Phospholipid metabolism and intracellular Ca2+ homeostasis in cultured rat hepatocytes intoxicated with cyanide ISA0 SAKAIDA, ANDREW P. THOMAS, AND JOHN L. FARBER Department of Pathology and Cell Biology, Thomas Jefferson University, Philadelphia, Pennsylvania 19107 Sakaida, Isao, Andrew P. Thomas, and John L. Farber . Phospholipid metabolismand intracellular Ca2+ homeostasisin cultured rat hepatocytesintoxicated with cyanide. Am. J. Physiol. 263 (Cell Physiol. 32): C684-C690, 1992.-The killing of cultured hepatocytes by 1 mM sodium cyanide was reducedby 100 PM chlorpromazine or cytochalasin B (25 pg/ml) or by loweringthe pH of the culture mediumto 6.0. In eachcase, ATP was depleteddespite the decreasednumber of dead cells. The cell killing by cyanide was accompaniedby an accelerated releaseof “H-labeled arachidonate from phospholipids.Depletion of ATP by oligomycin did not acceleratephospholipiddegradation or kill the hepatocytes. Chlorpromazine, cytochalasin B, and extracellular acidosisreduced the rate of phospholipid degradationin control cellsaswell asthe increasethat occurred with cyanide. The calcium ionophore A23187 increasedphospholipid degradation and killed the hepatocytes. Chlorpromazine and extracellular acidosis,but not cytochalasin B, protected the cellsand prevented the increasedlipid degradationin responseto A23187. After addition of cyanide, cytosolic free calcium ( [Ca2+]i) did not changefor 71 k 8 min, at which time it roseto a plateau of 683t 210nM within 10min. A secondand larger rise occurred after 84 t 8 min and beforethe death of the cells at 89 t 8 min. Treatment with 3.5 mM ethylene glycolbis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid, as well as removal of extracellular calcium, prevented theselate increases in [Ca2+]; without affecting the lossof viability. It is concluded that cyanide kills cultured hepatocytesby a mechanismthat is likely related to an accelerateddegradation of phospholipids. This change in lipid metabolismis not mediated by a rise in [Ca2+]i but rather may relate to an alteration in the interaction betweenthe cytoskeleton and the plasmamembrane. chlorpromazine; cytochalasin B; acidosis;furaCELL CULTURE MODELS are increasingly
being used to study the mechanisms by which ischemia-anoxia produces lethal injury. Anoxic rat hepatocytes displayed a similar alteration in phospholipid metabolism (5, 19) to that of the same cells made ischemic in the intact animal (3, 6, 20). Cultured neonatal rat myocardial cells were treated with metabolic inhibitors as a model of the biochemical consequences of ischemia-anoxia (1, 4, 9, 23). Release of arachidonate from phospholipids accompanied sarcolemmal defects, electrolyte derangements, and the loss of viability. Inhibition of phospholipid degradation by a steroidal diamine (U-26384) protected the myocytes (23). The release of arachidonic acid and the accompanying membrane injury were considered direct consequences of the ATP depletion and associated with progressive calcium overloading (1, 9). By contrast, we have argued previously (12, 15) that ATP depletion alone is not sufficient to irreversibly injure cultured rat hepatocytes during the same time course that anoxia or cyanide kills the cells. Treatment of cultured hepatocytes with cyanide and iodoacetate was followed by the appearance of surface C684
0363-6143/92
$2.00
Copyright
0
1992
blebs that accompanied the fall in ATP (7, 11, 13, 18). Again, the formation of these blebs was suggested to be a direct consequence of the depletion of ATP (7). The transition from reversible to irreversible injury correlated with the rupture of the blebs (11,13,18). Cytosolic free calcium ( [Ca2+];) did not change during bleb formation or before the loss of cellular viability (13, 18). By contrast, cytosolic calcium was reported to rapidly rise when a hepatoma cell line was treated with cyanide (16). These previous investigations have left a number of unresolved issues. In particular, it is not established that metabolic inhibitors kill cells by the same mechanism as does ischemia-anoxia. There are also conflicting claims as to whether ATP depletion is sufficient to produce lethal cell injury. Finally, conflicting results exist with respect to the role that alterations in calcium homeostasis play in the various models studied. The present study attempted to address these issues. First, we show that treatment of cultured hepatocytes with cyanide alters phospholipid metabolism in a manner similar to that with anoxia or with liver ischemia in the intact animal. Second, this alteration in lipid metabolism cannot be attributed to the depletion of ATP alone. Finally, the killing of cultured hepatocytes by cyanide is not a result of alterations in intracellular calcium homeostasis. MATERIALS AND METHODS Cultures and agents. Male Sprague-Dawleyrats (150-200 g) were obtained from CharlesRiver Breeding Laboratories (Wilmington, MA). All animalswere fed ad libitum and fasted 24 h before use. Hepatocytes were isolated by collagenase(Sigma) perfusion according to Seglen (22). Yields of 3-5 x lo8 cells/ liver with 90-95% viability by trypan blue exclusion were routinely obtained. The hepatocytes were plated in 25-cm2flasks (Corning GlassWorks, Corning, NY) at a density of 1.33 x lo6 cells/flaskin 3 ml of Williams E medium (GIBCO Laboratories) containing penicillin (10 IU/ml), streptomycin (10 pg/ml), gentamicin (0.5 mg/ml), insulin (0.02 U/ml), and 10% heat-inactivated (55°C for 15 min) fetal calf serum (FCS; Hazleton ResearchProducts, Lenexa, KS; completeWilliams E). After 2 h at 37°C in an atmosphereof 5% CO,-95% air, the cultures were rinsedtwice with a prewarmedN-2-hydroxyethylpiperazine-N’2-ethanesulfonicacid (HEPES; Sigma) buffer, pH 7.4 (0.14 M NaCl, 6.7 mM KCl, 1.2 mM CaCI,, and 2.4 mM HEPES) to remove unattached deadcells. CompleteWilliams E (5 ml) was replaced,and the cellswereincubated for 15-16 h. The cultures were then washedtwice with prewarmed HEPES buffer and incubated in Krebs-Ringer bicarbonate buffer (in mM: 120 NaCl, 4.8 KCl, 1.2KH2P04, 1.2MgSO,, 24 NaHCO,, 1.8 CaCl,, 10 HEPES, and 10 glucose)with the agents describedin the text. Chlorpromazine, cytochalasin B, ionomycin, and A23187 (Sigma) were dissolved in dimethyl sulfoxide (final concn of dimethyl sulfoxide wasalways 0.5%, which had no effect on the the AmericanPhysiological Society
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
toxicity of cyanides). Sodium cyanide (Sigma) was dissolved in 0.9% NaCl. Cell killing was assessed by the release of lactate dehydrogenase into the culture medium as described previously (2). Cellular ATP content was determined by the luciferin-luciferase method as described previously (12). Protein was measured by the method of Lowry et al. (14). Assay of phospholipid degradation. Phospholipid degradation was measured by the release of sH-labeled arachidonic acid as described previously (24, 25). Hepatocytes in culture for 2 h were washed and then placed in complete Williams E medium containing 0.5 &i [“Hlarachidonic acid (135 Ci/mmol, Amersham). After 24 h, the cells were washed twice with a HEPES buffer and incubated for 90 min in a Krebs-Ringer bicarbonate buffer (in mM: 140 NaCl, 4.8 KCl, 1.2 KH2P04, 1.2 MgSO*, 1.8 CaC1,, 24 NaHCO,,, 10 glucose, and 10 HEPES, pH 7.4) with the additions as described in the text. At the end of the incubation, the medium was immediately transferred to microcentrifuge tubes on ice. After a 5min centrifugation at 10,000 g in an Eppendorf 5415 centrifuge (Beckman) at 4”C, the radioactivity of supernatant was determined by liquid scintillation counting. The radioactivity released into the culture medium reflects the degradation of prelabeled phospholipids (25) and represents free arachidonic acid (70%) plus metabolites. There was no change in the total radioactivity present as free fatty acids in the hepatocytes under any of the conditions studied. Radioactivity released from the hepatocytes is expressed as a percent of total radioactivity in the hepatocytes at the beginning of the assay. Measurement of [Ca2+li. [Ca*+]i was measured by digital imaging fluorescence microscopy of hepatocytes loaded with fura- (21). The hepatocytes were plated on glass cover slips (VWR Scientific) coated with poly-D-lysine (5 pg/cm*) at a density of 2 x lo5 cells/Petri dish in 3 ml of Williams E medium (complete). The cells were incubated for 2 h, washed twice with prewarmed HEPES buffer, and incubated overnight in complete Williams medium. The hepatocytes were again washed twice and then placed in a modified Krebs-Ringer bicarbonate buffer (in mM: 10 HEPES, 121 NaCl, 4.7 KCl, 1.2 KH2P04, 1.2 MgSO,, 2.0 CaC12, 5 NaHC03, and 10 glucose and 2% FCS, pH 7.4) containing 1.6 PM fura-2/AM in 0.03% Pluronic F-127 (Molecular Probes) at 37°C for 30 