The RAG2 C terminus suppresses genomic instability and lymphomagenesis

Abstract

Misrepair of DNA double-strand breaks produced by the V(D)J recombinase (the RAG1/RAG2 proteins) at immunoglobulin (Ig) and T cell receptor (Tcr) loci has been implicated in pathogenesis of lymphoid malignancies in humans and in mice. Defects in DNA damage response factors such as ataxia telangiectasia mutated (ATM) protein and combined deficiencies in classical non-homologous end joining and p53 predispose to RAG-initiated genomic rearrangements and lymphomagenesis. Although we showed previously that RAG1/RAG2 shepherd the broken DNA ends to classical non-homologous end joining for proper repair, roles for the RAG proteins in preserving genomic stability remain poorly defined. Here we show that the RAG2 carboxy (C) terminus, although dispensable for recombination, is critical for maintaining genomic stability. Thymocytes from 'core' Rag2 homozygotes (Rag2(c/c) mice) show dramatic disruption of Tcrα/δ locus integrity. Furthermore, all Rag2(c/c) p53(-/-) mice, unlike Rag1(c/c) p53(-/-) and p53(-/-) animals, rapidly develop thymic lymphomas bearing complex chromosomal translocations, amplifications and deletions involving the Tcrα/δ and Igh loci. We also find these features in lymphomas from Atm(-/-) mice. We show that, like ATM-deficiency, core RAG2 severely destabilizes the RAG post-cleavage complex. These results reveal a novel genome guardian role for RAG2 and suggest that similar 'end release/end persistence' mechanisms underlie genomic instability and lymphomagenesis in Rag2(c/c) p53(-/-) and Atm(-/-) mice.

Figure 1. The C-terminus of RAG2 is a tumor suppressor in developing thymocytes

a. Kaplan-Meier tumor-free survival analysis for cohorts of control (WT, n=12 and Rag2^c/c, n=19), p53^−/− (n=32) and Rag2^c/c p53^−/− (n=25) mice. Animals were monitored for 50 weeks. The average age of death in weeks is shown for p53^−/− (22.8 weeks) and Rag2^c/c p53^−/− (12.1 weeks) genotypes with the P-value determined by the Wilcoxon rank sum test. b. Pie chart showing the tumor spectrum observed for Rag2^c/c p53^−/− (n=25) and p53^−/− mice (n=27). All Rag2^c/c p53^−/− animals (n=25) showed enlarged thymus. p53^−/− animals showed either enlarged thymus and/or spleen (n=18) or other non lymphoid tumor mass (n=9). c. Physical appearance of normal thymus (wild type) and thymic lymphoma (Rag2^c/c p53^−/−, arrow) of 3-month-old animals.

Figure 2. Rag2^c/c p53^−/− thymic lymphomas display recurrent translocations involving chromosomes that harbor antigen-receptor loci

Representative images of spectral karyotyping (1790T and 1745T) and G-band karyotyping (1779T) analysis of three Rag2^c/c p53^−/− T cell lymphomas. Metaphase number analyzed and translocations for each tumor sample are listed in the table.*All three tumors harbor clonal translocations involving chromosomes that carry Tcr (Chr.14: Tcrα/δ; Chr.6:Tcrβ) and/or Ig (Chr.12:Igh; Chr.6:Igκ; Chr.16:Igλ) loci.

Figure 3. Rag2^c/c p53^−/− thymocytes displayTcrα/δ and Igh-associated genomic instability

a. Top panel: schematic of the Tcrα/δ locus, with positions of the BACs used for generation of DNA FISH probes indicated. Bottom panels: representative metaphases from two Rag2^c/c p53^−/− thymic lymphomas using the Tcrα/δ V BAC probe (red signal) combined with chromosome 14 paint (green signal, top row) or with the Tcrα/δ C BAC probe (green signal, bottom row). Arrows point the amplification of the Tcrα/δ V region, arrow heads point the translocated chromosome 14. b. Top panel: schematic of the Igh locus, with positions of the BACs used for generation of DNA FISH probes indicated. Bottom panels: representative metaphases from the same two Rag2^c/c p53^−/− thymic lymphomas using the Igh C BAC probe (red signal) combined with chromosome 12 paint (green signal, top row), or with the Igh V BAC probe (green signal, bottom row). Combination of chromosome 12 (red) and chromosome 14 (green) paints is shown for both tumors in black boxes. Arrow heads point the translocated chromosome 12. c. Examples of confocal sections of three-dimensional Tcrα/δ DNA FISH on freshly isolated wild-type (top row) or Rag2^c/c (bottom rows) double positive thymocytes. Tcrα/δ V (Green) and C (Red) BAC probes were used. Scale bar = 1µm. d. Representative experiment showing the frequency at which Tcrα/δ V and/or Tcrα/δ C signals are lost in wild-type, p53^−/− and Rag2^c/c thymocytes (n>200, see Supplementary Fig. 11 for additional experiments and statistical analysis).

Figure 4. The C-terminus of RAG2 stabilizes the RAG post-cleavage complex

a. Biochemical end release assay. Purified GST-tagged core RAG1and non-tagged RAG2 (full length or core) proteins (yellow circles) cleave a 500bp DNA substrate at 37°C. Post-cleavage signal end complexes are thermally challenged at increasing temperatures to force the release of signal ends, which are detected after electrophoresis and gel staining. b. Representative gel for end release assays. Numbers above each lane indicate the temperatures the reactions were heated to before electrophoresis. PK, samples treated with proteinase K and SDS; SC, single cleavages; SE, signal ends; CE, coding ends. c. Quantification of SE release, measured as the combined amount of signal ends divided by the signal from the total amount of DNA in the lane, from six experiments using two different protein preparations (*P < 0.05, Student’s t-test).