ICE-ing the heart.

In this issue of Circulation Research, Syed et al 1 report that overexpression of caspase-1 (ICE) contributes to myocardial ischemia/reperfusion injury. Transgenic mice were generated using the alpha-myosin heavy chair promoter to drive caspase-1 expression in the heart. The mice expressed caspase-1 30-fold more than the nontransgenics. Despite this massive overexpression, no spontaneous caspase activation or apoptosis was noted, and there was no apparent cardiac phenotype. However, endotoxin exposure and ischemia/reperfusion were associated with increased caspase-1 and -3 activation in the transgenic mice, suggesting that caspase-1 contributes to myocardial pathology in these settings. Cell culture studies were used to bolster the notion that caspase-1 activated caspase-3 directly rather than through upstream caspases. Missing from this study was any analysis of cytokine processing, although the primary role of caspase-1 is to process interleukin-1-beta (IL-1 ) and IL-18, and potentially IL-1 , IL-6, and tumor necrosis factor alpha(TNF). Enhanced caspase-1 activity and IL-18 have been reported after myocardial infarction, suggesting a role for caspase-1 in postischemic inflammation and injury.2 Frantz et al who showed that targeted deletion of caspase-1 was associated with reduced postoperative mortality and less left ventricular dilatation after myocardial infarction.3 The authors also noted reduced IL-18 levels and decreased matrix metalloproteinase-3 (MMP-3) in the caspase-1-null animals after myocardial infarction, suggesting that cytokine signaling contributed to the injury process. Caspase-1 can be processed to the active form by caspase11,4 which is activated in response to ischemia or endotoxin exposure. Caspase-11 is also capable of activating caspase-3 directly,5 although its significance in the heart is unknown. Caspase-1 is activated by oligomerization with adaptor proteins via the Caspase Recruitment Domain (CARD) domain6 (Caspase-1 can interact with Ipaf, or with one of the 14 known members of the NALP family, which contain 3 domains: a leucine-rich repeat, a NACHT domain, and a pyrin domain, which is bridged to the CARD domain of caspase-1 via a second adaptor, ASC, which contains both a pyrin and a CARD domain. Caspase-5 is also recruited to this multiprotein complex and is believed to participate in caspase-1 activation and IL-1 processing. Formation of this caspase-1 activation complex, known as the inflammasome, is triggered by lipopolysaccharide binding to its receptor TLR-4. The inflammasome can also be activated by signaling through P2X7 receptors and hypotonic stress. Less is known about the formation of the inflammasome in response to hypoxia or oxidative stress. Whether the composition of the inflammasome complex can vary depending on the stimulus remains to be determined. The net effect is to trigger processing of inflammatory cytokines, particularly IL-1 ; some members of the NALP family also modulate NFB activation. Caspase-1 can also form a complex with Nod1 and RIP2/RICK/CARDIAK.7–9 One can expect that cytokinedependent inflammation and apoptosis of susceptible cells would ensue. This may be particularly relevant to heart failure, in which inflammatory processes and low levels of ongoing cell death prevail. Whether the caspase-1-dependent apoptosis after ischemia is due to extracellular signaling by the processed cytokines or to intracellular events (caspase-1 activation of caspase-3) remains an open question. The cellular studies conducted by Syed et al1 show that in procaspase-1 transfected cells subjected to hypoxia, there was caspase-3 cleavage in the absence of processing of caspase-8 or -9, which they argue reflects direct activation of caspase-3 by caspase-1. However, it is now widely recognized that caspase-8 and -9 can be active in the absence of proteolytic processing, and that ‘downstream’ caspases, however they become activated, can proteolytically process the initiator caspases. Thus it is difficult to infer activation of caspase-8 or -9 from the demonstration of proteolytic fragments.10 Caspase-1 has been shown to process Bid to its active form, thereby activating the mitochondrial pathway of apoptosis.11 Subsequent work showed that this pathway could be initiated in neuronal cells subjected to ischemia.12 Bid is well known to act as a protease sensor, and can respond to initiator caspases, Granzyme B, cathepsins, and calpains.13,14 Work by Kitsis et al showed marked infarct size reductions in Bid-null mice, pointing to the importance of this mediator regardless of the upstream protease.15 An additional consideration is whether caspase-1 enzymatic activity is required for initiation of apoptosis in this setting. It was shown by Lamkanfi et al16 that caspase-1 could activate NFB in the absence of protease activity. Given the controversial roles of NFB signaling in the ischemic and reperfused heart,17,18 future studies using a catalytically inactive mutant of caspase-1 will provide important mechanistic insights. What does this study tell us about the role of caspase-1 in the myocardium? Overexpression of caspase-1 allowed the investigators to convincingly demonstrate that endotoxin exposure and ischemia can lead to processing of caspase-3. However, such an overexpression study cannot provide inThe opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Scripps Research Institute, La Jolla, Calif. Correspondence to Dr Roberta A. Gottlieb, Department of Molecular and Experimental Medicine, The Scripps Research Institute MEM220, 10550 North Torrey Pines Road, La Jolla, CA. Email [email protected] (Circ Res. 2005;96:1036-1038.) © 2005 American Heart Association, Inc.