Major Histocompatibility Complex (MHC)

Dustin J Penn, University of Utah, Salt Lake City, Utah, USA

The major histocompatibility complex (MHC) contains the most diverse genes known in vertebrates, the class I and II loci. These highly polymorphic genes encode cell surface receptors that play a central role in controlling immunological self/nonself recognition, and subsequently tissue rejection, autoimmunity, and immune responses to infectious diseases. The polymorphisms of MHC genes has been maintained by natural selection over long periods of evolutionary time.

Introductory article
Article Contents
. Introduction . The History of MHC Research . MHC Genetic Architecture . Immunological Functions . Evolutionary Origins of MHC Genes . MHC Genes Influence Disease Resistance . MHC Polymorphisms . Darwinian Selection and MHC Polymorphisms . MHC Monomorphisms

Introduction
For the immune system to attack an invading parasite, it must be able to discriminate between the bodys own tissue (self) and foreign tissue (nonself). Immunological self/ nonself recognition is partly controlled by a set of genes in the major histocompatibility complex or MHC. The MHC is a large region of DNA, spanning about four million base pairs in humans, or about 0.1% of the human genome, and contains over 200 coding loci. Immunological self/nonself recognition is controlled by MHC class I and II MHC genes, which are the most polymorphic loci known in vertebrates. These MHC genes encode cell surface proteins (class I and II MHC molecules) that present small peptide antigens to T cells. Through antigen presentation, MHC molecules control all specic immunological responses, both cell- and antibody-mediated. Because MHC genes play such a central role in the immune system, they inuence resistance and susceptibility to infectious and autoimmune diseases. MHC genes were initially discovered because they also play a central role in controlling tissue transplant rejection. Today, MHC genes are the most intensively studied of all genetic systems because they inuence many important traits, including resistance to infectious diseases, autoimmunity, compatibility of tissue transplants, spontaneous abortion, odour and mating preferences. histocompatibility (histo means tissue) is genetically controlled. In 1933, Haldane suggested that an immune response leading to rejection of transplanted tumours was directed against normal cellular antigens belonging to a dierent strain rather than against some antigens unique to tumours. In the late 1930s, Gorer discovered four blood group antigens (he called antigen I, II, III and IV), and found that the growth or rejection of a tumour was associated with the expression of the particular antigen. During the 1940s, Medawar and his colleagues demonstrated that tissue graft rejection in rabbits was indeed due to an immune response attacking the foreign tissue graft. A few years later, Snell and his colleagues systematically bred dierent strains of laboratory mice that were virtually genetically identical, except at a single genetic region (congenic strains) that controlled the rejection of foreign tissue. Snell called the genes controlling tissue rejection histocompatibility (H) genes. Subsequent work has shown that there are several histocompatibility genes closely linked on the same chromosome that control tissue rejection, and so this region is now known as the major histocompatibility complex or MHC. (Unfortunately, the mouse MHC is sometimes known as H-2, which originated by combining Snells H-genes with Gorers antigen II, and the human MHC is also called HLA or human leucocyte antigen.) The actual immunological functions of the MHC remained a mystery until the 1970s. During this time, it was found that immune responses were controlled by interactions between MHC molecules and T lymphocytes. In 1975, Zinkernagel and Doherty found that T-cell responses were restricted not only to the antigen, but also to MHC molecules. They suggested that MHC heterozygotes would have greater T-cell responses than homozygotes, and this could potentially explain both the evolutionary advantage for gene duplication and diversity of MHC genes. Later, MHC molecules were found to be involved in antibody-mediated, as well as cellular
1

The History of MHC Research

MHC genes were identied through research on tissue rejection using domestic strains of mice and rabbits. In 1916, Little and Tyzzer performed tumour transplants between dierent strains of mice, and found that tumours could be transplanted among some strains of mice, but rejected among others. In 1927, Bover found that tissue transplants were not rejected if the donor and recipients were identical twins. These observations indicated that tissue compatibility between a donor and recipient or

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

responses. During the 1980s and 1990s, the structure of MHC molecules was revealed by X-ray crystallography studies.

