Long-Term caspase inhibition ameliorates apoptosis, reduces myocardial troponin-I cleavage, protects left ventricular function, and attenuates remodeling in rats with myocardial infarction

Caspase 3 activation

Therapy with the caspase inhibitor significantly reduced caspase 3 activation, the main effector caspase, in the LV myocardium from rats with MI. In Western blots, caspase 3 cleavage was reduced by 55% in caspase inhibitor treated compared with vehicle treated rats (Fig. 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

(Top)Representative Western blot showing reduced caspase 3 and troponin-I cleavage in caspase inhibitor treated rats. Arrowsidentify the cleaved fragment. (Bottom)Mean density units normalized to actin (n = 8 each group) for caspase 3 and troponin-I cleaved fragments. Control = vehicle treated rats.

Cardiac apoptosis

Caspase 3 positivity predominantly involved the cardiomyocytes, as evidenced by double staining-activated Caspase 3 with either alpha sarcomeric actin or hematoxylin and eosin, where the caspase 3 staining was found within the outlines of apoptotic cardiomyocytes (Fig. 2A, panels A and B). Immunohistochemistry (Fig. 2A, panels C and D) also showed that caspase inhibition reduced myocyte apoptosis; there were 60% fewer caspase 3 positive cells in the remote LV myocardium of caspase inhibitor treated rats than in vehicle treated rats (Fig. 2B) (Mann-Whitney Utest, p < 0.001). TUNEL (Fig. 2A, panels E and F) confirmed myocyte apoptosis and also showed that the caspase inhibitor reduced cardiomyocyte apoptosis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

(A)Immunostaining for activated caspases in cardiomyocytes. Representative sections of left ventricular myocardium (200 to 400×). The left panelshows Apologix carboxyfluorescein activated caspase stain (white spots in panel A)and the same field stained with alpha sarcomeric actin (B)below. Most of the apoptotic cells were myocytes. One fibroblast with apoptosis is also shown (arrowhead in panels A and B)and has a different morphology. The next two panelsshow the effect of caspase inhibitors (caspase Rx) on number of apoptotic cells measured with Apologix staining (seen as large white bright spots in panels C and D) or TUNEL (greenish color in panels E and F). There were fewer (panels C and E)activated caspase-positive cells (panel C)or TUNEL-positive cells (panel E)in the myocardium of animals treated with caspase inhibitors than in the myocardium of animals with vehicle treatment (no caspase Rx, panels D and F). (B)Mean data for the number of apoptotic cardiomyocytes from rats with myocardial infarction treated or not treated with the caspase inhibitor (n = 8 each group).

Troponin-I cleavage

Caspase activation cleaves contractile proteins, and this may have implications for contractile dysfunction. Caspase inhibition reduced troponin I cleavage 51% compared with vehicle treatment (Fig. 1).

Hemodynamics

In vivo LV end-diastolic pressure (8 ± 0.9 vs. 13 ± 1.6 mm Hg, p < 0.05) was lower, whereas the LV developed pressure trended to be higher (117 ± 7 vs. 100 ± 4 mm Hg, p = 0.04) in caspase inhibitor treated rats than in vehicle treated rats.

LV systolic function

Consistent with the reduced myocardial troponin I cleavage seen in Figure 1, the caspase inhibitor treatment significantly attenuated the decline in systolic function seen after MI (Figs. 3 and 4). ⇓⇓Ejection fraction (39% and 46% greater at 1 and 4 weeks after MI, p < 0.05) and fractional shortening (55% and 45% greater at 1 and 4 weeks after MI, p < 0.05) were better preserved (Fig. 3) and LV end-systolic dimensions were 26% smaller in caspase inhibitor treated animals than in either vehicle treated or saline treated animals with MI (Fig. 4, lower panel). Vehicle treated animals were no different from animals with MI not given any infusion. Left ventricular function, however, remained worse in caspase-inhibited animals than in sham-operated animals. The mean arterial pressure was similar in caspase inhibitor and vehicle treated animals (93 ± 2.6 vs. 100 ± 1.7 mm Hg, p = ns), suggesting that the differences in systolic function were not due to a vasodilator effect.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Systolic function was better preserved both at one week and four weeks after myocardial infarction (MI) in caspase inhibitor (CI) treated (MI + CI, n = 12) rats than in vehicle (dimethyl sulfoxide) treated (MI + DMSO, n = 9) rats or MI rats without any pump (untreated MI, n = 15). Systolic function still remained worse than in sham-operated animals (n = 17). EF = ejection fraction; FS = fractional shortening.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Left ventricular (LV) remodeling and function were improved in rats with myocardial infarction (MI) given a caspase inhibitor (MI + CI, n = 12) for four weeks compared with vehicle (dimethyl sulfoxide) treated (MI + DMSO, n = 9) rats or MI rats without any pump (untreated MI, n = 15). Left ventricular remodeling still remained worse than in sham-operated animals (n = 17).