min. The cover slips were washed three times with the same buffer and transferred to a chamber with 1 ml of the modified Krebs-Ringer buffer (no FCS) and mounted on the stage of a Zeiss IM-35 inverted microscope. The stage, ~16 oil-immersion objective, and chamber were thermostatically regulated at 37°C. Alternatively, the cells were loaded with fura- in the initial modified Krebs-Ringer buffer containing 3.5 mM ethylene glycol-bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA) for 30 min, washed with the same buffer, and treated with cyanide in the presence of EGTA. In the experiments with a low-calcium buffer (~2 PM total calcium), the cells were incubated with fura-2/AM for 30 min in modified Krebs-Ringer buffer that had been passed through a Chelex 100 (200-400 mesh) column (Bio-Rad). A liquid N,-cooled charge-coupled device camera (Photometrics) was used as the imaging device. Pairs of images were taken every 40 s over 90 min and digitized at 12-bit resolution, stored, and analyzed with a Heurikon HK68/M 10 computer. Fluorescence images were obtained at excitation wavelengths of 340 and 380 nm (lo-nm bandwidth) with an emission wavelength of 460-580 nm. The integration time for each image was 400 ms, and individual pixels were binned into 3 x 3 superpixels at readout from the charge-coupled device detector to improve signal-to-noise ratio. The time required for wavelength changes was 1 s. To minimize photobleaching, a computer-controlled
C685
HYPOXIA
shutter was used to limit the exposure of the cells to excitation light. The mean fluorescence intensity of individual cells was determined by selecting a region of the image covering each cell. The mean intensity over the same area for each of the successive images in a time series was measured to create a time course of fluorescence changes. At the end of each experiment, cells were exposed to ionomycin (20 PM) and MnCl, (2 mM) for 30 min. The residual fluorescence was measured over the same region of each cell as the Ca*+ -dependent fluorescence changes and subtracted from the fura- Ca*+ signals for that cell at the appropriate wavelength. Calibration of fura- fluorescence in terms of [Ca’+]i was calculated from the ratio of 340- to 380-nm fluorescence values (after subtraction of background fluorescence) as described by Grynkiewicz et al. (8). For each experiment, values for Rmin and R,,, were determined using solutions of fura-2free acid in the same microscope chamber used for the cell incubations. The dissociation constant for the fura-2-Ca*+ was taken as 224 nM (8). Statistical analysis. Analyses of variance using a repeatedmeasures design were performed on the data using the PC version of the SAS statistical package (version 6.04). RESULTS
Prevention of toxicity of cyanide by chlorpromazine, cytochalasin B, and extracellular acidosis. Table 1 docu-
ments that a number of manipulations reduced the killing of cultured hepatocytes by cyanide without sparing the depletion of ATP that occurs with this inhibitor of cytochrome oxidase. Chlorpromazine was shown previously to prevent the degradation of phospholipid that accompanies liver ischemia in the intact rat (3). Similarly, cytochalasin B was shown previously to reduce the killing of cultured hepatocytes by anoxia (19). Lastly, reducing the pH of the culture medium to 6.0 protects the hepatocytes (7, 10, 13, 17). Accelerated phospholipid degradation in cyanide-intoxicated hepatocytes. Hepatocytes were labeled with [3H] arachidonic acid during their first 24 h in culture. The
cells were then washed, resuspended in a Krebs-Ringer bicarbonate buffer, and treated with 1 mM cyanide. Table 1. Chlorpromazine, cytochalasin B, and extracellular acidosis do not prevent depletion of ATP induced by cyanide Cell
Death, %
ATP, nmol/mg protein
No additions 3tl 11.8OkO.28 Cyanide alone 65t5 0.65kO.25 Chlorpromazine alone 5+1 Chlorpromazine + cyanide 22t4 0.58t0.3 1 Cytochalasin B alone 4+1 Cyanide + cytochalasin B 25t2 0.61kO.21 pH 6.0 alone 5&l Cyanide + pH 6.0 5&l 0.46kO.23 Oligomycin 6t2 0.61kO.18 Values are means t SD; n = 3 separate cultures. Cultured hepatocytes were treated with 1 mM cyanide in presence or absence of 100 pM chlorpromazine or cytochalasin B (25 pg/ml). Cells treated with oligomycin received 0.10 pg/ml. Alternatively, cells were exposed to cyanide in culture medium at pH 6.0. Controls received no additions, chlorpromazine alone, or cytochalasin B alone or were exposed to culture medium at pH 6.0. After 15 min, cellular content of ATP was measured. Viability of cells was determined after 90 min.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
C686
LIPID METABOLISM
WITH
Phospholipid degradation was assessed by the release of arachidonic acid into the culture medium and was expressed as a percent of the total [3H]arachidonic acid incorporated into phospholipids. Table 2 indicates that in the absence of cyanide, the hepatocytes released -3% of total [3H]arachidonic acid incorporated into phospholipids within 90 min. With 1 mM cyanide, approximately twice as much [3H]arachidonic acid was released. Table 2 further details the effect of chlorpromazine, cytochalasin B, and extracellular acidosis on the extent of phospholipid degradation in cyanide-intoxicated hepatocytes. Chlorpromazine, cytochalasin B, and extracellular acidosis lowered the basal rate of phospholipid degradation and prevented the increase produced by cyanide. The accelerated phospholipid degradation in cyanideintoxicated hepatocytes is not a simple consequence of the depletion of ATP. Oligomycin reduced the ATP content to the same extent as did cyanide (Table 1) but did not affect the viability (Table 1) or lipid hydrolysis (Table 2) over a similar 1.5 h of incubation. Activation of phospholipid hydrolysis by the calcium ionophore A23167. The [Ca’+]; of cultured hepatocytes
can be abruptly increased and the cells killed by adding the calcium ionophore A23187 (24, 25). Table 3 shows that 15 PM A23187 increased the rate of phospholipid hydrolysis in parallel with the loss of viability of 70% of the cells within 90 min. Omission of calcium from the culture medium prevented both the cell killing and the increased rate of phospholipid degradation (data not shown). Interestingly, cytochalasin B did not protect the hepatocytes from the toxicity of A23187 and did not prevent the accelerated phospholipid degradation, despite a lowering of the basal rate of phospholipid degradation (Table 3). By contrast, chlorpromazine and extracellular acidosis lowered the basal rate of phospholipid degradation, prevented the increased degradation produced by A23187, and reduced the extent of the cell killing (Table 3). [ Ca2+]i in cyanide-intoxicated hepatocytes. The effect of cyanide on [Ca2+]; was studied at the single-cell level
CHEMICAL
HYPOXIA
Table 3. Phospholipid degradation and cell killing by A23187 Phospholipid
Degradation,
%
Cell
Death,
%
No additions 3.18k0.12 5t1 A23187 6.31t0.18 71t4 +Cytochalasin B 5.89kO.35 67k4 +Chlorpromazine 2.15t0.38 24k4 +pH 6.0 3.59k0.28 8t3 Cytochalasin B alone 1.88kO.08 4tl Chlorpromazine alone 2.45k0.46 11&l pH 6.0 alone 2.63k0.21 5tl Values are means t SD; n = 3 separate cultures. Cultured hepatocytes were treated with cytochalasin (25 pg/ml) B or 100 PM chlorpromazine with or without 15 PM A23187. Alternatively, cells were treated with A23187 in culture medium at pH 6.0. Controls received no additions, cytochalasin B alone, or chlorpromazine alone or were exposed to culture medium at pH 6.0. After 90 min, extent of phospholipid degradation and loss of cell viability of cells were measured.
by using digital imaging fluorescence microscopy. The ability of this methodology to detect changes in [Ca2+]i under the present conditions is documented in Fig. 1. The time course of the changes in fluorescence intensity is illustrated at both 340 and 380 nm after treatment of a
r
Table 2. Phospholipid degradation and prevention of toxicity of cyanide Phospholipid
Hydrolysis,
%
3.17t0.13 No additions Cyanide 5.55t0.60* Chlorpromazine 2.10t0.20* Cyanide + chlorpromazine 1.90+0.lO*jCytochalasin B 1.74t0.18* Cyanide + cytochalasin B 1.83+0.ll*t pH 6.0 1.19t0.13* Cyanide + pH 6.0 1.80k0.28*t 3.19kO.22 Oligomycin Values are means k SD; n = 3 separate cultures. Cultured hepatocytes were treated with 1 mM cyanide in presence or absence of 100 PM chlorpromazine or cytochalasin B (25 pg/ml). Alternatively, the cells were exposed to cyanide in culture medium at pH 6.0. Controls received no additions, chlorpromazine alone, or cytochalasin B alone or were exposed to culture medium at pH 6.0. Phospholipid degradation was measured after 90 min. * Significantly different from control cells (no additions) at P < 0.01. t Not significantly different from control of respective treatment alone (no cyanide).
