• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Feb 24, 2009; 106(8): 2758–2763.
Published online Feb 6, 2009. doi:  10.1073/pnas.0813253106
PMCID: PMC2650339

The C-terminal region of activation-induced cytidine deaminase is responsible for a recombination function other than DNA cleavage in class switch recombination


Activation-induced cytidine deaminase (AID) is an essential factor for the class switch recombination (CSR) and somatic hypermutation (SHM) of Ig genes. CSR and SHM are initiated by AID-induced DNA breaks in the S and V regions, respectively. Because truncation or frame-shift mutations at the carboxyl (C)-terminus of AID abolishes CSR but not SHM, the C-terminal region of AID likely is required for the targeting of DNA breaks in the S region. To test this hypothesis, we determined the precise location and relative amounts of AID-induced DNA cleavage using an in situ DNA end-labeling method. We established CH12F3–2 cell transfectants expressing the estrogen receptor (ER) fused with wild-type (WT) AID or a deletion mutant lacking the C-terminal 16 aa, JP8Bdel. We found that AID-ER, but not JP8Bdel-ER, caused a CSR to IgA from the addition of 4-hydroxy tamoxifen. In contrast, both WT AID and JP8Bdel induced DNA breaks in both the V and S regions. In addition, JP8Bdel enhanced c-myc/IgH translocations. Our findings indicate that the C-terminal domain of AID is not required for S-region DNA breaks but is required for S-region recombination after DNA cleavage. Therefore, AID does not distinguish between the V and S regions for cleavage, but carries another function specific to CSR.

Keywords: C-terminal deletion mutant, chromosomal translocation, synapse formation, target specificity

Activation-induced cytidine deaminase (AID) is essential for the generation of antibody memory, which depends on 2 genetic alterations of Ig genes: somatic hypermutation (SHM) and class switch recombination (CSR) (1, 2). AID induces DNA cleavage in the V and S regions to trigger SHM and CSR, respectively (3, 4). The molecular mechanism of DNA cleavage by AID has been debated extensively. The RNA-editing hypothesis assumes that AID edits unknown mRNAs to generate a novel mRNA encoding endonucleases or its cofactor that cleave target DNA (5). In contrast, the DNA deamination hypothesis assumes that AID directly targets DNA and deaminates C to U, generating U–G mismatches that are subsequently recognized by the base excision repair pathway (6). By this hypothesis, the repair mechanism would create phosphodiester bond cleavages as intermediates that are proposed to trigger SHM or CSR. Lines of evidence support each hypothesis, but neither has been proven.

The other question concerning AID function is how, with only 198 residues, it can distinguish between the V and S regions to initiate SHM or CSR appropriately. The RNA-editing hypothesis assumes that AID edits 2 different mRNAs, each specific for 1 of the 2 regions. The evidence supporting this hypothesis comes from studies on AID mutants in which alterations in the carboxyl (C)-terminal region cause a specific loss of CSR (7, 8). In addition, point mutations in the amino (N)-terminal region of AID drastically reduce SHM, and also compromise CSR to some extent (9). The DNA deamination hypothesis does not provide a formal explanation for target specificity but assumes that some cofactors are required to guide AID specifically to the V or S region.

Gaining insight into the mechanism of AID's target specificity requires a reliable cleavage assay. Previously, DNA cleavage was measured by γH2AX focus formation, which occurs immediately after a double-strand break (DSB) in DNA (10). But this assay is unlikely to reveal the precise target specificity between the V and S regions, because in yeast, phosphorylation of the H2AX histone occurs over large stretches of DNA (11), sometimes extending into the megabases. If this were also the case in mammals, then the γH2AX method could not distinguish the V from the Sμ region, because these regions are <10 kb apart. Consequently, we devised a specific cleaved-end–labeling method that uses the incorporation of biotin-labeled dUTP by terminal deoxynucleotidyl transferase (TdT) to permit the capture of the cleaved DNA on streptavidin beads (12). These collected cleaved ends are then analyzed by PCR to distinguish between cleavage of the V and S regions.

The C-terminal region of AID contains a highly conserved nuclear export signal (NES) (13, 14). Truncation of the C-terminal region induces the accumulation of AID in the nucleus and the loss of CSR activity, but increases SHM activity (13, 14). Although this truncation clearly affects the export of AID from the nucleus, whether the C-terminus of AID is required only for nuclear export or also for other functions has not been determined. In an accompanying paper (15), we show that the C-terminal region of AID is required for its association with poly(A)+ RNA, and that thus AID is likely to associate with poly(A)+ RNA and then to be exported along with the mRNA from the nucleus to the cytoplasm. In the present work, we analyzed the role of the C-terminus of AID by mutagenesis and a DNA cleavage assay. We found that C-terminally truncated AID can cleave both the V and S regions, but although the mutant can still mediate SHM, it cannot mediate CSR. We propose that a novel function of AID associated with its C-terminus is required for recombination of the cleaved S region.


Requirement for the Nuclear Export of CSR.

