Toxicology Letters, 63 (1992f271-281
211
0 1992 Elsevier Science Publishers B.V. All rights reserved 0378-4274/92/%5.00
TOXLET 02812
Independence and additivity of cultured hepatocyte 1. * killing by Ca2’ overload and ATP depletion
Lisa M. Kamendulis and George B. Corcoran Toxicology Program, College of Pharmacy, University of New Mexico, Albuquerque, NM (USA)
(Received 26 June 1992) (Accepted 26 August 1992) Key words: DNA fragmentation;
Ca”; A23 187; Chemical hypoxia; ATP; Cultured hepatocytes
SUMMARY In two competing models of toxic cell death, hepatocyte killing by chemical hypoxia (CN/IAA) is attributed to ATP depletion and killing by A23187 is attributed to Ca*‘-induced damage. The independence of these models can be questioned because CN/IAA elevates Ca*’ before killing lclcl hepatoma cells and because the ATP source fructose prevents hepatocyte killing by Br-A23187. In the present studies, cultured mouse hepatocytes were exposed to CN/IAA, A23187, or treatments in combination. A23187 produced toxicity proportional to Ca’+-activated DNA fragmentation. CN/IAA caused comparable toxicity but no fragmentation of DNA. Treatments in combination were more toxic than either treatment alone. Aurintricarboxylic acid, a Ca*‘-endonuclease inhibitor, decreased DNA fragmentation and the toxicity of A23187 and combination treatment without affecting CN/IAA toxicity. ATP plus oligomycin decreased CN/IAA and combination treatment toxicity but not that of A23187. These findings indicate that cultured mouse hepatocytes are killed through mechanisms that are independent and additive in their toxicities.
INTRODUCTION
Hepatic necrosis is an unplanned consequence of exposure to a wide range of xenobiotics, including beneficial drugs and useful industrial chemicals [ 11. Precisely how these agents cause irreversible cell damage remains unclear. At present, there are two competing hypotheses to explain necrosis (toxic cell death) in the liver. One states
Correspondence to: George B. Corcoran, Toxicology Program, College of Pharmacy, University of New Mexico, Albuquerque, NM, USA 87131-1066. *Portions of this work were presented at the annual meeting of the Federation of American Societies for Experimental Biology, Atlanta, GA, April 21-26, 1991, and published in abstract form (FASEB J. 5, A1277).
278
that death is the result of damage to Ca” regulation [2], whereas the second attributes cell death to energy depletion [3]. Many studies link hepatic necrosis in vivo and hepatocyte death in vitro to elevated concentrations of Ca” (reviewed in Refs. 4-8). although a vital target of elevated Ca*’ has not yet been identi~ed. Some results implicate attack on cytoskeletal proteins [9]. Cytoskeleton degradation by Ca”-activated proteases could cause lethal membrane blebbing. Other data suggest that phospholipids are the target [lo]. The activation of Ca” -stimulated lipases could lead to irreversible membrane dysfunction. Another possible target in toxic cell death is DNA. Ca’+ has been shown to activate an endonuclease that fragments DNA before liver cells die by necrosis and cultured hepatocytes are killed by acetaminophen and dimethylnitrosamine [I l-141. Fragmentation of DNA is thought to explain another type of cell death that is distinct from toxic cell death termed apoptosis, or programmed cell death [15]. This is a beneficial form of physiological cell death triggered during homeostatically controlled events such as normal tissue regression and immature l~phocyte selection (see Ref. 4). Cellular Ca”’ rises in apoptosis, triggering a Ca”“- activated endonudease that cleaves DNA into characteristic oligonucleosome-length fragments [ 161. The importance of Ca’“‘ and DNA fragmentation in apoptotic and necrotic cell death is further suggested by studies with aurintri~arboxyli~ acid. This Ca’+-endonuclease inhibitor decreases DNA fragmentation and toxic cell death caused by acetaminophen and dimethylnitrosamine [I 3,14,17], as well as the apoptotic death of thymocytes induced by glucocorticoids and TCDD [ 18,193. Results from the chemical hypoxia model of toxic cell death directly challenge the Ca” hypothesis. Cyanide (CN) and iodoacetic acid (IAA) cause near total depletion of ATP in hepatocytes, however, cells appear to maintain normal Ca2’ regulation until death [3.20]. The ability of the ATP source fructose to prevent cell death caused by chemical hypoxia and by a range of other toxic chemicals including mitochondrial inhibitors, ionophores, and oxidants has promoted the view that ATP depletion and changes in cellular pH rather than Ca” overload are critical factors in toxic cell death 121-231. However, DNA status has not been examined during chemical hypoxia and DNA fragmentation has not been excluded as a cause of cell death in this model. The possible importance of DNA is suggested by two observations: (i) Ca” is increased during lclc7 hepatoma ceI1 killing by chemical hypoxia, and (ii) Ca”” chelators delay the onset of Iclc7 cell killing without preventing ATP depletion [24]. Finally, a fimdamental observation in support of the Ca*’ hypothesis is the ability of Ca’+ ionophore A23187 to kill hepatocytes. However, fructose and oligomycin protect against Br-A23187 toxicity 1221,suggesting that A23187 also depletes ATP and that energy depletion explains cell death rather than Ca” effects on targets such as DNA. Although there is support for both the CaZ+ overload and ATP depletion hypotheses of toxic cell death, the independence of these mechanisms remains in question. In studies presented here, the toxicities of A23 187 and chemical hypoxia were measured in the same target cell, the cultured mouse hepatocyte. If these insults work through independent mechanisms, they should produce additive cell killing. Also, antagonists
219
that are selective for one mechanism should have no protective effect against cell killing by the second mechanism. MATERIALS AND METHODS
Materials A23187, ATP, collagenase (Type IV-S), diphenylamine, fructose, Hanks’ balanced salt solution, HEPES, N-lauroylsarcosine, insulin, iodoacetic acid, nicotinamide, oligomycin, penicillin G, potassium cyanide, streptomycin, Tris-HCl, Triton X-100, and William’s E media were purchased from Sigma Chemical Co. (St. Louis, MO). Falcon tissue culture dishes and fetal bovine serum were from Fisher Scientific (Fair Lawn, NJ) and Hyclone Laboratories, Inc. (Logan, UT), respectively. Agarose was from Bio-Rad Laboratories (Richmond, CA). All chemicals were of analytical grade or higher. Hepatocyte preparation and culture Hepatocytes were isolated from 30-35 g male ICR mice (Harlan Sprague-Dawley, Inc., Indianapolis, IN) by a two-step collagenase perfusion yielding ~95% cell viability [ 121. Hepatocytes were plated at a density of 1x1 O6per ml of complete William’s E media containing 15 mM nicotinamide and 10% fetal bovine serum [25]. After cells were allowed to adhere for 18 h, media was changed to William’s E lacking nicotinamide and fetal bovine serum, and toxin treatment was begun. Evaluation of cytotoxicity and DNA damage Release of lactate dehydrogenase into the culture media was assayed spectrophotometrically and used as a measure of toxic cell death [26]. Enzyme release in control cells is expressed in international units, where 1 IU represents oxidation of 1 ,uM substrate per min at 25°C. Treatment effects are reported as percent of control leakage. Quantitation of DNA fragmentation was by the diphenylamine assay [16]. Supernatant media was removed from plates and cells were detached using a lysis buffer containing 5 mM Tris, 20 mM EDTA, and 0.5% Triton X-100. The samples were centrifuged 20 min at 27 OOOxgto separate intact chromatin (pellet) from DNA fragments (supernatant). Pellets were resuspended in 0.5 M HClO,, and both pellet and supernatant assayed for DNA by reaction with diphenylamine [27]. DNA fragmentation was also assessed qualitatively by agarose gel electrophoresis [12]. Cells were isolated and treated for 20 min at 4°C with lysis buffer containing 0.5% Nlauroylsarcosine in place of Triton X-100. DNA was extracted with phenol-chloroform and precipitated with ethanol/sodium acetate [28]. DNA (5-10 pugllane) was separated on 1.4% agarose gels containing ethidium bromide. A Hind111 digest of A DNA provided molecular size standards.
