178, 617-624 (1977)
Homeostasis in Galactosamine-Intoxicated Liver Cells1
JOHN L. FARBER, SAMIR K. EL-MOFTY,” FRANCIS A. X. SCHANNE, JOSEPH J. ALEO, JR., AND ADA SERRONI Department
and the Fels Research Institute, Philadelphia,
Temple University 19140
School of Medicine,
Received June 21, 1976 Active transport of Ca *+ by isolated microsomes and mitochondria from galactosamine-intoxicated rat liver cells was studied. The aim was to determine the respective role of each organelle in the disturbed intracellular Ca 2+ homeostasis induced by this hepatotoxin. Calcium uptake by isolated microsomes is ATP dependent and oxalate augmented with a V of 1.45 nmol of Ca*+/mg of microsomal proteinimin at 25°C and an apparent K, for free Ca*+ of 2.4 PM. Concentrations of total Ca2+higher than 40 PM are inhibitory. Two hours after administration of galactosamine (200 or 400 mglkg), at a time when the total cell Ca*+ content has increased, microsomes isolated from the treated animals exhibited no impairment in calcium transport. The microsomal preparations from the galactosamine-treated animals also had the same content of cytochrome P-450 and the same specific activity of glucose 6-phosphatase as those from the control animals. Calcium uptake by isolated liver mitochondria is also ATP dependent but virtually completely inhibited by 5 mM sodium azide. The V is higher than that of the microsomes, 20.5 nmol of Ca2+/mg of proteinimin at 25”C, but the apparent K, for free Ca2+ is similar, 5.7 PM. There was no alteration in the Caz+ uptake activity of mitochondria isolated from galactosamine-treated animals. These results imply that the initial disturbance in intracellular Ca Z+ homeostasis induced by galactosamine is entirely a consequence of the previously described plasma membrane injury. The potential significance of the observed kinetic properties of microsomes and mitochondria with regard to their respective roles in intracellular Ca2+ homeostasis is discussed.
Administration of n-galactosamine to rats produces a marked disturbance in the control of intracellular calcium levels within the liver (l-3). The increased concentration of liver cell calcium, in turn, has been implicated as the mediator of the cell death that results from galactosamine intoxication. These alterations in calcium metabolism provide a convenient system for studying the control of intracellular calcium content. The changes are dose dependent and easily measureable. The disturbed calcium regulation can be readily 1This work was supported by Grants CA-12073 and AM-19154 from the National Institutes of Health. 2Present address: Department of Oral Pathology, Washington University School of Dentistry, St. Louis, Missouri 63110.
reversed by the administration of uridine 0). Our previous studies have suggested that the disturbed calcium homeostasis is related to an as yet poorly characterized alteration in the structure of the plasma membrane (1). Plasma membranes isolated within the first few hours from galactosamine-treated animals show a 40% reduction in 5’-nucleotidase activity and a twofold increase in maximum negative ellipticity determined by circular dichroism. Simultaneous administration of uridine prevents these alterations in the plasma membranes. The membrane alterations are reversed when uridine is administered for up to 2.5 h after galactosamine. Uridine will prevent and reverse the changes in calcium content in parallel to its ability 617
Copyright 0 1977by Academic press, Inc. All rights of reproduction in any form reserved.
