- Email: [email protected]
Protection by acidotic pH and fructose against lethal injury to rat hepatocytes from mitochondrial inhibitors, ionophores and oxidant chemicals
Vol.
167,
March
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
16, 1990
PROTECTION
BY ACIDOTIC
TO RAT HEPATOCYTES
Anna-Liisa
pH AND FRUCTOSE
AGAINST
LETHAL
FROM MITOCHONDRIAL INHIBITORS, AND OXIDANT CHEMICALS1 Nieminen, Thomas L. Dawson,
Toru Kawanishi3,
600-606
INJURY
IONOPHORES
Gregory J. Gores2,
Brian Herman and John J. Lemasters4
Laboratories for Cell Biology, Department of Cell Biology & Anatom , School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, N E 27599-7090 Received
January
11,
1990
The importance of mitochondrial ATP formation and extracellular acidosiswas evaluated in hepatocyte suspensionsafter different toxic treatments. Acidotic pH was protective against cell killing from all toxic treatments examined except for pronase, a toxic protease. Fructose, a substrate for glycolytic ATP formation, provided good protection against toxicity from cyanide, oli omycin, t-butyl hydro eroxide, menadione and cystamine. Protection by fructose against ECCP, gramicidin and E!r-A23187 required oligomycin. This indicated that these ionophores were causing cytotoxicity by uncoupling oxidative phosphorylation. Fructose provided little protection against pronase and HgC12,the latter compound being a potent inhibitor of glycolysis. In conclusion, disruption of mitochondrial ATP formation was a common event contributing to the toxicity of chemical oxidants and ionophores. Acidotic pH was generally protective under these conditions of impaired ATP generation. 0 1990 Academic Press.Inc.
Depletion of ATP is a typical feature of hypoxic and toxic injury (e.g., ref l-3). In hypoxia, ATP depletion is causedby inhibition of aerobic mitochondrial ATP formation, an effect which can be mimicked by the respiratory inhibitor, cyanide. Mitochondria may also be a target of injury by toxic chemicals (2-4). For example, oxidative injury by the strong oxidant, HgC12,appears to causeATP depletion by depolarizing the mitochondrial membrane lThis work was supported by Grants AGO7218 and DK30874 from the National Institutes of Health and Grant J-1433 from the Office of Naval Research. %he present addressof Dr. Gores is GI Research Unit, Mayo Clinic, Rochester, MN 55905. %‘he resent addressof Dr. Kawanishi is Biological Safety Research Center, National Institute of R ygienic Sciences,Tokyo 158, Japan. ?To whom correspondence should be addressed. Abbreviations used are: BSA, bovine serum albumin; t-BuOOH, tert-butyl hydroperoxide; CCCP, carbonyl cyanide m-chlorophenylh drazone; IAA, iodoacetic acid; KRH, Krebs-Ringers-Hepes buffer containing 115 mM KaC1, 5 mM KCl, 2 mM CaC12, 1 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHEPES buffer, pH 7.4; MPP+, l-methyl-C phenylpyridinium; MPTP, 1-methyl-4-phenyl-1,2,3@etrahydropyridine. 0006-291X/90 Copyright All rights
$1.50 0 1990 by Academic Press, of reproduction in any form
Inc. reserved.
600
Vol.
167,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
(3). However, it is not clear to what extent ATP depletion by itself is the principal factor leading to cell killing in different models of acute toxic injury. Previously, fructose was shown to protect against lethal cell injury from anoxia and cyanide in isolated rat livers (5,6) and hepatocyte suspensions (1,7). Fructose also protected hepatocytes against MPTP and its toxic metabolite, MPP+ (8). Fructose protection in all these circumstances was mediated by glycolytic formation of ATP. Other sugars which are not readily metabolized by liver such as glucose did not protect. Extracellular acidosis also protects against hypoxic cell death (1,9), an effect mediated by intracellular acidification (10,ll). Little is known, however, concerning protection by acidosis and fructose against toxic cell injury. Thus, the aim of this study was to evaluate fructose and1 acidosis as protective agents in models of acute toxic injury to hepatocyte suspensions. The results indicate that mitochondrial dysfunction and consequent disruption of ATP formation are common events leading to cell death caused by mitochondrial inhibitors, ionophores, and chemical oxidants. Acidosis substantially delays cell killing by these toxic treatments and may be generally protective after cellular ATP depletion. MATERIALS
AND METHODS
Hepatocyte kolation - Hepatocytes were isolated from 18-20 hour-fasted male Sprague-Dawley rats (200-250g) by collagenase perfusion as previously described (1). Viability of the cells was ~90% determined by trypan blue exclusion. Qtotoxicity and enzyme assays - Cell viability in hepatocyte suspensions was monitored using propidium iodide fluorescence (1). Freshly isolated hepatocytes were incubated in KRH buffer containing 1 PM propidium iodide and In experiments where monensin was employed, cells were incubated in 10 mM Na+, 105 mM choline-substituted KRH buffer, pH 7.4. Propidium iodide fluorescence was measured every 30 main with a Sequoia-Turner Model 450 filter fluorometer (Mountain View, CA) using 520 run excitation and 605 nm emission filters. Under these conditions, fluorescence is linearly proportional to loss of