Our immune systems have a tough task. They are engaged in a relentless battle to protect us from potentially harmful microbes and other parasites. Not only must we be protected against a bewildering array of pathogens, but, in addition, new forms of a pathogen can rapidly develop by mutation to provide new challenges to which human populations must continuously adapt. Different types of proteins work in the immune system, some as sentinels dedicated to recognizing antigens on the surface of invading microbes, some as signaling molecules that allow different immune system cells to communicate and co-operate, and some as executor proteins involved in mechanisms to deal with the invaders.
4 we covered the detail about the biology of the vertebrate immune system: the cells involved and some of the principal mechanisms used in immune system host defense. Here, our priority is to explain the extraordinary sequence diversity of some immune system proteins that act as sentinels by seeking out and binding foreign antigen. Although we will revisit some of the concepts involved to set the scene, interested readers might wish to reacquaint themselves with some of the wider background given in Section 3.4.
Recall that the vertebrate immune system has two arms. The innate immune system provides a first line of defense: it uses different types of barrier to reduce access by harmful pathogens to body tissues, and develops a rapid response based on general pattern recognition of invading microbes. The more sophisticated adaptive immune system is mobilized by components of the innate immune system when the latter fails to provide adequate protection against an invading pathogen. It has exquisite specificity, being able to discriminate between tiny differences in molecular structure of invading cells and viruses that pose a threat to the body. At the forefront of the adaptive immune system’s ability to recognize foreign antigen are four types of protein that show extraordinary sequence variation, as described below and in Figure 1.

Fig1. Extreme variation exhibited by four types of protein that are used to recognize foreign antigens. Immunoglobulins (Igs), T-cell receptors, and MHC (major histocompatibility complex) proteins are heterodimers with similar structures—globular domains (maintained by intrachain disulfide bridges) and N-terminal variable domain that bind foreign antigens (but they otherwise have conserved sequence, mostly constant domains in Igs). They are cell surface receptors (except that Igs in activated B cells become secreted antibodies). Only a few human genes encode each protein chain, but nevertheless many different proteins are made because of special genetic mechanisms in B and T lymphocytes and because of selection for heterozygosity of HLA proteins (see main text). β2-microglobulin (β2 M), the non-polymorphic light chain of class I HLA proteins, is encoded by a gene outside the major histocompatibility complex. * See Box 11.4 Figure 1 for genes encoding classic HLA proteins. ** The recombination mechanisms can generate a theoretical diversity of at least 5 × 1013 naive B-cell receptors and 1018 α:β T-cell receptors, but only a subset of this diversity is expected to be physiologically present in the actual repertoire of an individual.
• Immunoglobulins. Expressed on the surface of B cells in the bone marrow or secreted as soluble immunoglobulins (antibodies) by activated B cells, their main task is to recognize and bind specific foreign antigens. They can bind and neutralize toxins released by microbes, inhibit viruses from infecting host cells, and activate complement-mediated lysis of bacteria and phagocytosis.
• T-cell receptors. Displayed on the surface of T cells, they work in cell-mediated immunity, along with proteins encoded by the major histocompatibility complex (MHC; known in humans as the HLA complex).
• Class I MHC (HLA) proteins. They are expressed on the surface of almost all nucleated cells and enable cytotoxic T lymphocytes to recognize and kill host cells that are infected by a virus or other intracellular pathogen.
• Class II MHC (HLA) proteins. They are displayed on the surface of very few types of cell, notably immune system cells that present foreign antigen to be recognized by helper T lymphocytes.
The mechanisms that produce the extraordinary variety of immunoglobulins and T-cell receptors are different from those responsible for producing the different types of MHC (HLA) proteins. The extensive variety of MHC proteins is apparent at a population level. MHC proteins are highly polymorphic but a single person has a limited number of them, having at most two alleles at each of a small number of polymorphic MHC loci. By contrast, the huge sequence variation of immunoglobulins and T-cell receptors is generated at the level of the individual, as a result of post-zygotic DNA changes in B cells and T cells.
