Volume 4, Issue 5, Pages 870-878 (September 2013) RAG2’s Acidic Hinge Restricts Repair-Pathway Choice and Promotes Genomic Stability  Marc A. Coussens, Rebecca L. Wendland, Ludovic Deriano, Cory R. Lindsay, Suzzette M. Arnal, David B. Roth  Cell Reports  Volume 4, Issue 5, Pages 870-878 (September 2013) DOI: 10.1016/j.celrep.2013.07.041 Copyright © 2013 The Authors Terms and Conditions

Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure 1 Acidic Residues in the Hinge Region of RAG2 Regulate Repair-Pathway Choice (A) RAG2 schematic. (B–F) Quantification of fluorescence-activated cell sorting (FACS) data from three separate experiments for aNHEJ in CHO-K1 cells measuring mutations to known regulatory regions (B), truncation mutations (C), internal deletions (D), and neutralizing mutations (E and F). (G) XRCC4-deficient CHO cells were tested for their ability to bypass cNHEJ defects using CJGFP and SJGFP. (H) Recombination on the chromosomal substrate pMX-INV in SCID-MEFs with the indicated RAG2 mutations. Values significantly different from wild-type RAG1 and RAG2 are marked with an asterisk (Student’s t test, unpaired, two-tailed, equal variance). Means from three independent experiments are plotted; error bars reflect the SEM. See also Figures S1 and S2. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure 2 Charged Residues in the Acidic Region Stabilize the RAG-DNA PCC (A) Biochemical end-release assay. (B) Gels for end-release assays, with the challenge temperatures indicated. PK samples were treated with proteinase K and SDS. (C) Quantification of signal end release. Error bars represent the SEM (∗p < 0.05, Student’s t test). (D and E) Hybrid joint formation measured by semiquantitative PCR (D) and sequence analysis (E). See also Figure S2. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure 3 The Acidic Region in RAG2 Promotes Genomic Stability (A) RAG2 messenger RNA (mRNA) expression in complemented RAG2−/− pre-B cells. (B) Top: Igκ locus with BAC probes for DNA FISH. Bottom: schematic of normal and aberrant metaphases. (C) Metaphases from complemented RAG2−/− preB cells with the BAC probes indicated in (A) and an FITC-chr6 paint. (D) Percentages of aberrant metaphases identified at the Igκ locus. Results from three separate experiments (Figure S3) were collected and combined. Statistical analysis was conducted using a one-tailed Fisher’s exact test for comparison with the wild-type-complemented aberrant metaphases. Error bars reflect the SEM. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure 4 Human SVs Permit aNHEJ and Impair Genomic Stability (A) Structurally predictive software to measure the inherent disorder and flexibility of the acidic hinge was used to compare the identified human RAG2 mutations with the wild-type sequence. (B) Human RAG2 acidic region SVs were tested for their ability to permit aNHEJ. (C) Metaphases from cells expressing the indicated RAG2 mutants with the Igκ locus on chromosome 6. (D) Quantification of aberrant metaphases observed in (C). Results from separate experiments (Figure S4) were collected and combined. Statistical analysis was conducted using a one-tailed Fisher’s exact test for comparison with the wild-type-complemented aberrant metaphases. Error bars reflect the SEM. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure S1 Measuring the Effects of RAG2 Mutations on the Formation of Coding and Signal Joints, Related to Figure 1 (A) Extrachromosomal substrates used in this study for coding joints (left), signal joints (middle), and coding joints formed through alternative NHEJ (right). (B–E) Quantification of FACS data from three separate experiments in CHO-K1 cells 48 hr after transfection with the indicated Rag expression vectors and the reporter substrates measuring (B) mutations to known regulatory regions, (C) premature truncation mutations as indicated, (D) internal deletion mutants as indicated, and (E) neutralizing mutations within the acidic hinge. (F) Verification of XRCC4-deficient cells by measuring coding joint formation by complementation experiments with wild-type XRCC4. (G) Measuring coding joints (left) and AltGFP-coding joints (right) in the human 293T cell line. Error bars represent the SEM. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure S2 RAG2 Mutant Protein Levels and Conservation, Related to Figures 1 and 2 (A) A Western blot of indicated HA-tagged RAG2 mutants expressed in fibroblasts. 15ug of soluble cell lysate was separated by gel-electrophoresis and transferred to a nitrocellulose membrane and exposed antibody to the HA tag for detection. Lanes were arranged for presentation. (B) Sequence alignment from indicated species of the acidic region in RAG2 performed by ClustalW. Residues indicated in black are completely conserved, while those indicated in gray are charge-conserved (Boxshade 3.21), supplementing previously published sequenced alignment data. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure S3 Data from Separate Metaphase-FISH Experiments and Flexibility Predictions, Related to Figure 3 (A) Three metaphase-FISH experiments showing the frequency at which the 3′ and 5′ Igκ signals were lost in pre-B cells expressing wild-type, core, Neut370-383, and Scrambled#2 RAG2. (B) Predictions based on the IUPREDS, IUPREDL, and mFDP algorithms for Wild-type RAG2 (black), core RAG2 (red), Neut370-383 RAG2 (blue) and Scrambled#2 RAG2 (yellow). Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions

Figure S4 Data from Separate Metaphase-FISH Experiments Using Human SVs, Related to Figure 4 Separate metaphase-FISH experiments showing the frequency at which the 3′ and 5′ Igκ signals were lost in pre-B cells expressing wild-type, N364K, and N364K/D380Y/D400H RAG2. Cell Reports 2013 4, 870-878DOI: (10.1016/j.celrep.2013.07.041) Copyright © 2013 The Authors Terms and Conditions