Structure, Function, and Diversity of Class I Major Histocompatibility Complex Molecules

Abstract

Lymphocytes are the agents of antigenic specificity in the immune response. They divide into two groups: B cells that make immunoglobulins (Igs), and cells that perform a heterogenous set of functions including help for B cells, production of delayed-type hypersensitivity reactions, and the specific killing of virus-infected cells. Igs are the sole source of B cell specificity, whereas molecules from two diverse families of cell-surface glycoproteins, the T cell receptors (TcRs) and the major histocompatibility complex (MHC) glycoproteins, are the key elements of specificity in the T cell response to foreign antigens. TcRs, in common with the Ig receptors of B cells, are products of somatically rearranging genes and are clonally expressed (reviewed in Refs. 1, 2 and M. M. Davis, this volume). There are, however, two important differences in antigen recognition by these related groups of lymphocyte receptors. The first is a difference in the type of antigenic determinant (epitope) recognized: TcRs recognize short, linear, peptide determinants of 10-20 amino acids (3-5), the generation of which usually requires unfolding and proteolytic fragmentation ("processing") of the antigenic protein (6). By contrast, Igs commonly interact with epitopes formed by the three-dimensional structure of native proteins, although they can also be raised against peptides. The second difference in antigen recognition by Igs and TcRs is the involvement of a third molecule that performs the role of "presenting" the antigen to the receptor. For B cells such molecules do not exist; the antibody receptor forms a stable bimolecular complex with the antigenic protein. For T cells, the antigenic peptide must be bound by an MHC glycoprotein, and it is this complex of MHC molecule plus peptide that forms the structure rccognized by the TcR. MHC glycoproteins are thus peptide-binding proteins, and can be considered as antigen-presenting molecules. There are two structurally distinct, but related, families of MHC molecules that present antigens to two subsets of T cells: Class I MHC molecules present antigens to T cells that express the CD8 cell-surface glycoprotein, and class II MHC molecules present antigens to T cells that express the CD4 cell-surface glycoprotein (7, 8). Other less rigid distinctions between class I and II MHC molecules are in the categories of antigen presented and in the functional activities of the responding T cells. Class I molecules commonly present peptides derived from endogenously synthesized proteins, such as viral components produced upon virus infection, and result in the stimulation of CD8-bearing cytotoxic (killer) T cells (9). On the other hand, class II molecules generally present peptides derived from exogenously synthesized proteins and stimulate CD4-bearing helper T cells (10). This specificity in the antigens presented by class I and class II molecules is thought to result from cellular segregation of the compartments or sites at which class I and class II MHC molecules encounter peptide (11, 12) (Figure 1). That T cells have a dual requirement for recognition of both foreign antigen and a self-MHC molecule was first recognized by cellular immunologists in the early 1970s (13, 14): T cell recognition of antigen became known "MHC restricted," with considerable debate focussing upon its molecular interpretation and whether one or two receptor molecules were involved (15). Transfection experiments have now demonstrated that a single TcR recognizes both foreign antigen and MHC molecule (16). Discovery of the T cell requirement for fragmented protein antigens led to investigation of the antigen processing pathway, and revealed that incubation of antigen-presenting cells with short peptides circumvented the need for processing and provided the necessary target for the TcR (3, 17). Combining the concepts of MHC restriction and antigen processing led to the hypothesis that MHC molecules actually bind processed antigen fragments, thereby forming ligands for TcRs. The first direct evidence for this model came with the demonstration of binding of specific antigen peptides from hen egg lysozyme to a murine class II MHC molecule (18). The experiments that defined and characterized MHC restriction were only possible because MHC genes are polymorphic, with the products of different alleles being distinguishable by both antibodies and TcRs. Indeed, alloreactive responses in which the immune system of one individual makes antibodies and T cells specific for the MHC molecules of another, are the basis for the immunological rejection of transplanted tissues that led to the discovery of the Major Histocompatibility Complex (19). The number of alleles at MHC loci and the complexity of their patterns of substitutions are unique when compared to other mammalian genes, and the basis for this diversity and its functional significance are long-standing questions (20). Analyses of MHC restriction (14), of T cell epitopes (5), and of peptide binding (21) all showed that polymorphic differences between MHC proteins affected the peptide antigens presented and the T cells that could respond. Characterization of MHC alleles revealed specific regions of the encoded protein at which sequence diversity was concentrated (22), and the marked predominance of coding over silent substitutions within those regions strongly suggested that diversification was the result of natural selection (23, 24). A model for how MHC restriction might work was eventually developed (15, 25): a TcR was postulated to interact with a complex of a peptide bound to a particular class I or class II MHC molecule, and a T cell selected by one combination of peptide plus MHC would only be restimulated by that same or a closely related peptide-MHC complex. The specificity of restriction would then result from sequence variability in the MHC molecule, which determines both the peptides capable of binding and the TcR selected. Determination of the three-dimensional structure of HLA-A2a, human class I MHC molecule (26, 27), has confirmed the general principles of this model and revealed considerable insight into the details of the molecular interactions involved and of the role played by MHC polymorphism. In addition, the structure has helped to rationalize the kinetics of peptide binding to MHC molecules and the inaccessibility of the site in all but a fraction of purified MHC molecules (28, 29). More importantly, it has provided a mechanistic understanding immunological tolerance -- the discrimination between self and nonself -- which is arguably the most fundamental property of the immune system. This review is focussed on the three-dimensional structure of HLA-A2 and its interpretation in terms of the diversity of human class I MHC molecules, and their interactions with peptides, TcRs, and the CD8 glycoprotein. Some of the conclusions, but undoubtedly not all, will extend to class II MHC molecules.

Files

Additional details