reoxygenation injury to rat neonatal cardiac myocytes

Vol. 179, No. 2, 1991 September

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 798-803

16, 1991

PROTECTION

BY ACIDOTIC pH AGAINST ANOXLWREOXYGENATION RAT NEONATAL CARDIAC MYOCYTESf

INJURY

TO

John M. Bond, Brian Herman and John J. Lemasters2 Laboratories for Cell Biology, Department of Cell Biology & Anatomy, and Curriculum Toxicology, University of North Carolina at Chapel Hill, NC 27599 Received

August

1,

in

1991

We assessed the effect of acidosis on cell killing during anoxia and reoxygenation in cultured rat neonatal cardiac myocytes. After 4.5 hours of anoxia and glycolytic inhibition with Z-deoxyglucose, loss of viability was >90% at pH 7.4. In contrast, at pH 6.2-7.0, viability was virtually unchanged. To model changes of pH and oxygenation during ischemia and reperfusion, myocytes were made anoxic at pH 6.2 for 4 hours, followed by reoxygenation at pH 7.4. Under these conditions, reoxygenatton precipitated loss of viability to about half the cells. When pH was increased to 7.4 without reoxy enation, similar lethal injury occurred. No cell killin occurred after reoxygenation at pH t .2. We conclude that actdosis protects against letha Iganoxic injury, and that a rapid return from acidotic to physiologic pH contributes significantly to reperfusion injury to cardiac myocytes - a ‘pH paradox’. 0 1991Academ. Press, Inc.

A key event during myocardial ischemia is tissue acidosis which results from lactic acid accumulation and hydrogen ion release during ATP hydrolysis. Previous work from this and other laboratories demonstrated that acidosis protects hepatocytes and other non-muscle cells against cell death during anoxia, metabolic inhibition, and exposure to toxic chemicals (l-4). Protection is mediated by intracellular acidification (5). Early reports suggested that acidosis may also protect during myocardial ischemia (6,7), but most workers consider acidosis harmful because it disrupts normal myocardial physiology (8). Reperfusion injury is the exacerbation of tissue damage when blood flow is restored to an ischemic organ. A marked reperfusion injury occurs to ischemic myocardium, whose mechanisms remain unclear. Generation of free radicals (9), critical depletion of ATP lev-

‘This work was supported, in art, by Grants from the Office of Naval Research, the National Institutes of Health, and t Ii e Gustavus and Louise Pfeiffer Research Foundation. Portions of this work were presented at the 74th Annual Meetin of the Federation of American Societies for Experimental Biology, Washington, DC, Apt-i B l-5,1990 (24). 2To whom correspondence should be addressed. Abbreviations used: KRH Krebs-Rin er-HEPES buffer containin 115 mM NaCl, 5 mM KCl, 1 mM KH2PO4,1.2 n&f MgSO4, !mM CaQ, and 25 mM Na-!IEPES buffer, pH 7.4. WO6-291X/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

798

Vol.

179, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

els (lo), disruption of normal Ca*+ homeostasis (7), unmasking of latent injury (ll), and neutrophil infiltration (12) all may contribute to repel-fusion injury. During reperfusion of ischemic myocardium, both reoxygenation and a rapid return to physiologic pH occur. Much attention has been given to events associated with reoxygenation, but relatively little to changes of pH. Accordingly, the objective of this study was to assessthe role of changes of pH in anoxia/reoxygenation injury. To this end, we developed an in vitro model of reperfusion injury that simulates the changes of oxygenation and pH that occur during ischemia and reperfusion.

