MAJOR HISTOCOMPATIBILITY COMPLEX

Chapter 7

The Major Histocompatibility Complex (MHC) represents a tightly linked gene cluster first recognized for its role in self/nonself discrimination. The MHC plays a major role in the acceptance (histocompatible) or rejection (histoincompatible) of tissue transplants, and hence the name given to this region of multiple loci. We now know the MHC plays an important role in the development of an immune response, both humoral and cell mediated. MHC's central position in the formation of an immune response is derived from its role in presenting Ag to T-cells. Because T-cells can only see Ag presented by MHC, the specific set of MHC molecules presented by an individual influences the repertoire of Ags to which an individual can respond against. Therefore, the expression of MHC by an individual can in part define the individuals susceptibility to disease as well as the tendency for the development of autoimmunity and allergies. There is also information that the MHC may influence the mating process to increase variability of the population. In many animals MHC molecules are secreted in urine and other body secretions that attract animals to each other and that divergence in MHC produce more attractive scents. Even in humans certain characteristic body odors are associated with the expression of specific MHC types.

MHC Gene Organization:

• Gorer and Snell in the 1930's while studying inbred mice identified 4 groups of blood-cell Ags (I-IV) and eventually mapped rejection of transplanted tumors and other tissues to group II Ags. The histocompatibility-2 (H-2) gene cluster was named in recognition of this work for which Snell received the the Nobel Prize in 1980.
• The MHC gene complex is organized along chromosome 6 in humans (HLA) and 17 (H-2) in mice. The organization of the genes differs slightly between species, however, all contain genes that code or 3 classes of molecules (fig 7-1 Kuby)
• Class I MHC is encoded in regions designated K and D* in the H-2 locus, and Qa and Tla known as nonclassical class I in the closely linked Tla locus in mice. In humans the Class I are encoded in the A, B, & C regions of the HLA complex with nonclassical genes HLA-E through H near by. The products of Class I MHC genes are glycoproteins found on nearly all nucleated cells of the body. They serve to present Ags of altered self or internal agents (viruses) to Tc cells.
• Class II MHC genes include IA and IE regions of mice and the DP*, DQ*, and DR* regions in humans. Class II MHC gene products are glycoproteins that are expressed only on membranes of Antigen Presenting Cells (APCs). APCs including macrophage, dendritic cells, and B-cells, which present processed exogenous Ags to Th cells. Nonclassical Class II MHC genes also exist such as DM, involved in loading peptides into Class II molecules.
• Class III MHC is encoded in the S for soluble region in mice and the C2, C4, BF region, named for the soluble products coded for in this region. Class III MHC genes encode a diverse group of secreted proteins which function in the immune process. These include serum proteins, some components of the complement cascade (C4, C2), and tumor necrosis factors.

* Note that in in mice the D region codes Class I molecules and in humans the DP, DR, & DQ regions code Class II molecules.

The most recent attempt to reorganize the MHC groupings for Class I molecules is presented in the following table; however, it may be some time before this will be accepted. The organization represents the separation of classical and nonclassical MHC genes and the further separation of the nonclassical genes over evolution. Ref. Immunol Today, 20:22(1999)

	Class 1A	Class 1B	Class 1C	Class 1D
Human	HLA-A HLA-B HLA-C	HLA-E HLA-F HLA-G	MICA MICB Hfe	FcRN Zinc-a2-gp MR1
Mouse	H-2K H-2D H-2L	H-2Q region H-2T region H-2M region	Hfe	FcRN CD1

MHC Haplotypes:

• the loci that make up the MHC are highly polymorphic, with many alleles present at each locus.
• the MHC is closely linked (recombination frequency in H-2 is only 0.5%), therefore we inherit our alleles in sets, one from each parent. Each set of alleles is called a haplotype, with one haplotype donated by each parent.
• both haplotypes are expressed, the alleles are therefore codominantly expressed. Each nucleated cell in your body expresses both your fathers and mothers alleles. In inbred populations the maternal and paternal loci are the same and therefore the population becomes homogeneous
• inbred mouse strains (prototype strains) of defined haplotypes have been developed and designated by specific haplotype identifiers that are shorthand for the complete description of the set of alleles (table 7-1 Kuby). In crosses between 2 different prototype stains the F1 generation will express the haplotypes of both parents and will be an able recipients of grafts from either parent, but will not be able to donate grafts to either parent.
• SYNGENEIC & CONGENIC STRAINS Mice inbred to be identical at all loci are syngeneic, while congenic inbred strains are genetically identical except at a specific loci. Congenic mice that differ at H-2, have allowed for the detailed study of H-2. Further, recombinant congenic strains, resulting from cross-overs that differ from the congenic stain in only a few genes within the MHC (fig 7-3;7-4 Kuby) have refined these studies.