cell viability as measured by lactate dehydro enase release and nuclear staining with trypan blue (1). At the end of each experiment, 37 s PM digitonin was added to permeabilize all cells and obtain a maximal fluorescence reading corresponding to 100% cell death. Glyceraldehyde-3-phosphate dehydrogenase was assayed spectroph~a;;;~lly by NADH oxidation in the presence of 1,3-bisphosphoglycerate (12). - Bovine glyceraldehyde-3-phosphate dehydrogenase was obtained from Sigma (St. Louis, MO). BrA23187, the non-fluorescent derivative of the calcium ionophore, A23187, was obtained from Calbiochem (San Diego, CA). All other reagents were of analytical grade and obtained from the usual commercial sources. RESULTS Protection by jkctose against lethal cell injury - In isolated hepatocytes and intact perfused livers, glycolytic ATP formation utilizing endogenous glycogen or exogenous fructose protects against cell killing in models of hypoxic injury (l&7). In order to evaluate whether protection by fructose is a general phenomenon in hepatocellular injury, cell viability was measured during different toxic treatments in the presence and absence of 20 mM fructose (Fig. 1). Fructose provided substantial protection against cytotoxicity by the mitochondrial inhibitors, cyanide (2.5 mM) and oligomycin (10 &ml), but little protection against the uncoupler, CCCP (10 FM). Iodoacetic acid (0.5 mM), which inhibits the glycolytic enzyme 601
Vol.
167,
No.
2, 1990
BIOCHEMICAL
0
AND
7.4
BIOPHYSICAL
I
Fructose
Gramicidin
Br-A23187
D
COMMUNICATIONS
eZa 6.5
CCCP
Oligomycin
RESEARCH
KCNIIAA
Pronase
100
50
0 Menadione
t-BuOOH
Cystamine
Hg’X
Figure 1. Effect of acidosis andfnrctose on hepatocyte viability during treatments with various toxtc chemicals - Fresh1 isolatedhe atocyteswere suspended in unmodifiedKRH, KRH adjustedto pH 6.5 with ACl, or KR r-f supplemented with 20 mM fructose. After 30 min preincubation,toxic chemicalswere added,andviability wasassessed after variousperiodsof time asdescribedin MATERIALS AND METHODS. Concentrationsand timesof
aspercent of the viability of control cellsat pH cate determinationsfrom 2 or morecell isolations. glyceraldehyde-3-phosphate dehydrogenase, abolished protection by fructose against cyanide toxicity. Fructose alone provided little protection against the toxicity of the ionophores, gramicidin D (250 nM) and Br-A23187 (10 PM), and the protease, pronase (0.5 mg/ml). However, fructose provided substantial protection against toxicity by the oxidant chemicals, menadione (50 PM), t-BuOOH (50 PM) and cystamine (2.5 mM). Fructose did not protect against HgCl2 (50 PM) which at this concentration was a potent inhibitor of bovine glyceraldehyde-3-phosphate dehydrogenase(data not shown). Fructose provided good protection against cyanide and oligomycin toxicity but little protection against CCCP. An explanation may be that CCCP, but not cyanide or 602
Vol.
167,
No.
BIOCHEMICAL
2, 1990
100
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
A
e i50e
AND
50 0
f
--
P
6 t &
s
o-
C t 50-
O-
.-==Ys 0
30
Time
60
90
120
(min)
Fi re 2. Effect of fructoseand oligomycinon the cytotoxicityof CCCP,Br-A23187and gramzc m D - Freshly isolatedhepatocyteswere incubatedin KRH in the resenceor absenceof 20 mM fructose.After 30min, CCCP(10 PM, panelA B), and gramicidinD (250 nM, panel C) were added with or without 10 &ml ohgomycin. Cell viabihty was assessed asdescribedin Additions are: none, open circles;toxic chemical(CCCP, Br-A23187 or gramicidin D), closedcircles;toxic chemicalplusfructose,closedtrian es;toxic chemicalplusfructoseand oligomydn,closedsquares.Data representmeansf d.E. of triplicate determinationsfrom 3 or more cellisolations.
7i.F
oligomycin, stimulates ATP hydrolysis by the mitochondrial FIFO-ATPase. Such uncoupled ATP consumption would counteract the benefit of glycolytic ATP generation from fructose. To test this hypothesis, we examined whether oligomycin, a specific inhibitor of mitochondrial ATPase, might confer upon fructose the property of protecting against CCCP toxicity (Fig. 2). In confirmation, oligomycin (10 pg/ml) plus fructose conferred complete protection against CCCP toxicity. This occurred despite the fact that oligomycin by itself was toxic, causing 50% cell killing in 90 min (Fig. 1) and 90% cell killing in 120 min. Oligomycin plus fructose also1provided complete protection against toxicity by BrA23187 and substantial protection against gramicidin D toxicity (Fig. 2). Thus, cytotoxicity by BrA23187 and gramicidin D appeared to be mediated by uncoupling of oxidative phosphorylation and intracellular A’TP hydrolysis. Protection againsttoxic cell killing by acidoticpH - Previously, acidotic pH was shown to be protective in models of hypoxic cell injury (1,9-11). To determine whether acidosiswas generally protective in hepatocellular injury, cell killing at pH 7.4 and pH 6.5 was compared after exposure of hepatocytes to various toxic chemicals (Fig. 1). Acidotic pH protected against lethal cell injury from exposure to all toxic chemicalsexamined except pronase. Role of intracellular acidosisin protection by fnrctose - To determine if protection by fructose was due to lactic acid formation and subsequent intracellular acidosis,hepatocytes 603
Vol.
