Alternative End-Joining and Classical Nonhomologous End-Joining Pathways Repair Different Types of Double-Strand Breaks during Class-Switch Recombination

Elena M. Cortizas, Astrid Zahn, Maurice E. Hajjar, Anne-Marie Patenaude, Javier M. Di Noia and Ramiro E. Verdun
J Immunol December 1, 2013, 191 (11) 5751-5763; DOI: https://doi.org/10.4049/jimmunol.1301300

Abstract

Classical nonhomologous end-joining (C-NHEJ) and alternative end-joining (A-EJ) are the main DNA double-strand break (DSB) repair pathways when a sister chromatid is not available. However, it is not clear how one pathway is chosen over the other to process a given DSB. To address this question, we studied in mouse splenic B cells and CH12F3 cells how C-NHEJ and A-EJ repair DSBs initiated by the activation-induced deaminase during IgH (Igh) class-switch recombination (CSR). We show in this study that lowering the deamination density at the Igh locus increases DSB resolution by microhomology-mediated repair while decreasing C-NHEJ activity. This process occurs without affecting 53BP1 and γH2AX levels during CSR. Mechanistically, lowering deamination density increases exonuclease I recruitment and single-stranded DNA at the Igh locus and promotes C-terminal binding protein interacting protein and MSH2-dependent DSB repair during CSR. Indeed, reducing activation-induced deaminase levels increases CSR efficiency in C-NHEJ–defective cells, suggesting enhanced use of an A-EJ pathway. Our results establish a mechanism by which C-NHEJ and this C-terminal binding protein interacting protein/MSH2-dependent pathway that relies on microhomology can act concurrently but independently to repair different types of DSBs and reveal that the density of DNA lesions influences the choice of DSB repair pathway during CSR.

Introduction

Unrepaired DNA double-strand breaks (DSBs) are highly toxic lesions that can produce a permanent arrest of cell division or even cell death. Even though different DNA repair pathways can be recruited to sites of DNA strand breaks in a normal cell, the choice of an appropriate repair pathway is critical to avoid genomic instability that could lead to cell death or transformation.

Mammalian cells rely on nonhomologous end joining (NHEJ) and homologous recombination (HR) as the main pathways for repairing DSBs. Because HR depends on the presence of homologous sequences elsewhere in the genome, it is active during the S and G2/M phases of the cell cycle. In contrast, classical-NHEJ (C-NHEJ) can fuse DNA ends without any sequence homology and is active during all the cell-cycle phases (1, 2). Therefore, C-NHEJ is the preferred pathway for the repair of DSB in G1 cells, when sister chromatids are not available. This is illustrated by the essential role of C-NHEJ during the repair of DSBs induced by ionizing radiation (3). C-NHEJ is also critical for the repair of programmed DSBs generated during lymphocyte development. Under these circumstances, C-NHEJ repairs the DSBs that are physiological intermediates during V(D)J recombination of B and TCRs genes and plays a dominant role during Ig class-switch recombination (CSR) in B cells (4).

In mature B lymphocytes, CSR underpins the change of the Cμ C-region of the IgH (Igh) to a downstream C-region such as Cε, Cγ, or Cα. This process allows B lymphocytes to switch from the initially expressed IgM to a different isotype (IgA, IgG, or IgE) while keeping the same Ag specificity. At the molecular level, CSR is a deletional recombination reaction at the Igh locus between highly repetitive switch regions (S-regions), which are located just upstream of each C-region (4). CSR depends on activation-induced cytidine deaminase (AID), an enzyme that converts deoxycytidines to deoxyuridines within the donor and acceptor S-regions (5). The deaminase activity of AID in both the donor Sμ and the downstream acceptor S-region initiates noncanonical repair, leading to DSBs that are recombined to resolve through one of two end-joining pathways during the G1 phase of the cell cycle (5–10). The important role of C-NHEJ during CSR is supported by the drastically reduced level of CSR observed in cells with mutations in the core components of the C-NHEJ pathway, such as the DNA-end binding complex Ku70/Ku80 heterodimer or the DNA ligase complex XRCC4/ligase-4 (Lig4) (11–15). However, C-NHEJ–defective B cells still have significant levels of CSR, in which the switch junction sites show a preference for the use of microhomologies for the joining of DNA ends (13, 15–19). This end-joining mechanism, which is loosely defined by not relying on core components of the C-NHEJ pathway, has been termed alternative-NHEJ, microhomology-mediated-EJ, or alternative end-joining (A-EJ). Hence, although the C-NHEJ pathway joins DNA ends with short homology (1–3 nt) or no homology, A-EJ shows a preference for joining DNA ends that display microhomology (4–20 nt). Beyond this partial bias, it is still under discussion if A-EJ represents one or more pathways (20). However, it is accepted that at least microhomology-mediated A-EJ starts with limited 5′ to 3′ resection at the DSB to expose ssDNA to be used for annealing to a homologous sequence (21). In support of this model, A-EJ can use the C-terminal binding protein interacting protein (CtIP) for the DNA end-resection step, although CtIP can also boost more extensive end-resection for HR (22, 23). Moreover, a microhomology-mediated A-EJ seems to generate a majority of chromosomal translocations involving the Igh locus that occur as CSR byproducts (16, 19, 24, 25). Thus, A-EJ may be considered a potentially deleterious backup end-joining pathway that operates only in C-NHEJ–defective cells (16, 18, 19, 26). However, it has been reported that A-EJ plays an important role in the repair of DSBs in cells with an intact C-NHEJ pathway. Indeed, knockdown of CtIP produces an important decrease in the level of CSR in C-NHEJ–proficient B cells (27). Although the above-mentioned studies established a role for the A-EJ pathway in normal cells at least during CSR to IgA, it is still unclear how the choice between using A-EJ or C-NHEJ is made and whether A-EJ is in competition with C-NHEJ for repairing the same DSBs.

In this study, we investigate the order and interdependency of association of multiple DNA repair factors with the Igh locus during CSR. Our results reveal that both C-NHEJ and a CtIP/MSH2-dependent A-EJ pathway are simultaneously active during CSR, and, rather than competing, they preferentially repair different types of DSBs as defined by the density of deamination events. Our data support a model in which the spatial distance between AID-initiated breaks on opposite DNA strands determines the type of DSB requiring repair and that this requirement defines which end-joining pathway is most efficient in repairing the lesion. Hence, the topological nature of the DSB has a critical influence on determining the choice of usage between C-NHEJ and at least a CtIP/MSH2-dependent A-EJ pathway.

Materials and Methods

Mice, cell lines, and CSR analysis

CH12F3-2 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 0.05% 2-ME, and 5% NCTC 109 (Sigma-Aldrich). CH12F3-2 Lig4−/− cells were supplemented with 20% of conditional media from an exponentially growing CH12F3-2 culture.

CH12F3-2 cells stably expressing uracil-DNA glycosylase (UNG) inhibitor (Ugi) were obtained by retroviral delivery using pMIG vector. The supernatant of HEK293T cells cotransfected with pMIG or pMIG-Ugi and vectors expressing murine leukemia virus Gag-Pol and vesicular stomatitis virus-G envelope (3:1:1 ratio) was transferred to retronectin-coated (r-fibronectin CH-296; Takara Bio) plates and centrifuged at 2000 × g for 90 min at 32°C. After a PBS wash, 5 × 105 CH12F3-2 cells were added and spin-infected at 600 × g for 30 min at 32°C. Infected cells were sorted to obtain GFP+ homogeneous populations. Class-switching induction to IgA in CH12F3-2 cells was performed as previously published (28). CSR to IgA in CH12F3-2 cells was determined with an Accuri C6 Flow Cytometer (BD Biosciences). Class-switching induction and determination of IgG1 in resting B cells from Aicda−/− mice were performed as previously published (28).

Short hairpin RNA knockdowns

We generated stable short hairpin RNA (shRNA)-mediated knockdowns in CH12F3-2 cells following protocols from The RNAi Consortium library. Transfected cells were selected with puromycin at 2 μg/ml final concentration. For double knockdowns, CH12F3-2 cells were first transfected with a pLKO.1 neomycin version of the shRNA (shRNA against AID [shAID]-1, shAID2, or shGFP) and selected, then transduced with the second shRNA, and selected again with puromycin before analysis. The shRNAs used in this study are listed in Supplemental Table I.

Abs sources and Western blot

Western blot for the different proteins was performed as previously described (29). The following Abs were used in this study: 53BP1 (NB100-304; Novus Biologicals); Actin (A2066; Sigma-Aldrich); AID, a kind gift from J. Chaudhuri (Sloan-Kettering Institute), for Western blot; AID (39885; Active Motif) for Western blot and chromatin immunoprecipitation (ChIP); CtIP (sc-5970; Santa Cruz Biotechnology); DNA Ligase III (HPA006723; Sigma-Aldrich); exonuclease I (EXO1; NBP1-19709; Novus Biologicals); γH2AX (EMD Millipore, 05-636); H2AX (A300-083A; Bethyl Laboratories); Ku70 (sc1486; Santa Cruz Biotechnology); MSH2 (sc-22771; Santa Cruz Biotechnology); and replication protein A (RPA; NA19L; Millipore).