280
TABLE I LACTATE D~HYDROGENASE RELEASE AND DNA FRAGMENTATION FOLLOWING HEPATOCYTE EXPOSURE TO CN/IAA, A23187. OR COMBINATION TREATMENT Treatment
Time (min)
Lactate dehydrogenase release (% control)
DNA fragmentation (% control)
CNIIAA
60 120 180
182 ?r 5.1* 221 rt 4.0* 245 I: 6.3*
102 I!I0.6 93 + 1.7 102 + 0.7
A23187
60 120 180
221 rt 3.6* 363 t 25.6” 383 I 9.4*
203 I 1.0* 205 i 1.2* 229 ?I 2.2*
CNilAA + A23 187
60 120 180
269 L 7.5* 501 + 5.1* 541 i: 1.8*
169 I t.9* 162 + l.O* 180 + 2.5*
Mouse hepatocytes cultured 18 h in William’s E media were exposed to 0.4 mM cyanide/O.2 mM iodoacetate, 20 PM A23 187, or both treatments. Lactate dehydrogenase leakage was determined from the rate of NADH consumption, and DNA fragmentation was assessed by sedimentation and reaction with diphenylamine. Results are mean t SE of four culture plates and are representative of three independent experiments. In control cells, lactate dehydrogenase release at 60, 120, and 180 min was 37.5 F 3.9.39 f 2.5, and 37.8 k 1.5 IU/l respectively. DNA fragmentation was 10.3 t 0.8, 11.5t 1.2, and 9.6 i 1.6%. * Si~ifi~antly different from control value (P < 0.05).
Data analysis and statistics
Results are presented as mean + SE unless otherwise indicated. Data were compared by ANOVA followed by Scheffe’s multiple comparison test [29]. Differences were attributed to treatment rather than chance variation when P50.05. RESULTS
Chemical hypoxia and A23187 each produced significant toxicity to cultured mouse hepatocytes, as reflected in the release of lactate dehy~ogenase into the media at 60, 120 and 180 min (Table I). Enzyme activity increased to 245% and 383% of control at 180 min for chemical hypoxia and A23 187, respectively. The combination treatment produced a larger increase in lactate dehydrogenase activity than either individual treatment, suggesting that the two toxins were acting independently to produce damage. Although chemical hypoxia failed to cause DNA fragmentation, A23187 increased fragmentation to 229% of control (Table I). Interestingly, DNA fragmentation by A23187 plus chemical hypoxia was slightly less than with A23187 alone, suggesting that ATP may be required for full Ca”-endonuclease activity. The Ca2*-endonuclease inhibitor aurintricarboxylic acid had no effect on CN/IAA
281
A.
a
I El:
CN/IAA
% 2
225
A23187
B.