Somatic mechanisms allow cell-specific production of immunoglobulins and T-cell receptors
Human sperm cells and egg cells each have just three immunoglobulin genes—IGH, IGK, and IGL—and four T-cell receptor genes—TRA, TRB, TRD, and TRG. Nevertheless, each person can make huge numbers of different immunoglobulin proteins, and huge numbers of different T-cell receptors. That happens because immunoglobulin genes in maturing B cells, and T-cell receptor genes in maturing T cells, are programmed to undergo extraordinary DNA changes. There is a high degree of flexibility in how the changes occur, and they occur in a cell-specific way: in each person the precise DNA changes vary from one maturing B cell to the next, and from one maturing T cell to the next. Different mechanisms are responsible as described below.
Combinatorial diversity via somatic recombination
In germ-line DNA, each immunoglobulin and T-cell receptor gene is made up of a series of repeated gene segments that specify discrete segments of the protein. Take the exam ple of an immunoglobulin heavy chain. The constant domain defines the functional class of immunoglobulin (IgA, IgD, IgE, IgG, or IgM) and is encoded by repeated C gene segments (each with coding sequences split by introns). The variable domain is encoded by three types of repeated gene segment containing coding DNA only: a V (variable region) gene segment (encoding the majority of the variable domain, starting from the N-terminus) plus a D (diversity region) gene segment and a J (joining region) gene segment (Figure 2A). A similar arrangement applies to light-chain domains, except that the variable domain is encoded by two repeated gene segments only: V gene segments and J gene segments.

Fig2. Variation in the structure of the immunoglobulin heavy-chain gene in germ-line DNA and maturing B lymphocytes. Somatic DNA rearrangements occur in immunoglobulin genes and T-cell receptor genes in B and T cells, respectively, resulting in a change in gene organization, as in the human IGH gene shown here. (A) In germ-line DNA, and in all non-B cells, the IGH gene has multiple but slightly different repeats for each of four types of gene segment and is not expressed. V (variable region), D (diversity region), and J (joining region) gene segments have sequences that together can encode the variable domain of the immunoglobulin heavy chain; C (constant domain) gene segments have sequences that can encode the constant domain of the protein (note: individual C domain gene segments have multiple exons that are not shown here for clarity). (B) Somatic recombination in maturing B cells produces a change in gene organization that allows gene expression and the production of a single type of immunoglobulin heavy chain. As a result of recombination events, a single V gene segment is fused to individual D and J gene segments to produce a functional VDJ unit that activates transcription and splicing. The resulting VDJC mRNA is translated to give the heavy chain, with the VDJ sequence specifying the variable domain of the protein, and the C sequence specifying the constant domain. For the sake of clarity, the other gene segments are not shown.
Cells other than B lymphocytes have the same immunoglobulin gene organization as in germ-line DNA, and the immunoglobulin genes are not expressed. In maturing B cells, however, a single type of immunoglobulin protein is made. In the case of the immunoglobulin (Ig) heavy chain, somatic recombination brings about joining of one V gene segment to individual D and J gene segments to form a functional VDJ unit, essentially a novel exon, that activates transcription and RNA splicing. As a result, a VDJ exon is joined to a C sequence at the RNA level; the VDJ component of the mRNA specifies the variable domain of the protein and the C component specifies the constant domain (Figure 2B).
In much the same way, an immunoglobulin light chain is made after somatic recombination forms a novel functional VJ exon from the light-chain gene segments. T-cell receptor chains are formed in maturing T cells after individual gene segments are brought together by somatic recombination to form a functional VDJ exon (in the case of a T-cell receptor β chain) or a VJ exon (for a T-cell receptor α chain).
The key point about the somatic recombination events used to produce the functional VDJ or VJ exons is that they occur in a cell-specific way. From the two or three clusters of repeated segments encoding sequences for the variable domain, one segment only from each cluster is selected to be included to make a functional VDJ or VJ exon, and that choice is made randomly in each B or T cell in a person. We use the example of making an immunoglobulin heavy chain to illustrate this point in detail in Figure 3.