167,
No.
2, 1990
BIOCHEMICAL
0'
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
I 0
30
60
90
120
Time (min)
Figure 3. Effect of fructoseon cell viabilig in monensin-treated hepatocytes- Freshly isolated heuatocvtes were incubated in modified KRH at DH 7.4. Fructose (20 mMj wasincluded where i
were exposed to cyanide, oligomycin, and monensin (10 PM) at pH 7.4. Monensin catalyzes the exchange of Na+ for H+, and we have previously shown that intracellular pH is clamped to extracellular pH under these conditions (10,ll). Monensin alone did not produce cell killing, but cyanide and oligomycin together with monensin caused 92% cell killing after 120 min (Fig. 3). Fructose delayed this cell killing significantly. Thus, protection by fructose was not accounted for by intracellular acidosis secondary to accumulation of lactic acid. DISCUSSION In hypoxia, ATP formation from oxidative phosphorylation is inhibited due to lack of oxygen. Cyanide mimics this inhibition, and we have previously shown that cyanide causes ATP depletion and cell death in glycogen-depleted hepatocytes from fasted rats (1). Here, we show that mitochondrial dysfunction and subsequent disruption of ATP formation by oxidative phosphorylation are also common events in other types of toxic injury. Fructose, an efficient glycolytic substrate in liver, provided near complete protection against toxicity of the mitochondrial inhibitors, cyanide and oligomycin, but little protection against CCCP, Br-A23187, and gramicidin D. Although oligomycin was cytotoxic by itself, in the presence of fructose it confelred substantial protection against cell killing by CCCP, BrA23187 and gramicidin. Oligomycin is a specific inhibitor of mitochondrial ATPase, whereas fructose is a potent substrate for glycolysis. The observation of oligomycin plus fructose-dependent protection against cytotoxicity indicates, therefore, that an important mechanism contributing to cell killing by Br-A23187 and gramicidin is mitochondrial uncoupling and consequent disruption of ATP formation. In isolated mitochondria, both A23187 and gramicidin are well known to cause uncoupling in the presence of the appropriate cations (13,14). CCCP is a classical mitochondrial uncoupler which collapses the proton electrochemical gradient across the mitochondrial inner membrane. Recently, it was proposed that cell 604
Vol.
167,
No.
2, 1990
BIOCHEMICAL
killing with mitochondrial brane potential
AND
BIOPHYSICAL
toxins is better correlated upon oligomycin
COMMUNICATIONS
with loss of the mitochondrial
than with ATP depletion per se (15).
CCCP toxicity was dependent
RESEARCH
Here, fructose
protection
(Fig. 2). The synergism
memagainst
of fructose
and
oligomycin can only be explained by an increase in the availability of ATP, since neither fructose nor oligomycin will promote mitochondrial repolarization in the presence of CCCP. Thus, disruption of ATP generation rather than mitochondrial membrane depolarization appears to be the more critical event leading to cell killing by mitochondrial inhibitors. Many workers Olofsdottir
have employed
and coworkers
A23187 as a model of Ca2+-dependent
showed loss of mitochondrial
glutathione
cell killing.
and ensuing oxidation
of pyridine nucleotides preceding lethal cell injury from A23187 in rat hepatocytes (4). Shier et al. showed an early increase in hydrolysis of arachidonate-containing phospholipids in 3T3 mouse fibroblasts
by A23187 (16). Such changes have been postulated to result from
increases of cytosolic free Ca2+. However,
these events might also be the consequence of
A23187-induced
since uncoupling
mitochondrial
uncoupling,
tion and inhibits ATP-dependent
leads to mitochondrial
oxida-
reacylation of lysophosphatides.
Fructose also protected against several oxidant chemicals including menadione, cystamine and r-BuOOH (Fig. 1). Although each of these compounds has been shown to cause oxidative stress by a different mechanism (2,17,18), fructose protection suggests that ATP depletion is a common event leading to lethal cell injury for all. Fructose did not protect against toxicity of another oxidant chemical, HgC12. However, inhibitor
of the glycolytic
HgClz-treated tochondrial
enzyme
HgC12 proved to be a potent
glyceraldehyde-3-phosphate
dehydrogenase.
Thus,
cells could not utilize fructose as a source of glycolytic ATP. Nonetheless, injury may be important
in cell injury from HgC12, and we have recently shown
that HgC12 causes early mitochondrial to the onset of cell death (3). Extracellular
depolarization
and subsequent
ATP depletion prior
acidosis was highly protective in almost all models of toxic injury (Fig. 1).
