JEZ 860 THE JOURNAL OF EXPERIMENTAL ZOOLOGY 279:471–475 (1997) ATP Regulation of a Swelling-Activated Osmolyte Channel in Skate Hepatocytes NAZZARENO BALLATORI1* AND JAMES L. BOYER2 Department of Environmental Medicine, University of Rochester School of Medicine, Rochester, New York 14642 2 Department of Medicine and Liver Center, Yale University School of Medicine, New Haven, Connecticut 06520 1 ABSTRACT Hypotonic swelling of isolated skate hepatocytes activates a regulatory volume decrease (RVD) which is achieved in part by the release of taurine and other intracellular organic osmolytes. Volume-activated taurine efflux appears to be mediated by an anion channel that exhibits a taurine/chloride permeability ratio of approximately 0.2. Of significance, this channel was shown to be regulated by intracellular nucleotide. When intracellular ATP was decreased to about 50% of control levels, channel opening was completely prevented. Many putative ion channel blockers were found to inhibit the channel indirectly, by depleting intracellular ATP, rather than by directly interacting with the channel. Investigators using these channel blockers in whole cell preparations should be aware of this alternative mechanism. Cell swelling-activated taurine efflux was also inhibited by HgCl2, DIDS, and pyridoxal 5-phosphate, at concentrations of these agents that had no effect on intracellular ATP levels, suggesting additional mechanisms of inhibition and regulation of the volume-sensitive osmolyte channels. J. Exp. Zool. 279:471475, 1997. © 1997 Wiley-Liss, Inc. Mechanisms of regulatory volume decrease (RVD) in animal cells occur through the release of major intracellular inorganic ions (potassium and chloride) as well as the release of organic osmolytes. The relative contribution of the inorganic versus the organic osmolytes to the RVD differs between species and even among tissues of a given species. Most marine organisms have relatively high concentrations of organic osmolytes, and RVD in these animals is associated with a loss of compounds such as taurine, betaine, sorbitol, inositol, and tetramethylamines. Release of organic osmolytes avoids the potential perturbing effects associated with reductions in major intracellular ions. Our studies characterized the mechanism for RVD in hepatocytes isolated from the small skate (Raja erinacea), an osmoconforming marine elasmobranch. Skate hepatocytes are ideal models for these studies since they contain high concentrations of organic osmolytes, including high concentrations of taurine (65 mM; Ballatori and Boyer, ’92a). Skate hepatocytes are easily isolated and maintained in cell suspension with high viabilities, and are also easily amenable to patch clamp analysis. Our studies demonstrated a requirement for intracellular ATP in the activation of a volume-regulatory organic osmolyte channel. © 1997 WILEY-LISS, INC. MECHANISM FOR RVD IN SKATE HEPATOCYTES When exposed to an acute hypotonic shock, skate hepatocytes swell in proportion to the degree of dilution, behaving as nearly perfect osmometers (Ballatori and Boyer, ’92b). However, volume regulatory mechanisms are rapidly activated after cell swelling, leading to a gradual recovery of cell volume over the next 30–40 minutes to approach their original size. To examine whether intracellular K+ is involved in the RVD response in skate hepatocytes, cellular K+ content was measured by flame photometry, as well as in cells preloaded with 86Rb+ as a marker for K+. The results demonstrated that cell swelling had only small effects on K+ content: potassium content was decreased only ~5 and 10% by 40 and 50% dilution with water, respectively (Ballatori and Boyer, ’92b). Because these relatively small changes in K+ efflux in hypotonic medium could not account for the loss of osmolytes during skate hepatocyte RVD, additional studies examined the role of another major intracellular osmolyte, taurine, in cell *Correspondence to: Ned Ballatori, Ph.D., Department of Environmental Medicine, University of Rochester School of Medicine, Rochester, NY 14642. E-mail: email@example.com 472 N. BALLATORI AND J.L. BOYER volume regulation. In contrast to potassium, there was a marked stimulation of taurine release after cell swelling: cells diluted with 30, 40, or 50% H2O released 24, 41, and 65% of intracellular [14C]taurine over 2 h, respectively (Ballatori and Boyer, ’92b). Most of the taurine was released during the first 30 min after dilution, with rates of efflux approaching baseline rates by 60 min. Of significance, the time course of taurine release paralleled that of volume recovery, in that most of the RVD also occurred during the first 30 min after dilution. Direct amino