Manipulation of NK cytotoxicity by the IAP family member Livin.

Natural killer (NK) cells are part of the innate immune system, capable of killing tumor and virally infected cells. NK cells induce apoptosis in the target cell by either granule- or receptor-mediated pathways. A set of inhibitory and activation ligands governs NK cell activation. As transformed cells often attempt to evade NK cell killing, up-regulation of a potential anti-apoptotic factor should provide a survival advantage. The inhibitor of apoptosis protein (IAP) family can inhibit apoptosis induced by a variety of stimuli. We have previously described a new IAP family member, termed Livin, which has two splice variants (alpha and beta) with differential anti-apoptotic activities. In this study, we explore the ability of Livin to inhibit NK cell-induced killing. We demonstrate that Livin beta moderately protects against NK cell killing whereas Livin alpha augments killing. We show that Livin beta inhibition in Jurkat cells is apparent upon concomitant activation of an inhibitory signal, suggesting that Livin augments an extrinsic inhibitory signal rather than functioning as an independent inhibitory mechanism. Finally, we demonstrate that detection of both Livin isoforms in melanoma cells correlates with a low killing rate. To date, this is the first evidence that directly demonstrates the ability of IAP to protect against NK cell-induced apoptosis.

Subcellular localization determines the delicate balance between the anti- and pro-apoptotic activity of Livin.

Livin is a member of the Inhibitor of Apoptosis Protein family which inhibits apoptosis induced by a variety of stimuli. We previously identified Livin and demonstrated that following apoptotic stimuli, Livin is cleaved by effector caspases to produce a truncated form with paradoxical pro-apoptotic activity. In the present study, we reveal that while full-length Livin shows diffuse cytoplasmic localization, truncated Livin (tLivin) is found in a peri-nuclear distribution with marked localization to the Golgi apparatus. Using mutation analysis, we identified two domains that are crucial for the pro-apoptotic activity of tLivin: the N-terminal region of tLivin which is exposed by cleavage, and the RING domain. We demonstrate that, of the N-terminal sequence, only the first N-terminal glycine residue dictates the peri-nuclear distribution of tLivin. However, while the perinuclear localization of tLivin is essential, it is not sufficient for tLivin to exert its pro-apoptotic function. Once tLivin is properly localized, an intact RING domain enables its pro-apoptotic function.