HLA      MHC     Genetics      Evolution      Epidemiology     Biostatistics      Glossary       Homepage

 

Major Histocompatibility Complex

 

M.Tevfik Dorak, MD, PhD

 

The major histocompatibility complex (MHC) is a set of genes with immunological and non-immunological functions and present in all vertebrates studied so far 1;2. It was discovered during transplantation studies in mice (as the H-2 complex) by Peter A Issac Gorer in the Lister Institute in London in 1937 who later collaborated with George Snell of the Jackson Laboratories in Ben Harbor 3;4. Jean Dausset described the first human MHC antigen MAC (HLA-A2) as part of the Hu-1 system 5 followed by the discovery of the FOUR series 4a and 4b (HLA-Bw4 and -Bw6) by the Leiden group led by Jon van Rood in 1963 6;7. Rose Payne and Walter & Julia Bodmer identified the LA series (1964). Bernard Amos, who had originally worked with Gorer, organised the first International Histocompatibility Workshop in 1964 and the first WHO Nomenclature Committee Meeting in 1968 (see IHWG website; Marsh, 2004). See full references for early MHC-HLA work.

 

The function of the MHC can be described as pleiotropic, i.e., multiple unrelated ones 8-11. It is best known with its role in histocompatibility 12 and in immune regulation 13-17 with many other functions not much appreciated yet 2;18-22. The main function of the main MHC molecules is peptide binding and presentation of them to T lymphocytes. Among the non-immune functions, the noteworthy ones are interactions with other receptors on the cell surface 23;24, in particular with transferrin receptor (TfR), epidermal growth factor 25 and various hormone receptors 26-28, and signal transduction 29.

 

In nature, different taxa of multicellular organisms have unrelated compatibility systems such as the protozoan pheromone system 30;31, the fungal compatibility system 32-35, angiosperm (flowering plants) self-incompatibility system 33;36, and the invertebrate allorecognition systems 37-39. All of these systems are primarily involved in prevention of matings between genetically similar individuals to avoid the harmful effects of inbreeding. The MHC also prevents inbreeding through its influence on mate choice in mice 40;41 and humans 42;43; and on reproductive processes in rats 44, mice 45;46 and humans 47;48. The reproductive mechanisms are varied and range from selective fertilization to selective abortion. A major common feature of the compatibility systems is that they favour genetic dissimilarity between mates and the gametes (mate choice, selective fertilization); but similarity in co-operation (kin recognition, dual recognition, transplant matching) 49;50. All these functions are based on the provision of a phenotype for the genetic identity of the individual by the MHC: either cell surface molecules or chemosensory signals.

 

MHC structure

The MHC in humans is called Human Leukocyte Antigens (HLA). It is located on chromosome 6p21.31 and covers a region of about 3.6 Mbp depending on the haplotype 1;51. The longest haplotype is the HLA-DR53 group haplotypes because of the 110-160 kb extra DNA in their DR/DQ region 52-56. The HLA complex is divided into three regions: class I, II, and III regions as first proposed by Jan Klein in 1977 57. The telomeric region to the classical HLA complex is now called the class Ib region; and there has also been a suggestion for a class IV region located at the telomeric end of the class III region 2. The classical HLA antigens encoded in each region are HLA-A, -B, and -C in the class I region, and HLA-DR, -DQ and -DP in the class II region. All class I genes are between 3 and 6 kb, whereas, class II genes are 4-11 kb long 58. The 1998 Nomenclature Committee recognized more HLA genes all of which are in the class I and Ib regions: HLA-E, -F, -G, -H, -J, -K and -L  59. Among those, only HLA-E, -F and -G are expressed 60. The massive sequencing project of a human MHC haplotype has just been completed and the map positions of all of these genes are known 51. The class III region has the highest gene density but some of the genes are not involved in the immune system 2;61. Among the genes which are of interest, HSP70, TNF, C4A, C4B, C2, BF and CYP21 should be mentioned. The HSP70 genes encode cytosolic molecular chaperons and might have donated to the PBR region to the ancestor MHC gene 62. It has also been proposed that HSP70 may be the functional forerunners of MHC molecules because of their peptide binding and presenting abilities 63. By presenting intracellular contents of a cancer cells to the immune system, HSP70 behaves like a tumour rejection antigen 64-68 similar to the other molecular chaperons, calreticulin and grp94/gp96 67-69. An important feature of HSP70 alleles makes this locus a useful one in disease association studies. They show strong linkage disequilibrium (LD) with HLA-DR alleles 70-72. TNF(A) and TNFB (LTA) genes encode cachectin and lymphotoxin-a molecules, respectively 2;73. C2, C4A and C4B are the genes for some of the complement proteins, whereas, BF codes for factor B which is also involved in immune response 74;75. CYP21 is the gene for 21-hydroxylase which is an important enzyme in corticosteroid metabolism. Its complete deficiency causes congenital adrenal hyperplasia which was the first disease identified to be the result of a structural change in an HLA-linked gene 76. Other genes of interest in the class III region are the human homologue of the mouse mammary tumour integration site Int-3, NOTCH4, and the homologue of a homeobox gene similar to PBX1 involved in t(1;19) translocation in pre-B cell ALL encoded on chromosome 1q23, PBX2 (or HOX12) 2;77-79.

