PHYSIOLOGICAL REVIEWS   Vol. 78 No. 1 January 1998, pp. 247-306
Copyright ©1998 by the American Physiological Society

Functional Significance of Cell Volume Regulatory Mechanisms

FLORIAN LANG, GILLIAN L. BUSCH, MARKUS RITTER, HARALD VÖLKL, SIEGFRIED WALDEGGER, ERICH GULBINS, AND DIETER HÄUSSINGER

Institute of Physiology, University of Tübingen, Tübingen; Department of Internal Medicine, University of Düsseldorf, Düsseldorf, Germany; and Department of Internal Medicine and Institute of Physiology, University of Innsbruck, Innsbruck, Austria

I. INTRODUCTION
II. CELL VOLUME REGULATORY MECHANISMS
    A. Ions in Steady-State Maintenance of Cell Volume
    B. Volume Regulatory Ion Transport
    C. Osmolytes
    D. Further Metabolic Pathways Contributing to Cell Volume Regulation
III. INTRACELLULAR SIGNALING OF CELL VOLUME REGULATION
    A. Macromolecular Crowding
    B. Cytoskeleton
    C. Cell Membrane Stretch
    D. Cell Membrane Potential
    E. Cytosolic pH
    F. Calcium
    G. G Proteins
    H. Protein Phosphorylation
    I. Chloride
    J. Magnesium
    K. Eicosanoids
    L. pH in Acidic Cellular Compartments
    M. Gene Expression
IV. CHALLENGES OF CELL VOLUME CONSTANCY
    A. Alterations of Extracellular Osmolarity
    B. Alterations of Extracellular Ion Composition
    C. Energy Depletion
    D. Ion Transport Altered by Hormones and Transmitters
    E. Substrate Transport
    F. Metabolism
    G. Others
V. ROLE OF CELL VOLUME REGULATORY MECHANISMS IN CELL FUNCTIONS
    A. Erythrocyte Function
    B. Epithelial Transport
    C. Regulation of Metabolism
    D. Receptor Recycling
    E. Hormone and Transmitter Release
    F. Excitability and Contraction
    G. Migration
    H. Pathogen Host Interactions
    I. Cell Proliferation
    J. Cell Death
    K. Others
REFERENCES

	ABSTRACT

Top References

Lang, Florian, Gillian L. Busch, Markus Ritter, Harald Völkl, Siegfried Waldegger, Erich Gulbins, and Dieter Häussinger. Functional Significance of Cell Volume Regulatory Mechanisms. Physiol. Rev. 78: 247-306, 1998.  To survive, cells have to avoid excessive alterations of cell volume that jeopardize structural integrity and constancy of intracellular milieu. The function of cellular proteins seems specifically sensitive to dilution and concentration, determining the extent of macromolecular crowding. Even at constant extracellular osmolarity, volume constancy of any mammalian cell is permanently challenged by transport of osmotically active substances across the cell membrane and formation or disappearance of cellular osmolarity by metabolism. Thus cell volume constancy requires the continued operation of cell volume regulatory mechanisms, including ion transport across the cell membrane as well as accumulation or disposal of organic osmolytes and metabolites. The various cell volume regulatory mechanisms are triggered by a multitude of intracellular signaling events including alterations of cell membrane potential and of intracellular ion composition, various second messenger cascades, phosphorylation of diverse target proteins, and altered gene expression. Hormones and mediators have been shown to exploit the volume regulatory machinery to exert their effects. Thus cell volume may be considered a second message in the transmission of hormonal signals. Accordingly, alterations of cell volume and volume regulatory mechanisms participate in a wide variety of cellular functions including epithelial transport, metabolism, excitation, hormone release, migration, cell proliferation, and cell death.

	I. INTRODUCTION

Top Next References

With only few exceptions (416), the membranes of animal cells are highly permeable to water (212, 776). Animal cell membranes cannot tolerate substantial hydrostatic pressure gradients, and water movement across those membranes is in large part dictated by osmotic pressure gradients (406, 445, 608, 985, 1043). Thus any imbalance of intracellular and extracellular osmolarity is paralleled by respective water movement across cell membranes and subsequent alterations of cell volume.

As outlined below, most mammalian cells are bathed in extracellular fluid with almost constant osmolarity. Nevertheless, considerable alterations of extracellular osmolarity are encountered in a variety of diseases. Excessive alterations of extracellular osmolarity occur in kidney medulla during transition between antidiuresis and diuresis (63).

Even at constant extracellular osmolarity, cell volume constancy is compromised by alterations of intracellular osmolarity. A wide variety of metabolic pathways leads to cellular formation or dissipation of osmotically active substances. Moreover, transport across the cell membrane modifies cellular osmolarity and thus cell volume.