Taken together, our data suggest that acidotic pH is protective when the primary injury is disruption
mi-
of cellular ATP supply.
Only the toxicity of pronase, a mixture of pro-
teases which1 attack the cell surface (19), was independent hibited hepatocytes, we demonstrated
effect of
of pH. Previously, in cyanide-in-
that protection by extracellular
acidosis was mediated
by intracellular acidification (lO,ll), and it seems likely that intracellular acidification derlies the protective effect of extracellular acidosis in other models of toxic cell injury. Why do cells which are equally depleted of ATP die at different rates?
un-
For example,
rates of ATP depletion are virtually the same after exposure to HgCl2 as after cyanide plus iodoacetate (1,3), but HgC12 toxicity causes much faster of cell killing. The explanation may be the degree of intracellular acidosis caused by proton liberation during hydrolysis of ATP In cyanide plus iodoacetate-treated cells, intracellular and other high energy phosphates. pH falls by more than 1 unit and remains acidotic until the onset of cell death (10,ll). This persistent intracellular acidosis protects against cell killing and indicates that the plasma membrane remains highly impermeable
under these conditions.
Other toxins may promote
membrane leakiness and dissipate pH gradients across the plasma membrane. This leakiness will cause cytosolic pH to return to neutrality and accelerate cell killing. Such differ60.5
Vol.
167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
ences of intracellular pH may explain why rates of cell killing can be so different under conditions of nearly identical ATP depletion. Since acidotic pH was so potent in protecting against various types of toxic injury, we investigated whether fructose was providing protection because of intracellular lactic acidosis. However, when intracellular pH was clamped to extracellular pH with monensin, fructose retained its ability to protect against cell killing (Fig. 3). Moreover, acidosis does not diminish the rate of ATP depletion after treatment of hepatocytes with cyanide and iodoacetate (1). Thus, acidosis and fructose protect against cell killing by independent mechanisms. In conclusion, acidosis and fructose were highly protective against toxic hepatocellular injury in nearly all models examined. These observations suggest nutritional and metabolic interventions which might be therapeutically effective in toxic liver injury. REFERENCES 1. Gores, G.J., Nieminen, A.-L., Fleishman, K.E., Dawson, T.L., Herman, B., Lemasters, J.J. (1988) Am. J. Physiol. 255, C315-C322. 2. Masaki, N., Kyle, M.E., Serroni, A., Farber, J.L. (1989) Arch. Biochem. Biophys. 270, 672-680. 3. Nieminen, A.-L., Gores, G.J., Dawson, T.L., Herman, B., Lemasters, J.J. (1990) J. Biol. Chem. 265,2399-2408. Olafsdottir, K., Pascoe, G.A., Reed, D.J. (1988) Arch. Biochem. Biophys. 263,226-235. 4: Anundi, I., Kmg, J., Owen, D.A., Schneider, H., Lemasters, J.J., Thurman, R.G. (1987) Am. J. Physiol. 253, G390-G396. Younes, M., Strubelt, 0. (1988) Pharmacol. Toxicol. 63,382-385. ; Anundi, I., de Groot, H. 1989) Am. J. Ph siol. 257, G58-G64. 8: Di Monte, D., Sandy, hi .S., Blank, L., Hmith, M.T. (1988) Biochem. Biophys. Res. Commun. 153,734-740. 9. Bonventre, J.V., Cheung, J.Y. 1985) Am. J. Physiol. 249, C149-C159. 10. Gores! G.J., Nreminen, A.-L., 6 awson, T.L., Herman, B., Lemasters, J.J. (1988) In Integration of Mitochondrial Function (J.J. Lemasters, C.R. Hackenbrock, R.G. Thurman, H.V. Westerhoff, Eds. , pp. 421-428, Plenum Press, New York. 11. Gores, G.J., Nieminen, A.- I! ., Wray, B.E., Herman, B., Lemasters, J.J. (1989) J. Clin. Invest. 83,386-396. 12. Patnode, R., Bartle, E., Hill, E.J., LeQuire, V., Park, J.H. (1976) J. Biol. Chem. 251, 4468-4475. 13. Reed, P.W., Lardy, H.A. (1972) J. Biol. Chem. 247,6970-6977. Rottenberg, H., Hashimoto, K. (1986 Biochemist 25, 1747-1755. ::: Masaki, N., Thomas, A.P., Hock, J. B ., Farber, J.z . (1989) Arch. Biochem. Biophys. 272,152-161. Shier, W.T., Dubourdieu, D.J. (1985) Am. J. Pathol. 120,304-315. it: Thor, H., Smith, M.T., Hartzell, P., Bellomo, G., Jewell, S.A., Orrenius, S. (1982) J. Biol. Chem. 257,12419-12425. 18. Nicotera, P., Hartzell, P., Baldi, C., Svensson, S.-A, Bellomo, G., Orrenius, S. (1986) J. Biol. Chem. 261,14628-14635. 19. Mills, D.M., Zucker-Franklin, D. (1969) Am. J. Pathol. 54, 147-166.
