We have recently found that a large fraction of Mart-1 epitope specific primary CTLs, generated in an in vitro peptide-loaded DC-based CTL generation protocol, undergoes AICD after the very first secondary encounter of the cognate antigen. The AICD in these CTLs is not triggered by the usual external death receptor (FAS, TNFR, etc.)- mediated signaling and is not caspasedependent. Our studies indicated that these CTLs could be rescued from AICD by the c-jun Nterminal kinase (JNK) inhibitor, SP600125. In the process of rescuing, SP600125 interfered with their capacity to synthesize IFNg but did not block their cytolytic function. We have recently foundthat the AICD in these CTLs is mediated by the activation of a mitochondria! apoptotic machinery characterized by the release of the mitochondria-based apoptosis inducing factor (AIF) without any cytochrome c release. AIF then translocates onto the nuclei and causes large scale (around 50 kbp) single stranded DMAbreaks. The JNK inhibitor SP600125 blocks the AIF release. We have detected that a short phosphorylatedfragment of Bim, inhibitable by the JNK inhibitor SP600125, is generated in these CTLs during AICD. We have also found that JNK to be present on mitochondria and to interact with several mitochondria-based Bcl-2 family proteins and with the mitichondrial porin voltage dependent anion channel (VDAC). These observations have prompted us to hypothesize that AICD in self-but-melanoma epftope-specific CTLs mostly results from the release of the mitochondria-based apoptotic effector proteins such asAIF triggeredby the activation of the mitochondhalJNK-BH3-only prpapoptotic protein- VDAC axis and that SP600125 rescues some of these CTLs from AICD by blocking JNK activation. Thus, a long- lived CTLresponse to tumor epitopes might be orchestratedby interfering with this JNK-driven apoptotic pathway". Hence, the specific aims are :1) To define the rule(s) underlying AICD and rescue from AICD in "self but melanoma epitope specific CTLs and influenza MP (a non-self and dangerous antigen)-specific CTLs; 2) To study the mechanism underlying the mitochondria-based apoptotic machinery involved in AICD in the melanoma epitope specific primary CTLs with emphasis on elucidating a potential role of JNK-Bim/Bax-VDAC interaction as the trigger for AIF release; 3) To extend our findings of AICD and rescue from AICD in the melanoma epitope specific CTLs in a xenograft model in RagW- mice, in vivo. The work will be carried out with the respective antigen specific CTLs generated in an in vitro CTL generation protocol followed by assessing their sensitivity to AICD under different experimental conditions in vitro and in vivo in a xenograft model in Rag1-/- mice. The mechanism of AICD and rescue from AICD will be explored through pharmacologic, genetic (interfering RNA-based silencing of JNK, Bim), and biochemical approaches. These studies will provide much needed understanding of howCTLs generated against a relevant tumor associated antigen can be kept alive longer and will therefore facilitate the design of more effective tumor immunotherapy.