To avoid excessive alterations of cell volume, cells have developed and utilize a multitude of volume regulatory mechanisms including transport across the cell membrane and metabolism. These mechanisms are triggered by minute alterations of cell volume. They not only serve to readjust cell volume but profoundly modify a wide variety of cellular functions. Thus cell volume is an integral element within the cellular machinery regulating cellular performance.

It is the aim of this review to illustrate the functional significance of cell volume. To this end, a description of the volume regulatory mechanisms and the cellular functions sensitive to cell volume is followed by a discussion of the role of cell volume in several integrated cell functions.

This review does not consider volume regulatory mechanisms in prokaryotic cells or comparative aspects of cell volume regulation, and the reader may refer to the respective pertinent literature (84, 96, 128, 175, 225, 232, 236, 309, 336, 387-391, 395, 400, 604, 698, 763, 792, 852, 970, 1121, 1138, 1168).

Instead, the discussion focuses on the significance of cell volume for the performance of mammalian cells. Moreover, the paper stresses recent developments. For a more complete coverage of earlier literature, the reader may refer to previous reviews on cell volume regulation (114, 425, 539, 545, 682, 776, 778, 815, 817, 843, 911a, 971, 1061, 1168), osmolytes (63, 142, 144, 370), and the role of cell volume in regulation of cell function (495, 693).

For various functions, experimental evidence pointing to the involvement of cell volume and cell volume regulatory mechanisms is intriguing but far from conclusive. It is hoped that this review stimulates further experimental effort in this exciting area of research to clarify the many remaining questions.

	II. CELL VOLUME REGULATORY MECHANISMS

Top Previous Next References

Rapid changes of cell volume are usually caused by movement of water across the cell membrane (J_v), which is driven by a hydrostatic (p) and osmotic () pressure gradient and depends on the hydraulic conductivity of the cell membrane (L_p) depends on the effective concentration difference across the cell membrane (c) and the reflection coefficient () for each solute i where R and T are the gas constant and the absolute temperature, respectively. Application of the equations would require that the cytosol behaves as a dilute solution. This is not entirely true, as discussed elsewhere in detail (194-196, 357, 722). The cell membrane does not withstand hydrostatic pressure gradients exceeding 2 kPa (445), which is equivalent to 1 mmol/l c. Even though interaction of the cell membrane with the cytoskeleton may allow the hydrostatic gradient to become larger (382, 577, 828), the movement of water is mainly dictated by osmotic gradients across the cell membrane.

The L_p depends on the presence of water channels (aquaporins) in the cell membrane, a family of molecules, which are inserted into the cell membrane and allow the passage of water (6, 214, 270, 300, 349, 356, 494, 536, 565, 773, 980). The aquaporins are especially important in water transporting epithelia but may be expressed in nonepithelial cells. Even though osmotically driven water transport is an obvious requirement for osmotic cell swelling and cell volume regulation, water movement across the cell membrane is rarely a limiting factor in cell volume changes. Thus alterations of intra- or extracellular osmolarity are in general followed by the respective movements of water and alterations of cell volume.

Cell volume regulatory mechanisms are thus most conveniently disclosed by exposing cells to abrupt changes of extracellular osmolarity. If cells are exposed to hypotonic extracellular fluid, they initially swell as more or less perfect osmometers but then approach the original cell volume by so-called regulatory cell volume decrease (RVD). If cells are exposed to hypertonic extracellular fluid, they initially shrink like almost perfect osmometers but then may approach the original cell volume by so-called regulatory cell volume increase (RVI). It should be kept in mind, however, that exposure of cells to anisotonic extracellular fluid does not only modify cell volume but also the volume of intracellular organelles such as mitochondria (969). Furthermore, in parallel to cellular osmolarity, cellular ionic strength is altered even if extracellular ionic strength is kept constant. Thus the sequelae of osmotic alterations of cell volume are not necessarily identical to the consequences of isotonic alterations of cell volume.

Alterations of cell volume may be limited by constraints from extracellular tissue, as shown for the brain (1220), the heart (972), and renal proximal tubules (753, 755). More important, however, is the ability of cells to adjust intracellular osmolarity by ion movement across the cell membrane and by generation, breakdown, uptake, or release of organic substances.

Ions contribute the bulk of intracellular (mainly K⁺) and extracellular (mainly NaCl) osmolarity. Furthermore, ions contribute some two-thirds to cell volume regulation after rapid alterations of extracellular osmolarity (331, 1266). Thus ion transport across the cell membrane is of paramount importance for the regulation of cell volume. Largely because of volume regulatory ion transport, RVD and RVI are accomplished within minutes after exposure to anisotonic media.