:
Toxin Toxin
Abne + ATA
CN/IAk/A23 187
I
Ed:
:
Toxin Toxin
Alone + ATA
J
CN/KAA
A23187
CN/IAA/A23187
Fig. 1. Effects of aurintricarboxyhG acid on DNA fragmentation and cytotoxicity. Hepatocytes were treated as in TabIe I for 180 min, with or without 100 PM aur~t~carboxy~~ acid. (A) Lactate dehydrogenase leakage; (B) DNA fra~entation. Results are mea&SE of four cuhure plates and are representative of three independent experiments. An asterisk indicates the value is significantly different from toxin alone. The control lactate dehydrogenase leakage was 37.6k4.0 IUil and the control DNA fragmentation was 11.2f1.3%
toxicity but greatly reduced cytotoxicity and DNA fra~entation induced by A231 87 and the combination treatment (Fig. 1). A~int~~rboxyIic acid was unable to abolish combination treatment toxicity, supporting the view that the toxins acted independently in producing toxie cell death. DNA analysis by agarose gel electrophoresis confirmed that chemical hypoxia failed to damage DNA whereas A23187 and the combination treatment produced cleavage of genomic DNA into oligonucleosomelength fragments diagnostic of Ca2’-endonuclease attack (Fig. 2, upper panel), Aurintricarboxylic acid blocked the DNA fragmentation ladder caused by A23187 and combination treatment and maintained DNA in its native, undamaged form. Fructose has been used in past studies to prevent chemical hypoxia-induced cytotoxicity, and presumed to work by increasing ATP. In the present investigation, we
282
_A_ c
A23187 +
A23187
A__!2_ +
+A
+A0
HYPOX
tATA
+
+A
+
+ATA
HYPOXIA +
A23,,fHYPOX
+A0
t ATA
A23[HYPOX +
+A
+A0
Fig. 2. Qualitative analysis of DNA damage. Electrophoretograms are ethidium bromide-stained agarose gels ~epre~ntative of three independent experiments. Upper panel: effects of aurintricarboxylic acid on DN A damage. Lower panel: effects of ATP and oligomycin on DNA damage. Treatments were for 180 min and are described in Figs. 1 and 3. /2 = HindiKI digested 2 DNA size standards; C = untreated cant rol cells; HY POX = cyanide/iodoacetic acid; A23/f_IYPOX = A23187 plus ~yanide/iodoa~tic acid combin ation; + = tcaxin; +ATA = toxin plus aurintr~~rboxyli~ acid; +A = toxin plus ATP; +A0 = toxin plus PLTP and oligomycin.
283
Toxin Alone
I:
KQ: Toxin+ ATP U : Toxin + ATP and
CN/IAA ~~~-~_~
B.
a 1: $
A23187
I:
250
CN/IAA
CN/IAA/'A23187
ToxtnAlone
ES: Cl2
OIigomycir
:
Toxin Toxin
A23187
+ ATP + ATP and
Oligomycin
CN/IAA/A23187
Fig. 3. Effects of ATP and oligomycin on DNA fragmentation and cytotoxicity. Hepatocytes were treated as in Table I for 180 min, with or without 0.5 mM ATP or ATP plus 10 ,BM oiigomycin. (A) Lactate dehydrogenase leakage; (3) DNA fra~entation. Results are mea&SE of four culture plates and are ~pre~n~ti~e of three inde~ndent ex~r~ents. An asterisk indicates the value is si~~~~ntly different from toxin alone. The control lactate dehydrogenase leakage was 37.8i1.5 W/l and the control DNA fragmentation was 10.4+3.4%.