606
167,
March
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
16, 1990
PROTECTION
BY ACIDOTIC
TO RAT HEPATOCYTES
Anna-Liisa
pH AND FRUCTOSE
AGAINST
LETHAL
FROM MITOCHONDRIAL INHIBITORS, AND OXIDANT CHEMICALS1 Nieminen, Thomas L. Dawson,
Toru Kawanishi3,
600-606
INJURY
IONOPHORES
Gregory J. Gores2,
Brian Herman and John J. Lemasters4
Laboratories for Cell Biology, Department of Cell Biology & Anatom , School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, N E 27599-7090 Received
January
11,
1990
The importance of mitochondrial ATP formation and extracellular acidosiswas evaluated in hepatocyte suspensionsafter different toxic treatments. Acidotic pH was protective against cell killing from all toxic treatments examined except for pronase, a toxic protease. Fructose, a substrate for glycolytic ATP formation, provided good protection against toxicity from cyanide, oli omycin, t-butyl hydro eroxide, menadione and cystamine. Protection by fructose against ECCP, gramicidin and E!r-A23187 required oligomycin. This indicated that these ionophores were causing cytotoxicity by uncoupling oxidative phosphorylation. Fructose provided little protection against pronase and HgC12,the latter compound being a potent inhibitor of glycolysis. In conclusion, disruption of mitochondrial ATP formation was a common event contributing to the toxicity of chemical oxidants and ionophores. Acidotic pH was generally protective under these conditions of impaired ATP generation. 0 1990 Academic Press.Inc.
Depletion of ATP is a typical feature of hypoxic and toxic injury (e.g., ref l-3). In hypoxia, ATP depletion is causedby inhibition of aerobic mitochondrial ATP formation, an effect which can be mimicked by the respiratory inhibitor, cyanide. Mitochondria may also be a target of injury by toxic chemicals (2-4). For example, oxidative injury by the strong oxidant, HgC12,appears to causeATP depletion by depolarizing the mitochondrial membrane lThis work was supported by Grants AGO7218 and DK30874 from the National Institutes of Health and Grant J-1433 from the Office of Naval Research. %he present addressof Dr. Gores is GI Research Unit, Mayo Clinic, Rochester, MN 55905. %‘he resent addressof Dr. Kawanishi is Biological Safety Research Center, National Institute of R ygienic Sciences,Tokyo 158, Japan. ?To whom correspondence should be addressed. Abbreviations used are: BSA, bovine serum albumin; t-BuOOH, tert-butyl hydroperoxide; CCCP, carbonyl cyanide m-chlorophenylh drazone; IAA, iodoacetic acid; KRH, Krebs-Ringers-Hepes buffer containing 115 mM KaC1, 5 mM KCl, 2 mM CaC12, 1 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHEPES buffer, pH 7.4; MPP+, l-methyl-C phenylpyridinium; MPTP, 1-methyl-4-phenyl-1,2,3@etrahydropyridine. 0006-291X/90 Copyright All rights
$1.50 0 1990 by Academic Press, of reproduction in any form
Inc. reserved.
600
Vol.
167,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
(3). However, it is not clear to what extent ATP depletion by itself is the principal factor leading to cell killing in different models of acute toxic injury. Previously, fructose was shown to protect against lethal cell injury from anoxia and cyanide in isolated rat livers (5,6) and hepatocyte suspensions (1,7). Fructose also protected hepatocytes against MPTP and its toxic metabolite, MPP+ (8). Fructose protection in all these circumstances was mediated by glycolytic formation of ATP. Other sugars which are not readily metabolized by liver such as glucose did not protect. Extracellular acidosis also protects against hypoxic cell death (1,9), an effect mediated by intracellular acidification (10,ll). Little is known, however, concerning protection by acidosis and fructose against toxic cell injury. Thus, the aim of this study was to evaluate fructose and1 acidosis as protective agents in models of acute toxic injury to hepatocyte suspensions. The results indicate that mitochondrial dysfunction and consequent disruption of ATP formation are common events leading to cell death caused by mitochondrial inhibitors, ionophores, and chemical oxidants. Acidosis substantially delays cell killing by these toxic treatments and may be generally protective after cellular ATP depletion. MATERIALS
AND METHODS
Hepatocyte kolation - Hepatocytes were isolated from 18-20 hour-fasted male Sprague-Dawley rats (200-250g) by collagenase perfusion as previously described (1). Viability of the cells was ~90% determined by trypan blue exclusion. Qtotoxicity and enzyme assays - Cell viability in hepatocyte suspensions was monitored using propidium iodide fluorescence (1). Freshly isolated hepatocytes were incubated in KRH buffer containing 1 PM propidium iodide and In experiments where monensin was employed, cells were incubated in 10 mM Na+, 105 mM choline-substituted KRH buffer, pH 7.4. Propidium iodide fluorescence was measured every 30 main with a Sequoia-Turner Model 450 filter fluorometer (Mountain View, CA) using 520 run excitation and 605 nm emission filters. Under these conditions, fluorescence is linearly proportional to loss of cell viability as measured by lactate dehydro enase release and nuclear staining with trypan blue (1). At the end of each experiment, 37 s PM digitonin was added to permeabilize all cells and obtain a maximal fluorescence reading corresponding to 100% cell death. Glyceraldehyde-3-phosphate dehydrogenase was assayed spectroph~a;;;~lly by NADH oxidation in the presence of 1,3-bisphosphoglycerate (12). - Bovine glyceraldehyde-3-phosphate dehydrogenase was obtained from Sigma (St. Louis, MO). BrA23187, the non-fluorescent derivative of the calcium ionophore, A23187, was obtained from Calbiochem (San Diego, CA). All other reagents were of analytical grade and obtained from the usual commercial sources. RESULTS Protection by jkctose against lethal cell injury - In isolated hepatocytes and intact perfused livers, glycolytic ATP formation utilizing endogenous glycogen or exogenous fructose protects against cell killing in models of hypoxic injury (l&7). In order to evaluate whether protection by fructose is a general phenomenon in hepatocellular injury, cell viability was measured during different toxic treatments in the presence and absence of 20 mM fructose (Fig. 1). Fructose provided substantial protection against cytotoxicity by the mitochondrial inhibitors, cyanide (2.5 mM) and oligomycin (10 &ml), but little protection against the uncoupler, CCCP (10 FM). Iodoacetic acid (0.5 mM), which inhibits the glycolytic enzyme 601
Vol.