To maintain their metabolic functions, cells have to accumulate a number of substances, such as proteins, amino acids, or carbohydrate metabolites. The concentration of these substances is higher within the cells than in extracellular fluid. The excess cellular concentrations of these organic substances are counterbalanced by lower intracellular ion concentration. Most cells extrude Na⁺ in exchange for K⁺ by the Na⁺-K⁺-ATPase. The cell membrane is only poorly permeable to Na⁺, and the exclusion of impermeable Na⁺ outweighs the cellular osmolarity created by impermeant organic solutes (double Donnan hypothesis; Refs. 721, 776). On the other hand, the cell membrane is highly permeable to K⁺. The exit of K⁺ creates an outside-positive cell membrane potential, which drives Cl out of the cell. The low intracellular Cl concentration compensates for the excess intracellular concentration of organic substances.

Inhibition of the Na⁺-K⁺-ATPase with ouabain eventually leads to cell swelling (see Table 2) because of dissipation of the Na⁺ and K⁺ gradients, depolarization of the cell membrane, and subsequent entry of Cl into the cells (688, 779). However, inhibition of the Na⁺-K⁺-ATPase does not always lead to a rapid increase of cell volume, which may remain constant (407, 779, 924, 1039) or even transiently decrease (16, 631, 919, 1131). A sequence of events leading to cell shrinkage after inhibition of the Na⁺-K⁺-ATPase includes increase of intracellular Na⁺ activity, reversal of Na⁺/Ca²⁺ exchange, increase of intracellular Ca²⁺ activity, subsequent activation of Ca²⁺-sensitive K⁺ channels, hyperpolarization (despite decrease of intracellular K⁺ activity), and cellular KCl loss. Obviously, the time required by ouabain to eventually cause cell swelling depends on Na⁺ entry. In thick ascending limb (519, 520, 1178) or diluting segment (444) of the nephron, for instance, swelling can be delayed by inhibition of Na⁺-K⁺-2Cl cotransport.

Under the influence of ouabain, hepatocytes and renal cortical cells are apparently able to maintain their cell volume by electrolyte accumulation in intracellular vesicles, which are subsequently expelled by exocytosis (1038, 1039, 1257-1260). The electrolyte accumulation is accomplished by a H⁺-ATPase in parallel to Cl channels (1039). At least theoretically, a H⁺-ATPase in the plasma membrane could similarly maintain cell volume by creating a cell negative potential difference across the cell membrane, thus driving Cl extrusion.

As indicated above, ion transport across the cell membrane is the most efficient and rapid means of altering cellular osmolarity. During cell swelling, cells extrude ions, thus accomplishing RVD, whereas during cell shrinkage, cells accumulate ions to achieve RVI. The activation of ion release during RVD is paralleled by inhibition of ion uptake mechanisms, and the ion uptake during RVI is paralleled by inhibition of ion release mechanisms. Thus the simultaneous stimulation of ionic mechanisms for RVD and RVI is largely avoided (927, 938). A tremendous amount of work has been dedicated to the elucidation of the ion transport systems in different tissues. A synopsis of tissue-specific transport systems is beyond the scope of this review and has been reviewed in detail elsewhere (682).

1. Regulatory cell volume decrease

The transport systems most often activated by cell swelling are separate K⁺ and anion channels. In several studies, the anion channels activated by cell swelling have been found to be nonselective, allowing the passage not only of Cl but also of HCO₃ (

2. Regulatory cell volume increase

The major ion transport systems accomplishing electrolyte accumulation in shrunken cells are the Na⁺-K⁺-2Cl cotransporter (

The cellular accumulation of electrolytes after cell shrinkage is limited because high ion concentrations interfere with structure and function of macromolecules, including proteins (25, 139, 192, 193, 413, 491, 564, 573, 1376, 1381). Furthermore, alterations of ion gradients across the cell membrane would affect the respective transporters. An increase of intracellular Na⁺ activity, for instance, would reverse Na⁺/Ca²⁺ exchange and thus increase intracellular Ca²⁺ activity, which would in turn affect a multitude of cellular functions (1226).

To circumvent the untoward effects of disturbed ion composition, cells produce so-called osmolytes, molecules specifically designed to create osmolarity without compromising other cell functions (37, 44, 141, 371, 384, 484, 629, 630, 714, 875, 1138, 1377, 1378). Unlike ions, organic osmolytes even at high concentrations are compatible with normal macromolecular function. Thus the term compatible osmolytes has been coined (129).

Three groups of osmolytes are used in mammalian cells: polyalcohols, such as sorbitol and inositol; methylamines, such as glycerophosphorylcholine and betaine; and amino acids and amino acid derivatives, such as glycine, glutamine, glutamate, aspartate, and taurine (141, 368, 370, 371, 629, 630, 714, 719, 724, 1077, 1381). Tissue-specific utilization of the various osmolytes has been reviewed elsewhere in detail (682).

