; (F) hypoosmotic.
structural organization of some macromolecular component(s) of the so-called “ground material” of the cytoplasm; and (2) the organization of the nucleus, particularly of the chromatin, which shows an important condensation in hyperosmotic media and of a decondensation in hypoosmotic ones. These different modifications can be related to changes in the ions’ intracellular environment rather than to changes in cell hydration. They seem also to be of general occurrence, at least in mammalian cells, since they have been shown in rat pheochromocytoma cells of line PC12, mouse Ehrlich ascites tumor cells, rat intestinal and kidney tubular cells. It is worth noting that changes in chromatin organization, of the level of importance of those recorded in mammalian cells, cannot be recorded in the cells
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
235
of the crab Carcinus muenas submitted to variation in environmental osmolality either in the course of short-term in vitro experiments on isolated tissue preparation (Fig. 19) or of long-term in vivo acclimation of the animals to media of different salinities (not shown). Our crab muscle preparations, as the mammalian cells we have studied, however, show changes in ion concentration during in vitro osmotic challenges. The fact that there are only slight changes in chromatin organization, if any, cannot therefore be related to a lack of modification in ion levels. Possibly, euryhaline crustaceans might have evolved macromolecular components less sensitive to changes in ion levels. This would be an interesting adaptation for such species, which have to cope regularly with changes in salinity, inducing modifications in ion concentration. Another explanation could be found in the presence of high levels of amino compounds in the tissues of these species. These compounds would act as “compensatory solutes” against the disturbing effects of ions on macromolecular structures. More results are obviously needed to shed more light on this interesting problem. The results available to date nevertheless indicate that macromolecular structures of primary biological importance appear to be, directly or indirectly, far less sensitive to ion-disturbing effects in invertebrate marine poekilosmotic species than in mammalian homeosmotic ones. This difference is of much ecological interest since in the former species the cells have to cope with high and sometimes largely changing NaCl environmental levels, while in the latter, the cells never experience such challenges.
IV. CONCLUSIONS To briefly conclude this article, let us say that cells of euryhaline invertebrates, when submitted to osmotic challenges, show volume regulation responses that are most probably of ecological significance. For irstance, the volume regulatory response seen in tissues of the very euryhaline Chinese crab Eriocheir sinensis is much greater and faster than the one observed in the less euryhaline green crab Carcinus muenas, it being greater and faster than the one seen in the marine stenohaline lobster or stone crab Cancer pagurus. Volume regulatory responses are concomitant with changes in the intracellular levels of different osmotic effectors: essentially inorganic ions and amino compounds. So far as inorganic ions are concerned, K plays a particularly important role in the early, large volume decrease that immediately follows the osmotically induced swelling. The changes in the level of that ion are related to changes in efflux due to modifications in the activity of leak pathways. The amino compounds most studied to date have been the amino acids. Their intracellular concentration appears to be regulated essentially by control of their +
236
R. GILLES
output from the pool, this control being achieved mainly by two different mechanisms: (1) control of the efflux from the cells related to changes in permeability, and (2) control of the rate of metabolic exit from the pool by modulation of the enzymatic trans-deamination sequence. Nothing is actually known about the mechanisms of control of the leak pathways for both K + and the amino compounds. What are these pathways and what are the events that trigger their changes in structure and activity are surely fruitful topics for future research. Nothing is known about either the intracellular organization of the amino acid metabolism or of its modulation. Data on intracellular localization of enzymes, intracellular concentrations of substrates, coenzymes, and modulators, as well as on the ionic composition prevailing locally, are needed before a more detailed picture can be produced. On the other hand, we have shown that different intracellular macromolecular components, among which is chromatin, of marine euryhaline crustaceans appear, when studied in situ, to be far less sensitive to “disturbing” effects of monovalent ions than those of mammalian cells. This difference is obviously of much importance at the ecological level. It could result from a specific adaptation of the components themselves or from an indirect “compensatory” effect related to the presence in the cells of high amounts of different organic nitrogenous compounds that can act against the effects of ions. This also represents a most interesting topic for research on which more data would be welcome. REFERENCES Ahmad, I., and Hellebust, J. A. (1985). Osmoregulation in the euryhaline flagellate Brachiomonus submurim (chlorophyceae).Mar. Eiol. 87, 245-250. Ahokas, R. A., and Duerr, F. G. (1975). Tissue water and intracellular osmoregulation in two species of euryhaline teleosts, Culueu inconsruns and Fundulus diuphunus. Comp. Biochem. Physiol. 52A,449-454. Ahokas, R. A., and Sorg, G. (1977). The effect of salinity and temperature on intracellular osmoregulation and muscle free amino acids in Fundulus diuphunus. Comp. Biochem. Physiol. 56A, 101-105. Allen, J. A,, and Garrett, M. R. (1971). The excretion of ammonia and urea by Myu urenuriu L. (Mollusca: bivalvia). Comp. Biochem. Physiol. 39A, 633-642. Allen, K. (1961). The effect of salinity on the amino acid concentration in Rungiu cuneuru (Pelecypoda). Biol. Bull. 121, 419-424. Amende, L. M., and Pierce, S. K. (1980a). Cellular volume regulation in salinity stressed molluscs: The response of Noeriu ponderosu (Arcidae) red blood cells to osmotic variations. J . Comp. Physiol. 138, 283-289. Amende, L. M., and Pierce, S. K. (1980b). Free amino acid mediated volume regulation of isolated Noeriu ponderosu red blood cells: Control by Ca2+ and ATP. J. Comp. Physiol. 138,291-298. Andreoli. T. E., Grantham, J. J., and Rector, F. C. (1977). “Disturbances in Body Fluid Osmolality.” American Physiological Society, Bethesda, Maryland. Baginski, R. M., and Pierce, S. K. (1977). The time course of intracellular free amino acid accumulation in tissues of Modiolus demissus during high salinity adaptation. Comp. Biochem. Physiol. 57A, 407-412.
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
237
Baginski. R. M.,and Pierce, S. K. (1978). A comparison of amino acid accumulation during high salinity adaptation with anaerobic metabolism in the ribbed mussel, Modiolus dernissus demiS S U S . J . EXP. ZOO^. 203, 419-428. Balinsky, J . B., Dicker, S. E., and Elliot, A. B . (1972). The effect of long-term adaptation to different levels of salinity on urea synthesis and tissue amino acid concentrations in Rana cancrivora. Comp. Biochem. Physiol. 43B, 7 1-82. Bartberger, C. A . , and Pierce, S . K. (1976). Relationship between ammonia excretion rates and hemolymph nitrogenous compounds of a euryhaline bivalve during low salinity acclimation. Biol. Bull. 150, 1-14. Baxter, C. F., and Ortiz. C. L. (1966). Amino acids and the maintenance of osmotic equilibrium in brain tissue. Life Sci. 5 , 2321-2329. Bedford, J . J., and Leader, J. P. (1977). The composition of the haemolymph and muscle tissue of the shore crab tfemigrapsus edwardsi exposed to different salinities. Comp. Biochem. Physiol. 57A, 341-345. Benson, J. A , , and Treherne, J. E. (1978a). Axonal adaptations to osmotic and ionic stress in an invertebrate osmoconformer (Mercierella efligmura Fauvel). 11. Effects of ionic dilution on the resting and action potentials. J. Exp. Biol. 76, 205-219. Benson. J. A,. and Treherne, J. E. (1978b). Axonal adaptation to osmotic and ionic stress in an invertebrate osmoconformer (Mercierella enigmara Fauvel). 111. Adaptations to hypoosmotic dilution. J. Exp. Biol. 76, 221-235. Benson-Rodenbough, B.,and Ross Ellington, W. (1982). Responses of the euryhaline sea anemone (Bunodosoma cavernara (Bosc)) (Anthozoa, actinaria, actimidae) to osmotic stress. Comp. Biochem. Physiol. 72A, 731-735. Beres, L. S . , and Pierce, S. K. (1981). The effects of salinity stress on the electrophysiological properties of Mya arenuria neurons. J. Comp. Physiol. 144, 165-173. Bishop, S. H . (1976). Nitrogen metabolism and excretion: Regulation of intracellular amino acid concentration. In "Estuarine Processes" (M.Witey, ed.), Vol. 1. pp. 414-431. Academic Press, New York. Bishop, S. H., Greenwalt, D. E., and Burcham, J. M. (1981). Amino acid cycling in ribbed mussel tissues subjected to hyperosmotic shock. J. Exp. Zool. 215, 277-287. Bollette-Dugaillay, S . , and Schoffeniels, E. (1970). lsoenzymes de la deshydrogenase lactique et stress osmotique. Arch. Int. Physiol. Biochim. 78, 23-27. Bowlus, R. D., and Somero. G. N. (1979). Solute compatibility with enzyme function and structure: Rationales for the selection of osmotic agents and end-products of anaerobic metabolism in marine invertebrates. J. Exp. Zool. 208, 137-152. Boyd, T. A,, Cha, C. J., Forster, R. P., and Goldstein, L. (1977). Free amino acids in tissues of the skate Raja erinucea and the stingray Dasyaris sabinu: Effects on environmental dilution. J. Exp. ZOO^. 199, 435-442. Bricteux, S . , Jeuniaux, Ch., and Florkin, M . (1961). Role de la variation de la composante amino acide intracellulaire dans I'euryhalinitC de Leander serrurus et de Leander squillu. Arch. Inr. Physiol. Biochim. 69, 744. Bricteux-Gdgoire, S.,Duchateau-Bosson. Gh., Jeuniaux, Ch., and Florkin, M. (1962). Constituants osmotiquement actifs des muscles du crabe chinois Eriocheir sinensis adapt6 h I'eau douce ou h I'eau de mer. Arch. Inr. Physiol. Biochim. 70, 273-286. Bricteux-Gdgoire, S.,Duchateau-Bosson, Gh., Jeuniaux, Ch., Schoffeniels, E., and Florkin, M. (1963). Constituants osmotiquement actifs du sang et des muscles du scorpion Androctomus australis L. Arch. Int. Physiol. Biochim. 71, 393-400. Bricteux-Grkgoire, S.,Duchateau-Bosson, Gh., Jeuniaux, Ch., and Florkin, M. (1964a). Constituants osmotiquement actifs des muscles adducteurs d'Osrrea edulis adapt6e h I'eau de mer ou h I'eau saumltre. Arch. Int. Physiol. Biochim. 72, 267-275. Bricteux-G&goire, S., Duchateau-Bosson, Gh., Jeuniaux, Ch., and Florkin, M . (1964b).Consti-
R. GILLES tuants osmotiquement actifs des muscles adducteurs de Gryphea angulara adaptte a I’eau de mer ou a I’eau saumltre. Arch. Inr. Physiol. Biochim. 72, 835-842. Bricteux-Gkgoire, S., Duchateau-Bosson, Gh., Jeuniaux, Ch., and Florkin, M. (1964~).Constituants osmotiquement actifs des muscles adducteurs de Mytilus edulis adaptCe a I’eau de mer ou a I’eau saumltre. Arch. Int. Physiol. Biochim. 72, 116-123. Bricteux-Grtgoire, S . , Duchateau-Bosson, Gh., Jeuniaux, Ch., and Florkin, M. (1966). Les constituants osmotiquement actifs des muscles et leur contribution a la rtgulation isosmotique intracellulaire chez Limulus polyphemus. Comp. Biochem. Physiol. 19, 729-736. Burton, R. S., and Feldman, M. W. (1982). Changes in free amino acid concentrations during osmotic response in the intertidal copepod Tigriopus californicus. Comp. Biochem. Physiol.
73A, 441-445. Cala, P. M. (1977). Volume regulation by flounder red blood cells in anisotonic media. J . Gen. Physiol. 69, 537-552. Cala, P. M. (1980). Volume regulation by Amphiuma red blood cells. The membrane potential and its implications regarding the nature of the ion-flux pathways. J . Gen. Physiol. 76, 683-708. Cala, P. M. (1983). Volume regulation by red blood cells: Mechanisms of ion transport. Mol. Physiol. 4, 33-52. Cala, P. M. (1985). Volume regulation by Amphiuma red blood cells: Characteristics of volumesensitive K/H and Na/H exchange. Mol. Physiol. 8, 199-214. Chaplin, A. E., Huggins, A. K., and Munday, K. A. (1965). Ionic effects on glutamate dehydrogenase activity from beef liver, lobster muscle and crab muscle. Comp. Biochem. Physiol. 16,4962. Chaplin. A. E., Huggins, A. K., and Munday, K. A. (1967). The distribution of ~ - amia notransferases in Carcinus mueans. Comp. Biochem. Physiol. 20, 195- 198. Chaplin, A. E., Huggins, A. K., and Munday, K. A. (1970).The effect of salinity on the metabolism of nitrogen-containing compounds by Carcinus maenas. Int. J . Biochem. 1, 385-400. Cholette, C., and Gagnon, A. (1973). Isosmotic adaptation in Myxine glurinosa L. 11. Variations of the free amino acids, trimethylamine oxide and potassium of the blood and muscle cells. Comp. Biochem. Physiol. &A, 1009-1021. Cholette, C., Gagnon, A,, and Germain, P. (1970). Isosmotic adaptation in Myxine glutinosa. 1. Variations of some parameters and role of the amino acid pool of the muscle cells. Comp. Biochem. Physiol. 33, 333-346. Clark, M. E. (1968). A survey of the effect of osmotic dilution on free amino acids of various polychaetes. Biol. Bull. 134, 252-260. Clark, M. E. (1985). The osmotic role of amino acids: Discovery and function. In “Transport Processes-lono and Osmoregulation” (R. Gilles and M. Gilles-Baillien, eds.). pp. 412-423. Springer-Verlag, Berlin. Clark, M., and Zounes, M. (1977). The effects of selected osmolytes on the activity of lactate dehydrogenase from the euryhaline polychaete, Nereis succinea. Biol. Bull. 153, 468-484. Clark, M., Hinke, I . A. M., and Todd, M. E. (1981). Studies on water in barnacle muscle fibres. 11. Role of ions and organic solutes in swelling of chemically skinned fibres. J. Exp. Biol. 90,4363. Colley, L.,Rowe, W. C., Huggins, A. K., Elliott, A. B., and Dicker, S. E. (1972). The effect of short-term changes in the external salinity on the levels of the non-protein nitrogenous compounds and the ornithine urea cycle enzymes in Rana cancrivora. Comp. Biochem. Physiol. 41B, 307-322. Colley, L.,Fox, F. R., and Huggins, A. K. (1974). The effect of changes in external salinity on the non-protein nitrogen constituents of parietal muscle from Agonus cataphracrus. Comp. Biochem. Physiol. 48A, 757-763. Corman, L., and Kaplan. N. 0. (1967). Kinetic studies of dogfish liver glutamate dehydrogenase
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
239
with diphosphopyridine nucleotides and the effect of added salts. J . Eiol. Chem. 242, 28402846. Costa, C. J., and Pierce, S. K. (1982). Effect of divalent cations and metabolic inhibitors on Glyceru red coelomocyte volume regulation and solute balance during hypoosmotic stress. J. Comp. Physiol. 151, 133-144. Costa, C. J . , Pierce, S. K . , and Warren, M. K. (1980). The intracellular mechanism of salinity tolerance in polychaetes: Volume regulation by isolated Glyceru dibrunchiuru red coelomocytes. Eiol. Bull. 159, 626-638. Delpire, E., Duchene. C . , Coessens, G . , and Gilles, R. (1985a). Effects of osmotic shocks on the ultrastructure of different tissues and cell types. Exp. Cell Res. 160, 106-116. Delpire, E., Gilles, R., Duchene, C . , and Goessens, G . (1985b). Effects of osmotic shocks on the ion content and ultrastructure of rat pheochromocytoma cells of line PC12. Mol. Physiol. 8, 293-306. Dillon, T . , and Anderson. J. W. (1979). Ionic and osmotic adjustments of the clam Rangia cunearu fray in response to reduced salinity and mercury exposure. Comp. Biochem. Physiol. 62A,519526. Doljanski, F., Ben-Sasson, S., Reich, M., and Grover, N. B. (1974). Dynamic osmotic behavior of chick blood lymphocytes. J . Cell. Physiol. 84, 215-224. Duchateau, Gh., and Florkin, M. (1955). Concentration du milieu extCrieur et Ctat stationnaire du pool des acides aminks non protCiques des muscles d’Eriocheir sinensis, Milne Edwards. Arch. Int. Physiol. Biochim. 63, 249-25 1 . Duchateau-Bosson, Gh., and Florkin, M. (1961 ). Change in intracellular concentration of free amino acids as a factor of euryhalinity in the crayfish Asfucus usfucus. Comp. Biochem. Physiol. 3, 245-249. Duchateau, Gh, Florkin, M., and Jeuniaux, Ch. (1959). Composante amino-acide des muscles de Curcinus muenas L. lors du passage de I’eau de mer I’eau saumitre et au cours de la mue. Arch. Inf. Physiol. Eiochim. 67, 489-500. Duchateau-Bosson, Gh., Jeuniaux, Ch., and Florkin, M. (1961). Role de la variation de la composante amino-acide intracellulaire d m s I’euryhalinitt d’Arenicolu marina L. Arch. Inr. Physiol. Biochim. 69, 30-35. Du Paul, W. E., and Webb, K. L. (1970). The effect of temperature on salinity induced changes in the free amino acid pool of Myu arenuria. Comp. Biochim. Physiol. 32, 785-801. Du Paul, W. D., and Webb, K. L. (1974). Salinity induced changes in the alanine and aspartic amino transferase activity in the three marine bivalve molluscs. Arch. Inf. Physiol. Biochim. 82, 817822. Ellory, J . C., Hall, A. C., and Stewart, G. W. (1985a). Volume-sensitive cation fluxes in mammalian red cells. Mol. Physiol. 8, 235-246. Ellory, J. C., Hall, A. C., and Stewart, G , W. (1985b). Volume sensitive passive potassium fluxes in red cells. In “Transport Processes. Iono- and Osmoregulation” (R. Gilles and M. GillesBaillien, eds.), pp. 401-410. Springer-Verlag, Berlin. Emerson, D. N. (1969). Influence of salinity on ammonia excretion rates and tissue constituents of euryhaline invertebrates. Comp. Biorhem. Physiol. 29, 1115-1 133. Farmer, L., and Reeve, M. R . (1978). Role of the free amino acid pool of the copepod Acurriu fonsu in adjustment to salinity change. Mar. Biol. 48, 311-316. Fisher, R. S . , Persson, 6. E., and Spring, K. R. (1981). Epithelial cell volume regulationbicarbonate dependence. Science 214, 1357- 1359. Florkin, M. (1956). Vergleichende Betrachtung des stationiiren Zustandes der nichteiweissgebundenen Aminosaiiren der Tiere. Colloq. Ges. Physiol. Chem., 6th pp. S-62-99. Florkin, M. (1962). La rCgulation isosmotique intracellulaire chez les invertebrh marins euryhalins. Bull. Acud. Belg. Cl. Sci. 48, 687-694.