to reverse the plasma membrane alterations. This evidence linking the changes in the plasma membrane to the increases in liver cell calcium is only circumstantial. It is possible that there are effects on other intracellular organelles involved in the maintenance of calcium homeostasis. Damage to these structures may be a factor in the disturbed calcium metabolism induced by galactosamine. The binding and transport of calcium cations have been characterized second only to oxidative phosphorylation as one of the most ubiquitous energy-linked reactions of isolated mitochondria (4-7). It has been suggested that the transport of Ca2+may be an alternative to oxidative phosphorylation (8, 9) and may play an important regulatory role in cell calcium physiology (4, 10-16). Active efflux of calcium from rat liver cells has been reported (17, 18). Although attempts to demonstrate a calcium pump mechanism in isolated rat liver plasma membranes have been unsuccessful (19), energy-dependent calcium sequestration has been observed by microsomes derived from rat liver cells (20). In uiuo administration of carbon tetrachloride resulted in inhibition of the in vitro calcium uptake activity of isolated microsomes (21). Inhibition of this calcium pump mechanism was suggested as playing a role in the disturbed calcium homeostasis induced by carbon tetrachloride (21). In the present study we have examined the effect of galactosamine intoxication on the calcium transport activities of isolated mitochondria and microsomes. At a time when plasma membrane alterations and an associated increase in cellular calcium content are easily demonstrable, there are no detectable alterations in the active transport of calcium by mitochondria and microsomes. These observations would imply that the initial disturbance in intracellular calcium homeostasis induced by galactosamine is entirely a consequence of the plasma membrane injury with either an increased passive in&x of calcium or a failure of some as yet undetected active efflux mechanism present in the plasma membranes.
ET AL. MATERIALS
Female Wistar rats (Charles River Farms, Wilmington, Del.) weighing 150-170 g and fasted overnight were used in all the experiments except where indicated. n-Calactosamine (Sigma) was dissolved in 0.9% NaCl (20 mglml) and administered by intraperitoneal injection of either 200 or 400 mglkg body wt. Control animals received an equal volume of 0.9% saline. Ccl, (Fisher Scientific) was administered by stomach tube (2.5 ml/kg body wt). Preparation ofmicrosomes. Microsomes were prepared by modification of procedures previously described (22, 23). All steps were performed at 4°C. Animals were sacrificed by decapitation and the livers were removed immediately. Liver (4-6 g) was minced and homogenized by eight strokes of a motor-driven Teflon pestle in a Potter homogenizer in 4 vol of 100 mM KCl, 30 mM histidine-imidazole buffer, pH 6.6. The homogenate was centrifuged at 80008 for 20 min in a Sorvall RC-2B centrifuge. The supernatant was aspirated through an l&gauge cannula and centrifuged at 28,000g for 60 min. The supernatant was aspirated and discarded. The pellet was washed with 1 ml of 100 mM KCl, 30 mM histidine-imidazole buffer, pH 7.4, and suspended by hand-homogenization in a loose-fitting Dounce homogenizer in 8 ml of the same buffer. Preparation of mitochondria. Mitochondria were prepared by modification of the method of Schneider and Hogeboom (24). Three grams of liver was homogenized by eight strokes of a motor-driven Teflon pestle in a glass homogenizer in 27 ml of 0.25 M sucrose. The homogenate was centrifuged at SOOgfor 10 min. The supernatant was respun for 10 min at 85OOg. The resulting pellet was rinsed with 0.25 M sucrose, resuspended by gentle homogenization in a loose-fitting Dounce homogenizer in 7 ml of 0.25 M sucrose, and respun at 85OOg for 10 min. The pellet was resuspended and centrifuged again and the iinal pellet was resuspended in 8 ml of 100 mM KCl, 30 mM histidine-imidazole, pH 7.4. Measurement of calcium uptake activity. Calcium uptake by whole liver homogenates, purified