Osmolytes are specifically important for cell volume regulation in renal medulla, where extracellular osmolarity may become more than fourfold that of isotonicity (62, 64-66, 99, 112, 141, 369, 399, 442, 452, 525, 666, 718, 724, 855, 1075, 1076, 1078, 1352, 1353, 1356, 1377, 1378), and in the brain, where cell volume alterations cannot be tolerated due to the rigid skull and where alterations of ion composition would affect excitability (453, 524, 715-717, 745, 746, 915, 1155, 1166, 1167, 1170, 1220, 1221, 1266, 1270).

Osmolytes not only replace ions as osmotically active species but also stabilize macromolecules, thus counteracting the adverse effects of inorganic (such as K⁺, Na⁺, and Cl) and organic (such as spermine) ions (26, 138, 139, 213, 334, 662, 1271, 1351). Furthermore, betaine and glycerophosphorylcholine, and to a lesser extent inositol, counteract the destabilizing effect of urea on proteins (145, 203, 384, 483, 750, 879, 966, 1119, 1380, 1382). For normal cell function, an appropriate balance must be maintained between destabilizing (i.e., ions, urea) and stabilizing (i.e., counteracting osmolytes) forces (19, 267, 807, 863, 1272, 1376, 1383). Accordingly, an increase of urea concentration specifically favors the parallel increase of glycerophosphorylcholine (723, 855, 966).

Beyond their function in cell volume regulation, osmolytes are protective against the destructive effects of excessive temperatures (34, 292, 352, 383, 534, 579, 759, 926, 989, 1058, 1160, 1193, 1200) and dessication (169, 232). Furthermore, they have been found to ease cell membrane assembly (642).

Cellular osmolyte accumulation can be achieved by stimulated uptake, enhanced formation, or decreased degradation. Decrease of intracellular osmolyte concentration is accomplished by degradation or release. As compared with RVI accomplished by ions, accumulation of osmolytes is a slow process taking hours to days.

6. Amino acids

In addition to taurine, the cellular concentration of several other amino acids and amino acid metabolites is modified by cell volume, including glutamine, glutamate, glycine, proline, serine, threonine, -alanine, (N-acetyl)aspartate, and GABA (for review, see Refs.

In addition to amino acids, numerous organic metabolites contribute to cellular osmolarity. Several metabolic pathways known to be sensitive to cell volume may modify the concentrations of these metabolites and thus contribute to cell volume regulation. Cell swelling increases glycogen synthesis and inhibits glycolysis, thus decreasing the concentration of carbohydrate metabolites (12, 49, 50, 696, 819, 953). Furthermore, cell swelling has a relatively weak stimulatory effect on lipogenesis (51). As detailed in section VC, cell volume changes interfere with a great number of other metabolic functions that to some extent may modify cellular osmolarity. The overall impact of these effects on cellular osmolarity is probably modest, but the influence of cell volume on various metabolic pathways is of paramount importance for regulation of metabolic function (see sect. VC).

	III. INTRACELLULAR SIGNALING OF CELL VOLUME REGULATION

Top Previous Next References

Cell swelling and shrinkage exert profound effects on intracellular signaling mechanisms, which in turn modify a multitude of cellular functions including the volume regulatory mechanisms. A great deal of experimental effort has been spent in elucidating the intracellular machinery underlying cell volume regulation. Frequently, the result has been inconclusive for several reasons. 1) Not every effect of altered cell volume on intracellular signaling is related to regulation of cell volume. 2) Cells usually use several mechanisms in parallel, with different, partially overlapping cellular signaling mechanisms. 3) Different cells utilize distinct mechanisms, i.e., the information gained in any given cell cannot necessarily be generalized to other cells. 4) Volume regulation requires mechanisms that are themselves not modified by cell volume changes but rather permissive for activation of cell volume regulatory mechanisms.

In the simplest case, an intracellular mechanism serves cell volume regulation if it is modified by alterations of cell volume, and a qualitatively and quantitatively identical modification of this intracellular mechanism triggers the appropriate alterations of cell volume. This requirement frequently is not met. If the identical modification does not trigger respective alterations of cell volume, the participation of a mechanism in cell volume regulation still cannot be ruled out, since the mechanism may require the collaboration of other mechanisms to be effective. Similarly, the use of inhibitors, even if they are specific, does not lead to conclusive results. On the one hand, inhibition of cell volume regulation by elimination of a given element of intracellular signaling (e.g., inhibition of protein kinases or removal of Ca²⁺) does not discriminate between volume regulatory and permissive mechanisms. On the other hand, cell volume regulation may prove insensitive to elimination of a volume regulatory mechanism if other mechanisms operating in parallel are strong enough to replace the defect. Further examples could be given, each of which illustrates that the experimental elucidation of the complex machinery serving cell volume regulation is extremely difficult.

