Mechanisms of Recombination and Somatic Mutation
The immune system employs a DNA recombination/repair process to create genetic diversity in developing B cells. Immunoglobulin genes created by V(D)J recombination can further diversified by two processes; class switch recombination and somatic hypermutation. Class switch recombination changes antibodies produced by B cells from IgM to either IgG, IgE, or IgA and somatic hypermutation alters the variable region to increase their affinity against antigen. Despite its importance in generating immune responses, little is known about the molecular details of the switching mechanism. Tandemly repeated sequences located upstream of all H-chain antibody constant regions genes have generally been thought to be important in the targeting and/or the mechanism of class switch recombination.
Figure 1. This illustration diagrams the events that occur in both V(D)J and class switch recombination.
We have used gene targeting to delete the tandem repeats from the switch (S) region associated with the Cµ constant region in mice. Surprisingly, these mutant mice are still able to undergo isotype switching, although the efficiency of the process is reduced. We also find that switching to different isotypes is differentially affected by deletion of the Sµ tandem repeats. We are now examining if the mice lacking Sµ maintain normal IgA-switching. We are also evaluating switching to IgE in these mice. Potential molecular explanations for the different effects of Sµ on switching to the different isotypes are being investigated by characterizing switch recombination junction sites for isotypes that show the greatest and least effect due to Sµ deletion. Second, double-stranded DNA breaks within the JH-Cµ intron are being analyzed in wild type and Sµ mutant mice to assess whether DNA cleavage sites might be affected by Sµ deletion. Third, gene targeting is being used to delete additional sequences within the JH-Cµ intron to localize those sequences that are required for isotype switching. Lastly, we are considering introducing specific DNA sequences into an IgH allele that is defective for switching to investigate the minimal sequences that can restore the capacity for switching.
A second major area of interest relates to the way in which the recombination that occurs during isotype switching actually occurs. We are using the mice that lack the Sµ tandem repeats from the switch (S) region associated with the Cµ constant region for these studies. As noted earlier, these mutant mice are still able to undergo isotype switching, although the efficiency of the process is modestly reduced. Thus, the Sµ tandem repeats are not required for the switching process. More recent work indicates that the enzymes involved in DNA mismatch repair (MMR) enzymes also affect the efficiency and joining site selection of class switching. We have shown that this effect is particularly pronounced in the absence of the Sµ tandem repeats. We are now investigating the mutual relationships between the Sµ tandem repeats and MMR enzymes in the switching process by producing double-mutant animals that lack both the Sµ repeats and specific individual MMR proteins.
Figure 2. This drawing illustrates potential models by which MMR enzymes may affect class switching.
Our model proposes that Sµ tandem repeats and Msh2 are engaged in parallel pathways by which class switch-associated DNA lesions are resolved. DNA cleavages within Sµ tandem repeats are proposed to be closely spaced and, therefore, predominantly generate DNA ends that have short single-stranded extensions. These can be joined with downstream switch region DNA ends by DNA-PK/Ku complex-mediated nonhomologous end-joining (NHEJ) pathway independent of Msh2 (left panel). In contrast, sequences outside of the SµTR region (right panel) have more widely spaced cleavage sites, which can lead to flap DNA ends with longer 5' or 3' extensions. Msh2 is suggested to recruit nucleases to end-process the 3' flap DNA ends before they are recombined by the NHEJ mechanism.
We are also planning to investigate other aspects of the switch recombination process. For example, we want to determine whether RNA:DNA complexes can form in the ΔSµ JH-Cµ intron that lacks the tandem repeat element to explore the previously suggested importance of these complexes in the switching process. In addition, we plan to investigate the relative importance of the tandem repeats, in comparison to the tandem repeats found associated with downstream CH genes. For these experiments, we will produce, by gene targeting, mutant mice that lack either Sα or both Sµ and Sα. Analysis of switching in these mutant mice should indicate whether downstream S regions are much more important in controlling the recombination process as has been suggested by some studies of artificial switch substrates.