These results suggest that the EBNA1-derived HPV epitope may be a relevant target of EBV-specific CTL responses. To investigate the presentation of the HPV CTL epitope in EBV-positive cells, HLA-B35 or HLA-B53 positive LCLs and BL cells were used as targets of HPV-specific CTL buy Poziotinib cultures obtained from donors 5 and 7. We found, in the 5-hr 51Cr-release assay that unmanipulated HLA-B35- and HLA-B53-matched LCLs were lysed by HPV-specific CTL cultures whereas BL cells were not recognized, suggesting that the HPV epitope is poorly presented at the surface of BL cells (Fig. 2a,b). To exclude poor sensitivity

to lysis of BL lines, we evaluated the killing of BLs loaded with the synthetic HPV epitope by cytotoxic assay. We found that HPV-pulsed BL cells were recognized by HPV-specific CTLs, indicating that BL cells are sensitive to lysis and able to present the HPV T-cell epitope when exogenously added (Fig. 2b). The IFN-γ production assays have been mainly used in studies documenting the presentation of EBNA1-derived MHC-I-presented CTL epitopes because it is considered a more sensitive indicator

of target cell recognition.10–12 Therefore, we tested whether recognition of EBNA1-expressing BL cells could be revealed by monitoring IFN-γ release in ELISPOT assays. To this end, HPV-specific CTLs and matched LCLs and BL cells were seeded at an effector : target AZD3965 clinical trial ratio of 10 : 1, and the number of HPV-specific IFN-γ-producing cells was evaluated after 24 hr. As shown in Fig. 2(c), Florfenicol release of IFN-γ was specifically induced by HLA-B35-matched LCLs while HLA-B35-matched and HLA-B53-matched BL cells did not stimulate IFN-γ release, thereby confirming the poor presentation of this epitope in BL cell lines. As a whole, these results demonstrate that the EBNA1-derived HPV epitope is generated and presented in LCLs but not in BL cells. This suggests

that HPV generation does not exclusively depend on the presence of the GAr domain. Loss or down-regulation of HLA class I is one of the routes of immune escape in a variety of human tumours, including BL cell lines.25–28 Therefore, the surface expression of class I molecules in BL cells and LCLs was tested by indirect immunofluorescence. As shown in Fig. 3 and supplementary material, see Table S1, Jijoye cells expressed lower amounts of class I molecules whereas BJAB B95.8 cell lines showed similar levels of total HLA class I molecules, compared with LCLs. However, significant levels of lysis were achieved by the addition of HPV peptide to BL cells, thereby suggesting that sufficient levels of class I molecules were expressed at the cell surface (Fig. 2b).