Constitutive hsp70 Attenuates Hydrogen Peroxide-Induced Membrane Lipid Peroxidation

Biochemical and Biophysical Research Communications 265, 279 –284 (1999) Article ID bbrc.1999.1649, available online at http://www.idealibrary.com on

Constitutive hsp70 Attenuates Hydrogen PeroxideInduced Membrane Lipid Peroxidation Ching-Yuan Su, Kowit-Yu Chong, Kerry Edelstein, Sean Lille, Romesh Khardori, and Chen-Ching Lai Department of Medicine, Southern Illinois University School of Medicine, Springfield, Illinois 62794

Received October 7, 1999

Thermal pretreatment improves cardiac recovery from subsequent ischemia/reperfusion. Induction of heat shock proteins (hsps) may contribute to this protection. We have demonstrated that augmentation of the constitutive hsp70 (hsc70) in H9c2 heart myoblasts promotes oxidative resistance. We employed a model oxidant to explore potential target(s) of protection by hsc70. Upon exposure to 54 mM of hydrogen peroxide (H 2O 2), hsc70-overexpressing cells exhibited a lower lipid peroxidation than the sham-transfected control. Constituitive hsc70 overexpression, however, did not protect against H 2O 2-induced depletion of ATP and glutathione (GSH). Lipid protection also occurred in cells preconditioned at 39°C (selectively induces hsc70) during H 2O 2 exposure. Interestingly, the protection conferred by hsc70 was comparable in magnitude to that provided by a-tocopherol, and was followed with a reduced release of lactate dehydrogenase and a unaltered calcium uptake during H 2O 2 challenge. Collectively, our observations suggest that hsc70 may preserve membrane function via attenuation of lipid peroxidation during oxidative insult. © 1999 Academic Press

Key Words: lipid peroxidation; thiobarbituric acidreactive substances; 45Ca influx; ATP; lactate dehydrogenase; glutathione; constitutive 70-kD heat shock protein; hydrogen peroxide.

The restoration of blood flow following an ischemic event to the myocardium introduces a series of ultrastructural and molecular changes that can lead to cellular death. This inflicted, potentially reversible tissue damage has many clinical manifestations including myocardial infarction, postischemic function abnormalities and arrythmias. It has been postulated that an increased production of reactive oxygen species (ROS), such as the superoxide anion, H 2O 2 and the hydroxyl radical, are the predominant contributors to myocardial ischemia/reperfusion injury through either

direct toxic effects or stimulating signaling mechanisms which promote the induction of inflammatory mediators (1). Cardiomyocytes contain endogenous mechanisms for repairing or removing damaged cellular constituents during oxidative challenge. Augmentation of these mechanisms may have a clinical relevance to preserve the myocardium during ischemia/ reperfusion episodes. Thermal induction of a unique set of proteins, the hsps, has been shown to reduce cardiac dysfunctions following ischemia/reperfusion challenge (2). Moreover, artificial expression of a major heat-responsive hsp, the inducible hsp70, protects rat heart-derived H9c2 cells against simulated ischemia or hypoxia (3, 4), and this expression in transgenic mice limits post-ischemic myocardial injury upon reperfusion (5, 6). Using the H9c2 cell model, we have investigated whether other hsps play a similar protective role during ischemia/reperfusion. We found that pretreatment of cells with acute hyperthermia promoted H 2O 2 resistance, which was correlated with a simultaneous increase in the inducible hsp70 and its constitutive counterpart, hsc70 (7). Furthermore, genetic augmentation of intracellular hsc70 lowered cellular susceptibility to oxidative insult (8), and mild hyperthermia that selectively induced hsc70 promoted oxidative protection (9). These studies demonstrate that hsc70 is a protective protein against ROS-mediated cell injury. Our studies also indicate that hsc70 has a lower threshold for induction than the inducible hsp70 (9), and thus, have more clinical practicality due to the lower potential for generating adverse systemic mediators associated with the higher degree of hyperthermia required for the activation of the inducible hsp70. Constitutive hsc70, therefore, represents a novel protective mechanism to achieve oxidative protection. The pathogenic effects of H 2O 2 have been relatively well characterized. Once inside cells, this oxidant can be detoxified by either catalase or glutathione peroxidase at the expense of GSH. Excess intracellular H 2O 2 can also be transformed via the iron-catalyzed Fenton or