Structure -Function of MHC Molecules:

Class I and II MHC molecules are membrane bound members of the immunoglobulin superfamily, MHC class III gene products differ from Class I & II proteins and from each other.

Class I MHC:

• the Class I molecule is made up of one chain coded for in the MHC locus, call the alpha chain.
• the alpha chain is a polymorphic, 45 kDa glycoprotein coded for in A,B, or C in HLA or K and D/L of H-2.
• the alpha chain is associated with a second smaller (12 kDa) invariant chain, which is required for function, but which is not a product of MHC genes called beta-2 microglobulin.
• the alpha chain is organized into 3 external immunoglobulin-like domains alpha-1, alpha-2 and alpha-3 (90 amino acids each), a hydrophobic transmembrane region, and a short cytoplasmic tail .
• the alpha-3 domain and the beta-2 microglobulin share homology to each other and to the constant regions of immunoglobulins. They associate together in the Class I molecule to form the membrane-proximal domains, serving much like the constant region of Ig. The alpha-3 domain, like the beta-2 domain, shares homology with constant region domains of Ig. It is highly conserved in Class I molecules and contains a recognition site for CD8 binding.
• alpha-1 and alpha-2 give rise to the membrane-distal domains, which serves as the peptide-binding cleft. It can hold a peptide of 8-10 amino acids. They form a deep pocket of approximately 25Å X 10Å X 11Å in size constructed around a base of 8 antiparallel beta strands and 2 long alpha-helical regions. The final structure, formed by alpha-1 and alpha-2 coming together giving rise to the Ag-binding site, is also dependent upon interactions between beta-2 microglobulin and alpha-1 and alpha-2 to initiate the final folding of the molecule.
• the assembly of the Class I molecule occurs in the ER, requires the help of a set of chaperone proteins and is initiated by beta-2 microglobulin binding to the alpha-chain to form a metastable "empty dimmer" that requires the binding of Ag to maintain the structure and for transport to the membrane. Therefore, without each of the three components of the Class I complex expression of Class I on the membrane surface will not take place.

Class II MHC:

• Class II MHC is dimeric, with both peptides coded in the MHC. The alpha chain is 33 kDa while the beta chain is 28 kDa in size. Both chains contain 2 external immunoglobulin domains, a transmembrane domain, and a cytoplasmic domain.
• the alpha-2/beta-2 domains of Class II are similar to alpha-3/beta-2 microglobulin domains of Class I. They form the membrane-proximal domains, possess Ig-like folds making them part of the Ig-superfamily, and form the equivalent to the constant region of an Ig molecule
• the membrane-distal region of Class II is formed by alpha 1/beta-1 domains, which form the Ag-binding region. The structure of the Ag-binding region is similar to that seen in Class I molecules.The site contains 8 antiparallel beta strands and alpha helix structures forming the lips of the pocket. In crystal structures the Class II molecule is organized as a dimmer of heterodimers (alpha/beta). It is not known if this occurs in vivo.

Peptide Binding by MHC Molecules:

• each individual can express up to 6 different Class I molecules and up to 12 different Class II molecules. In spite of this limited expression of MHC these individuals are able to present a vast number of antigens to the immune system. This would suggest that the MHC molecules are able to present a large number of peptides each. It also means that Ag-binding to MHC doesn't follow the specificity seen in Ag-Ab interactions, as MHC is able to bind a number of different Ags. These interactions are said to be "promiscuous". Class I and Class II molecules have similar binding sites so it is expected that the peptides they bind will also be similar in structure. Both bind linear peptides that fit into the cleft. As the Class I cleft is closed at its ends it can only hold a peptide of 8-10 amino acids, while Class II, which is open at its ends, can hold a peptide of 13-18 amino acids. Equilibrium dialysis has shown the binding between MHC and peptide to have a KD in the order of 10-6 meaning that most MHC molecules will be associated with self or nonself peptides (See figs 7-10 to 7-13)

Class I MHC-Peptide Interactions:

• peptides of Class I are derived from endogenous proteins digested into peptides in the cytosol and moved to the ER by processing pathways discussed later.
• each Class I molecule binds a unique set of peptides, and each allelic variant of Class I (H-2Kk and H-2Kd) binds a distinct set of peptides.
• each cell, which has on its surface 105 copies of each Class I molecule, will display many different peptides on its surface at any one time. In healthy cells most of the peptides are self proteins and won't initiate a response by the immune system. It is estimated that each peptide will be present at 100-4000 copies/cell, with as few as 100 copies/cell being required to initiate an immune response.
• the bound peptides are normally nonamers, suggesting this length of peptide is most suitable for the peptide-binding cleft which is closed at both ends.
• peptides able to bind specific MHC types contain specific amino acids at specific anchor locations that define MHC-peptide interactions. All Class I binding peptides contain a carboxyl terminal anchor (the majority of which are hydrophobic) and an anchor at position 2 or 2 & 3 from the amino-terminal end. The rest of the peptide arches off the floor of the groove to interact with the TCR (See figs 7-10 to 7-13). Knowledge of these anchor residues may allow us to identify and/or design custom peptides to initiate an immune response to a complex antigen.

Class II MHC-Peptide Interactions:

• peptides bound by Class II MHC are generally exogenous self (associated with the plasma membrane or endocytic vesicles) or nonself proteins degraded within the endocytic processing system. Most are derived from membrane bound proteins internalized by phagocytosis or receptor-mediated endocytosis.
• peptides released from Class II MHC tend to be 13-18 amino acids in size. The binding site of Class II is open allowing for longer peptides to extend out like a hot dog in a bun. The peptide also lies flat in the binding grove, unlike Class I peptides. The peptides bound by a class II molecule have conserved motifs all along the molecule, rather than anchor residues at the ends of the peptides. The presence of core sequences made up of aromatic or hydrophobic amino acids, one at the amino terminus, 2-3 in the center and one at the carboxy terminus, account for the binding. Most peptides also contain proline residues at position 2 and another at the carboxy-terminal. The prolines may be frequent due to their presence at cut sites during processing.

Organization of Class I & II Genes:

• the cloning and sequencing of a number of Class I genes has demonstrated that separate exons encode each domain of the protein (fig 7-9, Kuby). The gene codes for a 5"-leader sequence, and exons coding for the alpha-1, alpha-2, and alpha-3 domains, the transmembrane domain and 2 exons coding the cytoplasmic domain.
• for Class II, the genes are again organized into a series of exons and introns, with the exon order in the same sequence as the structural domains of the proteins.
• detailed mapping of the Class I region has demonstrated the presence of nonclassical MHC genes. These are MHC-like and are called Class Ib molecules. They have less diversity and different tissue distribution. The role of these proteins are not well understood, but they may play a role in presenting peptides of intracellular prokaryotes such as M. tuberculosis, L. momocytogenes, B. abortus, and S.typhimurium.
• mapping of the Class II region has also defined nonclassical genes, including DM which seems to function in loading antigenic peptides into Class II MHC molecules. This region also contains LMP2 and LMP7 that encode proteosome subunits, and TAP1 and TAP2 that encode peptide-transporter subunits

Polymorphism of Class I and Class II Molecules:

• the diversity seen in the MHC is very different from the diversity seen in the generation of B-cell receptor (Ab) and T-cell receptor. In Ab and TCR gene rearrangement take place to create a changing diversity in each individual clone of cells. That is, within an individual there is diversity between his/her own B- and T-cells. In MHC the diversity is within a population, my cells MHCs are different than yours, however all the MHC molecules in an individual are identical. The diversity of MHC is not derived by gene rearrangement but rather by the presence of multiple alleles at a given genetic locus-POLYMORPHISM.
• the MHC is one of the most polymorphic genetic complexes known in higher vertebrates. In human Class I 59 A alleles, 111 B alleles, and 39 C alleles have been identified. In mice 55 K and 60 D alleles are known. MHC can therefore be said to be polygenic. Estimates of actual polymorphic structures based on serology and binding studies suggest >100 alleles at each locus. This would create a theoretical diversity in mice of:100(K) X 100(IA-alpha) X 100 (IA-beta) X 100 (IE-alpha) X 100 (IE-beta) X 100 (D) = 1012
• while the diversity within the MHC of a species is as high as one would expect to see with other loci between species (5-10% at the amino acid level),the diversity is not as great as would be predicted above. The fact that diversity is not as great as expected and that certain combinations of alleles combine more frequently than anticipated is termed LINKAGE DISEQUILIBRIUM. Three hypotheses for this are FOUNDER HYPOTHESIS, POSITIVE/NEGATIVE COMBINATIONS, & HOTSPOT CROSSOVER. Diversity is also not random throughout the molecule but is concentrated at the membrane-distal domains of both class I & II molecules (Fig 7-14a). GENE CONVERSION is believed to account for this divergence. Gene conversion involves the transfer of a short donor sequence, following its base-pairing with a partially homologous sequence of a recipient gene followed by excision, repair and replication mechanism which inserts the donor into the recipient DNA.The exact mechanism isn't understood but it is believed that the many pseudogenes in the MHC locus may provide the donor material.
• it is believed that the polymorphism is responsible for the different Ag binding patterns displayed by the MHC. As pointed out, the highly variable regions of MHC fall in the Ag binding membrane-distal domains.