MHC Genetic Architecture

The MHC is a large chromosomal region with over 200 coding loci (Figure 1). The MHC region contains class I and II genes, which control all specic immune responses, but it also contains many other genes that inuence growth, development, reproduction, odour and olfaction. Among the MHC loci that control the immune system are class I and II MHC loci (classical MHC genes), which are the most highly polymorphic genes known among vertebrates. Class I and II MHC loci are closely linked within the MHC cluster in many species, such as mice and humans. In humans there are six class I loci (e.g. A, B and C) and eight class II loci (e.g. DP, DQ and DR), and in mice there are three class I (K, D, and L) and 2 class II (A and E) loci. Each class II locus consists of multiple coding genes. In most species, the polymorphism of MHC varies from one to over 100 alleles per locus. The close linkage of MHC loci on the same chromosome means that each individual inherits their particular combination of MHC alleles as a single unit or haplotype (closely linked genes on a chromosome have a low chance of recombination separating them). The close linkage of MHC loci suggests that they have a common ancestor, and that the close physical linkage on the same chromosome has been maintained by natural selection. MHC loci are not always found in close linkage. In zebrash and African clawed frogs, for example, the class I and II loci are located on dierent chromosomes.

Their function is to present antigens that originate from inside the cell (endogenous antigens) to cytotoxic T lymphocytes (CTLs). Class II molecules are mainly found on specialized antigen-presenting cells of the immune system, such as dendritic cells, macrophages and B cells. Their job is to present antigens that enter the cell by endocytosis (exogenous antigens) to helper T lymphocytes. Class I molecules Class I molecules present small peptides to CTLs, eectively advertising the contents of the bodys cells to the immune system. These peptides are usually self peptides; however, if a cell becomes infected with a virus or other type of intracellular parasite (Figure 2a), then the infected cell presents foreign peptides from the invading parasite. CTLs probe the surface of cells, searching for evidence of an infection. Each CTL has a unique, highly specic T-cell receptor (TCR) on its surface that binds to MHCantigen complexes. CTLs also have CD8 adhesion molecules on their surface that stabilize the interaction between T cells and the presenting cell by binding to class I MHC molecules. If recognized as foreign, the CTL becomes activated, and once activated, the T cell proliferates and all of the clones search out and eliminate other infected cells presenting the foreign peptide. Class II molecules Class II molecules are expressed on specialized antigenpresenting cells (APCs), such as macrophages. Their function is to present peptides to helper T cells. Helper T cells play several crucial roles, including activating macrophages and B lymphocytes. Macrophages search out and engulf extracellular viruses, bacteria and fragments from other parasitic invaders (Figure 2b). They digest their prey and use class II molecules to present peptide samples to helper T cells. Helper T cells have CD4 molecules on their surface to stabilize their interaction during presentation. If the peptide is recognized as foreign, then the helper T cell is activated to search out and to activate other macrophages. Class II molecules are also expressed on B cells (Figure 2c). When an immunoglobulin on the surface of a B cell binds to a foreign antigen, the B cell undergoes proliferation and
Complement proteins

Immunological Functions
Antigen presentation
MHC molecules come in two dierent forms, class I and II molecules. Each of the nucleated cells of the body are covered with as many as 100 000 class I MHC molecules.

DP Human

DQ

DR

K Mouse

MHC class I

II

Figure 1 The genetic architecture of the major histocompatibility complex (MHC) in humans and house mice.

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

(a) (5) Activation TCR and proliferation (4) Recognition (3) Presentation Foreign peptide (2) Antigen binding (1) Pathogen infects cell Infected cell Cytotoxic T cell (6) Induction of cell death

(b) Helper T cell (4) Recognition (3) Presentation Foreign peptide (1) Phagocytosis of pathogen (5) Activation of macrophage

(c) Helper T cell (4) Recognition (3) Presentation (5) Activation of B-cell (6) Antibody secretion Class II MHC molecule (2) Antigen binding B cell

Class I MHC molecule

Class II MHC molecule (2) Antigen binding Macrophage

(1) Immunoglobulin binds and endocytoses pathogen

Cytotoxic T cell

Helper T cell

Helper T cell

T-cell receptor CD8 protein Antigenic peptide MHC class I CD4 protein

T-cell receptor Antigenic peptide MHC class II CD4

Infected cell

Macrophage

B cell (Antigen-presenting cell)

Figure 2 (a) Cytotoxic T cells recognize foreign peptides presented by class I MHC molecules on infected cells. Helper T cells recognize foreign peptides presented by class II MHC molecules presented by (b) macrophages and (c) B lymphocytes. TCR, T-cell receptor.

the clones become antibody-secreting plasma cells. The antibodies bind and tag foreign invaders so that the immune system can eciently eliminate them. However, before a B cell can become activated, it must present the antigen to helper T cells to ensure that the antigen is foreign.