240
R. GILLES
Forster, R. P., and Goldstein, L. (1976). Intracellularosmoregulatory role of amino acids and urea in marine elasmobranchs. Am. J. Physiol. 230, 925-931. Forster, R. P., and Hannafin, J. A. (1980). Osmotic and cell volume regulation in atrium and ventricle of the elasmobranch skate, Raja erinacea. Comp. Biochem. Physiol. 65A. 445-451. Forster, R. P.,Hannafin, J. A., and Goldstein, L. (1978). Osmoregulatory role of amino acids in brain of the elasmobranch Raja erinacea. Comp. Biochem. Physiol. MA, 25-30. Foskett, J. K.,and Spring, K. R. (1985). Involvement of calcium and cytoskeleton in gallbladder epithelial cell volume regulation. Am. J . Physiol. 248, C27-36. Fredericq, L. (1885). Influence du milieu ambiant sur la composition du sang des animaux aquatiques. Arch. Zool. Exp. Gen. I11 (sdrie 2), XXXIV-XXXVIII. Fredericq, L. (1901). Sur la concentration moltculaire du sang et des tissus chez les animaux aquatiques. Bull. Acad. R . Belg. CI. Sci. 428-454. Freel, R. W. (1978a). Patterns of water and solute regulation in the muscle fibres of osmoconforming marine decapod crustaceans. J . Exp. Biol. 72, 107-126. Freel, R. W. (1978b). Transmembrane activity gradients of K and CI in the muscle fibres of osmoconforming marine decapod crustaceans. J . Exp. Biol. 72, 127-140. Freel, R. W. (1982). Heterogeneity and dynamics of muscle fiber water in marine crustaceans. J . Comp. Physiol. 146, 113-121. Freel, R. W., Medler, S. G . , and Clark, M. E. (1973). Solute adjustments in the coelomic fluid and muscle fibers of a euryhaline polychaete, Neanrhes succinea, adapted to various salinities. Biol. Bull. 144, 289-307. Fyhn, H. J . (1976). A note on the hyperosmotic regulation in the brackish water clam Rangza cuneaia. J. C o w . Physiol. 107, 265-268. Gainer, H., and Grundfest, H. (1968). Permeability of alkali metal cations in lobster muscle. A comparison of electrophysiologicaland osmometric analyses. J. Gen. Physiol. 51, 399-425. Gerard, I. F. (1975). Volume regulation and alanine transport. Response of isolated axons of Callinecres sapidus Rathbun to hypo-osmotic conditions. Comp. Biochem. Physiol. SlA, 225229. Gerard, J. F., and Gilles, R. (1972a). Modification of the amino-acid efflux during the osmotic adjustment of isolated axons of Callinecres sapidus. Experienria 28, 863-864. Gtrard, J. F., and Gilles, R. (1972b). The free amino-acid pool in Callinecres sapidus (Rathbun) tissues and its role in the osmotic intracellularregulation. J. Exp. Mar. Biol. Ecol. 10, 125- 136. Gilles, R. (1961). Etude des relations existant entre la synthhse des acides amints A partir d’oses et de ddrivts d’oses et les phdnomknes osmokgulateurs au niveau de nerfs isolts de Homarus vulgaris (L.).Ann. SOC.R. Zool. Belg. 92, 191-193. Gilles, R. (1969). Effect of various salts on the activity of enzymes implicated in amino-acid metabolism. Arch. Inr. Physiol. Biochim. 77, 441-464. Gilles, R. (1970). Osmoregulation in the stenohaline crab “Libinia emarginara” Leach. Arch. Inr. Physiol. Biochim. 78, 91-99. Gilles, R. (1972a). Osmoregulation in three molluscs: Acanrhochirona discrepans (Brown), Glycymeris glycymeris (L.) and Myrilus edulis (L.).Biol. Bull. 142, 25-35. Gilles, R. (1972b). Amino-acid metabolism and isosmotic intracellular regulation in isolated surviving axons of Callinecres sapidus. Life Sci. 11, Part 11, 565-572. Gilles, R. (1973). Osmotic behaviour of isolated axons of a euryhaline and a stenohaline crustacean. Experienria 29, 1354- 1355. Gilles, R. (1974a). Mdtabolisme des acides amints et controle du volume cellulaire. Arch. Inr. Physiol. Biochim. 82, 423-589. Gilles, R. (1974b). Glucose metabolism during osmotic stress in isolated axons of Eriocheir sinensis. Life Sci. 15, 1363-1369. Gilles, R. (1974~).Studies on the effect of NaCl on the activity of Eriocheir sinensis glutamate dehydrogenase. Inr. J . Biochem. 5 , 623-628.
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
24 1
Gilles, R. (1975). Mechanisms of ion and osmoregulation. I n “Marine Ecology” (0.Kinne, ed.), Vol. 2, Part 1, pp. 259-347. Wiley (Intersciences), London, New York. Gilles, R. (1977). Effects of osmotic stresses on the proteins concentration and pattern of Eriocheir sinensis blood. Comp. Biochem. Physiol. 56A, 109-1 14. Gilles, R. (1978). Intracellular free amino acids and cell volume regulation during osmotic stresses. A. Benzon Symp. 11, 470-491. Gilles, R. (1979). Intracellular organic osmotic effectors. I n “Mechanisms of Osmoregulation in Animals” (R. Gilles, ed.), Chap. 4, pp. 11 1-153. Wiley (Interscience), London, New York. Gilles, R. (1980a). Cell volume control as studied in crustacean tissues. I n “Animals and Environmental Fitness” (R.Gilles, ed.), Vol. 1, pp. 3-24. Pergamon, Oxford. Gilles, R. (1980b). Mechanisms of volume regulation and control of the level of intracellular osmotic effectors. J. Parenter. Enteral Nurr. 4, 121-130. Gilles, R. (1983). Volume maintenance and regulation in animal cells: Some features and trends. Mol. Physiol. 4, 3-16. Gilles, R., and Gerard, J. F. (1974). Amino-acid metabolism during osmotic stress in isolated axons of Callinectes sapidus. Life Sci. 14, 1221- 1229. Gilles, R., and Jobsis, F. F. (1972). Isosmotic intracellular regulation and redox changes in the respiratory chain components of Cullinectes sapidus isolated muscle fibers. Life Sci. 11, 877885. Gilles, R., and Pkqueux, A. (1977). Effect of salinity on the free amino acids pool of the red alga Porphyridium purpureum (=P. Cruentum). Comp. Biochem. Physiol. 57A, 183-185. Gilles, R., and PCqueux, A. (1981). Cell volume regulation in crustaceans: Relationship between mechanisms for controlling the osmolality of extracellular and intracellular fluids. J. Exp. Zool. 215, 351-362. Gilles, R., and Pkqueux, A. (1983). Interactions of chemical and osmotic regulation with the environment. I n “The Biology of Crustacea” (F. J . Vernberg and W. B. Vernberg, eds.), Vol. 8, pp. 109-177. Academic Press, New York. Gilles, R., and Schoffeniels, E. (1964). Action de la veratrine, de la cocahe et de la stimulation Clectrique sur la synthbse et sur le pool des acides aminks de la chaine nerveuse ventrale du Homard. Biochim. Biophys. Acta 82, 525-537. Gilles, R., and Schoffeniels. E. (1969). Isosmotic regulation in isolated surviving nerves of Eriocheir sinensis Milne Edwards. Comp. Biochem. Physiol. 31, 927-939. Gilles, R., Hogue, P., and Keamey, E. B. (1971). Effect of various ions on the succinic dehydrogenase activity of Mytilus californianus. Life Sci. 10, Part 11, 1421-1427. Gilles, R., Duchene, C.. and Lambert, I. (1983). Effect of osmotic shocks on rabbit kidney cortex slices. Am. J. Physiol. 244, F696-705. Gilles, R., Delpire, E.,Duchene, C., M. Cornet, and Pequeux, A. (1986). The effect of cytochalasin B on the volume regulation response of isolated axons of the green crab Carcinus maenas submitted to hypo-osmotic media. Comp. Biochem. Physiol. Gilles-Baillien, M . (1973). Isosmotic regulation in various tissues of the diamondback terrapin Malaclemys cenrrata centrata (Latreille). J . Exp. Biol. 59, 39-43. Gordon, M. S. (1965). Intracellular osmoregulation in skeletal muscle during salinity adaptation in two species of toads. Biol. Bull. 128, 218-229. Gordon, M. S.. and Tucker, V. A. (1968). Further observations on the physiology of salinity adaptation in the crab-eating frog (Rana cancrivora). J . Exp. Biol. 49, 185-193. Grinstein, S., Goetz, J. D., Furuya, W., Rothstein, A,, and Gelfand, E. W. (1985). Mechanism of regulatory volume increase in osmotically shrunken lymphocytes. Mol. Physiol. 8, 185- 198. Hamabata, A,, and Von Hippel, P. H. (1973). Effects of vicinal groups on the specificity of binding of ions to amide groups. Biochemisrv 12, 1264- 127 1. Hams, R. R. (1976). Extracellular space changes in Carcinus maenas during adaptation to low environmental salinity. J. Physiol. (London) 258, 3 I -32P.
242
R. GILLES
Henry, R. P., and Mangum, C. P. (1980). Salt and water balance in the oligohaline clam Rungiu cuneuta. Ill. Reduction of the free amino acid pool during low salinity adaptation. J. Exp. Zool. 211, 25-32. Henry, R. P., Mangum, C. P., and Webb, K. L. (1980). Salt and water balance in the oligohaline clam, Rungiu cunearu. 11. Accumulation of intracellular free amino acids during high salinity adaptation. J. Exp. Zool. 211, 11-24. Hoffmann, E. K. (1977). Control of cell volume. In “Transport of Ions and Water in Animals’’ (B. L. Gupta, R. B. Moreton, J. Oschman. and B. J. Wall, eds.), pp. 285-332. Academic Press, London. Hoffmann, E. K.(1980). Cell volume regulation in mammalian cells. I n “Animals and Environmental Fitness” (R. Gilles, ed.), Vol. 1, pp. 43-60. Pergamon, Oxford. Hoffmann, E. K. (1985a). Regulatory volume decrease in Ehrlich ascites tumor cells: Role of inorganic ions and amino-compounds. Mol. Physiol. 8, 167-184. Hoffmann, E. K. (1985b). Cell volume control and ion transport in a mammalian cells. I n “Transport Process, Iono- and Osmoregulation” (R. Gilles and M. Gilles-Baillien, eds.), pp. 389400. Springer-Verlag, Berlin. Hoffmann, E. K., and Hendil, K. B. (1976). The role of amino acids and taurine is isosmotic intracellular regulation in Ehrlich ascites mouse tumour cells. J. Comp. Physiol. 108,279-286. Hoffmann, E. K., Sjoholm, C., and Ole Simonsen, L. (1983). N a + , CI- cotransport in Ehrlich ascites tumor cells activated during volume regulation (regulatory volume increase). J . Membr. Biol. 76, 269-280. Howard, C. L., and Kasschau, M. R. (1980). Effects of seasonal changes on the free amino acid pool of the sea anemone Bunodosoma cavernutu. Am. Zool. 20, 818 Hoyaux, J., Gilles, R., and Jeuniaux, Ch. (1976). Osmoregulation in molluscs of the intertidal zone. Comp. Biochem. Physiol. 53A, 361-365. Huggins, A. K., and Colley, L. (1971). The changes in non-protein nitrogenous constituents of muscle during the adaptation of the eel Anguillu unguilla L. from fresh water to sea water. Comp. Biochem. Physiol. 38B, 537-541. Huggins, A. K., and Munday, K. A. (1968). Cmstacean metabolism. Adv. Comp. Physiol. Biochem. 3, 271-378. Hughes, P. M., and MacKnight, A. D. C. (1976). The regulation of cellular volume in renal cortical slices incubated in hyposmotic medium. J. Physiol. (London) 257, 137-154. Jeuniaux, Ch., and Florkin, M. (1961). Modification de I’excrktion azotke du crabe chinois au cours de I’adaptation osmotique. Arch. Int. Physiol. Biochim. 69, 385-386. Jeuniaux, Ch., Duchiteau-Bosson, Gh., and Florkin, M. (1961a). Variation de la composante amino-acide des tissus et euryhalinitk chez Perinereis cultriferu Gr. et Nereis diversicolor (0.F. Miiller). J. Biochem. (Tokyo) 49, 527-53 1. Jeuniaux, Ch., Bricteux-Grkgoire, S., and Florkin, M. (1961b). Contribution des acides aminks libres la dgulation osmotique intracellulaire chez deux crustacbs euryhalins L e a d e r serrutus F. et Leunder squillu L. Cuh. Biol. Mar. 11, 373-380. Jeuniaux, Ch., Bricteux-Grbgoire, S . , and Florkin, M. (1962). Role osmorkgulateur intracellulaire du glycocolle et de la taurine chez I’btoile de mer. Arch. Inr. Physiol. Biochim. 70, 155-156. Jonsson, 0. (1971). Effects of variations in the extracellular osmolality on the permeability to nonelectrolytes of vascular smooth muscle. Acru Physiol. S c u d . 81, 528-539. Jurss, K. (1980). The effect of changes in external salinity on the free amino acids and two aminotransferases of white muscle from fasted Sulmo guirdneri Richardson. Comp. Biochem. Physiol. 65A, 501-504. Kaneshiro, E. S . , Holz, G.G., Jr. and Dunham, P. B. (1969). Osmoregulation in a marine ciliate, Miumiensis uvidus 11. Regulation of intracellular free amino acids. Biol. Bull. 137, 161-169. Kasschau, M. R. (1975). The relationship of free amino acids to salinity changes and temperature-
243
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
salinity interactions in the mud-flat snail, Nassarius obsolerus. Comp. Biochem. Physiol. SlA, 30 1-308. Kaushik, S., Harache, Y . , and Luquet, P. (1977). Variations in the total free amino acids level in rainbow trout muscle and blood during its adaptation to sea water. Ann. Hydrobiol. 8, 145-151. Ktvers, C., Ptqueux, A.. and Gilles, R . (1979a). Effects of hypo- and hyperosmotic shocks on the volume and ions content of Carcinus maenas isolated axons. Comp. Biochem. Physiol. 64A, 427-43 1 . Ktvers, C., Ptqueux, A., and Gilles, R. (1979b). Effects of an hypo-osmotic shock on N a f , K + and CI - levels in isolated axons of Carcinus maenas. J. Comp. Physiol. 129, 365-371. Kevers, C., Ptqueux, A . , and Gilles. R. (1981). Role of K + in the cell volume regulation response of isolated axons of Carcinus maenas submitted to hypo-osmotic conditions. Mol. Physiol. 1, 13-22. King, P. A., Cha. C., and Goldstein. L. (1980). Amino acid metabolism and cell volume regulation in the little skate, Raja erinacea. I. Oxidation. J . Exp. 2001. 212, 69-77. Kregenow, F. M. (1971). The response of duck erythrocytes to non-hemolytic hypotonic mediaEvidence for a volume controlling mechanism. J. Gen. Physiol. 58, 372-395. Kregenow, F. M. (1981). Osmoregulatory salt transporting mechanisms: Control of cell volume in anisotonic media. Annu. Rev. Physiol. 43, 493-505. Lang, M. A., and Gainer, H. (1969a). lsosmotic intracellular regulation as a mechanism of volume control in crab muscle fibers. Comp. Biochem. Physiol. 30, 44-456. Lang, M. A., and Gainer, H. (1969b). Volume control by the muscles of the blue crab. Volume readjustments in hypotonic salines. J. Gen. Physiol. 53, 323-340. Lange, R . (1963). The osmotic function of amino acids and taurine in the mussel, Myrilus edulis. Comp. Biochem. Physiol. 10, 173-179. Lange, R . ( 1964). The osmotic adjustment in the echinoderm srrongylocenrrorus droebachiensis. Comp. Biochem. Physiol. 13, 205-216. Lange, R., and Fugelli. K. (1965). The osmotic adjustment in euryhaline teleosts, the flounder Pleuronecfes Jesus L. and the three spined stickle-back, Gasrerosreus aculearus L. Comp. Biochem. Physiol. 15, 283-292. Lange. R., and Mostad, A. (1967). Cell volume regulation in osmotically adjusting marine animals. J. Exp. Mar. Biol. Ecol. 1, 209-219. Lasserre. P., and Gilles, R. (1971). Modification of the amino acid pool in the parietal muscle of two euryhaline teleosts during osmotic adjustment. Experienriu 27, 1434- 1435. Lauf, P.K. (1985). On the relationship between volume- and thiol-stimulated K , C1- fluxes in red cell membranes. Mol. Physiol. 8, 215-234. Law, R. (1985). Volume regulation by mammalian renal cells exposed to anisosmotic media. Mol. Physiol. 8, 143-160. Leader, J . P., and Bedford, J. J. (1978). Volume regulation in virro of muscle fibers of the crab, Hemigrapsus edwardsi. J. Comp. Physiol. 128, 153-159. Licht, P., Feder, M. E., and Bledsoe, S. (1975). Salinity tolerance and osmoregulation in the salamander Barrachoseps. J. Comp. Physiol. 102, 123- 134. Linshaw, M. A , , and Grantham. J. J . (1980). Effect of collagenase and ouabain on renal cell volume in hypotonic media. Am. J. Physiol. 238, F491-498. Liu. M. S.. and Hellebust, J. A. (1976). Effects of salinity and osmolarity of the medium on amino acid metabolism in Cvclorella cryprica. Can. J . Bor. 54, 938-948. L o w , P. S. (1985). Molecular basis of the biological compatibility of nature's osmolytes. I n "Transport Processes-lono- and Osmoregulation" (R. Gilles and M. Gilles-Baillien, eds.), pp, 469483. Springer-Verlag, Berlin. Lynch. M. P., and Wood. L. (1966). Effects of environmental salinity on free amino acids of Crassosrrea virginica (Gmelin). Comp. Biochim. Phvsiol. 19, 783-790. +