Class III Molecules:

• the Class III region codes for a number of different proteins which function in the immune system.These include:
  • the complement proteins C2, C4, and factor B
  • two steroid 21-hydroxylase enzymes
  • tumor necrosis factors alpha & beta
  • two heat-shock proteins
• it is not known why the Class III proteins are found in the MHC complex. There are interesting links to human disease and class III protein linkage with MHC molecules; for example ankylosing spondylitis is linked to the HLA-B27 allele, which is closely linked to TNF-alpha and TNF-beta genes, which may be responsible for the destruction of cartilage seen in this disease. There is also evidence in Lupus and in other autoimmune diseases of the involvement of other class III genes including heat-shock and complement genes.

Cellular Distribution of MHC Molecules:

• Class I MHC is found on most somatic cells but at different levels. It is highest on lymphocytes (1% of membrane protein or 5 X 105/cell), low levels are found on fibroblasts, muscle cells, liver hepatocytes, and neural cells, and is not expressed on neurons, sperm, and placenta. Class I molecules will include both maternal and paternal haplotypes for a total of 6 different class I MHC molecules.
• Class II MHC are only found on Ag-presenting cells (macrophage, dendritic cells and B-cells). The expression on these cells can change. Immature B-cells don't express Class II, however mature B-cells and dendritic cells constitutively express class II. Monocytes and macrophage express low levels of class II, but when activated by Ag express high levels of class II. The number of types of class II molecules expressed on the cell surface is greater than the number of genes for class II. First both haplotypes will be expressed (4 in mice 6 in humans) and because the class II molecules are composed of 2 peptides all combinations of the alpha and beta chains can be found producing 6 possible arrangements in humans. This is increased further by the presence of multiple beta chains in human and mouse MHC and multiple alpha chains in human. This would increase the numbers of peptides that can be presented to Th cells.
• regulation of class I & II expression is important. Two factors regulating the cell specific expression of class II MHC are CIITA (a transactivator) and RFX (transcription factor). Cytokines can also regulate MHC expression. IFN-gamma increases class I expression by up-regulating transcription of both the alpha chain and beta2-microglobulin, as well as LMP and TAP genes and it can induce expression of class II on cells not normally expressing class II (intestinal epithelia, vascular endothelia, skin) by inducing the expression of CIITA in these cells. IFN-gamma reduces class II expression on B-cells. Other cytokines can have a more limited effect on expression, ( IL-4 up-regulates class II on resting B-cells). Corticosteroids and prostaglandins can decrease class II expression, and many viruses also effect MHC expression in order to evade the immune system.

Immune Responsiveness:

• the ability to respond to an antigen, IMMUNE RESPONSIVENESS, has been mapped to the Class II MHC genes. This is due to the central role played by class II in presenting Ag to Th cells.
• · Two models have been advanced to account for the variability in responsiveness seen between haplotypes.
  • DETERMINANT-SELECTION MODEL- states that different class II MHC molecules will differ in their binding of peptide Ags, therefore MHC polymorphism will generate different patterns of responsiveness or nonresponsiveness to an Ag. Equilibrium dialysis studies have demonstrated different affinities of MHC for different peptides, and that a correlation exists between affinity and ability to respond to the Ag. Therefore, if you don't have an MHC molecule that can bind and display the peptide Ag you can't respond to the Ag.
  • HOLES-IN-THE-REPERTOIRE MODEL- T-cells capable of responding to the MHC-Ag have been eliminated, therefore it is not possible to respond. Again if you don't have MHC-Ag recognition, there is no activation of the immune response

MHC and Disease Susceptibility:

The predilection of certain diseases, such as some autoimmune diseases, some viral infections, some disorders in immune (complement) or neural systems and in hypersensitivity have been linked to specific MHC types. The Relative Risk associated with specific MHC alleles can be calculated by comparing allele frequency in a patient population against its frequency in the general population.