microsomes or mitochondria, was measured immediately after preparation of the respective samples. The standard assay contained in a total volume of 2.5 ml: 100 mM KCl, 5 mM MgCl,, 5 mM ammonium oxalate, 5 mM sodium azide, 5 mM ATP as the disodium salt (Sigma), 40 pM CaCl, (0.1 &i/ml of 45CaCl,, New England Nuclear), and 30 mM histidine-imidazole buffer, pH 7.4 (21). All reactions were started by addition of the tissue fraction. The final protein concentrations were 60-80 pg/ml for liver homogenates, 80-160 pg/ml for microsomes, and lo-60 pg/ ml for mitochondrial assays. Incubation was at 25°C for up to 60 min. Aliquots (500 ~1) were removed at the indicated times and filtered through 0.45-pm filters (Type HA, Millipore Corp.) and washed with
2 ml of ice-cold 0.25 M sucrose. The filters were prepared for use by presoaking in ice-cold 0.25 M sucrose. %a was measured by liquid scintillation spectrometry in 10 ml of Aquasol (New England Nuclear Corp.). Protein was determined by the method of Lowry et al. (25) with bovine serum albumin as the standard. Measurement of cytochrome P-450 and glucose 6phosphatase. The cytochrome P-450 content of isolated microsomes was determined by the method of Omura and Sato (26) and glucose 6-phosphatase by the method of Swanson (27). RESULTS
Calcium Uptake by Isolated Rat Liver Microsomes
Rat liver microsomes will actively take up and sequester calcium in vitro (21). Table I illustrates the basic properties of this system under our conditions. In the presence of ATP and 5 mM ammonium oxalate, the initial rate of calcium uptake is maintained at least for 60 min. In the absence of ATP there is very little calcium uptake. The residual activity probably represents binding of 45Ca2+to the microsomal membranes. Elimination of oxalate from the assay greatly reduces the extent of the ATP-dependent calcium uptake, while the calcium binding in the absence of ATP is unaffected by the presence of oxalate. The initial rate of ATP-dependent calcium uptake with and without oxalate is the same. However, in the absence of oxalate this initial rate is maintained for only a short time such that, at the longer time shown
in Table I, considerably more Ca2+ has been taken up in its presence. The increased Ca2+uptake with oxalate is probably due to the precipitation of calcium oxalate in the interior of the microsomal vesicles. Precipitation occurs as the concentration of calcium oxalate exceeds its solubility product with the active accumulation of calcium from the medium. Sodium azide (5 mM) in the incubation had little or no effect on the ATP-dependent calcium uptake. This readily distinguishes microsomal calcium sequestration from that by isolated mitochondria (Table I). The latter is 95% inhibited by the same concentration of NaN,. The dependency of the rate of microsoma1 calcium uptake on the total Ca2+concentration is shown in Fig. 1. With Ca2+ concentrations between 0 and 40 PM, there is increasing uptake with increasing Ca2+. Concentrations higher than 40 PM are inhibitory, with the rate falling to less than 50% of the maximum with 200 PM total calcium. From a double-reciprocal plot of the data between 0 and 40 PM in Fig. 1, the apparent V is 1.46 nmol of calcium/mg of microsomal proteimmin at 25°C. The K, for total calcium is 18.4 FM. This corresponds to a K,,, for free Ca”+ of 2.4 PM using the method of Katz et al. (28) to correct for the Ca”+ bound to adenine nucleotides. Two preliminary experiments were per-
TABLE I CALCIUM UFTAKE BY ISOLATED RAT LIVER MICROSOMES Conditions
Complete Minus ATP Minus oxalate Plus azide
Microsomal calcium uptake (nmol of *CaZ’/ mg of protein/45 min) 74.2 3.12 19.4 77.8
k ? + L
8.6 0.3 1.6 12.9
Mitochondrial calcium uptake (nmol of Ta*+/ mg of protein/l0 min) 205.4 13.1 49.5 9.3
2 ? + -t
23.7 6.7 1.6 5.0
a Rat liver microsomes and mitochondria were isolated and assayed for calcium uptake activity as described under Materials and Methods. Incubations were at 25°C for 45 min with microsomes and 10 min with mitochondria. Results are the means 2 SD of separate preparations from three animals.