279

0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

Vol. 265, No. 2, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Harber-Weiss reaction into more reactive metabolites, such as the hydroxyl radical. It has been shown that H 2O 2 overload induces severe membrane lipid peroxidation (10) and GSH exhaustion (11) in cardiomyocytes. In addition, due to a rapid inhibition of glyceraldehyde-3-phosphate dehydrogenase (12) and mitochondrial pyruvate dehydrogenase by H 2O 2 and/or its metabolites (13), ATP depletion was also observed. The complexity of H 2O 2 pathophysiology offers an opportunity to explore the target(s) of protection by hsc70. In this study, we examined the differences in metabolic dysfunctions incited by H 2O 2 between hsc70overexpressing and control cells. The results demonstrate that an increased level of hsc70 protein preferentially attenuated membrane lipid peroxidation, but offered no protection against H 2O 2-induced ATP and GSH depletion. Therefore, we conclude that biological membranes represent a selective target for hsc70mediated protection during oxidative challenge. MATERIALS AND METHODS Cell cultures. H9c2 cardiac myoblasts were obtained from the American Type Culture Collection (CRL-1446, ATCC, Rockville, MD). The cells were maintained in Dulbecco’s modified Eagle’s medium (DME) (Gibco-BRL, Gaithersburg, MD) supplemented with 10% fetal bovine serum, sodium bicarbonate, and antibiotics under 95% air and 5% CO 2 at 37°C. The hsc70-overexpressing (H9/hsc70) and sham-transfected control (H9/sham) cells have been described previously (8). The H9/hsc70 cells harbored the entire coding region of rat hsc70 under the control of a human cytomegaloviral enhancerpromoter, whereas the H9/sham cells contained the empty expression vector. Both clonal cells were maintained in the presence of G418 (Gibco-BRL) and repeatedly tested for hsc70 expression using western blot analysis. For oxidative challenge, cells were plated at a density of ;2 3 10 4 cells/cm 2 in 100 3 20 mm culture dishes or 6-well culture plates and used within 2–3 days. Thermal preconditioning was conducted in a 39 6 0.1°C cell incubator for 3 days prior to oxidative challenge (9). As for the pretreatment with vitamin E, dl-a-tocopherol (Sigma, St. Louis, MO) was dissolved in ethanol and added to the culture medium at the indicated concentrations for 24 h. The control was added with ethanol (0.1% v/v) for the same duration. Oxidative challenge. The preparation of H 2O 2-containing DME was as mentioned previously (8, 9). Cells were rinsed with phosphate buffered saline (PBS) and then exposed to serum-free DME containing H 2O 2 at 37°C for the indicated times. At the end of the treatment, the medium was collected for the measurements of lactate dehydrogenase (LDH) release and thiobarbituric acid-reactive substance (TBARS) formation. The cells were quickly rinsed with PBS twice and used for the remaining biochemical assays, such as the measurements of ATP, GSH, and protein content. A portion of cells was not exposed to H 2O 2, but was incubated with serum-free DME for the same period (1–3 h). The formation of TBARS, release of LDH, etc. in the latter cells was assessed in the same fashion. The derived data were defined as the untreated value. Protein, ATP, and GSH measurements. For protein assay, cells were scraped into 1% triton X 100-containing PBS followed with sonication. Subsequently, the protein concentration in the resultant lysate was determined using a BCA Protein Assay Reagent Kit (Pierce, Rockford, IL). For ATP or GSH assay, the cells were rapidly lysed in ice-cold 8% perchloric acid, neutralized with 1.75 M tribasic potassium phosphate, and centrifuged for 20 min at 10000 3 g to collect the supernatant. ATP content in the supernatant was as-

