ARCHIVES
OF BIOCHEMISTRY
Vol. 282, No. 2, November
AND
BIOPHYSICS
1, pp. 35%362,199O
Relationships between the Mitochondrial Transmembrane Potential, ATP Concentration, and Cytotoxicity in Isolated Rat Hepatocytes Ellen Y. WU,*~ Martyn
T. Smith,* Giorgio Bellomo,*
and Donato Di Montet”
*Department of Biomedical and Environmental Health Sciences, School of Public Health, University Berkeley, California 94720; and tcalifornia Parkinson’s Foundation, San Jose, California 95128
of California,
Received March 26,1990, and in revised form June 14,199O
The relationships between mitochondrial transmembrane potential, ATP concentration, and cytotoxicity were evaluated after exposure of isolated rat hepatocytes to different mitochondrial poisons. Both the neurotoxicant 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and its fully oxidized metabolite, the I-methyl-4-phenylpyridinium (MPP+) ion, caused a concentrationand time-dependent depolarization of mitochondrial membranes which followed ATP depletion and preceded cytotoxicity. The effect of MPTP, but not that of MPP+, was prevented by deprenyl, an inhibitor of MPTP conversion to MPP+ via monoamine oxidase type B. Addition of fructose to the hepatocyte incubations treated with either MPTP or MPP+ counteracted the loss of mitochondrial transmembrane potential. Fructose was also effective in protecting against the mitochondrial membrane depolarization as well as ATP depletion and cytotoxicity induced by antimycin A, carbonyl cyanide p-trifluoromethoxyphenyl hydrazone, and valinomycin. Data confirm the key role played by MPP+-induced mitochondrial damage in MPTP toxicity and indicate that (i) ATP produced via the glycolytic pathway can be utilized by hepatocytes to maintain mitochondrial electrochemical gradient, and (ii) a loss of mitochondrial membrane potential may occur only when supplies of ATP are depleted. CC1990 Academic
Press,
Inc.
Mitochondria are vulnerable targets for toxic injury by a variety of compounds because of their crucial role in maintaining cellular structure and function via aerobic 1 To whom correspondence should be addressed at California Parkinson’s Foundation, 2444 Moorpark Avenue, Suite 316, San Jose, CA 95128.
ATP production. Recent studies indicate that the cytotoxic action of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP),’ a neurotoxicant which selectively destroys the nigrostriatal system and thus produces a parkinsonian syndrome in humans and monkeys (l-4), may arise from an impairment of mitochondrial respiration. Indeed, MPTP toxicity is ultimately dependent upon its conversion to the 1-methyl-4-phenylpyridinium (MPP’) species (5, 6), which has been shown to inhibit NADH-linked substrate oxidation in brain and liver mitochondria (7,8). Different mechanisms characterize the action of mitochondrial poisons. Compounds such as rotenone, MPP+, and antimycin act by blocking the electron flow coupled to proton pumping and to the generation of the electrochemical gradient. This gradient, which is comprised by 80-90% of an electrical component (transmembrane potential) and by lo-20% of a chemical component (pH gradient), plays an essential role in the production of ATP. Other compounds, such as protonophores (2,4dinitrophenol or carbonyl cyanide p-trifluoromethoxyphenyl hydrazone), act as membrane transporters for H+, thus uncoupling the proton movement across the inner mitochondrial membrane from ATP synthesis. A loss of mitochondrial transmembrane potential and depletion of ATP would result from the toxic effects of all these agents. Studies on the relationship between these biochemical events and their role in triggering cytotoxicity are essential to clarify the mechanism of action of compounds such as MPTP and design strategies to counteract their damaging effects. Recent experiments showing a clear ’ Abbreviations used: MPTP, I-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; MPP+, l-methyl-4-phenylpyridinium; FCCP, p-trifluoromethoxyphenyl hydrazone; DMSO, dimethyl sulfoxide; MAO B, monoamine oxidase type B.
358 All
0003.9861/90 $3.00 Copyright 8 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
FRUCTOSE
PREVENTS
MITOCHONDRIAL
correlation between ATP depletion and cytotoxicity in isolated hepatocytes exposed to MPTP (9) have, for example, led to the successful use of fructose as a protective agent in this experimental system (10). On the other hand, the relationship between intracellular levels of ATP and mitochondrial transmembrane potential remains to be explored in more detail. The effectiveness of glycolytically produced ATP in maintaining the cellular membrane potential has been demonstrated in bacteria (11); although a similar mechanism may be involved in maintaining the mitochondrial electrical proton gradient in mammalian cells, this has yet to be demonstrated (12). In the present work, the effects of MPTP and its toxic metabolite MPP+ on mitochondrial transmembrane potential and intracellular levels of ATP were assessed and correlated with cytotoxicity. Studies on the relationship between these biochemical and toxic events were then extended using other mitochondrial poisons. Finally, the ability of fructose to prevent the decrease in mitochondrial transmembrane potential as well as cytotoxicity induced by all these compounds was investigated. Results emphasize the importance of maintaining ATP levels in counteracting a loss of mitochondrial electrochemical gradient and protecting against cell death. MATERIALS
AND
METHODS
Hepatocytes were isolated from male Sprague-Dawley rats (250300 g) by collagenase (Boehringer, Mannheim, West Germany) perfusion ofthe liver and incubated (lO’cells/ml) at 37°C in Krebs-Henseleit buffer (pH 7.4) under a 95% 0,/5% CO, atmosphere (13). Cell viability was assessed by trypan blue exclusion and was always greater than 90% at the beginning of the experiments. Isolated hepatocytes were treated with MPTP (Research Biochemicals, Inc., Natick, MA), MPP+ (Research Biochemicals), antimycin A (Sigma Chemical Co., St. Louis, MO), valinomycin (Aldrich Chemical Co., Milwaukee, WI), or carbonyl cyanide p-trifluoromethoxyphenyl hydrazone (FCCP, Aldrich). Antimycin A, valinomycin, and FCCP were dissolved in dimethyl sulfoxide (DMSO) and added to the incubation at a final v/v of 0.5%. In experiments where the effects of deprenyl (Research Biochemicals) or fructose were assessed, deprenyl (10 fiM) was preincubated for 10 min before the addition of MPTP, and fructose was added at the beginning of the incubation period (10 mM) and after 60 min (5 mM). ATP was extracted from aliquots of the incubation mixtures by the addition of 0.35 N perchloric acid containing 22 mM EDTA. After centrifugation, supernatants were neutralized with 2.0 N KOH and ATP was assayed by chemiluminescence using the enzyme-substrate system luciferin-luciferase (14) with a Turner luminometer (Model TD-20e). Mitochondrial transmembrane potential was measured as the accumulation of rhodamine 123 (Rho 123, Aldrich), a cationic fluorescent probe which distributes across mitochondrial membranes with respect to the transmembrane potential (15, 16). Aliquots of the incubation mixtures were taken at different time points; hepatocytes were rapidly separated from the incubation medium and resuspended (lO”cells/ml) in Krebs-Henseleit buffer (pH 7.4) containing Rho 123 (1.5 gM). After incubation for 10 min at 37”C, cells were pelleted and Rho 123 accumulation determined by subtracting the fluorescence of the supernatant from the fluorescence of the initial 1.5 pM Rho 123 solution. Preliminary studies demonstrated that Rho 123 concentration reaches equi-
DEPOLARIZATION
AND
CYTOTOXICITY
359
A
FIG. 1. Time- and concentration-dependent effects of MPTP (A) and MPP+ (B) on the mitochondrial transmembrane potential of isolated hepatocytes. Cells were incubated with no addition (0) or in the presence of MPTP (0.5 mM, n ; 1 mM, 0) or MPP’ (1 mM, A; 2 mM, A). Samples were taken at the indicated time points until levels of cytotoxicity exceeded 40% and assayed for accumulation of Rho 123 as described under Materials and Methods. Average of initial level of Rho 123 accumulation was 0.59 ? 0.12 nmol/million cells. Values represent the means (*SD) of three separate experiments.