244
R. GILLES
Machin, J., and J. O’Donnell (1977). Volume regulation in the coelomocytes of the blood worm Glyceru dibrunchiuru. J. Comp. Physiol. 117, 303-31 1. Mackay, M. A., Norton, R. S . , and Borowitzka, L. J. (1984). Organic osmoregulatory solutes in cyanobacteria. J. Gen. Microbiol. 130, 2177-2191. Mangum, C. P., Silverthorn, S. V., Harris, J. L., Towle, D. W., and Krall, A. R. (1976). The relationship between blood pH, ammonia excretion and adaptation to low salinity in the blue crab Cullinectes supidus. J. Exp. 2 0 1 . 195, 129-136. Measures, J. C. (1975). Role of amino acids in osmoregulation of non-halophilic bacteria. Nature (London) 257, 398-400. Ming, S. L., and Hellebust, J. A. (1976). Effects of salinity and osmolarity of the medium on amino acid metabolism in Cyclorellu crypricu. Can J . Bor. 54, 938-949. Moran, W. M . , and Pierce, S. K . (1984). The mechanism of crustacean salinity tolerance: Cell volume regulation by K + and glycine effluxes. Mar. Biol. 81, 41-46. Needham, A. E. (1957). Factors affecting nitrogen excretion in Curcinides muenus (Pennant). Physiol. Comp. Oecol. 4, 209-239. Otto, J., and Pierce, S. K. (1981). Water balance systems in Rungiu cuneuru: Ionic and amino acid regulation in changing salinities. Mar. Biol. 61, 185-192. Paillard, M., Leviel, F., and Gardin, J. P. (1979). Regulation of cell volume in separated renal tubules incubated in hypotonic medium. Am. J . Physiol. 236, F226-F23 1. Pang, P. K. T., Griffith, R. W., and Atz, J. W. (1977). Osmoregulation in elasmobranchs. Am. ZOO^. 17, 365-377. Parker, I. C. (1977). Solute and water transport in dog and cat red blood cells. In “Membrane Transport in Red Cells” (C. Ellory and L. Lew, eds.), pp. 427-465. Academic Press, New York. Parker, J. C., and Hoffman, J. F. (1976). Influences of cell volume and adrenalectomy on cation flux in dog red blood cells. Biochim. Biophys. Acru 433, 404-408. Pkqueux, A., Gilles, R., Pilwat, G., and Zimmermann, U. (1980). Pressure-induced variations of K + permeability as related to a possible reversible electrical breakdown in human erythrocytes. Experienriu 36, 565-566. Persoon, B. E., and Spring, K. R. (1982). Gallbladder epithelial cell hydraulic water permeability and volume regulation. J. Gen. Physiol. 79, 481-505. Peterson, M. B . , and Duen, F. G. (1969). Studies on osmotic adjustment in Tegulu funebrulis (Adams, 1854). Comp. Biochem. Physiol. 28, 633-644. Pichon, Y . , and Treheme, J. E. (1976). The effects of osmotic stress on the electrical properties of the axons of a marine osmoconformer (Muiu squinudu. Brachyura: Crustacea). J . Exp. Biol. 65, 553-563. Pierce, S . K. (1971). A source of solute for volume regulation in marine mussels. Comp. Biochem. Physiol. M A , 103-1 17. Pierce, S . K . (1982). Invertebrate cell volume control mechanisms: A coordinated use of intracellular amino acids and inorganic ions as osmotic solute. Biol. Bull. 163, 405-419. Pierce, S. K . , and Greenberg, M. J. (1972). The nature of cellular volume regulation in marine bivalves. J. Exp. B i d . 57, 681-692. Pierce, S. K.,and Greenberg, M. J. (1973). The initiation and control of free amino acid regulation of cell volume in salinity-stressed marine bivalves. J. Exp. Biol. 59, 435-440. Pierce, S. K . , and Minasian, L. L., Jr. (1974). Water balance of a euryhaline sea anemone, Diudumene leucolenu. Comp. Biochem. Physiol. 49A, 159- 167. Pierce, S. K., Warren, M. K., and West, H. H. (1983). Non amino acid mediated volume regulation in an extreme osmoconformer. Physiol. Zool. 56, 445-454. Pierce, S . K.,Edwards, S . C., Mazzocchi, P. H., Klinger, L. J., and Warren, M. K. (1984). Proline betaine: A unique osmolyte in an extreme euryhaline osmoconformer. Biol. Bull. 167, 495500.
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
245
Potts, W. T. W. (1958). The inorganic amino acid composition of some lamellibranch muscles. J. Exp. Biol. 35, 749-764. Potts, W. T. W . , and Parry, G . (1964). “Osmotic and Ionic Regulation in Animals.” Pergamon, Oxford. Powell, E. N., Kasschau, M., Chen, E.. Koenig, M . , and Pecon, J. (1982). Changes in the free amino acid pool during environmental stress in the gill tissue of the oyster Crassosrrea virginica. Comp. Biochem. Physiol. 71A, 591-598. Pressley, T. A., and Graves, J. S. (1983). Increased amino acid oxidation in the gills of blue crabs acclimated to dilute sea water. J. Exp. Zool. 226, 45-5 I . Pressley, T. A., Graves, 1. S., and Krall, A. R. (1981). Amiloride-sensitive ammonium and sodium ion transport in the blue crab. Am. J. Physiol. 241, R370-R378. Prior, D. J . (1981). Hydration related behavior and the effects of osmotic stress on motor function in the slugs, Limax maximus and L i m pseudofavus. Adv. Physiol. Sci. 23, 131-145. Prior, D. J., and Pierce, S. K. (1981). Adaptation and tolerance of invertebrate nervous systems in osmotic stress. J. Exp. Zool. 215, 237-245. Reuben, J. P., Girardier, L., and Grundfest, H. (1964). Water transfer and cell structure in isolated crayfish muscle fibers. J. Gen. Physiol. 47, 1141-1174. Reynoso, G . T., and de Gamboa, B. A. (1982). Salt tolerance in the freshwater algae Chlamydomonas reinhurdiic Effect of proline and taurine. Comp. Biochern. Physiol. 73A, 95-99. Roesijadi, G., Anderson, J . W., and Giam, C. S. (1976). Osmoregulation of the grass shrimp Palaemoneres pugio exposed to polychlorinated biphenyls (PCB’s). 11: Effects on free amino acids of muscle tissue. Mar. Biol. 38, 357-363. Rorive, G., and Gilles, R. (1979). lntracellular inorganic osmotic effectors. In “Mechanisms of Osmoregulation in Animals” (R. Gilles, ed.), pp. 83-109. Wiley (Interscience), London. Roti-Roti, L. W., and Rothstein, A. (1973). Adaptation of mouse leukemic cells (L5178Y) to anisotonic media. 1. Cell volume regulation. Exp. Cell Res. 79, 295-310. Schleich. T., and Von Hippel. P. H. (1970). Ion-induced water-proton chemical shifts and the conformational stability of macromolecules. Biochemisrv 9, 1059- 1066. Schmidt-Nielsen, B. (1975). Comparative physiology of cellular ion and volume regulation. J. Exp. Zool. 194, 207-220. Schobert, B. (1974). The influence of water stress on the metabolism of diatoms. I: Osmotic resistance and proline accumulation in Cvclotella meneghiniana. Z. Planrzenphysiol. 74, 106120. Schobert, B. (1980). Proline catabolism, relaxation of osmotic strain and membrane permeability in the diatom Phaeodacrylum rricornurum. Physiol. Planr. 50, 47-52. Schoffeniels, E., and Gilles, R. (1963). Effect of cations on the activity of L-glutamic acid dehydrogenase. Life Sci. 2, 834-839. Schoffeniels, E., and Gilles, R. (1970). Osmoregulation in aquatic arthropods. In “Chemical Zoology” (M. Florkin and B. T. Scheer, eds.), Vol. 5 , Part A, pp. 255-286. Academic Press, New York. Schoffeniels, E., and Gilles, R. (1972). lonoregulation and osmoregulation in Mollusca. In “Chemical Zoology” (M. Florkin and B. T. Scheer. eds.), Vol. VII, pp. 393-420. Academic Press, New York. Seel. M., Rorive, G., Pkqueux, A., and Gilles, R. (1980). Effect of hypo-osmotic shock on the volume and ion content of rat kidney cortex slices. Comp. Biochem. Physiol. 65A, 29-33. Shaw, J. (1958). Osmoregulation in the muscle fibers of Carcinus. J. Exp. B i d . 35, 920-929. Shick, J. M. (1976). Ecological physiology and genetics of the colonizing actinian Haliplanella luciae. In “Coelenterate Ecology and Behavior” ( G . 0. Mackie, ed.), pp. 137-146. Plenum, New York. Siebens, A . W., and Kregenow, F. M. (1978). Volume regulatory responses of salamander red cells incubated in anisosmotic media: Effect of amiloride. Physiologisr 21, 110
246
R. GILLES
Siebers, D., Lucu, C., Sperling, K. R., and Eberlin, K. (1972). Kinetics of osmoregulation in the crab Carcinus maenas. Mar. Biol. 17, 291-303. Skaer, H. L. (1974).The water balance in a serpulid polychaete, Mercierella enigmarica (Fauvel). 1. Osmotic concentration and volume regulation. J . Exp. Biol. 60,35 1-370. Skaer, H. L., Treheme, J. E., Benson, J. A,, and R. B. Moreton (1978). Axonal adaptations to osmotic and ionic stress in an invertebrate osmoconformer (Mercierella enigmara Fauvel). 1. Ultrastructural and electrophysiological observations on axonal accessibility. J. Exp. Biol. 76, 19 1-204. Spaargaren, D. H. (1979). Marine and brackish water animals. In “Comparative Physiology of Osmoregulation in Animals” (G. M. 0. Maloiy, ed.), Vol. I , pp. 83- 116. Academic Press, New York. Stephens, G. C., and Virkar, R. A. (1966). Uptake of organic material by aquatic invertebrates. IV. The influence of salinity on the uptake of amino acids by the brittle star, Ophiactis arenosa. Bi01. Bull. 131, 172-185. Stoner, L. C . , and Dunham, P. B. (1970). Regulation of cellular osmolarity and volume in Tetruhymena. J. Exp. Biol. 53, 391-399. Storey, R., and Wyn Jones, R. G. (1977). Quaternary ammonium compounds in plants in relation to salt tolerance. Phytochemisrry 16, 447-453. Strange, K. B., and Crowe, J. H. (1979a). Acclimation to successive short term salinity changes by the bivalve Modiolus demissus. I. Changes in hemolymph osmotic concentration, hemolymph ion concentration and tissue water content. J . Exp. Zool. ?lo, 221-226. Strange, K. B., and Crowe, J. H. (1979b). Acclimation to successive short term salinity changes by the bivalve Modiolus demissus. 11. Nitrogen metabolism. J . Exp. Zool. 210, 227-236. Tal, M., Rosenthal, I., Abramovitz, R., and Forti, M. (1979). Salt tolerance in Simmondsia chinensis: Water balance and accumulation of chloride, sodium and proline under low and high salinity. Ann. Bor. 43, 701-708. Thurston, J. H., Hauhart, R. E.. and J. A. Dirgo (1980). Taurine: A role in osmotic regulation of mammalian brain and possible clinical significance. Life Sci. 26, 1561-1568. Thurston, J. H., Hauhart, R. E., and Naccarato, E. F. (1981). Taurine: Possible role in osmotic regulation of mammalian heart. Science 214, 1373-1374. Treherne, J. E. (1980). Neuronal adaptations to osmotic and ionic stress. Comp. Biochem. Physiol. 67B, 455-463. Treheme, J. E. (1985). Neuronal adaptations to osmotic stress. In “Transport Processes, Ion0 and Osmoregulation” (R. Gilles and M. Gilles-Baillien, eds.), pp. 376-388. Springer-Verlag. Berlin. Treherne, J. E.,and Pichon, Y. (1978). Long term adaptation of Sabella giant axons to hypoosmotis stress. J. Exp. Biol. 15. 253-263. Turano, C . , Fasella, P., and Giartosio, A. (1962). On the effect of small ions on the activity of glutamic-aspartic transaminase. Biochim. Biophys. Acra 58, 255-261. Ussing, H. H. (1982). Volume regulation of frog skin epithelium. Acra Physiol. Scand. 114, 363369. Vincent-Manque, C., and Gilles, R. (1970a). Modification of the amino acid pool in blood and muscle of Eriocheir sinensis during osmotic stress. Comp. Biochem. Physiol. 35, 479-485. Vincent-Marique, C., and Gilles, R. (1970b). Changes in the amino-acid concentration in blood and muscle of Eriocheir sinensis during hypoosmotic stress. Life Sci. 9, 509-512. Virkar, R. A. (1966). The role of free amino acids in the adaptation to reduced salinity in the sipunculid Golfingia gouldii. Comp. Biochem. Physiol. 18, 61 7-625. Virkar, R. A., and Webb, K. L. (1970). Free amino acid composition of the soft shell clam Mya arenaria in relation to salinity of the medium. Comp. Biochem. Physiol. 32, 775-783. Vislie, T. (1980). Hyper-osmotic cell volume regulation in vivo and in virro in flounder (Plarichrhys flesus) heart ventricles. J . Comp. Physiol. 140, 185-191.
VOLUME REGULATION IN EURYHALINE INVERTEBRATES
247
Vislie. T . , and Fugelli, K. (1975). Cell volume regulation in flounder (Plarichthys Jesus) heart muscle accompanying an alteration in plasma osmolality. Comp. Biochem. Physiol. 52A, 4 15418. Von Hippel, P. H. (1976). Neutral salts effects on the conformational stability of biological macromolecules: Model studies. I n “L’eau et les systemes biologiques-colloques CNRS n”246” (A. Alfsen and A. J. Berteaud, eds.), pp. 19-26. CNRS, Paris. Von Hippel, P. H., and Schleich, T. (1969a). The effects of neutral salts on the structure and conformational stability of macromolecules in solution. In “Biological Macromolecules” (G. Fasman and S . Timasheff, eds.), Vol. 11, pp. 417-574. Dekker, New York. Von Hippel, P. H.. and Schleich, T. (1969b). Ion effects on the solution structure of biological macromolecules. Arc. Chem. Res. 2, 257-265. Von Hippel, H. T., and Wong, K. Y. (1963). The collagen-gelatin phase transition. Biochemisrry 2, 1387- 1398. Von Hippel, P. H., and Wong, K. Y. (1965). On the conformational stability of globular proteins: The ribonuclease transition. J. B i d . Chem. 240, 3909-3923. Wanson, S., Pkqueux, A., and Gilles, R. (1983). Osmoregulation in the stone crab Cancerpagurus. Mar. B i d . Lett. 4, 321-330. Warren, J. C., and Cheatum, S . G. (1966). Effect of neutral salts on enzyme activity and structure. Biochemistry 5 , 1702- 1707. Warren, J . C., Stowring, L., and Morales, M. (1966). The effect of structure-disrupting ions on the activity of myosin and other enzymes. J. B i d . Chem. 241, 309-316. Warren, M. K . , and Pierce, S. K. (1982). Two cell volume regulatory systems in the Limulus myocardium: An interaction ions and quaternary ammonium compounds. Biol. Bull. 163,504516. Webb, K . L., Schimpe, A. L., and Olmon, J. (1972). Free amino acid composition of scyphozoan polyps of Aurelia aurita, Chrysaora quiquecirrha and Cyanea capillara at various salinities. Comp. Biochem. Physiol. 43B, 653-663. Weber, R. E.. and Van Marrewijk, W. J. A. (1972). Free amino acids in the shrimp Crangon crangon and their osmoregulatory significance. Neth. J. Sea Res. 5 , 391-415. Wickes, M. A , , and Morgan, R. P. (1976). Effect of salinity on three enzymes involved in amino acid metabolism from the American oyster, Crassorrea virginica. Comp. Biochem. Physiol. 53B, 339-343. Willmer, P. G. (1978). Electrophysiological correlates of ionic and osmotic stress in an osmoconforming bivalve. J. Exp. Biol. 77, 181-205. Yancey. P. H., Clark, M. E., Hand, S. C., Bowlus, R. D., and Somero, G. N. (1982). Living with water stress: Evolution of osmolyte systems. Science 217, 1214- 12222. Zimmermann, V., Pilwat, G . , Pequeux, A,, and Gilles, R. (1980). Electro-mechanical properties of human erythrocytes membranes: The pressure-dependence of potassium permeability. J. Membr. B i d . 54, 103-113.