167,
No.
2, 1990
BIOCHEMICAL
0
AND
7.4
BIOPHYSICAL
I
Fructose
Gramicidin
Br-A23187
D
COMMUNICATIONS
eZa 6.5
CCCP
Oligomycin
RESEARCH
KCNIIAA
Pronase
100
50
0 Menadione
t-BuOOH
Cystamine
Hg’X
Figure 1. Effect of acidosis andfnrctose on hepatocyte viability during treatments with various toxtc chemicals - Fresh1 isolatedhe atocyteswere suspended in unmodifiedKRH, KRH adjustedto pH 6.5 with ACl, or KR r-f supplemented with 20 mM fructose. After 30 min preincubation,toxic chemicalswere added,andviability wasassessed after variousperiodsof time asdescribedin MATERIALS AND METHODS. Concentrationsand timesof
aspercent of the viability of control cellsat pH cate determinationsfrom 2 or morecell isolations. glyceraldehyde-3-phosphate dehydrogenase, abolished protection by fructose against cyanide toxicity. Fructose alone provided little protection against the toxicity of the ionophores, gramicidin D (250 nM) and Br-A23187 (10 PM), and the protease, pronase (0.5 mg/ml). However, fructose provided substantial protection against toxicity by the oxidant chemicals, menadione (50 PM), t-BuOOH (50 PM) and cystamine (2.5 mM). Fructose did not protect against HgCl2 (50 PM) which at this concentration was a potent inhibitor of bovine glyceraldehyde-3-phosphate dehydrogenase(data not shown). Fructose provided good protection against cyanide and oligomycin toxicity but little protection against CCCP. An explanation may be that CCCP, but not cyanide or 602
Vol.
167,
No.
BIOCHEMICAL
2, 1990
100
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
A
e i50e
AND
50 0
f
--
P
6 t &
s
o-
C t 50-
O-
.-==Ys 0
30
Time
60
90
120
(min)
Fi re 2. Effect of fructoseand oligomycinon the cytotoxicityof CCCP,Br-A23187and gramzc m D - Freshly isolatedhepatocyteswere incubatedin KRH in the resenceor absenceof 20 mM fructose.After 30min, CCCP(10 PM, panelA B), and gramicidinD (250 nM, panel C) were added with or without 10 &ml ohgomycin. Cell viabihty was assessed asdescribedin Additions are: none, open circles;toxic chemical(CCCP, Br-A23187 or gramicidin D), closedcircles;toxic chemicalplusfructose,closedtrian es;toxic chemicalplusfructoseand oligomydn,closedsquares.Data representmeansf d.E. of triplicate determinationsfrom 3 or more cellisolations.
7i.F
oligomycin, stimulates ATP hydrolysis by the mitochondrial FIFO-ATPase. Such uncoupled ATP consumption would counteract the benefit of glycolytic ATP generation from fructose. To test this hypothesis, we examined whether oligomycin, a specific inhibitor of mitochondrial ATPase, might confer upon fructose the property of protecting against CCCP toxicity (Fig. 2). In confirmation, oligomycin (10 pg/ml) plus fructose conferred complete protection against CCCP toxicity. This occurred despite the fact that oligomycin by itself was toxic, causing 50% cell killing in 90 min (Fig. 1) and 90% cell killing in 120 min. Oligomycin plus fructose also1provided complete protection against toxicity by BrA23187 and substantial protection against gramicidin D toxicity (Fig. 2). Thus, cytotoxicity by BrA23187 and gramicidin D appeared to be mediated by uncoupling of oxidative phosphorylation and intracellular A’TP hydrolysis. Protection againsttoxic cell killing by acidoticpH - Previously, acidotic pH was shown to be protective in models of hypoxic cell injury (1,9-11). To determine whether acidosiswas generally protective in hepatocellular injury, cell killing at pH 7.4 and pH 6.5 was compared after exposure of hepatocytes to various toxic chemicals (Fig. 1). Acidotic pH protected against lethal cell injury from exposure to all toxic chemicalsexamined except pronase. Role of intracellular acidosisin protection by fnrctose - To determine if protection by fructose was due to lactic acid formation and subsequent intracellular acidosis,hepatocytes 603
Vol.