Myocardial interstitial collagen

The extent of myocardial fibrosis in the vehicle treated animals was similar to that described by other investigators in rats with moderately large MIs (21). However, there was a 46% reduction in the percentage of LV myocardium occupied by collagen in the caspase inhibitor treated compared with the vehicle treated rats (8% ± 0.7% vs. 15% ± 1.7%, p < 0.05).

LV remodeling

Caspase inhibition attenuated LV remodeling, as assessed by a number of measures. Heart weight was increased in both groups as expected after MI. However, heart weight and heart-weight-to-body-weight ratio were significantly lower in the caspase inhibitor than in the vehicle treated group (3.17 ± 0.14 vs. 4.7 ± 0.3, p = 0.028). The body weights were not significantly different (336 ± 3 vs. 342 ± 2 g, p = ns, in caspase inhibitor vs. vehicle treated animals). Echocardiographic LV end-diastolic dimensions were lower in the caspase inhibitor treated animals (Fig. 4, upper panel). These differences persisted even when normalized to body weight. Echocardiographic LV mass tended to be lower in the caspase inhibitor treated animals (0.57 ± 0.04 g vs. 0.64 ± 0.04 g), once again suggesting reduced remodeling, but this was not statistically significant.

Cell loss and progression to HF after MI

The syndrome of congestive HF is characterized by a relentless progression and high mortality. Although the exact mechanisms responsible for progression are not totally clear, it appears that myocardial structural remodeling is an important predictor of adverse prognosis in HF. Remodeling after MI is a progressive, often self-perpetuating process, and cell loss through apoptosis may be an important component in its genesis (5–7). Increased wall stress and neurohormonal/cytokine activation are known to adversely influence remodeling. There is evidence that these triggers can also mediate apoptosis. Apoptotic cell loss may contribute to ventricular dilation and increased wall stress, which may further perpetuate HF (5,22). Thus, interruption of apoptosis may have a therapeutic potential.

Apoptosis in MI and HF

Apoptosis has been shown to be prominent in both animal (2,3)and human (4)MI. The amount of apoptosis is variable and time dependent. There is evidence for an increase in many of the apoptosis intermediates, including a co-localization of caspase 3 and apoptosis in the ischemic heart (8,23). Apoptosis is also found in the chronically failing heart (6,7), albeit in a lesser amount, and a progressive loss of myocytes through apoptosis has been thought to be important for progression in this syndrome (5). Some investigators have also found chronic ongoing apoptosis in the hibernating heart, although this is controversial (24).

Caspases and HF

Caspases, which are activated after MI and in various stages of HF, are the key effector molecules for apoptosis. Cellular caspases exist as inactive precursors and need proteolytic cleavage for activation. They seem to be a crucial step in myocardial apoptosis after MI (8). Caspase activation co-localizes to apoptotic areas and precedes DNA degradation and the development of apoptotic morphology (23). Furthermore, caspase can also influence the contractile machinery of myocytes through cleavage of troponin (25,26). This can result in contractile dysfunction. Indeed, over-expression of caspase 3 induces contractile dysfunction in mice (27). However, caspase inhibition has not been studied in depth for modifying the course of the failing heart.

Caspase inhibition in hf and mi

Although caspase inhibitors have been tested in animal models of cardiac and non-cardiac ischemia, there are no published studies that have evaluated chronic caspase inhibition in the failing heart. A few studies report the short-term effects of caspase inhibition in MI (11,28). Caspase inhibition reduces apoptosis in experimental models of ischemia-reperfusion, but the degree of infarct size reduction is controversial (10,11). None of these studies has evaluated longer-term effects of reducing apoptosis. Our data is perhaps the first demonstration that long-term caspase inhibition reduces cardiomyocyte apoptosis in the non-infarct myocardium, attenuates systolic dysfunction, and reduces ventricular remodeling after MI.