ChIP assays

We adapted previously published protocols (29) for the ChIP assays with minor modifications, except coimmunoprecipitated DNA was analyzed using real-time PCR with SYBR Green (Applied Biosystems). Briefly, cells were cross-linked with 1% formaldehyde for 20 min at room temperature, and reaction was stopped by addition of glycine to 125 mM final concentration. Cells were washed twice with cold PBS, harvested, and kept at −80°C overnight. Samples were resuspended in RIPA buffer (150 mM NaCl, 1% [v/v] Igepal CA-630, 0.5% [w/v] sodium deoxycholate, 0.1% [w/v] SDS, 50 mM Tris-HCl [pH 8], 5 mM EDTA, protease, and phosphatase inhibitors) and sonicated to generate DNA fragments <500 bp using a Bioruptor Next Gen (Diagenode). Samples were then clarified by centrifugation at 20,000 rpm (4°C), and their protein content was measured using the BCA method (Bio-Rad). For immunoprecipitation, 0.5 mg (2 μg/μl) protein extract was precleared for 2 h with 30 μl 50% G protein-Sepharose slurry before addition of indicated Abs. Between 2 and 5 μg each Ab was added to the samples and incubated overnight at 4°C.

Thirty microliters 50% Agarose A/G plus (Santa Cruz Biotechnology) slurry per immunoprecipitation were preblocked overnight at 4°C with 1 mg/ml BSA and 0.3 mg/ml tRNA before addition of indicated Abs. Beads were washed twice (15 min at 4°C) with RIPA buffer, three times with ChIP Wash Buffer (100 mM Tris-HCl [pH 8.5], 0.5 M LiCl, 1% [v/v] Igepal CA-630, and 1% [w/v] sodium deoxycholate), and once with 1× TE. Immunocomplexes were eluted for 10 min at 65°C with 100 μl Elution buffer (1% [w/v] SDS), and cross-linking was reversed by adjusting to 200 mM NaCl, 1 mM EDTA, and 1 mM DTT and incubating overnight at 65°C in the presence of 5 μg proteinase K. DNA was purified using QIAquick PCR purification kit (Qiagen), and DNA resuspended in 60 μl Tris-HCl (pH 8) was used as template in real-time PCR reactions. Primers against multiple regions of the mouse (C57BL/6) Igh locus were designed with Primer Express 2.0 (Applied Biosystems). PCR products ranging between 57 and 123 bp in size were cloned and sequenced to confirm identity. Ten-microliter PCR reactions containing 1× SYBR Green Mix (Applied Biosystems), 1/60 fraction of the ChIP-enriched DNA, and 100 nM primers were set up in 96-well plates. Standard curves using different quantities of the input extracts were run alongside ChIP samples for each individual primer, and plates were read in an Applied Biosystems StepOnePlus Real-time PCR instrument. IgG and input DNA values were used to subtract/normalize the values from ChIP samples. All primer sequences used for the ChIP analyses are available upon request.

Switch junctions analysis

Genomic DNA was purified from CH12F3 cells after switching to IgA with anti-CD40 Ab (BD Biosciences), TGF-β (R&D Systems), and IL-4 (PeproTech). Fifty micrograms was used for the PCR amplification step following previously published protocols (30), except Taq Gold polymerase (Applied Biosystems) was used. The PCR products were cloned in a TOPO system and sequenced with an automated sequencer at the University of Miami Oncogenomics facility. The primers used for the amplification of Sμ-Sα are: 5′-CCAGACAGAGAAAGCCAGACTCATAAAGCT-3′ and 5′-AGATAGTAGTCTTCACACTGCCCATCTCGA-3′.

The junctions were analyzed by aligning the PCR products with the Sμ and Sα sequences. Analysis of mutation pattern at the switch junctions was performed as previously described (30, 31).

DNA breaks assay

The DNA break assay was performed as previously described (32) with minor modifications. The HindIII digestion of the genomic DNA was replaced by sonication to DNA fragments of 500–1000 bp using a Bioruptor Next Gen (Diagenode), and the biotin-labeled DNA fragments were purified using Dynabeads-Streptavidin (Invitrogen) following the manufacturer's instructions. The levels of biotinilated DNA were determined via quantitative PCR using the Sμ primers indicated in Supplemental Table I.

Intra-S deletion analysis

CH12F3 cells expressing shAID-1 or shGFP were stimulated for class switching during 96 h, and clones were isolated by serial dilution. After identifying IgM-secreting clones via dot blot/Western blot analysis and validating by flow cytometry analysis, the selected IgM+/IgA− clones were expanded in 24-well plates. The extracted genomic DNA from the selected clones was used as template for PCR amplification of the Sμ region using these primers: forward, 5′-CAGTCATTGCTTTAGGGGGAGAAAGAGACA-3′ and reverse, 5′-GCCACTGCACACTGATGTCT-3′. The PCR products were analyzed by Southern blot using a 32P-labeled Cμ probe as described previously (33). The primer set used for the Cμ probe is: OJ417 5′-AGCCCCTCCCCCACCTCCACCTACCTATTAC-3′ and OJ418 5′-CACGAACTTCACTCCAAAACCGATCACAG-3′.

Statistical analysis

A two-tailed paired Student t test was used for the analyses of CSR efficiency, and a two-tailed Fisher exact test was used for analyses of microhomology length at switch junctions in the different mutants.

Results

Ku70 and CtIP bind independently to the Igh locus during CSR

To determine whether C-NHEJ and A-EJ compete for the repair of DSBs during CSR, we used ChIPs to study the chronological order underlying the recruitment of key components from each end-joining pathway to the Igh locus. We set up a system using the mouse lymphoma cell line CH12F3-2, which is capable of efficient cytokine-induced AID expression and CSR to IgA in vitro through the same molecular mechanisms used by primary lymphocytes (5, 18, 27, 34). Time-course experiments showed that AID and CSR are rapidly induced after cytokine stimulation, as detected by immunoblotting, and by the presence of IgA at the cell surface, as observed by flow cytometry (Fig. 1A, 1B). Accordingly, we found detectable AID at the Sμ as early as 8–10 h poststimulation, reaching the maximum level at ∼16 h postinduction (Fig. 1C).

FIGURE 1.
FIGURE 1.

AID expression and binding to the Igh locus. (A) Western blot showing AID expression in CH12F3-2 cells during CSR. (B) FACS analysis of CSR after staining of surface IgA during the first 24 h postcytokine stimulation in CH12F3-2 cells expressing the indicated shRNAs. Percentage of IgA-positive cells is indicated. (C) AID binding to the Igh locus at different times after inducing CSR to IgA in CH12F3-2 cells expressing the indicated shRNAs was analyzed by ChIP and the Sμ-U amplicon signal plotted. The average values obtained from at least three independent experiments at the different time points are represented in the bar graph. (D) Top panel, Schematic representation of Igh showing the location in Sμ and Sα of primers sets used for ChIP assays. First nucleotide in schematics corresponds to nt 180969 of National Center for Biotechnology Information reference sequence NG_005838.1. Bottom panel, ChIP experiments showing AID distribution in Sμ and Sα 16 h after cytokine stimulation of CH12F3-2 cells expressing the indicated shRNAs. (E) Comparison of switch-associated mutation in m.2 region of Sμ in CH12F3-2 cells expressing Ugi and either shGFP or shAID-3. The average number of mutations per 10 kb sequenced is shown (>98% of mutations were G to A or C to T transitions). Error bars represent SD from at least three independent experiments.

The highest AID levels were observed at the S-regions, more specifically in the region immediately upstream to the Sμ repeat (Pre-Sμ region) and the 3′ region within the Sα repeat (Fig. 1D), consistent with previous reports that analyzed the AID footprint at the Igh (35, 36). For the negative control, we used cytokine-stimulated cells with stable knockdown of AID via shRNA (Fig. 1C, 1D; denoted as shAID-3, see below). The timing of AID occupancy correlated with the activity of AID at the pre-Sμ region [m.2 (36)] (Fig. 1E), as revealed by the presence of G-to-A or C-to-T transitions in CH12F3-2 cells expressing the potent Ugi, which inhibits base excision repair (35–38). These experiments showed that AID was not only rapidly induced in stimulated CH12F3-2 but also immediately recruited to the S-regions, where it deaminated dC and triggered CSR.

We then analyzed the recruitment of C-NHEJ and A-EJ in this system. Although there are no known factors that are exclusive for A-EJ to use as markers, CtIP has been clearly implicated in a microhomology-mediated pathway for CSR to IgA in CH12F3-2 cells (27), so we analyzed it as a possible marker or at least this particular case of A-EJ. To determine whether there was a chronological order of recruitment of DNA repair proteins to the Igh locus during CSR, we performed kinetic analyses for Ku70, a core component of the C-NHEJ pathway (39), and CtIP. We found that CtIP and Ku70 were both present in high levels at the Sμ region when AID recruitment was maximal (14–16 h; Fig. 2C, 2E). However, CtIP association with chromatin preceded that of Ku70 and was found even when AID was at low concentration in the cells and showed lower Igh occupancy (8–12 h; Fig. 2C, 2E). Control experiments with CH12F3-2 cells expressing shAID-3 demonstrated that the recruitment patterns we observed were AID dependent at all time points (Fig. 2C, 2E).