 

Classical genetics

A highly relevant feature of the MHC antigens is their co-dominant expression. Since both alleles contribute to the phenotype equally, it is important to investigate the genotypes in disease association studies rather than the alleles on their own. If susceptibility to a disease is a recessive trait, allelic association studies may not yield a positive result. Also important is the fact that the MHC is inherited en bloc as a haplotype with the exception of the rare recombinational events. Recombination occurs at 1-3% frequency mostly at the HLA-A or HLA-DP ends, i.e., in 100 meiosis the haplotype will be broken and reconstituted in one to three of them. The large segment from HLA-B to HLA-DQB is almost always inherited as a whole. This also has important implications in disease associations. A haplotypical association is usually stronger and more meaningful than an allelic association.

 

The co-dominant expression and haplotypical transmission have an important consequence: within a family, HLA-identical sibling frequency should be 25% according to Mendelian expectations. This has been, however, found to be higher than that in leukaemia 80-84. This would suggest preferential transmission of leukaemia-associated HLA haplotypes 85. The fact that HLA-identical sibling frequency is higher than 25% in leukemic families should not be confused with the overall chance of having an HLA-identical sibling which is correlated with the family size (equal to [1 - (0.75)n] where n is the number of siblings). This probability may go up to 55% in areas where families are traditionally large 86.

 

Despite the enormous number of alleles at each expressed loci, the number of haplotypes observed in populations is much smaller than theoretical expectations. This is to say that certain alleles tend to occur together on the same haplotype rather than randomly segregating together. This is called linkage disequilibrium (LD) and quantitated by a D value 87;88.

 

The public specificities, also called supertypes and sometimes wrongly broad specificities, group a number of private specificities. In the HLA class I region, all HLA-B private specificities are grouped into two supertypical families: HLA-Bw4 and -Bw6. In recent years, the nature of HLA-B supertypes has been better understood. They are not encoded by a different gene. The antigens HLA-Bw4 and -Bw6 reside on a unique epitope on each HLA-B molecule and are distinctly different from the epitopes that determine the HLA-B specificity. Each HLA-B molecule expresses either the Bw4 or Bw6 supertype (residues 74 to 83 of the a 1 helix) in addition to a (private) HLA-B specificity. The amino acid residues 80 IALR 83 represent the Bw4 specificity, whereas, 80 NLRG 83 represent Bw6 (Ref 89).

 

Likewise, HLA-DR alleles are also associated with supertypes. However, the HLA-DR supertypes are not allelic with each other 90. They are encoded by separate genes (HLA-DRB3, -B4, -B5) and are distinct molecules (HLA-DR52, -DR53, -DR51, respectively). Only one or none of these genes occurs on a haplotype.

The private specificities in each supertypical family are as follows:

 

DR51 (DRB5): DR2 (DRB1*15/16)

DR52 (DRB3): DR3 (17/18; DRB1*03), DR5 (DRB1*11/12), DR6 (DRB1*13/14)

DR53 (DRB4): DR4 (DRB1*04), DR7 (DRB1*07), DR9 (DRB1*09)

 

Although all HLA-DR4 / 7 / 9 haplotypes carry the structural gene HLA-DRB4, not all of them express the HLA-DR53 molecule 91. The non-expression, however, is restricted to the HLA-(B57) : DR7 (Dw11): DQ9 haplotype 92 due to a G to A substitution in the acceptor splice site at the 3' end of the first intron, changing the normal AG dinucleotide to AA 93;94. In fact, the null allele of the HLA-DRB4 gene is expressed but it is an aberrant protein 95. An exception has been reported as an unexpected expression of HLA-DR53 in a DR7 (Dw11) : DQ9 - positive leukaemia patient 96. A difference between HLA-B and -DR supertypes is that not all DR alleles are associated with a supertype. These are HLA-DR1, -DR8 and -DR10. Thus, no supertypical gene is present on these haplotypes.