FIG. 1. Dependency of microsomal calcium uptake on the concentration of total calcium. Microsomes were isolated and assayed for calcium uptake activity as described under Materials and Methods. All values are corrected for the amount of Va*+ bound to the microsomal membranes by subtracting the 0 time values. Results are the means -+ SD of separate preparations from three animals.
formed in preparation for examining the effect of galactosamine on the activity of the microsomal calcium pump. Microsomes prepared from the livers of male rats had greater calcium uptake activity than microsomes prepared from female rats. Calcium uptake activity of microsomes prepared from female rats given 2.5 ml of carbon tetrachloride/kg body wt by stomach tube 30 min prior to sacrifice were inhibited by 85%. These observations are consistent with previous reports (20, 21) and indicate our ability to detect differences in microsomal calcium pump activity. Effect of Galactosamine Calcium Pump Activity
It is helpful to compare the reaction of rat liver cells to different doses of n-galactosamine (1). A relatively low dose, 200 mg/kg, produces inhibition of RNA and protein synthesis with little or no cell death. Administration of a larger dose (400 mg/kg) produces liver cell death in addition. During the first 2 h, the calcium rises in both groups. However, between the second and third hours, there is a significant difference between the two doses. The calcium level continues to rise in the livers of TABLE
ACTIVE SEQUESTRATION OF CALCIUM BY MICROSOMES AND MITOCHONDRIA Microsomal calMitochondrial Conditions calcium uptake’ cium u take” (mu01 0P“Ca2+/ (nmol of ‘Ta*+/ mg of protein/45 mg of protein/l0 min) min) Control Galactosamine (200 mg/kg) Galactosamine (400 mg/kg)
77.0 2 8 82.0 k 4
203 + 12 230 + 12
02.2 k 4
220 + 4
a Rat liver microsomes were isolated and assayed for calcium uptake activity as described under Materials and Methods. Incubations were at 25°C for 45 min. All assays were linear for 60 min. Galactosamine at the doses indicated was given 2 h prior to sacrifice. Results are the means ? SD of separate preparations from three animals. *Rat liver mitochondria were isolated and assayed for calcium uptake activity as described under Materials and Methods. Incubations were at 25°C for 10 min. Results are the means ? SD of separate preparations from three animals.
those animals treated with the high dose, while it falls sharply to the control level in those animals treated with the low dose (1). Table II gives the calcium pump activity of microsomes isolated from animals treated with 200 and 400 mg/kg of galactosamine, respectively. In both cases there is no effect of the galactosamine treatment on the activities illustrated. The significance of these negative data with respect to the mechanism of Ca2+ accumlation in the injured cells necessitated several controls to ensure their validity. It was necessary to show that an effect on the microsomes was not masked by differences in the composition of the microsomal preparations obtained from the control and galactosamine-treated animals. Azide-resistant, calcium sequestration activity was measured on liver homogenates from which the microsomes are prepared. Homogenates from control animals had a specific activity of 1.52 * 0.19 nmol of Ca2+/10 min/mg wet wt of liver; with 200 mg/kg of galactosamine it was 1.42 ? 0.36, and with 400 mg/kg of galactosamine, 1.33 2 0.09. The glucose 6-phosphatase activity of control microsomes was 0.39 pm01 of POde3 liberated/min/mg of microsomal protein and 0.40 + 0.03 and 0.41 + 0.05 for microsomes prepared from animals given 200 and 400 mg/kg of galactosamine, respectively. Lastly, the cytochrome P-450 content of the respective microsomal preparations was measured. Again there were no significant differences between the control and experimental preparations: 0.645 ? 0.079 nmol of cytochrome P-45O/mg of microsomal protein for the controls and 0.788 + 0.106 and 0.663 -+ 0.102 for microsomes prepared from animals treated with 200 and 400 mg/kg of galactosamine, respectively. Calcium Uptake by Isolated Rat Liver Mitochondria
