Different Signaling Pathways Induce Apoptosis in Endothelial Cells and Cardiac Myocytes During Ischemia/Reperfusion Injury
This article has a correction. Please see:
Apoptosis contributes, with necrosis, to the cardiac cell loss after ischemia/reperfusion injury. The apoptotic cascade is initiated either by mitochondrial damage and activation of caspase-9 or by death receptor ligation and activation of caspase-8. In the present study, performed in the isolated rat heart exposed either to ischemia alone or ischemia followed by reperfusion, cleavage of caspase-9 was observed primarily in endothelial cells. Conversely, caspase-8 cleavage was only found in cardiomyocytes, where it progressively increased throughout reperfusion. Addition of a specific caspase-9 inhibitor to the perfusate before ischemia prevented endothelial apoptosis, whereas preischemic infusion of a specific caspase-8 inhibitor affected only myocyte apoptosis. Additionally, caspase-8–mediated BID processing was observed only during reperfusion. Production of tBID then sustains mitochondrial injury and perpetuates caspase-9 activation.
Apoptosis is an active form of cell suicide affecting both endothelial cells and cardiac myocytes during ischemia/reperfusion injury. Mitochondrial damage leads to activation of the initiator protease, caspase-9, which then propagates the cascade of downstream caspases. In contrast, ligation of death receptors, such as Fas, activates caspase-8, which also then processes downstream effector enzymes.1 Enzymatically active caspase-8 cleaves BID, and the truncated protein, tBID, relocates to the mitochondria where it induces caspase-9 processing.2
In this study, we have addressed three questions in the isolated rat heart: the level of enzymatic activity of the two initiator caspases during ischemia versus reperfusion; the selective contribution of these caspases to apoptosis of endothelial cells and myocytes; and the role played by BID in linking the two pathways.
Materials and Methods
Isolated Langendorff-perfused rat hearts3 were randomized to 11 groups, each of at least 6 animals. The care and use of the animals in this study were in accordance with the Guidance of the Animal (Scientific Procedures) Act 1986, UK. The animals were purchased from Charles River Italia S.p.a. (a division of Charles River Laboratories, Wilmington, Mass). The control group was buffer-perfused for 60 minutes; 4 ischemia/reperfusion control groups and as many treated groups were exposed to 35 minutes of global ischemia either alone or followed by 5, 60, and 120 minutes of reperfusion, respectively. Before induction of ischemia, treated hearts were perfused for 20 minutes either with a specific and irreversible inhibitor of caspase-8 (Z-IEDT.fmk; C8i) or caspase-9 (Z-LEHD.fmk; CASPASE-9i), at a dose of 0.07 μmol/L. Finally, 20 minutes before ischemia, two other treated groups subjected to 35 minutes of global ischemia and 120 minutes of reperfusion received either a pan-caspase inhibitor (Z-VAD.fmk; 0.1 μmol/L) or a specific and irreversible inhibitor of caspase-3 (Ac-DEVD.cmk; 0.07 μmol/L; C3i). All caspase inhibitors were from Calbiochem.
Caspase-8 and Caspase-9 Enzymatic Activity Measurement
Cardiac activation of caspase-8 and caspase-9 was evaluated in tissue extracts using commercial kits (BioVision; see online data supplement available at http://www.circresaha.org).
Myocardial sections were stained with antibodies recognizing the cleaved active form of caspase-8 and caspase-9 (BioVision; see online data supplement). Other sections were stained with TUNEL, labeled with anti-desmin (Research Diagnostics Inc) or anti–von Willebrand antibodies (Boehringer Mannheim Biochemica), counterstained with propidium iodide, and finally analyzed by confocal fluorescent microscopy.3
Antibodies anti–cytochrome c (Cytc) and anti-BID (Santa Cruz Biotechnology) were used to process frozen samples from each heart by Western blotting (see online data supplement).
Significance was evaluated using the ANOVA test. A value of P<0.05 was considered significant.
An expanded Materials and Methods section can be found in the online data supplement available at http://www.circresaha.org.
In agreement with our earlier results in cultured myocytes, 4 enzymatic assay shows that caspase-9 activation starts during ischemia and increases during reperfusion. In contrast, caspase-8 is functionally active only in hearts exposed to ischemia/reperfusion (Figure 1A).