.-” I.OC 1
s:
E 0
0.5 u 0’
0
I
I 4
I
I 8
I
I 12
1
I 18
Time (mid Fig. 1. Fluorescence changes and calibrated cytosolic free Ca*+ concentration ([Ca”‘] J response in a single hepatocyte treated with ionomytin. Representative tracings of fluorescence signals and calculated [Ca*+]i increase observed in 1 fura-2-loaded hepatocyte after addition of 15 PM ionomycin are shown. Fluorescence values were measured from a sequential series of images as described in MATERIALS AND METHODS. Top: time courses of fluorescence changes observed at excitation wavelengths of 340 and 380 nm. Integration time for each image was 400 ms, with a l-s delay between wavelength changes. Bottom: same data calibrated in terms of [Ca*+]i calculated from ratio of 340- to 380-nm fluorescence values as described in MATERIALS AND METHODS. Arrow, time of addition of ionomycin.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
fura-2-loaded hepatocyte with 15 PM ionomycin, a calcium ionophore (Fig 1, top). Immediately upon addition of ionomycin, the 340-nm fluorescence (calcium-bound fura-2) increased and the 380-nm fluorescence (free fura-2) decreased. Over the ensuing 6-8 min, fura- fluorescence changed only slightly. The rapid loss of fluorescence at both 340 and 380 nm after -9 min reflected the loss of fura- from the cell, an effect signifying the death of the cell. There was a calculated rise in [Ca2+]i in response to ionomycin (Fig. 1, bottom). The calcium concentration rose rapidly within seconds and then increased slowly up to the point of the death of the cell. Table 4 summarizes the changes in [Ca2+]; induced by ionomycin in the nine hepatocytes studied. An almost eightfold rise in [Ca2+]; occurred within 20 s of adding the ionomycin. Figure 2 illustrates the time course of the effect of 1 mM cyanide on [ Ca2+]; in a representative fura-2-loaded hepatocyte. For >70 min after addition of cyanide, there was no change in [Ca2+]i. After -70 min, there was a rise in [Ca2+]i that reached a plateau within 10 min. This late rise was observed in all 43 cells studied that died and began at 71.4 t 7.7 min to reach a plateau value of 683 t 210 nM. Interestingly, this plateau concentration of calcium is close to the maximal level generally induced by hormones in these cells (21). Shortly after the plateau concentration was reached, there was an abrupt increase in [Ca’+]; (Fig. 2). This second rise was also seen in all 43 cells studied that died and occurred 83.5 t 8.0 min after the addition of cyanide. After this second rise in [Ca2+];, fura- fluorescence at both 340 and 380 nm (Fig. 2, top) was rapidly lost as the cells died. Table 5 summarizes the [Ca2+]i 45 min after exposure of hepatocytes to 1 mM cyanide. Cyanide did not produce a significant increase in [Ca”+]; after 45 min in the 43 cells studied. To examine the effects of cyanide on cellular autofluorescence, experiments were carried out using cells not loaded with fura-2. Under these conditions, the hepatocytes did not show any significant changes of fluorescence at either 340 or 380 nm upon addition of 1 mM cyanide (data not shown). [Ca2+]i in hepatocytes treated with cyanide in presence
of EGTA. To explore the relevance of the late alterations in intracellular calcium homeostasis to the cell killing, the cells were incubated with cyanide in the presence of 3.5 mM EGTA. Figure 3 illustrates that in the presence of EGTA there were no late increases in [Ca2+]i. Thus neither the delayed plateau response nor the later abrupt Table 4. Calcium homeostasisin hepatocytes treated with 15 PM ionomycin [ Ca’+]i,
nM T peak)
Time
0
S
Peak
206k55 1,573+505 21t7 Values are means k SD; n = 9 individual hepatocytes. hepatocytes were loaded with fura-2, and cytosolic free Ca2+ was determined for individual cells from ratio images of furacence. Tpeak, time to half height of abrupt rise in [Ca”+]i after of ionomycin.
Cultured ([Ca’+]i) fluoresaddition
700
s
HYPOXIA
C687
r
100
3
25.
;ts
2.0
8 1.5 t .-0 I.0 2 co 0.5 0” 0 0
20
40 60 80 100 Time (min) Fig. 2. Effect of cyanide on fluorescence changes and calibrated [Ca2+li response in a single hepatocyte. Representative tracings of fluorescence signals (top) and calculated [Ca2+]i (bottom) observed in 1 fura-2-loaded hepatocyte after addition of 1 mM cyanide are shown. Experiment was carried out as for Fig. 1.
Table 5. Calcium homeostasis in hepatocytes intoxicated with cyanide [Ca2+];,
nM
n Time
0
45 min
Plateau
T dead7 min
KCN 43 199t54 217t66 683t210 89t8 KCN + EGTA 19 88t46 90t58 88tl KCN (low Ca2+) 9 166t38 217t56 85k17 Values are means t SD; n, no. of separate measurements made on individual hepatocytes. Cultured hepatocytes were loaded with fura-2, and [Ca2+]i was determined for individual cells from ratio images of fura- fluorescence at time 0,45 min after treatment of cells with 1 mM cyanide, and at plateau. T dead,time to half point of rapid loss of fluorescence intensity at 340 nm accompanying death of cells.
increase seen in Fig. 2 was observed. In most cells, [Ca2+]i actually decreased just before the loss of viability (Fig. 3). Interestingly, this decrease in the presence of EGTA occurred at a time similar to that at which the large Ca2+ increase occurred in physiological Ca2+ medium (Fig. 2). Table 5 documents that EGTA lowered the basal level of cytosolic calcium and the [Ca2+]; 45 min after treatment with cyanide. However, there was no plateau phase (Table 5 and Fig. 3), and the time to the death of the cells was the same as in the presence of extracellular calcium (Table 5). EGTA had no effect on phospholipid degradation or on the extent of cell killing by cyanide (Table 6). Similar results were obtained by simply treating the hepatocytes with 1 mM cyanide in a low-calcium buffer (~2 PM free Ca2+). Table 5 indicates that the basal level
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
r
WITH CHEMICAL
HYPOXIA
DISCUSSION
The data presented in this report identify a number of conditions that substantially reduced the toxicity of cyanide for cultured rat hepatocytes. In turn, these conditions were used to assess the relationship between the cell killing by cyanide and an accompanying accelerated degradation of phospholipids. Finally, digital imaging fluorescence microscopy was used to establish that an elevation of [Ca2+]i is not the cause of the enhanced lipid degradation. Cyanide inhibits cytochrome oxidase and thereby prevents oxygen reduction. Continued electron transport and thus mitochondrial energization is prevented. The mitochondrial membrane potential is lost (15), and ATP stores are rapidly depleted. In a manner that has not been clearly defined, these initial consequences of cyanide intoxication are followed by a loss of the permeability barrier function of the plasma membrane and the death of the cell. The nature of the events coupling the loss of mitochondrial function to the disruption of plasma mem0 brane integrity was the focus of the present study. 0’ 0 20 40 60 80 100 The conditions that were found to reduce the extent of cell killing by cyanide most likely act to modify those Time (mid events that couple the loss of mitochondrial energization Fig. 3. Effect of EGTA on fluorescence changes and calibrated [Ca2+]i to the death of the cells. With chlorpromazine, cytocharesponse in a single cultured hepatocyte treated with cyanide. Overlasin B, and extracellular acidosis, ATP was depleted (Tanight-cultured hepatocytes were washed and loaded with fura- in modified Krebs-Ringer bicarbonate buffer containing 3.5 mM EGTA for 30 ble 1) despite the reduction in the number of cells that min. After cells were washed, they were placed in the same buffer died (Table 1). Thus they must act independently of the containing 3.5 mM EGTA. Representative tracings of fluorescence sigloss of the ATP-generating capacity of the mitochondria. nals (top) and calculated [Ca”] i (bottom) observed in a single fura-2The killing of the cultured hepatocytes by cyanide was loaded hepatocyte after addition of 1 mM cyanide are shown. accompanied by an accelerated release of [3H]arachidonate from cellular phospholipids (Table 2). Chlorpromazine, cytochalasin B, and extracellular acidosis reduced Table 6. Phospholipid hydrolysis and cell killing this phospholipid degradation (Table 2) in parallel with by cyanide in presence or absence their preservation of the viability of the hepatocytes (Taof extracellular calcium ble 1). These data would suggest that phospholipid degradaPhospholipid Cell Death, % Hydrolysis, % tion is related to the loss of membrane integrity that determines the fate of the cells. Favoring such an interControl cells + Ca2+ 3.04t0.04 5tl pretation is the fact that chlorpromazine, cytochalasin B, Cyanide + Ca2+ 5.52t0.25 63t5 and extracellular acidosis affected the rate of lipid degraControl cells + EGTA 3.07Iko.10 8t3 74k7 Cyanide + EGTA 5.55t0.35 dation in the control cells. In other words, the protective 4tl Control cells - Ca2+ 3.02t0.14 action of chlorpromazine, cytochalasin B, and extracell66t6 Cyanide - Ca2+ 5.64t0.23 ular acidosis can be directly related to their ability to Values are means t SD; IZ = 3 separate cultures. Cultured hepatoaffect phospholipid degradation and thus to inhibit its cytes were treated with 1 mM cyanide in a Krebs-Ringer bicarbonate stimulation by cyanide intoxication. buffer