Fig3. Cell-specific immunoglobulin chains are produced by somatic recombination in B cells. Here, as in Figure 11.15, we use the example of the human IHG gene. The fusion of single V, D, and J gene segments to make an active VDJ unit is cell-specific, and the chosen single gene segments are shown in dark coloring for emphasis. In this B cell we imagine that the second of the V gene segments fuses with the third D gene segment and the second J gene segment. This comes about by two sequential somatic recombinations: first, D–J joining; then addition of the chosen V gene segment. The mature, functional VDJ coding sequence unit is effectively a large novel exon that will encode the variable domain of the protein. In this example, the successful combination is V2 D3 J2 , but the choice of combinations is cell-specific: in a neighboring maturing B cell it could be V24 D19 J5 , for example. Once a functional VDJ exon has been assembled, transcription is initiated starting with this exon and RNA splicing joins the VDJ coding sequence to coding sequences initially in the constant domain (C) gene segment, in this case Cμ and then by alternative splicing to Cδ . Another type of somatic recombination (known as class-switching, but not shown here) can change the position of C gene segments so that other C gene segments can be used to give alternative classes of immunoglobulin (see text).
RNA splicing fuses the VDJ sequence initially to a C sequence from the nearest C gene segment, initially Cμ and then, through alternative splicing, either Cμ or Cδ . The first immunoglobulins to be made by a B cell, therefore, are membrane-bound IgM and then IgD. Subsequently, as B cells are stimulated by foreign antigen and helper T lymphocytes, they secrete IgM antibodies.
Later in the immune response, B cells undergo another type of somatic recombination called class-switching (or isotype-switching) to produce different antibody classes. The recombinations (not shown here) can position an alternative C gene segment to be nearest to the J gene segments: either a Cγ , Cε , or Cα gene segment to produce, respectively, an IgG, IgE, or IgA antibody.
Additional diversity generation The use of somatic recombination alone would allow each person to produce diverse sets of both immunoglobulins (potentially over 2 million types; see Table 1) and T-cell receptors. As listed below, two or three additional mechanisms are responsible for generating sequence diversity; together with VDJ and VJ recombination, they endow each of us with the potential for making huge numbers of different antigen-binding sites, both for immunoglobulins and T-cell receptors. As required, individual B cells and T-cell receptors that successfully recognize foreign antigen are induced to proliferate to make identical clones with the same antigen specificity as the original cell.
• Protein chain combinatorial diversity. Immunoglobulins and T-cell receptors are heterodimers and diversity is compounded by unique combinations of two unique protein chains. Take the example of immunoglobulins. Although a B cell, being diploid, has six immunoglobulin genes, each B cell is said to be monospecific: it makes only one type of antibody. That happens because in each B cell only one of the two IGH alleles is (randomly) selected to make a heavy chain (allelic exclusion), and because only one of the four light-chain genes is ever used to make a light chain (a combination of light-chain exclusion—to select either a kappa or lambda light chain—plus allelic exclusion). That raises the number of different immunoglobulins by a factor of eight (see Table 1).
• Junctional diversity. The somatic recombination mechanisms that bring together different gene segments in immunoglobulin or T-cell receptor genes variably add or subtract nucleotides at the junctions of the selected gene segments.
• Somatic hypermutation. This mechanism applies to immunoglobulins and is used to further increase variability in the variable domain after somatic recombinations have produced functional VDJ or VJ exons. When B cells are stimulated by foreign antigen, an activation-induced cytidine deaminase is produced by the activated B cell that deaminates cytidine to uridine. The uridines are variably repaired so that multiple nucleotides in the variable domain are mutated.

Table1. MILLIONS OF DIFFERENT IMMUNOGLOBULINS CAN BE MADE USING JUST CELL-SPECIFIC SOMATIC RECOMBINATION AND PROTEIN CHAIN COMBINATORIAL DIVERSITY
Functions of major histocompatibility complex (MHC) proteins and the concept of MHC restriction
The HLA complex is the human major histocompatibility complex (MHC). The latter name came from the observation that certain MHC genes are the primary determinants in transplant rejection. That is an artificial situation: the normal function of MHC genes largely consists of helping T cells to identify host cells harboring an intracellular pathogen such as a virus.
Some MHC genes—called classic MHC genes—are extremely polymorphic. They are deployed on the cell surface as heterodimers (see Figure 1), where they serve to bind peptide fragments derived by intracellular degradation of pathogen proteins. After being displayed on the surface of host cells, individual MHC–peptide fragment complexes can be recognized by T cells with a suitable T-cell receptor (antigen presentation). Appropriate immune reactions are then initiated to destroy infected host cells. There are two major classes of classic MHC protein, as detailed below.