FIGURE 2.
FIGURE 2.

Ku70 and CtIP recruitment to the Igh locus during CSR. (A) Western blot analysis of Ku70 and CtIP knockdown in CH12F3-2 cells expressing the indicated shRNAs. (B) FACS analysis of CSR after staining of surface IgA during 24 and 48 h postcytokine stimulation in CH12F3-2 cells expressing the indicated shRNAs. Percentage of IgA-positive cells is indicated. (C) CtIP binding to the Igh locus at different times after inducing CSR to IgA in CH12F3-2 cells expressing the indicated shRNAs was analyzed by ChIP. The ChIP results obtained using the Sμ-U set of primers is plotted. The average values obtained from at least three independent experiments at the different time points are represented in the bar graph. (D) ChIP experiments showing AID distribution in Sμ and Sα 16 h after cytokine stimulation of CH12F3-2 cells expressing the indicated shRNAs. (E) As in (C) for Ku70. (F) As in (D) for Ku70 in CH12F3-2 cells. Error bars represent SD from at least three independent experiments. (G) As in (D) for Ku70 in G1-enriched CH12F3 cells expressing the indicated shRNAs. (H) As in (G) for CtIP. Error bars represent SE from two independent experiments. Student t test, *p = 0.01, **p = 0.006, ***p = 0.009.

The different timing observed for the recruitment of Ku70 and CtIP to the Igh locus suggested that these factors do not compete for binding to the S-regions during CSR. Therefore, we tested whether respective knockdown of either CtIP or Ku70 would modify the magnitude of recruitment for the other factor. In agreement with previous reports (17, 27), knockdown of Ku70 and CtIP using specific shRNAs reduced CSR to different extents (∼60% reduction for shKu70 and ∼25% for shCtIP), and simultaneous knockdown of both factors abolish almost completely CSR (Fig. 2A, 2B). This reduction was not due to nonspecific effects on cell proliferation, AID protein expression, or germline transcript levels (Supplemental Fig. 1). These experiments validated the choice of CtIP as a component of an A-EJ pathway for CSR in CH12F3-2 cells because the other pathway in which CtIP has been implicated, HR, would not produce CSR. Importantly, CtIP knockdown reduced its concentration at the S-regions to background levels during CSR (Fig. 2C) without affecting the presence or distribution of Ku70 at the S-regions (Fig. 2E, 2F). Similar conclusions were obtained regarding the presence and distribution of CtIP at the S-regions in cells depleted of Ku70 (Fig. 2C, 2D).

Recruitment of Ku70 to the Igh locus presumably occurs during G1 phase of the cell cycle during the C-NHEJ repair of the AID-induced DSBs (40), whereas a recent report showed that CtIP is involved in microhomology-mediated end joining of DSBs during S and G2 phases (41). Therefore, to address whether CtIP and Ku70 could be both present at the Igh in the same cell-cycle stage, we performed ChIP assays for Ku70 and CtIP in G1-enriched CH12F3 cells after class-switching stimulation to IgA. We found that CtIP was present at the Sμ in G1-enriched CH12F3 cells and that its enrichment and distribution were not altered by the knockdown of Ku70 (Fig. 2G). Analogous results were obtained for Ku70 in G1-enriched CH12F3 cells (Fig. 2H), thus indicating that the presence of CtIP and Ku70 were not marking cells in different stages of the cell cycle.

These results strongly suggest that these components of the C-NHEJ and microhomology-mediated A-EJ pathways do not compete for association to the Igh locus and suggest that both pathways are independently recruited to the Igh S-regions during CSR, although CtIP was preferentially recruited at early time points (Fig. 2C), suggesting it might be present when AID levels and activity at the Igh are lower (Fig. 1).

The use of C-NHEJ during CSR requires high deamination density of the S-regions

Previous reports have shown that Ku70 knockdown increases the use of microhomologies at the switch junctions during CSR, whereas inhibition of CtIP decreases the use of microhomologies to favor direct joins at the switch junctions (17, 27). We have obtained the same results in this study (Fig. 3A–C). However, when we mapped the location of the junctions within the Sμ genomic region, we noticed a striking difference in the distribution of junctions between Ku70- and CtIP-depleted cells. Indeed, most junctions in the Ku70-depleted cells were found in the regions further away from the Sμ, where the preferred AID substrate motifs (WRCY) are sparser (area a in Fig. 3D). Thus, compared with control cells, Ku70-depleted cells preferentially lost junctions nearby or inside the Sμ repeats, which are rich in AID hot spots (area b in Fig. 3D). Conversely, junctions from CtIP-depleted cells were clustered adjacent to the Sμ repeat and very few closer to Iμ, where AID hot spots are less abundant (Fig. 3D). These results suggest that DSBs in different areas of the Sμ region can be preferentially processed by the CtIP-dependent A-EJ or C-NHEJ pathway and that this correlates with the distribution of WRCY motifs and AID deamination density found in this region.

FIGURE 3.
FIGURE 3.

Effect of Ku70 and CtIP knockdown on pattern of Sμ-Sα junctions and breakpoints location. (A) Examples of alignment of Sμ-Sα with germline sequences. The position of the primers used for the amplification of the junctions is represented in the schematic. Arrowheads indicate breakpoints in Sμ and Sα. (B) Comparison of Sμ-Sα junctions in control CH12F3-2 cells (shGFP) or CH12F3-2 cells expressing shRNA against Ku70 (shKu70-1 or shKu70-2) or CtIP (shCtIP-2 or shCtIP-3). Two-tailed Fisher exact test was used for p value test. (C) Percentage of switch junctions with the indicated length of microhomology identified as the largest perfect matches to the germline sequences. Data shown are based on 79 or more independent sequences per genotype that were obtained from three independent experiments. (D) Plot showing the position and frequency of breakpoints, determined as shown in (A), with respect to the pre-Sμ and Sμ after analysis of switch junctions from the indicated CH12F3-2 cell lines. The positions of predicted AID deamination hot spots (W(A/T)R(A/G)CY(C/T)) are indicated. *p = 0.010, **p = 0.014.

We therefore asked whether AID levels affect the likelihood of CSR proceeding through the C-NHEJ or CtIP-dependent pathways. To answer this question, we established stable CH12F3-2 cell lines expressing different shRNAs that decreased AID to various levels (Fig. 4A). For each shRNA, the decrease in AID levels observed by Western blot was proportional to the reduction in IgA switching in each cell line (Fig. 4C). Germline transcripts, cell division, and Ku70 or CtIP expression were not affected by shRNA knockdown of AID (Fig. 4A, 4B, Supplemental Fig. 1). Furthermore, we analyzed the AID footprint in CH12F3-2 cells with low levels of AID (i.e., with shAID-1) and control cells (shGFP) expressing Ugi. This analysis confirmed that in cells with low levels of AID, there was a ∼60% reduction in the total number of mutations in the Pre-Sμ region and also an increase in the distance between mutations located in opposite DNA strands (Fig. 4E, Supplemental Fig. 2; similar results were observed for shAID-2, not shown). Next, we analyzed the sequence of the switch junctions after inducing CSR in two independent shAID-1– or shAID-2–expressing CH12F3-2 cell lines (referred to as AIDlow). Both lines showed a clear decrease in direct junctions and an increase in the use of microhomologies at the switch-junctions (Fig. 4F, shAID-2 not shown). These results are in agreement with a previous report showing that human patients heterozygous for AID show increased usage of microhomologies at the Sμ-Sα switch-junctions (30). Our results suggested that partial AID knockdown might promote the use of microhomologies and possibly A-EJ over C-NHEJ in the repair of DSBs during the CSR. In support of these conclusions, knockdown of CtIP but not Ku70 significantly reduced CSR in AIDlow cells, confirming that most CSR was proceeding via a CtIP-dependent pathway in these cells (Fig. 4D). Furthermore, we found a significant decrease in intra-S recombination at the Sμ in AIDlow cells when compared with control cells (shGFP) (Supplemental Fig. 2B), further indicating that the level of AID influences the choice of joining mechanism.

FIGURE 4.
FIGURE 4.

C-NHEJ activity depends on a high density of deaminations in the S-regions. (A) Left panel, Western blot analysis of AID in CH12F3-2 cells expressing the indicated shRNAs against AID. Middle panel shows the quantification of the Western blot in the left panel. Right panel shows the proliferation analysis of CH12F3-2 cells expressing the indicated shRNAs against AID. (B) Western blot analysis of Ku70 or CtIP in CH12F3-2 cells expressing the indicated shRNAs against AID. (C) Representative flow cytometry analysis of CSR after staining of surface IgA in CH12F3-2 cells expressing the indicated shRNAs 24 h after stimulation. Percentage of IgA+ cells is indicated at top right of boxed areas. CSR efficiency relative to control cells (shGFP) after 48 h is indicated in the right panel as average + SD for five experiments. Fisher exact test, *p = 0.001. (D) FACS analysis of CSR after staining of surface IgA 24 h postcytokine stimulation in CH12F3-2 cells expressing the indicated shRNAs. Percentage of IgA-positive cells is indicated at top right of boxed areas. CSR efficiency relative to control cells (shGFP) after 48 h is indicated in the right panel. t test, *p = 0.01, **p = 0.006. (E) Left panel, Comparison of mutation load in the m.2 region of Sμ in CH12F3-2 cells expressing Ugi and the indicated shRNAs 20 h poststimulation. The average number of transitions at C:G per 10 kb sequenced is shown. Student t test, *p = 0.003. Right panel, Plot showing the position and frequency of transitions in the indicated region of pre-Sμ. (F) Comparison of Sμ-Sα junctions in control CH12F3-2 cells (shGFP) or CH12F3-2 cells expressing shAID-1. Percentage of switch junctions with the indicated length of microhomology identified as the largest perfect matches to the germline sequences. Data shown are based on ≥84 independent sequences per genotype that were obtained from three independent experiments. Error bars represent SD. *p = 0.001.