Even though our understanding of the intracellular machinery mediating cell volume regulation is still incomplete, knowledge of the interaction between cell volume, elements of cellular signaling, and cell volume regulatory mechanisms is mandatory for understanding the role of cell volume for cell function.

Cell swelling leads to dilution and cell shrinkage to concentration of cellular constituents including proteins. The concentration of intracellular proteins markedly influences their function (131, 353, 376, 834, 836, 937, 1011). In erythrocytes, the volume regulatory set point can indeed be varied by manipulation of intracellular protein concentration (835, 837, 1397). The set points of both the KCl symport (835, 838) and the Na⁺/H⁺ exchanger (209, 210, 940) appear to be determined by macromolecular crowding. It has been suggested that among the enzymes sensitive to ambient protein concentration is a kinase that is inactivated by protein dilution during cell swelling and activated by protein crowding during cell shrinkage (835, 838). This kinase may inhibit the volume regulatory KCl cotransport. Its inactivation during cell swelling would then disinhibit volume regulatory KCl efflux.

Because of interaction of proteins with ambient electrolytes, macromolecular crowding is reduced by increasing ionic strength, which indeed shifts the volume regulatory set point to smaller volumes (942).

Similarly, urea decreases the thermodynamic activity of proteins and thus reduces macromolecular crowding (211, 574, 723, 837, 978, 984, 1376). Urea has been shown to activate erythrocyte KCl transport (293, 596, 936), erythrocyte Na⁺/Ca²⁺ exchange (936), and hepatocyte K⁺ channels (474) and to inhibit Na⁺/H⁺ exchanger in erythrocytes (936) and adenosine 3',5'-cyclic monophosphate (cAMP) production (58), Na⁺/H⁺ exchanger (732), and Na⁺-K⁺-2Cl cotransport (595) in thick ascending limb cells, thus leading to cell shrinkage. In erythrocytes, the effect of urea was reversed by okadaic acid, pointing to the involvement of phosphorylation (936).

A variety of ion channels are activated by cell membrane stretch, i.e., increased tension of the cell membrane (1040, 1043). Stretch increases the open probability of the channels without affecting single-channel conductance or selectivity of the channels (1043).

The stretch-activated channels may be selective for K⁺ or for anions, thus directly serving cell volume regulation (857, 1043). Most of these channels, however, are nonselective cation channels, allowing the passage of K⁺, Na⁺, and Ca²⁺ (for review, see Refs. 682, 1043). Because of the cell-negative cell membrane potential, the respective electrochemical gradients favor the cellular accumulation of Na⁺ and Ca²⁺ rather than cellular loss of K⁺. Thus these channels are not likely to directly serve cell volume regulation, and inhibition of these channels by gadolinium has been shown to decrease osmotic swelling and favor regulatory decrease of cell volume (1173). On the other hand, Ca²⁺ entering the cells through these channels is thought to activate Ca²⁺-sensitive K⁺ channels (187, 1194, 1234).

The mechanism linking membrane stretch to activation of the channels has not been clearly defined (1043). Under discussion are 1) release of fatty acids from the stretched membrane and subsequent activation of stretch-sensitive channels by these fatty acids (917) and 2) stretch-induced activation of some element of the cytoskeleton, such as spectrin (1133). Because stretch enhances channel open probability in the cell-free excised patch configuration (1043), cytosolic components are apparently not required for channel activation.

It is debatable whether stretch-activated channels participate in the fine-tuning of cell volume, since considerable stretch is required to activate these channels (1043). Possibly, these channels may represent a last line of defense against excessive cell swelling but are not involved in the response to moderate changes of cell volume.

The influence of cell swelling on cell membrane potential depends on the ion channels preferentially activated or inactivated and on the potential difference before cell swelling. Activation of K⁺ channels and a low initial cell membrane potential favor hyperpolarization, whereas activation of anion or nonselective cation channels and a high initial cell membrane potential would favor depolarization. After cell swelling, hyperpolarization of the cell membrane is seen in hepatocytes (406), depolarization of the cell membrane in Ehrlich ascites tumor cells (680, 691), Madin-Darby canine kidney (MDCK) cells (947), opossum kidney cells (1235), lymphocytes (418, 419, 1060), pancreatic -cells (124), astrocytes (627), neuroblastoma cells (313), and vascular smooth muscle cells (685). In some cells, a transient hyperpolarization due to activation of K⁺ channels is followed by a more sustained depolarization due to activation of anion channels (516-518, 1002).

