Basic Res Cardiol 93: Suppl 3, 8 – 12 (1998)© Steinkopff Verlag 1998

P. AnversaW. ChengY. LiuA. LeriG. RedaelliJ. Kajstura

Apoptosis and myocardial infarction

involved 2.8 million cells at 2 hoursafter coronary artery occlusion andnecrosis only 90,000 cells. Myocyteapoptosis continued to represent themajor form of cell death, affecting 6.6million cells at 4.5 hours, whereasmyocyte necrosis peaked at 1 day,including 1.1 million cells. Apoptoticmyocyte cell death was also present inthe surviving portion of the wall adja-cent to and remote from the infarctedmyocardium where it peaked at 1–2days. At this interval, 700/106 and110/106 myocyte nuclei were under-going apoptosis in the non-infarctedtissue bordering on and away from theischemic area, respectively. Myocytenecrosis was absent in the viablemyocardium after infarction. Sincemechanical forces produced by patho-logic loads may activate apoptosis,papillary muscles were exposed to high

levels of resting tension in vitro and themagnitude of cell death in these sam-ples was determined. Overstretchingresulted in a 21-fold increase in apop-totic myocyte cell death which wascoupled with the formation of reactiveoxygen species, side-to-side slippage ofmyocytes, and depressed tension gener-ation of the myocardium. In conclu-sion, apoptotic myocyte cell deathplays a major role in ventricular re-modeling after infarction, but whetherphysical forces, oxidant stress, archi-tectural rearrangement of myocytes,and impaired force development of themyocardium in vivo are causaly relatedrequires further investigation.

Key words Programmed myocyte celldeath – necrotic myocyte dell death –myocyte slippage – infarct size – wallrestructuring

Piero Anversa, M.D. (Y) · W. Cheng · Y. Liu A. Leri · G. Redaelli · J. KajsturaDepartment of MedicineVosburgh Pavilion, Room 302New York Medical CollegeValhalla, New York 10595

This work was supported by Grants HL-38132, HL-39902, PO-1-43023, and AG-15746 from the National Institutes ofHealth, and by Grant-in Aid 950321 from theAmerican Heart Association

Abstract Myocardial infarction wasproduced in rats and the contribution ofapoptotic and necrotic myocyte celldeath was measured quantitatively.Myocyte cell death by apoptosis

Apoptosis and necrotic myocyte dell death

Programmed cell death in the myocardium has been linked toischemia reperfusion injury (9) as well as to excessive physi-cal forces associated with increases in ventricular loading (4).Moreover, hypoxia activates the suicide program of cardiacmyocytes in vitro (20) and in vivo (19) and decreases in pHtrigger apoptosis (10). Occlusion of a major epicardial coro-nary artery leads within 1 minute to loss of contraction in thesupplied ischemic zone which is subjected to the mechanicalstimuli generated by systolic and diastolic cavitary pressures(12). The activation of anaerobic glycolysis and the formation

of lactate result in a reduction in intracellular pH (13, 17) andthe subsequent accumulation of CO2 and lactic acid in theinterstitium (7, 23). These changes are accompanied by severeacidification of the extracellular compartment as well. Thus,the combination of these factors has raised the possibility thatcell death by apoptosis may be implicated in the infarctedheart. Recent findings obtained in our laboratory have docu-mented that programmed myocyte cell death occurs acutelyafter myocardial infarction in both the region of the left ven-tricle supplied by the occluded coronary artery (12) and thesurviving portion of the wall adjacent to and remote from theinfarcted myocardium (3). Apoptotic myocyte cell death pre-

P. Anversa et al. 9Apoptosis in the Infarcted Heart

cedes cell necrosis and is the major determinant of infarct size(12).

In the infarcted portion of the wall, myocytes with DNAstrand breaks were detected first at 2 hours after coronaryartery occlusion; at this interval, approximately 2.7 millionmyocytes were apoptotic. A value of 6.6 million cells wasreached at 4.5 hours, indicating that from 2 to 4.5 hours therewas a 2.4-fold increase in the absolute number of apoptoticmyocytes in the left ventricular free wall (12). The magnitudeof apoptosis decreased progressively at the later time intervals,reaching a value of 43,000 at 7 days. Necrotic myocyte celldeath also appeared at 2 hours and increased continuouslyfrom 2 hours to 1 day following coronary artery ligation (12).At this time point, nearly 1.1 million myocytes were involved,whereas 90,000 cells were undergoing necrotic cell death at2hours. From 1 to 2 days, this phenomenon was markedly

attenuated, and only 3,000 myocytes were affected by necro-sis at 7 days. These observations document that apoptotic andnecrotic myocyte cell death are independent contributing vari-ables of infarct size, but apoptosis accounted for 86 % of thetotal loss of myocytes and necrosis for only 14 %. Althoughquantitative data remain to be obtained, myocyte apoptosisplays an important role in the infarcted heart in humans (11,16). In a manner comparable to the animal model, pro-grammed myocyte cell death extensively involves the infarctedmyocardium.

