Apoptosis in Human Acute Myocardial Infarction
Background After reopening of the infarct-related coronary artery, cardiomyocytes continue to die during reperfusion. The mechanisms of cell death have been subject to debate. We studied whether an apoptotic type of cell death occurs in human acute myocardial infarction (AMI).
Methods and Results We studied myocardial samples of eight patients who died of AMI and had patent infarct-related arteries at autopsy. Six of the patients had received initially successful thrombolysis. Extensive formation of DNA strand breaks, the typical biochemical feature of apoptosis, was detected with the use of the in situ DNA end-labeling method. Apoptotic cardiomyocytes were observed particularly in the border zones of histologically infarcted myocardium, whereas very few apoptotic cells were present in the remote noninfarcted myocardium. Internucleosomal fragmentation was confirmed by agarose gel electrophoresis of DNA isolated from the representative myocardial areas.
Conclusions This study provides evidence that in addition to overt necrosis, a subset of myocytes undergo apoptosis during ischemia-reperfusion injury. Apoptosis may provide a new target for cardioprotection during evolving AMI in humans.
Opening of the infarct-related coronary artery is a valued therapeutic goal in acute myocardial infarction (AMI). However, cardiomyocytes continue to die during reperfusion. The mechanisms of cell death during this phase of AMI have been subject to debate.1 Apoptosis is a morphologically distinct, genetically controlled type of cell death.2 3 Extensive fragmentation of genomic DNA into nucleosome-sized fragments by endogenous endonucleases is the biochemical hallmark of apoptosis.4
Various physiological regulators and exposures to noxious stimuli can activate apoptosis.3 Experimental evidence suggests that cardiomyocytes are able to undergo apoptosis during hypoxia,5 myocardial infarction,6 heart failure,7 and ischemia-reperfusion.8 To explore the clinical relevance of these findings, we studied whether myocardial apoptosis occurred in human AMI after reopening of the coronary artery, the clinical counterpart of ischemia-reperfusion injury. Apoptosis was demonstrated by detecting DNA strand breaks with the in situ end-labeling (ISEL) technique, and internucleosomal fragmentation was confirmed by agarose gel electrophoresis of isolated DNA. Histological features of ISEL-positive areas were characterized by van Gieson staining of corresponding myocardial sections.
Tissue samples were obtained from autopsies of 14 patients. In 8 patients (Table⇓), AMI was diagnosed clinically by symptoms, ECG changes, and elevated serum levels of cardiac enzymes. Intravenous thrombolytic therapy was administered in 6 patients and resulted in initial alleviation of chest pain and ST-segment elevations as well as a rapid leakage of cardiac enzymes. Two patients (patients 7 and 8 in the Table⇓) presented with chest pain, deep ST-segment depressions, and elevated cardiac enzymes but did not receive thrombolysis. In all AMI patients, autopsy revealed multivessel coronary artery disease and severe but not total obstruction of the infarct-related artery, supporting the clinical judgment of initially successful thrombolysis (patients 1 through 6) or suggesting spontaneous subendocardial ischemia-reperfusion (patients 7 and 8, who also showed hemorrhagic infarction in histological sections). Six patients who died of noncardiac causes (Table⇓) served as control patients. The tissue collection procedures were approved by the National Board of Medicolegal Affairs in Finland.
Transmural myocardial tissue samples from infarcted, border-zone, and remote areas of infarcted hearts and from the left ventricles of noninfarcted hearts were fixed in 10% neutral buffered formalin for 24 hours, embedded in paraffin, and cut in 4-μm sections for ISEL and van Gieson staining. Comparable samples were frozen in liquid nitrogen for isolation of DNA.
Detection of Apoptosis
DNA strand breaks in tissue sections were 3′-end labeled with digoxigenin-conjugated dideoxy-UTP by terminal deoxynucleotidyltransferase and detected immunohistochemically with digoxigenin antibody conjugated to alkaline phosphatase, as described previously.8 9 Sections were pretreated with sodium citrate and proteinase K to improve the accessibility of DNA. The intensity of immunohistochemical staining was monitored continuously by light microscopy to confirm optimal sensitivity of the assay. The maximal incubation time was determined by the first appearance of positivity in samples from control patient hearts. Sections pretreated with DNAase I (1 U/mL) for 30 minutes at 37°C were intensely positive at this time. Reagents were purchased from Boehringer Mannheim, except dideoxy-ATP from Pharmacia and DNAase I from Sigma Chemical Co.
To confirm the apoptotic nature of DNA fragmentation, we performed autoradiographic analysis of electrophoretically fractionated and [32P]dideoxy-ATP (Amersham) end-labeled DNA according to previously described methodology.10 DNA was isolated from ISEL-positive and -negative areas in two patients and from areas of infarction, border-zone, and remote myocardium in another two patients.
The number of ISEL-positive myocytes and their percentage of total cardiocytes were counted with the use of a microscope with an eyepiece grid (magnification ×200). An average of 1800 fields were analyzed per patient. Histological features of ISEL-stained sections were determined by comparison with adjacent, serial sections stained with van Gieson. Samples from all AMI patients contained histological features of recent infarction: neutrophil infiltration, eosinophilia, cellular edema, nuclear changes, contraction bands, and coagulation necrosis.