librium intra- and extracellularly after incubation for 10 min. Nonspecific binding of Rho 123 was determined for each experiment by incubating dead cells in the presence of the fluorescent cation and was subtracted from the accumulation values. Fluorescence readings were taken on a Turner spectrophotometer (Model 430) with the excitation wavelength at 498 nm and the emission wavelength at 525 nm.
RESULTS
Incubation of isolated rat hepatocytes in the presence of either MPTP or MPP+ resulted in a marked decrease in mitochondrial transmembrane potential (Fig. 1). This decrease was time- and concentration-dependent and, at equimolar concentrations of the two compounds, occurred more rapidly after exposure to MPTP. Pretreatment of hepatocytes with 10 PM deprenyl, a specific inhibitor of monoamine oxidase type B (MAO B), was ineffective against MPP+, but significantly antagonized the MPTP-induced decrease in Rho 123 accumulation (Fig. 2). The temporal relationship between this loss of mitochondrial transmembrane potential, the concentration of ATP, and cytotoxicity was evaluated next. MPTP (0.5 mM) caused a rapid depletion of ATP levels which were less than 10% of the initial value after incubation for 80 min (Fig. 3A). At this time point, accumulation of Rho 123 was decreased by more than 60%; on the other hand, a significant increase in cytotoxicity was observed only after incubation for 120 min. A similar sequence of events occurred after exposure of hepatocytes to 1 mM MPP+, although the time course was significantly less rapid (Fig. 3B). Thus, both MPTP and MPPt caused depletion of ATP and a decrease in mitochondrial transmembrane potential before the onset of cell death.
360
WU ET AL.
FIG. 2. Effect of deprenyl on the decrease in accumulation of Rho 123 induced by MPTP or MPP+. Hepatocytes were preincubated for 10 min in the absence (open symbols) or presence (filled symbols) of 10 1M deprenyl. Experiments were then started with the addition of 0.5 mM MPTP (squares) or 1 mM MPP+ (triangles) or no addition (circles). Values represent the means (*SD) of three separate experiments.
The capability of ATP to counteract both the loss of mitochondrial transmembrane potential and cytotoxicity was assessed by adding fructose (10 mM at 0 time and 5 mM at 60 min) to incubations of hepatocytes exposed to MPTP or MPP+. Comparison of Fig. 3 with Fig. 4 revealed that, after an initial rapid drop of ATP levels in the presence of fructose, ATP depletion caused by MPTP and MPP+ was prevented by fructose-mediated activation of glycolysis. Furthermore, fructose was effective in protecting against the loss of Rho 123 accumulation. With 0.5 mM MPTP, Rho 123 accumulation was decreased by 80% at the 120-min time point (Fig. 3A); in the presence of fructose. this accumulation was
B
FIG. 3. Relationship between ATP levels, accumulation of Rho 123, and cytotoxicity in hepatocytes treated with MPTP (A) or MPP’ (B). Cells were incubated with 0.5 mM MPTP (0) or 1 mM MPP+ (A). Samples were taken at the indicated time points and assayed for ATP levels (filled symbols), accumulation of Rho 123 (open symbols), and cytotoxicity (dotted line). Averages of initial values for Rho 123 accumulation and ATP levels were 0.63 + 0.11 nmol/million cells and 12.0 f 2.1 nmol/million cells, respectively. Values represent the means (*SD) of three separate experiments.
FIG. 4. Effect of fructose on ATP depletion, loss of accumulation of Rho 123, and cytotoxicity caused by MPTP (A) and MPP’ (B). Fructose (10 mM at 0 min and 5 mM at 60 min) was supplemented to the incubation medium of hepatocytes treated with 0.5 mM MPTP (0) or 1 mM MPP+ (a). Samples were taken at the indicated time points and assayed for ATP levels (filled symbols), accumulation of Rho 123 (open symbols), and cytotoxicity (dotted line). Values represent the means (*SD) of three separate experiments.