This Page Intentionally Left Blank
Part 111
Physicochemical Prospectives
This Page Intentionally Left Blank
CURRENT TOPICS Pi MEMBRANES AND TRANSPORT, VOI.UME .W
Non-Donnan Effects of Organic Osmolytes in Cell Volume Changes MARY E . CLARK Department of Biology College of Sciences San Diego Stare University San Diego. Calfornia 92182
1.
INTRODUCTION
This article concerns itself with the changes in cell volume that are attributable not only to the total number of osmotically active substances in a cell, but to the specific chemical nature of those substances. Strictly speaking, changes in cell volume could come about through changes in the total mass of the molecular species contained within it, or through changes in the specific gravity of one or more species: that is, by changing the density of the system. Since the compressibility of the liquid and solid components of cells is negligible even at high pressures, it is assumed for the purpose of this article that changes in specific gravity are inconsequential relative to changes in mass of the cell components, particularly the major intracellular species, water. We shall therefore assume that the significant changes in volume that intact cells may undergo are primarily responses to disparities in water activity between the intracellular and extracellular compartments. Furthermore, our concern in this article will be restricted to the physics of the equilibrium conditions of the cell interior, leaving aside any metabolic processes by which these conditions are achieved. As numerous other articles in this volume have indicated, most living cells maintain a steady state volume that is not at thermodynamic equilibrium with the surroundings owing to various ongoing, energy-consuming processes. In the past, perhaps because the classical osmotic studies were done on simple spherical cells such as the unfertilized eggs of marine invertebrates (Heilbrunn, 1943, p. 1 lo), it was concluded that cells behave as reasonably good osmometers and that the intracellular water activity is predictable from a summation of the 251
Copynghi 6 IYR7 by Academic Press, Inc All nghi- of mpmduclion in any form mxrved
252
MARY E. CLARK
soluble components in the cell, suitably corrected for their deviations from ideality (see, for example, Shaw, 1958). Such a theory assumes first that all solute particles, regardless of their species, perform similarly, in a thermodynamic sense, in terms of volume regulation, and second that all of the water in the cell serves as solvent for such solutes. That this view is indeed oversimplified should not surprise us, given our current knowledge of the complexity of the cell interior. The notion of a cell as an osmometer, however, provides us only with information on relative changes in volume, but not with why a particular cell should have a particular initial volume. For this purpose, a second and related physical model has been invoked, namely the theory of Donnan, where the interior of the cell is viewed not as a simple solution, but as a liquid or semisolid crystal of large, physically constrained polyions, the total volume of which is determined by the balance between attractive forces (van der Waals forces and, in the case of the myofilaments of muscle about which we will be concerned here, the Cproteins that vertically connect myosin filaments, and other elastic constraining elements) and repulsive forces (the density of exposed negative charges along the length of the polyions). The only function of the intervening solution is as a dielectric that modulates the repulsive force of the charges, but otherwise does not interact with the polyions. Volume changes are then predicted according to changes in the dielectric properties of the intervening solution which result from changes in ionic strength. (A good description of this model is given by Elliott, 1968; Elliott and Rome, 1969.) The shortcomings of this model as a complete description of cell volume changes is the subject of this article. Owing to their unique architecture, muscle cells are especially suited for testing this theory and we shall emphasize studies on volume changes in muscle fibers to indicate why neither regarding the cell as a simple osmometer nor regarding it as a charged gel in a state of Donnan equilibrium is a sufficient explanation of cell volume changes, although both of course contribute. A third factor involves the mutual interaction among solvent, solutes, and large molecules, which appears to influence the quality of macromolecular hydration. There is evidence that these interactions, which depend not only on the number but also on the properties of the solute molecules, also contribute to the activity of intracellular water. Finally, suggestions are made regarding potentially fruitful lines of further investigation.
II. DONNAN THEORY AS A MODEL A “Donnan equilibrium” is defined as a state of no net movement of solutes or of water between two phases: one phase consists only of mobile solvent (e.g., water) and of diffusible solutes, some or all of which are charged (e.g., ions),
253
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
and the other phase, of mobile solvent and solutes plus nonmobile, multiply charged molecules (e.g., proteins). The nonmobility of the latter is the consequence either of a bounding membrane that they cannot penetrate, or of intermolecular bonds that maintain their aggregated state (Fig. I ) , or of some combination of these. In the absence of either of these constraints on motion, a Donnan system does not exist. Instead, the macromolecular polyions and their counterions distribute themselves randomly among all of the mobile solvent and solute molecules. Such uniformity would have two consequences. First, the typical boundary (Donnan) potential that exists between the two phases would disappear. [One could, of course, treat each polyion as a separate phase and discuss the unequal distribution of mobile counter- and coions in its sphere of influence, as does Fumio Oosawa (1971), but this model clearly brings us to a different level of organization than that appropriate for a discussion of cell volume regulation. Nevertheless, certain considerations appropriate to Oosawa’s model are also appropriate in considerations of cell volume regulation, as we shall see.] Second, of course, without a phase boundary one can hardly speak of “volume” regulation. At any given ionic strength, a Donnan system always results in a greater concentration of mobile counterions within the phase containing the immobile polyions than in the bathing medium. This results in a tendency for that phase to swell, a tendency that is resisted by whatever forces are constraining the polyions. (Since the excess ions are mainly counterions of the polyions, there is also a Donnan potential at the phase boundary, as shown in Fig. 1, but this is of no direct importance to volume considerations.) If the ionic strength of the surrounding medium is increased, the osmotic differential between the external medium and the internal medium becomes relatively less, and hence the tendenA
B
FIG. 1 . Two simple models of Donnan systems: (A) Membrane bounded; (B)intermolecularly bonded aggregates. In each case, circles represent nonmobile polyanions, + and - signs are mobile counter- and coions. Note the existence of a Donna potential at the phase boundary.
254
MARY E. CLARK
0
0.5
1.0
1.5
2.0
2.5
Relative Osmolarity
FIG. 2. Theoretical volume changes in either a simple osmometer or a Donnan system in response to varying osmolarity or ionic strength of the bathing medium. When plotted against osmolarity-I, such a curve yields a positive linear relationship.
cy for swelling is also reduced. Provided that the total immobile charges remain constant, that the activity (yi) and osmotic ( c $ ~ )coefficients of the diffusible solutes are the same in both phases, and that the chemical potential of the macromolecules remains unchanged, then by application of the Boyle-van’t Hoff equation (e.g., Nobel, 1970), we can obtain the theoretical relation shown in Fig. 2; there is an inverse relationship between volume and ionic strength that strongly resembles that seen between volume and osmotic pressure for simple osmometers containing nondiffusible uncharged particles. How closely do muscle cells follow such a theoretical model?
A. Intact Muscle Fibers Over a decade ago, April and his colleagues (1972), using X-ray diffraction methods, examined the responses of the volume of the myofilament lattice in intact crayfish muscles in relation to the ionic strength of the bathing medium. (In all cases, the bathing medium employed was buffered potassium propionate.) Like the whole cell, the lattice unit cell behaved as an osmometer. The solid circles shown in Fig. 3 give a curve closely resembling the theoretical curve of
255
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
4.0
).
0
d.5
1.0
1.5
2.0
215
Relative Osmolarity
FIG. 3. Changes in the relative spacing in myofilament lattices in intact (0)and skinned (A) crayfish muscle cells in response to changes in osmolarity of potassium propionate bathing medium. Note that intact fibers behave as "good" osmometers, but skinned fibers show quite anomalous behavior. The points with error bars are the volumes in media isosmolar with intact cells. Note that upon removal of its membrane, a fiber swells by around 35%. [From April er al. ( 1972). Reproduced from The Journal of Cell Biology, 1972, Vol. 53, p. 59 by copyright permission of The Rockefeller University Press.]
Fig. 2. Thus one could conclude that the changes in cell volume were due to changes in the water content of the myofilament lattice, and that the lattice was obeying standard osmometric and Donnan expectations.
B. Skinned Muscle Fibers When the cell membrane is removed, either mechanically or chemically, from muscle fibers, the myofilaments remain intact and capable of generating tension. If cell volume changes are, indeed, due to changes in lattice volume, then removal of the membrane should not alter cell volume [except for a slight swelling owing to the entry of cations formerly excluded by active ion pumps (MacKnight and Leaf, 1977)l.After that, the system should still respond similarly to the curve in Fig. 2 with changing ionic strength. As the curve joining the triangles in Fig. 3 shows, however, this is not at all the case; the system swells,
256
MARY E. CLARK
rather than shrinks, with increasing ionic strength. (The swelling is not as marked in these experiments as in our later studies on barnacle muscle, to be discussed below, because propionate, a protein-stabilizing organic solute, was used as the anion in the bathing medium.) Clark et al. (1981) obtained similar results on chemically skinned muscle
0.5: I: SAL1 CONCEWTRATIOI
0:
1
1
0.1
0.2
I
3:
21
(I
BASIC SOUI1101)
1
0.3 0.1 MA0 (HOL/LIlER)
1
0.5
I 0.6
FIG.4. Effects of ions in the absence and presence of organic solutes on properties of barnacle muscle fibers. (A) Effect of osmotic strength on water content of skinned fibers equilibrated for 40 hr to salt solutions with and without 0.5 M TMAO. Basic (1 X ) salt solution was 50 mM NaC1, 150 mM KCI, 10 mM MgC12, which was halved, doubled, or tripled to give 0.5 X , 2.0X. and 3.0X solutions, respectively. All solutions were Ca2+ and H + buffered with 5 mM EGTA and 25 mM Trispropionate (pH 6.9-7.1 at termination). (B) Effects of increasing TMAO (in 1 X salt solution) on the water content of membrane-damaged fibers. TMAO has an approximately inverse linear effect on fiber swelling induced by neutral salts. (C) Changes in fixed charge (zrvr)on nonmobile myofilament proteins in response to ionic strength. (z, = valence and Y, = concentration of charged polyions;
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
257
hence, z,v, is the product of charged sites times concentrationsof the myofilament proteins.) In the presence of rising concentrations of salt, the fixed charge increases (whether or not TMAO is present) as protons are dissociated. (D) Effect of increasing salt concentration on excess ion accumulation in a barnacle muscle fiber. When only salt is present, the anions are excluded in proportion as cations are included to compensate for the growing negative fixed charge on myofilaments. If 0.5 M TMAO is added, a large increase of free cations occurs which "cancels" the eltchorepulsive effects of the deprotonation of myofilaments induced by salt. TMAO facilitates counterion binding to fixed charge sites. [All from Clark er a/. (1981). courtesy of the Journal ofExperimenrar Biology.]
258
MARY E. CLARK
500 nm FIG.5 . Electron micrographs of cross-sections of barnacle muscle myofibrils in the region of thick-thin filament overlap. Top, intact fiber. Rest are skinned fibers exposed 40 hr to varying ionic strengths, with or without 0.5 M TMAO. [After Clark et al. (1981), courtesy of the Journal of Experimental Biology. ]
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
259
fibers of the giant barnacle, Balanus nubilis. They equilibrated fibers with bathing media containing the chlorides of potassium, sodium, and magnesium in the relative ratios in which these cations occur in the cell, and observed changes in their water and ion content relative to the nonsoluble (i.e., myofilament) dry weight. As Fig. 4A shows, there is considerable swelling with increasing ionic strength. Ion analyses of the fibers indicate that the net myofilament charge increases with ionic strength, owing to the expected (see Oosawa, 1971, Article 8) deprotonation of weak acid residues (Fig. 4C), and that this charge increase is met by an approximately equal exclusion of anions and inclusion of cations within the myofilament lattice (Fig. 4D), as expected by Donnan theory when both ions have similar mobilities. These results are compatible with the expected tendency toward salting in of the major myofilament proteins (von Hippel and Schleich, 1969), an effect that was indeed confirmed by electron microscopy (Fig. 5). The initial modest swelling in isosmotic solutions, observed whenever cell membranes are damaged, is generally ascribed to the influx of cations previously excluded by the sarcolemma, as already noted above. This swelling, however, may also be due, along with the abnormal changes in cell volume observed with changing ionic strength, to the substitution of perturbing solutes (ions) for nonperturbing or stabilizing solutes (various organic molecules) that normally account for a large proportion of the cell’s osmotic activity.
111.
PERTURBING AND STABILIZING SOLUTES
The concept of stabilizing solutes was first proposed by Clark and Zounes (1977) and by Brown and Borowitzka (1979), the latter using the term “compatible solutes.” It was Clark and Zounes who noted that amino acids, taurine, and methylamines that account for up to 70% of the total intracellular osmotic activity in marine invertebrates may exert their stabilizing effects through their resemblance to the stabilizing cations [(CH,),N +,NH,+] and anions (SO,2-, CH,COO-) of the Hofmeister or lyotropic series. Relative to such neutral salts as sodium and potassium chloride, these favorable ions tend to cause salting out of proteins and to stabilize their native conformations (von Hippel and Schleich, 1969). In the intervening years, numerous tests of this hypothesis have been undertaken, and the results uniformly show that such organic solutes stabilize both the native conformation and the function of a variety of enzymes and other proteins (see review by Yancey et a l . , 1982). We continue here with the case of barnacle muscle as an example.
A. Muscle Cell Components An analysis of the dry-weight components of the muscle of Balanus nubilis is shown in Table I. Note that only 50% of the fiber dry weight is nonsoluble
260
MARY E. CLARK TABLE I INTRACELLULAR DRYWEIGHT C o M m N m s OF Balanus nubilis MUSCLEF I B E R S ~ . ~
Component
mOsm
(4)
Dissolved solids Low molecular weight Ions Na
Organics Amino acids + taurine TMAO + betaine Unidentified (by difference) Subtotal, organics Subtotal, osmotically active solutes
Grand total
1.29 7.04 1.24 9.51
Percentage dry weight
0.37 2.02 0.36 2.75
244 (0.9) 503 (1.0) 94 (1.15) 149 (1.0)
503 82 149 (est)
50.9 9.3 25.2
14.63 2.67 7.23
746
134 (est)
85.4
24.53
95.0
27.3
45.0 35.1 80.I
12.9 10.1 23 .O
990
1005
High molecular weight Protein Other Subtotal, high-molecularweight solutes Total: All dissolved solids Nonsolute solids Fat Mg Ca Other (mainly myofilament protein) Total: Nonsolute solids
g/kg H20
35-56 158- 180 30-35 27 1
K CI Subtotal, ions
mmollkg H20
16 1
175
50.3
11.00 0.39 0.16 161.5
3.2 0.1 0.04 46.4
173
49.7
348
100.0
a Based on 75% total fiber water content and 6% of water and nonsolute dry weight in extracellular space. b Adapted from Clark er al. (1981), courtesy of the Journal ofExperimenta1 Biology.
matter, mainly myofilament proteins. Of the soluble matter, fully half is small organic osmolytes. Although the inorganic ions comprise approximately onequarter of the osmolarity (244 mOsm), they contribute to less than 3% of the dry weight. About three-quarters of the osmolarity (746 mOsm) is due to amino acids, methylamines, and metabolic intermediates. The remaining soluble matter
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
26 1
is cytoplasmic protein and other macromolecules, either free or conjugated (Clark and Hinke, 1981). By means of the skinned muscle preparation, Clark and colleagues (1981) examined the role of the organic solutes vis-a-vis that of ions in the hydration of muscle fibers and the stabilization of myofilament architecture. Of the naturally occurring organic solutes, trimethylamine oxide (TMAO) was chosen as primary model since it is uncharged and hence requres no counterion.
6. Organic Solutes Offset Salting In The most startling result of adding 0.5 M TMAO to each of the various salt solutions in which skinned fibers were equilibrated was the total elimination of the observed swelling with increasing ionic strength (Fig. 4A). At constant ionic strength, this effect was approximately linear with TMAO concentration (Fig. 4B). Thus, the TMAO titrates the effects of neutral salts on myofilament hydration. As Fig. 5 shows, the salt-induced hydration is coupled with a gradual disruption of myofilament architecture that the presence of TMAO almost entirely prevents. The effect of TMAO does not seem to be mediated by any significant inhibition of the salt-induced deprotonation of myofilament proteins (Fig. 4C; other experiments showed no change at all in net myofilament charge, zp,, upon addition of TMAO to the various bathing solutions). On the other hand, the presence of TMAO had an unexpected effect on the distribution of cations and anions within the fiber. Instead of an approximately equal exclusion of anions and inclusion of cations at all ionic strengths, the addition of TMAO resulted in a massive increase in cations, but no change in the exclusion of anions (Fig. 4D). This leads to an apparent paradox: despite this TMAO-induced excess of supposedly osmotically active particles in the fiber relative to the bath, the fiber does not swell. The failure of these excess ions to exert an osmotic effect is discussed below. Clark et al. (1981) carried out similar studies using other organic solutes that accumulate in the cells of various water-stressed organisms (see Yancey er al., 1982, for species): the amino acids glycine, alanine, serine, and proline; glycerol; and urea. Of these, urea proved ineffective in offsetting the perturbations caused by neutral salts. [This result is expected, given the general destabilizing effect of urea on macromolecular structure and function (Yancey and Somero, 1979)l. Indeed, in species where urea is accumulated intracellularly, it is almost always offset by the presence of srabilizing solutes, particularly methylamines (Yancey and Somero, 1979, 1980). The remaining solutes, however, shared in varying degree with TMAO its ability to offset ion-induced swelling and to cause excess ion accumulation in the fibers. The organic solutes themselves were slightly excluded from the fiber relative to the bath (0.88X to
262
MARY E. CLARK
0.92x), except for TMAO which was slightly included ( 1 . 0 5 ~ (Clark ) et al., 1981).