Class I MHC
proteins Class I MHC proteins are expressed on almost all nucleated host cells. Their job is to help cytotoxic T cells (CTLs) to recognize and kill host cells that have been infected by a virus or other intracellular pathogen. When intracellular pathogens synthesize protein within host cells, a proportion of the protein molecules get degraded by proteasomes in the cytosol. The resulting peptide fragments are transported into the endoplasmic reticulum (Figure 4A). Here, a newly formed class I MHC protein binds a peptide and is exported to the cell surface where it is recognized by a CTL with a suitable receptor.

Fig4. MHC peptide binding, antigen presentation, and MHC restriction. (A) Class I MHC proteins (known as class I HLA proteins in human cells) serve to bind peptides and display them on the cell surface. The peptides are produced by the degradation of any protein synthesized within the cell (either a host-cell protein or one made by a virus or other intracellular pathogen). Peptide fragments are produced within the proteasome and transported into the endoplasmic reticulum (ER). Here they are snipped by an endoplasmic reticulum aminopeptidase (ERAP) to the proper size needed for loading on to a partly unfolded class I HLA protein. Once the peptide has been bound, the HLA protein completes its folding and is transported to the plasma membrane with the bound peptide displayed on the outside. (B) Receptors on cytotoxic T cells bind class I MHC–peptide complexes; those on helper T cells bind class II MHC–peptide complexes. (C) MHC restriction. T cells have cell-specific receptors that recognize a combination of a specific peptide and a specific MHC protein. (Adapted from Murphy K [2011] Janeway’s Immunobiology, 8th edn. Garland Science. With permission from WW Norton.)
Because of cell-specific somatic recombinations, individual CTLs make unique T-cell receptors that recognize specific class I MHC–peptide combinations (Figure4B). If the bound peptide is derived from a pathogen, the CTL induces killing of the host cell. Note that a proportion of normal host-cell proteins also undergo degradation in the cytosol and the resulting self-peptides are bound by class I MHC proteins and displayed on the cell surface. But there is normally no immune response (starting in early fetal life, CTLs that recognize MHC–self-peptide are programmed to be deleted to minimize auto immune responses).
Class II MHC
proteins Class II MHC proteins are expressed in professional antigen-presenting cells: dendritic cells, macrophages, and B cells. These cells also express class I MHC proteins but, unlike most cells, they make co-stimulatory molecules needed to initiate lymphocyte immune responses.
Whereas class I MHC proteins bind peptides from endogenous proteins (that were made within the cytosol, such as a viral protein made after infection of that cell), class II MHC proteins bind peptides derived from exogenous proteins that were transported into the cell (by endocytosis of a bacterium or other microbial cell or its products) and delivered to an endosome where limited proteolysis occurs. The resulting peptide fragments are bound by previously assembled class II MHC proteins and transported to the cell surface so that a helper T lymphocyte with an appropriate receptor recognizes a specific class II MHC–peptide combination (see Figure 4B). (Helper T cells play critical roles in co-ordinating immune responses by sending chemical signals to other immune system cells.)
MHC restriction
T cells recognize foreign antigens only after they have been degraded and become associated with MHC molecules (MHC restriction; Figure 4C). A proportion of all nor mal proteins in a cell are also degraded and the resulting peptides are displayed on the cell surface, complexed to MHC molecules. MHC proteins cannot distinguish self from nonself, and even on the surface of a virus-infected cell the vast majority of the many thousands of MHC proteins are associated with bound peptides derived from host cell proteins rather than from virus proteins.
The rationale for MHC restriction is that it provides a simple and elegant solution to the problem of how to detect intracellular pathogens—it allows T cells to survey a pep tide library derived from the entire set of proteins in a cell but only after the peptides are displayed on the cell surface.
What is the basis of the extraordinary polymorphism of MHC proteins?
The polymorphism of the classic MHC genes is unusual in several respects. There is a very high number of alleles at each locus (Table 2), and because of the generally low allele frequencies, individuals are often heterozygotes at the classic MHC loci. There is also a very high frequency of nonsynonymous base substitution, consistent with some type of positive selection for new variants. Importantly, the substituted amino acids are not just restricted to the variable domains (which form the peptide binding groove), but are clustered at specific sites, occupying positions that line the peptide-binding groove.