The alteration of cell membrane potential may influence the activity of additional ion channels. A depolarization of the cell membrane may open voltage-sensitive ion channels. In lymphocytes, RVD involves n-type K⁺ channels (273, 429), which are activated by cell membrane depolarization. Depolarization of the cell membrane may further activate voltage-sensitive Ca²⁺ channels, as observed in pancreatic -cells (124) and vascular smooth muscle cells (685). The increase of intracellular Ca²⁺ could trigger a variety of further mechanisms, as illustrated in section V, E and F.

As discussed in section IIB, increase of extracellular osmolarity may either depolarize or hyperpolarize cells due to activation of unspecific cation channels or inhibition of K⁺ and/or Cl channels.

Cell swelling leads to cytosolic acidification (216, 397, 479, 610, 616, 685, 757, 864, 1002, 1089, 1143, 1279), which has been explained by the exit of HCO₃ through anion channels (1334), by release of H⁺ from acidic intracellular compartments (see sect. IIIL), and by enhancement of Cl/HCO₃ exchange due to decreasing cellular Cl activity (757). This latter mechanism, however, should be impeded by inhibition of the exchanger at acidic cytosolic pH (645), and cellular acidosis is not inhibited by removal of extracellular Cl, at least in osteosarcoma cells (864).

Cell shrinkage is frequently observed to alkalinize cells, at least partially due to activation of volume regulatory Na⁺/H⁺ exchange (see sect. IIB).

The impact of intracellular pH on volume regulatory mechanisms has not been explored. After cell swelling, the cytosolic acidification may impede activation of volume regulatory K⁺ channels (783) and may contribute to inhibition of glycolysis and thus to the decreased release of lactic acid (696). The alkalinization after cell shrinkage should stimulate glycolysis (696).

After cell swelling, intracellular Ca²⁺ concentration ([Ca²⁺]_i) increases in a variety of cells, whereas it remains apparently constant in others (for review, see Refs. 682, 815). Swelling may increase [Ca²⁺]_i by both activation of Ca²⁺-permeable channels in the cell membrane and Ca²⁺ release from intracellular stores. Calcium-permeable channels may be activated by cell membrane stretch (see sect. IIIC ), cell membrane depolarization (see sect. IIID), and/or protein kinase C (1066). Calcium release from intracellular stores is presumably triggered by inositol phosphates (52, 72, 1180, 1288, 1289) or Ca²⁺-induced Ca²⁺ release (516, 518). The regulation of [Ca²⁺]_i by cell volume may interfere with signaling of Ca²⁺-recruiting hormones, as shown for gastric parietal cells (885) and HT-29 cells (330). In those cells, agonist-induced entry of Ca²⁺ was further stimulated by cell swelling (330, 885) and inhibited by cell shrinkage (885).

The increase of [Ca²⁺]_i accounts for the activation of Ca²⁺-sensitive K⁺ channels, as shown in Ehrlich ascites tumor cells (188), MDCK cells (1334), proximal tubule cells (290, 605), thick ascending limb cells (1194), choroid plexus epithelial cells (187), and neuroblastoma cells (313). In MDCK cells, [Ca²⁺]_i did not appreciably increase upon moderate osmotic cell swelling, even though the Ca²⁺-sensitive K⁺ channels were already activated by this treatment (1002). Possibly small, localized increases of [Ca²⁺]_i are sufficient to activate the K⁺ channels but are not detected by fluorescence measurements (1357).

Despite the activation of Ca²⁺-sensitive K⁺ channels, RVD is apparently not mediated by a rise of [Ca²⁺]_i in Ehrlich ascites tumor cells (1204), and swelling-induced K⁺ efflux was virtually unaffected by the inhibitors of the Ca²⁺-sensitive K⁺ channels, clotrimazole and charybdotoxin (489). Because these K⁺ channels are inwardly rectifying (1334), K⁺ exit through these channels during cell swelling may be limited by the depolarization of the cell membrane. Other less inwardly rectifying K⁺ channels may thus be more important for cell volume regulation in those cells (539). In a variety of tissues, increase of [Ca²⁺]_i was not required for RVD (for review, see Ref. 682). Clearly, activation of K⁺ channels by Ca²⁺ may contribute to, but is frequently not crucial for, RVD (539).