The Ca2+ uptake activity of isolated rat liver mitochondria, measured under the same conditions as the microsomes, is characterized in Table I. Several features distinguish mitochondrial from microsoma1 Ca2+ uptake. The initial rate of Ca2+ uptake by mitochondria per milligram of
protein is about tenfold higher than that of microsomes. There is a marked inhibition of the mitochondrial activity by 5 mM sodium azide. There is a dependency on added oxalate, but not quite as great as that of the microsomes. Calcium uptake activity of both mitochondria and microsomes is quite dependent on an exogenous source of ATP. Figure 2 illustrates the relationship between Ca2+ uptake activity and the concentration of total Ca2+in the assay. From the double-reciprocal plot of the same data, the apparent K, is 43.5 PM total calcium. Correcting for Ca2+bound to adenine nucleotides, this corresponds to a free calcium concentration of 5.7 PM. The data presented in Fig. 2 were based on the extent of calcium uptake in 5 min. The time course of the Ca’+ accumulation was linear at both a low (8 ,UM)and high (40 PM) total Ca2+ concentration for the 5 min used in these studies. Effect of Galactosamine on Mitochondrial Calcium Uptake Activity
Table II gives the uptake of Ca2+ by mitochondria isolated from control animals and animals treated with 200 and 400 mg/kg of galactosamine, respectively. As with the microsomal calcium uptake activity, there is no apparent effect of the galactosamine treatment on this in vitro activ225 t
FIG. 2. Dependency of mitochondrial calcium uptake on the concentration of total calcium. Mitochondria were isolated and assayed for calcium uptake activity as described under Materials and Methods. All values are corrected for the amount of *%a2+ bound to the mitochondrial membranes by subtracting the 0 time value. Results are the means ? SD of separate preparations from three animals.
ity of isolated mitochondria. Again, this is at a time (2 h) when there are readily detectable alterations in calcium homeostasis within the liver cells. DISCUSSION
The data presented in this paper would seem to exclude either a microsomal or a mitochondrial alteration as a factor contributing to the initial disturbance in Ca2+ homeostasis induced by galactosamine. Microsomes and mitochondria isolated from galactosamine-treated animals were just as active in the uptake of calcium as those from control animals. Furthermore, these negative results cannot be explained by our having analyzed dissimilar microsomal preparations. The calcium uptake activity of whole liver homogenates was the same from control and galactosaminetreated animals, measured so that mitochondrial calcium uptake was eliminated by 5 mM sodium azide. Two additional in vitro functions of the isolated microsomes were also unaffected. The similar cytochrome P-450 contents were important. This determination depends upon the formation and detection of stable complexes with carbon monoxide (26). The reaction is probably nonenzymatic and can be used as a measure of the amount rather than the activity of this particular function. The same content of cytochrome P-450 per milligram of membrane protein isolated from both control and galactosamine-treated animals allows interpretation of the glucose 6-phosphatase and calcium pump activities as showing no functional disturbance attributable to galactosamine. Confirming this observation is the recent report in which no effect of galactosamine (400 mg/kg) could be detected on bilirubin conjugation by microsomal glucuronyl transferase (29). The disturbance in calcium homeostasis produced by galactosamine, at least in its initial development, is probably entirely a consequence of the plasma membrane alterations described previously (1). The increased cell calcium content must result from an accelerated influx across the plasma membrane following breakdown of the marked permeability barrier that this membrane normally presents. Or, there
FARBER ET AL.