containing no calcium or 1.8 mM CaC12. Other cells were preChlorpromazine lowered the basal rate of phospholipid treated with 1 mM EGTA in a calcium-free Krebs-Ringer buffer for 45 min before addition of 1 mM cyanide. After 90 min, phospholipid hy- degradation and prevented the increase occurring with drolysis and loss of cell viability were measured. cyanide (Table 2). Because phospholipase A activities have an alkaline pH optimum, by acidifying the cytosol (7)) extracellular acidosis would inhibit phospholipid degof cytosolic calcium was reduced somewhat in the low- radation, an effect that can account for both the lowered basal rate and the inhibition of the increase seen with calcium medium (P < 0.05). However, 45 min after treatcyanide (Table 2). ment with 1 mM cyanide in the low-calcium buffer, Cytochalasin B similarly lowered the basal rate of lipid [ Ca2+]; was not significantly different from that in 1.8 mM calcium. Furthermore, the time to the death of the degradation and prevented the increase with cyanide. There is an intimate association between the cytoskelecells was not changed in the low-calcium medium. Table 6 documents that the low-calcium medium did not affect ton and the plasma membrane of the hepatocyte. It is of actin the increased phospholipid degradation produced by cy- suspected that as a result of the depolymerization microfilaments, cytochalasin B inhibits an interaction anide nor the extent of cell killing within 90 min.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
between the cytoskeleton and the plasma membrane that determines, at least in part, the basal rate of lipid turnover. Interestingly, such an hypothesis implies that cyanide intoxication may activate phospholipid degradation by an effect on the cy-toskeleton and its interaction with cellular membranes. The enhanced phospholipid degradation in hepatocytes intoxicated with cyanide cannot be attributed to a rise in [Ca’+]i. An elevated [Ca”+]i can induce phospholipid degradation (24,25), and raising the calcium content of the cultured hepatocytes by treating with the calcium ionophore A23187 increased the release of [3H]arachidonate and killed the cells (Table 3). Chlorpromazine and extracellular acidosis but not cytochalasin B prevented the increased phospholipid hydrolysis occurring with A23187 intoxication and protected the cells. These data distinguish the toxicity of ionomycin from that of cyanide and suggest that an elevated [Ca2+]i is not the mechanism of the increased lipid degradation associated with cyanide intoxication. Such a conclusion was confirmed by directly measuring the effect of cyanide on the [Ca2+]; of cultured hepatocytes. There was no change in [Ca2+]; in cyanide-intoxicated hepatocytes until just before the cells died (Fig. 1 and Table 5), a result that confirms a similar finding reported previously (11, 13, 16). The late increases in [Ca2+]i most likely represent an influx of calcium ions across an injured plasma membrane. Removal of extracellular calcium ions prevented the late increases in calcium without affecting the loss of viability (Tables 5 and 6). Cytosolic calcium was reduced by >50% by treating the cells with 3.5 mM EGTA. There was again no rise in [Ca”+]; with cyanide and no effect on the extent of cell killing (Fig. 3 and Tables 5 and 6). With EGTA there was actually a late decrease in [Ca2+]i just before the death of the cells. This most likely represents an efflux of calcium across a damaged plasma membrane from a higher cytosolic to a lower extracellular calcium ion concentration. In the presence of a physiological extracellular calcium concentration, the gradient is in the reverse direction, and calcium enters the cells. In either case, the late changes in [Ca2+]; do not affect the fate of the cells. Finally, the data here and those published previously (11, 13, 18) differ from the reported effects of cyanide on [Ca2+]i in a hepatoma cell line (16). Upon addition of cyanide, [Ca2+]i started to rise and reached a plateau sevenfold higher than the basal level within 10 min. The discrepancy here in the effect of cyanide most likely reflects the fact that the hepatoma cell line studied is no longer an hepatocyte. One is comparing the effect of cyanide on two very different kinds of cells. It may be difficult to extrapolate conclusions with respect to the behavior of the normal hepatocyte from studying a malignant variant. This work was supported by National Institutes of Health Grants DK-38305, DK-38422, and AA-07186. Present address of I. Sakaida: First Department of Internal Medicine, Yamaguchi University School of Medicine, Ube 755, Japan. Address for reprint requests: J. L. Farber, Dept. of Pathology, Room 251, Jeff Alumni Hall, Thomas Jefferson Univ., Philadelphia, PA 19107.
C689
HYPOXIA
Received 22 October 1991; accepted in final form 21 April 1992. REFERENCES 1. Buja, L. M., H. K. Hagler, D. Parsons, K. R. Chien, R. C. Reynolds, and J. T. Willerson. Alterations of ultrastructure
and elemental composition in cultured neonatal rat cardiac myocytes after metabolic inhibition with iodoacetic acid. Lab. Inuest. 53: 397-412, 1985. 2. Casini, A., M. Giorli, and J. L. Farber.
R. J. Hyland,
A. Serroni,
D. Gilfor,
Mechanisms of cell injury in the killing of cultured hepatocytes by bromobenzene. J. Biol. Chem. 257: 67216728, 1982.
3. Chien, Farber.
K. R., J. Abrams,
A. Serroni,
J. T. Martin,
and J. L.
Accelerated phospholipid degradation and associated membrane dysfunction in irreversible liver cell injury. J. Biol. Chem. 253: 4809-4817, 1978.
4. Chien, K. R., A. Sen, D. Gunn, L. M. Buja,
R. Reynolds, A. Chang, and J. T. Willerson.
Y. M. Kim,
M.
Release of arachidonate from membrane phospholipids in cultured neonatal rat myocardial cells during adenosine triphosphate depletion. Correlation with the progression of cell injury. J. Clin. Inuest. 75: l7701780, 1985. 5. Farber, J. L., and E. E. Young. Accelerated phospholipid degradation in anoxic rat hepatocytes. Arch. Biochem. Biophys. 211: 312-320, 1981. 6. Finkelstein, S. D., D. Gilfor, and J. L. Farber. Alterations in the metabolism of lipids in ischemia of the liver and kidney. J. Lipid Res. 26: 726-734, 1985. 7. Gores, G. J., A.-L. J. J. Lemasters.
Nieminen,
B. E. Wray,
B. Herman,
and
Intracellular pH during “chemical hypoxia” in cultured rat hepatocytes. Protection by intracellular acidosis against the onset of cell death. J. Clin. Invest. 83: 386-396, 1989. 8. Grynkiewicz, G., M. Poenie, and R. Y. Tsien. A new generation of Ca*+ indicators with greatly improved fluorescent properties. J. BioZ. Chem. 264: 3440-3450, 1985. 9. Gunn, Buja,
M. and
D., A. Sen, Y. M. Kim, J. T. Willerson, K. R. Chien. Mechanisms of accumulation
L. M.
of arachionic acid in cultured myocardial cells during ATP depletion. Am. J. Physiol. 249 (Heart Circ. Physiol. 18): H1188-H1194, 1985. 10. Harrison, D. C., J. J. Lemasters, and B. Herman. A pHdependent phospholipase A2 contributes to loss of plasma membrane integrity during chemical hypoxia in rat hepatocytes. Bichem. Biophys. Res. Commun. 174: 654-659, 1991. 11. Herman, B., A.-L. Nieminen, G. J. Gores, and J. J. Lemasters. Irreversible injury in anoxic hepatocytes precipitated by an abrupt increase in plasma membrane permeability. FASEB J. 2: 146-151,1988. 12. Kane, A. B.,
D. Petrovich,
R. 0.
Stern,
and
J. L. Farber.
ATP depletion and the loss of cell integrity in anoxic hepatocytes and silica-treated P388Dl macrophages. Am. J. Physiol. 249 (Cell Physiol. 18): C256C266, 1985. 13. Lemasters, Herman.
J. J.,
J. DiGuiseppi,
A.-L.
Nieminen,
and
B.
Blebbing, free Ca*+ and mitochondrial membrane potential preceding cell death in hepatocytes. Nature Lond. 325: 78-81, 1987.
14. Lowry, Randall.
0.
15. Masaki,
N.,
H.,
N.
J.
Rosebrough,
A.
L.
Farr,
and
R. J.
Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275, 1951. A. P. Thomas,
J. B. Hoek,
and
J. L. Farber.
Intracellular acidosis protects cultured hepatocytes from the toxic consequences of a loss of mitochondrial energization. Arch. Biothem. Biophys. 272: 152-161, 1989. 16. Nicotera, P., H. Thor, and S. Orrenius. Cytosolic-free Ca*+ and cell killing in hepatoma lclc7 cells exposed to chemical anoxia. FASEB J. 3: 59-64, 1989. 17. Nieminen, B. Herman,
A.-L., and
T. L. Dawson, J. J. Lemasters.
G. J. Gores,
T. Kawanishi,
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, 1990.
18. Nieminen, A.-L., J. J. Lemasters.
G. J. Gores,
B. E. Wray,
B. Herman,
and
Calcium dependence of bleb fomration and cell death in hepatocytes. Cell Calcium 9: 237-248, 1989.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
C690
LIPID METABOLISM
WITH CHEMICAL
19. Okayasu, T., M. T. Curtis, and J. L. Farber. Cytochalasin delays but does not prevent cell death from anoxia. Am. J. Pathol. 117: 163-167,
1984.
Petrovich, D. R., S. Finkelstein, A. J. Waring, and J. L. Farber. Liver ischemia increases the molecular order of microsoma1 membranes by increasing the cholesterol-to-phospholipid ratio. J. Biol. Chem. 259: 13217-13223, 1984. 21. Rooney, T. A., E. J. Sass, and A. P. Thomas. Characterization of cytosolic calcium oscillations induced by phenylephrine and vasopressin in single fura-2-locaded hepatocytes. J. Biol. Chem. 264: 17131-17141, 1989. 22. Seglen, P. 0. Preparation of isolated rat liver cells. Methods Cell Biol. 13: 29-83, 1976.