In contrast to volume regulatory K⁺ channels in many tissues, volume regulatory Cl channels have been found to be insensitive to Ca²⁺ in intestinal cells (517, 657), ciliary epithelial cells (1385), cardiac myocytes (1227), T84 colon carcinoma cells (1134, 1364), airway epithelial cells (361, 1134), MDCK cells (48, 1029, 1334), lymphocytes (421), Ehrlich cells (188), and chromaffin cells (287). Yet, as shown in Ehrlich ascites tumor cells, Ca²⁺ triggers the formation of leukotrienes, which do activate the channels (539). In kidney cortex, Ca²⁺-sensitive Cl channels have been found in endosomes and proposed to serve cell volume regulation (991). Calcium is also thought to trigger the fusion of vesicles, allowing the release of sorbitol (629), whereas it is apparently not required for release of GPC (629) or taurine (1051).

Calmodulin antagonists and inhibitory peptides against Ca²⁺/calmodulin-dependent kinase (116) have been shown to inhibit RVD or activation of cell volume regulatory ion channels (71, 263, 421, 424, 543, 546, 815), and it has been concluded that calmodulin/Ca²⁺ complexes are important for activation of RVD. In other cells, however, Ca²⁺-sensitive K⁺ channels involved in cell volume regulation did not require calmodulin and were actually activated by calmodulin antagonists (950).

Beyond its putative role during RVD, calmodulin has been implicated in activation of the Na⁺-K⁺-2Cl cotransport in RVI (583).

Inhibitors of G proteins such as pertussis toxin or cholera toxin have been shown to blunt the RVD (539) as well as swelling-induced osmolyte efflux (1037), increases of intracellular Ca²⁺ concentration (30), mitogen-activated protein kinase (MAPK) activity (895, 1072), vesicular acidification (1088), and swelling-induced stimulation of taurocholate excretion (895), suggesting that G proteins do mediate some effects of cell swelling. Activation of the Na⁺/H⁺ exchanger during cell shrinkage has similarly been claimed to involve G proteins (108, 249).

In addition to heterotrimeric G proteins, small G proteins have been implicated in cell volume regulation. Clostridium botulinus C₃ exoenzyme, which depolymerizes the actin filament network by ADP-ribosylation of rho (8), blunts the volume regulatory anion efflux (1209). In neurons, osmotic cell shrinkage stimulates the expression of ₁-chimerin (286), a GTPase-activating protein that inactivates the small G protein Rac. The impact of this effect on cell volume regulation is not explored.

Decreased intracellular Cl activity is apparently required for full activation of Na⁺-K⁺-2Cl cotransport during cell shrinkage (539, 548, 1245, 1370). Beyond its influence on one of the driving forces of Na⁺-K⁺-2Cl cotransport, a decreased intracellular Cl concentration is thought to play a permissive role for the activation of both Na⁺-K⁺-2Cl cotransport (119, 734, 735, 1009) and Na⁺/H⁺ exchange (108, 250, 933, 934, 1009). If extracellular osmolarity is made hypertonic by increased extracellular NaCl concentration, the increase of intracellular Cl activity could thus impede RVI (539). Accordingly, some cells are unable to regulate their volume during exposure to hypertonic extracellular fluid (308, 437, 522, 539, 544, 545, 634). If intracellular Cl is lowered by prior RVD (308, 420, 522, 634) or by activation of Cl channels with cAMP (437, 1177) or vasopressin (351, 519, 1177), the same cells do accomplish RVI. Moreover, if cells are exposed to short-chain fatty acids, they swell by accumulation of the acids along with Na⁺ and are forced to release Cl for RVD (see Table 2). In the presence of these acids, they display RVI after exposure to hypertonic extracellular fluid (1016).

Intracellular Cl inhibits the glycogen synthase phosphatase (408), and the decrease of intracellular Cl activity participates in the stimulation of glycogen synthesis during osmotic or glutamine-induced cell swelling (555, 819).

The dilution and concentration of intracellular solutes during cell swelling or shrinkage lead to the respective alterations of Mg²⁺ concentration. Magnesium has been described to inhibit volume regulatory KCl cotransport (78, 295). During cell swelling, the decrease of intracellular Mg²⁺ concentration may partially account for the activation of KCl cotransport (78, 295). Conversely, an increase of intracellular Mg²⁺ activity stimulates Na⁺/H⁺ exchange (944) and Na⁺-K⁺-2Cl cotransport (794) and may thus participate in regulatory cell volume increase.

Cell swelling has been shown to activate phospholipase A₂ (542, 800), possibly in part through decrease of macromolecular crowding (539). Metabolites of arachidonic acid include the products of cyclooxygenase (e.g., prostaglandins), 15-lipoxygenase (such as hepoxilin A₃), 5-lipoxygenase [e.g., leukotriene (LT) D₄], and epoxygenase (epoxyeicosatrienoic acids) (675). On the other hand, as shown in Ehrlich ascites tumor cells, swelling stimulates the formation of the leukotrienes, namely, LTD₄ , at the expense of prostaglandins such as prostaglandin (PG) E₂ (675). Both phospholipase A₂ and 5-lipoxygenase are activated by Ca²⁺ (542, 675). Thus an increase of intracellular Ca²⁺ concentration during cell swelling could participate in the activation of these two enzymes during cell swelling. Along these lines, LTD₄ may overcome the inhibitory effect of the calmodulin antagonist pimozide on cell volume regulation, suggesting that LTD₄ is a signal downstream of Ca²⁺ (673).