167,
No.
2, 1990
BIOCHEMICAL
0'
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
I 0
30
60
90
120
Time (min)
Figure 3. Effect of fructoseon cell viabilig in monensin-treated hepatocytes- Freshly isolated heuatocvtes were incubated in modified KRH at DH 7.4. Fructose (20 mMj wasincluded where i
were exposed to cyanide, oligomycin, and monensin (10 PM) at pH 7.4. Monensin catalyzes the exchange of Na+ for H+, and we have previously shown that intracellular pH is clamped to extracellular pH under these conditions (10,ll). Monensin alone did not produce cell killing, but cyanide and oligomycin together with monensin caused 92% cell killing after 120 min (Fig. 3). Fructose delayed this cell killing significantly. Thus, protection by fructose was not accounted for by intracellular acidosis secondary to accumulation of lactic acid. DISCUSSION In hypoxia, ATP formation from oxidative phosphorylation is inhibited due to lack of oxygen. Cyanide mimics this inhibition, and we have previously shown that cyanide causes ATP depletion and cell death in glycogen-depleted hepatocytes from fasted rats (1). Here, we show that mitochondrial dysfunction and subsequent disruption of ATP formation by oxidative phosphorylation are also common events in other types of toxic injury. Fructose, an efficient glycolytic substrate in liver, provided near complete protection against toxicity of the mitochondrial inhibitors, cyanide and oligomycin, but little protection against CCCP, Br-A23187, and gramicidin D. Although oligomycin was cytotoxic by itself, in the presence of fructose it confelred substantial protection against cell killing by CCCP, BrA23187 and gramicidin. Oligomycin is a specific inhibitor of mitochondrial ATPase, whereas fructose is a potent substrate for glycolysis. The observation of oligomycin plus fructose-dependent protection against cytotoxicity indicates, therefore, that an important mechanism contributing to cell killing by Br-A23187 and gramicidin is mitochondrial uncoupling and consequent disruption of ATP formation. In isolated mitochondria, both A23187 and gramicidin are well known to cause uncoupling in the presence of the appropriate cations (13,14). CCCP is a classical mitochondrial uncoupler which collapses the proton electrochemical gradient across the mitochondrial inner membrane. Recently, it was proposed that cell 604
Vol.
167,
No.
2, 1990
BIOCHEMICAL
killing with mitochondrial brane potential
AND
BIOPHYSICAL
toxins is better correlated upon oligomycin
COMMUNICATIONS
with loss of the mitochondrial
than with ATP depletion per se (15).
CCCP toxicity was dependent
RESEARCH
Here, fructose
protection
(Fig. 2). The synergism
memagainst
of fructose
and
oligomycin can only be explained by an increase in the availability of ATP, since neither fructose nor oligomycin will promote mitochondrial repolarization in the presence of CCCP. Thus, disruption of ATP generation rather than mitochondrial membrane depolarization appears to be the more critical event leading to cell killing by mitochondrial inhibitors. Many workers Olofsdottir
have employed
and coworkers
A23187 as a model of Ca2+-dependent
showed loss of mitochondrial
glutathione
cell killing.
and ensuing oxidation
of pyridine nucleotides preceding lethal cell injury from A23187 in rat hepatocytes (4). Shier et al. showed an early increase in hydrolysis of arachidonate-containing phospholipids in 3T3 mouse fibroblasts
by A23187 (16). Such changes have been postulated to result from
increases of cytosolic free Ca2+. However,
these events might also be the consequence of
A23187-induced
since uncoupling
mitochondrial
uncoupling,
tion and inhibits ATP-dependent
leads to mitochondrial
oxida-
reacylation of lysophosphatides.
Fructose also protected against several oxidant chemicals including menadione, cystamine and r-BuOOH (Fig. 1). Although each of these compounds has been shown to cause oxidative stress by a different mechanism (2,17,18), fructose protection suggests that ATP depletion is a common event leading to lethal cell injury for all. Fructose did not protect against toxicity of another oxidant chemical, HgC12. However, inhibitor
of the glycolytic
HgClz-treated tochondrial
enzyme
HgC12 proved to be a potent
glyceraldehyde-3-phosphate
dehydrogenase.
Thus,
cells could not utilize fructose as a source of glycolytic ATP. Nonetheless, injury may be important
in cell injury from HgC12, and we have recently shown
that HgC12 causes early mitochondrial to the onset of cell death (3). Extracellular
depolarization
and subsequent
ATP depletion prior
acidosis was highly protective in almost all models of toxic injury (Fig. 1).