By immunocytochemistry, in hearts receiving only ischemia, expression of cleaved caspase-9 is significantly increased in cardiomyocytes and even more in endothelial cells (4.7±0.65% and 6.8±0.54%, respectively; P<0.05 versus controls) (Figure 1B). In the same hearts exposed to ischemia alone, consistent with the level of caspase enzymatic activity, cleavage of caspase-8 is not observed in any cell type. The proportion of endothelial cells positive for activated caspase-9 rises dramatically in ischemic/reperfused hearts, peaking after 1 hour of reperfusion (35.9±3.1%; P<0.001 versus control). In contrast, cleavage of caspase-9 in cardiomyocytes remains stable after 5 and 60 minutes of reperfusion and halves at 120 minutes of reperfusion. Finally, in hearts exposed to ischemia/reperfusion, cleavage of caspase-8 is not observed in endothelial cells. However, the proportion of cardiomyocytes positive for cleaved caspase-8 progressively increases throughout reperfusion, reaching its maximum value after 120 minutes (8.9±1.6%; P<0.01 versus control).
The percentage of TUNEL-positive endothelial cells and cardiomyocytes in the hearts pretreated with caspase-8 and caspase-9 inhibitors (i) is reported in Figure 1C. Preischemic infusion of caspase-9i dramatically reduces endothelial apoptosis at all 3 time points of reperfusion. The decrease of cardiomyocyte apoptosis, in contrast, is less pronounced and becomes statistically significant only after 60 minutes of reperfusion. The administration of caspase-8i before ischemia consistently prevents TUNEL positivity in cardiomyocytes throughout reperfusion, without significantly affecting endothelial cell death.
Cleavage of BID assessed by Western blotting is seen in ischemic/reperfused hearts, but not in hearts exposed to ischemia alone (Figure 2A). Additionally, BID processing is greatly reduced by either Z-VAD or caspase-8i given before ischemia but is not affected by preischemic administration of either caspase-9i or caspase-3i (Figure 2B). To test whether tBID mediates communication between activated C8 and the mitochondrial death machinery in the intact heart, we evaluated the time kinetics of both BID processing and mitochondrial release of Cytc throughout ischemia/reperfusion in hearts pretreated with caspase-8i and caspase-9i (Figures 2C and 2D). After inhibition of caspase-8, processing of BID is not observed. However, leakage of Cytc, reflecting the extent of direct mitochondrial injury in the absence of tBID, is seen after 5 and 60 minutes of reperfusion but disappears by 120 minutes (Figure 2C). In contrast, when caspase-9 is inhibited and only caspase-8 is active, BID processing proceeds throughout reperfusion and Cytc relocates not only at 5 and 60 minutes of reperfusion but also after 120 minutes (Figure 2D; for ischemia/reperfusion control, see online data supplement). At this time, tBID production parallels the peak of cytosolic relocation of Cytc, which is not observed when BID processing is prevented by inhibition of caspase-8.
Similar results are found evaluating the proportion of cleaved caspase-8 and caspase-9 in hearts subjected to ischemia/reperfusion after their selective inhibition (Figure 3A). Pretreatment with caspase-8i consistently reduces not only cleavage of caspase-8 in cardiomyocytes but also that of caspase-9 in both endothelial cells and cardiomyocytes. In contrast, the caspase-9i diminishes activation of caspase-9 in endothelial cells and cardiomyocytes but does not affect the proportion of cells positive for cleaved caspase-8. Hearts pretreated with caspase-8i, although not caspase-9i, similarly show reduction of caspase-8 and caspase-9 enzymatic activities (Figures 3B and 3E). Additionally, whereas both endothelial cells and cardiomyocytes are positive for active caspase-9 after ischemia alone and after ischemia/reperfusion, cleaved caspase-8 staining is only seen in cardiomyocytes after ischemia/reperfusion (Figures 3C and 3D). Figure 3D confirms this identification of the cell types by staining the same sections with anti–von Willebrand and anti-desmin antibodies.