50% of the initial value after incubation for 180 min (Fig. 4A). No increase in cell death was observed in MPTPand MPP+-treated hepatocyte incubations as compared to untreated cells in the presence of fructose (Fig. 4). The relationship between the loss of mitochondrial transmembrane potential, ATP depletion, and cytotoxicity was further examined by exposing isolated hepatocytes to the following mitochondrial poisons: antimycin A, an inhibitor of electron transport, FCCP, an uncoupler, and valinomycin, a potassium ionophore. At their respective concentrations, all these compounds produced a dramatic ATP depletion and a decrease in Rho 123 accumulation; cytotoxicity consistently followed these events (Table I). For example, ATP levels and Rho 123 accumulation were decreased to 5% and 38% of the initial values, respectively, after 60 min of incubation in the presence of 2 yM FCCP. At this time point, an increase in cell death in FCCP-treated hepatocytes as compared to untreated cells began to be observed. Addition of fructose to preparations of hepatocytes exposed to antimycin A, FCCP, or valinomycin had effects similar to those seen with MPTP or MPP’: it counteracted both ATP depletion and the loss of Rho 123 accumulation and protected against cytotoxicity (Table I). Control incubations with DMSO (final v/v, 0.5%), the solvent used with antimycin A, FCCP, and valinomycin ruled out any effect of this compound on ATP levels and mitochondrial transmembrane potential of hepatocytes. DISCUSSION
The relationship between ATP concentration, mitochondrial transmembrane potential, and cytotoxicity was studied by adding MPTP or its fully oxidized metabolite MPP+ to isolated rat hepatocytes. MPTP causes a
FRUCTOSE
PREVENTS
MITOCHONDRIAL
DEPOLARIZATION TABLE
AND
361
CYTOTOXICITY
I
Effect of Fructose on the Decrease in Rhodamine 123 Accumulation, ATP Depletion, and Cytotoxicity Induced by Antimycin A, FCCP, and Valinomycin Rho 123 accumulation (% of initial value)
Treatment Control t fructose Antimycin (20 pM) + fructose FCCP (2 /AM) + fructose Valinomycin (5 nM) + fructose
60 min 106.3 107.1 35.7 75.8 38.1 69.3 71.6 95.8
k + 2 + f 2 + -t
13.6 15.7 7.7 10.4 4.8 8.1 10.3 10.6
120 min 92.1 116.0 26.3 89.6 18.9 85.2 28.4 98.1
f lir + + + + t k
Toxicity (Ul, cell death)
ATP concentration (% of initial value)
10.8 16.2 5.8 8.9 5.2 11.2 12.4 6.6
60 min 93.1 43.3 3.5 73.2 4.6 25.7 10.8 21.9
* 10.1 III 10.6 * 2.0 + 11.7 k 1.5 f 8.8 f 2.3 f 7.1
120 min 90.1 65.1 2.2 54.1 2.7 17.0 4.6 28.1
* 11.2 k 9.8 + 1.7 k 12.5 + 1.1 -c 5.8 k 2.0 2 8.2
60 min
120 min
11 f4 15 -t 3 32 f 7 9t5 24 f 6 9t-6 14 + 5 18f4
14 f 3 12 f 2 89 + 4 18 f 7 89 f 3 14+8 71&8 11+5
(5 nM) in the absence or presence of fructose (10 mM Note. Hepatocytes were exposed to antimycin A (20 PM), FCCP (2 PM), or valinomycin at 0 min and 5 mM at 60 min). Averages of initial values for Rho 123 accumulation and ATP concentration were 0.62 k 0.09 and 11.3 + 1.8 nmol/million cells, respectively. Data represent the means (&SD) of three separate experiments.
parkinsonian syndrome in humans virtually indistinguishable from idiopathic Parkinson’s disease (2) and may represent a striking example of the importance of mitochondria as possible targets for the action of toxic compounds. The deleterious effects of MPTP depend upon its metabolic activation to MPP+ (5, 6), mediated via mitochondrial MAO B (17). MPP+ is likely to act by blocking mitochondrial electron flow at the level of Complex I (7) and, thus, inhibiting proton motive force (18) and synthesis of ATP (9). The crucial role played by this impairment of mitochondrial function in MPTP/ MPP’-induced cytotoxicity is suggested by the fact that, in this study, both MPTP and MPPf caused a loss of mitochondrial transmembrane potential before killing isolated hepatocytes. At equimolar concentrations of the two compounds, MPTP was markedly more effective than MPP+ in decreasing Rho 123 accumulation, due to the lipophilic chemical structure of MPTP, which facilitates its access into the cell (19). Furthermore, the effect of MPTP, but not that of MPP+, was prevented by the MAO B inhibitor deprenyl, confirming the role of the intracellular generation of MPP+ in mediating MPTP-induced impairment of mitochondrial function. Protection by deprenyl was not complete, probably because MAO B inhibition does not completely block MPPi production in isolated hepatocytes (20). The loss of mitochondrial transmembrane potential caused by MPTP and MPP+ was clearly related to ATP depletion and cytotoxicity; indeed, it followed the decline of ATP levels and preceded the onset of cell death. Furthermore, addition of fructose to the incubation mixtures counteracted the decrease in Rho 123 accumulation as well as ATP depletion and cytotoxicity induced by MPTP and MPP+. Similar results were obtained using other compounds able to disrupt mitochondrial elec-
trochemical gradient and ATP synthesis. Antimycin A, FCCP, and valinomycin all caused ATP depletion, a decrease in Rho 123 accumulation, and hepatocyte death, which were prevented by activation of glycolytic ATP production by fructose. Taken together, these findings support the conclusion that a loss of mitochondrial transmembrane potential occurs only when supplies of ATP are depleted, since maintenance of adequate concentrations of ATP appears to be very effective in preventing the disruption of the mitochondrial electrochemical gradient. On the other hand, it is important to note that ATP depletion does not necessarily result in a decrease in mitochondrial transmembrane potential; this has been clearly shown using oligomycin, a specific inhibitor of mitochondrial ATPase (2 1). Results of this study also reveal that ATP produced via the glycolytic pathway can be utilized by hepatocytes to maintain the mitochondrial transmembrane potential. Previous work on the effect of anoxia on the mitochondrial electrochemical gradient of isolated hepatocytes has led to the conclusion that retention of the protonmotive force in this experimental condition is not due to energy provided by glycolysis (12). Thus, various mechanisms may play different roles in protecting against the loss of mitochondrial transmembrane potential during hypoxic and toxic injury. Whether disruption of ATP generation or mitochondrial membrane depolarization is more directly correlated with cell death is quite controversial, since evidence supporting the former (22) or the latter (21) hypothesis has been put forward. In all the experimental conditions used in this study, a decrease in Rho 123 accumulation mirrored the decline in ATP levels, suggesting that both these events may be responsible for hepatocyte death. However, as discussed above, the loss of mitochondrial transmembrane potential seems to occur
362
WU ET AL.