attempted to protect hepatocytes with exogenous ATP. Directly supplying cells with ATP prevented chemical hypoxia-induced toxicity and substantially reduced combination treatment toxicity. ATP did not protect against A23187-induced damage (Fig. 3), again suggesting separate ‘routes to toxic cell death for the two treatments. Oligomycin, which is thought to prevent ATP destruction by uncoupled mitochondria, augmented ATP protection against chemical hypoxia and the combination treatment. Neither ATP nor ATP plus oligomycin affected DNA fra~entation by any treatment in the present studies, or in other experiments in which higher concentrations of toxins caused greater toxicity and more extensive DNA fragmentation (data not shown). In additional studies, fructose alone and fructose plus oligomycin were able
284
TABLE II EFFECTS OF FRUCTOSE AND OLIGOMYCIN TREATMENTS ON LACTATE DEHYDROGENASE RELEASE AND DNA FRAGMENTATION AFTER 180 min OF HEPATOCYTE EXPOSURE TO CN/IAA. A23187, OR COMBINATION TREATMENT Treatment
CNIIAA
Fructose
Oligomycin
LDH release (% control)
DNA fragmentation (% control)
_
-
906 k 38.7* 379 * 45.5* 147 k 25.6
105 k 0.6 111 + 0.5 88 + 1.5
665 k 24.1* 560 + 2.5* 566 ? 36.9*
222 * 1.1* 202 + 1.8* 182 f 2.4*
844 k 25.0* 686 k 10.7* 469 + 14.2*
216 i 2.3* 208 f 2.0* 191 k 1.5*
+ + A231 87
CNiIAA + A23187
f
_ + +
_ _
+ +
_ _
+
+
Hepatocytes were co-treated with 20 mM fructose and/or IO @g/ml oligomycin. Other conditions are as described in Table I. In control cells, lactate dehydrogenase release at 180 min was 36.2 f 1.1 IUil and DNA fragmentation was 11.2 + 2.5%. In cells treated only with fructose or oligomycin, lactate dehydrogenase release was 38.4 k 0.8 and 79.6 k 3.2 IUII, and DNA fragmentation was 9.0 k 0.1% and 16.6 k O.l%, respectively. * Significantly different from control value (P < 0.05).
to mimic the effects of ATP alone and ATP plus oligomycin (Table II). This indicates probable uptake of ATP by cultured mouse hepatocytes. Finally, electrophoresis of DNA confirmed that neither ATP nor ATP plus oligomycin altered the degree of DNA damage produced by A23187 or the combination treatment (Fig. 2, lower panel). DISCUSSION
The chemical hypoxia model of toxic cell death has been used to investigate the potentially lethal effects of ATP depletion [3,30,31]. Cyanide blocks mitochondrial ATP production by inhibiting cytochrome oxidase and iodoacetic acid inhibits glycolytic ATP synthesis by alkylating glyceraldehyde-3-phosphate dehydrogenase. Although the model was developed to mimic the profound and presumably lethal effects of ATP depletion, recent evidence of Ca2’ deregulation has raised questions about the actual mechanism of cell killing in this model. Several groups detect no Ca2’ increase during chemical hypoxia injury to hepatocytes and myocytes [3,30,31]. However, chemical hypoxia-induced killing of lclc7 hepatoma cells is preceded by increased
285
Ca2’. Moreover, the death of lclc7 cells is delayed by Ca” chelators, while ATP supplies remain depleted [24]. The present studies establish that chemical hypoxia did not stimulate Ca*+-endonuclease mediated fragmentation of DNA in cultured mouse hepatocytes before toxic cell death (Table I, Figs. 1 and 2). Consistent with this, the Ca*‘-endonuclease inhibitor aurintricarboxylic acid failed to prevent chemical hypoxia toxicity, but greatly decreased the toxicity of A23187, whether alone or in combination with CN/IAA. Conversely, ATP plus oligomycin had no effect on toxicity of A23 187, yet efficiently protected against chemical hypoxia-induced toxicity. These findings support the conclusion that cultured mouse hepatocytes can be killed by independent mechanisms involving ATP depletion or Ca*’ overload. The greater combined toxicity of A23 187 plus chemical hypoxia suggests that the killing mechanisms are independent and additive in their toxic effects. The contrasting effects of chemical hypoxia on cultured hepatocytes and hepatoma cells may reflect the fact that killing mechanisms are different for transformed vs. normal liver cells. The finding from the present studies that separate killing insults can be independent and additive in cultured hepatocytes indirectly supports this view. Cell preference for one killing mechanism over another could be controlled by several factors. Hepatoma cells possess high glycolytic capacity and are relatively resistant to hypoxic killing, probably because they exhibit increased efficiency of ATP utilization during anaerobiosis [32]. As a result, the toxic effects of CNlIAA on other sites such as ion pumps may be proportionally greater in hepatoma cells than cultured hepatocytes, leading to the predominance of Ca*‘-induced cell killing over killing by ATP depletion. We conclude from the present studies that A23187 and chemical hypoxia kill cultured hepatocytes through distinct mechanisms that can act independently and in concert to produce toxic cell death. ACKNOWLEDGEMENT
This work was supported in part by Grant No. GM-41564 from the National Institute of General Medical Sciences, National Institutes of Health. REFERENCES 1 Zakim, D. and Boyer, T.D. (Eds.) (1990) Hepatology, A Textbook of Liver Disease, 2nd Edn., W.B. Saunders, Philadelphia. 2 Schanne, F.A.X., Kane, A., Young, E. and Farber, J.L. (1979) Calcium dependence of toxic cell death: a final common pathway. Science 206, 700-702. 3 Lemasters, J.J., DiGuiseppi, J., Nieminen, A. and Herman, B. (1987) Blebbing, free Ca*’ and mitochondrial membrane potential preceding cell death in hepatocytes. Nature 325,78-81. 4 Corcoran, G.B. and Ray, S.D. (1992) The role of the nucleus and other compartments in toxic cell death produced by alkylating hepatotoxicants. Toxicol. Appl. Pharmacol. 113, 167-183.