sessed using a luciferin/luciferase-based ATP assay kit (Molecular Probes, Inc., Eugene, OR), while that of GSH was measured by performing a highly specific enzymatic assay, which follows the formation of S-lactoyl glutathione from methylglyoxal catalyzed by glyoxlase I at 240 nm (14). The data of ATP and GSH are expressed as nmol/mg protein. Consistent with a prior report by Ishikawa and Sies (15), we found that oxidized glutathione was transported out of the cells. The intracellular level of this metabolite was, hence, not measured. Assay of LDH activity. The activity of LDH present in the collected medium was assayed as described (9). The total activity of LDH in untreated cells was also determined. LDH leakage is expressed as percentage of total LDH activity. Assay of TBARS. Almost all of the accumulated TBARS during H 2O 2 exposure were released into an extracellular space (16, and our observation). Therefore, we measured only TBARS present in the medium. The conditions previously reported by Wey et al. (16) were followed with some modifications. In brief, catalase (5000 U/ml) and butylated hydroxytoluene (0.01%) were added to the medium to terminate H 2O 2 toxicity and the progression of lipid peroxidation, respectively. Subsequently, ice-cold trichloroacetic acid was added at a final concentration of 5.5% to precipitate cellular proteins. The whole mixture was centrifuged at 1000 3 g for 10 min. The clear supernatant was transferred to a tube followed with an addition of an equal volume of thiobarbituric acid (0.6%) plus EDTA (1 mM). After boiling for 20 min, TBARS in the mixture were extracted with 4 ml of n-butanol/pyridine (15:1, v/v). Fluorescence was measured in a Perkin-Elmer LS-50 Spectrofluorometer with excitation wavelength of 515 nm and emission wavelength of 553 nm. A standard curve for TBARS was generated by dissolving tetraethoxypropane in ethanol followed with a serial dilution in 0.01 N HCl. The level of TBARS in serum-free DME was also measured. Data after subtraction of the background were expressed as nmol TBARS/mg protein. Measurement of 45Ca uptake. The conditions previously employed by Murphy et al. (17) were followed with minor modifications. For this experiment, the preceding oxidative challenge was performed in the presence of H 2O 2-containing DME buffered with N-2-hydroxyethyl-piperazine-N9-2-ethanesulfonic acid (HEPES) (25 mM at pH 7.4). At the indicated times, the H 2O 2-containing DME was aspired and replaced with HEPES-buffered DME supplemented with 45Ca 21 (5 mCi/ml, specific activity ;30 mCi/mg, Dupont NEN, Boston, MA) prewarmed at 37°C. The cells were allowed to incubate with Ca 21 for 5 min to label the rapidly exchangeable pool. Afterward, the 45Ca 21containing DME was aspired, and the cells were washed five times for 15 sec each with an ice-cold HEPES-buffered balanced salt solution, consisting of 109.5 mM NaCl, 5.4 mM KCl, 1.36 mM CaCl 2, 0.8 mM MgSO 4, and 1 mM of lanthanum chloride. Lanthanum chloride was added to displace 45Ca 21 from the interstitial space and to rapidly inhibit calcium influx. The cells were then dissolved in 0.1% of SDS. A portion of the resultant lysate was used for the measurement of protein content as mentioned above, while the rest was transferred to a scintillation tube, to which 10 ml of Scintisafe Plus 50% (Fisher, Chicago, IL) was added, and the incorporated radioactivity was measured. Statistics. The data were expressed as mean 6 S.D. (n 5 4 – 8). Statistical differences were analyzed by Student’s t-test for unpaired data.

