Analysis of V(D)J recombination in NHEJ-defective human cells & a structure:function analysis of mammalian Ku86.

Abstract

V(D)J recombination is a site-specific DNA recombination process required for the assembly of immunoglobulin (Ig) and T cell receptor (TCR) genes, which is accomplished through the use of DNA double strand breaks (Bassing et al., 2002). The resulting hairpinned coding ends and the signal ends are rejoined primarily by the canonical non-homologous end joining (C-NHEJ) pathway which consists of at least seven factors: DNA-PKcs, Ku70, Ku86, Artemis, LigIV, XLF, XRCC4 (Rooney et al., 2004). Deficiency of any these NHEJ factors leads to a phenotype of ‘Severe Combined Immune Deficiency’ (SCID), which is characterized by a profound immune dysfunction resulting from aberrant V(D)J recombination (Bassing et al., 2002; Rooney et al., 2004; Zha et al., 2007). The accompanying blockage in B and T cell development, increased ionizing radiation sensitivity, and developmental defects can all be explained by the absence of the C-NHEJ pathway (Bassing et al., 2002; Rooney et al., 2004; Zha et al., 2007). Using rAAV-mediated gene targeting (Kohli et al., 2004), our laboratory has generated loss-of-function mutations for all of the genes involved in NHEJ with the exception of Artemis and XRCC4 (in press or manuscripts in preparation). We have characterized these human cell lines that are either heterozygously or homozygously defective for one of the NHEJ factors, to assess the role of each factor in V(D)J recombination with three episomal substrates; pGG49 (signal joint), pGG51 (coding joint) and pGG52 (inversion construct). We have found that there is a dramatic defect in the frequency of V(D)J recombination in all null cell lines with all three substrates. However, there is variation in the recombination frequency and the sequences utilized at joints depending on which substrate and which cell line is used. This implies a different role for each of the NHEJ factors in signal, coding or hybrid joint formation in V(D)J recombination. We also observed similar recombination efficiency and sequence variation patterns at the joint junction in wild type and all heterozygous cell lines suggesting that residual NHEJ activity is enough for V(D)J recombination. This contrasts with the role of these same genes in generalized DNA double-strand break

(DSB) repair for which a haploinsufficiency is often observed (Fattah et al., 2008b; Ruis et al., 2008). In chapter 2, we pursued the idea suggested by our work in chapter 1 that CNHEJ factors may have different functions in different pathways. To this end, we investigated a separation-of-function series of mutations for Ku86. For example, the Ku heterodimer not only recognizes the broken ends of DNA at V(D)J recombinationmediated DSBs and promotes their reassembly but also protects telomeres from undergoing fusions. How Ku performs these seemingly opposite functions on different, but structurally similar, substrates is unknown. As a precedent for the successful application of this experimental approach, however, separation-of-function mutations for Ku have been identified in yeast Ku (Bertuch and Lundblad, 2003; Palmbos et al., 2005; Ribes-Zamora et al., 2007; Roy et al., 2004; Stellwagen et al., 2003). Previously, several Ku86 mutant cell lines (sxi-1-4) derived from Golden Syrian hamster V79-4 cells were isolated by forward and molecular genetic screens (He et al., 1996b; Zdzienicka, 1995). All of these cell lines shared common features including elevated sensitivity to radiation, defective DNA DSB repair, and aberrant V(D)J recombination. By functionally complementing these mutant cell lines with wild-type Ku86 cDNAs, it was determined that the absence of Ku86 caused these aberrant phenotypes (Errami et al., 1996; Smider et al., 1994). Here, we used the Ku86-defective sxi-3 cell line to determine if separation-offunction of Ku86 in terms of DNA DSB repair, recombination and telomere maintenance is conserved in mammalian cells. We introduced six point mutations associated with defective telomere maintenance into the hamster Ku86 cDNA. These cDNAs were then stably introduced into sxi-3 cells and various DNA repair and V(D)J recombination assays were used to interrogate the lines. Domain-specific functions of Ku86 were unveiled by these experiments. In summary, we have used mutant human cell lines and mutant cDNAs to elaborate the role of C-NHEJ factors — in general and Ku86 specifically — in a plethora of DNA metabolic reactions. Our results have improved our understanding of C-NHEJ reactions in humans.