MATERL4LS

AND METHODS

Isolation and culhue of myocytes - Cardiac myocytes from 2-3 day old rat neonates were dispersed by digestion with pancreatin and collagenase (13), and purified by flowelutriation (14), as previously described (15 . Punfied myocytes were cultured on polystyrene coverslips at a density of 106 cells/3 z mm culture dish in Eagle’s MEM fortified with 5% fetal calf serum, 10% horse serum, 10 U/ml penicillin, and 10 &ml streptomycin at 37°C in humidified sir/5% carbon dioxide. Myocytes contracted spontaneously after about 3 days in culture and began to show striations after 5 days. At this time, the majority of cells exhibited vigorous, synchronous contractions at rates exceeding 60 beats/minute. Anoxia and reoaygenation - Myocytes were mounted in an airtight perfusion chamber and infused with an anoxic suspension of submitochondrial particles (1 mg of protein/ml) and succinate (5 mM) in KRH buffer, as previously described (16). Respiration by submitochondrial particles actively removed oxygen entering the chamber by back diffusion from the atmos here. Although respiration by submitochondrial particles was inhibited by approximately s 0% at pH 6.2, their concentration was sufficiently high to maintain oxygen at below limits of detectton of a Clark-type oxygen electrode (cl torr) at all values of H studied. In these experiments, Zdeoxyglucose (20 mM) was added to inhibit glycolysis. F or reoxygenation, submitochondrial partrcles were replaced with aerobic buffer not containing succinate or 2-deoxyglucose. In some experiments, anoxic submitochondrial particles were infused a second time to change pH without reoxygenation. Loss of cell viability was monitored by fluorescent labelling of nuclei with propidntm iodide (5 PM) (16,17). At the end of each experiment, digitonin (30 PM) was added to label nuclei of all cells in order to calculate percent loss of viability. Temperature was regulated at 37°C with an air curtain incubator (Laboratory Products, Boston, MA). Microscopy - Phase contrast and fluorescence images were collected with a Zeiss IM3.5 inverted microscope (Thornwood, NY). Propidium iodide fluorescence was imaged with 546~nm excitation, 580~nm dichroic, and 590-nm emission filters. Experiments were routinely recorded wrth a half-inch time-lapse video cassette recorder. h4ateriaZ.s- Eagle’s MEM, enicillin and stre tomycin were obtained from the Univer. of North Carolma Tissue 8 ulture Center ( & ape1 Hill, NC); pancreatin from Gibco 78and Island, NY); collagenase D from Boehringer Mannheim (Indianapolis, IN); and 2deoxyglucose from Sigma (St. Louis, MO). Other reagent-grade chemicals were obtained from standard commercial sources.

RESULTS Loss of viability of purified myocytes during anoxia was assessed at various values of extracellular pH. Cultured myocytes on polystyrene cover-slips were mounted into a gastight chamber. Initially, cells were pre-equilibrated in aerobic KRH, pH 7.4, for lo-20 minutes. During this period, myocytes exhibited vigorous, spontaneous contractions. Anoxia was produced by infusing an anoxic suspension of submitochondrial particles and succinate in KRH buffer at pH values ranging between 6.2 and 7.4. After l-2 minutes of anoxia, spon799

Vol.

179, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

012345 Time

(hr)

F&J. Rote&n againstanoxickilling of culturedneonatalcardiacmyocytesby acidotic extracelhdarpH- Culturesof 6-12 day old myocyteswere mountedin a gas-tightchamber

and incubated in aerobic KRH buffer, pH 7.4, for 10-15 minutes. Anoxia was started by in-

fusionof an anoxicsuspension of submrtochondrialparticles(1 m of protein/ml succinate 5 mM , and Zdeoxyglucose(20 mM) in KRH buffer at pH va!ues betweentand74 by pro idium iodide stainin of myocyte nuclei. Ckli Loss od cell viability was assessed killing at H 6.2-7.0wassignificantlylesstKan at pH 7.4 (p
taneous contractions ceased, and the cells entered a quiescent phase that lasted throughout the experiment. After 4.5 hours at pH 7.4, more than 90% of nuclei were labelled with propidium iodide (Fig. 1). Nuclear labelling was progressive and nearly linear with time. 50% loss of viability occurred after about 3 hours. By contrast, at acidotic pH (pH 6.2-7.0), few nuclei took up propidium iodide (Figs. 1 and 2A). Binding of propidium iodide to nuclear chromatin was not itself pH-dependent, as nuclei labelled brightly after addition of digitonin over the entire range of pH studied (Fig. 2B). This concurs with previous studies in hepatocytes (3).

,Fi& 2. Propidiumiodidelabellingof anoxicmyocytes- 12-da my submrtochondrial articles,succinate,and Zdeoxyglucosein K&-I bxF;:e8;:d:: scribedin Fig. 1. xft er 5 hours,labellin of nucleiwith ropidium iodide 2 FM) wasminimal PanelA). Subseuent addition of 5tgttonin (30 p l&f) led to stainingoI all nuclei (Panel 9. k epresentativefie3dsare shown.

800

Vol.