RESULTS Treatment of H9c2 cells with H 2O 2 at graded doses resulted in a similar pattern of metabolic dysfunctions (preliminary work). A moderate dose of H 2O 2 (54 mM) was chosen, since it allowed resolution of the induced metabolic dysfunctions within a reasonable time

280

Vol. 265, No. 2, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. Overexpression of hsc70 inhibited lipid peroxidation during H 2O 2 challenge. Stable clones that either overexpressed hsc70 protein (H9/hsc70) or harbored an empty expression vector (H9/ sham) were subjected to serum-free DME containing 54 mM of H 2O 2 for various times. At the end of exposure, both the medium and cells were harvested and processed as described in Materials and Methods for the measurements of TBARS (A), GSH (B), ATP (C), and total proteins. In parallel, a portion of the clonal cells was not exposed to H 2O 2, but was incubated with serum-free DME for the same duration of time (1–3 h). (U) Data derived from these H 2O 2-untreated cells. Mean 6 S.D. (n 5 3–5). * indicates the differences between the hsc70 and control cells at P , 0.05, and ** at P , 0.01.

frame. Figure 1 shows that H 2O 2 overload provoked a time-dependent formation of TBARS, a rapid decline of cellular GSH, as well as a drastic depletion of the high-energy phosphate ATP. Compared with the H9/ sham control, the H9/hsc70 cells exhibited a slightly lowered formation of TBARS prior to oxidative challenge (Fig. 1A). Upon oxidative challenge, a distinct lipid protection by hsc70 overexpression was visible. TBARS formations in the H9/sham and H9/hsc70 cells were 1.11 6 0.20 and 0.65 6 0.08 nmol/mg, respectively, after the first hour of H 2O 2 treatment and 1.93 6 0.10 and 0.91 6 0.15 nmol/mg, respectively, after the second hour of treatment. For both groups of cells, the accumulation of TBARS after the third hour

of H 2O 2 treatment was not significantly different from that detected after the second hour, indicating that membrane lipid oxidation may reach plateau after 2 hours of incubation of cells with H 2O 2. In contrast to the observed lipid protection, hsc70 overexpression did not diminish the damaging effect of H 2O 2 on cellular ATP or GSH. In both H9/sham and H9/hsc70 cells, ATP concentration declined by ;85% during the first hour of H 2O 2 exposure; afterward, this high-energy phosphate pool was completely drained (Fig. 1C). Although less drastic, GSH was also remarkably exhausted after a prolonged incubation of both groups of cells with H 2O 2 (Fig. 1B). This suggests that hsc70 overexpression does not afford a general protection against all metabolic dysfunctions evoked by H 2O 2. It is of interest to know whether hsc70 augmentation by other approaches provides a similar lipid protection. To assess this, parental H9c2 cells were preconditioned at 39°C for 3 days, which selectively induces the hsc70 protein (9). Prior to oxidative challenge, the basal formation of TBARS was similar for both preconditioned and unconditioned cells (0.23 6 0.06 vs. 0.29 6 0.02 nmol/mg). After H 2O 2 challenge for 2 h, the accumulation of TBARS in the preconditioned cells was considerably lower than that of the unconditioned cells (1.28 6 0.04 vs. 2.38 6 0.46 nmol/mg, P , 0.05). Based upon this observation and that of the hsc70overexpressing cells, we conclude that lipid protection is associated with an increased level of hsc70 protein. We next examined whether hsc70-mediated attenuation of lipid peroxidation improves membrane integrity during oxidative challenge. Figure 2A shows that H 2O 2 treatment promoted in the H9/sham cells a marked release of LDH. This enzyme leakage was temporally linked to, and correlated with the advanced lipid peroxidation observed in the cells (Fig. 1A). In contrast, LDH leakage was delayed and reduced in severity in the H9/hsc70 cells after the same challenge, which was again in accordance with the preceding observation of diminished lipid peroxidation in the H9/ hsc70 cells (Fig. 1A). Lipid protection by hsc70 may also offer an advantage to maintain cellular ion homeostasis. Before oxidative challenge, both H9/sham and H9/hsc70 cells exhibited a low basal level of calcium uptake (Fig. 2B). After exposure to H 2O 2 for 3 h, the H9/sham cells had a large increase in calcium influx, indicating severe membrane damage. This damage, however, was absent in the H9/hsc70 cells. It should be noted that for this experiment, cells were removed from a 37°C incubator for aspiration of preexisting medium and the addition of radioisotope-containing medium. Although efforts had been attempted to prewarm the fresh medium at 37°C, temperature control during the immediately followed short-term (5 min) measurement of calcium influx may not be as stringent as a continuous incubation