 

An interesting group of MHC haplotypes is the ancestral or extended haplotypes (also called supratypes). These are specific HLA-B, -DR, BF, C2, C4A and C4B combinations in significant linkage disequilibrium in chromosomes of unrelated individuals. They extend from HLA-B to DR and have been conserved en bloc 97-101. In some Caucasian populations, the extended haplotypes constitute 25-30% of all MHC haplotypes and together with recombinants between any two of them, they account for almost 75% of unselected haplotypes 97;98;100. Particular extended haplotypes are identical by descent. The evidence for this is that in one study, 22 of 26 unrelated extended-haplotype-matched subjects had similar mixed lymphocyte reactivity to HLA-identical siblings 99. Matching for extended haplotypes significantly improves survival in kidney transplantation 102. In Caucasians, there are 10 to 12 common extended haplotypes that show significant linkage disequilibrium. They are relatively population-specific 101;102 and are believed to represent the original MHC haplotypes of our ancestors which are still segregating unchanged. They are easily recognized from their characteristic class III polymorphisms called complotypes 100;102-104. Disease associations with extended haplotypes are generally stronger than allelic associations 100. The best examples of extended haplotype associations are those with rheumatoid arthritis 105, multiple sclerosis 106, insulin-dependent diabetes mellitus 100;107;108, and systemic lupus erythematosis 109.

 

Polymorphism

One of the main characteristics of the MHC is its extreme polymorphism. Among the expressed loci, the MHC has the greatest degree of polymorphism in the human genome. The numbers of alleles recognized at the classical loci by December 1998 are presented in Table 1 (for the latest number of alleles, follow the link at the end).

 

Table 1. Number of alleles at the classical HLA loci

Locus

DNA-level Alleles

Serological Equivalents

HLA-A

119

40

HLA-B

245

88

HLA-C

74

9

HLA-DRB1

201

80

HLA-DQB1

39

7

HLA-DPB1

84

(-)

Data from Refs 59,91,110

 

This is at such a degree that it is theoretically possible for each human to possess a different set of MHC alleles. This feature of the MHC is shared by other compatibility systems in different taxa (such as the fungal mating types, invertebrate allorecognition system and plant self-incompatibility system). It is, however, important to recognise that within the allelic polymorphism at the DNA level which seems endless, there are ancient lineages which predate speciation and maintain themselves in closely related species. This is the basis of the trans-species polymorphism theory proposed by Jan Klein and has found widespread support 111. Allelic lineages may be shared by related species, such as human and apes 112;113 or even human and mice 114, having been present in their common ancestor. However, when primate and human HLA alleles are compared, there is no identical (private) class I allele in great apes and humans despite the similarities in polymorphic motifs 113;115-120. The only similarity is that the human class I supertypes Bw4 and Bw6 are cross-reactive with chimpanzees and gorillas (and even with rhesus monkeys) 113;117;121-123. This conservation throughout hominoid evolution is attributed to the functional importance of these two epitopes in CTL immunomodulation 89 and NK cell function 124-126.

 

Similarly, in the HLA-DRB loci, while no private specificity has an equivalent in another species, the DRB3 / 4 / 5 loci seem to have remained as they are in all primates or even in rhesus monkeys 127;128. Interestingly, these loci encode the class II supertypes HLA-DR52, -53, and -51. The most ancient polymorphic class II locus appears to be HLA-DQA1 129;130. The polymorphism of this locus also correlates to the MHC class II supertypical groupings 131;132.