286
5 Fawthrop, 437444.
D.J., Boobis,
6 Nicotera, 3,484494.
P., Bellomo,
A.R. and Davies,
G. and Orrenius,
7 Reed, D.J. (1990) Review Toxicol.
of cell death.
Arch.
Toxicol.
65.
S. (1990) The role of Ca*’ in cell killing. Chem. Res. Toxicol.
of the current
status
of calcium
and thiols in cellular
injury.
Chem.
Res.
3, 495-502.
8 Farber,
J.L. (1990) The role of calcium
9 Jewel1 S., Bellomo, during
D.S. (1991) Mechanisms
G., Thor,
drug metabolism
in lethal cell killing. Chem. Res. Toxicol.
H., Orrenius,
is caused
S. and Smith,
by disturbances
3, 503-508.
M. (1982) Bleb formation
in thiol and calcium
in hepatocytes
ion homeostasis.
Science 217,
1257-1259. 10 Chien
K.R.,
Pfau,
chlorpromazine
R.G.
and
Farber,
of an accelerated
Am. J. Pathol.
with accumulation
of calcium
icity in cultured
J.L. and Corcoran,
cell injury:
acetaminophen-induced
L.M., Ray, S.D. and Corcoran.
with cell death. Toxicol.
icity in cultured tion inhibitors
correlation
prevention
membrane
by
dysfunction. DNA in vivo
liver injury. Toxicol.
Appl. Pharm.
mouse hepatocytes:
Appl.
J.J. and Duke,
lymphocytes
hepatic
16 Wyllie. A.H. (1980) Glucocorticoid-induced activation.
Nature
DNA repair,
Toxicol.
Appl. Pharmacol.
D., Hartzell,
tion kills immature 19 McConkey,
DNA fragmenta-
Effects of Ca”-endonuclease
and poly(ADP-
Exp. Ther. 263, 3877394. activation
of a calcium-dependent
endonuclease
D., Hartzell,
in
132, 3842. thymocyte
apoptosis
is associated
with endogenous
endo-
284, 5555556.
P., Nicotera,
thymocytes.
dibenzo-p-dioxin (TCDD) Science 242.256-258.
deple-
112, 3240.