Immunological self/nonself recognition

To attack an invading pathogen, the immune system must be able to distinguish self from foreign tissues. MHC molecules present both self and nonself peptides. T cells do not actually distinguish between self and nonself peptides. It is simply that T cells that recognize self peptides are eliminated or inactivated during the development of the

immune system. Before birth, immature, undierentiated T lymphocytes arise in the bone marrow and then migrate to the thymus where they mature and undergo thymic selection (hence the term T cell). Within the thymus, a massive proliferation of T cells occurs and an enormous diversity of T cells is generated by random rearrangements of TCR genes. Over 1011 types of T cells are generated, but since they are randomly generated, many are likely to bind to self antigens, and therefore, 95% are eliminated or inactivated. The only T cells to mature and enter circulation are those that recognize self peptideMHC complexes (positive selection) below a threshold avidity. Those that do not bind above the threshold are killed or inactivated (negative selection). Thus, through thymic selection MHC genes inuence the development of an individuals particular T-cell repertoire.
3

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

(a) Polymorphic domains Antigen binding site

(b)

Antigen binding site 2 microglobulin 3

Figure 3 (a) A class I MHC molecule and (b) the antigen-binging site of a class I MHC molecule.

Structure of MHC molecules The structure of MHC molecules has been revealed by Xray crystallography. Class I molecules are composed of two separate polypeptide chains, a heavy a chain and a light b chain (Figure 3a). The a chain is encoded by a class I gene within the MHC, but the b chain is encoded by a gene (b2microglobulin) on another chromosome. The b chain plays a role in stabilizing the molecule; therefore, mice genetically engineered (through targeted gene disruption) that have no b2-microglobulin gene do not express any class I MHC molecules. Class II molecules are very similar to class I except that the a and b chains are both encoded in the MHC. The lower part of an MHC molecule is anchored in the cell membrane, whereas the upper portion contains a small groove, which is the antigen-binding site (Figure 3b). Here small peptides are bound and presented to T cells. In the next section, we will see that most of the diversity of MHC alleles occurs in the antigen-binding site (ABS) and that the ABS is under natural selection.

Class I Class II Class II 2-microglobulin

Immunoglobulins
Figure 4 MHC genes belong to the immunoglobulin multigene superfamily, which includes T-cell receptors and antibodies, suggesting that these molecules evolved from a common ancestor.

multigene families, including genes for antibodies, T-cell receptors and class I and II MHC molecules.

Evolutionary Origins of MHC Genes

The similar structure and function of class I and II molecules indicate that these proteins evolved from a common ancestor (Figure 4). Indeed, MHC molecules belong to the immunoglobulin multigene superfamily, which all apparently evolved from a common ancestor. The superfamily has diversied into single-gene and
4

MHC Genes Influence Disease Resistance

Genetic variation at the MHC inuences resistance and susceptibility to a wide variety of infectious diseases, including viruses, bacteria, yeasts, protozoans and intestinal worms. For example, certain MHC alleles in a feral population of sheep in Scotland are associated with increased resistance to an intestinal parasitic nematode,

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

and this resistance to the parasite is associated with increased survivorship. Humans also show many MHC disease associations, such as with malaria, hepatitis, and human immunodeciency virus (HIV). Interestingly, most of the diseases associated with certain MHC alleles are autoimmune diseases, such as lupus and arthritis, rather than infectious diseases. Most experiments on MHC-dependent resistance to infectious diseases have been conducted using domestic mice and chickens. Following Snell, MHC researchers have bred many mouse strains that are virtually genetically identical except at particular MHC loci. These MHCcongenic mouse strains have made it possible to isolate the eects of MHC genes on disease resistance, and to rule out the eects of other genes. More recently, mice have been genetically engineered that lack all expression of either class I or II MHC genes (MHC knockout or transgenic mice). By experimentally disrupting one of the genes encoding one of the subunits of a class I molecule (b2microglobulin), class I molecules are not expressed on the cell surface (these mice also have reduced numbers of CD8 1 class I MHC-restricted cytotoxic T lymphocytes). Experimentally knocking out the expression of MHC genes, especially class II, reduces the ability of mice to clear many types of infections, increases the severity of the disease, and increases the mortality of infected mice MHC molecules encoded by dierent alleles bind and present dierent sets of peptides to the immune system. The dierential antigen binding of MHC alleles should help to explain both the immunological basis of MHC-dependent resistance and the evolution of MHC polymorphisms.