only with consumption of ATP supplies; in this respect, the latter biochemical change may be regarded as a critical step in the events leading to cytotoxicity. The importance of ATP depletion in the toxic events assessed in this study is demonstrated by the experiments with fructose. This glycolytic substrate has already been shown to increase the rate of lactate formation when supplemented to hepatocytes exposed to MPTP or MPP+ (10). In the present study, fructose counteracted both the loss of mitochondrial transmembrane potential and the cytotoxicity caused by all compounds tested. However, its protective effects seemed to be dependent upon at least two experimental variables. First, the concentration of fructose added to the incubation mixtures should not be greater than lo-20 mM since the initial phosphorylation of this carbohydrate itself causes a sharp drop in ATP levels. For this reason, we found it most effective to treat hepatocytes with two sequential additions of fructose, at the beginning of the incubations (10 mM) and after 1 h (5 mM). Second, it is important to consider that the protective effects of fructose become evident when the rate of production of ATP via glycolysis overcomes the rate of ATP consumption, which may vary in different experimental conditions. For example, with FCCP, the rate of ATP consumption may be particularly high as a consequence of stimulation of ATP hydrolysis by mitochondrial ATPase. Thus, fructose was found in this study to prevent ATP depletion, mitochondrial transmembrane potential, and cytotoxicity only when relatively low concentrations of FCCP were used. Similarly, with the uncoupler CCCP, the protective effect of fructose against toxicity has been reported to be augmented in the presence of oligomycin, which decreases consumption of ATP by blocking mitochondrial ATPase (22). The ability of fructose to counteract cytotoxicity might be attributed, at least in part, to the generation of lactic acid and consequent decrease in intracellular and extracellular pH. Indeed, acidosis has been found to protect against toxic injury to liver cells (23, 24). However, recent studies have demonstrated that the mechanisms of protection by fructose and acidosis are different and independent, since, for example, acidosis does not counteract ATP depletion (24) and fructose prevents cytotoxicity even after alkalinization of the cytosol by monensin (22). Therefore, the protective effects of fructose against cell death and the loss of mitochondrial transmembrane potential in this study can be attributed solely to its ability to replenish supplies of ATP. This reemphasizes the key role played by ATP depletion in the cytotoxicity of MPTP and in the biochemical and toxic events induced by mitochondrial poisons. ACKNOWLEDGMENTS This work was supported by the Health Effects Component of the University of California Toxic Substances Program and the California
Parkinson’s Foundation. Dr. Di Monte was a recipient of the Lillian Schorr Research Fellowship from the Parkinson’s Disease Foundation, New York.
REFERENCES 1. Davis, G. C., William, A. C., Markey, S. P., Ebert, M. H., Caine, E. D., Reichert, C. M., and Kopin, I. J. (1979) Psychiatry Res. 1,
949-954. 2. Langston, J. W., Ballard, P., Tetrud, J. W., and Irwin, I. (1983) Science 219,979-980. 3. Burns, R. S., Chiueh, C. C., Markey, S. P., Ebert, M. H., Jacobowitz, D. M., and Kopin, I. J. (1983) Proc. N&l. Acad. Sci. USA 80,
4546-4550. 4. Langston, J. W., Forno, L. S., Rebert, C. S., and Irwin, I. (1984) Brain Res. 292,390-394. 5. Langston, J. W., Irwin, I., Langston, E. B., Neurosci. Lett. 48, 87-92. 6. Markey, S. P., Johannessen, J. N., Chiueh, Herkenham, M. A. (1984) Nature (London) 7. Nicklas, W. J., Vyas, I., and Heikkila, R.
and Forno, L. S. (1984) C. C., Burns, R. S., and
311,464-467. E. (1985) Life Sci. 36,
2503-2508. 8. Ramsay, R. R., Salach, J. I., Dadgar, J., and Singer, T. P. (1986) Biochem. Biophys. Res. Commun. 135,268-275. 9. Di Monte, D., Jewell, S. A., Ekstrom, G., Sandy, M. S., and Smith, M. T. (1986) Biochem. Biophys. Res. Commun. 137,310-315. 10. Di Monte, D., Sandy, M. S., Blank, L., and Smith, M. T. (1988) Biochem. Biophys. Res. Commun. 153,734-740. 11. Harold, F. M. (1977) Curr. Top. Bioenerg. 6,83-149. 12. Andersson, B. S., Aw, T. Y., and Jones, D. P. (1987) Amer. J. Physiol. 252, C349-C355. 13. Moldeus, P., Hogberg, J., and Orrenius, S. (1978) in Methods in Enzymology (Fleischer, S., and Packer, L., Eds.), Vol. 52, pp. 6071, Academic Press, San Diego. 14. Lemasters, J. J., and Hackenbrock, C. R. (1979) in Methods in Enzymology (Fleischer, S., and Packer, L., Eds.), Vol. 56, pp. 530544, Academic Press, San Diego. 15. Emaus, R. K., Grunwald, R., and Lemasters, J. J. (1986) Biochim. Biophys. Acta 850,436-448. 16. Johnson, L. V., Walsh, M. L., Bockus, B. J., and Chen, L. B. (1981) J. Cell Biol. 88,526-535. 17. Chiba, K., Trevor, A., and Castagnoli, N., Jr. (1984) Biochem. Biophys. Res. Commun. 128,1228-1232. 18. Singh, Y., Bhatnagar, R., Sidhu, G. S., Batra, J. K., and Krishna, G. (1989) Arch. Biochem. Biophys. 271,217-222. 19. Di Monte, D., Ekstrom, G., Shinka, T., Smith, M. T., Trevor, A. J., and Castagnoli, N., Jr. (1987) Chem:Biol. Interact. 62,105116. 20. Di Monte, D., Shinka, T., Sandy, M. S., Castagnoli, N., Jr., and Smith, M. T. (1988) Drug Metab. Dispos. 16,250-255. 21. Masaki, N., Thomas, A. P., Hoek, J. B., and Farber, J. L. (1989) Arch. Biochem. Biophys. 272,152-161. 22. Nieminen, A., Dawson, T. L., Gores, G. J., Kawanishi, T., Herman, B., and Lemasters, J. J. (1990) Biochem. Biophys. Rex Commun., 167,600p606. 23. Gores, G. J., Nieminen, A., Wray, B. E., Herman, B., and Lemasters, J. J. (1989) J. Clin. Znuest. 83,386-396. 24. Gores, G. J., Nieminen, A., Fleishman, K. E., Dawson, T. L., Herman, B., and Lemasters, J. J. (1988) Amer. J. Physiol. 255, C315-
C322.