Cellular targets for improving ventricular remodeling

It is interesting that caspase inhibition, apart from attenuating ventricular dilation, also reduced myocardial fibrosis. Caspase inhibition may have affected a variety of cells in the remodeling myocardium, including cardiomyocytes, fibroblasts, endothelial cells, and even the inflammatory cell infiltrate. Although anti-apoptotic effects on each of these cell populations are likely to have complex and possibly discordant effects on ventricular remodeling, the neteffect in our study was an attenuated LV remodeling. At the time we measured apoptosis (4 weeks after MI), most of the apoptotic cells appeared to be cardiomyocytes. This is consistent with data in other models of HF (29)and even in hearts with risk factors for myocardial ischemia but no failure (30).

There are no studies in literature to clarify how anti-apoptotic therapy might affect various components of ventricular remodeling. Our study also does not allow us to determine how caspase inhibition therapy affected other cell populations involved in ventricular remodeling and their interactions in the remodeling process. Although reduction in myocyte apoptosis could explain attenuated ventricular remodeling, it is not clear whether blocking myofibroblast apoptosis would be beneficial or harmful. Inhibiting myofibroblast apoptosis could make the fragile granulation tissue survive longer (31)and might make the scar tissue stronger. This might prevent excessive ventricular dilation, and indeed, there is very recent evidence (32)supporting this using ischemia-reperfusion—a model that, however, is significantly different from our model. Anti-apoptotic effects on fibroblasts early after MI have been suggested to retard ventricular remodeling (33). Secondly, fibroblast death affects the interconnection between myocytes, which would be expected to increase myocyte slippage, enhance mural thinning and chamber dilation, and increase wall stress (29,33). All of these features worsen remodeling and could be attenuated if anti-apoptotic therapy prolonged myofibroblast survival. This is supported by studies showing that inhibiting collagen degradation attenuates LV remodeling (34)and enhancing collagen degradation hastens the development of HF (35). Thus the attenuated remodeling seen in our model could be the result of beneficial effects on both myocyte and non-myocyte components.

The reduction in non-infarct myocardial fibrosis that we found at four weeks after MI in the caspase inhibitor group may not, at first glance, be consistent with reduced fibroblast or myofibroblast apoptosis. However, the initial stimulus to myofibroblast proliferation and increased fibrosis may arise from increased regional wall stress or stretch and neurohormonal activation (36). The latter would be attenuated if anti-apoptotic therapy reduced cardiomyocyte loss. Moreover, anti-apoptotic effects on myofibroblasts, in the critical initial days after MI, could have protected against ventricular dilation and wall stress and thus reduced the stimulus for fibrosis. Whereas our studies concentrated on the net effectsof anti-apoptotic therapy and were not designed to tease out these individual elements, one could speculate that reduced myocyte apoptosis (and possibly reduced myofibroblast apoptosis in the early post-MI period) might have reduced the stimuli for fibrous tissue proliferation.

Study limitations

There are several limitations of this study. Measurements including those for apoptosis were made at four weeks after MI. Even though we did find significant anti-apoptotic effects, earlier measurements might have shown a larger difference in apoptosis. Caspases also have beneficial housekeeping actions, and caspase inhibition may result in unwanted consequences. These were not evaluated in this study. Caspase inhibition may not block all pro-apoptotic stimuli, and combined strategies, attacking multiple pathways, may be more effective but remain subject to future studies. We chose the dose from previous studies in literature (13), but it is possible that a higher dose may have had a greater effect on remodeling. Finally, it is uncertain whether remodeling would worsen once caspase inhibition is terminated. This is the subject of a more long-term investigation.

Conclusions

We conclude that long-term caspase inhibition, started in the peri-MI period and continued for four weeks, significantly inhibited non-infarct myocardial caspase 3 activation and contractile protein cleavage in rats with a large anterior wall MI. This was accompanied by better preserved LV function and attenuated LV remodeling. Caspase inhibition might have a therapeutic role in improving the natural history of MI.