could exist an as yet uncharacterized active efllux mechanism localized in the plasma membrane affected by galactosamine. However, it has not been possible to detect a calcium-stimulated, magnesiumdependent ATPase in rat liver plasma membranes (19). The surface area of the endoplasmic reticulum (approximately 63,000 pm2/hepatocyte) is about 37 times that of the plasma membrane (30). Assuming that it can lead to Ca’+ extrusion from the cell, an active calcium efIlux mechanism in microsomal membranes makes it difficult to see the advantage of a similar pump in the plasma membrane. Injury to the plasma membrane alone contrasts galactosamine-induced liver cell injury with that produced by carbon tetrachloride. With Ccl, there is microsomal (21) as well as plasma membrane3 damage. Each agent produces liver cell death accompanied by marked disturbances in calcium homeostasis. However, the pattern of disturbed calcium metabolism is different, which possibly can be attributed to inhibition of the microsomal calcium pump by Ccl,. Calcium accumulation is biphasic following Ccl, poisoning, first reversibly entering almost all liver parenchymal cells between 15 min and 2 h, and then, between 4 and 8 h, increasing again to reach levels more than 20 times the normal value by the end of the first day (31, 32). The early, reversible increase in cellular calcium could result from the rapid inhibition of the microsomal calcium pump. The normal cytoplasmic Ca2+ concentration within rat liver cells is not known. However, based on the levels within other cells (33) it is probably lop7 to lo-” M. This is below the apparent Km and considerably below the level of Ca2+ that gives maximal pump activity. If the pump is inhibited less than quantitatively, the consequent rise in cellular Ca2+concentrations would raise the residual pump activity with elimination of the excess Ca2+and restoration of the steady state, as occurs during the first 2 h following administration of Ccl,. The subsequent, irreversible increases in Ca2+would represent the con3El-Mofty, S. K., and Farber, J. L., unpublished data.
sequence of the plasma membrane injury, as occurs with galactosamine. A recent report by van Rossum et al. (34) supports this interpretation. Inhibition of the active mechanism extruding calcium from liver cell slices by incubating them at 0°C permitted entry of Ca2+ into the cytosol from the exterior and its accumulation by the mitochondria, known to remain active at 0°C. Returning the slices to 38°C led to reactivation of the efflux mechanism (? the microsomal calcium pump) with prompt return of total cellular and mitochondrial Ca2+concentrations to the original values. This rise and subsequent decline in the level of calcium within the liver cells is analogous to the early alterations in calcium homeostasis induced by Ccl,. The two preceding examples and a third, the early rise and fall of the Ca2+ content with CCL, the restoration of a normal Ca2+ content in liver slices previously incubated at O”C, and the decrease in the elevated cell Ca2+ following uridine treatment of galactosamine-injured cells (1)) indicate that elevated total cellular, including mitochondrial , Ca2+ concentrations can be readily reversed. This would imply that cell Ca2+ is primarily regulated not by the mimchondria, which in all the above cases lose Ca2+, but by the permeability of the plasma membrane and an active efflux system, presumably the microsomal calcium pump. Several previous studies have suggested that mitochondria interact with Ca2+with a Km in the micromolar range. The uptake of Ca2+ by liver mitochondria takes primacy over the phosphorylation of ADP, even when the concentration of added Ca2+ is 10 PM or less, and that of added ADP is 100 times greater (9). Carafoli and Azzi (35) have estimated the affinity of liver mitochondria for Ca’+ using the oxidationreduction shift of cytochrome b and Ca2+EGTA buffers to obtain stabilized concentrations of free calcium. Under these conditions, they measured a K, of the order of 2 to 3 PM. Spencer and Bygrave (36) measured the affinity of the ATP-supported uptake system by assaying the initial rates of translocation as a function of both Ca2+ and ATP concentrations. After correcting for the Ca2+ complexed to adenine