17 Corcoran, G.B., Ray, S.D., Sorge, C. and Kamendulis, L.M. (1991) DNA fragmentation nuclear Ca” precede dimethylnitrosamine-induced hepatic necrosis in mice. Toxicologist 18 McConkey,
cytotox-
and glutathione
G.B. (1992) Ca”-Activated
necrosis:
R.C. (1984) Glucocorticoid J. Immunol.
cytotox-
and early DNA fragmen-
G.B. (1992) Acetaminophen-induced
L.M. and Corcoran.
in mice. J. Pharmacol.
leads to cell death.
accumulation
111, 2422254.
and cell death.
tion and dimethylnitrosamine-induced inhibitors
Ca”
effects of Ca”-endonuclease,
on DNA fragmentation
ribose) polymerase
G.B. (1991) Acetaminophen-induced
of nuclear
L.M., Ray, S.D. and Corcoran,
14 Ray, S.D., Sorge, C.L., Kamendulis.
nuclease
myocardial and associated
G.B. (1990) Early loss of large genomic
in the nucleus during
mouse hepatocytes:
13 Shen, W., Kamendulis,
15 Cohen,
Ischemic
degradation
106, 346-351.
12 Shen, W.. Kamendulis, tation
(1979)
97. 505-530.
11 Ray, S.D., Sorge, C.L., Raucy, Pharmacol.
J.L.
phospholipid
P. and Orrenius,
FASEB
P., Duddy,
S. (1989) Calcium-activated
and increased 11, 102.
DNA fragmenta-
J., 3. 184331849.
SK..
Hakasson.
kills immature
thymocytes
H. and Orrenius, by calcium
S. (1988) 2,3,7,8-Tetrachloro-
mediated
endonuclease
activation.
20 Lemasters, J.J.. Gores, G.J.. Nieminen, A.-L., Dawson. T.L.. Wray, B.E. and Herman, B. (1990) Multiparameter digitized video microscopy of toxic and hypoxic injury in single cells. Environ. Health Persp. 84, 83-94. 21 Anundi. prevents 22 Nieminen,
I.. King, J., Owen, D.A., Schneider, H.. Lemasters, J.J. and Thurman, hypoxic cell death in liver. Am. J. Physiol. 253, G390-G396. A.-L.,
Dawson,
T.L., Gores,
G.J., Kawanishi,
T., Herman,
R.G. (1987) Fructose
B. and Lemasters.
J.J. (1990)
Protection by acidotic pH and fructose against lethal injury to rat hepatocytes from mitochondrial inhibitors, ionophores, and oxidant chemicals. Biochem. Biophys. Res. Commun. 167, 600-606. 23 Herman, B., Gores, G.J., Nieminen, A.-L., Kawanishi, T., Harman, A. and Lemasters, J.J. (1990) Calcium
and pH in anoxic
and toxic injury.
Crit. Rev. Toxicol.
24 Nicotera, P., Thor, H. and Orrenius, S. (1989) Cytosolic-free cells exposed to chemical anoxia. FASEB J. 3, 59-64.
21, 127-148. Ca*+ and cell killing in hepatoma
25 Paine, A.J., Williams, L.J. and Legg, R.F. (1979) Apparent maintainance nicotinamide in primary cultures of rat hepatocytes. Life Sci. 24, 218552192.
of cytochrome
lclc7
P-450 by
27 Burton, K. (1956) A study of the conditions and mechanism of the diphenyi~ine reaction for the calorimetric ~t~ation of deoxyribonucleic acid. B&hem. J. 62, 315-322. 28 Sambr~k, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning, A Laboratory Manual, 2nd Edn., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 29 Zar, J.H. (1984) Biostatisti~I Analysis, 2nd Edn., ~enti~Hal1, Englewood Cliffs, NJ. 30 Cobbofd, P., Boume, P. and Cuthbertson, KS. (1985). Evidence from aequorin for injury ofmetabohtally inhibited myocytes independently of free Ca*‘. Basic Res. Cardiol. 80, 155458. 31 Allen, D.G. and Orchard, C.H. (1983). Intracellular calcium concentration during hypoxia and metabolic inhibition in mammalian ventricular muscle. J. Physiol. 339, 107-122. 32 Hugo-Wissemann~ D., Anundi, I., Lauchart, W., Viebahn, R. and de Groot, H. (1991) Differences in glycolytic capacity and hypoxia tolerance between hepatoma cells and hepatocytes. Hepatology 13, 297-303.