Human Major Histocompatibility Complex Genes: Class I Antigens and Tumor Necrosis Factors

In this paper I discuss two topics relating to genes of the human major histocompatibility complex (MHC) in which new information has recently been obtained. They are (1) the use of cytotoxic T lymphocyte (CTL) variants and site-directed mutagenesis to map func-tional sites in class I MHC molecules and (2) the regu-lation of expression of class I genes by tumor necrosis factor (TNF) and the fact that TNFa and the closely related lymphotoxin (TNF/3) are themselves encoded in the MHC. The human class II genes are encoded at the cen-tromeric end of the human MHC and the class I genes are encoded at the telomeric end on chromosome 6 in the region of 6p21.1-.3. The elucidation of the protein structure of the human class I antigen showed that it was composed of two chains, a and/~, comprising four extracellular domains, three of them in the ct chain (a 1, c~2, and a3) and the fourth in the/ ~ chain (/~2 micro-globulin, or/32m). This domain structure was later sup-ported by the elucidation of the class I gene structure that showed that each domain is represented by a sep-arate exon. A physical map of the class I region of the MHC is not yet completed, but at the present time 17 different class I genes, all pseudogenes xcept hose en-coding HLA-A,-B, and-C, have been identified. The human homologs of the murine Qa and Tla genes have not yet been found (B. Koller and H. Orr, pers. comm.). In the class II proteins the four domains are linked differently in that the c ~ and/ ~ chains each contain two domains. In addition, only one chain of class I proteins spans the membrane whereas both chains of class II proteins span the membrane. Our knowledge of the biochemistry is not yet sufficiently sophisticated to tell us how these seemingly small differences in structure are reflected in the rather large differences in function of these two classes of molecule