Table 1 Evidence that MHC polymorphism is maintained by natural selection 1. 2. 3. 4. 5. 6. 7. High number of alleles Uniform allelic frequencies Deciencies of homozygotes Linkage disequilibrium among loci High nonsynonymous substitution rates at the antigenbinding site codons Ancient allelic lineages (transspecies evolution) Disassortative mating preferences (sexual selection)

Adapted from Apanius V, Penn D, Slev P, Ru LR and Potts WK (1997) The nature of selection on the major histocompatibility complex. Critical Reviews in Immunology 17: 179224.

ABS 25 Nonsynonymous minus synonymous nucleotide substitutions (%) 20 15 10 5 0 5 Class I Class II Class II non ABS

MHC Polymorphisms
Class I and II MHC genes are the most highly polymorphic genes known in vertebrates. Most other genes that are considered polymorphic have only a few alleles, whereas MHC loci have as many as 170 alleles per locus in humans and 100 alleles per locus in house mice. The enormous polymorphism of MHC alleles is dicult to explain because natural selection should eliminate all but the most disease-protective allele. For example, a particular human MHC allele confers resistance to malaria in Africa. This process of directional selection should lead to all of the susceptible alleles becoming extinct, leaving only the most resistant allele in the population (xation). Therefore, there must be some other evolutionary pressure that maintains MHC diversity.

Figure 5 In both humans and house mice, the antigen-binding site (ABS) of class I and II MHC molecules (light blue) have a high rate of nonsynonymous versus synonymous nucleotide substitutions, which is the opposite pattern for genes under purifying selection, such as nonantigenbinding sites of MHC molecules (dark blue). Adapted from Potts WK and Wakeland EK (1990) Evolution of diversity at the major histocompatibility complex. Trends in Ecology and Evolution 5: 181 187.

Natural selection maintains MHC polymorphisms

Much evidence indicates that MHC polymorphisms are maintained by natural selection (Table 1). First, the high

number of alleles implies that selection is maintaining MHC polymorphisms because neutral alleles, which are not under selection, tend to go to xation through random genetic drift. Second, MHC alleles often show more uniformity in the proportions of alleles within populations than one would expect by chance alone. Third, populations of humans and mice have fewer MHC homozygotes than would be predicted by random mating expectations, which can only be explained by selection. Fourth, MHC alleles at dierent loci are found associated together in individuals more often than by chance expectations (linkage disequilibrium). Fifth, the ratio of nonsynonymous to synonymous substitutions in the antigen-binding site of MHC genes indicates that selection maintains MHC diversity (Figure 5). If MHC genes were not under selection, then the rate of
5

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

synonymous and nonsynonymous substitutions would be equal throughout the molecule, and if MHC genes were under purifying (diversity-eliminating) selection, then most nonsynonymous mutations would be harmful, and therefore selected against compared to synonymous substitutions. However, the ratio of nonsynonymous to synonymous substitutions is greater than expectations of both neutrality or purifying selection, which indicates that positive Darwinian selection is maintaining the diversity at the antigen-binding site of class I and II MHC alleles. Sixth, the diversity of MHC alleles in some species is ancient, predating the origin of some species, such as humans and chimpanzee evolutionary divergence (transspecies polymorphisms). In other genes, alleles at a particular genetic locus usually dier by a few nucleotide substitutions, but some MHC alleles dier by over 100 substitutions. This indicates that MHC alleles diverged long ago. The persistence of genetic diversity over long periods of evolutionary time can only be explained by natural selection. In the next section, we will see that infectious diseases may provide the selective force maintaining MHC diversity.

selection within groups. However, if natural selection is acting to reduce MHC diversity within groups, then there will be no diversity left for group selection to favour in the long term. Thus, group selection is an unlikely explanation for MHC diversity.

Overdominant selection
One way that natural selection can maintain MHC diversity over evolutionary time is if heterozygotes are more resistant to infectious diseases than homozygotes (heterozygote advantage or overdominant selection). The evidence for heterozygote advantage is rather mixed however. Doherty and Zinkernagel found that MHCheterozygotes present more antigens to the immune system than homozygotes; yet, the infected heterozygous mice in their experiment all died, apparently from over-responsiveness. It has recently been found that MHC heterozygosity is associated with an ability to clear hepatitis B virus, and slower progression to AIDS and death among HIV infected people. Other observational studies on malaria in humans and parasitic worms in feral sheep, along with experimental infection studies, have failed to nd that heterozygotes are more disease resistant than homozygotes. Furthermore, even if heterozygotes are protected, theoretical models indicate that overdominant selection is insucient to maintain the high diversity of MHC alleles.