IV. MECHANISMS OF ACTION OF STABILIZING SOLUTES The explanation of these phenomena is undoubtedly complex, depending on a number of interacting factors among the myofilaments, the solvent, the perturbing ions, and the stabilizing solutes. Various parts of the answer may be hypothesized by focusing on one or another particular interaction, although it is too soon to attempt a quantitative explanation that permits assigning exact chemical potentials to the several components. Here we focus on three phenomena that may be important for the action of stabilizing solutes.
A. Counterion Binding The increased excess of univalent cations in the fiber in the presence of organic solutes gave rise to the proposal that organic solutes may enhance the ’
1 .o
0.9
0.a
B 0.7
0.6
0.5
0
0.1
0.1
0.3
0.5
0.4
(P
FIG. 6 . The degree of dissociation (p) of counterions from fixed charges on polyions (e.g., myofilarnent proteins) as a function of their apparent partial volume, As p decreases (a larger fraction of fixed charges on the polyions have counterions bound to them), the fractional space of the polyions increases. The relationship varies with Q, a parameter incorporating the density of charges along the polyion and the dielectric constant of the intervening medium. For myofilament proteins in a 1 X salt solution, Q = 0.5. The point corresponds to the water content of a fiber in 1 X salt + 0.1 M TMAO and the arrowhead corresponds to that of a fiber in 1 X salt + 0.6 M TMAO (from Fig. 4B). [After Clark et al. (1981), courtesy of the Journal ofExperimenta1 Biology.]
+.
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
263
association of these cations with the fixed charge sites on the myofilaments, thus decreasing charge repulsion. Employing the relationship provided by Oosawa (1971) between the degree of dissociation (p) of mobile counterions and the apparent volume fraction (+) of cylindrical charged macromolecules in an aqueous system, Clark et al. (1981) demonstrated that the increase in cations in the fiber in the presence of TMAO, if only modestly bound to the fixed charges of actin and myosin, was sufficient to account for the observed decrease in fiber hydration (Fig. 6). Counterion binding was shown to be a potentially much more significant factor in fiber hydration than any changes in dielectric constant of the medium insofar as effects on charge repulsion are concerned. In the case of myofilament proteins then, at least some of the osmolyte effects may be mediated directly at the biopolymer surface.
B. Solute Exclusion from the Protein Domain It has been shown that stabilizing solutes, including salting-out salts of the Hofmeister series [e.g., (NH,),SO,], glycerol, and TMAO and other methylamines, have algebraically additive effects vis-a-vis destabilizing solutes, such as perturbing ions (e.g., Li , Ba2 , I - , CNS - , and guanidinium ion) and urea (see Yancey et al., 1982, for discussion). For example, approximately one TMAO molecule offsets the effect of two urea molecules over a wide range of concentrations, in terms of stabilizing protein unfolding, maintaining a given melting temperature, and keeping constant the kinetic parameters (V,,, and K,) of enzymes (Yancey and Somero, 1979, 1980). Methylamines also favor the native aggregated state of oligomeric proteins (Hand and Somero, 1982). The unifying relationship in all these stabilizing effects is the reduction of the protein surface requiring hydration to a minimum, which coincides with its most compact, native, folded state. The nature of these generalized solute effects on macromolecular hydration has been explored mainly in the laboratory of Serge N. Timasheff. Whereas denaturing organic solutes tend to “bind” to those groups normally sequestered in the interior of proteins (alcohols to hydrophobic groups, and guanidinium ion and urea to peptide groups), certain stabilizing solutes such as glycerol tend to make water less available for protein hydration, and hence to minimize disaggregation or unfolding. Densitometric studies on the equilibrium exclusion of glycerol from g1ycerol:water mixtures containing varying concentrations of four soluble proteins were carried out by Gekko and Timasheff (1981a,b). The water layer surrounding a protein is structurally unsuited as a solvent for glycerol, the extent of the exclusion being dependent on the nature of the protein surface; less polar proteins exclude glycerol more strongly. The result is essentially a twophase system (although likely without a sharp boundary) where the glycer+
+
264
MARY E. CLARK
o1:water mixture is rich in water near the protein and rich in glycerol further from its surface. The negative entropy of such a separation is thermodynamically unfavorable. The chemical potential of the protein has effectively been raised, leading to whatever folding or aggregation of subunits that result in a reduction of its hydrated surfaces. Similar results have been obtained with other polyalcohols, including the sugars sucrose, lactose, and glucose (Lee and Timasheff, 1981; Gekko and Morikawa, 1981a,b; Arakawa and Timasheff, 1982a). This technique has more recently been applied to the study of exclusion of various salts (Arakawa and Timasheff, 1984a,b) and organic osmolytes (Arakawa and Timasheff, 1983, 1985) from the hydration domain of proteins. In the case of salts, the Hofmeister series was followed for both cations and anions, with the most stabilizing ions the ones most strongly excluded from the hydration sphere of the protein. Although the divalent cations tend to be weakly excluded from this sphere, their ability to bind tightly to charged sites on the protein surface readily overcomes this weak stabilizing effect (so the authors argue), causing a decrease in preferential hydration and a tendency toward unfolding and denaturation (Arakawa and Timasheff, 1984a). Algebraic additivity occurs in the case of guanidine sulfate, where the strongly stabilizing sulfate ion more than offsets the destabilizing effect of the guanidinium ion, resulting in net stabilization of protein (bovine serum albumin) in the presence of this neutral salt (Arakawa and Timasheff, 1984b). A similar result was obtained for several organic solutes: the amino acids glycine, a-and p-alanine, proline, serine, y-aminobutyric acid, and taurine; the methylamines sarcosine and betaine (Arakawa and Timasheff, 1983, 1985). All are excluded from the hydration sphere of the protein (lysozyme), and all are effective in raising the enzyme’s thermal denaturation temperature, indicating a stabilizing effect on its structure. A similar slight exclusion from the domain of the myofilaments in barnacle muscle was also observed for amino acids and glycerol, although not for TMAO (Clark et af., 1981). Two other solutes have given anomalous results, however. The amino acid valine (never a prominent amino acid intracellularly in species having high osmoticities) is excluded from the protein’s domain, but does not stabilize its structure (Arakawa and Timasheff, 1985). And TMAO, which has repeatedly been found to stabilize proteins, rather than being excluded, was found to be slightly concentrated within the barnacle fiber (Clark et af., 1981). To hypothesize about these observations we need first to turn to the structure of water. C. Water Structure and Protein Hydration Although there is not space here to review all of the studies bearing on this topic, it is pertinent to make two or three salient points regarding the interactions
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
265
between proteins and water that could bear on the ways that stabilizing solutes bring about their effects. First, theory regarding the structure of liquids, including water, although still far from satisfactory, has been gradually advancing. Of particular interest is the recent principle of multiple potential energy minima put forth by Stillinger and Weber (1984). In the old model, the short-range order and long-range disorder in a highly structured liquid such as water was envisioned as due to slightly imperfect tetrahedrons of nearest neighbors surrounding each molecule, with a statistical distribution of degrees of imperfection spread randomly throughout the system. There was, however, a single basic structure toward which the system tended and from which it was constantly disturbed by random thermal kinetic forces. The new principle is much more interesting for our purposes since it assumes that many different ordered structures are possible in liquids and that they all have such similar free energy minima that there is little thermodynamic preference among them. They randomly coexist in interchanging patches. Water, for example, is thus a mosaic of many structural tendencies. Furthermore, these various forms of ordering are all more or less equally perturbed by heat and so persist throughout the liquid state. Finally, having similar kinetic energies, the molecules in these variously structured arrangements would have similar rotational correlation times and hence would not be readily distinguishable in nuclear magnetic or dielectric relaxation studies, a point of importance as we shall see shortly. This principle immediately provides biologists with a new way of thinking about protein hydration. If different kinds of solutes can be incorporated easily into one kind of water structure but not another, then a variety of phases of quite similar free energy may become stabilized and coexist without readily mixing; this of course raises the free energy of the overall system, creating a thermodynamically less favored state. We can thus imagine a protein surface creating around itself a large hydration domain comprising one or two types of water structure to accommodate its polar and nonpolar patches. Most of this water would not need to be tumbling measurably slower than the water in the surrounding bulk phase. In fact, this concept neatly explains the great differences that are observed between the amount of severely motionally restricted “bound” water measured by NMR relaxation methods (Bumell et al., 1981) and the amount of thermodynamically “bound” water measured, for instance, by vapor adsorption (4 times greater) (Bull and Breese, 1968), or the amount of “bound“ water detected by self-diffusion coefficient measurements (10 times greater) (Clark et al., 1982). We can now examine hypothetical mechanisms by which osmolytes can influence protein hydration, and hence cell volume. A number of possibilities exists, but none seems to unequivocally account for all cases.
266
MARY E. CLARK
1. CHANGES IN SURFACE FREEENERGY
Arakawa and Timasheff (1985) propose that most of the stabilizing solutes are acting through their effects on water structure by forming stronger hydrogen bonds, on average, with water molecules than the latter do with each other; this is reflected in a rise in the surface tension of the solution, as solvent molecules are drawn more strongly into the solution. In this view, the stabilization of native protein structure is explained by a competition between protein and added solute for the available water. This, however, is an incomplete statement since it does not explain why “competition” should exist. What is the basis for the frequently observed exclusion of solute from the domain of the protein? The water around the protein must be differently structured from the bulk water or else it would accommodate protein and solute almost equally well. Only solutes that cannot “substitute” reasonably well for water molecules are capable of competing with proteins for hydration water. An even more curious problem is the fact that the amino acid valine is excluded from the protein domain, yet is not at all stabilizing of protein structure. Is hydrophobicity at one of its poles a sufficient explanation? There are, after all, strong structural resemblances between valine and glycine betaine, which is also excluded from the protein domain but has a strong stabilizing effect. There is also the fact that strongly destabilizing salts such as MgCl, and especially CaCl,, although they, too, raise the surface tension of aqueous solutions, are neither excluded from the domain of the protein nor help maintain its native structure (Arakawa and Timasheff, 1982b). Finally, there is the fact that molecules such as glycerol and betaine, that have no effect at all on surface tension (or may even decrease it) are both excluded from the domain of the protein and act to stabilize native protein structure. In what sense are they “competing” for hydration water? Although it is possible to generate more than one hypothesis to explain these discrepancies, this seems a futile exercise at this stage. It is surely more useful to explore first the interactions between the solvent water and various stabilizing and destabilizing solutes in order to form a clearer picture of the kinds of water structure accommodated by each. In this way it may eventually be possible to evaluate more fully the structural and thermodynamic interactions among solvent, osmolytes, and macromolecules. Discovering different and noncompatible forms of liquid water structure is not a simple task, yet one approach has been to look at self-diffusion coefficients of solvent in the presence of various solutes.
2. SELF-DIFFUSION AND WATERSTRUCTURE The inability of molecules such as the amino acids and polyalcohols to freely enter the hydration domain of a protein implies a strongly ordered interaction
267
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
between water and such solutes which in some way differs from the interaction between water and proteins. Here we turn again to the concept of Stillinger and Weber of multiple, thermodynamically almost equivalent yet geometrically incompatible domains of water structure. Identifying differences in water structure among such domains will not be simple since thermodynamically dependent properties (e.g., vapor pressure) will be indistinguishable, given that the water molecules are at the same chemical potential. More productive may be studies of molecular motions, in particular those of translational motion which may provide clues to different water structures. Along these lines is the work of Clark et al. (1982), who by measuring proton magnetic resonance relaxation rates in the presence of applied field gradients, estimated the relative self-diffusion coefficient (9/9,Jof water in solutions of various combinations of sarcoplasmic solutes, as well as in variously treated barnacle muscle fibers. Their most significant finding was that not only does the water associated with soluble proteins (e.g., bovine serum albumin) exhibit an expected reduction in 9/Ch0owing to an ordering of water molecules in its hydration domain, but so do solutions of various organic solutes, such as propionate ion, glycerol, TMAO, and a mixture of eight amino acids that are dominant in the sarcoplasm (Fig. 7). Apparently only translational motions are affected in the presence of these organic solutes; since spin-spin (T,) relaxation times are reduced only in the presence of exchangeable solute protons, rotational motions are evidently not affected. In one molar solutions of small organics, diffusional motions of water molecules are reduced between 20 and 30%. A similar finding of a reduction in 9/9was ,,made on mixtures of water and organic liquids by Goldammer and Hertz (1970). They showed that this reduction is not due to the formation of long-lived hydration spheres. Zeidler (1974)
I
o
., 0.2
0.4
0.6
0.8
1.0
Solute Concentration (M)
FIG. 7 . Relative self-diffusion coefficient of water, 9/9d,, as a function of organic solute concentration. A, TMAO; @, glycerol; amino acid mixture similar to that in intact barnacle sarcoplasm. (Modified from Clark er al.. 1982. Reproduced from the Biophysical Journal, 1982, Vol. 39, pp. 294 and 295 by copyright permission of the Biophysical Society.)
268
MARY E. CLARK
has suggested that organic solutes are able to rearrange the hydrogen bonding of solvent molecules over long distances. Thus there emerges a picture of large numbers, perhaps as many as a hundred, water molecules being moderately affected in their translational motions by each organic solute molecule. The effects of organic solutes on water translational motion can be contrasted with that of neutral inorganic salts. The ions most commonly found intracellularly (K+ ,C1- ,Na+) have little effect on 9/9at ,concentrations less than 1 M ,and among the ions that do markedly reduce 9/53,, the anions are stabilizing (acetate, sulfate) and the cations are destabilizing (Ca2+, Mg2+) in the Hofmeister series (McCall and Douglass, 1965). There thus appear to be several different kinds of water ordering in the presence of various solutes. In their studies, Clark et al. (1982) found that when various sarcoplasmic solutes (inorganic salts, small organic molecules, and proteins) are combined, their effects on 9l9, of water are additive. This permits the reassessment of “bound” water around proteins in the absence and presence of organic solutes, where bound water is now that water within a specially organized domain characterized by reduced translational motion. These workers found in both intact fibers and in skinned fibers exposed only to buffered water that this aqueous domain around the myofilament proteins amounted to 0.65 g/g. (This is in marked contrast to the 0.07 g/g of severely motionally restricted “bound” water calculated from the relaxation studies of Burnell et al., 1981.) As might be expected, this domain increased markedly (to about 1S O g/g) when fibers were exposed to an inorganic salt solution, a finding consistent with a partial unfolding of protein and exposure of surfaces that require further hydration. What is interesting is that in the presence of TMAO (with or without added salt), although total fiber water increased only marginally, the “bound” water presumably associated with the myofilaments increased some 50% (to about 0.95 g/g). This suggests a possible nonadditive, three-way interaction among this particular stabilizing solute, protein, and water whereby an increase in total water ordering is imposed, leading to a less favorable (higher) free energy state. 3. INTERACTIONSBETWEEN SALTSAND ORGANIC SOLUTES
Although most of the effects of solutes on protein hydration are probably mediated indirectly through changes in water structure, an examination of the possible interaction of stabilizing solutes with potentially destabilizing ions should not be ignored. As noted, TMAO, for example, has an enormous ability to offset their destabilizing effects on myofilament proteins. In this respect, preliminary unpublished studies in our laboratory suggest that in 0.5 M TMAO solutions containing various concentrations of KCl, the K+ activity is reduced by 30% and C1- activity by about 5% when measured with ion-selective electrodes. This effect seems most likely due to the sequestering of ions, especially cations, in the vicinity of this enormously polar molecule.
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
269
Thus, TMAO may be having a double effect on the denaturing capacity of ions. By sequestering ions in its own vicinity, TMAO makes them less able to interact with the protein. Second, by making the water around the protein more specifically structured, TMAO may be increasing ion activities in the vicinity of the protein surface, thus favoring counter-ion binding. Clearly, these are highly speculative hypotheses.
V.