281

Vol. 265, No. 2, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1

Metabolic Profile in a-Tocopherol-Pretreated Cells after H 2O 2 Exposure

Vehicle a-Tocopherol

LDH release (% of total)

ATP content (nmol/mg)

GSH content (nmole/mg)

48 6 6 11 6 2**

1.3 6 2.3 3.0 6 2.6

7.3 6 6.7 5.0 6 4.6

Note. The H9/sham cells were preloaded with 20 mM of a-tocopherol or vehicle (0.1% ethanol) for 24 h. Subsequently, these cells were rinsed with PBS twice and then subjected to 54 mM of H 2O 2 for 2 h. The resultant changes in cellular ATP and GSH as well as LDH release were determined as mentioned previously. Mean 6 S.D. (n 5 3). ** at P , 0.01.

DISCUSSION

FIG. 2. Overexpression of hsc70 reduced LDH leakage and calcium influx during H 2O 2 challenge. Two more cell injury parameters were employed to specifically assess membrane integrity in the H9/ sham and H9/hsc70 cells during oxidative insult. (A) The time course of LDH activity released from the H 2O 2-treated cells. (B) The time course of calcium uptake by the H 2O 2-treated cells. (U) Data derived from the H 2O 2-untreated clonal cells. Mean 6 S.D. (n $ 5). ** at P , 0.01. Note that both LDH leakage and calcium influx in the control cells were detected after TBARS accumulation, suggesting that the former two events may be the result of lipid peroxidation-mediated membrane damage.

of cells at 37°C. Therefore, the onset and/or severity of calcium overload in the target cells may be underestimated. Finally, the efficacy of membrane protection conferred by hsc70 was compared with that contributed by a-tocopherol. a-Tocopherol has an antioxidant action to inhibit membrane lipid peroxidation by scavenging lipid peroxyl radicals, and has been shown to have an effect to limit myocardial ischemia/reperfusion injury (18). The H9/sham cells were preloaded with a-tocopherol at 0 –20 mM for 24 h, and then subjected to H 2O 2 exposure accordingly. We found that a-tocopherol ameliorated H 2O 2-induced LDH leakage (Table 1) and TBARS formation (Fig. 3), but exerted no beneficial effect against the decline of GSH and ATP provoked by H 2O 2 exposure (Table 1). Based upon the data of TBARS formation, we conclude that a 2-fold overexpression of hsc70 in the H9/hsc70 cells (8) protects against lipid peroxidation as effective as preloading of the cells with 5–10 mM of a-tocopherol (Fig. 3). Therefore, augmentation of hsc70 protein is a sufficient means to preserve the membrane during oxidative insult.

In this study, we explored the potential target(s) of protection by hsc70 in H 2O 2-treated H9c2 heart myoblasts. H9c2 myoblasts mimic adult cardiomoycytes in several aspects (19, 20). Upon oxidative insult, these cells respond similarly to cardiomyocytes from cell culture or isolated heart models. Similar to the observations in the more complicated myocardial models, H 2O 2 evoked severe lipid peroxidation, ATP depletion, and GSH loss in H9c2 cells. The drastic fall of ATP in the H9c2 cells after one hour of H 2O 2 exposure (Fig. 1C) supports Chatham and coworkers’ (12) findings that H 2O 2 (150 mM) perfusion of the isolated rat heart for 40 min reduced ATP store by 75%. In addition, the kinetics of TBAR formation and LDH release in the H 2O 2treated H9c2 cells (Figs. 1A and 2A) were consistent

FIG. 3. The antiperoxidative effect of hsc70 overexpression was comparable to that achieved by a-tocopherol. The control clonal cells were preloaded with either vehicle (0.1% ethanol) or a-tocopherol as previously mentioned. Subsequently these cells were subjected to 54 mM of H 2O 2 for 2 h. The degrees of attenuation of TBARS formation by various doses of a-tocopherol were used to estimate the biopotency of a 2-fold increment in hsc70 protein. Mean 6 S.D. (n 5 3).

