T cell response to myelin basic protein in the context of the multiple sclerosis-associated HLA-DR15 haplotype: peptide binding, immunodominance and effector functions of T cells.

In this study, we evaluated the role of the two functional HLA-DR heterodimers, DR2a (DR alpha paired with the beta chain encoded by DRB5*0101) and DR2b (DR alpha paired with the beta chain encoded by DRB1*1501), that are coexpressed in the multiple sclerosis (MS)-associated haplotype HLA-DR15 Dw2, in presenting myelin basic protein (MBP) peptides to MBP-specific T cell lines (TCL). Our results show that both HLA-DR molecules serve as restriction elements for HLA-DR15-restricted TCL. Slightly higher numbers of TCL use DR2a as restriction element, and the epitopes contained in the immunodominant C-terminal region (131-159) are uniquely restricted by DR2a. The immunodominant middle epitope (81-99) is recognized in the context of both DR2a and DR2b, but this specificity strongly dominates the DR2b-restricted T cell response. Overall, immunodominance in the MBP-specific T cell response correlated well with peptide binding to DR2a or DR2b, demonstrating that the affinity of MHC-peptide interactions is important for shaping the T cell response to this autoantigen. Furthermore, we show that binding of the middle MBP peptide to HLA-DR15 molecules prevents cleavage by cathepsin D, a protease abundantly found in endosomal processing compartments, and thus contributes to its immunodominance. Surprisingly, the restriction element employed by MBP-specific T cell clones influenced the effector function (i.e., cytotoxic activity) of T cells irrespective of their peptide fine specificity.

Differences in specificity of "DPw2-specific" cytotoxic T cell clones revealed with HLA-mutant lines: evidence that non-DP HLA genes influence recognition by some clones.

Seven allospecific cytotoxic T lymphocyte (CTL) clones derived from DPw2-specific bulk populations were characterized by three approaches to obtain a more detailed understanding of the T cell recognition of the HLA-DPw2 molecule. All seven of the clones were DPw2 specific and indistinguishable in specificity by three approaches: (a) patterns of lysis of panels of targets from normal donors; (b) inhibition of lysis by anti-class II monoclonal antibodies (mAb); (c) lysis of mutant lymphoblastoid B cell lines (LCL) with isolated loss of DP2 alpha or DP2 beta gene expression (as a result of selection for resistance to DPw2-specific CTL). However, only four of the seven CTL clones (which we designate "orthodox") lysed all mutant DPw2+ LCL tested; the other three ("heterodox") CTL clones showed reduced or no lysis of particular LCL which expressed DPw2 but had been mutagenized and selected for loss of DR expression. Cold target blocking experiments with the mutant LCL confirmed differences in: (a) specificity among CTL clones and (b) CTL-defined phenotype among serologically indistinguishable DR-DPw2+ mutant LCL. Differences were not explained by different levels of DP expression by the mutant LCL. Given the nature of the mutagens and mutations, it is highly improbable that point mutations in DP account for differences in recognition. These data suggest that non-DP HLA genes influence recognition by some DP-specific clones, potentially due to corecognition of HLA-DR alpha or another non-DP HLA product in the context of a DPw2 alpha/beta dimer.