OF BIOCHEMISTRY
Vol. 282, No. 2, November
AND
BIOPHYSICS
1, pp. 35%362,199O
Relationships between the Mitochondrial Transmembrane Potential, ATP Concentration, and Cytotoxicity in Isolated Rat Hepatocytes Ellen Y. WU,*~ Martyn
T. Smith,* Giorgio Bellomo,*
and Donato Di Montet”
*Department of Biomedical and Environmental Health Sciences, School of Public Health, University Berkeley, California 94720; and tcalifornia Parkinson’s Foundation, San Jose, California 95128
of California,
Received March 26,1990, and in revised form June 14,199O
The relationships between mitochondrial transmembrane potential, ATP concentration, and cytotoxicity were evaluated after exposure of isolated rat hepatocytes to different mitochondrial poisons. Both the neurotoxicant 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and its fully oxidized metabolite, the I-methyl-4-phenylpyridinium (MPP+) ion, caused a concentrationand time-dependent depolarization of mitochondrial membranes which followed ATP depletion and preceded cytotoxicity. The effect of MPTP, but not that of MPP+, was prevented by deprenyl, an inhibitor of MPTP conversion to MPP+ via monoamine oxidase type B. Addition of fructose to the hepatocyte incubations treated with either MPTP or MPP+ counteracted the loss of mitochondrial transmembrane potential. Fructose was also effective in protecting against the mitochondrial membrane depolarization as well as ATP depletion and cytotoxicity induced by antimycin A, carbonyl cyanide p-trifluoromethoxyphenyl hydrazone, and valinomycin. Data confirm the key role played by MPP+-induced mitochondrial damage in MPTP toxicity and indicate that (i) ATP produced via the glycolytic pathway can be utilized by hepatocytes to maintain mitochondrial electrochemical gradient, and (ii) a loss of mitochondrial membrane potential may occur only when supplies of ATP are depleted. CC1990 Academic
Press,
Inc.
Mitochondria are vulnerable targets for toxic injury by a variety of compounds because of their crucial role in maintaining cellular structure and function via aerobic 1 To whom correspondence should be addressed at California Parkinson’s Foundation, 2444 Moorpark Avenue, Suite 316, San Jose, CA 95128.
ATP production. Recent studies indicate that the cytotoxic action of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP),’ a neurotoxicant which selectively destroys the nigrostriatal system and thus produces a parkinsonian syndrome in humans and monkeys (l-4), may arise from an impairment of mitochondrial respiration. Indeed, MPTP toxicity is ultimately dependent upon its conversion to the 1-methyl-4-phenylpyridinium (MPP’) species (5, 6), which has been shown to inhibit NADH-linked substrate oxidation in brain and liver mitochondria (7,8). Different mechanisms characterize the action of mitochondrial poisons. Compounds such as rotenone, MPP+, and antimycin act by blocking the electron flow coupled to proton pumping and to the generation of the electrochemical gradient. This gradient, which is comprised by 80-90% of an electrical component (transmembrane potential) and by lo-20% of a chemical component (pH gradient), plays an essential role in the production of ATP. Other compounds, such as protonophores (2,4dinitrophenol or carbonyl cyanide p-trifluoromethoxyphenyl hydrazone), act as membrane transporters for H+, thus uncoupling the proton movement across the inner mitochondrial membrane from ATP synthesis. A loss of mitochondrial transmembrane potential and depletion of ATP would result from the toxic effects of all these agents. Studies on the relationship between these biochemical events and their role in triggering cytotoxicity are essential to clarify the mechanism of action of compounds such as MPTP and design strategies to counteract their damaging effects. Recent experiments showing a clear ’ Abbreviations used: MPTP, I-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; MPP+, l-methyl-4-phenylpyridinium; FCCP, p-trifluoromethoxyphenyl hydrazone; DMSO, dimethyl sulfoxide; MAO B, monoamine oxidase type B.
358 All
0003.9861/90 $3.00 Copyright 8 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
FRUCTOSE
PREVENTS
MITOCHONDRIAL
correlation between ATP depletion and cytotoxicity in isolated hepatocytes exposed to MPTP (9) have, for example, led to the successful use of fructose as a protective agent in this experimental system (10). On the other hand, the relationship between intracellular levels of ATP and mitochondrial transmembrane potential remains to be explored in more detail. The effectiveness of glycolytically produced ATP in maintaining the cellular membrane potential has been demonstrated in bacteria (11); although a similar mechanism may be involved in maintaining the mitochondrial electrical proton gradient in mammalian cells, this has yet to be demonstrated (12). In the present work, the effects of MPTP and its toxic metabolite MPP+ on mitochondrial transmembrane potential and intracellular levels of ATP were assessed and correlated with cytotoxicity. Studies on the relationship between these biochemical and toxic events were then extended using other mitochondrial poisons. Finally, the ability of fructose to prevent the decrease in mitochondrial transmembrane potential as well as cytotoxicity induced by all these compounds was investigated. Results emphasize the importance of maintaining ATP levels in counteracting a loss of mitochondrial electrochemical gradient and protecting against cell death. MATERIALS
AND
METHODS
Hepatocytes were isolated from male Sprague-Dawley rats (250300 g) by collagenase (Boehringer, Mannheim, West Germany) perfusion ofthe liver and incubated (lO’cells/ml) at 37°C in Krebs-Henseleit buffer (pH 7.4) under a 95% 0,/5% CO, atmosphere (13). Cell viability was assessed by trypan blue exclusion and was always greater than 90% at the beginning of the experiments. Isolated hepatocytes were treated with MPTP (Research Biochemicals, Inc., Natick, MA), MPP+ (Research Biochemicals), antimycin A (Sigma Chemical Co., St. Louis, MO), valinomycin (Aldrich Chemical Co., Milwaukee, WI), or carbonyl cyanide p-trifluoromethoxyphenyl hydrazone (FCCP, Aldrich). Antimycin A, valinomycin, and FCCP were dissolved in dimethyl sulfoxide (DMSO) and added to the incubation at a final v/v of 0.5%. In experiments where the effects of deprenyl (Research Biochemicals) or fructose were assessed, deprenyl (10 fiM) was preincubated for 10 min before the addition of MPTP, and fructose was added at the beginning of the incubation period (10 mM) and after 60 min (5 mM). ATP was extracted from aliquots of the incubation mixtures by the addition of 0.35 N perchloric acid containing 22 mM EDTA. After centrifugation, supernatants were neutralized with 2.0 N KOH and ATP was assayed by chemiluminescence using the enzyme-substrate system luciferin-luciferase (14) with a Turner luminometer (Model TD-20e). Mitochondrial transmembrane potential was measured as the accumulation of rhodamine 123 (Rho 123, Aldrich), a cationic fluorescent probe which distributes across mitochondrial membranes with respect to the transmembrane potential (15, 16). Aliquots of the incubation mixtures were taken at different time points; hepatocytes were rapidly separated from the incubation medium and resuspended (lO”cells/ml) in Krebs-Henseleit buffer (pH 7.4) containing Rho 123 (1.5 gM). After incubation for 10 min at 37”C, cells were pelleted and Rho 123 accumulation determined by subtracting the fluorescence of the supernatant from the fluorescence of the initial 1.5 pM Rho 123 solution. Preliminary studies demonstrated that Rho 123 concentration reaches equi-
DEPOLARIZATION
AND
CYTOTOXICITY
359
A
FIG. 1. Time- and concentration-dependent effects of MPTP (A) and MPP+ (B) on the mitochondrial transmembrane potential of isolated hepatocytes. Cells were incubated with no addition (0) or in the presence of MPTP (0.5 mM, n ; 1 mM, 0) or MPP’ (1 mM, A; 2 mM, A). Samples were taken at the indicated time points until levels of cytotoxicity exceeded 40% and assayed for accumulation of Rho 123 as described under Materials and Methods. Average of initial level of Rho 123 accumulation was 0.59 ? 0.12 nmol/million cells. Values represent the means (*SD) of three separate experiments.