nucleotides they calculated a value not far from 1 PM. Recently, however, Scarpa and Graziotti (37) have indicated that the Km for the energy-linked interaction of Ca2+ with heart mitochondria, measured spectrophotometrically with the murexide indicator, is between 1 and 2 orders of magnitude higher than that found by other methods. They could never measure Km values lower than 30 to 40 PM. Under comparable conditions, we have measured the kinetic properties of the Ca” uptake activity of mitochondria and microsomes. Mitochondria do not seem to be at any particular kinetic advantage over the microsomes with respect to calcium accumulation. The K,,, for free Ca2+ was 5.7 and 2.4 PM for mitochondria and microsomes, respectively. The V of the mitochondrial calcium pump is tenfold greater than that of the microsomes. However, it has been estimated that the surface area of the endoplasmic reticulum is about 8.5 times the surface area of the outer mitochondrial membrane (30). These numbers taken together would imply that in an intact cell the microsomes and mitochondria should compete equally for free Ca2+ ions. In contrast to the calcium-extruding system, liver mitochondria retain accumulated Ca2+ in a dynamic steady state, in which a presumably passive efflux of Ca2+ is counterbalanced by active influx of Ca2+ driven by resting or state 4 electron transport. Under limited loading conditions, irreversible sequestration of the accumulated calcium does not appear to take place (38). These considerations would suggest respective roles for the mitochondria and microsomes in the control of Ca2+ concentrations within liver cells. We are assuming that the active efIlux mechanism extruding Ca2+ from the cells is localized in the endoplasmic reticulum and corresponds to the in vitro sequestration activity of isolated microsomes. This envisions calcium eMux as a “secretion” analogous to the release of albumin. Under normal steadystate conditions, calcium influx by passive diffusion across the plasma membrane along its steep concentration gradient (33) is balanced by an equal rate of active ef-
flux. The cytoplasmic concentration and the total cell Ca2+ content are kept at a very low level. Changes in the permeability properties of the plasma membrane brought about by phsyiological (hormones, nutrition, etc.) or pathological (hepatotoxins, etc.) alterations in the cell environment produce rapid increases in the Ca2+ concentration in the cytoplasm. This increases the function of both mitochondrial and microsomal Ca2+ pumps by substrate activation. The Ca2+ content of the mitochondria rises, and the rate of active efflux increases, thereby lowering the critical level of the free Ca2+ in the cytoplasm. If the increased rate of influx is not too great, a new steady state will be reached where the rate of efflux will be equal to the rate of influx. The mitochondria will then lose their sequestered Ca2+ as the cytoplasmic concentration falls with stimulation of the efflux mechanism. The mitochondrial Ca2+ transport system can react to changes in Ca2+concentration in cytoplasm. Its action helps to lower acute elevation in cytoplasmic Ca”+. However, the mitochondrial Ca2+content is viewed as being a function primarily of the difference between the rates of Ca2+ influx and efllux, rising with changes in influx and falling as efIlux rises to equal and/or temporarily exceed influx. An irreversible and self-aggravating situation would result when the rate of efflux cannot rise to meet a new and persistent rate of influx. In this situation there would be a progressive rise in the concentration and content of total liver cell Ca2+ with marked sequestration of Ca”+ by the mitochondria. Massive calcium accumulation by the mitochondria ultimately interferes with their structure and function (39). A rising calcium concentration in the cytoplasm would also eventually lead to inhibition of the active efflux mechanism (Fig. 1). These effects on the mitochondria and microsomes produce a sort of positive feedback situation with a progressive loss of the ability to maintain calcium homeostasis characteristic of cell death. 1. EL-MOFTY, NICOLINI,
REFERENCES S. K., SZRUTTON, M. C., SERRONI, A., C., AND FARBER, J. L. (1975) Amer.