It should be emphasized that the apoptotic myocytes in theinfarcted myocardium subsequently undergo necrotic changesconsistent with the appearance of myosin labeling. This is theconsequence of the persistence of this interruption of coronaryblood flow to the ischemic area of the wall leading to secon-dary modifications associated with necrosis and healing. The

Fig. 1 Detection of DNA strandbreaks in myocyte and non-myocyte nuclei by the terminaldeoxynuclotidyl transferase assayof an infarcted portion of leftventricular free wall 3 hours aftercoronary artery occlusion (A).Myocytes were identified by a-sarcomeric actin antibody label-ing (B). Magnification: X350. For details see reference (12).

evolution of the process, therefore, involves the combinationof myocyte necrosis and apoptosis in the majority of cells.However, the primary event remains apoptosis which isresponsible for the death of 86 % of myocytes following coro-nary artery occlusion. This condition applies to large infarctsaffecting an average 60 % of myocytes of the left ventricularwall and overt cardiac failure (12). Whether smaller infarctscharacterized by a better preservation of cardiac pump func-tion may differ in the proportion of necrotic and apoptoticmyocytes is an important unanswered question.

The morphologic documentation of apoptosis, based on theterminal deoxynucleotidyl transferase assay (8), allows theapplication of morphometric methods to the assessment of theextent and distribution of this form of cell death in the ven-tricular wall. Similarly, myosin antibody labeling permits theidentification and quantification of necrotic myocytes in themyocardium (2). However, these approaches have to be com-plemented with the biochemical analysis of the DNA todemonstrate internucleosomal DNA fragmentation and/orDNA diffusion (6, 22). During apoptosis, activation of anendogenous endonuclease occurs and this event results inendonucleolysis (5). DNAdegradation is specific of the spacerregions, leaving the DNA associated with the nucleosomesintact (6, 22). The detection in the cells of DNA fragments ofsize equivalent to the mono- and oligonucleosomes is consid-ered the trademark of apoptosis. In contrast, with cell necro-sis, loss of plasmamembrane integrity and the release of lyso-somal proteases lead to the degradation of histones in thenucleosomes which results in the loss of DNA protection andits exposure to endonuclease. In essence, lysosomal proteasesdegrade nucleosomal histones so that the unprotected DNAcan be cleaved at random by endonuclease (6). This is coupledwith the generation of DNA fragments with large variations inlength as a diffuse pattern, like a smear, on agarose gel elec-trophoresis. On this basis, the characteristics of myocyte cell

death after infarction have been analyzed morphologically andbiochemically (12).

Apoptosis and myocyte slippage

In recent years, several studies have indicated that shortly aftermyocardial infarction there is a marked increase in diastolicwall stress that is associated with dilation of the ventricularchamber and mural thining (14, 15, 18, 21). This form of car-diac restructuring has been shown to be mediated by an archi-tectural rearrangement of myocytes, consisting of side-to-sideslippage of cells within the wall (14, 15). Such a reorganiza-tion of the myocyte compartment in the overloaded heartappears to account for the augmentation in cavitary volumeand the reduction in wall thickness acutely after myocardialinfarction (15). However, the phenomenon of myocyte slip-page has raised questions because it implies that myocytes arecapable of undergoing side by side translocation altering car-diac anatomy. If myocyte bundles oriented circumferentiallyto the transverse chamber diameter are involved, the immedi-ate consequence is mural thinning and an increase in the trans-verse and longitudinal axes of the heart. Conversely, if slip-page of cells occurs within myocytes oriented circumferen-tially to the longitudinal axis of the heart, this can be expectedto result in thinning of the wall and augmentation in the trans-verse chamber diameter, but with no change in the longitudi-nal axis of the ventricle. In both conditions, the ring of cellsthat moves radially toward the epicardial region must enlargeto adapt to the new circumference associated with a greaterchamber diameter.