 

The haplotypical structure, phylogenetic analysis and sequence comparisons agree on the presence of five major haplotypical groups in the HLA class II region 59;128;133-135. These are HLA-DR1, HLA-DR51, HLA-DR52, HLA-DR8, and HLA-DR53. It appears that the oldest lineages are HLA-DRB1*04 (represented by the exon 2 motif 9 EQVKH 13) and HLA-DRB1*03 (the motif 9 EYSTS 13) 136. The analysis of intron sequences suggest that HLA-DRB1*03 first diverged from HLA-DRB1*04 more than 85 million years ago and later gave rise to HLA-DRB1*15 (Refs 133,137,138).

 

MHC class I and class II supertypes are biologically as functional as private specificities. MHC restriction of peptide presentation has been shown for HLA-Bw4 and -Bw6 139;140, HLA-DR52 141;142 and HLA-DR53 143-147. HLA-Bw4 appears to be more immunogenic than HLA-Bw6 judged by the antibody response in the case of mismatching in transplantation 148. Similar to the cross-reactive HLA-B supertypes between humans and chimpanzees 117, HLA-DR supertypes are cross-reactive with chimpanzee 142 and even with mouse class II supertypes 149. Particularly striking is the cross-reactions between HLA-DR53 and H-2Ek 145;150. Furthermore, HLA-DR53 has its own peptide binding motif 151;152. The most abundant peptide eluted from the DR53 molecule is derived from an intracellular protein, calreticulin, which is involved in MHC class I biosynthesis, heat shock response and tumour rejection 69;153-155.

 

HLA Biosynthesis (Animations)    MHC Animations (Serotec)    KEGG Antigen Processing and Presentation Pathways

See also Immunology in a Nutshell PowerPoint Presentation

 

Thomson Lab    LGD @ NCI    Trowsdale Lab    Parham Lab    Kulski Lab 

Cytokine Gene Polymorphisms    Cytokine Reference    Cytokine Immunogenetics Group (Gallagher)    HUMIGEN

 

References

1. Trowsdale J. "Both man & bird & beast": comparative organization of MHC genes [Review]. Immunogenetics 1995; 41: 1-17.

2. Gruen JR, Weissman SM. Evolving views of the major histocompatibility complex. Blood 1997; 90: 4252-4265.

3. Gorer PA. The genetic and antigenic basis of tumour transplantation. Journal of Pathology & Bacteriology 1937; 44: 691-697.

4. Gorer PA, Lyman S, Snell GD. Studies on the genetic and antigenic basis of tumour transplantation. Linkage between a histocompatibility gene and "fused" in mice. Proceedings of the Royal Society of London - Series B: Biological Sciences 1948; 135: 499-505.

5. Dausset J. Iso-leuco-anticorps. Acta Haematologica 1959; 20: 156-166 (see Commentary).

6. van Rood, J. J. Leukocyte Grouping: A Method and Its Application. PhD dissertation, 1962. University of Leiden, Netherlands.

7. van Rood JJ, van Leeuwen A. Leukocyte grouping. A method and its application. Journal of Clinical Investigation 1963; 42: 1382-1390.

8. Bodmer WF. Evolutionary significance of the HLA system [Review]. Nature 1972; 237: 139-145.

9. Meruelo D, Edidin M . The biological function of the major histocompatibility complex: hypotheses [Review]. Contemporary Topics in Immunobiology 1980; 9: 231-253.

10. Jonker M, Balner H . The major histocompatibility complex: a key to a better understanding of evolution. Transplantation Proceedings 1980; 12: 575-581.

11. Dausset J. The major histocompatibility complex in man. Science 1981; 213: 1469-1474.

12. Snell GD. Studies in histocompatibility. Science 1981; 213: 172-178.

13. Benacerraf B, McDevitt HO. Histocompatibility-linked immune response genes. Science 1972; 175: 273-279.

14. Zinkernagel RM, Doherty PC. Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 1974; 248: 701-702.

15. Benacerraf B. Role of MHC gene products in immune regulation. Science 1981; 212: 1229-1238.

16. Doherty PC, Zinkernagel RM. A biological role for the major histocompatibility antigens. Lancet 1975; 1: 1406-1409.

17. Zinkernagel RM. Cellular immune recognition and the biological role of major transplantation antigens. Scandinavian Journal of Immunology 1997; 46: 421-436.

18. Bonner JJ. Major histocompatibility complex influences reproductive efficiency: evolutionary implications. Journal of Craniofacial Genetics and Developmental Biology 1986; Suppl 2: 5-11.