CONCLUSIONS
When we consider that three-quarters of the dry weight of most cells is protein which in turn normally alters significantly the structure of at least 0.5 to 1.0 g water/g protein, it becomes clear that factors affecting protein-water interactions can have an important effect on cell volume. Although the extreme predictions of G . N. Ling (e.g., Ling et al., 1973) on the structured nature of cell water are clearly incorrect, it is equally incorrect to regard the cytoplasm as a uniform solution in which solvent molecules can be assigned the properties of ordinary bulk water or of dilute salt solutions. The existence of numerous, partially incompatible water domains each with its own characteristic concentration of various solutes now seems a more accurate image of the cell interior. In regard to the roles of stabilizing solutes that maintain a minimum protein hydration, it is not clear that all are excluded from the hydration domain of proteins. Clark er al. (1981), for example, found TMAO was preferentially located in skinned muscle fibers, although these findings bear repetition. Furthermore, exclusion of a solute does not guarantee that it will have a stabilizing effect on proteins (e.g., valine, CaCl,, MgCl,). Third, although many excluded, stabilizing molecules raise the surface tension of water, this is not true in the case of two of the most effective stabilizing molecules, betaine and glycerol. Likewise, although a number of stabilizing organic solutes have been examined for their ability to reduce the self-diffusion coefficient of water, it is by no means certain that this will prove the case for all of them. Nor is it self-evident that this parameter will be linearly related to the stabilizing capabilities of various solutes. One can easily imagine different forms of structured water, all with similar values of 9/9l0but with dissimilar compatibilities with the hydration sphere of proteins. Last, there exists the possibility for direct interactions between perturbing and stabilizing solutes, as between salts and TMAO, that has barely been explored. At this stage it is not at all clear whether we are looking at a single, nonspecific thermodynamic pattern of varying intensity that can explain the effects of all compatible solutes, or whether there are several patterns, all with functionally similar consequences. What is required is a methodical examination of families of related solutes, looking for correlations arrived at from equilibrium dialysis
270
MARY E. CLARK
studies, such as those from Timasheff‘s laboratory; the physical properties of solutions (surface tension, dielectric constant, osmotic pressure, and solvent and ); from interactions with perturbing solutes, esmotions, especially 9/9,, pecially ions. Cell volume regulation can no longer be regarded simply as a matter of accumulating the requisite number of solute particles to return the cell to some “preferred” volume. Rather, what counts is the nature of the solutes themselves. They are critical for preserving the proper solvation necessary for macromolecular structure and function, a point that, as Stillinger and Weber (1984) point out, “is usually disregarded, or at best incorporated in some simple averaged way” by those investigating biopolymer conformational problems. Cell volume regulation is thus a consequence of maintaining the milieu inrrucelluluire in a viable state-one that is compatible with macromolecular function. It is a matter of “quality,” not mere “quantity. ” REFERENCES April, E. W., Brandt, P. W., and Elliott, G. F. (1972). The myofilament lattice. Studies on isolated fibers 11. The effects of osmotic strength, ionic concentration, and pH upon the unit cell volume. 1. Cell Biol. 53, 53-65. Arakawa, T., and Timasheff, S . N. (1982a). Stabilization of protein structure by sugars. Biochemistry 21, 6536-6544. Arakawa, T., and Timasheff, S. N. (1982b). Preferential interactions of proteins with salts in concentrated solutions. Biochemistry 21, 6545-6552. Arakawa, T., and Timasheff, S . N. (1983). Preferential interactions of proteins with solvent components in aqueous amino acid solutions. Arch. Biochem. Biophys. 224, 169-177. Arakawa, T., and Timasheff, S. N. (1984a). Mechanism of protein salting in and salting out by divalent cation salts: Balance between hydration and salt binding. Biochemistry 23,5912-5923. Arakawa, T., and Timasheff, S. N. (1984b). Protein stabilization and destabilization by guanidinium salts. Biochemistry 23, 5924-5929. Arakawa, T., and Timasheff, S. N. (1985). The stabilization of proteins by osmolytes. Biophys. J . 47, 41 1-414. Brown, A. D., and Borowitzka, L. J. (1979). Halotolerance of Dunuliellu. In “Biochemistry and Physiology of Protozoa” (M. Levandowsky and S . H. Hutner, eds.). 2nd Ed., Vol. 1, pp. 139190. Academic Press, New York. Bull, H. B., and Breese, K. (1968). Protein hydration. I. Binding sites. Arch. Biochem. Biophys. 128, 488-496. Burnell, E. E., Clark, M. E., Chapman, N. R., and Hinke, J. A. M. (1981). Water in barnacle muscle. 111. NMR studies of fresh fibers and membrane-damaged fibers equilibrated with selected solutes. Biophys. J. 33, 1-26. Clark, M. E., and Hinke, J. A. M. (1981). Studies on water barnacle muscle fibres. I. The dry weight components of fresh fibres. J. Exp. Biol. 90, 33-41. Clark, M. E., and Zounes, M. (1977). The effects of selected cell osmolytes on the activity of lactate dehydrogenase from the euryhaline polychaete, Nereis succineu. B i d . Bull. 153, 468-484. Clark, M., Hinke, J. A. M.,and Todd, M. E. (1981). Studies on water in barnacle muscle fibres. 11. Role of ions and organic solutes in swelling of chemically-skinned fibres. J. Exp. Biol. 90,4363.
NON-DONNAN EFFECTS OF ORGANIC OSMOLYTES
271
Clark, M. E.. Bumell, E. E., Chapman, N. R., and Hinke, J. A. M. (1982). Water in barnacle muscle. IV. Factors contributing to reduced self-diffusion. Biophys. J. 39, 289-299. Elliott, G . F. (1968). Force-balances and stability in hexagonally-packed polyelectrolyte systems. J. Theor. Biol. 21, 71-87. Elliott, G. F., and Rome, E. M. (1969). Liquid-crystalline aspects of muscle fibers. Mol. Cryst. Liq. Cryst. 8, 215-218. Gekko, K., and Morikawa, T. (1981a). Preferential hydration of bovine serum albumin in polyhydric alcohol-water mixtures. J . Biochem. 90, 39-50. Gekko, K., and Morikawa, T. (1981b). Thermodynamics of polyol-induced thermal stabilization of chymotrypsinogen. J. Biochem. 90, 51-60. Gekko, K., and Timasheff, S. N. (1981a). Mechanism of protein stabilization by glycerol: Preferential hydration in glycerol-water mixtures. Biochemistry 20, 4667-4676. Gekko, K., and Timasheff, S. N. (1981b). Thermodynamic and kinetic examination of protein stabilization by glycerol. Biochemistry 20, 4677-4686. Goldammer. E. V . , and Hertz, H. G . (1970). Molecular motion and structure of aqueous mixtures with nonelectrolytes as studied by nuclear magnetic relaxation methods. J. Phys. Chem. 74, 3734-3755. Hand, S. C., and Somero, G . N. (1982). Urea and methylamine effects on rabbit muscle phosphofructokinase. J . Biol. Chem. 257, 734-741. Heilbrunn, L. V. (1943). “An Outline of General Physiology,” 2nd Ed. Saunders, Philadelphia. Lee, J. C., and Timasheff, S . N. (1981). The stabilization of proteins by sucrose. J . Biol. Chem. 256, 7193-7201. Ling, G. N., Miller, C., and Ochsenfeld, M. M. (1973). The physical state of solutes and water in living cells according to the association-induction hypothesis. Ann. N . Y . Acud. Sci. 204, 6-47. McCall, D. W., and Douglass, D. C. (1965). The effect of ions on the self-diffusion of water. I. Concentration dependence. J. Phys. Chem. 69, 2001-201 1. MacKnight, A. D. C., and Leaf, A. (1977). Regulation of cellular volume. Physiol. Rev. 57, 510573. Nobel, P. S. (1970). “Plant Cell Physiology.” Freeman, San Francisco. Oosawa, F. (1971). “Polyelectrolytes.” Dekker, New York. Shaw, J. (1958). Osmoregulation in the muscle fibres of Curcinus maenas. J . Exp. Biol. 35, 920929. Stillinger, F. H., and Weber, T. A. (1984). Packing structures and transitions in liquids and solids. Science 225, 983-989. von Hippel, P. H., and Schleich. T. (1969). The effects of neutral salts on the structure and conformational stability of macromolecules in solution. In “Structure and Stability of Biological Macromolecules” (S. N. Timasheff and G. D. Fasman, eds.). pp. 417-574. Dekker, New York. Yancey. P. H., and Somero, G. N. (1979). Counteraction of urea destabilization of protein structure by methylamine osmoregulatory compounds of elasmobranch fishes. Biochem. J . 183, 3 17323. Yancey, P. H., and Somero, G. N. (1980). Methylamine osmoregulatory solutes of elasmobranch fishes counteract urea inhibition of enzymes. J . Exp. Zool. 212, 205-213. Yancey, P. H., Clark, M. E., Hand, S. C., Bowlus, R. D., and Somero, G. N. (1982). Living with water stress: Evolution of osmolyte systems. Science 217, 1214- 1222. Zeidler, M. D. (1973). NMR spectroscopic studies. I n “Water: A Comprehensive Treatise. Vol. 2, Water in Crystalline Hydrates; Aqueous Solutions of Simple Nonelectrolytes” (F. Franks, ed.), pp. 529-584. Plenum, New York.
This Page Intentionally Left Blank
Index
A
Absolute volume in fixed specimens, 4 Acetazolamide, 110, 114-115 Actin, see also Cytoskeleton; F-actin cell membrane and, 77-78 contractile mechanism and, 80-83 nonmuscle, 82 cytochalasins and, 76 in lower vertebrates, 198 shape and volume, 79 Activation mechanisms in cultured cells. 167I69 Active transport in animal cells, 1280, 129I30 Adenylate cyclase, 83 in MDCK cells, 87 ADH, microtubules and, 85 f3-Adrenergic receptors, 83 P-Alanine in lower vertebrates, 186 skate and, 189-190 Amiloride MDCK cells and, 94 RVI and, 139-141 Amino acids in euryhaline invertebrates. 221 -231 in lower vertebrates, 183, 188-189; see also Solute composition of cells in lower vertebrates RVD and, 164-166 TMOA and, 184 Amphibian oocytes, I I ; see also Lower vertebrates, cell volume regulation in Amphiuma, 108 activation mechanisms in, 167
VRD in, 116, 163-164 VRI in, 114, 139-141 Anions, medium, 32-37 Anisosmotic conditions in cultured cells, see Cultured cells, anisosmotic conditions and epithelia and, see Epithelia, anisosmotic conditions and euryhaline invertebrates and, see Euryhaline invertebrates, cell volume regulation in lower vertebrates and, see Lower vertebrates, cell volume regulation in Anisotonic media, human erythrocytes and, 81 Ankyrin, 77, 80-81; see also Cytoskeleton in erythrocyte, 83 microtubules, 84 Aquatic invertebrates, see Euryhaline invertebrates, cell volume regulation in Anemia cysts, I I Arterial wall tissue, ion binding and, 13 Association-induction hypothesis, 12- I3 ATP, 68-69 ATPase inhibitor, 68-69 Avian red cells, RVI in, 132-134 Avian salt gland, 67-68 Axolemma, axoplasmic microtubules and. 84
0
Balanus nubilis, 259-26 1 BALBlc 3T3 cells, RVI in, 139 Barnacle muscle fibers, cell volume changes in, 256-261 Basement membrane, 18 Basolateral membrane area, 107
273
274
INDEX
Blood osmoregulation in euryhaline invertebrates, 205-286 Bufo murinus, 182; see also Lower vertebrates, cell volume regulation in Bufo viridis, 182, 1830; see also Lower vertebrates, cell volume regulation in Bulk water and nuclear water, 11 Bumetanide, 108-1 10, 150, 170 RVI and, 132-139
C
Calianussa crassipes, 208; see also Euryhaline invertebrates, cell volume regulation in Callinectes sapidus, 220, 223-225, 227-228; see also Euryhaline invertebrates, cell volume regulation in Calmodulin, RVD and, 149-150, 154-155, 157 CAMP, 78, 94; see also db CAMP in MDCK cells chloride secretion and, 94 in Ehrlich ascites tumor line, 169 in MDCK cell line, 85, 87-89, 168 S49 lymphoma cells and, 83 Canaliculi, exocytosis of materials into, 61-63 Capping and patching, 77 Carbonyl cyanide m-chlorophenylhydrazone (CCP), 68 Carcinus maenus, 208, 2 0 9 0 , 2 1 0 0 cytochalasin B and, 213 Cardiac glycosides in isosmotic volume maintenance, 22-23 CCP, 68 Cell organization cell ions and, 12-13 cell water and, 1 I cytoplasm and, 14-16 Cell shape cell volume and, 78-80 cytoskeleton and, see Cytoskeleton, isosmotic volume maintenance and Cell solute composition in lower vertebrates, 183- 187 Cellular binding, 13 Cellular damage, potassium and, 6 Chloride, see also Ions cellular swelling and, 22 in lower vertebrates, 183; see also Solute
composition of cells in lower vertebrates transepithelial transport and, 109- 110, I I I V ) , 113-116, 118-119 Chondroitinase ABC, 13 Chromatin organization, effects of ions on, 234-235 Colchicine, 6O(t), 61 cell shape and, 78 cytoskeleton and, 76-77 MDCK cells and, 89-90, 93(t) microtubules and, 84-85 Colloid osmotic lysis, 125- 126 Colloid osmotic pressure, ouabain and, 18 Compartmentalization in isosmotic volume maintenance cellular binding and, 13 vesicular structures and, 46-51 Composition of cells in lower vertebrates during osmotic stress, 187-189 Contractile mechanism in nonmuscle cells, 8083 Contractile vacuoles, 47-48 Cotransport, see also Potassium-chloride cotransport of anions and cations, 128-130 NEM-stimulated, 160, 162- 163 RVI and, 132-139 activation, 134-137 sites, number of, 138 Coulter counting, 4 Counterions, see also Ions binding of, 262-263 Donnan theory and, 253 Crab-eating frog, cell volume regulation in, 182; see also Lower vertebrates, cell volume regulation in Crassostrea virginica, 229; see also Euryhaline invertebrates, cell volume regulation in Crustaceans, see Euryhaline invertebrates, cell volume regulation in Cryptic pump, 52 Cultured cells, anisosmotic conditions and activation mechanisms in, 167-169 amiloride-sensitive Na + / H + exchange in, 139-144 amino acids and taurine in, 164-166 conductance pathways in, 146-157 furosemide and bumetanide and, 132- 139 general considerations for, 125- 127
275
INDEX identification of membrane constituents in, 167- I7 I mechanisms of ion transport in, 128-130 perspectives on, 166-167 potassium-chloride cotransport in, 157- 163 potassium-hydrogen exchange in, 163- 164 RVD in, 144-146 RVI in, 131, 1320) Cytochalasin B, see also Cytochalasins euryhaline invertebrates and, 212-213 Cytochalasins, 6 0 ( r ) , 61; see also Cytochalasin B cytoskeleton and, 76, 83 MDCK cells and, 91, 93 neural tube formation in, 82 Cytoplasm as gel versus solution, 14 nature of, 14-16 Cytoplasmic vesicles, isosmotic volume maintenance and general considerations for, 69-7 I ion transport and, 68-69 in vertebrates, 51-52 liver, 52-63, 69(r) other tissues, 63-68 water movement and compartmentalization and contractile vacuoles, 47-48 tonoplast of plant cells, 48 vertebrate tissue, 48-5 I Cytoskeleton anisosmotic conditions and, 197-201 isosmotic volume maintenance and cell shape. 76-77 cell volume, 78-80 general considerations, 75-76, 96 MDCK cell line, 85-96 mechanism of cell volume control, see Mechanisms of cell volume control, cytoskeleton and membrane interactions with, 77-78 Cytosolic proteins, see Cytoskeleton, isosmotic volume maintenance and
D
db CAMP in MDCK cells, 87-96 DCCD, 68 Dibutyryl CAMP, 78