282

Vol. 265, No. 2, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

with those shown by Janero et al. (21) using cultured cardiomyocytes and 50 –250 mM of H 2O 2. Furthermore, 10 –20 mM of a-tocopherol has been demonstrated to protect cultured cardiomyocytes against lipid peroxidation evoked by H 2O 2 (21) or other ROS (22). We also found that these doses of a-tocopherol abolished H 2O 2mediated TBARS formation in the H9c2 cells (Fig. 3). These findings suggest that the oxidative sensitivity of H9c2 cells resembles that of cardiomyocytes or the whole heart. Therefore, the information derived from this cell model should have value for extrapolating hsc70-mediated protection in the more complex, mature myocardial milieu. The demonstration of lipid protection by hsc70 overexpression is of particular interest. In general, hsc70 has been recognized as a chaperone assisting the folding of nascent polypeptides or the disassembly of damaged proteins by metabolic stress (23). Our data, nevertheless, indicate an additional lipochaperoning effect of hsc70. It is noteworthy that hsc70 overexpression coordinately reduced TBARS formation and abolished H 2O 2 induced LDH release and calcium uptake. Calcium influx after ischemia/reperfusion predisposes the myocardium to arrhythmia and facilitates hypercontraction (24), and LDH leakage is a well recognized index for irreversible myocardial injury (25). Our findings, thus, may have clinical implications regarding the management of myocardial injury during ischemia/ reperfusion. Our prior work indicates that hsc70 overexpression affects neither the endogenous concentrations of other hsps, such as hsp27, hsp60, hsp70, and hsp90, nor the activity of catalase or glutathione peroxidase (8 and unpublished observations). Indeed, there was no visible difference regarding GSH depletion or ATP exhaustion between the H9/sham and H9/hsc70 cells after H 2O 2 challenge, implicating that detoxification of intracellular H 2O 2 may proceed at a similar rate in both groups of cells, and that hsc70 overexpression may not interfere with H 2O 2 transformation into more reactive metabolites. Under such conditions, it is intriguing that hsc70 augmentation by either genetic manipulation or thermal preconditioning confers lipid protection. This pattern of oxidative protection by hsc70 is similar to that afforded by a-tocopherol, which attenuated TBARS formation but not ATP/GSH depletion. Strikingly, the magnitude of lipid protection conferred by a moderate (2-fold) increment in hsc70 approximated that provided by pretreatment of cells with 5–10 mM of a-tocopherol. It has been reported that in rat hepatocytes, preincubation with these doses of a-tocopherol enriched the cellular content of this antiperoxidant by 30 – 60 fold (26). This information suggests that hsc70 is a protein of great biopotency that either directly or indirectly defends cells against lipid oxidative modification. At present, the exact mecha-