Table2. SEQUENCE VARIATION AT THE SIX MOST POLYMORPHIC HLA LOCI
In the past, selection may have driven gene duplication in the MHC: by producing multiple classic class I and class II loci, more scope was provided for generating protein diversity. But it is clear that natural selection also works to promote the very high levels of polymorphism at the classic MHC loci, generating protein variants that exhibit different peptide-binding specificities.
The prevailing theory is that selection is pathogen-driven. Balancing selection (also called overdominant selection) has been invoked to favor heterozygotes ( heterozygote advantage). Heterozygosity at classic MHC loci might be expected to confer greater protection from pathogens than homozygosity, and so a higher chance of transmitting genotypes to future generations. Diversifying frequency-dependent selection has alternatively been proposed, whereby selection is skewed toward initially rare MHC alleles that confer a fitness advantage in response to new pathogenic variants.
Any selection operating on MHC genes must be longstanding, and is likely to have originated before the speciation event leading to evolutionary divergence from the great apes. MHC polymorphism is exceptional in showing trans-species polymorphism: some HLA protein variants are more closely related to the equivalent chimpanzee protein variants than they are to other protein variants at the same HLA locus. For example, human HLA-DRB1*0701 and HLA-DRB1*0302 show 31 amino acid differences out of 270 amino acid positions, but human HLA-DRB1*0701 and chimpanzee Patr-DRB1*0702 show only two differences out of the 270 positions.
An additional component seems to be MHC-driven disassortative mating. In many species, the choice of mate seems to be influenced by the MHC: individuals tend to choose MHC-different mates to have progeny with. If the differences in the MHC of the two parents are great, there is a greater chance that progeny are heterozygous at MHC loci. The basis of this type of sexual selection is not well understood. If it were to be based on olfaction, it may not be so significant in some contemporary human cultures where perfumes, aftershave lotions, and so on may be expected to obscure the natural olfactory cues.
The medical importance of the HLA system
The HLA system is medically important for two principal reasons. First, the high degree of HLA polymorphism poses problems in organ and cell transplantation. Second, certain HLA alleles are risk factors for individual diseases, notably many autoimmune diseases and certain infectious diseases; other HLA alleles are protective factors, being negatively correlated with individual diseases.
Transplantation and histocompatibility testing
Following organ and cell transplantation, the recipient’s immune system will often mount an immune response against the transplanted donor cells (the graft), which carry different HLA proteins to those of the host cells. The immune reaction may be sufficient to cause rejection of the transplant (but corneal transplants produce minimal immune responses—the cornea is one of a few immune privileged sites that actively protect against immune responses in several ways, including having much-reduced expression of class I HLA proteins).
Bone marrow transplants and certain stem cell transplants can also result in graft versus-host disease (GVHD) when the graft contains competent T cells that attack the recipient’s cells. GVHD can even occur when donor and recipient are HLA-identical because of differences in minor (non-HLA) histocompatibility antigens.
Immunosuppressive drugs are used to suppress immune responses following trans plantation, but transplant success largely depends on the degree of HLA matching between the cells of the donor and the recipient. Histocompatibility testing (also called tissue typing) involves assaying HLA alleles in donor tissues so that the best match can be found for prospective recipients. The key HLA loci are the most polymorphic ones: HLA-A, -B, -C, -DRB1, -DQB1, and -DPB1 (see Table 2).
HLA disease associations
By displaying peptide fragments on host-cell surfaces, HLA proteins direct T cells to recognize foreign antigens and initiate an immune response against cells containing viruses or other intracellular pathogens. Because HLA proteins differ in their ability to recognize specific foreign antigens, people with different HLA profiles might be expected to show different susceptibilities to some infectious diseases.
In autoimmune diseases, the normal ability to discriminate self-antigens from foreign antigens breaks down and autoreactive T cells launch attacks against certain types of host cell. Certain HLA proteins are very strongly associated with individual diseases, such as type 1 diabetes and rheumatoid arthritis; in general, genetic variants in the HLA complex are the most significant genetic risk factors that determine susceptibility to autoimmune diseases. Determining to what extent HLA variants are directly involved in the pathogenesis and how much is contributed by other variants that lie in the immediate vicinity of the HLA genes (and outside of the HLA complex) is a major area of research.