Darwinian Selection and MHC Polymorphisms

Explaining how natural selection maintains MHC polymorphisms is sometimes called the holy grail of MHC biology. There have been several hypotheses proposed to explain how MHC polymorphisms are maintained by natural selection.

Frequency-dependent selection
Haldane rst pointed out that rapidly evolving parasites should adapt to their hosts MHC alleles and subsequently provide a selective force favouring rare MHC alleles. As an MHC allele resistant to a particular parasite becomes more common in the host population, this creates increasing selection on the parasite to adapt to the common MHC allele. For example, a new strain of parasite may arise that is able to evade or abolish presentation of an MHC molecule. The evolution of such a new strain of parasite will cause the common MHC allele to lose its advantage to rare alleles. This hostparasite coevolutionary arms race is called frequency-dependent selection because an alleles advantage depends on its frequency or how common it is in the host population. This type of selection can potentially maintain the diversity of MHC alleles indenitely, and there is some indirect evidence to support this idea.

Group selection
It is often assumed that MHC polymorphisms are maintained by natural selection because they enable a population or species to recognize a wider range of infectious diseases. Haldane rst suggested that it is an advantage to the species to be biochemically diverse, and even to be mutable as regards genes concerned in disease resistance. Doherty and Zinkernagal also suggested that the existence of multiple polymorphisms thus minimizes the risk of there being a general unresponsiveness throughout the population. A population that is polymorphic at the MHC may be less likely to become extinct than a monomorphic one. There is evidence that genetically diverse crops are more resistant to being decimated by infectious diseases than monocultures. The problem is that the dierential survival of groups over the long term (group selection) is a relatively weak force compared with the dierential survival of individuals within a group in the short term (individual selection). The group selection hypothesis requires that selection between groups favouring diversity is greater than diversity-reducing (purifying)
6

Sexual selection
Another way that MHC genetic diversity could be maintained is through nonrandom mating (sexual selection). Several studies have found that house mice and humans prefer mates carrying dissimilar MHC genes. This mating preference apparently involves olfactory cues, since

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net

Major Histocompatibility Complex (MHC)

many studies have shown that MHC genes inuence individual odours. Interestingly, mice learn the odours of their parents and avoid mating with mice having their parents particular MHC-determined odour-type (chemosensory imprinting). MHC-dependent mating preferences may function to increase the disease resistance of an individuals progeny by producing heterozygous ospring. MHC-dependent mating preferences also may function to recognize kin and avoid inbreeding, since MHC-similar individuals are likely to be close kin.

apparently due to inbreeding. Low MHC diversity is not necessarily associated with low genetic variation overall, however. European beavers (Castor ber), Sei whales (Balaenoptera borealis), Fin whales (Balaenoptera physalus), and southern elephant seals (Mirounga leonina) all have low MHC variation, but normal genetic variation at other loci. This indicates that some selective factor, such as infectious diseases, has reduced the MHC diversity of these species. OBrien and others have suggested that reduced MHC diversity may leave an endangered population or species susceptible to epidemic diseases in future, but this idea has not been tested.

MHC Monomorphisms
Not all species show high MHC polymorphisms; many show limited or no MHC diversity. For example, cheetahs (Aconyx jubatus) readily accept skin grafts from unrelated individuals, indicating that these African cats have no MHC diversity. Asiatic lions (Panthera leo persica) also show no MHC variation. Since these lions also show low genetic diversity overall, their low MHC diversity has been attributed to having undergone a population or genetic bottleneck. Similarly, European and American moose (Alces alces) exhibit low diversity class II MHC loci, which has been attributed to a population bottleneck that occurred before the divergence of Old and New World subspecies. Naked mole rats (Heterocephalus glaber) show low MHC and overall genetic diversity in general,

Further Reading
Beauchamp GK, Yamazaki K and Boyse EA (1985) The chemosensory recognition of genetic individuality. Scientic American 253: 8692. Edwards SV and Hedrick PW (1998) Evolution and ecology of MHC molecules: from genomics to sexual selection. Trends in Ecology and Evolution 13: 305311. Janeway CA, Jr. (1993) How the immune system recognizes invaders. Scientic American 269: 7279. Klein J (1986) Natural History of the Histocompatibility Complex. New York: Wiley. Potts W and Wakeland E (1990) Evolution of diversity at the major histocompatibility complex. Trends in Ecology and Evolution 5: 181186. von Boehmer H and Kisielow P (1990) How the immune system learns about self. Scientic American 265: 7481.

ENCYCLOPEDIA OF LIFE SCIENCES / & 2002 Macmillan Publishers Ltd, Nature Publishing Group / www.els.net