The MHC and the recognition of self and altered self in experimental and rheumatoid arthritis

Sammanfattning: The Major histocompatibility complex (MHC) is a highly polymorphic and genedense region on chromosome 6 in humans and 20 in the rat. Genes in the MHC are the major risk factor for the development of autoimmunity. Rheumatoid arthritis (RA) is a common autoimmune disease with a strong association to a specific subset of HLADRB1 alleles, which encode a shared amino acid motif in the HLA-DR! chain. This shared epitope (SE) contributes to the formation of the P4 pocket in the peptide-binding groove of the HLA-DR complex. Recently, two weaker associations to the MHC class I gene HLA-B and to a second MHC class II gene, HLA-DPB1, have been identified. Whereas the dissection of the genetic association between autoimmunity and the MHC has progressed rapidly in the last decade, the functional role of the MHC genes in the development of autoimmunity has not yet been resolved to the same extent. The work presented in this thesis represents some of our efforts to contribute to this understanding. Paper I describes our attempts to induce arthritis with "-enolase and citrullinated "-enolase in HLA-DR4 transgenic mice. Based on the strong genetic association between antibodies to citrullinated "-enolase peptide 1 (CEP-1) and HLA-DRB1 SE alleles, we aimed to identify if anti-CEP-1 antibodies can be induced in mice transgenic for a human HLA-DRB1 SE allele, and if this would depend on a citrullinated HLADRB1' 0401 restricted T cell epitope. We could neither prove that native or citrullinated "-enolase induced arthritis nor did we observe a citrulline specific B or T cell response. In Paper II we investigated the relevance and the extent of joint-directed anticitrulline immunity in RA. We identified two citrullinated antibody epitopes on type II collagen (CII) present in 17% and 21% of RA patients. Anti-citrullinated CII reactivity partly overlapped with the reactivity towards CEP-1, however only antibodies directed to citrullinated CII bound to RA cartilage specimens. This suggests that antibodies directed to citrullinated CII may contribute to the inflammatory process in the joints. Paper III-V summarizes our work concerning the influence of allelic variation in the rat MHC on T cell selection, MHC expression and susceptibility to autoimmune arthritis. Paper III introduces our panel of intra-MHC congenic strains and demonstrates that an interaction between the RT1-A genes in the MHC class I and Tap2 in the MHC class II region regulates the negative selection of CD8 T cells. The RT1-A genes are part of a haplotype designated T cell selection QTL-1 (Tcs1), whereas Tap2 is in linkage disequilibrium with the MHC class II genes, in the locus designated T cell selection QTL-2 (Tcs2). In Paper IV we evaluated the impact of these two QTL on the regulation of Pristane-induced arthritis (PIA). Allelic variation in Tcs1 did not influence PIA, whereas Tcs2 regulated the onset and the severity of PIA. A comparison of the amino acid polymorphisms between the haplotypes as well as functional studies suggested a major contribution of the HLA-DQ homolog RT1-B to the development of PIA. It has earlier been demonstrated that pristane-primed CD4 T cells transfer arthritis in naive recipients. To investigate how the MHC-II haplotype affects T cell priming and the subsequent PIA development, we characterized the T cell compartment after pristane administration. The frequency of Th1 cells correlated with an early onset of PIA and reduced arthritis severity in one haplotype was associatedwith a high ratio of T regulatory to T effector cells. These results are presented in Paper V.