J. Path. 79, 579496. 2. FARBER,
Amer. J. Path. 81, 237-250. 3. EL-MOFTY, S. K., G~RIN, P., WENDLING, W. W., SERRONI, A., AND FARBER, J. L. (1977) Amer. J. Path., in press. 4. LEHNINGER, A. L. (1970) Biochem. J. 119, 129138. 5. CARAFOLI, E., AND LEHNINGER, A. L. (1971)Biothem. J. 122, 681-690. 6. CHEN, C. H., ANDLEHNINGER, A. L. (1973)Arch. B&hem. Biophys. 157,449-459. 7. CHEN, C. H., ANDLEHNINGER,A. L. (1973)Arch. Biochem. Biophys. 157, 183-196. 8. BRIERLY, G. P. (1963) in Energy-Linked Functions of Mitochondria (Chance, B., ed.), pp. 237-251, Academic Press, New York. 9. ROSSI, C. S., AND LEHNINGER, A. L. (1964) J. Biol. Chem. 239, 3971-3980. 10. CARAFOLI, E. (1967) J. Gen. Physiol. 50, 18491864. 11. CARAFOLI, E., PATRIARCA, P., AND ROSSI, C. S. (1969) J. Gen. Physiol. 74, 17-30. 12. HAUGAARD, N., HAUGAARD, E. S., LEE, N. H., AND HORN, R. S. (1969) Fed. Proc. 28, 16571662. 13. CARAFOLI, E. (1973) Biochimie 55, 755-762. 14. BORLE, A. B. (1973) Fed. Proc. 32, 1944-1950. 15. FISHER, A. B., SCARPA, A., LANOUE, K. F., BASSETT, D., AND WILLIAMSON, J. R. (1973) Biochemistry 12,1438-1445. 16. PENG, C. F., PRICE, D. W., BHUVANESWARAN, C., AND WADKINS, C. L. (1974) Biochem. Biophys. Res. Commun. 56, 134-141. 17. JUDAH, J. D., AND AHMED, K. (1964) Biol. Rev. 39, 160-193. 18. VAN ROSSUM, G. D. V. (1970) J. Gen. Physiol. 55, 18-32. 19. CHAMBUT, A. M., FRANCOISE, L. P., FELDMANN, G., AND HANOUNE, J. (1974) J. Gen. Physiol. 64, 104-126. 20. MOORE, L., CHEN, T., KNAPP, H. R., AND LANDON, E. J. (1975) J. Biol. Chem. 250, 45624568.
ET AL. 21. MOORE, L., DAVENPORT, G. R., AND LAND~N, E. J. (1976) J. Biol. Chem. 251, 1197-1201. 22. MARTONOSI, A., AND FRETAS, R. (1964) J. Biol. Chem. 239, 648-658. 23. ALONSO, G. L., BAZEQUE, P. M., AND ARRIGO, D. M. (1971) J. Gen. Physiol. 58, 340-350. 24. SCHNEIDER, W. C., AND HOGEBOOM, G. H. (1950) J. Biol. Chem. 183, 123-128. 25. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem. 193, 265-275. 26. OMURA, T., AND SATO, R. (1964) J. Biol. Chem. 239, 2370-2378. 27. SWANSON, M. A. (1955) in Methods in Enzymology (Colowick, N. O., and Kaplan, N. O., eds.), Vol. 2, pp. 541-543, Academic Press, New York. 28. KATZ, A. M., REPKE, D. I., UPSHAW, J. E., AND POLASCIK, M. A. (1970) Biochim. Biophys. A& 205, 473-490. 29. OTANI, G., ABOU-EL-MAKAREM, M. M., AND BOCK, K. W. (1976) Biochem. Pharmacol. 25, 1293-1297. 30. WEIBEL, E. R., STAUBLI, W., GNAGI, H. R., AND HESS, F. A. (1969) J. Cell Biol. 42, 68-91. 31. REYNOLDS, E. S. (1964) Lab. Invest. 13, 14571470. 32. SMUCKLER, E. A. (1966) Lab. Invest. 15,157-166. 33. RASMUSSEN, H. (1970) Science 170, 404-412. 34. VAN ROSSUM, G. D. V., SMITH, K. P., AND BEETON, P. (1976) Nature (London) 260, 335-337. 35. CARAFOLI, E., AND Azz~, A. (1972) Experientia 28, 906-908. 36. SPENCER, T., AND BYGRAVE, F. L. (1973) Bioenergetics 4, 347-362. 37. SCARPA, A., AND GRAZIOTTT, P. (1973) J. Gen. Physiol. 62, 756-772. 38. DRAHOTA, Z., CAF~AFOLI, E., ROSSI, C. S., GAMBLE, R. L., AND LEHNINGER, A. L. (1965) J. Biol. Chem. 240, 2712-2720. 39. GREENAWALT, J. W., ROSSI, C. S., AND LEHNINGER, A. L. (1964) J. Cell Biol. 23, 21-38.