4. The diversity of germline information

24.3 Immunoglobulin genes are assembled from their parts in lymphocytes

Key terms defined in this section

C genes code for the constant regions of immunoglobulin protein chains.
V gene is sequence coding for the major part of the variable (N-terminal) region of an immunoglobulin chain.

A remarkable feature of the immune response is an animal’s ability to produce an appropriate antibody whenever it is exposed to a new antigen. How can the organism be prepared to produce antibody proteins each designed specifically to recognize an antigen whose structure cannot be anticipated?

For practical purposes, we usually reckon that a mammal has the ability to produce 10⁶-10⁸ different antibodies. Each antibody is an immunoglobulin tetramer consisting of two identical light chains (L) and two identical heavy chains (H). If any light chain can associate with any heavy chain, to produce 10⁶ V10⁸ potential antibodies requires 10³ V10⁴ different light chains and 10³ V10⁴ different heavy chains.

Figure 24.17 Immunoglobulin type and function is determined by the heavy chain. J is a joining protein in IgM; all other Ig types exist as tetramers.

There are 2 types of light chain and ~10 types of heavy chain. Different classes of immunoglobulins have different effector functions. The class is determined by the heavy chain constant region, which exercises the effector function (see Figure 24.17).

Figure 24.4 Heavy and light chains combine to generate an immunoglobulin with several discrete domains.

The structure of the immunoglobulin tetramer is illustrated in Figure 24.4. Light chains and heavy chains share the same general type of organization in which each protein chain consists of two principal regions: the N-terminal variable (V) region; and the C-terminal constant (C) region. They were defined originally by comparing the amino acid sequences of different immunoglobulin chains. As the names suggest, the variable regions show considerable changes in sequence from one protein to the next, while the constant regions show substantial homology.

Corresponding regions of the light and heavy chains associate to generate distinct domains in the immunoglobulin protein:

The variable (V) domain is generated by association between the variable regions of the light chain and heavy chain. The V domain is responsible for recognizing the antigen. An immunoglobulin has a Y-shaped structure in which the arms of the Y are identical, and each arm has a copy of the V domain. Production of V domains of different specificities creates the ability to respond to diverse antigens. The total number of variable regions for either light- or heavy-chain proteins is measured in hundreds. So the protein displays the maximum versatility in the region responsible for binding the antigen.

The number of constant regions is vastly smaller than the number of variable regions Xtypically there are only 1 V10 C regions for any particular type of chain. The association of constant regions in the immunoglobulin tetramer generates several individual C domains. The first domain results from association of the constant region of the light chain (C_L) with the C_H1 part of the heavy-chain constant region. The two copies of this domain complete the arms of the Y-shaped molecule. Association between the C regions of the heavy chains generates the remaining C domains, which vary in number depending on the type of heavy chain.

Comparing the characteristics of the variable and constant regions, we see the central dilemma in immunoglobulin gene structure. How does the genome code for a set of proteins in which any individual polypeptide chain must have one of <10 possible C regions, but can have any one of several hundred possible V regions? It turns out that the number of coding sequences for each type of region reflects its variability. There are many genes coding for V regions, but only a few genes coding for C regions.

In this context, "gene" means a sequence of DNA coding for a discrete part of the final immunoglobulin polypeptide (heavy or light chain). So V genes code for variable regions and C genes code for constant regions, although neither type of gene is expressed as an independent unit. To construct a unit that can be expressed in the form of an authentic light or heavy chain, a V gene must be joined physically to a C gene. In this system, two "genes" code for one polypeptide. To avoid confusion, we shall now refer to these units as "gene segments" rather than "genes."

The sequences coding for light chains and heavy chains are assembled in the same way: any one of many V gene segments may be joined to any one of a few C gene segments. This somatic recombination occurs in the B lymphocyte in which the antibody is expressed. The large number of available V gene segments is responsible for a major part of the diversity of immunoglobulins. However, not all diversity is coded in the genome; some is generated by changes that occur during the process of constructing a functional gene (for review see Tonegawa, 1983; Alt et al., 1987).