Taken together, our data suggest that acidotic pH is protective when the primary injury is disruption
mi-
of cellular ATP supply.
Only the toxicity of pronase, a mixture of pro-
teases which1 attack the cell surface (19), was independent hibited hepatocytes, we demonstrated
effect of
of pH. Previously, in cyanide-in-
that protection by extracellular
acidosis was mediated
by intracellular acidification (lO,ll), and it seems likely that intracellular acidification derlies the protective effect of extracellular acidosis in other models of toxic cell injury. Why do cells which are equally depleted of ATP die at different rates?
un-
For example,
rates of ATP depletion are virtually the same after exposure to HgCl2 as after cyanide plus iodoacetate (1,3), but HgC12 toxicity causes much faster of cell killing. The explanation may be the degree of intracellular acidosis caused by proton liberation during hydrolysis of ATP In cyanide plus iodoacetate-treated cells, intracellular and other high energy phosphates. pH falls by more than 1 unit and remains acidotic until the onset of cell death (10,ll). This persistent intracellular acidosis protects against cell killing and indicates that the plasma membrane remains highly impermeable
under these conditions.
Other toxins may promote
membrane leakiness and dissipate pH gradients across the plasma membrane. This leakiness will cause cytosolic pH to return to neutrality and accelerate cell killing. Such differ60.5
Vol.
167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
ences of intracellular pH may explain why rates of cell killing can be so different under conditions of nearly identical ATP depletion. Since acidotic pH was so potent in protecting against various types of toxic injury, we investigated whether fructose was providing protection because of intracellular lactic acidosis. However, when intracellular pH was clamped to extracellular pH with monensin, fructose retained its ability to protect against cell killing (Fig. 3). Moreover, acidosis does not diminish the rate of ATP depletion after treatment of hepatocytes with cyanide and iodoacetate (1). Thus, acidosis and fructose protect against cell killing by independent mechanisms. In conclusion, acidosis and fructose were highly protective against toxic hepatocellular injury in nearly all models examined. These observations suggest nutritional and metabolic interventions which might be therapeutically effective in toxic liver injury. REFERENCES 1. Gores, G.J., Nieminen, A.-L., Fleishman, K.E., Dawson, T.L., Herman, B., Lemasters, J.J. (1988) Am. J. Physiol. 255, C315-C322. 2. Masaki, N., Kyle, M.E., Serroni, A., Farber, J.L. (1989) Arch. Biochem. Biophys. 270, 672-680. 3. Nieminen, A.-L., Gores, G.J., Dawson, T.L., Herman, B., Lemasters, J.J. (1990) J. Biol. Chem. 265,2399-2408. Olafsdottir, K., Pascoe, G.A., Reed, D.J. (1988) Arch. Biochem. Biophys. 263,226-235. 4: Anundi, I., Kmg, J., Owen, D.A., Schneider, H., Lemasters, J.J., Thurman, R.G. (1987) Am. J. Physiol. 253, G390-G396. Younes, M., Strubelt, 0. (1988) Pharmacol. Toxicol. 63,382-385. ; Anundi, I., de Groot, H. 1989) Am. J. Ph siol. 257, G58-G64. 8: Di Monte, D., Sandy, hi .S., Blank, L., Hmith, M.T. (1988) Biochem. Biophys. Res. Commun. 153,734-740. 9. Bonventre, J.V., Cheung, J.Y. 1985) Am. J. Physiol. 249, C149-C159. 10. Gores! G.J., Nreminen, A.-L., 6 awson, T.L., Herman, B., Lemasters, J.J. (1988) In Integration of Mitochondrial Function (J.J. Lemasters, C.R. Hackenbrock, R.G. Thurman, H.V. Westerhoff, Eds. , pp. 421-428, Plenum Press, New York. 11. Gores, G.J., Nieminen, A.- I! ., Wray, B.E., Herman, B., Lemasters, J.J. (1989) J. Clin. Invest. 83,386-396. 12. Patnode, R., Bartle, E., Hill, E.J., LeQuire, V., Park, J.H. (1976) J. Biol. Chem. 251, 4468-4475. 13. Reed, P.W., Lardy, H.A. (1972) J. Biol. Chem. 247,6970-6977. Rottenberg, H., Hashimoto, K. (1986 Biochemist 25, 1747-1755. ::: Masaki, N., Thomas, A.P., Hock, J. B ., Farber, J.z . (1989) Arch. Biochem. Biophys. 272,152-161. Shier, W.T., Dubourdieu, D.J. (1985) Am. J. Pathol. 120,304-315. it: Thor, H., Smith, M.T., Hartzell, P., Bellomo, G., Jewell, S.A., Orrenius, S. (1982) J. Biol. Chem. 257,12419-12425. 18. Nicotera, P., Hartzell, P., Baldi, C., Svensson, S.-A, Bellomo, G., Orrenius, S. (1986) J. Biol. Chem. 261,14628-14635. 19. Mills, D.M., Zucker-Franklin, D. (1969) Am. J. Pathol. 54, 147-166.
606