librium intra- and extracellularly after incubation for 10 min. Nonspecific binding of Rho 123 was determined for each experiment by incubating dead cells in the presence of the fluorescent cation and was subtracted from the accumulation values. Fluorescence readings were taken on a Turner spectrophotometer (Model 430) with the excitation wavelength at 498 nm and the emission wavelength at 525 nm.
RESULTS
Incubation of isolated rat hepatocytes in the presence of either MPTP or MPP+ resulted in a marked decrease in mitochondrial transmembrane potential (Fig. 1). This decrease was time- and concentration-dependent and, at equimolar concentrations of the two compounds, occurred more rapidly after exposure to MPTP. Pretreatment of hepatocytes with 10 PM deprenyl, a specific inhibitor of monoamine oxidase type B (MAO B), was ineffective against MPP+, but significantly antagonized the MPTP-induced decrease in Rho 123 accumulation (Fig. 2). The temporal relationship between this loss of mitochondrial transmembrane potential, the concentration of ATP, and cytotoxicity was evaluated next. MPTP (0.5 mM) caused a rapid depletion of ATP levels which were less than 10% of the initial value after incubation for 80 min (Fig. 3A). At this time point, accumulation of Rho 123 was decreased by more than 60%; on the other hand, a significant increase in cytotoxicity was observed only after incubation for 120 min. A similar sequence of events occurred after exposure of hepatocytes to 1 mM MPP+, although the time course was significantly less rapid (Fig. 3B). Thus, both MPTP and MPPt caused depletion of ATP and a decrease in mitochondrial transmembrane potential before the onset of cell death.
360
WU ET AL.
FIG. 2. Effect of deprenyl on the decrease in accumulation of Rho 123 induced by MPTP or MPP+. Hepatocytes were preincubated for 10 min in the absence (open symbols) or presence (filled symbols) of 10 1M deprenyl. Experiments were then started with the addition of 0.5 mM MPTP (squares) or 1 mM MPP+ (triangles) or no addition (circles). Values represent the means (*SD) of three separate experiments.
The capability of ATP to counteract both the loss of mitochondrial transmembrane potential and cytotoxicity was assessed by adding fructose (10 mM at 0 time and 5 mM at 60 min) to incubations of hepatocytes exposed to MPTP or MPP+. Comparison of Fig. 3 with Fig. 4 revealed that, after an initial rapid drop of ATP levels in the presence of fructose, ATP depletion caused by MPTP and MPP+ was prevented by fructose-mediated activation of glycolysis. Furthermore, fructose was effective in protecting against the loss of Rho 123 accumulation. With 0.5 mM MPTP, Rho 123 accumulation was decreased by 80% at the 120-min time point (Fig. 3A); in the presence of fructose. this accumulation was
B
FIG. 3. Relationship between ATP levels, accumulation of Rho 123, and cytotoxicity in hepatocytes treated with MPTP (A) or MPP’ (B). Cells were incubated with 0.5 mM MPTP (0) or 1 mM MPP+ (A). Samples were taken at the indicated time points and assayed for ATP levels (filled symbols), accumulation of Rho 123 (open symbols), and cytotoxicity (dotted line). Averages of initial values for Rho 123 accumulation and ATP levels were 0.63 + 0.11 nmol/million cells and 12.0 f 2.1 nmol/million cells, respectively. Values represent the means (*SD) of three separate experiments.
FIG. 4. Effect of fructose on ATP depletion, loss of accumulation of Rho 123, and cytotoxicity caused by MPTP (A) and MPP’ (B). Fructose (10 mM at 0 min and 5 mM at 60 min) was supplemented to the incubation medium of hepatocytes treated with 0.5 mM MPTP (0) or 1 mM MPP+ (a). Samples were taken at the indicated time points and assayed for ATP levels (filled symbols), accumulation of Rho 123 (open symbols), and cytotoxicity (dotted line). Values represent the means (*SD) of three separate experiments.