179,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

I

60 -

I

0 Reoxy, A Anoxio. n Reoxy,

RESEARCH

COMMUNICATIONS

I

pH 7.4 pH 7.4 pH 6.2

40 -

0

20

40

Time

60

80

(min)

F&& Repe@sion injury to cultured neonatal myoqtes - To simulate the acidosis and gen deprivation of &hernia, myo eswere made anoxic at pH 6.2, as described in Fig. 1. IT waschanged: 1) myocytes were reoxygenated, and pH wasincreased to with reoxygenation (reoxy, pH 7.4); 2) pH wasincreased to 7.4 without reoxygenatiau (anoxia, H 7.4); 3 myocytes were reoxygenated while maintainin pH.at 6.2 (reoxy, pH 6.2). Loss o Pcell viabrI*rty wasassessedby nuclear. staining withypi 2 mm iodide as a function of time after reoxygenatiorr/PH change. Cell krllmg at pH 6. wrth reoxygenation was significantly less that at pH 7.4 wrth or without reoxygenation (~~0.01 by Student’s t-test, n=5-9 per group).

This study also sought to evaluate the role of changes of pH in reperfusion injury to myocytes. Accordingly, we exposed cultured rnyocytes to anoxia and glycolytic inhibition at pH 6.2 to simulate ischemia. Subsequently, we reoxygenated at pH 7.4 to simulate reperfusion. After reoxygenation at pH 7.4, nuclei progressively became labelled with the propidium iodide. Nearly half of the myocytes lost viability within 80 minutes (Fig. 3). During simulated reperfusion, myocytes were reoxygenated and pH was returned to ‘7.4 simultaneously. To determine whether reoxygenation or an increase of pH was precipitating loss of cell viability, myocytes were reoxygenated at pH 6.2, or pH was returned to 7.4 without reoxygenation. After reoxygenation at pH 6.2, virtually no loss of cell viability occurred. In contrast, when we increased pH to 7.4 without reoxygenation, myocytes lost viability at virtually the same rate as after reoxygenation at pH 7.4. Thus, in this model of ischemia/reperfusion injury, the return of pH from acidotic to physiologic precipitated lethal injury. This paradoxical worsening of injury associated with the return of normal pH is a ‘pH paradox’, analogous to the calcium and oxygen paradoxes described previously for heart (18,19).

DISCUSSION

The hallmarks of myocardial ischemia are ATP depletion and acidosis. ATP depletion is caused by inhibition of mitochondrial oxidative phosphorylation, whereas acidosis is due tloth to lactic acid accumulation from glycolysis and to release of 2 hydrogen ions for every molecule of ATP hydrolyzed to AMP. Here, we show that this naturally occurring acidosis in myocardial ischemia protects against lethal anoxic injury to cardiac myocytes. Even rela801

Vol.

179, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

tively mild acidosis (pH 7.0) exerted substantial protection. tend previous findings in non-muscle cells (l-6).

RESEARCH COMMUNICATIONS

These findings confirm and ex-

In these experiments, 2-deoxyglucose was used to inhibit glycolysis, because carryover of glucose from the culture medium might support glycolytic ATP formation during anoxia. Glycolysis is generally not significant for maintenance of ATP levels during myocardial ischemia, because endogenous glycolytic substrates are limited and rapidly depleted (20). Thus, the inclusion of 2-deoxyglucose allowed us to better simulate conditions of myocardial ischemia. Upon reperfusion of ischemic myocardium, reoxygenation and a return of normal extracellular pH occur simultaneously. Reperfusion exacerbates tissue injury. Several mechanisms, particularly oxygen radical generation, have been proposed to cause reperfusion injury. The present findings indicate that pH changes associated with ischemia/reperfusion also contribute to the pathogenesis of reperfusion injury. To simulate ischemia, we exposed cultured myocytes to anoxia at acidotic pH. To simulate reperfusion, we reoxygenated at physiologic pH. Under these conditions, the return of pH, not reoxygenation, precipitated cell killing. This pH paradox may be a general phenomenon in ischemia/reperfusion injury. In recent studies, we also observed a pH paradox after hypoxic and ischemic injury to isolated hepatocytes, perfused livers, and hepatic sinusoidal endothelial cells (3,21,22). Propidium iodide was used to label the nuclei of nonviable myocytes. Labelling by propidium iodide is identical to that by trypan blue (3). It signifies breakdown of the plasma membrane permeability barrier, with consequent collapse of all ion and electrical gradients and loss of metabolic intermediates and cytosolic enzymes. However, irreversible cell injury, namely an injury from which a cell cannot recover, may occur before the outright expression of cell death. Previously, we related blebbing, hypercontracture, and propidium iodide labelling of myocytes to long-term recovery from metabolic inhibition with cyanide and 2-deoxyglucose (‘chemical hypoxia’). Propidium iodide-positive cells never recovered, whereas propidium iodide-negative cells recovered fully. Even in blebbed and hypercontracted cells, removal of inhibitors led to resumption of spontaneous contractions. After 24 hours, these myocytes were indistinguishable from untreated cells. Thus, propidium iodide labelling appears to signify the transition from reversible to irreversible injury in cultured cardiac myocytes. The mechanisms responsible for the pH paradox are unknown. Our working hypothesis is that the rise of pH after reperfusion triggers the activity of degradative enzymes with neutral or alkaline pH optima, such as non-lysosomal proteases and phospholipases ($23). Action of these enzymes on membrane proteins and lipids would then lead to sarcolemmal breakdown and cell death. Alternatively, return to physiologic pH may lead to cytosolic Ca’+ overload by concerted action of sarcolemmal Na+/Ca2+ and Na+/H+ exchangers. The rise of Ca2+ would then lead to mitochondrial Ca2* loading and uncoupling, activation of Ca2+-dependent degradative enzymes, and formation of rigor complexes, culminating in cell death. These hypotheses are currently under investigation. 802