20.
HYPOXIA
23. Sen, A., J. C. Miller, R. Reynolds, J. T. Willerson, L. M. Buja, and K. R. Chien. Inhibition of the release of arachidonic acid prevents the development of sarcolemmal membrane defects in cultured rat myocardial cells during adenosine triphosphate depletion. J. Clin. Invest. 82: 1333-1338, 1988. 24. Shier, W. T. Serum stimulation of phospholipase A2 and prostaglandin release in 3T3 cells is associated with platelet-derived growth-promoting activity. Proc. Natl. Acad. Sci. USA 77: 137141, 1980.
25. Shier, W. T., and D. J. DuBourdieu. Role of phospholipid hydrolysis in the mechanism of toxic cell death by calcium and ionophore A23187. Biochem. Biophys. Res. Commun. 109: 106112, 1982.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
being used to study the mechanisms by which ischemia-anoxia produces lethal injury. Anoxic rat hepatocytes displayed a similar alteration in phospholipid metabolism (5, 19) to that of the same cells made ischemic in the intact animal (3, 6, 20). Cultured neonatal rat myocardial cells were treated with metabolic inhibitors as a model of the biochemical consequences of ischemia-anoxia (1, 4, 9, 23). Release of arachidonate from phospholipids accompanied sarcolemmal defects, electrolyte derangements, and the loss of viability. Inhibition of phospholipid degradation by a steroidal diamine (U-26384) protected the myocytes (23). The release of arachidonic acid and the accompanying membrane injury were considered direct consequences of the ATP depletion and associated with progressive calcium overloading (1, 9). By contrast, we have argued previously (12, 15) that ATP depletion alone is not sufficient to irreversibly injure cultured rat hepatocytes during the same time course that anoxia or cyanide kills the cells. Treatment of cultured hepatocytes with cyanide and iodoacetate was followed by the appearance of surface C684
0363-6143/92
$2.00
Copyright
0
1992
blebs that accompanied the fall in ATP (7, 11, 13, 18). Again, the formation of these blebs was suggested to be a direct consequence of the depletion of ATP (7). The transition from reversible to irreversible injury correlated with the rupture of the blebs (11,13,18). Cytosolic free calcium ( [Ca2+];) did not change during bleb formation or before the loss of cellular viability (13, 18). By contrast, cytosolic calcium was reported to rapidly rise when a hepatoma cell line was treated with cyanide (16). These previous investigations have left a number of unresolved issues. In particular, it is not established that metabolic inhibitors kill cells by the same mechanism as does ischemia-anoxia. There are also conflicting claims as to whether ATP depletion is sufficient to produce lethal cell injury. Finally, conflicting results exist with respect to the role that alterations in calcium homeostasis play in the various models studied. The present study attempted to address these issues. First, we show that treatment of cultured hepatocytes with cyanide alters phospholipid metabolism in a manner similar to that with anoxia or with liver ischemia in the intact animal. Second, this alteration in lipid metabolism cannot be attributed to the depletion of ATP alone. Finally, the killing of cultured hepatocytes by cyanide is not a result of alterations in intracellular calcium homeostasis. MATERIALS AND METHODS Cultures and agents. Male Sprague-Dawleyrats (150-200 g) were obtained from CharlesRiver Breeding Laboratories (Wilmington, MA). All animalswere fed ad libitum and fasted 24 h before use. Hepatocytes were isolated by collagenase(Sigma) perfusion according to Seglen (22). Yields of 3-5 x lo8 cells/ liver with 90-95% viability by trypan blue exclusion were routinely obtained. The hepatocytes were plated in 25-cm2flasks (Corning GlassWorks, Corning, NY) at a density of 1.33 x lo6 cells/flaskin 3 ml of Williams E medium (GIBCO Laboratories) containing penicillin (10 IU/ml), streptomycin (10 pg/ml), gentamicin (0.5 mg/ml), insulin (0.02 U/ml), and 10% heat-inactivated (55°C for 15 min) fetal calf serum (FCS; Hazleton ResearchProducts, Lenexa, KS; completeWilliams E). After 2 h at 37°C in an atmosphereof 5% CO,-95% air, the cultures were rinsedtwice with a prewarmedN-2-hydroxyethylpiperazine-N’2-ethanesulfonicacid (HEPES; Sigma) buffer, pH 7.4 (0.14 M NaCl, 6.7 mM KCl, 1.2 mM CaCI,, and 2.4 mM HEPES) to remove unattached deadcells. CompleteWilliams E (5 ml) was replaced,and the cellswereincubated for 15-16 h. The cultures were then washedtwice with prewarmed HEPES buffer and incubated in Krebs-Ringer bicarbonate buffer (in mM: 120 NaCl, 4.8 KCl, 1.2KH2P04, 1.2MgSO,, 24 NaHCO,, 1.8 CaCl,, 10 HEPES, and 10 glucose)with the agents describedin the text. Chlorpromazine, cytochalasin B, ionomycin, and A23187 (Sigma) were dissolved in dimethyl sulfoxide (final concn of dimethyl sulfoxide wasalways 0.5%, which had no effect on the the AmericanPhysiological Society
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
toxicity of cyanides). Sodium cyanide (Sigma) was dissolved in 0.9% NaCl. Cell killing was assessed by the release of lactate dehydrogenase into the culture medium as described previously (2). Cellular ATP content was determined by the luciferin-luciferase method as described previously (12). Protein was measured by the method of Lowry et al. (14). Assay of phospholipid degradation. Phospholipid degradation was measured by the release of sH-labeled arachidonic acid as described previously (24, 25). Hepatocytes in culture for 2 h were washed and then placed in complete Williams E medium containing 0.5 &i [“Hlarachidonic acid (135 Ci/mmol, Amersham). After 24 h, the cells were washed twice with a HEPES buffer and incubated for 90 min in a Krebs-Ringer bicarbonate buffer (in mM: 140 NaCl, 4.8 KCl, 1.2 KH2P04, 1.2 MgSO*, 1.8 CaC1,, 24 NaHCO,,, 10 glucose, and 10 HEPES, pH 7.4) with the additions as described in the text. At the end of the incubation, the medium was immediately transferred to microcentrifuge tubes on ice. After a 5min centrifugation at 10,000 g in an Eppendorf 5415 centrifuge (Beckman) at 4”C, the radioactivity of supernatant was determined by liquid scintillation counting. The radioactivity released into the culture medium reflects the degradation of prelabeled phospholipids (25) and represents free arachidonic acid (70%) plus metabolites. There was no change in the total radioactivity present as free fatty acids in the hepatocytes under any of the conditions studied. Radioactivity released from the hepatocytes is expressed as a percent of total radioactivity in the hepatocytes at the beginning of the assay. Measurement of [Ca2+li. [Ca*+]i was measured by digital imaging fluorescence microscopy of hepatocytes loaded with fura- (21). The hepatocytes were plated on glass cover slips (VWR Scientific) coated with poly-D-lysine (5 pg/cm*) at a density of 2 x lo5 cells/Petri dish in 3 ml of Williams E medium (complete). The cells were incubated for 2 h, washed twice with prewarmed HEPES buffer, and incubated overnight in complete Williams medium. The hepatocytes were again washed twice and then placed in a modified Krebs-Ringer bicarbonate buffer (in mM: 10 HEPES, 121 NaCl, 4.7 KCl, 1.2 KH2P04, 1.2 MgSO,, 2.0 CaC12, 5 NaHC03, and 10 glucose and 2% FCS, pH 7.4) containing 1.6 PM fura-2/AM in 0.03% Pluronic F-127 (Molecular Probes) at 37°C for 30 min. The cover slips were washed three times with the same buffer and transferred to a chamber with 1 ml of the modified Krebs-Ringer buffer (no FCS) and mounted on the stage of a Zeiss IM-35 inverted microscope. The stage, ~16 oil-immersion objective, and chamber were thermostatically regulated at 37°C. Alternatively, the cells were loaded with fura- in the initial modified Krebs-Ringer buffer containing 3.5 mM ethylene glycol-bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA) for 30 min, washed with the same buffer, and treated with cyanide in the presence of EGTA. In the experiments with a low-calcium buffer (~2 PM total calcium), the cells were incubated with fura-2/AM for 30 min in modified Krebs-Ringer buffer that had been passed through a Chelex 100 (200-400 mesh) column (Bio-Rad). A liquid N,-cooled charge-coupled device camera (Photometrics) was used as the imaging device. Pairs of images were taken every 40 s over 90 min and digitized at 12-bit resolution, stored, and analyzed with a Heurikon HK68/M 10 computer. Fluorescence images were obtained at excitation wavelengths of 340 and 380 nm (lo-nm bandwidth) with an emission wavelength of 460-580 nm. The integration time for each image was 400 ms, and individual pixels were binned into 3 x 3 superpixels at readout from the charge-coupled device detector to improve signal-to-noise ratio. The time required for wavelength changes was 1 s. To minimize photobleaching, a computer-controlled