19. Lerner SP, Finch CE. The major histocompatibility complex and reproductive functions [Review]. Endocrine Reviews 1991; 12: 78-90.

20. Powis SH, Geraghty DE. What is the MHC? Immunology Today 1995; 16: 466-468.

21. Zavazava N, Eggert F. MHC and behavior. Immunology Today 1997; 18: 8-10.

22. Penn DJ, Potts WK. The evolution of mating preferences and major histocompatibility complex genes. American Naturalist 1999; 153: 145-164.

23. Svejgaard A, Ryder LP. Interaction of HLA molecules with non-immunological ligands as an explanation of HLA and disease association. Lancet 1976; ii: 547-549.

24. Edidin M. Function by association? MHC antigens and membrane receptor complexes. Immunology Today 1988; 9: 218-219.

25. Schreiber AB, Schlessinger J, Edidin M. Interaction between major histocompatibility complex antigens and epidermal growth factor receptors on human cells. Journal of Cell Biology 1984; 98: 725-731.

26. Phillips ML, Moule ML, Delovitch TL, Yip CC. Class I histocompatibility antigens and insulin receptors: evidence for interactions. Proceedings of the National Academy of Sciences USA 1986; 83: 3474-3478.

27. Solano AR, Cremaschi G, Sbnchez ML, Borda E, Sterin-Borda L, Podestb EJ. Molecular and biological interaction between major histocompatibility complex class I antigens and luteinizing hormone receptors or beta-adrenergic receptors triggers cellular response in mice. Proceedings of the National Academy of Sciences USA 1988; 85: 5087-5091.

28. Verland S, Simonsen M, Gammeltoft S, Allen H, Flavell RA, Olsson L. Specific molecular interaction between the insulin receptor and a D product of MHC class I. Journal of Immunology 1989; 143: 945-951.

29. Schafer PH, Pierce SK, Jardetzky TS. The structure of MHC class II: a role for dimer of dimers. Seminars in Immunology 1995; 7: 389-398.

30. Weiss MS, Anderson DH, Raffioni S, et al. A cooperative model for receptor recognition and cell adhesion: evidence from the molecular packing in the 1.6-A crystal structure of the pheromone Er-1 from the ciliated protozoan Euplotes raikovi. Proceedings of the National Academy of Sciences USA 1995; 92: 10172-10176.

31. Vallesi A, Giuli G , Bradshaw RA, Luporini P. Autocrine mitogenic activity of pheromones produced by the protozoan ciliate Euplotes raikovi. Nature 1995; 376: 522-524.

32. Metzenberg RL. The role of similarity and difference in fungal mating. Genetics 1990; 125: 457-462.

33. Hiscock SJ, Kues U , Dickinson HG. Molecular mechanisms of self-incompatibility in flowering plants and fungi - different means to the same end. Trends in Cell Biology 1996; 6: 421-428.

34. Wendland J, Vaillancourt LJ, Hegner J, et al. The mating-type locus B alpha 1 of Schizophyllum commune contains a pheromone receptor gene and putative pheromone genes. EMBO Journal 1995; 14: 5271-5278.

35. Kothe E. Tetrapolar fungal mating types: sexes by the thousands. FEMS Microbiological Reviews 1996; 18: 65-87.

36. Haring V, Gray JE, McClure BA, Anderson MA, Clarke AE. Self-incompatibility: a self-recognition system in plants [Review]. Science 1990; 250: 937-941.

37. Scofield VL, Schlumpberger JM, West LA, Weissman IL. Protochordate allorecognition is controlled by a MHC-like gene system. Nature 1982; 295: 499-502.

38. Grosberg RK. The evolution of allorecognition specificity in clonal invertebrates. Quarterly Review of Biology 1988; 63: 377-412.

39. Weissman IL, Saito Y, Rinkevich B. Allorecognition histocompatibility in a protochordate species: is the relationship to MHC somatic or structural? Immunological Reviews 1990; 113: 227-241.

40. Williams JR, Lenington S. Factors modulating preferences of female house mice for males differing in t-complex genotype: role of t-complex genotype, genetic background, and estrous condition of females. Behavior Genetics 1993; 23: 51-58.

41. Potts WK, Manning CJ, Wakeland EK. Mating patterns in seminatural populations of mice influenced by MHC genotype. Nature 1991; 352: 619-621.