The regulation of caspase activity by inhibitor of apoptosis proteins

Members of the inhibitor of apoptosis protein (IAP) family have the unique ability to regulate apoptosis induced by numerous stimuli. The defining feature of the members of the IAP family is the presence of a least one baculoviral IAP repeat (BIR) domain, and elucidating the characteristics of these domains is the unifying principle of my thesis research. The X-linked IAP (XIAP) is the most well studied member of the family and has been established a potent inhibitor of caspases-3, 7, and 9. Due to the high degree of similarity, and initial enzymatic analysis, it has often been assumed that other IAPs have ability to inhibit caspases. Furthermore, the binding specificities of BIR domains have been generalized for fit a simple consensus sequence. These two assumptions have lead many to assume that most IAPs have overlapping and redundant functions. In the first aim of my thesis, I sought to determine the requirements for IAP-mediated caspase inhibition. In my dissection of these interactions it has become apparent that IAPs, other than XIAP, do not function as physiologic caspase inhibitors. I have found that the BIR domains of many IAPs are suitable for caspase binding; yet the flanking regions possess surfaces or conformations that disallow potent caspase inhibition. Therefore, the ability to potentially inhibit the catalytic activity of caspases is unique to XIAP. This marks an important revision in the field and opens several questions as to the mechanism apoptotic occlusion by these other IAPs. Secondly, I sought to investigate the other generalization of the IAP field, that being the equivalent specificities of the BIRs domain. Many BIR domains have a surface grove that endows them with the ability to interact with caspases and IAP-antagonists. Interestingly, the interaction motif of these pro-apoptotic proteins is conserved and thus it is assumed that this groove maintains the same specificity across many BIRs. In my second aim I developed a method to profile the binding specificity of this groove. I have found that many BIRs have binding signatures that closely resemble the classic motif. However, there are several other BIRs in which the signature is quite varied. These studies allow for prediction of cellular binding partners and may provide insight into the mechanism of apoptotic inhibition by some IAPs. Lastly, I turned my attention toward one of the least studied IAPs, the neuronal apoptosis inhibitory protein (NAIP). Although it was the first human IAP identified little is known about the true function of this IAP. Like most IAPs, NAIP has been suggested to be a caspase inhibitor. However, based on my dissections of IAP -mediated caspase inhibition in the first aim of this thesis it is clear that the BIR domains of NAIP lack conservation of the key caspase inhibitory surfaces. Interestingly, many have suggested that NAIP plays a role in innate immune signaling, however, the only data for this is based on the mouse studies suggesting it plays a significant role in restriction of the replication of intracellular bacteria. As predicted, I have found that NAIP is unable potently inhibit caspases. Intriguingly, I found that NAIP functions as an activator of the inflammatory caspases-4 and 5. The mechanism of activation seems to be quite unique and uncharacterized within the caspase family. These finding demonstrate for the first time that NAIP indeed has a role in regulating immune signaling