N.N’-Dicyclohexylcarbodiimide(DCCD), 68 DIDS, 169 Dogfish, 185; see also Lower vertebrates, cell volume regulation in Dog red cells, 116 RVI in, 141-142 Donnan equilibrium, 252-254; see also Gibbs-Donnan rule; Non-Donnan effects of organic osmolytes in cell volume changes in animal cells, 125-126 Double Donnan model of volume regulation, 22 Dry-weight components of Balanus nubilis muscle fibers, 259-261 Duck red cells RVD and, 157-158 volume regulatory decrease in, 115- I16
E Echinoderm sperm, cell shape and volume and, 79-80 Ehrlich ascites tumor cells activation mechanisms in. 167-169 NEM-stimulated cotransport in, 162-163 VRD in, 115, 147-156 amino acids and taurine, 164-166 VRI in, 132(r), 134-138 Elasmobranchs, 1 8 3 0 , 185- 186; see also Lower vertebrates, cell volume regulation in Electrodiffusion, I28 Electroneutral exchange of ions, 128- 129 in animal cells, 125 Electronic sizing techniques, 4 Environmental considerations in cell volume regulation in lower vertebrates, 182- I83 Enzyme activity. effects of ions on, 232-233 Epithelia anisosmotic conditions and general considerations, 105- 106 measurement methods, 11 1-1 12 mechanisms, 115-1 17 Necturus. 108, 110-115. 117-120 osmotic perturbations, 110, 1 I IV, physiological significance, 120 steady-state cell water permeability, 106-107
276
INDEX
polarity of solute movements and cell volume changes, 108 solute flow and steady-state cell volume, 107-108 transepithelial NaCl transport, 108-1 10 volume regulatory increase, 112-1 15 isosmotic volume maintenance in, 81 Eriocheir sinensis, 205, 208-21 1, 214, 229230; see also Euryhaline invertebrates, cell volume regulation in Erythrocyte, membrane-cytoskeleton interaction in, 77, 80-81; see also Human erythrocyte ghosts N-Ethylmaleimide (NEM), REV and, 160, 162-163
Euryhaline invertebrates, cell volume regulation in general considerations for, 205-206, 235236
macromolecules and changes in ion content in, 231-235 volume readjustment responses in, 207-21 I mechanisms, 211-231 Evolutionary considerations in cell volume regulation in lower vertebrates, 181-182 Exocytosis, 46-47, 5 6 0 , 57, 61, 85 Experimental approaches to isosmotic volume regulation, 4-10 Extracellular fluid regulation in lower vertebrates, 189-190 Extracellular osmolality in vertebrates, 3 Extracellular water, experimental approaches to isosmotic volume maintenance and. 7
F
F-actin, in MDCK cell line, 85-96; see also Actin Fibroblasts, cell shape in, 76 Ficopafamus. 212; see also Euryhaline invertebrates, cell volume regulation in Fixation and processing of tissue, alterations during, 4 Flounder, 1830, 184; see also Lower vertebrates, cell volume regulation in Fodrin, 78, 83-84; see also Cytoskeleton Forskolin, F-actin and, 89
Frog skin epithelium, volume regulatory decrease in, 115 Furosemide RVI and, 132-139 second sodium pump and, 52, 59, 61, 70
0
Gibbs-Donnan rule, see also Donnan equilibrium in cultured cells, 125-126 in lower vertebrates, 191-192, 195-196 Glial cells, RVI in, 139 Glucose, solute entry and, 107 Glutaraldehyde, 141 Glycera dibranchiata, 220; see also Euryhaline invertebrates, cell volume regulation in Glycerinated cells, nonmuscle contraction and, 82-83
Golgi system cytoplasmic vesicles and, 55-56, 68, 69(r), 70-71
microtubules and, 85 Goose erythrocytes, volume and shape in, 79 Gramicidin, RVD and, 148v). 149-151, 153154, 156
Guinea pig, 93-94
H
Hagfish, 184; see also Lower vertebrates, cell volume regulation in Hepatocytes, 63, 64v) Hepatoma 3924A, 62v), 63, 65 Homurus gammarus, see Lobster Human erythrocyte ghosts, 78-79 Hydrochlorthiazide, 108 Hydrostatic pressure, 15-16, 125 ouabain and, 18 Hyperosmotic media, volume response to in euryhaline invertebrates, 209-21 1 Hypoosmotic media, volume response to in euryhaline invertebrates, 207-209 Hypotonicity, decrease in cell volume and, 1 I5
INDEX
277 I
Identification of membrane constituents in cultured cells, 169-171 Impermeant cellular solutes, 3. 16-20 in mammalian tissue, 20-21 Impermeant univalent anion gluconate, 18- 19 Inorganic ions, see also Ions, in euryhaline invertebrates, 215-220 Intestinal epithelium, membrane-cytoskeleton interaction in, 81 lntracellular osmolytes in lower vertebrates, 191-197 Jntracellular reference-phase technique, cell water and, I I Inulin, 4 In v i m studies of isosmotic volume maintenance, 5-7 I n vivo studies of isosmotic volume maintenance, 5-6 Ion binding, 13; see also Ions; Ion transport Ionophore A23187, 79 lonophore, RVD and, 150-151, 153. 155-156 Ions, 12-13; see also Ion binding; Ion transport; Univalent ions cellular swelling and, 21-22 in euryhaline invertebrates, 215-220, 231237 permeabilities to, 126 in RVD, 118. 144-146 Ion transport, see also Cotransport; Ion binding; Ions isosmotic volume regulation and cytoskeleton, see Cytoskeleton, isosmotic volume maintenance and intracellular membranes, 68-69 mechanisms of, 128- 130 lsobutylmethylzanthine, 89, 9 0 0 Isosmotic conditions, volume maintenance in cell organization and, see Cell organization cellular swelling and limiting factors, 31-32 sodium, 2 1-30 cytoplasmic vesicles and, see Cytoplasmic vesicles, isosmotic volume maintenance and cytoskeleton and, see Cytoskeleton experimental approaches to study of, 4- 10 general considerations for, 3-4, 37-38
impermeant cellular soJutes and description, 20-21 osmotic forces, 16-20 medium anions and, 33-37
J J774.2 macrophage, shape and volume changes in, 78 K Kinetic studies of RVI, 133-134
L Latice unit cell in muscle fibers, 254-255 Leukotriene synthesis in cultured cells, 168 Leupeptin, fodrin and, 84 Light microscopy for epithelia, 110-1 I 1 Limulus polyphemus. 219 Liver, ouabain-resistant volume regulation in, 52-63, 69(1) Lobster, 208, 2 0 9 0 ; see also Euryhaline invertebrates, cell volume regulation in Lower vertebrates, cell volume regulation in cell solute composition and, 183-187 changes in cell composition during osmotic stress and, 187-189 cytoskeleton and, 197-201 environmental considerations and, 182- I83 evolutionary considerations and, 181- 182 extracellular fluid regulation and, 189- 190 intracellular osmolytes and, 191-197 Lung, cytoplasmic vesicles and, 67 Lymphocytes RVD in, 115, 156-157 RVI in, 142-144 Lysis in animal cells, 125- I26
M Macromolecules, 125- 126 Macrophages, 78, 81-82
INDEX Madin-Darby canine kidney (MDCK) cell line, cytoskeleton and cell volume in, 76, 85-96 RVI in, 138-139 Mammalian tissue, impermeant solutes in, 2021 Marker, extracellular, 4-5 MDCK, see Madin-Darby canine kidney cell line Measurement methods for cell volume changes in epithelia, 1 11-1 12 Mechanicochemical hypothesis of nonmuscle contraction, 80 Mechanisms of cell volume control, cytoskeleton and membrane proteins in, 83-85 nonmuscle contraction in, 80-83 Membrane-cytoskeletal interactions, 77-78; see also Cytoskeleton Membrane proteins, cytoskeletal control of activity and number of, 83-85 Mercierella enigemarica, 205;see also Euryhaline invertebrates, cell volume regulation in Microscopy for cellular volume, 4 epithelia and, 1 10-1 11 Microtubules, 84-85 in MDCK cell line, 85, 87 Modiolus demissus, 225;see also Euryhaline invertebrates, cell volume regulation in Monensin, 68 Morphometric methods for measuring cell volume changes, Ill-I12 Motile macrophages, volume changes in, 78 Mouse diaphragm, 17-18 osmolality ratios and, 19 Muscle fibers, cell volume changes in, 254262 Myofilament proteins, 256-2570. 260-261, 268 Myoinositol in lower vertebrates, 201 Myometrium, 66-67 Myosin, 82 Myxine glurinosu, see Hagfish
N NaCl transport Necrurus, 108-109 transepithelial, 108- 110
Necrurus gallbladder, 107- 108 anisosmotic conditions and activation mechanisms, 167- 168 epithelium ion movements, 110 NaCl transport, 108-109 ouabain, 110 VRD, 117-120 VRI, 111-115 isosmotic volume maintenance and CAMP, 93-94 cell ions, 12-13 vesicle insertion mechanism, 84-85 NEM-stimulated KCI cotransport, 160, 162163 Neural tube formation, 82 NMR spectroscopy, I 1 Nomarski optics, 4 Non-Donnan effects of organic osmolytes in cell volume changes Donnan theory as model and, 252-259 general considerations for, 25 1-252, 269270 muscle fibers and, 254-259 stabilizing solutes and, 259-262 mechanisms of action, 262-269 Nonerythmid cells, cell membrane linkage in, 77 Noninulin space ion concentrations, 6 1 0 Noninulin space water content, potassium and, 4-5 Norepinephrine, RVI and, 133 Nuclear magnetic resonance (NMR) spectroscopy, 1 1 Nuclear water and bulk water. 11
0 Oocytes, amphibian, 11 Optical, light microscopic methods for volume studies of epithelia, 1 I I Organelles, 1 1 , 14 Organic osmotic effectors, 220-23 1 Osmolality in euryhaline invertebrates, 205206 Osmometer, notion of cell as, 251-252 Osmotic effectors in euryhaline invertebrates, 215-231 Osmotic equilibrium in animal cells, 125
INDEX
279
Osmotic perturbations, epithelial cell volume and, 110, 1 1 I0 Ouabain anisosmotic conditions and, 110 volume regulatory decrease, I19 isosmotic volume maintenance and, 8-9, I8 renal cortical tissue, 25-26 Oyster, 229; see also Euryhaline invertebrates, cell volume regulation in
Protein kinase C , cell shrinking and, 169 Proteins, cytosolic, see Cytoskeleton Proximal tubule cells, 106 Pump-leak concept, 21, 126 in isosmotic volume maintenance, 22-23 experimental studies, 10 lower vertebrates and, 191
Q
P Paralichthys lethosrigma, 188 Parallel exchangers, epithelia and, 109-1 10 Patching and capping, 77 Peptide hormones, cell volume and, 78 Permeant solutes, 3 PH shape and volume change and, 78-79 sodium-chloride transport mechanisms and, 120 sodium-hydrogen countertransport system and, 22 Phase contrast, 4 Physiochemical prospectives, see Non-Donnan effects of organic osmolytes in cell volume changes Physiological significance of volume regulation, 120 Pimozide, 154, 1 5 5 0 , 157 Plant cell, tonoplast of, 46, 48 Pleuronectesflesus, 1 8 3 0 , 184 Polarity of solute movements and cell volume changes, 108 Polyanionic glycosaminoglycans, 13 Potassium, see also Ions; Solute composition of cells in lower vertebrates in lower vertebrates, 183 transepithelial transport and, 108-1 10, 113I20 Potassium-chloride cotransport in Ehrlich ascites tumor cells, 159-160 NEM stimulated, 160. 162-163 in red blood cells, 157-159 VRD and, 116, 119-120 Prostaglandin synthesis in cultured cells, 168 Protein domain, solute exclusion from, 263264 Protein hydration, water structure and, 264269
QNS, I I Quasielastic neutron scattering (QNS), I 1 Quinine, RVD and, 1 4 8 0 , 149-150, ISlY), I53
R
Rabbit renal cortical tissue, 8-9 tubules, 8 Raja erinacea, see Skate Rana cancrivora, 182; see also Lower vertebrates, cell volume regulation in Rat renal cortical slices, 9; see also Renal cortical slices skeletal muscle, cell ions in, 12-13 Red blood cells, NEM-stimulated cotransport in, 160-161 Regulatory volume decrease (RVD), see Cultured cells, anisosmotic conditions and, RVD in Regulatory volume increase (RVI). in vertebrate cells, see Cultured cells, anisosmotic conditions and, RVI in Renal cortical slices anisosmotic conditions and, 116-1 17 isosmotic volume maintenance and, 7- 10 cytoplasmic vesicles, 65-66 impermeant solutes, 17(r), 19 mammalian, 13 ouabain, 25-26 rabbit. 17-19 rat, 24(n, 25 Reptiles, 182, 183; see also Lower vertebrates, cell volume regulation in RVD, see Regulatory volume decrease
INDEX RVI, see Cultured cells, anisosmotic conditions and, RVI in
S
S49 lymphoma cells, 83 Salt solutions, organic solutes and, 261-262, 268-269 Second sodium pump, 52; see also Sodium Pump Secretion in vesicles in vertebrate tissue, 4851 Self-diffusion and water structure, 266-268 Serosal bath, epithelia and, 107 Sheep red cells, volume regulatory decrease in, 116 Shuttle hypothesis for vessel insertion, 84-85 Skate, 1830, 184; see also Elasmobranchs; Lower vertebrates, cell volume regulation in Skeletal muscle, isosmotic volume regulation in impermeant solutes, 1 7 0 , 19 mammalian, 22 mouse, 6-7 Smooth muscle cells, CAMP and, 94 Sodium, see also Ions; Pump-leak concept; Second sodium pump; Sodium-chloride; Sodium pump epitheliaand, 108-110, 111(n, 113-114. 116-120 in lower vertebrates, 183; see also Solute composition of cells in lower vertebrates in mammalian skeletal muscle, 22 squid giant axon and, 84 Sodium-chloride, 108-1 10; see also NaCl transport epithelia and, I10 pH and, 120 VRD and, 116-1 17 Sodium-hydrogen countertransport system in mammalian skeletal muscle, 22 Sodium pump, 18- 19; see also Second sodium pump; Sodium; Sodium-chloride furosemide and, 52, 59. 61, 70 ouabain and, 23-25 Solute composition of cells in lower vertebrates, 183- I87
Solute flow and steady-state volume, 107-108 Solutes, stabilizing and perturbing, 259-262 Spectrin, 77; see also Cytoskeleton Squalus acanrhias, see Dogfish Squid giant axon, sodium current in, 84 Stabilizing solutes, 259-262 Steady state epithelial cell volume during polarity, 108 solute flow, 107-108 water permeability, 106-107 isosmotic volume maintenance and impermeant cellular solutes, 16 water activity, 15 Stress fibers in MDCK cells, 8 6 0 , 87-88 nonmuscle contraction and, 82-83
T
Taurine in lower vertebrates, 186, 190 RVD and, 164-166 T ~ o l 76-77 , Teleosts, 186; see also Lower vertebrates, cell volume regulation in Terrestrial amphibia, 183; see also Lower vertebrates, cell volume regulation in Tissue slices, isosmotic volume maintenance in, 7-10; see also speciJic tissue slices T-lymphoma cell line, 77 TMAO, non-Donnan effect of organic osmolytes and, 256-258. 2580, 260(t), 261-264, 267-269; see Trimethylamine oxide Toads, 182-184, 187-188; see also Lower vertebrates, cell volume regulation in urinary bladder in ADH, 85 shuttle hypothesis, 84 Transcellular water movement, 50-5 1 Transepithelial NaCl transport, 108-1 10 Tributyl tin, 68 Trimethylamine oxide (TMAO) in lower vertebrates, 183-189, 196, 201 non-Donnan effects of organic osmolytes and, 256-258. 2580, 260(1), 261264, 267-269 Tris, 69(r)
INDEX Triton, 79 Tubules, isosmotic volume maintenance and, 7-8, 10, 18; see also Microtubules Tubulin, 84
Vesicle insertion mechanism, 95-96 Vesicle insertion/retrieval process, cytoskeleton and, 84-85 Vesicle-mediated secretion, microtubules and, 85 Volume-activated CI - and K conductance pathways, 146-157 Volume readjustment responses, see ulso Volume regulatory decrease; Volume regulatory increase in euryhaline, invertebrates, 206 mechanisms of components. 21 1-214 intracellular osmotic effectors, 2 15-23 I VRD in, 206-209 VRI in, 209-21 1 Volume regulatory decrease (VRD) in epithelia mechanisms, 115- I17 Necturus gallbladder, 117- 120 in euryhaline invertebrates, 206-209 Volume regulatory increase (VRI) in epithelia, 112-1 15 in euryhaline invertebrates, 209-21 I +
U
Univalent anion gluconate, 18- 19 Univalent ions, 12-13 Urea, 261 in lower vertebrates, 183-184; see also Solute composition of cells in lower vertebrates protein domain and, 263
V
Vacuolar degeneration, 46 Valinomycin, 68, 69(r) RVD and, 150, 152, 154 RVI and, 133 Van’t Hoff equation, 17- 18 Vertebrates, isosmotic volume regulation in avian salt gland and, 67-68 general considerations for, 5 1-52 hepatoma 3924A and, 63, 65 liver and, 52-63, 69(r) lung and, 67 myometrium and, 66-67 renal cortex and, 65-66 vesicles and, 48-51 Vesicle-apical cell membrane fusion events, ADH and, 85