50% of the initial value after incubation for 180 min (Fig. 4A). No increase in cell death was observed in MPTPand MPP+-treated hepatocyte incubations as compared to untreated cells in the presence of fructose (Fig. 4). The relationship between the loss of mitochondrial transmembrane potential, ATP depletion, and cytotoxicity was further examined by exposing isolated hepatocytes to the following mitochondrial poisons: antimycin A, an inhibitor of electron transport, FCCP, an uncoupler, and valinomycin, a potassium ionophore. At their respective concentrations, all these compounds produced a dramatic ATP depletion and a decrease in Rho 123 accumulation; cytotoxicity consistently followed these events (Table I). For example, ATP levels and Rho 123 accumulation were decreased to 5% and 38% of the initial values, respectively, after 60 min of incubation in the presence of 2 yM FCCP. At this time point, an increase in cell death in FCCP-treated hepatocytes as compared to untreated cells began to be observed. Addition of fructose to preparations of hepatocytes exposed to antimycin A, FCCP, or valinomycin had effects similar to those seen with MPTP or MPP’: it counteracted both ATP depletion and the loss of Rho 123 accumulation and protected against cytotoxicity (Table I). Control incubations with DMSO (final v/v, 0.5%), the solvent used with antimycin A, FCCP, and valinomycin ruled out any effect of this compound on ATP levels and mitochondrial transmembrane potential of hepatocytes. DISCUSSION
The relationship between ATP concentration, mitochondrial transmembrane potential, and cytotoxicity was studied by adding MPTP or its fully oxidized metabolite MPP+ to isolated rat hepatocytes. MPTP causes a
FRUCTOSE
PREVENTS
MITOCHONDRIAL
DEPOLARIZATION TABLE
AND
361
CYTOTOXICITY
I
Effect of Fructose on the Decrease in Rhodamine 123 Accumulation, ATP Depletion, and Cytotoxicity Induced by Antimycin A, FCCP, and Valinomycin Rho 123 accumulation (% of initial value)
Treatment Control t fructose Antimycin (20 pM) + fructose FCCP (2 /AM) + fructose Valinomycin (5 nM) + fructose
60 min 106.3 107.1 35.7 75.8 38.1 69.3 71.6 95.8
k + 2 + f 2 + -t
13.6 15.7 7.7 10.4 4.8 8.1 10.3 10.6
120 min 92.1 116.0 26.3 89.6 18.9 85.2 28.4 98.1
f lir + + + + t k
Toxicity (Ul, cell death)
ATP concentration (% of initial value)
10.8 16.2 5.8 8.9 5.2 11.2 12.4 6.6
60 min 93.1 43.3 3.5 73.2 4.6 25.7 10.8 21.9
* 10.1 III 10.6 * 2.0 + 11.7 k 1.5 f 8.8 f 2.3 f 7.1
120 min 90.1 65.1 2.2 54.1 2.7 17.0 4.6 28.1
* 11.2 k 9.8 + 1.7 k 12.5 + 1.1 -c 5.8 k 2.0 2 8.2
60 min
120 min
11 f4 15 -t 3 32 f 7 9t5 24 f 6 9t-6 14 + 5 18f4
14 f 3 12 f 2 89 + 4 18 f 7 89 f 3 14+8 71&8 11+5
(5 nM) in the absence or presence of fructose (10 mM Note. Hepatocytes were exposed to antimycin A (20 PM), FCCP (2 PM), or valinomycin at 0 min and 5 mM at 60 min). Averages of initial values for Rho 123 accumulation and ATP concentration were 0.62 k 0.09 and 11.3 + 1.8 nmol/million cells, respectively. Data represent the means (&SD) of three separate experiments.
parkinsonian syndrome in humans virtually indistinguishable from idiopathic Parkinson’s disease (2) and may represent a striking example of the importance of mitochondria as possible targets for the action of toxic compounds. The deleterious effects of MPTP depend upon its metabolic activation to MPP+ (5, 6), mediated via mitochondrial MAO B (17). MPP+ is likely to act by blocking mitochondrial electron flow at the level of Complex I (7) and, thus, inhibiting proton motive force (18) and synthesis of ATP (9). The crucial role played by this impairment of mitochondrial function in MPTP/ MPP’-induced cytotoxicity is suggested by the fact that, in this study, both MPTP and MPPf caused a loss of mitochondrial transmembrane potential before killing isolated hepatocytes. At equimolar concentrations of the two compounds, MPTP was markedly more effective than MPP+ in decreasing Rho 123 accumulation, due to the lipophilic chemical structure of MPTP, which facilitates its access into the cell (19). Furthermore, the effect of MPTP, but not that of MPP+, was prevented by the MAO B inhibitor deprenyl, confirming the role of the intracellular generation of MPP+ in mediating MPTP-induced impairment of mitochondrial function. Protection by deprenyl was not complete, probably because MAO B inhibition does not completely block MPPi production in isolated hepatocytes (20). The loss of mitochondrial transmembrane potential caused by MPTP and MPP+ was clearly related to ATP depletion and cytotoxicity; indeed, it followed the decline of ATP levels and preceded the onset of cell death. Furthermore, addition of fructose to the incubation mixtures counteracted the decrease in Rho 123 accumulation as well as ATP depletion and cytotoxicity induced by MPTP and MPP+. Similar results were obtained using other compounds able to disrupt mitochondrial elec-
trochemical gradient and ATP synthesis. Antimycin A, FCCP, and valinomycin all caused ATP depletion, a decrease in Rho 123 accumulation, and hepatocyte death, which were prevented by activation of glycolytic ATP production by fructose. Taken together, these findings support the conclusion that a loss of mitochondrial transmembrane potential occurs only when supplies of ATP are depleted, since maintenance of adequate concentrations of ATP appears to be very effective in preventing the disruption of the mitochondrial electrochemical gradient. On the other hand, it is important to note that ATP depletion does not necessarily result in a decrease in mitochondrial transmembrane potential; this has been clearly shown using oligomycin, a specific inhibitor of mitochondrial ATPase (2 1). Results of this study also reveal that ATP produced via the glycolytic pathway can be utilized by hepatocytes to maintain the mitochondrial transmembrane potential. Previous work on the effect of anoxia on the mitochondrial electrochemical gradient of isolated hepatocytes has led to the conclusion that retention of the protonmotive force in this experimental condition is not due to energy provided by glycolysis (12). Thus, various mechanisms may play different roles in protecting against the loss of mitochondrial transmembrane potential during hypoxic and toxic injury. Whether disruption of ATP generation or mitochondrial membrane depolarization is more directly correlated with cell death is quite controversial, since evidence supporting the former (22) or the latter (21) hypothesis has been put forward. In all the experimental conditions used in this study, a decrease in Rho 123 accumulation mirrored the decline in ATP levels, suggesting that both these events may be responsible for hepatocyte death. However, as discussed above, the loss of mitochondrial transmembrane potential seems to occur
362
WU ET AL.