nism by which hsc70 protein achieves lipid protection is not clearly known. A few hsps have been documented to protect membrane lipids during oxidative challenge. Liu et al. (27) showed that preinduction of the 78- (grp78) and 94-kD (grp94) glucose regulated proteins rendered renal epithelial cells less susceptible to lipid peroxidation induced by iodoacetamide. It is suggested that these hsps modulate endoplasmic reticulum calcium store to maintain calcium homeostasis, thereby inhibiting mitochondrial calicum uptake, ROS formation and membrane peroxidation. In addition, Mehlen and coworkers (28) illustrated that overexpression of hsp27 and aBcrystallin in murine L929 fibroblasts conferred lipid protection evoked by tumor necrosis factor. The presence of these hsps raises total GSH level. It is thought that these small hsps abolish tumor necrosis factorinduced bursts of ROS at the expense of GSH. More recently, Wong et al. (29) have reported that overexpression of the inducible hsp70 lowered hyperoxiamediated lipid peroxidation and ATP loss in human lung adenocarcinoma A549 cells. The protective basis for the latter hsp, however, remains to be elucidated. Apparently, hsc70 protein is one of the hsps involved in lipid protection during oxidative stress. To our knowledge, hsc70 has not been demonstrated to have a direct effect to scavenge ROS in a cell-free system. Instead of acting like the above-mentioned grp78, grp94, and small hsps to buffer endoplasmic reticulum calcium or elevate cellular GSH, hsc70 appears to preferentially “intercept” the toxicity of ROS near or in the membrane. Several lines of evidence support the view of hsc70 as a lipochaperone. First, the hsc70 protein can bind to biological membranes (30, 31). The portion of membrane-bound hsc70 is increased during cellular stress such as heat shock (32). Recently, To¨ro¨k et al. (33) demonstrated that an association between artificial membranes and the bacterial homologue of hsp60, GroEL, increases the structural stability of the lipid bilayer. Both hsc70 and hsp60 proteins are involved in mitochondrial protein transfer and/or folding (23). Presumably, hsc70 may share this property of hsp60 to raise lipid order, and the increased lipid order slows down the interaction between oxidizable lipids and ROS and curtails the progression of lipid peroxidation. In addition, H 2O 2 overload triggers the activation of lipases, including phospholipase A 2 (34). The highly homologous inducible hsp70 has been shown to inhibit phospholipase A 2 activation by tumor necrosis factor (35). We speculate that hsc70 may interfere with the function of this type of membrane enzyme. If hsc70 is capable of blocking H 2O 2-mediated lipase activation, the susceptibility of free fatty acids to the action of ROS will be reduced, thereby limiting lipid peroxidation as well. In summary, we have demonstrated that hsc70 overexpression can selectively stop H 2O 2 induced cytotoxic

283

Vol. 265, No. 2, 1999

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

cascade at the biological membrane. This membrane lipid protection is independent of the cellular status of GSH or ATP, but is associated with an improved membrane integrity to retain cellular enzymes or exclude undesirable ion influx. The fact that hsc70 is an effective protein to curtail lipid peroxidation implicates that this protein may act through some powerful mechanisms to interfere with the reaction of lipid oxidation. Further investigation is needed to elucidate the exact mechanism underlying hsc70-mediated lipid protection. ACKNOWLEDGMENTS Sean Lille is a scientist from the Department of Surgery, Southern Illinois University School of Medicine. This work is supported by Grant HL52214-04 from National Institute of Health.

REFERENCES 1. Opie, L. H. (1991) in The Heart: Physiology and Metabolism (Opie, L.H., Ed.), pp. 469 – 483, Raven Press, New York. 2. Currie, R. W., Karmazyn, M., Kloc, M., and Mailer, K. (1988) Circ. Res. 63, 543–549. 3. Mestril, R., Chi, S-H., Sayen, M. R., O’Reilly, K., and Dillmann, W. H. (1994) J. Clin. Invest. 93, 759 –767. 4. Heads, R. J., Yellon, D. M., and Latchman, D. S. (1995) J. Mol. Cell Cardiol. 27, 1669 –1678. 5. Marber, M. S., Mestril, R., Chi, S-H., Sayen, M. R., Yellon, D. M., and Dillmann, W. H. (1995) J. Clin. Invest. 95, 1446 –1456. 6. Plumier, J. C., Ross, B. M., Currie, R. W., Angelidis, C. E., Kazlaris, H., Kollias, G., and Pagoulatos, G. N. (1995) J. Clin. Invest. 95, 1854 –1860. 7. Su, C.-Y., Chong, K.-Y., Owen, O. E., Dillmann, W. H., Chang, C., and Lai, C.-C. (1998) J. Mol. Cell Cardiol. 30, 587–598. 8. Chong, K.-Y., Lai, C.-C., Lille, S., Chang, C., and Su, C.-Y. (1998) J. Mol. Cell. Cardiol. 30, 599 – 608. 9. Su, C.-Y., Chong, K.-Y., Chang, J.-X., Ryter, S., Khardori, R., and Lai, C.-C. (1999) J. Mol. Cell. Cardiol. 31, 845– 855. 10. Davies, S. W., Underwood, S. M., Wickens, D. G,. Feneck, R. O., Dormandy, T. L., and Walesky, R. K. (1990) Br. Heart J. 64, 236 –240. 11. Weisel, R. D., Mickle, D. A., Finkle, C. D., Tumiati, L. C., Madonik, M. M., Ivanov, J., Burton, G. W., and Ingold, K. U. (1989) Circulation 80, 14 –18. 12. Chatham, J. C., Gilber, H. F., and Radda, G. K. (1989) Eur. J. Biochem. 184, 657– 662.