We next analyzed the presence of CtIP as well as of C-NHEJ core elements in the Sμ during CSR in both CH12F3-2 cells and primary mouse B cells. As expected, a partial knockdown of AID (with shAID-1 or shAID-2) yielded a partial reduction (∼40–50%) in its occupancy of the Sμ (Fig. 5A, shAID-2 not shown). Notably, ChIP analysis showed that the concentration of Ku70 and CtIP at the S-regions showed opposite trends under partial AID knockdown. In AIDlow cells, the presence of Ku70 was drastically decreased, whereas presence of CtIP was significantly increased at the Sμ, relative to control cells (Fig. 5A, 5B). Interestingly, we also observed at the Sμ of AIDlow cells an increase of DNA ligase III, which has also been implicated in the microhomology-mediated A-EJ pathway in the absence of C-NHEJ (42, 43). Importantly, overexpression of human AID protein reverted the phenotype, confirming that the behavior of Ku70, CtIP, and DNA ligase III in AIDlow cells was not due to unspecific effects of shAID1 or shAID-2 (Fig. 5B). Similar results were obtained when we compared the presence of Ku70 and CtIP at the Sμ in stimulated primary splenic B cells from Aicda+/+ versus Aicda+/− mice (Fig. 5C), showing that this is not a CH12F3-2–specific phenomenon.

FIGURE 5.
FIGURE 5.

CtIP and DNA ligase III (LigIII) increase their concentration in the Igh when AID levels are reduced during CSR. (A) AID, CtIP, Ku70, and LigIII recruitment to the Igh locus during CSR analyzed by ChIPs in CH12F3-2 cells expressing the indicated shRNAs. Average + SD values obtained at each time point from three independent experiments are represented in the graph. (B) As in (A) but showing the distribution in pre-Sμ and Sμ regions in CH12F3-2 cell lines expressing the indicated shRNAs and human AID (hAID). Error bars represent SD from at least three independent experiments. (C) As in (B) but in naive splenic B cells obtained from the indicated mouse strains after class-switching stimulation with LPS and IL-4. Error bars represent SE from two independent experiments. (D) Western blot analysis indicating the AID expression levels in the indicated cell lines after 16 h of CSR induction.

Taken together, these data indicate that the distribution of AID deaminations in the S-regions (Fig. 4E), and therefore the density of single-strand breaks (SSBs) in each strand, can influence the choice of end-joining pathway for CSR by determining a different quality of DSBs (i.e., blunt versus staggered).

The mismatch repair pathway is essential for CSR in AIDlow cells

Our results in this study show that a lower density of AID-initiated lesions correlates with an elevated dependence on a CtIP-dependent, A-EJ pathway that preferentially used microhomologies for CSR. The mismatch repair (MMR) pathway can trigger CSR (38), possibly by processing distal SSBs to generate DSBs (44), thus indicating MMR might be more important in this case than for C-NHEJ (45). We confirmed the importance of MMR for CSR in CH12F3-2 cells by reducing MSH2 expression by shRNA, which significantly decreased the switching from IgM to IgA by ∼45% (Fig. 6A). However, MSH2 knockdown in the AIDlow cells reduced CSR more severely, by ∼80%, without affecting cell division, germline transcripts, or AID protein levels (Fig. 6A and Supplemental Fig. 1). To characterize the role of MMR in the shift from C-NHEJ to A-EJ usage in AIDlow cells, we examined the recruitment of the MMR factors MSH2 and EXO1 to S-regions during CSR. In control cells, both proteins were recruited to the Sμ in an AID-dependent fashion (Fig. 6B). Notably, we found EXO1 and MSH2 at the Igh locus from the same early time points at which AID and CtIP were first detected through later times when Ku70 was also present (compare Fig. 6B to Fig. 2C, 2E). This pattern did not change for MSH2 in AIDlow cells. However, AIDlow cells showed a significant increase in the magnitude of EXO1 recruitment to the Sμ at later time points (Fig. 6B), consistent with the need for DNA resection for the microhomology-mediated joins. Accordingly, we also found an increase in the accumulation of RPA at the Igh in AIDlow cells when compared with control cells (Fig. 6C, 6D). As RPA accumulates in regions where ssDNA is exposed, these results suggest increased DNA resection in AIDlow cells compared with the wild-type (WT) cells, subsequent to the processing of AID-initiated lesions by EXO1.

FIGURE 6.
FIGURE 6.

The MMR pathway is essential for CSR in AIDlow cells. (A) Analysis of CSR at the indicated time points for CH12F3-2 cells expressing the indicated shRNAs. Student t test; *p = 0.008, **p = 0.001. (B) AID, EXO1, and MSH2 association to the Sμ during CSR analyzed by ChIPs in CH12F3-2 cells expressing the indicated shRNAs. The average values obtained at the different time points are shown. (C) As in (B) for RPA. (D) ChIP experiments showing RPA distribution in Sμ and Sα 20 h after cytokine stimulation of CH12F3-2 cells. Error bars represent SD from at least three independent experiments.

Collectively, these results further support the notion that a CtIP-dependent microhomology-mediated A-EJ, in which MMR plays a major role, predominates for CSR to IgA in AIDlow cells

Reducing AID levels does not affect the DNA damage signaling at the Igh locus

The DNA damage signaling triggered after detection of a DSB is well characterized, with 53BP1 and the phosphorylated form of histone H2AX (ser129-H2AX or γH2AX) being two of the main players (46). Both factors are important for CSR, especially 53BP1, which is absolutely required for CSR (4). We showed that lower AID levels lead to more frequent usage of microhomology-mediated A-EJ to resolve CSR, but AIDlow cells display fewer mutations in the S-regions, which may translate into fewer DSBs that could affect the strength of DNA damage signaling during CSR. However, the results obtained via ChIP experiments did not show significant differences in the concentration or the distribution of γH2AX or 53BP1 at the Igh locus between control and AIDlow cells during CSR (Fig. 7A–D and not shown). In addition, direct quantitative comparison of DNA breaks by end-labeling followed by immunoprecipitation and real-time PCR (7) showed that there was no significant difference in the level of breaks at the Sμ between AIDlow and control cells (shGFP) before or after class-switching stimulation (Fig. 7C). Furthermore, CSR was ataxia telangiectasia mutated dependent in control as well as AIDlow cells, as demonstrated by incubating the cells with the inhibitor KU-55933 (Fig. 7D). Similarly, γH2AX levels were not affected in AIDlow cells during switching (Fig. 7E). Thus, these results indicate that the DSBs induced during CSR in AIDlow cells trigger a DNA damage response similar to that found in the control cells.

FIGURE 7.
FIGURE 7.

Reducing AID levels does not change the level of DNA damage factors at the Igh locus. (A) ChIPs showing γH2AX presence at the Igh locus during CSR in CH12F3-2 cells expressing the indicated shRNAs. Average values obtained at each time point are represented in the graph. (B) ChIPs showing the distribution of γH2AX in Sμ and Sα after 16 h of cytokine stimulation. (C) DSB assay in CH12F3-2 cells stimulated for class switching and expressing the indicated shRNAs. DSBs were labeled with TdT and biotin–deoxyuridine triphosphate as previously described (32). Biotin-labeled fragments were purified, and its distribution at the Sμ was analyzed via PCR using the primer sets indicated. (D) Representative flow cytometry analysis of CSR to IgA 24 h postcytokine stimulation in CH12F3-2 cells expressing the indicated shRNAs. DMSO or 2 μM ataxia telangiectasia mutated inhibitor KU55933 (ATMi) was added at time 0. Percentage of IgA+ cells is indicated at top right of boxed areas. (E) Western blot of total H2AX and γH2AX in CH12F3-2 cells expressing the indicated shRNAs, stimulated or not for CSR. The bars represent the average percentage for each group in the indicated cell lines. (F) As in (D) for CH12F3-2 cells expressing the indicated shRNAs. CSR efficiency relative to control cells (shGFP) after 48 h is indicated. Error bars represent SD from at least three independent experiments. t test, *p = 0.02.