only with consumption of ATP supplies; in this respect, the latter biochemical change may be regarded as a critical step in the events leading to cytotoxicity. The importance of ATP depletion in the toxic events assessed in this study is demonstrated by the experiments with fructose. This glycolytic substrate has already been shown to increase the rate of lactate formation when supplemented to hepatocytes exposed to MPTP or MPP+ (10). In the present study, fructose counteracted both the loss of mitochondrial transmembrane potential and the cytotoxicity caused by all compounds tested. However, its protective effects seemed to be dependent upon at least two experimental variables. First, the concentration of fructose added to the incubation mixtures should not be greater than lo-20 mM since the initial phosphorylation of this carbohydrate itself causes a sharp drop in ATP levels. For this reason, we found it most effective to treat hepatocytes with two sequential additions of fructose, at the beginning of the incubations (10 mM) and after 1 h (5 mM). Second, it is important to consider that the protective effects of fructose become evident when the rate of production of ATP via glycolysis overcomes the rate of ATP consumption, which may vary in different experimental conditions. For example, with FCCP, the rate of ATP consumption may be particularly high as a consequence of stimulation of ATP hydrolysis by mitochondrial ATPase. Thus, fructose was found in this study to prevent ATP depletion, mitochondrial transmembrane potential, and cytotoxicity only when relatively low concentrations of FCCP were used. Similarly, with the uncoupler CCCP, the protective effect of fructose against toxicity has been reported to be augmented in the presence of oligomycin, which decreases consumption of ATP by blocking mitochondrial ATPase (22). The ability of fructose to counteract cytotoxicity might be attributed, at least in part, to the generation of lactic acid and consequent decrease in intracellular and extracellular pH. Indeed, acidosis has been found to protect against toxic injury to liver cells (23, 24). However, recent studies have demonstrated that the mechanisms of protection by fructose and acidosis are different and independent, since, for example, acidosis does not counteract ATP depletion (24) and fructose prevents cytotoxicity even after alkalinization of the cytosol by monensin (22). Therefore, the protective effects of fructose against cell death and the loss of mitochondrial transmembrane potential in this study can be attributed solely to its ability to replenish supplies of ATP. This reemphasizes the key role played by ATP depletion in the cytotoxicity of MPTP and in the biochemical and toxic events induced by mitochondrial poisons. ACKNOWLEDGMENTS This work was supported by the Health Effects Component of the University of California Toxic Substances Program and the California
Parkinson’s Foundation. Dr. Di Monte was a recipient of the Lillian Schorr Research Fellowship from the Parkinson’s Disease Foundation, New York.
REFERENCES 1. Davis, G. C., William, A. C., Markey, S. P., Ebert, M. H., Caine, E. D., Reichert, C. M., and Kopin, I. J. (1979) Psychiatry Res. 1,
949-954. 2. Langston, J. W., Ballard, P., Tetrud, J. W., and Irwin, I. (1983) Science 219,979-980. 3. Burns, R. S., Chiueh, C. C., Markey, S. P., Ebert, M. H., Jacobowitz, D. M., and Kopin, I. J. (1983) Proc. N&l. Acad. Sci. USA 80,
4546-4550. 4. Langston, J. W., Forno, L. S., Rebert, C. S., and Irwin, I. (1984) Brain Res. 292,390-394. 5. Langston, J. W., Irwin, I., Langston, E. B., Neurosci. Lett. 48, 87-92. 6. Markey, S. P., Johannessen, J. N., Chiueh, Herkenham, M. A. (1984) Nature (London) 7. Nicklas, W. J., Vyas, I., and Heikkila, R.
and Forno, L. S. (1984) C. C., Burns, R. S., and
311,464-467. E. (1985) Life Sci. 36,
2503-2508. 8. Ramsay, R. R., Salach, J. I., Dadgar, J., and Singer, T. P. (1986) Biochem. Biophys. Res. Commun. 135,268-275. 9. Di Monte, D., Jewell, S. A., Ekstrom, G., Sandy, M. S., and Smith, M. T. (1986) Biochem. Biophys. Res. Commun. 137,310-315. 10. Di Monte, D., Sandy, M. S., Blank, L., and Smith, M. T. (1988) Biochem. Biophys. Res. Commun. 153,734-740. 11. Harold, F. M. (1977) Curr. Top. Bioenerg. 6,83-149. 12. Andersson, B. S., Aw, T. Y., and Jones, D. P. (1987) Amer. J. Physiol. 252, C349-C355. 13. Moldeus, P., Hogberg, J., and Orrenius, S. (1978) in Methods in Enzymology (Fleischer, S., and Packer, L., Eds.), Vol. 52, pp. 6071, Academic Press, San Diego. 14. Lemasters, J. J., and Hackenbrock, C. R. (1979) in Methods in Enzymology (Fleischer, S., and Packer, L., Eds.), Vol. 56, pp. 530544, Academic Press, San Diego. 15. Emaus, R. K., Grunwald, R., and Lemasters, J. J. (1986) Biochim. Biophys. Acta 850,436-448. 16. Johnson, L. V., Walsh, M. L., Bockus, B. J., and Chen, L. B. (1981) J. Cell Biol. 88,526-535. 17. Chiba, K., Trevor, A., and Castagnoli, N., Jr. (1984) Biochem. Biophys. Res. Commun. 128,1228-1232. 18. Singh, Y., Bhatnagar, R., Sidhu, G. S., Batra, J. K., and Krishna, G. (1989) Arch. Biochem. Biophys. 271,217-222. 19. Di Monte, D., Ekstrom, G., Shinka, T., Smith, M. T., Trevor, A. J., and Castagnoli, N., Jr. (1987) Chem:Biol. Interact. 62,105116. 20. Di Monte, D., Shinka, T., Sandy, M. S., Castagnoli, N., Jr., and Smith, M. T. (1988) Drug Metab. Dispos. 16,250-255. 21. Masaki, N., Thomas, A. P., Hoek, J. B., and Farber, J. L. (1989) Arch. Biochem. Biophys. 272,152-161. 22. Nieminen, A., Dawson, T. L., Gores, G. J., Kawanishi, T., Herman, B., and Lemasters, J. J. (1990) Biochem. Biophys. Rex Commun., 167,600p606. 23. Gores, G. J., Nieminen, A., Wray, B. E., Herman, B., and Lemasters, J. J. (1989) J. Clin. Znuest. 83,386-396. 24. Gores, G. J., Nieminen, A., Fleishman, K. E., Dawson, T. L., Herman, B., and Lemasters, J. J. (1988) Amer. J. Physiol. 255, C315-
C322.