Since stretching of sarcomeres in myocytes does not con-tribute significantly to increasing cell length and ventriculardimension, discrete myocyte cell death has been postulated to

10 Basic Research in Cardiology, Vol. 93, Suppl. 3 (1998)© Steinkopff Verlag 1998

Fig. 2 Effects of loading, 7–8mN/mm2 (low level of stretching:n = 6) and 50 mN/mm2 (highlevel of stretching: n = 6) for aperiod of 3 hours on papillarymuscle diameter (A) and numberof myocytes within the thicknessof the muscle (B). Results arepresented as mean ± SD. *Indicates a value that is statis-tically significantly different,p < 0.05. For details see reference(4).

P. Anversa et al. 11Apoptosis in the Infarcted Heart

occur; apoptosis of individual cells may rupture myocyte ringsallowing translocation of cells and the enclosing of largercavitary volumes (1). In the absence of multiple cell death, thesliding of myocyte bundles from the inner to the outer layer ofthe wall would not be possible, limiting mural thinning andchamber dilation. This hypothesis is supported by the obser-vation that programmed myocytes cell death takes place in thenon-infarcted tissue at 3 hours, reaching its maximum value at1–2 days (3). Importantly, side-to-side slippage of myocytesis present at this time and affects more the region bordering theinfarct than the remote healthy myocardium (15). Consis-tently, there are 700/106 and 100/106 myocytes undergoingapoptosis in the viable myocardium adjacent to and away fromthe ischemic area at 1–2 days after coronary artery occlusion.On this basis, the possibility may be advanced that diastolicstress, programmed myocyte cell death, and myocyte slippageare causally related in the dilation of the heart acutely aftermyocardial infarction (3).

In an attempt to document a cause and effect relationshipbetween mechanical forces, on the one hand, and myocyteapoptosis and myocardial restructuring, on the other, in vitroexperiments were performed in which normal papillary mus-cles were exposed to pathologic levels of resting tensions,mimicking diastolic Laplace overloading in vivo (4). Underthis setting, the force generating ability of the myocardiumdecreased along the entire length-tension curve, and this alter-

ation was accompanied by several anatomical changes. Over-stretching was characterized by a decrease in muscle diame-ter, apoptotic myocyte and non-myocyte cell death, and areduction in the number of muscle cells within the thicknessof the papillary muscle (4). Moreover, endogenous superoxideproduction was increased and intervention attenuating such aresponse markedly reduced programmed myocyte cell death.Thus, these observations are consistent with the notion thatscattered myocyte cell death may be required for the reorga-nization of muscle cell layers in the myocardium followingsevere increases in diastolic stress.

In summary, programmed myocyte cell death is the pre-vailing form of myocardial damage produced by occlusion ofa major epicardial coronary artery, whereas necrotic myocytecell death follows apoptosis and contributes minimally to theprogressive loss of myocytes after infarction. Programmedmyocyte cell death also affects the non-infarcted portions ofthe left ventricular wall and interventricular septum and thisphenomenon may be implicated in side-to-side slippage ofmyocytes, mural thinning and chamber dilation. Finally,abnormal levels of resting tension appear to be associated withthe activation of the suicide programm of myocytes and reduc-tion in muscle mechanical performance, suggesting thatmyocyte apoptosis may represent an important determinant ofthe unfavorable outcome of the cardiomyopathic heart ofischemic origin.

1. Anversa P, Olivetti G, Meggs LG, Sonnen-blick EH, Capasso JM (1993) Cardiacanatomy and ventricular loading after myo-cardial infarction. Circulation 87: VII22–VII27

2. Benjamin IJ, Jalil JE, Tan LB, Cho K,Weber KT, Clark WA(1989) Isoproterenol-induced myocardial fibrosis in relation tomyocyte necrosis. Circ Res 67: 657–670

3. Cheng W, Kajstura J, Nitahara JA, Li B,Reiss K, Liu Y, Clark WA, Krajewski S,Reed JC, Olivetti G, Anversa P (1996) Pro-grammed myocyte cell death affects theviable myocardium after infarction in rats.Exp Cell Res 226: 316–327

4. Cheng W, Li B, Kajstura J, Li P, Wolin MS,Sonnenblick EH, Hintze TH, Olivetti G,Anversa P (1995) Stretch-induced pro-grammed myocyte cell death. J Clin Invest96: 2247–2259

5. Darzynkiewicz Z, Bruno S, Del Bino G,Gorczyca W, Hotz MA, Lassota P,Traganos F (1992) Features of apoptoticcells measured by flow cytometry. Cyto-metry 13: 795–808