To further analyze the influence of γH2AX over the A-EJ during CSR, we decreased the levels of H2AX in control and AIDlow cells via shRNA. Although partial knockdown of H2AX (not shown) led to a decrease of switching in normal cells, it did not further decrease CSR in AIDlow cells (Fig. 7F, Supplemental Fig. 1E–G), in stark contrast to what we observed for MSH2 or CtIP (Figs. 4D, 6A). Hence, the AIDlow cells still show γH2AX and 53BP1 in the S-regions during CSR, but this does not prevent the activity of the CtIP/MSH2-mediated A-EJ pathway.

Reducing AID levels improve CSR efficiency in C-NHEJ–deficient cells

Based on the results described above, we envisage a model in which DSBs generated during CSR in regions with lower density of AID-mediated deamination would require nucleolytic processing and, at least for CSR to IgA, be preferentially repaired by a CtIP/MSH2-dependent A-EJ pathway. By contrast, C-NHEJ would take care of blunt DSBs that would more often arise in regions with higher density of deamination (Fig. 8A).

FIGURE 8.
FIGURE 8.

Reducing AID levels improves CSR in C-NHEJ mutant cells. (A) Model for C-NHEJ and A-EJ activity during CSR in cells with WT levels (AID) or reduced levels of AID (AID+/−). Dotted areas mark regions with low or AID-induced high density of DNA SSBs. (B) Western blot analysis of AID knockdown in CH12F3-2 cells expressing the indicated shRNAs. (C) As in (B) for Lig4−/− CH12F3-2 cells. (D) Representative flow cytometry analysis of CSR to IgA 24 h postcytokine stimulation in CH12F3-2 cells expressing the indicated shRNAs. Percentage of IgA+ cells is indicated at top right of boxed areas. Relative CSR efficiency to control cells (shGFP) after 48 h is indicated. Two-tailed Student t test was used for p value test. *p = 0.003. (E) As in (D), but with Lig4−/− CH12F3-2 cells or control cells (Lig4+/+) and after 48 h of class-switching stimulation. t test, *p = 0.005, **p = 0.004. Error bars represent SD from at least three independent experiments.

A testable prediction of our model is that if the nature of the DSB determines the choice of end-joining pathway, then lowering the density of deamination should increase the activity of this A-EJ by providing more suitable DSBs arising from distal nicks in opposite strands. For this critical test, we measured CSR in AIDlow cells lacking an active C-NHEJ pathway due to Ku70 or Ku80 knockdown (Fig. 8B, 8D). As we predicted, there was ∼50% increase in CSR in both the Ku70 AIDlow and Ku80 AIDlow cells compared with the Ku70 and Ku80 expressing WT AID levels (Fig. 8D). Next, we tested DNA Lig4 knockout CH12F3-2 cells, which undergo CSR to IgA using exclusively A-EJ (18). Lig4-null CH12F3-2 cells showed impaired proliferation when compared with control cells as previously reported (Supplemental Fig. 1I) (18). We addressed this problem by using conditional media and evaluating class switching 48 h poststimulation. CSR was increased by 2- to 3-fold in Lig4−/− cells simply by expressing shRNAs that reduced AID levels to ∼50% (Fig. 8C, 8E) without affecting the proliferation of these cell lines (Supplemental Fig. 1I). As would be expected from an increase in microhomology-mediated A-EJ activity, AIDlow Ku70 knockdown cells showed increased association of CtIP to the Sμ during CSR when compared with CH12F3-2 expressing WT levels of AID (not shown). In any case, our results indicate that WT levels of AID activity favor C-NHEJ by creating suitable DSBs and thereby reducing the usage of the CtIP/MSH2-dependent A-EJ.

Discussion

In this study, we examined whether the C-NHEJ and a microhomology-mediated A-EJ pathway process the same DSBs during CSR. Our results are consistent with a scenario in which C-NHEJ and at least this particular CtIP/MSH2-dependent microhomology-mediated A-EJ pathway do not compete for the repair of DSBs but act in parallel to repair different types of breaks initiated by AID during CSR to IgA in CH12F3-2 cells. Because A-EJ is undefined and probably includes more than one pathway, we can only conclude about the particular pathway that we show in this study is dependent on CtIP and MSH2 and is active even in C-NHEJ–competent cells. We demonstrate that the density of lesions in the S-regions can determine the usage of C-NHEJ or this A-EJ to resolve CSR (Figs. 2–5). Accordingly, diminishing the density of deamination events in the S-regions by controlling the levels of AID leads to an increase in CSR in C-NHEJ–defective cells, entirely catalyzed by the CtIP/MSH2-dependent A-EJ (Fig. 8). Hence, we propose a model for resolving CSR in which the nature of the lesion determines the choice of end-joining pathway: C-NHEJ preferentially repairs DSBs formed from nearby SSBs in opposite DNA strands, whereas A-EJ pathways would repair more staggered DSBs generated by the processing of distant SSBs (Fig. 8A).

Several studies have shown how the initial processing of DSBs influences the occurrence of an end-resection step at the break, which seems to control the decision of whether the DSB will be repaired by homology-dependent pathways (HR or microhomology-mediated A-EJ) or by C-NHEJ (47, 48). Recent reports have shown that 53BP1 and γH2AX inhibit the resection of DNA ends, which favors the C-NHEJ pathway over the A-EJ pathway for the repair of blunt DSBs in G1 cells (47, 49, 50). Our findings demonstrate that at least the CtIP-mediated A-EJ pathway is active even though γH2AX and 53BP1 are present at high concentration in the S-regions, suggesting this pathway is not significantly affected by these DNA damage factors during CSR, at least when the breaks are staggered (see Figs. 2, 7). Moreover, partial knockdown of H2AX did not affect the level of CSR in AIDlow cells, which rely mainly on this A-EJ for CSR (Fig. 7). Our conclusions do not contradict previous reports indicating that γH2AX and 53BP1 prevent end resection. The observed differences can be readily explained due to the different nature of the systems used in each study. Most previous studies on the influence of 53BP1 or γH2AX over the A-EJ pathway used systems in which DSBs were generated by either RAG1/2 during V(D)J recombination or by I-SceI meganuclease in modified S-regions (47, 49–51). In both cases, the DSBs generated are either blunt or have very short overhangs (4 nt) by the nature of the nicking endonucleases. However, AID initiates the formation of SSBs for CSR through DNA processing by multiple enzymes and in which components of the BER pathway (UNG and APE1) play a central role (4). Only those SSBs that are sufficiently near to each other in opposite DNA strands can form a spontaneous DSB and are likely protected from resection by 53BP1 and γH2AX. In contrast, SSBs in opposite strands resulting from deamination of two relatively distant dC need processing by the MMR pathway to be converted into DSBs, which might only then trigger the accumulation of γH2AX and 53BP1 at the damage site (44, 45, 52). Accordingly, MMR is particularly important to allow CSR using the pre-Sμ when the Sμ tandem repeats are deleted (53), because the lower AID hot spot frequency in the pre-Sμ will most preferentially accommodate staggered breaks. The proposed model for processing SSBs by MMR during CSR predicts that the U:G mismatch is recognized by the MSH2/MSH6 complex. After this, EXO1 together with the MLH1-PMS2 heterodimer are recruited to the damage site and induce the excision of the mutated strand from a nearby SSB toward the mismatch. EXO1 requires a nick and would continue until it finds an SSB in the opposite strand (44, 54), but recent evidence indicates that EXO1 functions as a scaffold in CSR for an unknown exonucleases that would then generate a DSB with a single-strand end or overhang. Mammalian EXO1 encodes both structural and catalytic functions that play distinct roles in essential biological processes (55). In support of the proposed role for the MMR pathway during CSR, the increase in EXO1 and RPA recruitment in the S-regions observed in AIDlow cells occurs at later times and correlates with the decrease in the density of mutations in the S-regions found in these cells (Figs. 4, 6), which suggests that exonucleolytic processing needs to be recruited to produce the DSBs in these cells. MMR-generated overhangs can therefore expose homologous sequences in both DNA ends, helping the A-EJ pathway to bypass the need for CtIP-dependent resection activity that is inhibited by γH2AX and 53BP1 as previously shown in other systems (47, 49–51). Although the use of EXO1 could bypass the need of end-resection by CtIP, AIDlow cells still depend on CtIP for CSR (Fig. 4). One potential explanation for this is that during CSR, CtIP is required to maintain the resected ends generated by the MMR pathway, as was recently suggested in another system (56). Alternatively, this could be explained by the scaffold role of EXO1 in CSR (55). Our results in AIDlow cells agree with previous evidence showing decreased microhomology in switch junctions from MSH2- and EXO1-deficient mice (57, 58) as well as with more recent evidence that MMR functions upstream of the DSB in C-NHEJ mutant cells (45) (Fig. 6). Thus, we propose a model in which the spacing of the DNA lesions dictates an MMR-dependent processing that fails to be inhibited by 53BP1 or γH2AX and facilitates repair by a microhomology-mediated A-EJ. Nevertheless, it is important to point out that the microhomology-mediated A-EJ pathway we analyzed still depends on UNG activity because UNG inhibition almost completely abolishes CSR (Supplemental Fig. 3). This is in agreement with a model in which the main role of MMR is to convert distant SSBs in opposite strands into DSBs as previously suggested (44, 52).

