In this study we followed the fate
In this study, we followed the fate of Ag-specific CD8+ T 487 australia by directly visualizing them using MHC class I tetramers coupled with ovoalubumin (OVA)257–264 following EG.7 inoculation. We also examined the mechanism of apoptosis in exhausted Ag-specific CD8+ T cells during tumor immune response using FasL mutant mice or Bcl-2 transgenic mice. Our results demonstrate that the survival of exhausted CD8+ T cells is critically dependent on Fas/FaL signaling, but not on the overexpression of Bcl-2. These findings provide an insight into a new approach that could be developed to augment an effective immune response in cancer immunotherapy.
Materials and methods
Discussion In this study, we examined the fate of Ag-specific activated CD8+ T cells during the tumor immune response using EG.7 cells expressing OVA as a model tumor antigen. We found that Ag-specific activated CD8+ T cells were generated in the TILs and lymphoid organs following EG.7 inoculation, but thereafter decreased by apoptosis in association with the upregulation of Fas in the activated CD8+ T cells, during a prolonged tumor immune response. Ag-specific activated CD8+ T cells protected the majority of activated T cells from death in FasL-dysfunctional gld mice following EG.7 inoculation, whereas the enforced expression of Bcl-2 failed to rescue apoptosis in activated CD8+ T cells. These results suggest that Fas/FasL signaling plays a critical role in the apoptosis of exhausted CD8+ T cells during the tumor immune response. We initially followed the fate of Ag-specific CD8+ T cells by directly visualizing them using MHC class I tetramers coupled with the TRP-2-SVYDFFVWL-peptide, as a tumor antigen, following inoculation with the B16F10 melanoma cell line. However, although some TRP-2 specific CD8+ T cells were detected in the TILs on day 20 after B16F10 inoculation at the peak of the immune response, it was difficult to analyze the kinetics of tumor-Ag specific CD8+ T cells due to their limited numbers. Therefore, we selected EG.7 cell lines expressing OVA as a model tumor antigen. A tiny nodule was visualized in the lateral flank of mice from day 7 to day 10 after EG.7 inoculation, but we could not detect the OVA257–264 specific CD8+ T cells in their TILs using the OVA257–264 peptide tetramer system (data not shown), and we were unable to detect OVA257–264 specific CD8+ T cells in the LN cells or splenocytes at these times. Furthermore, CD44high CD8+ T cells were also not detected in the tiny tumor nodules in the TILs, indicating that activated CD8+ T cells including OVA257–264 specific CD8+ T cells were not infiltrating the tumor site during the generation of the tiny tumor nodules. In contrast, from day 14 to day 21 after tumor inoculation, we found that Ag-specific CD8+ T cells were generated in the TILs and lymphoid organs, the numbers of which were correlated with the level of tumor growth. Thereafter, the number of Ag-specific CD8+ T cells decreased in both the TILs and lymphoid organs during a prolonged tumor immune response. Therefore, we speculate that cell death was induced in these exhausted Ag-specific CD8+ T cells by constitutive TCR stimulation. We also detected the apoptosis of Ag-specific CD8+ T cells following the peak of the immune response with EG.7, indicating that apoptosis was induced in exhausted Ag-specific CD8+ T cells during the tumor immune response in this model. However, it should be noted that we only examined the apoptosis of CD8+ T cells with one immunodominant epitope. Therefore, the generality of this finding and its possible implications awaits further analysis with different epitopes. AICD is mainly triggered through cell surface proteins of the TNFR family, including Fas (CD95) (Ju et al., 1995; Sytwu et al., 1996; Nagata and Suda, 1995). Fas/FasL signaling has been shown to mediate apoptosis of activated T cells elicited by TCR restimulation in vitro (Alderson et al., 1995) and is known to be critical for the apoptosis of Ag-specific effector CD8+ T cells during persistent virus infections in vivo (Zhou et al., 2002). Therefore, it is possible that the apoptosis of exhausted Ag-specific CD8+ T cells is induced via Fas/FasL signaling during the tumor immune response under prolonged Ag stimulation. In this study, we found that Fas expression was gradually induced in activated CD8+ T cells following the peak of the immune response against EG.7 and was correlated with the induction of apoptosis. We also found that exhausted Fas-expressing CD8+ T cells could survive in the TILs and lymphoid organs of FasL-dysfunctional gld mice following EG.7 inoculation. These results suggest that the exhausted Fas-expressing CD8+ T cells were susceptible to FasL-mediated apoptosis following EG.7 inoculation. It has previously been reported that Fas-expressing T cells are susceptible to apoptosis by FasL in melanoma cells in metastatic lesion, indicating that FasL in tumor cells may contribute to the immune privilege of tumors (Hahne et al., 1996). However, we were unable to detect FasL expression in the EG.7 cell line during the tumor immune response using flow cytometry and immunohistochemistry (data not shown). Therefore, it is most likely that the susceptibility of Fas-expressing exhausted CD8+ T cells to apoptosis is caused by the action of FasL on the host cells. We believe that the induction of apoptosis is a mechanism for preventing a severe inflammatory response during constitutive TCR stimulation, rather than for the tumor to escape the immune system, but further experiments will be required to confirm this.