Evolution of class II genes: role of selection in both the maintenance of polymorphism and the retention of non-expressed alleles

Characterization of polymorphism within the H2-M MHC class II loci

The products of the class II-like H2-M genes of the major histocompatibility complex are required for class II antigen processing. We sequenced H2-Ma and Mb from several mouse strains to determine whether these genes are polymorphic like the classical H2-A and E genes, or are oligomorphic, like H2-O. Both Mb loci appear to be transcribed and are distinct from each other. Mb1 and Mb2 differ by about 11% at the nucleotide level and are most dissimilar in their second exons (corresponding to the beta 1 domain). Relative to the published Mb1d haplotype sequence, the products of the b, g7, f, and k2 alleles of Mb1 from Mus musculus domesticus and the separate mouse species Mus spretus differ by only one to four amino acids. The majority of the changes occurred in the second exon of Mb1, in contrast to HLA-DMB, the human orthologue. Little polymorphism was seen for Mb2, and Ma was invariant in all strains tested. The similarity of the g7 allele to those from other haplotypes makes it unlikely that the M class II genes play a role in the autoimmune diabetes of NOD strain mice. The M genes are regulated in a manner similar to classical class II genes, in that they are upregulated by IFN-gamma in macrophages, and to a lesser extent by IL4 in B cells. When modeled on the crystal structure of the HLA-DR1 class II molecule, nearly all of the differences between M beta 1 and M beta 2 affect residues facing away from the putative peptide binding groove.

The minimal polymorphism of class II E alpha chains is not due to the functional neutrality of mutations

Given the extensive allelic amino acid sequence polymorphism present in the first domain of A alpha, A beta, and E beta chains and its profound effects on class II function, the minimal polymorphism in the mouse E alpha chain (and in its human homologue DR alpha) is paradox. Two possible explanations for the lack of polymorphism in E alpha are: (1) the E alpha chain plays such a uniquely critical structural/functional role in antigen presentation, T-cell activation, repertoire selection, and/or pairing with E beta or other proteins for expression that it cannot vary, and mutations are selected against; (2) the E alpha chain plays a less significant role than the outer domains of other major histocompatibility complex (MHC) proteins in determining the interactions with processed peptides or with T-cell receptor (TCR), so there is no selective pressure to maintain new mutations. To explore this question we compared the ability of transfectants expressing wild type (wt) E alpha E beta d and mutant E alpha wt E beta d proteins to present peptides and bacterial superantigens to T-cell hybridomas. Mutations at the E alpha amino acid positions 31, 52, and 65&66, to residues that represent allelic alternatives in A alpha chains, significantly reduced activation of peptide-specific T hybridomas, and mutations at 71 sometimes enhanced T-cell stimulation. None of the E alpha mutations reduced, and some enhanced, superantigen stimulation of T-cell hybridomas. These results argue against the hypothesis that E alpha chains are minimally polymorphic because mutations in E alpha are functionally neutral.

Competitive PCR quantification of CD4, CD8, ICAM-1, VCAM-1, and MHC class II mRNA in the central nervous system during development and resolution of experimental allergic encephalomyelitis

We used competitive polymerase chain reaction to quantify messenger RNA for the lymphocyte antigens CD4 and CD8, the adhesion molecules ICAM-1 and VCAM-1, and the MHC class II I-A molecule in the spinal cords of SJL/J mice at multiple times during the development and resolution of experimental allergic encephalomyelitis (EAE). CD4 and CD8 were not quantifiable at baseline, became detectable at 5 days after immunization, and increased steadily to a peak during clinical disease. I-A increased after CD4 and CD8, but before onset of disease. ICAM-1 and VCAM-1 did not increase until after onset of clinical disease. CD4, CD8, and I-A remained elevated long after recovery from disease. These results suggest that infiltration of CD4 and CD8 cells into the spinal cord and subsequent upregulation of I-A mRNA play an important role in the development of EAE, but reversal of these processes is not necessary for recovery. Upregulation of ICAM-1 and VCAM-1 mRNA does not appear to be important for development of disease.