We find less intra-S recombination at the Sμ in cells expressing lower levels of AID. Very limited analysis in WT B cells suggests that both C-NHEJ and A-EJ can mediate intra-S deletions but the situation is complex (59). A limitation of CH12F3 cells is that they may not undergo intra-S deletions through C-NHEJ and A-EJ in the same proportion than WT B cells; hence, we acknowledge that the physiological relevance (or mechanistic insight) of our result is limited. However, it does suggest that lowering the levels of AID changes the pathway of DNA repair used at the Sμ, as our other data indicate.

There are surely other factors that contribute to the choice of end-joining pathway. As expected for a mechanism that relies on DNA annealing, the level of sequence homology present near the breaks is most likely to influence how effective the microhomology-mediated A-EJ can be and how frequently it can be used. In this respect, we would expect the A-EJ to have a different efficiency in compensating for the absence of C-NHEJ depending on the isotype and accordingly to the microhomology that each S-region shares with the Sμ. In the mouse, this would be maximal for IgA and minimal for IgG1, with 14% identity between Sμ and Sγ1 and 33% identity between Sμ and Sα (19, 60). Indeed, when we analyzed Sμ-Sγ1 switch joins from Aicda+/− mice, the increase in microhomology use was less evident (data not shown), in line with previous reports in human B cells (61–63). A recent report showing that CtIP does not affect class switching to IgG1 in mouse splenic B cells is also compatible with these conclusions (64). Interestingly, some MMR mutants show a similar behavior, with the observed CSR deficiency varying with the isotype (45, 65, 66), which is consistent with our finding of A-EJ more critically relying on MMR (Fig. 6). Although it is conceivable that reduced levels of AID in AIDlow cells can reduce the possibility of generating simultaneous DSBs in both Sμ and Sα, we would expect this to prevent CSR altogether rather than choice between C-NHEJ and the microhomology-mediated A-EJ pathways. The usage of microhomology used as one of our readouts to infer the use of C-NHEJ versus microhomology-mediated A-EJ would not be affected by this issue because we can assume that in those cells in which CSR took place, the simultaneous DSBs did occur. Those data also support a model of differential use according to the level of deamination (Fig. 1F) (30).

In conclusion, our data demonstrate that a microhomology-biased A-EJ pathway functions during CSR to IgA in C-NHEJ–competent cells and that it processes DSBs that cannot be joined by the C-NHEJ pathway. We suggest that by determining the type of DNA breaks through the density of DNA deamination, AID activity influences the end-joining pathway that will be used for CSR. However, further work is needed to determine if A-EJ and C-NHEJ factors can bind to the same S-regions and whether they do so simultaneously. AID seems to indirectly influence the recruitment UNG and MSH2 (67), which could also contribute to determine the quality of DSBs at low concentrations of AID. We cannot rule out that AIDlow cells may attract a lower density of C-NHEJ components to the Sμ if AID were to directly recruit Ku70/80 or other critical C-NHEJ factor, although to date there is no solid evidence for this. Despite these unresolved issues, our results support a molecular mechanism that could explain the increased usage of microhomology in the Sμ-Sα joins that is characteristic in human AID haploinsufficient individuals (30). It is likely that CSR still preferentially proceeds by C-NHEJ in normal B cells for most isotypes (11–15), but it would be interesting to determine whether there is some type of intrinsic or extrinsic DNA lesion for which a CtIP-dependent microhomology-mediated A-EJ might be the preferred choice. For instance, the preferential use of A-EJ in processes such as chromosome translocations (25) may reflect the complex nature of the DSBs used for the translocations. It is tempting to speculate that by producing a deamination density at the Sμ that favors C-NHEJ over A-EJ, the physiological levels of AID have been optimized during evolution to reduce the risk of chromosomal translocations during CSR.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Tasuku Honjo for CH12F3-2 cells and AID−/− mice, Jayanta Chaudhuri for anti-AID Ab, Kefei Yu for CH12F3-2 Lig4−/− cells, and Jan Karlseder, Daniel Durocher, Marcela Raices, Carlos Perez-Stable, and Denise Muñoz for comments on the manuscript.

Footnotes

  • This work was supported by operating grants from a Stanley J. Glaser Foundation research award (to R.E.V.) and Canadian Cancer Society Research Institute Grant 2010-700348 (to J.M.D.N.). J.M.D.N. is supported by a Canada Research Chair tier 2.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    A-EJ
    alternative end-joining
    AID
    activation-induced deaminase
    ChIP
    chromatin immunoprecipitation
    C-NHEJ
    classical nonhomologous end-joining
    CSR
    class-switch recombination
    CtIP
    C-terminal binding protein interacting protein
    DSB
    double-strand break
    EXO1
    exonuclease I
    HR
    homologous recombination
    Lig4
    ligase-4
    MMR
    mismatch repair
    NHEJ
    nonhomologous end-joining
    PA
    replication protein A
    shAID
    short hairpin RNA against activation-induced deaminase
    shRNA
    short hairpin RNA
    S-region
    switch region
    SSB
    single-strand break
    Ugi
    uracil-DNA glycosylase inhibitor
    UNG
    uracil-DNA glycosylase
    WT
    wild-type.

  • Received May 15, 2013.
  • Accepted September 24, 2013.