C685
HYPOXIA
shutter was used to limit the exposure of the cells to excitation light. The mean fluorescence intensity of individual cells was determined by selecting a region of the image covering each cell. The mean intensity over the same area for each of the successive images in a time series was measured to create a time course of fluorescence changes. At the end of each experiment, cells were exposed to ionomycin (20 PM) and MnCl, (2 mM) for 30 min. The residual fluorescence was measured over the same region of each cell as the Ca*+ -dependent fluorescence changes and subtracted from the fura- Ca*+ signals for that cell at the appropriate wavelength. Calibration of fura- fluorescence in terms of [Ca’+]i was calculated from the ratio of 340- to 380-nm fluorescence values (after subtraction of background fluorescence) as described by Grynkiewicz et al. (8). For each experiment, values for Rmin and R,,, were determined using solutions of fura-2free acid in the same microscope chamber used for the cell incubations. The dissociation constant for the fura-2-Ca*+ was taken as 224 nM (8). Statistical analysis. Analyses of variance using a repeatedmeasures design were performed on the data using the PC version of the SAS statistical package (version 6.04). RESULTS
Prevention of toxicity of cyanide by chlorpromazine, cytochalasin B, and extracellular acidosis. Table 1 docu-
ments that a number of manipulations reduced the killing of cultured hepatocytes by cyanide without sparing the depletion of ATP that occurs with this inhibitor of cytochrome oxidase. Chlorpromazine was shown previously to prevent the degradation of phospholipid that accompanies liver ischemia in the intact rat (3). Similarly, cytochalasin B was shown previously to reduce the killing of cultured hepatocytes by anoxia (19). Lastly, reducing the pH of the culture medium to 6.0 protects the hepatocytes (7, 10, 13, 17). Accelerated phospholipid degradation in cyanide-intoxicated hepatocytes. Hepatocytes were labeled with [3H] arachidonic acid during their first 24 h in culture. The
cells were then washed, resuspended in a Krebs-Ringer bicarbonate buffer, and treated with 1 mM cyanide. Table 1. Chlorpromazine, cytochalasin B, and extracellular acidosis do not prevent depletion of ATP induced by cyanide Cell
Death, %
ATP, nmol/mg protein
No additions 3tl 11.8OkO.28 Cyanide alone 65t5 0.65kO.25 Chlorpromazine alone 5+1 Chlorpromazine + cyanide 22t4 0.58t0.3 1 Cytochalasin B alone 4+1 Cyanide + cytochalasin B 25t2 0.61kO.21 pH 6.0 alone 5&l Cyanide + pH 6.0 5&l 0.46kO.23 Oligomycin 6t2 0.61kO.18 Values are means t SD; n = 3 separate cultures. Cultured hepatocytes were treated with 1 mM cyanide in presence or absence of 100 pM chlorpromazine or cytochalasin B (25 pg/ml). Cells treated with oligomycin received 0.10 pg/ml. Alternatively, cells were exposed to cyanide in culture medium at pH 6.0. Controls received no additions, chlorpromazine alone, or cytochalasin B alone or were exposed to culture medium at pH 6.0. After 15 min, cellular content of ATP was measured. Viability of cells was determined after 90 min.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
C686
LIPID METABOLISM
WITH
Phospholipid degradation was assessed by the release of arachidonic acid into the culture medium and was expressed as a percent of the total [3H]arachidonic acid incorporated into phospholipids. Table 2 indicates that in the absence of cyanide, the hepatocytes released -3% of total [3H]arachidonic acid incorporated into phospholipids within 90 min. With 1 mM cyanide, approximately twice as much [3H]arachidonic acid was released. Table 2 further details the effect of chlorpromazine, cytochalasin B, and extracellular acidosis on the extent of phospholipid degradation in cyanide-intoxicated hepatocytes. Chlorpromazine, cytochalasin B, and extracellular acidosis lowered the basal rate of phospholipid degradation and prevented the increase produced by cyanide. The accelerated phospholipid degradation in cyanideintoxicated hepatocytes is not a simple consequence of the depletion of ATP. Oligomycin reduced the ATP content to the same extent as did cyanide (Table 1) but did not affect the viability (Table 1) or lipid hydrolysis (Table 2) over a similar 1.5 h of incubation. Activation of phospholipid hydrolysis by the calcium ionophore A23167. The [Ca’+]; of cultured hepatocytes
can be abruptly increased and the cells killed by adding the calcium ionophore A23187 (24, 25). Table 3 shows that 15 PM A23187 increased the rate of phospholipid hydrolysis in parallel with the loss of viability of 70% of the cells within 90 min. Omission of calcium from the culture medium prevented both the cell killing and the increased rate of phospholipid degradation (data not shown). Interestingly, cytochalasin B did not protect the hepatocytes from the toxicity of A23187 and did not prevent the accelerated phospholipid degradation, despite a lowering of the basal rate of phospholipid degradation (Table 3). By contrast, chlorpromazine and extracellular acidosis lowered the basal rate of phospholipid degradation, prevented the increased degradation produced by A23187, and reduced the extent of the cell killing (Table 3). [ Ca2+]i in cyanide-intoxicated hepatocytes. The effect of cyanide on [Ca2+]; was studied at the single-cell level
CHEMICAL
HYPOXIA
Table 3. Phospholipid degradation and cell killing by A23187 Phospholipid
Degradation,
%
Cell
Death,
%
No additions 3.18k0.12 5t1 A23187 6.31t0.18 71t4 +Cytochalasin B 5.89kO.35 67k4 +Chlorpromazine 2.15t0.38 24k4 +pH 6.0 3.59k0.28 8t3 Cytochalasin B alone 1.88kO.08 4tl Chlorpromazine alone 2.45k0.46 11&l pH 6.0 alone 2.63k0.21 5tl Values are means t SD; n = 3 separate cultures. Cultured hepatocytes were treated with cytochalasin (25 pg/ml) B or 100 PM chlorpromazine with or without 15 PM A23187. Alternatively, cells were treated with A23187 in culture medium at pH 6.0. Controls received no additions, cytochalasin B alone, or chlorpromazine alone or were exposed to culture medium at pH 6.0. After 90 min, extent of phospholipid degradation and loss of cell viability of cells were measured.
by using digital imaging fluorescence microscopy. The ability of this methodology to detect changes in [Ca2+]i under the present conditions is documented in Fig. 1. The time course of the changes in fluorescence intensity is illustrated at both 340 and 380 nm after treatment of a
r
Table 2. Phospholipid degradation and prevention of toxicity of cyanide Phospholipid
Hydrolysis,
%
3.17t0.13 No additions Cyanide 5.55t0.60* Chlorpromazine 2.10t0.20* Cyanide + chlorpromazine 1.90+0.lO*jCytochalasin B 1.74t0.18* Cyanide + cytochalasin B 1.83+0.ll*t pH 6.0 1.19t0.13* Cyanide + pH 6.0 1.80k0.28*t 3.19kO.22 Oligomycin Values are means k SD; n = 3 separate cultures. Cultured hepatocytes were treated with 1 mM cyanide in presence or absence of 100 PM chlorpromazine or cytochalasin B (25 pg/ml). Alternatively, the cells were exposed to cyanide in culture medium at pH 6.0. Controls received no additions, chlorpromazine alone, or cytochalasin B alone or were exposed to culture medium at pH 6.0. Phospholipid degradation was measured after 90 min. * Significantly different from control cells (no additions) at P < 0.01. t Not significantly different from control of respective treatment alone (no cyanide).