Essentially the same description applies to the formation of functional genes coding for the protein chains of the T-cell receptor. Two types of receptor are found on T cells, one consisting of two types of chain called α and β, the other consisting of γ and δ chains. Like the genes coding for immunoglobulins, the genes coding for the individual chains in T-cell receptors consist of separate parts, including V and C regions, that are brought together in an active T cell (see later; for review see Hood et al., 1985).

The crucial fact about the synthesis of immunoglobulins, therefore, is that the arrangement of V gene segments and C gene segments is different in the cells producing the immunoglobulins (or T-cell receptors) from all other somatic cells or germ cells (Hozumi and Tonegawa, 1976).

The construction of a functional immunoglobulin or T-cell receptor gene might seem to be a Lamarckian process, representing a change in the genome that responds to a particular feature of the phenotype (the antigen). At birth, the organism does not possess the functional gene for producing a particular antibody or T-cell receptor. It possesses a large number of V gene segments and a smaller number of C gene segments. The subsequent construction of an active gene from these parts allows the antibody/receptor to be synthesized so that it is available to react with the antigen. The clonal selection theory requires that this rearrangement of DNA occurs before the exposure to antigen, which then results in selection for those cells carrying a protein able to bind the antigen. The entire process occurs in somatic cells and does not affect the germline; so the response to an antigen is not inherited by progeny of the organism.

Recombination between V and C gene segments to give functional loci occurs in a population of immature lymphocytes. A B lymphocyte usually has only one productive rearrangement of light-chain gene segments and one of heavy-chain gene segments; a T lymphocyte productively rearranges an α gene and a β gene, or one δ gene and one γ gene. The antibody or T-cell receptor produced by any one cell is determined by the particular configuration of V gene segments and C gene segments that has been joined.

There are two families of immunoglobulin light chains, κ and λ , and one family containing all the types of heavy chain (H). Each family resides on a different chromosome, and consists of its own set of both V gene segments and C gene segments. This is called the germline pattern, and is found in the germline and in somatic cells of all lineages other than the immune system.

But in a cell expressing an antibody, each of its chains Xone light type (either κ or λ ) and one heavy type Xis coded by a single intact gene. The recombination event that brings a V gene segment to partner a C gene segment creates an active gene consisting of exons that correspond precisely with the functional domains of the protein. The introns are removed in the usual way by RNA splicing.

The principles by which functional genes are assembled are the same in each family, but there are differences in the details of the organization of the V and C gene segments, and correspondingly of the recombination reaction between them. In addition to the V and C gene segments, other short DNA sequences (including J segments and D segments) are included in the functional somatic loci (for review see Yancopoulos and Alt, 1986; Blackwell and Alt, 1989).

Figure 24.5 The lambda C gene segment is preceded by a J segment, so that V-J recombination generates a functional lambda light-chain gene.

A λ light chain is assembled from two parts, as illustrated in Figure 24.5. The V gene segment consists of the leader exon (L) separated by a single intron from the variable (V) segment. The C gene segment consists of the J segment separated by a single intron from the constant (C) exon.

The name of the J segment is an abbreviation for joining, since it identifies the region to which the V segment becomes connected. So the joining reaction does not directly involve V and C gene segments, but occurs via the J segment; when we discuss the joining of "V and C gene segments" for light chains, we really mean V-JC joining.

The J segment is short and codes for the last few (13) amino acids of the variable region, as defined by amino acid sequences. In the intact gene generated by recombination, the V-J segment constitutes a single exon coding for the entire variable region.

Figure 24.6 The kappa C gene segment is preceded by multiple J segments in the germ line. V-J joining may recognize any one of the J segments, which is then spliced to the C gene segment during RNA processing.

