Peter Parham

Board Member: 2007 -2012

Scholar: 1982

Awarded Institution
Stanford University
Department of Structural Biology


Research Interests

Structure, Function, and Evolution of Major Histocompatibility Complex Class I Polymorphism

Major Histocompatibility Complex (MHC) class I molecules are one of four families of antigen binding molecules that give the vertebrate immune system specificity and memory, properties harnessed in the medical practice of vaccination. The primary responsibility of MHC class I molecules is to reveal to cytolytic T lymphocytes those cells of the body that are compromised by infection or malignancy. To achieve this end MHC class I molecules bring to the cell surface an array of peptide fragments that derive from the proteins found within the cytoplasm and endoplasmic reticulum of the cell. Each peptide is bound by a single MHC class I molecule and such complexes form potential ligands for the antigen receptors of cytolytic T cells During development T cells are selected and "tolerized" to be unresponseive to the peptides derived from normal proteins but will react to peptides derived from the proteins of intracelllular pathogens or the altered proteins of cancer cells.

Certain pathogens, particularly viruses, have evolved mechanisms for evading the cytolytic T cell response of the host that involve disruption of expression of MHC class I molecules. Analogously, during the growth of tumors there is selection for cells that lose class I expression. To counter such strategies the host's immune system uses a second type of cytolytic lymphocyte known as the natural killer cell. These cells possess receptors that on interaction with polymorphic determinants of MHC class I molecules inhibit the cytolytic machinery.

Diversity of antigen binding by MHC class I molecules is built upon three principles: firstly, each class I molecule has the capacity to bind peptides whith many different sequences; secondly, each organism expresses a number of different class I genes; and thirdly, the class I genes are polymorphic such that most individuals within a population are heterozygous. A consequence of the polymorphism of MHC class I molecules is that they are major immunological factors causing rejection of tissue transplanted between unrelated individuals. For this reason, typing of human MHC class I polymorphisms (alleles of the HLA-A, B and C genes) is used clinically to match donor and recipient tissues for transplantation.

A major thrust of our research has been to build upon the results of clinical HLA typing and define HLA-A, B and C polymorphism at the level of nucleotide sequence. Over two hundred alleles have been studied so far and their comparison has produced general understanding of how the human system works. Amongst the allotypes of a given locus (HLA-A, B or C) differences are concentrated within the peptide binding site and serve to modify its specificity. Allotypes can differ by just one or many amino acid substitutions; the repertoires of peptides bound by two allotypes diverge commensurately with the allotypic differences. The main selection for HLA polymorphism appears to be for heterozygosity, giving individuals two distinctive peptide binding specificities per locus. Most new alleles are formed by recombination between existing alleles which reassort small segments of polymorphic sequence. Point mutations less commonly form new alleles, but once fixed, a point mutation is rapidly recombined with existing substitutions to form new recombinants.

The study of native populations in North and South America has shown that within relatively short periods of time (10,000 - 35,000 years) HLA class I genes can evolve rapidly. Moreover the extent of the evolution can vary between populations. Thus, HLA-A, B, and C in North Amerindian populations has remained stable while in South Amerindians there has been considerable evolution, particularly for alleles of the HLA-B locus. A burning question is the extent to which these differences depend upon different histories of population structure or pathogen exposure. Although many new HLA class I alleles have evolved in South Amerindian populations, the overall number of alleles for a give population has not increased because of commensurate loss of "old" alleles. That Amerindians and other indigenous populations have small numbers of HLA alleles (3-8 alleles per locus) compared to urban populations indicates that the very large numbers of alleles found in the city are the result, not of natural selection upon the immune system, but of the economic forces which have formed cities through the mixing of previously separated populations. This effect is perhaps most vividly illustrated by the cosmopolitan populations of North American cities, and as a consequence it is in these populations that the matching of HLA type for transplantation is most challenging.

To place our knowledge of humans in context, we are investigating the MHC class I polymorphism of chimpanzees, our closest relatives. The organization of the MHC locus in chimpanzees is similar to that of humans, and genes corresponding to HLA-A, B and C are present. Although many individual substitutions and sequence motifs are shared by the MHC class I alleles of humans and chimpanzees, no alleles are held in common, again testifying to the continual formation of new alleles and loss of old alleles. In collaborative studies with Dr. Christopher Walker of Chiron Corporation, we are analyzing the cytolytic lymphocyte response in chimpanzees that have been infected with human hepatitis C virus. The resulting disease in chimpanzees parallels the human condition in that most individuals develop a chronic condition, in which the liver undergoes cycles of destruction and regeneration, while only a few individuals resolve the infection. Cytolytic T cell responses are instrumentl in the disease process, though whether their effect is beneficial or detrimental is unclear. Resolution of these issues is a current goal. Of particular interest is the discovery of a structurally unusual class I allele that is shared by the few animals that resolved infection.

The chimpanzee cytolytic T cells we have found are all directed to peptides derived from hepatitis C virus that are presented by allotypes of the A and B loci. The lack of involvement of the C locus is consistent with observations on other viral infections. In contrast to this relative quiescence towards T cells is the emerging impression that allotypes of C are active in regulating natural killer cells. The HLA-B and C loci have features in common, suggesting they were formed by duplication from a common ancestral gene. Since that event, the two genes have diverged in terms of structure, levels of gene expression and functional focus. Ongoing investigations are aimed at determining which of the structural differences are responsible for the functional specialization of HLA-B and C.