The percentages of apoptotic cardiomyocytes in the infarcted, border-zone, and remote areas of the AMI hearts, expressed as mean±SD, were compared by use of Student's t test for paired data. Differences were considered significant at a value of P<.05.
In samples from each AMI patient, intense ISEL-staining of nuclei in some cardiomyocytes was observed, indicating extensive formation of DNA strand breaks. Electrophoretically separated DNA, isolated from the ISEL-positive areas, showed the characteristic ladder pattern in autoradiography (Fig 1a⇓), thus confirming the internucleosomal fragmentation typical of apoptosis. On the other hand, comparison of DNA fragmentation in the border zones versus central areas of infarction revealed more unspecific degradation in the central areas, visible as a smear in the autoradiogram (Fig 1b⇓). No detectable ladders were seen in the remote, noninfarcted areas. In all cases, clusters of apoptotic myocytes were detected particularly in the border zones of histologically recent infarction (Fig 2⇓, a through d). Many apoptotic cells were also present adjacent to scars of previous infarcts (Fig 2⇓, e and f) and in the endocardial regions adjacent to infarction. Some ISEL-positive inflammatory cells were detected, but their distribution showed no relation to the infarction and border-zone areas.
Fig 3⇓ shows that the percentage of apoptotic cells was significantly higher in the border zones compared with the central infarction areas (0.806±0.391% versus 0.039±0.027%; P=.0007). The amount of apoptosis in the remote noninfarcted segments (0.005±0.003%) did not differ from the control patient hearts (0.007±0.004%).
Experimental studies have shown that cardiomyocyte apoptosis is induced during hypoxia5 and ischemia-reperfusion.8 A specific association with reperfusion injury has even been suggested.8 The occurrence of cardiomyocyte apoptosis in the border zone of human myocardial infarction has also been described.11 The present study provides evidence that in addition to overt necrosis, a subset of myocytes undergoes apoptosis after AMI in patients with patent infarct-related arteries. These patients represent the clinical counterpart of experimental ischemia-reperfusion injury. Factors known to be associated with both ischemia-reperfusion injury and apoptosis include oxygen-derived free radicals,1 3 alterations of intracellular calcium homeostasis,1 3 increased mechanical stretch,6 and inflammatory reaction.3 12
The biochemical features of apoptosis have been described previously in histologically infarcted human myocardium.13 We found apoptotic cells to be significantly more numerous in the border zones of infarcted tissue, where percentages as high as 5.1% per microscopic field were occasionally observed. Because the time needed for a single cell to undergo apoptosis is probably very short, this process may allow considerable loss of cardiomyocytes in the vulnerable myocardial areas during the postinfarction recovery period.2 6 9 This possibility is further supported by the fact that we found the apoptotic type of cell death to be present days after initial reperfusion therapy as well as in the early phases of AMI.
Consistent with previous studies on myocardial ischemia-reperfusion,8 we did not detect clear apoptotic bodies of cardiomyocytic origin in routine histological sections. We suggest two possible explanations. First, apoptotic cells may be rapidly phagocytosed by the neighboring inflammatory cells.3 8 12 Second, the late stages of cardiomyocyte apoptosis and necrosis may share common morphological features, as was recently suggested by Kajstura and coworkers.6 However, in our sections, ISEL-positive cells showed no features of coagulative necrosis, which was otherwise detected in the central areas of infarction.
In conclusion, we have shown that cardiomyocyte apoptosis occurs in human AMI. Because apoptosis represents a potentially preventable form of cell death owing to its active nature, this finding may have important clinical implications as new cardioprotective treatment strategies are developed.
This work has been supported financially by the Academy of Finland and the Finnish Heart Association.
- Received September 30, 1996.
- Revision received November 18, 1996.
- Accepted November 20, 1996.
- Copyright © 1997 by American Heart Association
Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science. 1995;267:1456-1462.
Tanaka M, Hiroshi I, Adachi S, Akimoto H, Nishikawa T, Kasajima T, Marumo F, Hiroe M. Hypoxia induces apoptosis with enhanced expression of Fas antigen messenger RNA in cultured neonatal rat cardiomyocytes. Circ Res. 1994;75:426-433.
Gottlieb RA, Burleson KA, Kloner RA, Babior BM, Engler RL. Reperfusion injury induces apoptosis in rabbit cardiomyocytes. J Clin Invest. 1994;94:1621-1628.
Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol. 1992;119:493-501.
Quaini F, Cigola E, Sala R, Andreoli AM, Giordano G, Lagrasta C, Maestri R, Olivetti G. Apoptosis in the infarcted human heart. BAM. 1996;6:241-249.
Entman ML, Michael L, Rossen RD, Dreyer WJ, Anderson DC, Taylor AA, Smith CW. Inflammation in the course of early myocardial ischemia. FASEB J. 1991;5:2529-2537.