References

  1. ↵
    1. Karanam K.,
    2. R. Kafri,
    3. A. Loewer,
    4. G. Lahav
    . 2012. Quantitative live cell imaging reveals a gradual shift between DNA repair mechanisms and a maximal use of HR in mid S phase. Mol. Cell 47: 320–329.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Rothkamm K.,
    2. I. Krüger,
    3. L. H. Thompson,
    4. M. Löbrich
    . 2003. Pathways of DNA double-strand break repair during the mammalian cell cycle. Mol. Cell. Biol. 23: 5706–5715.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Lieber M. R.
    2010. The mechanism of double-strand DNA break repair by the nonhomologous DNA end-joining pathway. Annu. Rev. Biochem. 79: 181–211.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Stavnezer J.,
    2. J. E. Guikema,
    3. C. E. Schrader
    . 2008. Mechanism and regulation of class switch recombination. Annu. Rev. Immunol. 26: 261–292.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Muramatsu M.,
    2. K. Kinoshita,
    3. S. Fagarasan,
    4. S. Yamada,
    5. Y. Shinkai,
    6. T. Honjo
    . 2000. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102: 553–563.
    OpenUrlCrossRefPubMed
    1. Chaudhuri J.,
    2. M. Tian,
    3. C. Khuong,
    4. K. Chua,
    5. E. Pinaud,
    6. F. W. Alt
    . 2003. Transcription-targeted DNA deamination by the AID antibody diversification enzyme. Nature 422: 726–730.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Dickerson S. K.,
    2. E. Market,
    3. E. Besmer,
    4. F. N. Papavasiliou
    . 2003. AID mediates hypermutation by deaminating single stranded DNA. J. Exp. Med. 197: 1291–1296.
    OpenUrlAbstract/FREE Full Text
    1. Petersen-Mahrt S. K.,
    2. R. S. Harris,
    3. M. S. Neuberger
    . 2002. AID mutates E. coli suggesting a DNA deamination mechanism for antibody diversification. Nature 418: 99–103.
    OpenUrlCrossRefPubMed
    1. Pham P.,
    2. R. Bransteitter,
    3. J. Petruska,
    4. M. F. Goodman
    . 2003. Processive AID-catalysed cytosine deamination on single-stranded DNA simulates somatic hypermutation. Nature 424: 103–107.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Revy P.,
    2. T. Muto,
    3. Y. Levy,
    4. F. Geissmann,
    5. A. Plebani,
    6. O. Sanal,
    7. N. Catalan,
    8. M. Forveille,
    9. R. Dufourcq-Labelouse,
    10. A. Gennery,
    11. et al
    . 2000. Activation-induced cytidine deaminase (AID) deficiency causes the autosomal recessive form of the Hyper-IgM syndrome (HIGM2). Cell 102: 565–575.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Casellas R.,
    2. A. Nussenzweig,
    3. R. Wuerffel,
    4. R. Pelanda,
    5. A. Reichlin,
    6. H. Suh,
    7. X. F. Qin,
    8. E. Besmer,
    9. A. Kenter,
    10. K. Rajewsky,
    11. M. C. Nussenzweig
    . 1998. Ku80 is required for immunoglobulin isotype switching. EMBO J. 17: 2404–2411.
    OpenUrlAbstract/FREE Full Text
    1. Manis J. P.,
    2. Y. Gu,
    3. R. Lansford,
    4. E. Sonoda,
    5. R. Ferrini,
    6. L. Davidson,
    7. K. Rajewsky,
    8. F. W. Alt
    . 1998. Ku70 is required for late B cell development and immunoglobulin heavy chain class switching. J. Exp. Med. 187: 2081–2089.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Pan-Hammarström Q.,
    2. A. M. Jones,
    3. A. Lähdesmäki,
    4. W. Zhou,
    5. R. A. Gatti,
    6. L. Hammarström,
    7. A. R. Gennery,
    8. M. R. Ehrenstein
    . 2005. Impact of DNA ligase IV on nonhomologous end joining pathways during class switch recombination in human cells. J. Exp. Med. 201: 189–194.
    OpenUrlAbstract/FREE Full Text
    1. Rooney S.,
    2. J. Chaudhuri,
    3. F. W. Alt
    . 2004. The role of the non-homologous end-joining pathway in lymphocyte development. Immunol. Rev. 200: 115–131.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Soulas-Sprauel P.,
    2. G. Le Guyader,
    3. P. Rivera-Munoz,
    4. V. Abramowski,
    5. C. Olivier-Martin,
    6. C. Goujet-Zalc,
    7. P. Charneau,
    8. J. P. de Villartay
    . 2007. Role for DNA repair factor XRCC4 in immunoglobulin class switch recombination. J. Exp. Med. 204: 1717–1727.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Boboila C.,
    2. M. Jankovic,
    3. C. T. Yan,
    4. J. H. Wang,
    5. D. R. Wesemann,
    6. T. Zhang,
    7. A. Fazeli,
    8. L. Feldman,
    9. A. Nussenzweig,
    10. M. Nussenzweig,
    11. F. W. Alt
    . 2010. Alternative end-joining catalyzes robust IgH locus deletions and translocations in the combined absence of ligase 4 and Ku70. Proc. Natl. Acad. Sci. USA 107: 3034–3039.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Boboila C.,
    2. C. Yan,
    3. D. R. Wesemann,
    4. M. Jankovic,
    5. J. H. Wang,
    6. J. Manis,
    7. A. Nussenzweig,
    8. M. Nussenzweig,
    9. F. W. Alt
    . 2010. Alternative end-joining catalyzes class switch recombination in the absence of both Ku70 and DNA ligase 4. J. Exp. Med. 207: 417–427.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Han L.,
    2. K. Yu
    . 2008. Altered kinetics of nonhomologous end joining and class switch recombination in ligase IV-deficient B cells. J. Exp. Med. 205: 2745–2753.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Yan C. T.,
    2. C. Boboila,
    3. E. K. Souza,
    4. S. Franco,
    5. T. R. Hickernell,
    6. M. Murphy,
    7. S. Gumaste,
    8. M. Geyer,
    9. A. A. Zarrin,
    10. J. P. Manis,
    11. et al
    . 2007. IgH class switching and translocations use a robust non-classical end-joining pathway. Nature 449: 478–482.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Alt F. W.,
    2. Y. Zhang,
    3. F. L. Meng,
    4. C. Guo,
    5. B. Schwer
    . 2013. Mechanisms of programmed DNA lesions and genomic instability in the immune system. Cell 152: 417–429.
    OpenUrlCrossRefPubMed
  21. ↵
    1. McVey M.,
    2. S. E. Lee
    . 2008. MMEJ repair of double-strand breaks (director’s cut): deleted sequences and alternative endings. Trends Genet. 24: 529–538.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Bennardo N.,
    2. A. Cheng,
    3. N. Huang,
    4. J. M. Stark
    . 2008. Alternative-NHEJ is a mechanistically distinct pathway of mammalian chromosome break repair. PLoS Genet. 4: e1000110.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Yun M. H.,
    2. K. Hiom
    . 2009. CtIP-BRCA1 modulates the choice of DNA double-strand-break repair pathway throughout the cell cycle. Nature 459: 460–463.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Zhang Y.,
    2. M. Gostissa,
    3. D. G. Hildebrand,
    4. M. S. Becker,
    5. C. Boboila,
    6. R. Chiarle,
    7. S. Lewis,
    8. F. W. Alt
    . 2010. The role of mechanistic factors in promoting chromosomal translocations found in lymphoid and other cancers. Adv. Immunol. 106: 93–133.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Zhang Y.,
    2. M. Jasin
    . 2011. An essential role for CtIP in chromosomal translocation formation through an alternative end-joining pathway. Nat. Struct. Mol. Biol. 18: 80–84.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Zhu C.,
    2. K. D. Mills,
    3. D. O. Ferguson,
    4. C. Lee,
    5. J. Manis,
    6. J. Fleming,
    7. Y. Gao,
    8. C. C. Morton,
    9. F. W. Alt
    . 2002. Unrepaired DNA breaks in p53-deficient cells lead to oncogenic gene amplification subsequent to translocations. Cell 109: 811–821.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Lee-Theilen M.,
    2. A. J. Matthews,
    3. D. Kelly,
    4. S. Zheng,
    5. J. Chaudhuri
    . 2011. CtIP promotes microhomology-mediated alternative end joining during class-switch recombination. Nat. Struct. Mol. Biol. 18: 75–79.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Orthwein A.,
    2. A. M. Patenaude,
    3. B. Affar,
    4. A. Lamarre,
    5. J. C. Young,
    6. J. M. Di Noia
    . 2010. Regulation of activation-induced deaminase stability and antibody gene diversification by Hsp90. J. Exp. Med. 207: 2751–2765.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Verdun R. E.,
    2. L. Crabbe,
    3. C. Haggblom,
    4. J. Karlseder
    . 2005. Functional human telomeres are recognized as DNA damage in G2 of the cell cycle. Mol. Cell 20: 551–561.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Kracker S.,
    2. K. Imai,
    3. P. Gardès,
    4. H. D. Ochs,
    5. A. Fischer,
    6. A. H. Durandy
    . 2010. Impaired induction of DNA lesions during immunoglobulin class-switch recombination in humans influences end-joining repair. Proc. Natl. Acad. Sci. USA 107: 22225–22230.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Pan Q.,
    2. C. Petit-Frére,
    3. A. Lähdesmäki,
    4. H. Gregorek,
    5. K. H. Chrzanowska,
    6. L. Hammarström
    . 2002. Alternative end joining during switch recombination in patients with ataxia-telangiectasia. Eur. J. Immunol. 32: 1300–1308.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Doi T.,
    2. L. Kato,
    3. S. Ito,
    4. R. Shinkura,
    5. M. Wei,
    6. H. Nagaoka,
    7. J. Wang,
    8. T. Honjo
    . 2009. The C-terminal region of activation-induced cytidine deaminase is responsible for a recombination function other than DNA cleavage in class switch recombination. Proc. Natl. Acad. Sci. USA 106: 2758–2763.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Reina-San-Martin B.,
    2. S. Difilippantonio,
    3. L. Hanitsch,
    4. R. F. Masilamani,
    5. A. Nussenzweig,
    6. M. C. Nussenzweig
    . 2003. H2AX is required for recombination between immunoglobulin switch regions but not for intra-switch region recombination or somatic hypermutation. J. Exp. Med. 197: 1767–1778.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Nakamura M.,
    2. S. Kondo,
    3. M. Sugai,
    4. M. Nazarea,
    5. S. Imamura,
    6. T. Honjo
    . 1996. High frequency class switching of an IgM+ B lymphoma clone CH12F3 to IgA+ cells. Int. Immunol. 8: 193–201.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Longerich S.,
    2. U. Storb
