A study of lymphocytes that lack a DNA-repair enzyme challenges long-standing dogma about the spatial separation of processes that rearrange antibody genes, and provides clues about the origins of B-cell cancers.
Long-lived organisms are constantly being attacked by a myriad of pathogens that have evolved mechanisms to evade the host immune system. To counter this onslaught, vertebrate T and B lymphocytes have an extraordinarily diverse repertoire of surface receptors that recognizes an array of foreign antigens. The generation of this wide range of surface B-cell receptors (membrane-bound immunoglobulin) takes place in developing B lymphocytes in the bone marrow through a process that involves breakage and recombination of variable (V), diversity (D) and joining (J) segments of immunoglobulin genes. Mature B cells in peripheral tissues (the spleen and lymph nodes) also rearrange immunoglobulin genes by DNA breakage and repair, but through a different mechanism — class-switch recombination.
In an exciting study in this issue (page 231), Wang et al.1 find that a special type of V(D)J recombination — receptor editing — can take place in the periphery in mature B cells that are simultaneously undergoing class-switch recombination. In the absence of a DNA-repair enzyme, these cells experienced frequent chromosome translocations at the sites of immunoglobulin genes. These findings refute the long-standing belief that receptor editing and class-switch recombination are restricted to distinct anatomical locations and specific stages of B-cell development, and provide insight into the mechanism of gene translocations.
The immunoglobulin molecule (antibody) consists of two heavy-chain proteins that are noncovalently bound to two light chains (either two λ- or two κ-light chains). The genes encoding the heavy chain undergo V(D)J recombination, and those encoding the light chains VJ recombination, to form the V(D)J exon, which encodes the region of the immunoglobulin molecule that determines its specificity. This reaction is initiated by the RAG enzyme complex, which induces double-stranded DNA breaks in the V, D and J regions (Fig. 1a). After recombination, the breaks are repaired through a process known as non-homologous end-joining (NHEJ). In receptor editing, developing B cells in the bone marrow undergo successive rounds of RAG-mediated V(D)J recombination to exchange the light chains of an autoreactive immunoglobulin molecule so that it is no longer autoreactive2,3.
Class-switch recombination, which changes the effector function of immunoglobulins, is a process by which the exons for the constant domain of the heavy chain of IgM are swapped with downstream exons to generate different classes of antibody, such as IgG, IgE or IgA. Class-switch recombination is also initiated by DNA breaks, in this case induced by the AID enzyme (Fig. 1b). These breaks can also be repaired by NHEJ or by an alternative end-joining (A-EJ) pathway. Another AID-mediated mechanism of gene rearrangement in peripheral B cells is somatic hypermutation. Here, mutations accumulate in the rearranged immunoglobulin genes, potentially increasing antibody-binding specificity.
That immunoglobulin-gene rearrangements are associated with double-stranded DNA breaks underscores the enormous selective pressures driving the evolution of these processes — when not repaired correctly, DNA breaks can lead to chromosome translocations, which predispose to cancer. Indeed, certain types of human B-cell tumour (lymphomas) frequently contain translocations that merge antigen-receptor genes with a proto-oncogene (a gene with the potential to promote cancer).
There are several biological mechanisms that reduce the tumour-causing potential of DNA breaks in B cells, including restricting these processes to distinct tissues, such as the bone marrow for V(D)J recombination and peripheral tissues for class-switch recombination. However, as Wang et al.1 show, these apparent safeguards underestimate the plasticity of B cells.
The authors examined mice in which the NHEJ double-strand-break-repair protein XRCC4 is deleted in mature B cells. They report that a subset of activated peripheral B cells with defective NHEJ simultaneously harbour double-stranded DNA breaks associated with V(D)J recombination and class-switch recombination. Surprisingly, when these cells are activated by signals that lead to class-switch recombination in the DNA locus encoding the immunoglobulin heavy chain (Igh), they also re-initiate V(D)J recombination at the immunoglobulin-λ light-chain locus (Igl).
In Wang and colleagues' study, the splenic B cells that reactivate V(D)J recombination are not undergoing the conventional V(D)J recombination used by developing B cells to generate the initial immunoglobulin repertoire. Instead, the authors argue, these cells are undergoing receptor editing, which was thought to be confined to immature B cells in the bone marrow. It has previously been suggested4 that editing can occur in peripheral B cells during the generation of memory B cells in specific regions of the spleen — the germinal centres — through a mechanism termed receptor revision. However, the peripheral editing in the B cells in Wang et al.'s study1 seems to be distinct from the receptor-revision mechanism, because the B cells lack germinal-centre markers and are not activated by signals that normally lead to germinal-centre formation.
Incorrect repair of the breaks initiated by receptor editing and class switching frequently resulted in chromosome translocations1 involving Igh and Igl (Fig. 2, overleaf). Although neither Igh nor Igl are chromosomal regions that promote cancer, translocations involving Igh or Igl with a proto-oncogene, such as c-myc, can result in lymphomas. Indeed, deletion of both Xrcc4 and p53 (a gene encoding a tumour-suppressor protein) in mature B cells in mice leads to lymphomas, known as CXP lymphomas, the cells of which reveal evidence of receptor editing and class switching5. Thus it is likely that the XRCC4-depleted cells in Wang and colleagues' study are the progenitors of mouse CXP lymphoma cells, and that B cells with similar mutations may contribute to some human B-cell lymphomas.
B cells that undergo both class-switch recombination and receptor editing harbour AID- and RAG-dependent DNA breaks, and Wang et al.1 capitalize on this characteristic to identify factors that may enhance translocations between immunoglobulin loci. One such factor may be the A-EJ pathway of DNA double-strand-break repair. As the XRCC4-depleted cells cannot repair DNA breaks with NHEJ and are forced to use the A-EJ pathway, the increase in translocations may reflect a propensity of the A-EJ mechanism to generate such translocations.
The authors found a strong correlation between translocations and proximity of Igh and Igl in the nucleus of B cells at interphase, a cell-cycle stage during which much of the gene expression occurs. Similarly, co-localization of c-myc and Igh in the nucleus correlated with translocations between these loci, although the rate-limiting factor was the frequency of breaks at c-myc, which is strongly AID-dependent6.
That the B cells studied by Wang et al.1 are the progenitors of mouse CXP lymphomas suggests that peripheral editing occurs in vivo and may contribute to the development of such cancers. Paradoxically, in the authors' study, the signals activating V(D)J recombination in B cells are typically associated with activation of class-switch recombination rather than autoreactivity — the trigger for receptor editing in the bone marrow. So, if not the revision of an autoreactive receptor, what is achieved by replacing the light chain in this subset of peripheral B cells? To fully understand the significance of this phenomenon, it will be important to determine the frequency of receptor editing in peripheral B cells and its physiological function. Finally, this novel B-cell population can be exploited to elucidate the mechanisms that promote translocations between antigen-receptor loci and proto-oncogenes.
Wang, J. H. et al. Nature 460, 231–236 (2009).
Gay, D., Saunders, T., Camper, S. & Weigert, M. J. Exp. Med. 177, 999–1008 (1993).
Tiegs, S. L., Russell, D. M. & Nemazee, D. J. Exp. Med. 177, 1009–1020 (1993).
Hertz, M. & Nemazee, D. Curr. Opin. Immunol. 10, 208–213 (1998).
Wang, J. H. et al. J. Exp. Med. 205, 3079–3090 (2008).
Robbiani, D. F. et al. Cell 135, 1028–1038 (2008).