acid analysis confirmed that there was a net loss of taurine from the hepatocytes (Ballatori and Boyer, ’92b). PROPERTIES OF THE SWELLINGACTIVATED TAURINE EFFLUX PATHWAY IN SKATE HEPATOCYTES To distinguish whether volume-activated taurine efflux was occurring by a carrier mediated mechanism or by activation of a channel, additional studies examined [14C]taurine uptake and efflux during RVD in cells incubated in media containing taurine concentrations from 0.1–100 mM. Despite increasing concentrations of taurine in the medium, rate coefficients for [14C]taurine uptake and efflux were unchanged (Ballatori et al.,’94). This kinetic pattern is consistent with the activation of a channel rather than a carrier-mediated process. Studies carried out in collaboration with Paul Jackson and Kevin Strange (Ballatori et al., ’95; Jackson et al., ’96) confirmed the presence of a volume-activated anion channel in skate hepatocytes. When the patch pipette contained taurine at pH 8.2 so that significant amounts of this amino acid were in anionic form, an inward rectifying taurine current was demonstrated. Permeability and ion conductances for various anions showed characteristics identical to volume activated anion currents in glia cells previously demonstrated by Jackson and Strange (’93; Jackson et al., ’94). The relative Ptaurine/PCl in skate hepatocytes was 0.17. Studies in the isolated perfused skate liver demonstrated that the channel was localized to the basolateral membrane of the liver cell (Ballatori et al.,’94). The channel was immediately inactivated when isotonicity was restored and swollen hepatocytes rapidly returned to their normal volumes, but was unaffected by the transmembrane Na+, Cl–, or K+ gradients (Ballatori and Boyer, ’92b; Ballatori et al., ’94), indicating that neither Na+, Cl–, nor membrane potential are directly involved in volume-stimulated taurine transport. ROLE OF INTRACELLULAR ATP IN ACTIVATION OF THE TAURINE CHANNEL Volume-activated taurine efflux from skate hepatocytes was nearly completely blocked by decreasing the incubation temperature from 15° to 4°C, and by several metabolic inhibitors (Ballatori and Boyer, ’92b). Cells pretreated for 30 min with 2,4-dinitrophenol, oligomycin, CCCP, antimycin A, iodoacetate, KCN, sodium azide, or a combination of iodoacetate plus KCN or iodoacetate plus azide, released relatively little [14C]taurine following hypotonic cell swelling. The role of intracellular ATP in volume-activated taurine transport was studied further by examining the effects of ATP depletion at different times after the hypotonic stimulus. Administration of 2,4-dinitrophenol, a highly membrane-permeant metabolic inhibitor nearly completely prevented any further release of taurine, when added at different times after cell swelling, indicating that taurine release requires the continual presence of intracellular ATP (Ballatori et al., ’94). Comparable effects were noted with two other metabolic inhibitors, antimycin A and the combination of KCN plus iodoacetate. The quantitative relation between cellular ATP content and volume-activated [14C]taurine efflux was assessed in skate hepatocytes exposed to increasing concentrations of 2,4-dinitrophenol (Fig. 1). 14C-Taurine efflux exhibited a roughly sigmoidal relationship with cellular ATP levels (Ballatori et al.,’95). Efflux was inhibited by 50% when cellular ATP declined from approximately 7 nmol/ mg protein (~2 mM) to 4 nmol/mg protein, with essentially complete inhibition when ATP levels reached 3 nmol/mg protein. When taurine efflux was plotted against cellular ATP/ADP ratios, a sigmoidal relation was also observed (Fig. 1). There was a gradual inhibition of taurine efflux until the ATP/ADP ratio reached ~2, at which point there was a precipitous decrease in swelling-activated efflux (Ballatori et al., ’95). Dilution alone produced no effect on ATP levels or ATP/ADP ratios compared to hepatocytes maintained in isotonic medium. Whole cell patch recording of volume-activated anion current in skate hepatocytes demonstrated that these effects of 2,4-dinitrophenol are not directly related to an interaction of this metabolic poison with the channel (Ballatori et al., ’95). In contrast, swelling-activated anion current was critically dependent on the presence of intracellular ATP, as demonstrated in studies where ATP AN ATP-REGULATED OSMOLYTE CHANNEL 473 Fig. 1. Relation between volume-activated [ 14C]taurine efflux and ATP content (A), ADP content (B), and ATP/ADP ratio (C), in isolated skate hepatocytes exposed to increasing concentrations of 2,4-dinitrophenol, an uncoupler of oxidative phosphorylation. Hepatocytes preloaded with [14C]taurine were incubated for 30 