13. Janero, D. R., Hreniuk, D., and Sharif, H. M. (1994) Am. J. Physiol. 266, C179 –C188. 14. Bergmeyer, H. U. (1974) in Methods of Enzymatic Analysis (Bermeyer, H. U., Ed.), pp. 481– 482. Academic Press, New York. 15. Ishikawa, T., and Sies, H. (1984) J. Biol. Chem. 259, 3838 –3843. 16. Wey, H. E., Pyron, L., and Woolery, M. (1993) Toxicol. Appl. Pharmacol. 120, 72–79. 17. Murphy, J. G., Smith, T. W., and Marsh, J. D. (1987) J. Mol. Cell. Cardiol. 19, 271–279. 18. Massey, K. D., and Burton, K. P. (1989) Am. J. Physiol. 256, H1192–H1196. 19. Hescheler, J., Meyer, R., Plant, S., Krautwurst, D., Rosenthal, W., and Schultz, G. (1991) Circ. Res. 69, 1476 –1486. 20. Gerrelli, D., Huntriss, J. D., and Latchman. D. S. (1994) J. Mol. Cell Cardiol. 26, 713–719. 21. Janero, D. R., Hreniuk, D., and Sharif, H. M. (1991) J. Cell. Physiol. 149, 347–364. 22. Burton, K. P., Morris, A. C., Massey, K. D., Buja, L. M., and Hagler, H. K. (1990) J. Mol. Cell. Cardiol. 22, 1035–1047. 23. Su, C.-Y., Chang, C., and Lai, C.-C. (1996) in Pharmacology in Exercise and Sports (Somani, S. M., Ed.), pp. 147–167. CRC. 24. Atar, D., Gao, W. D., and Marban, E. (1995) J. Mol. Cell Cardiol. 27, 783–791. 25. Kehrer, J. P., and Starnes, J. W. (1989) Pharmacol. Ther. 44, 123–145. 26. Glascoatt, P. A. Jr., and Farber, J. L. (1999) Methods Enzymol. 300, 78 – 88. 27. Liu, H., Bowes, R. C. III., van de Water, B., Sillence, C., Nagelkerke, J. F., and Stevens, J. L. (1997) J. Biol. Chem. 272, 21751–21759. 28. Mehlen, P., Kretz-Remy, C. F., Pre´ville, X., and Arrigo, A-P. (1996) EMBO J. 15, 2695–2706. 29. Wong, H. R., Menendez, I. Y., Ryan, M. A., Denenberg, A. G., and Wispe´, J. R. (1998) Am. J. Physiol. 275, L836 –L841. 30. Hughes, E. N., and August, J. T. (1982) Proc. Natl. Acad. Sci. USA 79, 2305–2309. 31. Lithgow, T., Ryan, M., Anderson, R. L., Høi, P. B., and Hoogenraad, N. J. (1993) FEBS Lett. 332, 277–281. 32. Gudi, T., and Gupta, C. M. (1993) J. Biol. Chem. 268, 21344 – 21350. 33. To¨ro¨k, Z., Horva´th, I., Goloubinoff, P., Kova´cs, E., Glatz, A., Balogh, G., and Vı´gh, L. (1997) Proc. Natl. Acad. Sci. USA 94, 2192–2197. 34. Chakraborti, S., Gurtner, H., and Michael, J. R. (1989) Am. J. Physiol. 257, L430 –L437. 35. Ja¨a¨ttela¨, M. (1993) J. Immunol. 151, 4286 – 4294.

284