33 to 36 (sequences 33–47 to 36–50), peptides no. 48 to 58 (sequences 48–62 to 58–72), peptides no. 117 to 123 (sequences 117–131 to 123–137), peptides no. 151 to 166 Midostaurin cost (sequences 151–165 to 166–180), and peptides no. 261 to 263 (sequences 288–302 to 292–306). Conversely, some epitopes were specific for a particular HLA subtype, such as the determinant encompassing peptides no. 1 to 9 (sequences 1–15 to 9–23), which was specific for DR*0101 (Fig. 1). We additionally used the TEPITOPE program to predict the nonamer core sequences
binding to HLA DR*0101 and *0401 as well as to DR*0404 molecules. TEPITOPE identified 31 core epitopes; of these, 19 are listed in the column 2 of Table 1 because they were also binding in our assay. The 12 additional core sequences, predicted as poor binders by TEPITOPE, are listed in Supporting Information. The detailed analysis of hnRNP-A2 peptides binding to RA-associated molecules described above showed that these epitopes were too numerous to be tested with human https://www.selleckchem.com/products/epz-6438.html cells. Thus, T-cell epitope candidates were selected stepwise as follows: (i) When multiple overlapping nonameric peptide frames were found and/or predicted to interact with
RA-associated HLA molecules, the peptide length was determined to include all possible peptide frames within the sequence. Using these parameters, we selected and synthesized a set of 16 peptide sequences of 17–23 amino-acid length (see Table 1). These peptides were further tested in binding assays
to determine their relative Edoxaban affinity to HLA molecules compared to influenza hemagglutinin control peptides. The results obtained showed that hnRNP-A2 peptides are relatively poor binders compared to the control peptides (Supporting Information Fig. 1). The best binders were peptides 289–306 for DR*0401, 177–193 and 152–170 for DR*0404, and 3–19 for DR*0101, respectively (Table 1 and Supporting Information Fig. 1). There were some discrepancies between the binding assays and the binding prediction given by the TEPITOPE program: for example, peptide 120–133 was predicted to bind well to DR*0404 but appeared to be an extremely weak binder, at the limit of sensitivity of our assay (Table 1). If one postulates that a determinant intrinsically linked to RA pathogenesis should be presented by most RA-associated HLA molecules, i.e. by DR*0101, 0401 and 0404, peptides binding to these three molecules would represent the best candidates. The four peptides 10–26, 50–70, 120–133, and 152–170 were found to fulfill this criterium, although 10–26 bound weakly to DR*0101, 120–133 weakly to DR*0404, and 152–170 weakly to DR*0401. Therefore, these epitopes, followed by peptides 3–19, 177–193, and 289–306, were considered best candidates to detect hnRNP-A2 specific T cells in patients with RA. To verify that peptides binding to DR*0401 in vitro are also immunogenic, DR*0401-Tg mice were immunized subcutaneously with individual hnRNP-A2 peptides (Fig. 2).