Vol.

179, No. 2, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

REFERENCES 1. Penttila A. and Trum B.F. 1974 Science l&277-278. Bonvenfre > V and &eun$ .Y. 21985) Am. J. Physiol. 249, C149C159. 5: Gores, G.J., ‘Ncminen, A.- , and Lemasters, J.J. (1988) Am. J. Physiol. 255, C315C322. 4. Nieminen, A-L., Dawson, T.L., Gores, G.J., Kawanishi, T., Herman, B., and Lemasters, J.J. (1990 Biochem. Biophys. Res. Commun. 167,600-606. 5. Gores, G.J., rl ieminen, A.-L., Wray, B.E., Herman, B., and Lemasters, J.J. (1989) J. Clin. Invest. 83,386-3%.

6. Bing, O.H.L., Brooks, W.W., and Messer, J.V. 1973) Science 180,1297-1298. Altschuld, R.A., Hostetler, J.R., and Brierley, d .P. (1981) Circ. Res.49,307-316. i: Williamson, J.R., Schaffer, S.W., Ford, C., and Safer, B. (1976) Circulation 53 (Suppl. I), 13-114. 9. Korthuis, R.J., and Granger, D.N. (1986) In PhysioZogyof Oxygen Radicals (Taylor, A.E., Matalon, S., and Ward, P.A., Eds. , p. 217-249. Williams & Wilkins, Baltunore. 10. Jennings, R.B., Hawkins, H.K., Lowe, II!. ., Hill, M.L., Klotman, S., and Reimer, K.A. 1978) Am. J. Pathol. 92,187-214. b anote, C.E. (1983)J. Mol. Cell. Cardiol. 15,67-73. ::: Romson, J-L., Hook, B.G., Kunkel, S.L., Abrams, G.D., Schork, M.A., and Lucchesi, B.R. (1983) Circulation 67,1016-1023. Harary, I., and Farley, B. (1963) Exp. Cell Res. 29,451-465. 4: Ulrich, R., Elli et, K.A., and Rosnick, D.K. (1988)J. T&s. Cult. Meth. 11,217-221. 15: Bond, J.M., If erman, B., and Lemasters, J.J. (1991) Res. Comm. Chem. Pathol. Phannacol, 71,195-208. 16. Herman, B., Nieminen, A.-L., Gores, G.J., and Lemasters, J.J. (1988) FASEBJ. 2, 146151. 17. grnyters, J.J., DiGuiseppi, J., Nieminen, A.-L., and Herman, B. (1987) Nature 325, - .

18. Stern, M.D., Chien, A.M., Capogrossi, MC., Pelto, D.J., and Lakatta, E.G. (1985) Cir. Res. 56,899-903.

19. Grinwald, P.M., and Naylor, W.G. (1981) J. Mol. CeZL CdioL 13,867-880. Hearse, D.J., Garlick, P.B., and Humphre , S.M. (1977) Am. J. CardioZ.39,986-993. it Currin, R.T., Gores, G.J., Thurman, R. e ., and Lemasters, J.J. (1990) FASEB J. 3, A626. 22. Lernasters, J.J., Caldwell-Kenkel, J.C., Gao, W., Nieminen, A.-L., Herman, B., and Thurman, R.G. (1991) In Puthophysiologv of Reperfusion Injury, D.K. Das, Ed., CRC Press? in press. 23. Harrtson, D.C., Lemasters, J.J., and Herman, B. (1991) Biochem. Biophys. Rex Commm 174,654-659. 24. Bond, J.M., Herman, B., and Lemasters, J.J. (1990) FASEBJ. 4, A622.

803