Arachidonic acid has been shown to inhibit glial cell volume regulation (1052) and to inhibit volume regulatory Cl channels (403, 673, 679, 1047). On the other hand, it may increase Ca²⁺ concentration in renal collecting duct and activate K⁺ channels in neurons (622, 1213). Moreover, the 15-lipoxygenase product of arachidonic acid hepoxilin A₃ activates volume regulatory K⁺ channels in platelets (798-801), and the 5-lipoxygenase product LTD₄ activates volume regulatory K⁺ and Cl channels (547, 592, 674, 675, 679) and volume regulatory taurine release (592, 678) in Ehrlich ascites tumor cells, as well as taurine release in fish erythrocytes (1207). 5-Lipoxygenase products similarly appear to mediate regulatory cell volume decrease in colonic epithelium (277, 281, 282), chromaffin cells (287), MDCK cells (947), and human fibroblasts (808). However, in most of these cell types, the evidence comes largely from effects of 5-lipoxygenase inhibitors such as nordihydroguaiaretic acid (NDGA). In Ehrlich ascites tumor cells (679) and proximal renal tubules (Völkl and Lang, unpublished observations), on the other hand, the inhibitory effect of NDGA was not overcome by the addition of LTD₄ , pointing to an additional effect of the drug not related to 5-lipoxygenase inhibition. It may more directly inhibit the volume regulatory Cl channels (438) or be effective by increasing arachidonic acid concentration (1052).

The enhanced formation of leukotrienes in swollen Ehrlich ascites tumor cells parallels a decreased formation of PGE₂ , an effect possibly accounting for the inhibition of Na⁺ channels (679). Those channels are thought to be stimulated by PGE₂ (679). On the other hand, in ciliary epithelial cells, PGE₂ was thought to mediate the activation of volume regulatory K⁺ channels during cell swelling (191). Prostaglandin E₂ similarly activates K⁺ channels in MDCK cells (1153) and erythrocytes (743).

In collecting duct principal cells, inhibition of phospholipase A₂ with quinacrine blunted the activation of volume regulatory Ca²⁺-sensitive K⁺ channels, which, on the other hand, were activated by arachidonic acid (751). In LLC-PK₁ cells, however, volume regulatory rubidium flux was not modified by arachidonic acid, even though it was inhibited by an arachidonic acid antagonist (262).

Ketoconazole, an inhibitor of epoxygenase (cytochrome P-450), impedes volume regulatory efflux of osmolytes, such as sorbitol, betaine, myo-inositol, or amino acids from renal papillary cells (354), MDCK cells (38), and C6 glioma cells (818, 1171). However, the inhibitory effect is not reversed by addition of hydroxyeicosatetraenoic acids (818), indicating that the inhibitory effect on osmolyte flux is not due to inhibition of epoxygenase. Similarly, in Necturus gallbladder, ketoconazole does not prevent regulatory KCl efflux but stimulates NaCl entry, leading to cell swelling (615).

In addition to their influence on volume-sensitive ion channels, phospholipase A₂ and a cytochrome P-450 product of arachidonic acid have been invoked in mediating the swelling-induced cellular release of sorbitol (354).

The fatty acid composition of the cell membrane can be modulated by dietary polyunsaturated fatty acids, which lead to enhanced formation of leukotrienes and thus to acceleration of RVD in Ehrlich ascites tumor cells (712).

As evidenced from acridine orange and fluorescein isothiocyanate-dextran fluorescence, hepatocyte swelling leads to alkalinization of acidic cellular compartments, whereas cell shrinkage enhances the acidity in those compartments (156, 683, 1088, 1089, 1279, 1280).

The lysosomal proteases are known to have their pH optimum in the acidic range, and alkalinization of the lysosomes is well known to inhibit hepatic proteolysis (859, 860). Thus the alkalinizing effect on acidic cellular compartments could at least in theory contribute to the antiproteolytic action of cell swelling. Along these lines, alkalinization of acidic cellular compartments parallels the cell swelling and antiproteolytic effect of transforming growth factor-1 on LLC-PK₁ cells (752). However, the contribution of vesicular alkalinization to the antiproteolytic effect of cell swelling remains to be proven. At least in liver cells, swelling alkalinizes prelysosomal rather than lysosomal compartments (766, 1090). Moreover, inhibition of tyro