This study shows that caspase-9 is activated during ischemia and remains activated throughout reperfusion in the intact rat heart exposed to ischemia/reperfusion injury. In contrast, processing of caspase-8 is only triggered during reperfusion. Apoptosis of endothelial cells seems to be mediated completely by caspase-9 activation. Cardiomyocyte apoptosis relies on active caspase-9 during ischemia and early in reperfusion but shows an increasing dependence on caspase-8 activation later in reperfusion. Therefore, different initiator caspases are processed during different phases of ischemia/reperfusion injury and differentially trigger apoptosis in endothelial cells and cardiomyocytes.
The apparent resistance of endothelial cells to apoptosis after death receptor ligation and caspase-8 activation may reflect their high levels of expression of FLIP (FLICE inhibitory protein), an endogenous inhibitor of caspase-8 activation.5 Synthesis of death receptor ligands such as FasL and tumor necrosis factor-α has been demonstrated in isolated rat hearts early in reperfusion, and hearts from mice with a dysfunctional Fas receptor have fewer apoptotic cells after reperfusion.6 This suggests that cardiomyocyte apoptosis during reperfusion may be mediated by Fas ligation and is consistent with our observation of caspase-8 activation only during reperfusion. However, others have shown a critical role for free radicals and mitochondrial damage in the cardiomyocyte apoptosis induced by ventricular pacing,7 and the relative contribution of these two mechanisms to myocyte death remains to be fully clarified.
Our results on differential initiator caspase activation depend partly on the use of selective caspase inhibitors. Although the absolute specificity of these may be questioned, their use at submicromolar concentrations together with their in vivo effects (see Figures 3A and 3B) suggests that specific inhibition is being obtained under the experimental conditions used. Similarly, the previously documented specificity of the antibodies for the cleaved active forms of the respective caspases is supported by the general agreement between the immunostaining data and the enzymatic activity assays.
Our data further suggest that sustained activation of caspase-9 seen during reperfusion may depend on caspase-8–mediated cleavage of BID. Other caspases, such as caspase-3, as well as non–caspase proteases such as calpain,8 have also been shown to cleave BID, resulting in its mitochondrial translocation. Although the caspase-8 inhibitor reduces BID cleavage, the caspase-3 inhibitor is ineffective. Our data do not, however, exclude a contribution by calpain to BID cleavage.
These data, showing differences in initiator caspase activation over time, and between endothelial cells and cardiomyocytes, suggest that apoptosis after ischemia/reperfusion injury is not a homogenous process. Further understanding of the differential contribution of caspase-8 and caspase-9 may reveal new selective targets for minimizing cell loss after infarction.
This work was supported in part by the British Heart Foundation (BHF). Tiziano M Scarabelli is supported by a University College London Graduate School Research Studentship. Anastasis Stephanou is a BHF Intermediate Fellow.
Original received November 5, 2001; revision received February 21, 2002; accepted February 26, 2002.
- ↵Scarabelli TM, Stephanou A, Rayment N, Pasini E, Comini L, Curello S, Ferrari R, Knight RA, Latchman DS. Apoptosis of endothelial cells precedes myocyte cell apoptosis in ischemia/reperfusion injury. Circulation. 2001; 104: 253–256.
- ↵Suhara T, Mano T, Oliveira BE, Walsh K. Phosphatidylinositol 3-kinase/Akt signaling controls endothelial cell sensitivity to Fas-mediated apoptosis via regulation of FLICE-inhibitory protein (FLIP). Circ Res. 2001; 89: 13–19.
- ↵Jeremias I, Kupatt C, Martin-Villalba A, Habazettl H, Schenkel J, Boekstegers P, Debatin KM. Involvement of CD95/Apo1/Fas in cell death after myocardial ischemia. Circulation. 2000; 102: 915–920.
- ↵Cesselli D, Jakoniuk I, Barlucchi L, Beltrami AP, Hintze TH, Nadal-Ginard B, Kajstura J, Leri A, Anversa P. Oxidative stress-mediated cardiac cell death is a major determinant of ventricular dysfunction and failure in dog dilated cardiomyopathy. Circ Res. 2001; 89: 279–286.
- ↵Chen M, He H, Zhan S, Krajewski S, Ree JC, Gottlieb RA. BID is cleaved by calpain to an active fragment in vitro and during myocardial ischaemia/reperfusion. J Biol Chem. 2001; 276: 30724–30728.