W
Water activity, isosmotic volume maintenance and, 15 Water content of cell, 4 Water permeability of epithelial cell membrane, 106-107 Water physical state in euryhaline invertebrates, 211-212 Water structure, protein hydration and, 264269
This Page Intentionally Left Blank
Contents of Recent Volumes Volume 20
Na .K + -ATPase MITHAEL. KASHCAHIAN, DANIEI. 111 BIEMESlXRFliR. A N D BLISSFORBUSH Monoclonal Antibodies as Probes of Epithelial Cell Polarity Gl.OHGk K. OJAKlAN A N D DORISA. H~KZLINGW lmmunolabeling of Frozen Thin Sections and Its Application to the Study of the Biogenesis of Epithelial Cell Plasma Membranes IVAN EMANLJILO IVANOV, V HMDE PI-L-SKtN,DAVID D. SAHA'TINI, A N I ) MICHAELJ . RINDLER Development of Antibodies to Apical Membrane Constituents Associated with the Action of Vasopressin J A M ~ B. S WADE, V I C T O K I A GUCKIAN, A N D INGEBORC KOEPPEN Molecular Modification of Renal Brush Border Maltasr with Age: Monoclonal AntibodySpecific Forms of the Enzyme BERTRAMS A C K T O R A N D UZl RElSS +
Molecular Approaches to Eplthelial Transport PART I . FREQUENCY DOMAIN ANALYSIS OF ION TRANSPORT Fluctuation Analysis of Apical Sodium Transport T . HOSHIKO Impedance Analysis of Necturus Gallbladder Epithelium Using Extra- and Intracellular Microelectrodes J. 1. LIM. G . KOTTRA,L. KAMPMANN, A N D E. FKOMTER Membrane Area Changes Associated with Proton Secretion in Turtle Urinary Bladder Studied Using Impedance Analysis Techniques CHHlS Cl.ALiStN AN13 TROYE. DIXON Mechanisms of Ion Transport by the Mammalian Colon Revealed by Frequency Domain Analysis Techniques N. K . W1i.1.s Analysis of Ion Transport Using Frequency Domain Measurements SIMON A. LEWIS AND WILLIAM P. Ai.u<s Use of Potassium Depolarization to Study Apical Transport Properties in Epithelia LAWRENCE G . PALMER PART 11. USE OF ANTIBODIES TO EPITHELIAL MEMBRANE PROTEINS Biosynthesis of Na ,K + -ATPase in Amphibian Epithelial Cells B. C. ROSSIEK Use of Antibodies in the Study of N a + .K t. ATPase Biosynthesis and Structure ALICIAM. MCDONO~IGH Encounters with Monoclonal Antibodies to +
283
PART I l l . BIOCHEMICAL CHARACTERIZATION OF TRANSPORT PROTEINS Sodium-u-Glucose Corransport System: Biochemical Analysis of Active Sites R. KINNE,M. E. M . DA CRUZ,AND J . T. LIN Probing Molecular Characteristics of Ion Transport Proteins DARRELLD. FANESTIL,RALPHJ. KESSLER.AND CHUNSIK PARK Aldosterone-Induced Proteins in Renal Epithelia MALCOLMCox A N D MICHAEL. GEHEB Development of an isolation Procedure for Brush Border Membrane of an Electrically
284
CONTENTS OF RECENT VOLUMES
Tight Epithelium: Rabbit Distal Colon MICHAEL C. GUSTINAND DAVIDB. P. GOODMAN Index
Volume 21 Ion Channels: Molecular and Physiologlcal Aspects Ionic Selectivity of Channels at the End Plate PETERH. BARRYA N D PETERW. GAGE Gating of Channels in Nerve and Muscle: A Stochastic Approach RICHARDHORN The Potassium Channel of Sarcoplasmic Reticulum CHRISTOPHER MILLER,JOAN E. BELL AND ANAMARIAGARCIA Measuring the Properties of Single Channels in Cell Membranes H.-A. KOLB Kinetics of Movement in Narrow Channels DAVIDG . LEVIT Structure and Selectivity of Porin Channels R. BENZ Channels in the Junctions between Cells WERNERR. LOBWENSTEIN Channels across Epithelial Cell Layers SIMON A. LEWIS,JOHNw. HANRAHAN, AND W . V A N DRIESSCHE Water Movement through Membrane Channels ALANFINKELSTEIN Channels with Multiple Conformational States: Interrelations with Carriers and Pumps P. LAUGER Ion Movements in Gramicidin Channels S. B. HLADKYA N D D. A. H A Y D ~ N Index
Volume 22 The Squid Axon PART I. STRUCTURE Squid Axon Ultrastructure GLORIA M. VILLEGAS AND R A I M U N W VILLEGAS
The Structure of Axoplasm RAYMOND J. LASEK PART II. REGULATION OF THE AXOPLASMIC ENVIRONMENT Biochemistry and Metabolism of the Squid Giant Axon HAROLDGAINER, PAULE. GALLANT, ROBERTGOULD,A N D HARISHC. PANT Transport of Sugars and Amino Acids P. F. BAKERA N D A. CARRUTHERS Sodium Pump in Squid Axons Luis BEAU& Chloride in the Squid Giant Axon JOHN M. RUSSELL Axonal Calcium and Magnesium Homeostasis P. F. BAKERA N D R. DIPOLO Regulation of Axonal pH WALTERF. BORON Hormone-Sensitive Cyclic Nucleotide Metabolism in Giant Axons of Loligo P. F. BAKERA N D A. CARRUTHERS PART 111. EXCITABILITY Hodgkin-Huxley: Thirty Years After H. MEVES Sequential Models of Sodium Channel Gating CLAYM. ARMSTRONG A N D DONALD R. MAITESON Multi-Ion Nature of Potassium Channels in Squid Axons TEDBECENISICH AND CATHERINE SMITH Noise Analysis and Single-Channel Recordings FRANCOCONTI Membrane Surface Charge DANIELL. GILBERTAND GERALD EHRENSTEIN Optical Signals: Changes in Membrane Structure, Recording of Membrane Potential, and Measurement of Calcium LAWRENCE B. COHEN,DAVID LANDOWNE, LESLIEM. LOEW,AND BRIANM. SAUBERG Effects of Anesthetics on the Squid Giant Axon D. A. HAYDON,J. R. ELLIOIT, AND B. M. HENDRY
285
CONTENTS OF RECENT VOLUMES Pharmacology of Nerve Membrane Sodium Channels TOSHIONARAHASHI PART IV. INTERACTION BETWEEN GIANT AXON AND NEIGHBORING CELLS The Squid Giant Synapse RODOLPOR. LLlNAS Axon-Schwann Cell Relationship JORCE VILLEGAS Index
Volume 23 Genes and Membranes: Transport Proteins and Receptors PART I. RECEKORS AND RECOGNITION PROTEINS Sensory Transduction in Bacteria MELVIN1. SIMON,ALEXANDRA KRIKOS, NORIHIROMUTOH.A N D ALANBOYD Mutational Analysis of the Structure and Function of the Influenza Virus Hemagglutinin MARY-JANE GETHING,CAROLYN DOYLE, MICHAELROTH, AND JOE SAMBROOK
A Study of Mutants of the Lactose Transport System of Escherichiu coli T. HASTINCSWILSON, DONNASETOYOUNG,SYLVIE BEDU. RESHAM. PUTZRATH, A N D BENNOMULLER-HILL The Proton- ATPase of Escherichiu coli A. E. SENIOR The Kdp System: A Bacterial K + Transport ATPase WOLIGANG EPSTEIN Molecular Cloning and Characterization of a Mouse Ouabain Resistance Gene: A Genetic Approach to the Analysis of the Na ,K ATPase ROBERTLEVENSON +
+
Index
Volume 24 Membrane Protein Blosynthesls end Turnover Application of the Signal Hypothesis to the Incorporation of Integral Membrane Proteins TOMA. RAFQPORT AND MARTIN WIEDMANN
Structure and Function of the Signal Peptide G U YD. DUPFAUD, SUSANK. PART 11. CHANNELS LEHNHARDT, PAULE. MARCH,AND MASAYORI INOUYE Ca2 Channels of Paramecium: A The Use of Genetic Techniques to Analyze Multidisciplinary Study Protein Export in Escherichia coli CHINO KUNGAND YOSHIRO SAlMl VYTASA. BANKAITIS, J. PATRICK RYAN, Studies of Shaker Mutations Affecting a K + A N D PHILIPJ. BETHA. RASMUSSEN. Channel in Drosophila BASSFORD,JR. LILYYEH J A N , SANDRA BARBEL,LESLIE Structural and Thermodynamic Aspects of the TIMPE,CHERYL LAFFER,LAWRENCE Transfer of Proteins into and across SALKOPF,PATRICKO'FARRELL,A N D Membranes YUH NUNGJAN GUNNAR VON HEIJNE Sodium Channels in Neural Cells: Molecular Mechanisms and Functional Role of Properties and Analysis of Mutants WILLIAM A. CATTERALL, TOHRLJ GONOI, Glycosylation in Membrane Protein Synthesis SHARONS. KRAC AND M A R I A COSTA Protein Sorting in the Secretory Pathway ENRIQUE RODRIGUEZ-BOULAN, D A V I D E. PART 111. TRANSPORT SYSTEMS MISEK, DORAVEGA DE SALAS, PEDRO J. 1. SALAS, AND ENXI BARD The Histidine Transport System of Salmonellu Transport of Proteins into Mitochondria typhirnurium GRAEMEA. REID GIOVANNA FERRO-LUZZI AMES +
CONTENTS OF RECENT VOLUMES Assembly of the Sarcoplasmic Reticulum during Muscle Development DAVID H. MACLENNAN, ELIZABETH ZUBRZYCKA-GAARN. AND ANNELISE 0. JORGENSEN Receptors as Models for the Mechanisms of Membrane Protein Turnover and Dynamics H. STEVENWILEY The Role of Endocytosis and Lysosomes in Cell Physiology YVES-JACQUES SCHNEIDER. JEAN-NOEL OCTAVE,AND ANDRETROUET Regulation of Glucose Transporter and Hormone Receptor Cycling by Insulin in the Rat Adipose Cell IAN A. SIMPSON AND SAMUEL w. CUSHMAN Index
ADlL E. SHAMOO,INDUs. AMBUDKAR. MARCs. JACOBSON, A N D JEANBIDLACK Role of Calmodulin in the Regulation of Muscle Contraction ANNIEMOLLA,SIDNEY KATZ, AND JACQUESG. DEMAILLE Calcium Release from Sarcoplasmic Reticulum MAKOTOENDO The Regulation of Mitochondria1 Calcium Transport in Heart MARTIN CROMFTON PART Ill. CELLULAR ION REGULATION AND DISEASE Cellular Ion Regulation and Disease: A Hypothesis BENJAMIN F. TRUMPAND IRENE K. BEREZESKY Index
Volume 25 Volume 26 Regulation of Calclum Transport across Muscle Membranes Overall Regulation of Calcium Transport in Muscle ADILE. SHAMOO
Na +-H + Exchange, lntracellular pH, and Cell Function PART I. GENERAL ASPECTS OF INTRACELLULAR pH REGULATION AND Na+ -H EXCHANGE +
PART I . REGULATION OF CALCIUM TRANSPORT IN PLASMA MEMBRANES Sarcolemmal Enzymes Mediating P-Adrenergic Effects on the Heart LARRYR. JONES Properties of Myocardial Calcium Slow Channels and Mechanisms of Action of Calcium Antagonistic Drugs NICK SPERELAKIS, GORDON M. WAHLER, AND GHASSAN BKAlLY The Sarcolemmal Sodium-Calcium Exchange System JOHN P. REEVES PART 11. REGULATION OF CALCIUM TRANSPORT IN SUBCELLULAR ORGANELLES Regulation of Calcium Transport in Cardiac Sarcoplasmic Reticulum
lntracellular pH Regulation by Leech and Other Invertebrate Neurons R. C. THOMASAND W. R. SCHLUE Approaches for Studying Intracellular pH Regulation in Mammalian Renal Cells WALTERF. BORON Aspects of pHi Regulation in Frog Skeletal Muscle ROBERTW. PUTNAM AND ALBERTRoos Molecular Properties and Physiological Roles of the Renal Na+-H+ Exchanger PETERS. ARONSONAND PETER GARA AS HI PART II. Na+-H+ EXCHANGE AND CELL VOLUME REGULATION Volume-Sensitive Alkali Metal-H Transport in Amphiuma Red Blood Cells
PETERM. CALA
CONTENTS OF RECENT VOLUMES Na-Proton Exchange in Dog Red Blood Cells JOHN C. PARKER Activation of the Na +-H Antiport by Changes in Cell Volume and by Phorbol Esters; Possible Role of Protein Kinase S. GRINSTEIN. S. COHEN,J. D. GOETZ, A. ROTHSTEIN, A. MELLORS,A N D E. W. GELPAND +
PART 111. Na+-H+ EXCHANGE AND CONTROL OF CELL GROWTH The Generation of Ionic Signals by Growth Factors W. H. MOOLENAAR. L. H. K . DEFIZE, P. T . V A N DER SAAG, A N D S. W . DE LAAT Control of Mitogenic Activation of Na+ -H Exchange D. CASSEL,P. ROTHENBERC, 8. WHITELEY, D. MANCUSO,P. SCHLESSINGER, L. R ~ u s s E. , 1. CRAGOE, A N D L. GLASER Mechanisms of Growth Factor Stimulation of Na+-H Exchange in Cultured Fibroblasts MITCHELL. VILLEREAL. LESLIEL. MIXMULDOON, LUCIAM. VICENTINI, GORDONA. JAMIESON, JR., A N D NANCY E. OWEN B Lymphocyte Differentiation: Role of Phosphoinositides, C Kinase, and Na+-H + Exchange PHILIPM. ROSOFFA N D LEWISC. CANTLEY Na’--H Exchange and Growth Control in Fibroblasts: A Genetic Approach JACQUESPOUYSSEGUR, ARLETTE FRANCHI. MICHIAKI KOHNO,GILLES L’ALLEMAIN. AND SONIA PARIS
Proximal Tubule: Studies in Microvillus Membrane Vesicles JULIANL. SEIWERA N D RAYMOND C. HARRIS The Role of lntracellular pH in Insulin Action and in Diabetes Mellitus RICHARDD. MOORE The Proton as an Integrating Effector in Metabolic Activation WILLIAM B. BUSA Index
Volume 27 The Role of Membranes In Cell Growth and Dlfferentlatlon
+
+
PART I. DESCRIPTION OF ION TRANSPORT SYSTEMS IN ACTIVATED CELLS Mitogens and Ion Fluxes Luis REUSS,DANCASSEL,PAUL ROTHENBERG, BRIANWHITELFY, DAVID MANCUSO,A N D LUISGLASER Na+ -H and Na -Ca2 Exchange in Activated Cells MITCHELL VILLERLAL Chlonde-Dependent Cation Cotransport and Cellular Differentiation. A Comparative Approach PtTER K LAW +
+
+
+
PART IV. ROLE OF Na -H EXCHANGE IN HORMONAL AND ADAPTIVE RESPONSES +
+
PART I I . TRIGGERS FOR INCREASED TRANSPORT DURING ACTIVATION External Triggers of Ion Transport Systems: Fertilization, Growth, and Hormone Systems JOANBELL,LORERA NIELSEN,SARAH A N D DALEBENOS SARIBAN-SOHRABY, Early Stimulation of Na+-H + Antiport, Na -K Pump Activity, and Ca2 Fluxes in Fibroblast Mitogenesis ENRIQUE ROZENGURT AND STANLEY A. MENDOZA Volume-Sensitive Ion Fluxes in Amphiuma Red Blood Cells: General Principles Governing +
Hormonal Regulation of Renal Na -H Exchange Activity BERTRAM SACKTORA N D JAMES L. KINSELLA Adaptation of Na -H Exchange in the +
+
+
+
+
+
CONTENTS OF RECENT VOLUMES
288 Na-H and K-H Exchange Transport and C1HCO, Exchange Coupling PETERM. CALA PART 111. CONSEQUENCES O F THE ALTERATIONS IN ION TRANSPORT OBSERVED DURING ACTIVATION Intracellular Ionic Changes and Cell Activation: Regulation of DNA. RNA, and Protein Synthesis KATHIGEERING Energy Metabolism of Cellular Activation, Growth, and Transformation LAZAROJ. MANDEL Index
Volume 28 Potasalum Transport: Physiology and Pathophyslology PART I: CELL MECHANISM OF POTASSIUM TRANSPORT Role of Potassium in Epithelial Transport Illustrated by Experiments on Frog Skin Epithelium H. H. USSING Na+ , K , and Rb Movements in a Single Turnover of the Na/K Pump BLISSFORBUSH III Properties of Epithelial Potassium Channels DAVIDC. DAWSON Role of Potassium in Cotransport Systems ROLFKINNEAND ERICHHEINZ Functional Roles of Intracellular Potassium in Red Blood Cells JOSEPHF. HOFFMAN +
+
PART 11: RENAL AND EXTRARENAL CONTROL OF POTASSIUM: PHYSIOLOGY
Overview: Renal Potassium Transport along the Nephron FREDS. WRIGHT
Potassium Recycling REX L. JAMISONAND ROLANDMULLERSUUR
Cell Models of Potassium Transport in the Renal Tubule GERHARD H. GIEBISCH Adrenal Steroid Regulation of Potassium Transport ROGERG. O’NEIL Potassium and Acid-Base Balance RICHARD L. TANNEN Renal Potassium Adaptation: Cellular Mechanisms and Morphology BRUCEA. STANTON Quantitative Analysis of Steady-State Potassium Regulation DAVIDB. YOUNG PART Ill: RENAL AND EXTRARENAL CONTROL O F POTASSIUM: DIURETICS AND PATHOPHYSIOLOGY Regulation of Extrarenal Potassium Homeostasis by Insulin and Catecholamines RALPHA. DEFRONZO Diuretics and Potassium S . WILCOX CHRISTOPHER Pathogenesis and Pathophysiological Role of Hypoaldosteronism in Syndromes of Renal Hyperkalemia A N D ANTHONY MORRISSCHAMBELAN SEBASTIAN Hypokalemia JORDANJ. COHEN Metabolism and Potassium JAMESP. KNOCHEL
PART IV: POTASSIUM TRANSPORT IN MUSCLE AND COLON Effects of Potassium Deficiency on Na,K Homeostasis and Na ,K -ATPase in Muscle TORBEN CLAUSEN AND KELD KJELDSEN Relationship between Cell Potassium and Hydrogen Ion in Muscle SHELD~N ADLER Electrophysiology of Active Potassium Transport across the Mammalian Colon +
+
CONTENTS OF RECENT VOLUMES
N. K. WILLS,C. CLAUSEN,AND W. C. CLAUSS Potassium Adaptation in Mammalian Colon JOHNP. HAYSLEIT, HENRYJ. BINDER, AND MICHAELKASHGARIAN Index
Volume 29 Membrane Structure and Functlon Current Views of Membrane Structure ALANM. KLEINFELD Ultrastructural Studies of the Molecular Assembly in Biomembranes: Diversity and Similarity SEK-WENHUI
The Thermodynamics of Cell Adhesion MICAHDEMBOAND GEORGE1. BELL Rotational and Lateral Diffusion of Membrane Proteins and Lipids: Phenomena and Function MICHAELEDIDIN Biosynthesis and Distribution of Lipids KENNETHJ. LONGMUIR Lipid Exchange: Transmembrane Movement, Spontaneous Movement, and Protein-Mediated Transfer of Lipids and Cholesterol ELIEZAR A. D A W I ~ W I C Z Membrane Fusion ROBERTBLUMENTHAL The Control of Membrane Traffic on the Endocytic Pathway IRA MELLMAN,CHRISTINEHOWE, AND ARI HELENIUS Index
This Page Intentionally Left Blank