Vaccination with tumour-associated antigens (TAAs)-derived peptides designed to stimulate specific T cells has been a practicable approach evaluated in clinical trials [4–6]. Over the past decades, more than 60 TAAs, which can be recognized by CTLs and therefore can be used as tumour Maraviroc vaccine candidates, have been identified [7–9]. Cyclooxygenase-2 (COX-2) is over-expressed in various types

of human malignancies, including oesophageal carcinoma, breast cancer, gastric cancer, colon cancer and so on, but is hardly detected in most normal tissues at both mRNA and protein levels [10–12]. COX-2 is involved in the occurrence and development of many solid tumours via a variety of pathogenic mechanisms [13, 14]. These results indicated that COX-2 could be a useful target antigen to cancer immunotherapy [15]. Several widely expressed TAAs, including Survivin, Melan-A/MART-1, carcino-embryonic antigen (CEA) and gp100, represent self-proteins and as a result are poorly immunogenic because of immune tolerance. This may explain the failure of clinical trials in which self-proteins were used as immunogens [16]. One potential strategy to solve this problem is to design altered

peptide ligands (APLs). This approach has been applied with success for several HLA-A2 peptides derived from melanoma antigens and for gp100-derived epitopes [17, 18]. In 1993, Ruppert et al. [19] determined that buy PD-0332991 HLA-A2.1 binding motif could be defined as a leucine (L) or Methione (M) at position 2 (P2) and a leucine (L), valine (V), or isoleucine (I) at position 9 (P9). Tourdot et al. [20] showed that substitution of P1 by a tyrosine was a general strategy to enhance immunogenicity of HLA-A2-restricted epitopes. These results suggested that APL could

be used to exploit a potential capacity of the T cell repertoire to respond more effectively than that of native epitope. Our previous study see more has demonstrated that the cytotoxic T lymphocyte (CTL) epitope p321 (ILIGETIKI) from COX-2 could induce a moderate antitumour immune response in vitro [15], but could not induce antitumour immune response in vivo. In this study, we designed the analogues of p321 by altering p321 with a tyrosine at position 1 (1Y), and/or a leucine at position 9 (9L). Then, we performed peptide-MHC binding affinity and stability assay to determine their affinities to the HLA-A*0201 molecule. Subsequently, IFN-γ release ELISPOT assay and lactate dehydrogenase (LDH) release cytotoxic assay were employed to test its abilities to induce CTL responses in vitro. Finally, HLA-A2.1/Kb transgenic mice were used in this study to investigate the immune response elicited by naturally processing of COX-2-specific CTL epitope and its analogues in vivo. Peptide synthesis.