min with 2,4-dinitrophenol at concentrations of 0, 5, 10, 25, 50, 100, 200, 500, and 700 µM, and aliquots of the cell suspension were removed for ATP and ADP analysis and to assess 14C content of the cells. Hypotonicity was then induced by diluting the remaining cell suspensions 40% with either H2O or Ringer (control), and cellular 14 C content measured 30 min later. Values are means ± SEM of four to six experiments at each concentration of 2,4-dinitrophenol. Reprinted with permission of Williams & Wilkins from Ballatori et al. (’95). was omitted from the patch pipette (Jackson et al., ’96). When a nonhydrolyzable analog of ATP was included in the pipette, full activation of anion current was observed (Jackson et al., ’96), indicating that channel activation was not dependent on ATP hydrolysis, but rather was related to the binding of ATP to the channel or some regulatory protein. Most importantly, when various channel blockers were examined which inhibited volume-activated [14C]taurine efflux, no effect on volume-activated anion current was observed with a number of these compounds. Only DIDS and pyridoxal 5phosphate significantly inhibited volume-activated current, whereas compounds such as 2,4-dinitrophenol, ketoconazole and even NPPB produced little or no inhibition of volume-activated current (Ballatori et al., ’95). The mechanism by which these agents were inhibiting [14C]taurine efflux without affecting swelling-activated anion current became apparent when cellular ATP levels were compared (Fig. 2). NPPB, glibenclamide, DPC, ketoconazole, gossypol, niflumic acid, quinine, and phenylarsine oxide all decreased cellular ATP concentrations and ATP/ADP ratios at the same concentrations that they inhibited taurine efflux (Fig. 2; Ballatori et al.,’95). Thus, these compounds were inhibiting the swelling-activated channel indirectly, that is by depleting cellular ATP, which in turn prevented channel activation. In contrast, DIDS and pyridoxal 5-phosphate inhibited taurine efflux (Fig. 2) and whole cell anion current (Ballatori et al.,’95), but had no effect on cellular ATP levels or ATP/ADP ratios. The DIDS inhibition of whole cell anion current was rapid (<60 sec) and voltage-dependent, suggesting that DIDS interacts directly with the channel protein. Pyridoxal 5-phosphate also inhibited whole cell anion current, albeit in a different manner than DIDS. When swollen cells were exposed directly to the drug, whole cell anion current dropped very slowly or was largely unaffected. However, if the cells were pretreated with the drug for 15–30 min and then swollen, activation of whole cell current was dramatically inhibited. Anion current measured 3 min after induction of swelling in cells pretreated with pyridoxal 5-phosphate was inhibited 92% compared to untreated control cells. 474 N. BALLATORI AND J.L. BOYER Fig. 2. Effects of ion channel blockers and other transport inhibitors on volume-activated [14C]taurine efflux, cellular ATP content, and ATP/ADP ratios. Skate hepatocytes were loaded with [14C]taurine, then treated for 30 min with the indicated concentrations of inhibitors at 15°C. Control cells received an equal volume of vehicle (dimethylsulfoxide or elasmobranch Ringer). Aliquots of the cell suspensions were then removed for ATP and ADP analysis and to assess 14C content of the cells, and hypotonicity was induced by diluting the remaining cell suspensions 40% with either H2O or Ringer (control). Cellular 14C content was measured 30 min after dilution. Inhibitors were not removed prior to dilution, but remained (40% diluted) throughout the efflux period. Values are means ± SEM of three to six cell preparations, each performed in duplicate. Reprinted with permission of Williams & Wilkins from Ballatori et al. (’95). The slow inhibition of whole cell current by pyridoxal 5-phosphate suggests that the drug may be acting from an intracellular site. Studies of ion transporting ATPases have shown that pyridoxal 5-phosphate covalently modifies ATP binding sites (Hinz and Kirley, ’90; Tamura et al., ’86). Although the mechanism by which pyridoxal 5phosphate inhibits anion current and taurine efflux is not known, it may be through an interaction with an ATP binding site on the channel itself or an associated regulatory protein. MERCURY INHIBITION OF THE SWELLING-ACTIVATED TAURINE CHANNEL A major mechanism by which inorganic mercury and some organomercurial compounds produce cytotoxicity is by altering ion and nonelectrolyte transport, and cell volume regulation (Kinter and Pritchard, ’77). The inability to regulate cell volume disrupts transmembrane ion, solute, and electrical gradients, which in turn disrupt cellular communication, homeostasis, and