    . 2005. The contested role of uracil DNA glycosylase in immunoglobulin gene diversification. Trends Genet. 21: 253–256.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Xue K.,
    2. C. Rada,
    3. M. S. Neuberger
    . 2006. The in vivo pattern of AID targeting to immunoglobulin switch regions deduced from mutation spectra in msh2-/- ung-/- mice. J. Exp. Med. 203: 2085–2094.
    OpenUrlAbstract/FREE Full Text
    1. Di Noia J.,
    2. M. S. Neuberger
    . 2002. Altering the pathway of immunoglobulin hypermutation by inhibiting uracil-DNA glycosylase. Nature 419: 43–48.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Rada C.,
    2. J. M. Di Noia,
    3. M. S. Neuberger
    . 2004. Mismatch recognition and uracil excision provide complementary paths to both Ig switching and the A/T-focused phase of somatic mutation. Mol. Cell 16: 163–171.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Ma Y.,
    2. H. Lu,
    3. B. Tippin,
    4. M. F. Goodman,
    5. N. Shimazaki,
    6. O. Koiwai,
    7. C. L. Hsieh,
    8. K. Schwarz,
    9. M. R. Lieber
    . 2004. A biochemically defined system for mammalian nonhomologous DNA end joining. Mol. Cell 16: 701–713.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Petersen S.,
    2. R. Casellas,
    3. B. Reina-San-Martin,
    4. H. T. Chen,
    5. M. J. Difilippantonio,
    6. P. C. Wilson,
    7. L. Hanitsch,
    8. A. Celeste,
    9. M. Muramatsu,
    10. D. R. Pilch,
    11. et al
    . 2001. AID is required to initiate Nbs1/gamma-H2AX focus formation and mutations at sites of class switching. Nature 414: 660–665.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Truong L. N.,
    2. Y. Li,
    3. L. Z. Shi,
    4. P. Y. Hwang,
    5. J. He,
    6. H. Wang,
    7. N. Razavian,
    8. M. W. Berns,
    9. X. Wu
    . 2013. Microhomology-mediated End Joining and Homologous Recombination share the initial end resection step to repair DNA double-strand breaks in mammalian cells. Proc. Natl. Acad. Sci. USA 110: 7720–7725.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Lieber M. R.
    2010. NHEJ and its backup pathways in chromosomal translocations. Nat. Struct. Mol. Biol. 17: 393–395.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Simsek D.,
    2. E. Brunet,
    3. S. Y. Wong,
    4. S. Katyal,
    5. Y. Gao,
    6. P. J. McKinnon,
    7. J. Lou,
    8. L. Zhang,
    9. J. Li,
    10. E. J. Rebar,
    11. et al
    . 2011. DNA ligase III promotes alternative nonhomologous end-joining during chromosomal translocation formation. PLoS Genet. 7: e1002080.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Stavnezer J.,
    2. C. E. Schrader
    . 2006. Mismatch repair converts AID-instigated nicks to double-strand breaks for antibody class-switch recombination. Trends Genet. 22: 23–28.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Eccleston J.,
    2. C. Yan,
    3. K. Yuan,
    4. F. W. Alt,
    5. E. Selsing
    . 2011. Mismatch repair proteins MSH2, MLH1, and EXO1 are important for class-switch recombination events occurring in B cells that lack nonhomologous end joining. J. Immunol. 186: 2336–2343.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Downs J. A.,
    2. M. C. Nussenzweig,
    3. A. Nussenzweig
    . 2007. Chromatin dynamics and the preservation of genetic information. Nature 447: 951–958.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Bothmer A.,
    2. D. F. Robbiani,
    3. M. Di Virgilio,
    4. S. F. Bunting,
    5. I. A. Klein,
    6. N. Feldhahn,
    7. J. Barlow,
    8. H. T. Chen,
    9. D. Bosque,
    10. E. Callen,
    11. et al
    . 2011. Regulation of DNA end joining, resection, and immunoglobulin class switch recombination by 53BP1. Mol. Cell 42: 319–329.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Symington L. S.,
    2. J. Gautier
    . 2011. Double-strand break end resection and repair pathway choice. Annu. Rev. Genet. 45: 247–271.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Bunting S. F.,
    2. E. Callén,
    3. N. Wong,
    4. H. T. Chen,
    5. F. Polato,
    6. A. Gunn,
    7. A. Bothmer,
    8. N. Feldhahn,
    9. O. Fernandez-Capetillo,
    10. L. Cao,
    11. et al
    . 2010. 53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks. Cell 141: 243–254.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Helmink B. A.,
    2. A. T. Tubbs,
    3. Y. Dorsett,
    4. J. J. Bednarski,
    5. L. M. Walker,
    6. Z. Feng,
    7. G. G. Sharma,
    8. P. J. McKinnon,
    9. J. Zhang,
    10. C. H. Bassing,
    11. B. P. Sleckman
    . 2011. H2AX prevents CtIP-mediated DNA end resection and aberrant repair in G1-phase lymphocytes. Nature 469: 245–249.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Bothmer A.,
    2. D. F. Robbiani,
    3. N. Feldhahn,
    4. A. Gazumyan,
    5. A. Nussenzweig,
    6. M. C. Nussenzweig
    . 2010. 53BP1 regulates DNA resection and the choice between classical and alternative end joining during class switch recombination. J. Exp. Med. 207: 855–865.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Schrader C. E.,
    2. J. E. Guikema,
    3. E. K. Linehan,
    4. E. Selsing,
    5. J. Stavnezer
    . 2007. Activation-induced cytidine deaminase-dependent DNA breaks in class switch recombination occur during G1 phase of the cell cycle and depend upon mismatch repair. J. Immunol. 179: 6064–6071.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Min I. M.,
    2. C. E. Schrader,
    3. J. Vardo,
    4. T. M. Luby,
    5. N. D’Avirro,
    6. J. Stavnezer,
    7. E. Selsing
    . 2003. The Smu tandem repeat region is critical for Ig isotype switching in the absence of Msh2. Immunity 19: 515–524.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Kunkel T. A.,
    2. D. A. Erie
    . 2005. DNA mismatch repair. Annu. Rev. Biochem. 74: 681–710.
    OpenUrlCrossRefPubMed
  55. ↵
    1. Schaetzlein S.,
    2. R. Chahwan,
    3. E. Avdievich,
    4. S. Roa,
    5. K. Wei,
    6. R. L. Eoff,
    7. R. S. Sellers,
    8. A. B. Clark,
    9. T. A. Kunkel,
    10. M. D. Scharff,
    11. W. Edelmann
    . 2013. Mammalian Exo1 encodes both structural and catalytic functions that play distinct roles in essential biological processes. Proc. Natl. Acad. Sci. USA 110: E2470–E2479.
    OpenUrlAbstract/FREE Full Text
  56. ↵
    1. Kousholt A. N.,
    2. K. Fugger,
    3. S. Hoffmann,
    4. B. D. Larsen,
    5. T. Menzel,
    6. A. A. Sartori,
    7. C. S. Sørensen
    . 2012. CtIP-dependent DNA resection is required for DNA damage checkpoint maintenance but not initiation. J. Cell Biol. 197: 869–876.
    OpenUrlAbstract/FREE Full Text
  57. ↵
    1. Bardwell P. D.,
    2. C. J. Woo,
    3. K. Wei,
    4. Z. Li,
    5. A. Martin,
    6. S. Z. Sack,
    7. T. Parris,
    8. W. Edelmann,
    9. M. D. Scharff
    . 2004. Altered somatic hypermutation and reduced class-switch recombination in exonuclease 1-mutant mice. Nat. Immunol. 5: 224–229.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Schrader C. E.,
    2. J. Vardo,
    3. J. Stavnezer
    . 2002. Role for mismatch repair proteins Msh2, Mlh1, and Pms2 in immunoglobulin class switching shown by sequence analysis of recombination junctions. J. Exp. Med. 195: 367–373.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    1. Boboila C.,
    2. F. W. Alt,
    3. B. Schwer
    . 2012. Classical and alternative end-joining pathways for repair of lymphocyte-specific and general DNA double-strand breaks. Adv. Immunol. 116: 1–49.
    OpenUrlCrossRefPubMed
  60. ↵
    1. Hackney J. A.,
    2. S. Misaghi,
    3. K. Senger,
    4. C. Garris,
    5. Y. Sun,
    6. M. N. Lorenzo,
    7. A. A. Zarrin
    . 2009. DNA targets of AID evolutionary link between antibody somatic hypermutation and class switch recombination. Adv. Immunol. 101: 163–189.
    OpenUrlCrossRefPubMed
  61. ↵
    1. Du L.,
    2. M. van der Burg,
    3. S. W. Popov,
    4. A. Kotnis,
    5. J. J. van Dongen,
    6. A. R. Gennery,
    7. Q. Pan-Hammarström
    . 2008. Involvement of Artemis in nonhomologous end-joining during immunoglobulin class switch recombination. J. Exp. Med. 205: 3031–3040.
    OpenUrlAbstract/FREE Full Text
    1. Péron S.,
    2. A. Metin,
    3. P. Gardès,
    4. M. A. Alyanakian,
    5. E. Sheridan,
    6. C. P. Kratz,
    7. A. Fischer,
    8. A. Durandy
    . 2008. Human PMS2 deficiency is associated with impaired immunoglobulin class switch recombination. J. Exp. Med. 205: 2465–2472.
    OpenUrlAbstract/FREE Full Text
  63. ↵
    1. Sekine H.,
    2. R. C. Ferreira,
    3. Q. Pan-Hammarström,
    4. R. R. Graham,
    5. B. Ziemba,
    6. S. S. de Vries,
    7. J. Liu,
    8. K. Hippen,
    9. T. Koeuth,
    10. W. Ortmann,
    11. et al
    . 2007. Role for Msh5 in the regulation of Ig class switch recombination. Proc. Natl. Acad. Sci. USA 104: 7193–7198.
    OpenUrlAbstract/FREE Full Text
  64. ↵
    1. Bothmer A.,
    2. P. C. Rommel,
    3. A. Gazumyan,
    4. F. Polato,
    5. C. R. Reczek,
    6. M. F. Muellenbeck,
    7. S. Schaetzlein,
    8. W. Edelmann,
    9. P. L. Chen,
    10. R. M. Brosh Jr..,
    11. et al
    . 2013. Mechanism of DNA resection during intrachromosomal recombination and immunoglobulin class switching. J. Exp. Med. 210: 115–123.
    OpenUrlAbstract/FREE Full Text
  65. ↵
    1. Martomo S. A.,
    2. W. W. Yang,
    3. P. J. Gearhart
    . 2004. A role for Msh6 but not Msh3 in somatic hypermutation and class switch recombination. J. Exp. Med. 200: 61–68.
    OpenUrlAbstract/FREE Full Text
  66. ↵
    1. Schrader C. E.,
    2. W. Edelmann,
    3. R. Kucherlapati,
    4. J. Stavnezer
    . 1999. Reduced isotype switching in splenic B cells from mice deficient in mismatch repair enzymes. J. Exp. Med. 190: 323–330.
    OpenUrlAbstract/FREE Full Text
  67. ↵
    1. Ranjit S.,
    2. L. Khair,
    3. E. K. Linehan,
    4. A. J. Ucher,
    5. M. Chakrabarti,
    6. C. E. Schrader,
    7. J. Stavnezer
    . 2011. AID binds cooperatively with UNG and Msh2-Msh6 to Ig switch regions dependent upon the AID C terminus. J. Immunol. 187: 2464–2475.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section