.-” I.OC 1
s:
E 0
0.5 u 0’
0
I
I 4
I
I 8
I
I 12
1
I 18
Time (mid Fig. 1. Fluorescence changes and calibrated cytosolic free Ca*+ concentration ([Ca”‘] J response in a single hepatocyte treated with ionomytin. Representative tracings of fluorescence signals and calculated [Ca*+]i increase observed in 1 fura-2-loaded hepatocyte after addition of 15 PM ionomycin are shown. Fluorescence values were measured from a sequential series of images as described in MATERIALS AND METHODS. Top: time courses of fluorescence changes observed at excitation wavelengths of 340 and 380 nm. Integration time for each image was 400 ms, with a l-s delay between wavelength changes. Bottom: same data calibrated in terms of [Ca*+]i calculated from ratio of 340- to 380-nm fluorescence values as described in MATERIALS AND METHODS. Arrow, time of addition of ionomycin.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
fura-2-loaded hepatocyte with 15 PM ionomycin, a calcium ionophore (Fig 1, top). Immediately upon addition of ionomycin, the 340-nm fluorescence (calcium-bound fura-2) increased and the 380-nm fluorescence (free fura-2) decreased. Over the ensuing 6-8 min, fura- fluorescence changed only slightly. The rapid loss of fluorescence at both 340 and 380 nm after -9 min reflected the loss of fura- from the cell, an effect signifying the death of the cell. There was a calculated rise in [Ca2+]i in response to ionomycin (Fig. 1, bottom). The calcium concentration rose rapidly within seconds and then increased slowly up to the point of the death of the cell. Table 4 summarizes the changes in [Ca2+]; induced by ionomycin in the nine hepatocytes studied. An almost eightfold rise in [Ca2+]; occurred within 20 s of adding the ionomycin. Figure 2 illustrates the time course of the effect of 1 mM cyanide on [ Ca2+]; in a representative fura-2-loaded hepatocyte. For >70 min after addition of cyanide, there was no change in [Ca2+]i. After -70 min, there was a rise in [Ca2+]i that reached a plateau within 10 min. This late rise was observed in all 43 cells studied that died and began at 71.4 t 7.7 min to reach a plateau value of 683 t 210 nM. Interestingly, this plateau concentration of calcium is close to the maximal level generally induced by hormones in these cells (21). Shortly after the plateau concentration was reached, there was an abrupt increase in [Ca’+]; (Fig. 2). This second rise was also seen in all 43 cells studied that died and occurred 83.5 t 8.0 min after the addition of cyanide. After this second rise in [Ca2+];, fura- fluorescence at both 340 and 380 nm (Fig. 2, top) was rapidly lost as the cells died. Table 5 summarizes the [Ca2+]i 45 min after exposure of hepatocytes to 1 mM cyanide. Cyanide did not produce a significant increase in [Ca”+]; after 45 min in the 43 cells studied. To examine the effects of cyanide on cellular autofluorescence, experiments were carried out using cells not loaded with fura-2. Under these conditions, the hepatocytes did not show any significant changes of fluorescence at either 340 or 380 nm upon addition of 1 mM cyanide (data not shown). [Ca2+]i in hepatocytes treated with cyanide in presence
of EGTA. To explore the relevance of the late alterations in intracellular calcium homeostasis to the cell killing, the cells were incubated with cyanide in the presence of 3.5 mM EGTA. Figure 3 illustrates that in the presence of EGTA there were no late increases in [Ca2+]i. Thus neither the delayed plateau response nor the later abrupt Table 4. Calcium homeostasisin hepatocytes treated with 15 PM ionomycin [ Ca’+]i,
nM T peak)
Time
0
S
Peak
206k55 1,573+505 21t7 Values are means k SD; n = 9 individual hepatocytes. hepatocytes were loaded with fura-2, and cytosolic free Ca2+ was determined for individual cells from ratio images of furacence. Tpeak, time to half height of abrupt rise in [Ca”+]i after of ionomycin.
Cultured ([Ca’+]i) fluoresaddition
700
s
HYPOXIA
C687
r
100
3
25.
;ts
2.0
8 1.5 t .-0 I.0 2 co 0.5 0” 0 0
20
40 60 80 100 Time (min) Fig. 2. Effect of cyanide on fluorescence changes and calibrated [Ca2+li response in a single hepatocyte. Representative tracings of fluorescence signals (top) and calculated [Ca2+]i (bottom) observed in 1 fura-2-loaded hepatocyte after addition of 1 mM cyanide are shown. Experiment was carried out as for Fig. 1.
Table 5. Calcium homeostasis in hepatocytes intoxicated with cyanide [Ca2+];,
nM
n Time
0
45 min
Plateau
T dead7 min
KCN 43 199t54 217t66 683t210 89t8 KCN + EGTA 19 88t46 90t58 88tl KCN (low Ca2+) 9 166t38 217t56 85k17 Values are means t SD; n, no. of separate measurements made on individual hepatocytes. Cultured hepatocytes were loaded with fura-2, and [Ca2+]i was determined for individual cells from ratio images of fura- fluorescence at time 0,45 min after treatment of cells with 1 mM cyanide, and at plateau. T dead,time to half point of rapid loss of fluorescence intensity at 340 nm accompanying death of cells.
increase seen in Fig. 2 was observed. In most cells, [Ca2+]i actually decreased just before the loss of viability (Fig. 3). Interestingly, this decrease in the presence of EGTA occurred at a time similar to that at which the large Ca2+ increase occurred in physiological Ca2+ medium (Fig. 2). Table 5 documents that EGTA lowered the basal level of cytosolic calcium and the [Ca2+]; 45 min after treatment with cyanide. However, there was no plateau phase (Table 5 and Fig. 3), and the time to the death of the cells was the same as in the presence of extracellular calcium (Table 5). EGTA had no effect on phospholipid degradation or on the extent of cell killing by cyanide (Table 6). Similar results were obtained by simply treating the hepatocytes with 1 mM cyanide in a low-calcium buffer (~2 PM free Ca2+). Table 5 indicates that the basal level
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
r
WITH CHEMICAL
HYPOXIA
DISCUSSION
The data presented in this report identify a number of conditions that substantially reduced the toxicity of cyanide for cultured rat hepatocytes. In turn, these conditions were used to assess the relationship between the cell killing by cyanide and an accompanying accelerated degradation of phospholipids. Finally, digital imaging fluorescence microscopy was used to establish that an elevation of [Ca2+]i is not the cause of the enhanced lipid degradation. Cyanide inhibits cytochrome oxidase and thereby prevents oxygen reduction. Continued electron transport and thus mitochondrial energization is prevented. The mitochondrial membrane potential is lost (15), and ATP stores are rapidly depleted. In a manner that has not been clearly defined, these initial consequences of cyanide intoxication are followed by a loss of the permeability barrier function of the plasma membrane and the death of the cell. The nature of the events coupling the loss of mitochondrial function to the disruption of plasma mem0 brane integrity was the focus of the present study. 0’ 0 20 40 60 80 100 The conditions that were found to reduce the extent of cell killing by cyanide most likely act to modify those Time (mid events that couple the loss of mitochondrial energization Fig. 3. Effect of EGTA on fluorescence changes and calibrated [Ca2+]i to the death of the cells. With chlorpromazine, cytocharesponse in a single cultured hepatocyte treated with cyanide. Overlasin B, and extracellular acidosis, ATP was depleted (Tanight-cultured hepatocytes were washed and loaded with fura- in modified Krebs-Ringer bicarbonate buffer containing 3.5 mM EGTA for 30 ble 1) despite the reduction in the number of cells that min. After cells were washed, they were placed in the same buffer died (Table 1). Thus they must act independently of the containing 3.5 mM EGTA. Representative tracings of fluorescence sigloss of the ATP-generating capacity of the mitochondria. nals (top) and calculated [Ca”] i (bottom) observed in a single fura-2The killing of the cultured hepatocytes by cyanide was loaded hepatocyte after addition of 1 mM cyanide are shown. accompanied by an accelerated release of [3H]arachidonate from cellular phospholipids (Table 2). Chlorpromazine, cytochalasin B, and extracellular acidosis reduced Table 6. Phospholipid hydrolysis and cell killing this phospholipid degradation (Table 2) in parallel with by cyanide in presence or absence their preservation of the viability of the hepatocytes (Taof extracellular calcium ble 1). These data would suggest that phospholipid degradaPhospholipid Cell Death, % Hydrolysis, % tion is related to the loss of membrane integrity that determines the fate of the cells. Favoring such an interControl cells + Ca2+ 3.04t0.04 5tl pretation is the fact that chlorpromazine, cytochalasin B, Cyanide + Ca2+ 5.52t0.25 63t5 and extracellular acidosis affected the rate of lipid degraControl cells + EGTA 3.07Iko.10 8t3 74k7 Cyanide + EGTA 5.55t0.35 dation in the control cells. In other words, the protective 4tl Control cells - Ca2+ 3.02t0.14 action of chlorpromazine, cytochalasin B, and extracell66t6 Cyanide - Ca2+ 5.64t0.23 ular acidosis can be directly related to their ability to Values are means t SD; IZ = 3 separate cultures. Cultured hepatoaffect phospholipid degradation and thus to inhibit its cytes were treated with 1 mM cyanide in a Krebs-Ringer bicarbonate stimulation by cyanide intoxication. buffer containing no calcium or 1.8 mM CaC12. Other cells were preChlorpromazine lowered the basal rate of phospholipid treated with 1 mM EGTA in a calcium-free Krebs-Ringer buffer for 45 min before addition of 1 mM cyanide. After 90 min, phospholipid hy- degradation and prevented the increase occurring with drolysis and loss of cell viability were measured. cyanide (Table 2). Because phospholipase A activities have an alkaline pH optimum, by acidifying the cytosol (7)) extracellular acidosis would inhibit phospholipid degof cytosolic calcium was reduced somewhat in the low- radation, an effect that can account for both the lowered basal rate and the inhibition of the increase seen with calcium medium (P < 0.05). However, 45 min after treatcyanide (Table 2). ment with 1 mM cyanide in the low-calcium buffer, Cytochalasin B similarly lowered the basal rate of lipid [ Ca2+]; was not significantly different from that in 1.8 mM calcium. Furthermore, the time to the death of the degradation and prevented the increase with cyanide. There is an intimate association between the cytoskelecells was not changed in the low-calcium medium. Table 6 documents that the low-calcium medium did not affect ton and the plasma membrane of the hepatocyte. It is of actin the increased phospholipid degradation produced by cy- suspected that as a result of the depolymerization microfilaments, cytochalasin B inhibits an interaction anide nor the extent of cell killing within 90 min.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
LIPID METABOLISM
WITH CHEMICAL
between the cytoskeleton and the plasma membrane that determines, at least in part, the basal rate of lipid turnover. Interestingly, such an hypothesis implies that cyanide intoxication may activate phospholipid degradation by an effect on the cy-toskeleton and its interaction with cellular membranes. The enhanced phospholipid degradation in hepatocytes intoxicated with cyanide cannot be attributed to a rise in [Ca’+]i. An elevated [Ca”+]i can induce phospholipid degradation (24,25), and raising the calcium content of the cultured hepatocytes by treating with the calcium ionophore A23187 increased the release of [3H]arachidonate and killed the cells (Table 3). Chlorpromazine and extracellular acidosis but not cytochalasin B prevented the increased phospholipid hydrolysis occurring with A23187 intoxication and protected the cells. These data distinguish the toxicity of ionomycin from that of cyanide and suggest that an elevated [Ca2+]i is not the mechanism of the increased lipid degradation associated with cyanide intoxication. Such a conclusion was confirmed by directly measuring the effect of cyanide on the [Ca2+]; of cultured hepatocytes. There was no change in [Ca2+]; in cyanide-intoxicated hepatocytes until just before the cells died (Fig. 1 and Table 5), a result that confirms a similar finding reported previously (11, 13, 16). The late increases in [Ca2+]i most likely represent an influx of calcium ions across an injured plasma membrane. Removal of extracellular calcium ions prevented the late increases in calcium without affecting the loss of viability (Tables 5 and 6). Cytosolic calcium was reduced by >50% by treating the cells with 3.5 mM EGTA. There was again no rise in [Ca”+]; with cyanide and no effect on the extent of cell killing (Fig. 3 and Tables 5 and 6). With EGTA there was actually a late decrease in [Ca2+]i just before the death of the cells. This most likely represents an efflux of calcium across a damaged plasma membrane from a higher cytosolic to a lower extracellular calcium ion concentration. In the presence of a physiological extracellular calcium concentration, the gradient is in the reverse direction, and calcium enters the cells. In either case, the late changes in [Ca2+]; do not affect the fate of the cells. Finally, the data here and those published previously (11, 13, 18) differ from the reported effects of cyanide on [Ca2+]i in a hepatoma cell line (16). Upon addition of cyanide, [Ca2+]i started to rise and reached a plateau sevenfold higher than the basal level within 10 min. The discrepancy here in the effect of cyanide most likely reflects the fact that the hepatoma cell line studied is no longer an hepatocyte. One is comparing the effect of cyanide on two very different kinds of cells. It may be difficult to extrapolate conclusions with respect to the behavior of the normal hepatocyte from studying a malignant variant. This work was supported by National Institutes of Health Grants DK-38305, DK-38422, and AA-07186. Present address of I. Sakaida: First Department of Internal Medicine, Yamaguchi University School of Medicine, Ube 755, Japan. Address for reprint requests: J. L. Farber, Dept. of Pathology, Room 251, Jeff Alumni Hall, Thomas Jefferson Univ., Philadelphia, PA 19107.