The consequences of the κ joining reaction are illustrated in Figure 24.6. A κ light chain also is assembled from two parts, but there is a difference in the organization of the C gene segment. A group of five J segments is spread over a region of 500 V700 bp, separated by an intron of 2 V3 kb from the C_k exon. In the mouse, the central J segment is nonfunctional (ΨJ3). A V_k segment may be joined to any one of the J segments (Max et al., 1979).

Whichever J segment is used becomes the terminal part of the intact variable exon. Any J segments on the left of the recombining J segment are lost (J1 has been lost in the figure). Any J segment on the right of the recombining J segment is treated as part of the intron between the variable and constant exons (J3 is included in the intron that is spliced out in the figure).

All functional J segments possess a signal at the left boundary that makes it possible to recombine with the V segment; and they possess a signal at the right boundary that can be used for splicing to the C exon. Whichever J segment is recognized in DNA joining uses its splicing signal in RNA processing.

Heavy chain construction involves an additional segment. The D (for diversity) segment was discovered by the presence in the protein of an extra 2 V13 amino acids between the sequences coded by the V segment and the J segment. An array of >10 D segments lies on the chromosome between the V_H segments and the 4 J_H segments

Figure 24.7 Heavy genes are assembled by sequential joining reactions. First a D segment is joined to a J segment; then a V gene segment is joined to the D segment.

V-D-J joining takes place in two stages, as illustrated in Figure 24.7. First one of the D segments recombines with a J_H segment; then a V_H segment recombines with the DJ_H combined segment. The reconstruction leads to expression of the adjacent C_H segment (which consists of several exons). (We come later to the use of different C_H gene segments; it suffices for now to consider the reaction in terms of connection to one of several J segments that precede a C_H gene segment.)

The D segments are organized in a tandem array. The mouse heavy-chain locus contains 12 D segments of variable length; the human locus has ~30 D segments (not all necessarily active). Some unknown mechanism must ensure that the same D segment is involved in the D-J joining and V-D joining reactions. (When we discuss joining of V and C gene segments for heavy chains, we assume the process has been completed by V-D and D-J joining reactions.)

The V gene segments of all three immunoglobulin families are similar in organization. The first exon codes for the signal sequence (involved in membrane attachment), and the second exon codes for the major part of the variable region itself (<100 codons long). The remainder of the variable region is provided by the D segment (in the H family only) and by a J segment (in all three families).

The structure of the constant region depends on the type of chain. For both κ and λ light chains, the constant region is coded by a single exon (which becomes the third exon of the reconstructed, active gene). For H chains, the constant region is coded by several exons; corresponding with the protein chain shown in Figure 24.4, separate exons code for the regions C_H1, hinge, C_H2, and C_H3. Each C_H exon is ~100 codons long; the hinge is shorter. The introns usually are relatively small (~300 bp).

Reviews

Alt, F. W., Blackwell, T. K., and Yancopoulos, G. D. (1987). Development of the primary antibody repertoire. Science 238, 1079-1087.

Blackwell, T. K. and Alt, F. W. (1989). Mechanism and developmental program of immunoglobulin gene rearrangement in mammals. Ann. Rev. Genet. 23, 605-636.

Hood, L., Kronenberg, M., and Hunkapiller, T. (1985). T cell antigen receptors and the immunoglobulin supergene family. Cell 40, 225-229.

Tonegawa, S. (1983). Somatic generation of antibody diversity. Nature 302, 575-581.

Yancopoulos, G. D. and Alt, F. W. (1986). Regulation of the assembly and expression of variable-region genes. Ann. Rev. Immunol. 4, 339-68.

Research

Hozumi, N. and Tonegawa, S. (1976). Evidence for somatic rearrangement of immunoglobulin genes coding for variable and constant regions. Proc. Nat. Acad. Sci. USA 73, 3628-3632.

Max, E. E., Seidman, J. G., and Leder, P. (1979). Sequences of five potential recombination sites encoded close to an immunoglobulin k constant region gene. Proc. Nat. Acad. Sci. USA 76, 3450-3454.