Apoptosis in heart failure: release of cytochrome c from mitochondria and activation of caspase-3 in human cardiomyopathy

Abstract

Apoptosis has been shown to contribute to loss of cardiomyocytes in cardiomyopathy, progressive decline in left ventricular function, and congestive heart failure. Because the molecular mechanisms involved in apoptosis of cardiocytes are not completely understood, we studied the biochemical and ultrastructural characteristics of upstream regulators of apoptosis in hearts explanted from patients undergoing transplantation. Sixteen explanted hearts from patients undergoing heart transplantation were studied by electron microscopy or immunoblotting to detect release of mitochondrial cytochrome c and activation of caspase-3. The hearts explanted from five victims of motor vehicle accidents or myocardial ventricular tissues from three donor hearts were used as controls. Evidence of apoptosis was observed only in endstage cardiomyopathy. There was significant accumulation of cytochrome c in the cytosol, over myofibrils, and near intercalated discs of cardiomyocytes in failing hearts. The release of mitochondrial cytochrome c was associated with activation of caspase-3 and cleavage of its substrate protein kinase C delta but not poly(ADP-ribose) polymerase. By contrast, there was no apparent accumulation of cytosolic cytochrome c or caspase-3 activation in the hearts used as controls. The present study provides in vivo evidence of cytochrome c-dependent activation of cysteine proteases in human cardiomyopathy. Activation of proteases supports the phenomenon of apoptosis in myopathic process. Because loss of myocytes contributes to myocardial dysfunction and is a predictor of adverse outcomes in the patients with congestive heart failure, the present demonstration of an activated apoptotic cascade in cardiomyopathy could provide the basis for novel interventional strategies.

Figures

Figure 1

Apoptosis in end-stage heart failure. Analysis of apoptosis was performed by end-labeling of DNA fragments in rat mammary tissue (A; used as a positive control) and myocardial specimens from normal (B) and cardiomyopathic (C and D) hearts. The end-labeling of the DNA fragments was obtained by addition of biotinylated dUTP in the presence of terminal deoxynucleotidyltransferase (TdT) (Trevigen, Gaithersburg, MD). The nucleosomal DNA fragments were visualized with the help of avidin–biotin complex followed by chromogen substrate, which stained the apoptotic nuclei blue (A, C, and D, arrowheads, which appear black here). The apoptotic nuclei (arrowheads) can be seen in acinar cells in the mammary lobules (A). In contrast, apoptotic cells are not observed in the normal myocardial tissue (B) obtained from a victim of motor vehicle accident. Eosin counterstaining demonstrates myocardial silhouette, and the unstained nuclear regions appear clear (open arrows). In a myocardial specimen from a cardiomyopathic heart (C), two myocytes with apoptosis are shown by blue-stained nuclei (arrowheads, which appear black here). The occurrence of apoptosis in myocytes (arrowheads) is further confirmed by double staining for α-muscle actin with antibody HHF-35 (brown product, which appears gray here) (D).

Figure 2

Ultrastructural localization of cytochrome c. High magnifications of normal myocardial specimen from a donor heart (Top) and ischemic cardiomyopathic hearts (Middle and Bottom) demonstrate striking difference in cytochrome c immunoreactivity. Localization of cytochrome c is represented by (black) gold particles. The cytochrome c is predominantly localized in mitochondria (M) in normal hearts (Top). On the other hand, cytochrome c is substantially reduced in mitochondria (Middle, M) in cardiomyopathic heart and is concentrated either over Z-bands (Middle, open arrows) or near intercalated disc (Bottom, open arrows). Small arrows indicate the presence of normal sarcolemma (Middle).

Figure 3

Accumulation of cytochrome c in the cytosol. (A) Proteins from the cytosolic S-100 fractions were separated by 12.5% SDS/PAGE and were analyzed by immunoblotting (IB) with anti-cytochrome c antibody. (Top) Normal (Lanes 1 and 2), IDCM (Lanes 3 and 4), and ISCM (Lanes 5 and 6) hearts. Total lysates also were analyzed by immunoblotting with anti-cytochrome c (Middle) and anti-actin (Bottom) antibodies. (B) Neonatal rat cardiomyocytes (NRC) cultured in normoxic (Lane 1) or hypoxic conditions for 6 (Lane 2) and 12 (Lane 3) hours (Left), U-937 control cells (Lane 4), and cells treated with 20 Gy ionizing radiation and harvested at 6 hours (Lane 5). S-100 cytosolic extracts were prepared and analyzed by immunoblotting with cytochrome c antibody. Cytosolic accumulation of cytochrome c was only observed in cardiomyopathic myocardial specimens, irradiated U-937 cells, and hypoxic cardiomyocytes. Normal myocardial tissues and control cells show no evidence of cytosolic cytochrome c.

Figure 4

Activation of caspase-3 in cardiomyopathic hearts. (A) Proteins from total cell lysates of normal, IDCM, and ISCM were separated by SDS/PAGE, were transferred to nitrocellulose, and were analyzed by immunoblotting (IB) with anti-CPP32, anti-PARP, and anti-PKCδ antibodies. (B) Similarly, both control and irradiated U-937 cell lysates were analyzed by immunoblotting with anti-CPP32, anti-PARP, and anti-PKCδ antibodies. Cleavage of CPP32 and PKCδ is seen in cardiomyopathic hearts similar to U937 cells. PARP cleavage is not observed in cardiomyopathic hearts, unlike U-937 cells. Control myocardial lysates and untreated U-937 cells show only uncleaved intact precursors (FL or full-length).