6. Gerschenson LE, Rotello RJ (1992) Apop-tosis: A different type of cell death. FASEBJ 6: 2450–2455

7. Gettes LS (1986) Effect of ischemia on car-diac electrophysiology. In: Fozzard HA,Haber E, Yennings RB, Katz AM, MorganHE (eds) The Heart and CardiovascularSystem. Raven Press: New York, pp 1317–1341

8. Gorczyca W, Bruno S, Darzynkiewicz RJ,Gong J, Darzynkiewicz Z (1992) DNAstrand breaks occurring during apoptosis:Their early in situ detection by the terminaldeoxynucleotidyl transferase and nicktranslation assays and prevention by serineprotease inhibitors. Int J Oncol 1: 639–648

9. Gottlieb RA, Burleson KO, Kloner RA,Bablor BM, Engler RL (1994) Reperfusioninjury induces apoptosis in rabbit car-diomyocytes. J Clin Invest 94: 1621–1628

10. Gottlieb RA, Gruol DL, Zhu JY, Engler RL(1996) Preconditioning in rabbit cardio-myocytes. Role of pH, vacuolar protonATPase, and apoptosis. J Clin Invest 97:2391–2398

11. Itoh G, Tamura J, Suzuki M, Susuki Y,Ikeda H, Koike M, Nomura M, Jie T, Ito K(1995) DNA fragmentation of humaninfarcted myocardial cells demonstrated bythe nick end labelling method and DNAagarose gel electrophoresis. Am J Pathol146: 1325–1331

12. Kajstura J, Cheng W, Reiss K, Clark WA,Sonnenblick EH, Krajewski S, Reed JC,Olivetti G, Anversa P (1996) Apoptotic andnecrotic myocyte cell deaths are indepen-dent contributing variables of infarct size inrats. Lab Invest 74: 86–107

13. Natel S, Elharrar V, Zipes DP, Bailey JC(1981) pH dependent electrophysiologicaleffects of quinidine and lidocaine on caninecardiac Purkinje fibers. Circ Res 48: 55–61

References

14. Olivetti G, Capasso JM, Meggs LG, Son-nenblick EH, Anversa P (1991) Cellularbasis of chronic ventricular remodelingafter myocardial infarction in rats. Circ Res68: 856–869

15. Olivetti G, Capasso JM, Sonnenblick EH,Anversa P (1990) Side-to-side slippage ofmyocytes participates in ventricular wallremodeling acutely after myocardial infarc-tion in rats. Circ Res 67: 23–34

16. Olivetti G, Quaini F, Sala R, Lagrasta C,Corradi D, Bonacina E, Gambert SR,Cigola E, Anversa P (1996) Acute myocar-dial infarction in humans is associated withactivation of programmed myocyte celldeath in the surviving portion of the heart.J Mol Cell Cardiol 28: 2005–2016

17. Peiper GM, Todd GL, Wu ST, Salhany JM,Clayton FC, Eliot RS (1980) Attenuation ofmyocardial acidosis by propranolol duringischaemic arrest and reperfusion: Evidencewith 31P nuclear magnetic resonance. Car-diovasc Res 14: 646–653

18. Pfeffer MA, Braunwald E (1990) Ventricu-lar remodeling after myocardial infarction.Circulation 81: 1161–1172

19. Sharov VG, Sabbah HN, Shimoyama H,Goussev AV, Lesch M, Goldstein S (1996)Evidence of cardiocyte apoptosis inmyocardium of dogs with chronic heartfailure. Am J Pathol 148: 141–149

20. Tanaka M, Ito H, Adachi S, Akimoto H,Nishikawa T, Kasajima T, Marumo F, HiroeM (1994) Hypoxia induces apoptosis withenhanced expression of Fas antigen mes-senger RNA in cultured neonatal rat car-diomyocytes. Circ Res 75: 426–433

21. Weisman HF, Bush DE, Mannisi JA, Weis-feldt ML, Healy B (1988) Cellular mecha-nisms of myocardial infarct expansion. Cir-culation 78: 186–201

22. Wyllie AH, Morris RG, Smith AL, DunlopD (1984) Chromatin cleavage in apoptosis:Association with condensed chromatinmorphology and dependence on macro-molecular synthesis. J Pathol 142: 67–77

23. Yan GX, Kleber AG (1992) Changes inextracellular and intracellular pH ischemicrabbit papillary muscle. Circ Res 71:460–470

12 Basic Research in Cardiology, Vol. 93, Suppl. 3 (1998)© Steinkopff Verlag 1998