metabolic func- AN ATP-REGULATED OSMOLYTE CHANNEL tions. Our early studies with skate hepatocytes confirmed the effects of mercuric chloride on cell swelling, and further demonstrated that HgCl2 prevents the normal RVD observed after cells are swollen in hypotonic media (Ballatori et al., ’88a). RVD in isolated skate hepatocytes was essentially abolished by pretreatment with micromolar concentrations of mercuric chloride. To examine the mechanism for inhibition of RVD, we examined whether mercury and other sulfhydryl-reactive reagents could also inhibit volume-activated taurine release. All sulfhydryl reagents tested inhibited taurine release during RVD (Ballatori and Boyer, ’92b). Pretreatment with 1 mM N-ethylmaleimide or 10 mM diamide nearly completely prevented [14C]taurine release, whereas 10 mM t-butyl hydroperoxide and 25 µM HgCl2 were somewhat less effective. In contrast to metabolic inhibitors and sulfhydryl reagents, neither ouabain (2 mM) nor furosemide (1 mM) had any effect on [14C]taurine release during RVD (Ballatori and Boyer, ’92b). Ouabain at 2 mM effectively blocks 86Rb+ uptake, and leads to a gradual depolarization of membrane potential in skate hepatocytes (Ballatori et al., ’88a). Thus, neither Na+-K+-ATPase nor the cell membrane potential appear to be directly involved in volume-stimulated taurine transport. The effects of mercury were also not explained by a decrease in cell ATP levels (Ballatori and Boyer, ’96). Swelling activated [14C]taurine efflux was progressively inhibited in the presence of increasing concentrations of HgCl2; however, ATP levels and ATP/ADP ratios remained essentially unchanged over a range of HgCl2 concentrations from 0–40 µM. These studies indicate that HgCl2 may be interacting directly with the channel; however, additional studies are needed to test this hypothesis. ACKNOWLEDGMENTS Supported in part by National Institutes of Health Grants DK48823, DK34989, and ES01247, and by Grant ES03828 to the Center for Membrane Toxicity Studies at the Mount Desert Island Biological Laboratory. 475 LITERATURE CITED Ballatori, N., C. Shi, and J.L. Boyer (1988a) Altered plasma membrane ion permeability in mercury-induced cell injury: Studies in hepatocytes of elasmobranch Raja erinacea. Toxicol. Appl. Pharmacol., 95:279–291. Ballatori, N., R. Wondergem, and J.L. Boyer (1988b) Effects of L-alanine on membrane potential, potassium (86Rb) permeability and cell volume in hepatocytes from Raja erinacea. Biochim. Biophys. Acta, 946:261–269. Ballatori, N., and J.L. Boyer (1992a) Taurine transport in skate hepatocytes: I. Uptake and efflux. Am. J. Physiol., 262:G445–G450. Ballatori, N., and J.L. Boyer (1992b) Taurine transport in skate hepatocytes: II. Volume activation, energy and sulfhydryl dependence. Am. J. Physiol., 262:G451–G460. Ballatori, N., T.W. Simmons, and J.L. Boyer (1994) A volumeactivated taurine channel in skate hepatocytes: Membrane polarity and role of intracellular ATP. Am. J. Physiol., 267:G285–G291. Ballatori, N., A.T. Truong, P.S. Jackson, K. Strange, and J.L. Boyer (1995) ATP depletion and inactivation of an ATP-sensitive taurine channel by classic ion channel blockers. Mol. Pharmacol., 48:472–476. Ballatori, N., and J.L. Boyer (1996) Disruption of cell volume regulation by mercuric chloride is mediated by an increase in sodium permeability and inhibition of an osmolyte channel in skate hepatocytes. Toxicol. Appl. Pharmacol., 140:404–410. Haussinger, D., F. Lang, and W. Gerok (1994) Regulation of cell function by the cellular hydration state. Am. J. Physiol., 267:E343–E355. Hinz, H.R., and T.L. Kirley (1990) Lysine 480 is an essential residue in the putative ATP site of lamb kidney (Na,K)ATPase. J. Biol. Chem., 265:10260–10265. Jackson, P.S., and K. Strange (1993) Volume-sensitive anion channels mediate swelling-activated inositol and taurine efflux. Am. J. Physiol., 265:C1489–C1500. Jackson, P.S., R. Morrison, and K. Strange (1994) The volume-sensitive organic osmolyte-anion channel VSOAC is regulated by nonhydrolytic ATP binding. Am. J. Physiol., 267:C1203–C1209. Jackson, P.S., K. Churchwell, N. Ballatori, J.L. Boyer, and K. Strange (1996) Swelling-activated anion conductance in skate hepatocytes: Regulation by cell chloride and ATP. Am. J. Physiol., 270:C57–C66. Kinter, W.B., and J.B. Pritchard (1977) Altered permeability of cell membranes. In: Handbook of Physiology — Reactions to Environmental Agents, D.H.K. Lee, ed. Amer. Physiol. Soc., Baltimore, MD. Section 9, Chap. 36, pp 563–576. Tamura, J.K., R.D. Rakov, and R.L. Cross (1986) Affinity labeling of nucleotide-binding sites on kinases and dehydrogenases by pyridoxal 5'-diphospho-5'-adenosine. J. Biol. Chem., 261:4126–4133.