C689
HYPOXIA
Received 22 October 1991; accepted in final form 21 April 1992. REFERENCES 1. Buja, L. M., H. K. Hagler, D. Parsons, K. R. Chien, R. C. Reynolds, and J. T. Willerson. Alterations of ultrastructure
and elemental composition in cultured neonatal rat cardiac myocytes after metabolic inhibition with iodoacetic acid. Lab. Inuest. 53: 397-412, 1985. 2. Casini, A., M. Giorli, and J. L. Farber.
R. J. Hyland,
A. Serroni,
D. Gilfor,
Mechanisms of cell injury in the killing of cultured hepatocytes by bromobenzene. J. Biol. Chem. 257: 67216728, 1982.
3. Chien, Farber.
K. R., J. Abrams,
A. Serroni,
J. T. Martin,
and J. L.
Accelerated phospholipid degradation and associated membrane dysfunction in irreversible liver cell injury. J. Biol. Chem. 253: 4809-4817, 1978.
4. Chien, K. R., A. Sen, D. Gunn, L. M. Buja,
R. Reynolds, A. Chang, and J. T. Willerson.
Y. M. Kim,
M.
Release of arachidonate from membrane phospholipids in cultured neonatal rat myocardial cells during adenosine triphosphate depletion. Correlation with the progression of cell injury. J. Clin. Inuest. 75: l7701780, 1985. 5. Farber, J. L., and E. E. Young. Accelerated phospholipid degradation in anoxic rat hepatocytes. Arch. Biochem. Biophys. 211: 312-320, 1981. 6. Finkelstein, S. D., D. Gilfor, and J. L. Farber. Alterations in the metabolism of lipids in ischemia of the liver and kidney. J. Lipid Res. 26: 726-734, 1985. 7. Gores, G. J., A.-L. J. J. Lemasters.
Nieminen,
B. E. Wray,
B. Herman,
and
Intracellular pH during “chemical hypoxia” in cultured rat hepatocytes. Protection by intracellular acidosis against the onset of cell death. J. Clin. Invest. 83: 386-396, 1989. 8. Grynkiewicz, G., M. Poenie, and R. Y. Tsien. A new generation of Ca*+ indicators with greatly improved fluorescent properties. J. BioZ. Chem. 264: 3440-3450, 1985. 9. Gunn, Buja,
M. and
D., A. Sen, Y. M. Kim, J. T. Willerson, K. R. Chien. Mechanisms of accumulation
L. M.
of arachionic acid in cultured myocardial cells during ATP depletion. Am. J. Physiol. 249 (Heart Circ. Physiol. 18): H1188-H1194, 1985. 10. Harrison, D. C., J. J. Lemasters, and B. Herman. A pHdependent phospholipase A2 contributes to loss of plasma membrane integrity during chemical hypoxia in rat hepatocytes. Bichem. Biophys. Res. Commun. 174: 654-659, 1991. 11. Herman, B., A.-L. Nieminen, G. J. Gores, and J. J. Lemasters. Irreversible injury in anoxic hepatocytes precipitated by an abrupt increase in plasma membrane permeability. FASEB J. 2: 146-151,1988. 12. Kane, A. B.,
D. Petrovich,
R. 0.
Stern,
and
J. L. Farber.
ATP depletion and the loss of cell integrity in anoxic hepatocytes and silica-treated P388Dl macrophages. Am. J. Physiol. 249 (Cell Physiol. 18): C256C266, 1985. 13. Lemasters, Herman.
J. J.,
J. DiGuiseppi,
A.-L.
Nieminen,
and
B.
Blebbing, free Ca*+ and mitochondrial membrane potential preceding cell death in hepatocytes. Nature Lond. 325: 78-81, 1987.
14. Lowry, Randall.
0.
15. Masaki,
N.,
H.,
N.
J.
Rosebrough,
A.
L.
Farr,
and
R. J.
Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275, 1951. A. P. Thomas,
J. B. Hoek,
and
J. L. Farber.
Intracellular acidosis protects cultured hepatocytes from the toxic consequences of a loss of mitochondrial energization. Arch. Biothem. Biophys. 272: 152-161, 1989. 16. Nicotera, P., H. Thor, and S. Orrenius. Cytosolic-free Ca*+ and cell killing in hepatoma lclc7 cells exposed to chemical anoxia. FASEB J. 3: 59-64, 1989. 17. Nieminen, B. Herman,
A.-L., and
T. L. Dawson, J. J. Lemasters.
G. J. Gores,
T. Kawanishi,
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, 1990.
18. Nieminen, A.-L., J. J. Lemasters.
G. J. Gores,
B. E. Wray,
B. Herman,
and
Calcium dependence of bleb fomration and cell death in hepatocytes. Cell Calcium 9: 237-248, 1989.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
C690
LIPID METABOLISM
WITH CHEMICAL
19. Okayasu, T., M. T. Curtis, and J. L. Farber. Cytochalasin delays but does not prevent cell death from anoxia. Am. J. Pathol. 117: 163-167,
1984.
Petrovich, D. R., S. Finkelstein, A. J. Waring, and J. L. Farber. Liver ischemia increases the molecular order of microsoma1 membranes by increasing the cholesterol-to-phospholipid ratio. J. Biol. Chem. 259: 13217-13223, 1984. 21. Rooney, T. A., E. J. Sass, and A. P. Thomas. Characterization of cytosolic calcium oscillations induced by phenylephrine and vasopressin in single fura-2-locaded hepatocytes. J. Biol. Chem. 264: 17131-17141, 1989. 22. Seglen, P. 0. Preparation of isolated rat liver cells. Methods Cell Biol. 13: 29-83, 1976.
20.
HYPOXIA
23. Sen, A., J. C. Miller, R. Reynolds, J. T. Willerson, L. M. Buja, and K. R. Chien. Inhibition of the release of arachidonic acid prevents the development of sarcolemmal membrane defects in cultured rat myocardial cells during adenosine triphosphate depletion. J. Clin. Invest. 82: 1333-1338, 1988. 24. Shier, W. T. Serum stimulation of phospholipase A2 and prostaglandin release in 3T3 cells is associated with platelet-derived growth-promoting activity. Proc. Natl. Acad. Sci. USA 77: 137141, 1980.
25. Shier, W. T., and D. J. DuBourdieu. Role of phospholipid hydrolysis in the mechanism of toxic cell death by calcium and ionophore A23187. Biochem. Biophys. Res. Commun. 109: 106112, 1982.
Downloaded from www.physiology.org/journal/ajpcell at Midwestern Univ Lib (132.174.254.157) on February 13, 2019.