To explore whether or not the nuclear export of AID is required for CSR, we tested the effect of leptomycin B (LMB), a specific inhibitor of nuclear export factor CRM1, on CSR in AID-expressing cells. A fusion protein of AID and the hormone-binding domain of the estrogen receptor (AID-ER) can be activated quickly by adding the hormone analogue 4-hydroxy tamoxifen (4-OHT) to induce CSR within 3–6 h, as assessed by digestion circularization PCR (DC-PCR) (16). We took advantage of this system to avoid prolonged incubation with LMB, which has cytotoxic effects caused by inhibition of the nuclear export of proteins essential for many cellular functions, such as cell cycle progression.

AID-ER was retrovirally introduced into Aicda/ B cells, and CSR was induced by adding 4-OHT 1 day after the infection. Between 3 and 6 h later, CSR between the Sμ and Sγ1 regions was detected by DC-PCR. The addition of LMB 1 h before the addition of 4-OHT completely blocked CSR (Fig. 1A); however, the addition of LMB simultaneously with or 1 h after 4-OHT showed limited inhibitory effects on CSR. Thus, only a short period of AID activation was sufficient to counteract the inhibitory effects of LMB on CSR. This is reminiscent of our previous findings on CSR, in which the inhibitory effects of cycloheximide were overcome by a 1-h preactivation of AID-ER (16). We confirmed that LMB did not affect the germline transcription of Sμ and Sγ1 (Fig. 1B). The results suggest that the nuclear export of AID is required for CSR.

Fig. 1.
LMB inhibited CSR induced by AID-ER in Aicda−/− B cells. A retroviral expression construct for AID-ER was introduced into Aicda−/− B cells. One day later, 1 μM 4-OHT was added to induce CSR at time 0. LMB (10 ng/mL) ...

Point Mutations in the NES Reduce the CSR Activity of AID.

To confirm the role of the nuclear export of AID in the CSR activity of AID, we evaluated the correlation between the nuclear localization and CSR activity of a series of C-terminal point mutants of AID, which included mutations in the conserved NES motif shown in Fig. 2A. To examine the cytoplasmic localization of each mutant, we generated them as fusion proteins with GFP at the C-terminus, and then introduced these constructs into NIH 3T3 cells using a retroviral vector. Wild-type (WT) AID-GFP was localized mostly to the cytoplasm, as reported previously (Fig. 2B). As expected, the mutations that affected the NES consensus amino acids (L189A, F193A, L196A, and L198A) changed the predominant localization of AID from the cytoplasm to the nucleus, whereas all of the other point mutants remained predominantly in the cytoplasm.

Fig. 2.
Cytoplasmic localization of AID was required for CSR but not for SHM. (A) Alignment of the C-terminal part of AID from several animals. Red characters indicate NES consensus amino acids. Numbers indicate the amino acid position. (B) GFP fusion AID constructs ...

To monitor the CSR and SHM activity, we constructed another retroviral vector that expressed WT or mutant AID together with CD8α. We infected Aicda/ spleen cells with these retroviruses and, 3 days later, analyzed CSR to IgG1 by FACS. All of the NES mutants drastically reduced the CSR activity compared with WT AID (Fig. 2C). In contrast, mutations outside the NES motif residues (L172A, R190A, D191A, A192G, R194A, T195A, and G197A) exhibited only slight or milder effects on CSR. The loss of CSR was more drastic in a L172A/G197A double mutant. This mutant is similar to a C-terminal insertion mutant P20, which loses CSR activity with intact NES (7).

To test the effects on SHM, we introduced these constructs into NTZ-PI cells (i.e., NIH 3T3 cells carrying a GFP mutation reporter construct with a stop codon) (17). We assessed SHM activity by assaying the reversion to WT GFP as well as by sequence analysis 3 days after the infection. In contrast to the effects on CSR, SHM activity was enhanced in all of the NES mutants of AID, in agreement with previous reports (7, 8) [Fig. 2D and supporting information (SI) Table S1]. These results indicate that CSR, but not SHM, is dependent on the nuclear export of AID. This observation is consistent with the conclusion reached previously by us and another group that AID regulates SHM and CSR differentially (79).

NES Mutants Induce DSB in the S Region.

The C-terminal function of AID purportedly is important in targeting DNA breakage to the S region. To test this hypothesis directly, we evaluated the DNA breakage activity of AID mutants in the S region by in situ DNA end-labeling using TdT and biotinylated dUTP (12). Because apoptosis-associated DNA breaks in primary B cells increased the background in the DNA break assay, we established mutant AID-ER–expressing CH12F3–2 cells, which, when expressing WT AID-ER, efficiently performed CSR to IgA without cytokine stimulation after the addition of 4-OHT.

In addition to the point mutations in the NES coding sequence, we previously made and analyzed AID-ER mutants with C-terminal truncation (ΔC) or insertion (JP8Bdel or P20, respectively), N-terminal single replacement (G23S), and catalytic center triple replacements (KSS) (4, 7, 9, 18). We previously reported that JP8Bdel and P20 are CSR-defective, whereas G23S is more deficient in SHM than in CSR activity (7, 9). We used these mutants in our current analysis. Here we found that the protein expression levels of our AID-ER mutants were comparable to the level of WT AID-ER (Fig. 3A), and that WT AID-ER demonstrated IgA CSR in response to 4-OHT. In contrast, all of the AID-ER mutants had much lower levels of IgA CSR (Fig. 3B), and KSS-ER had no activity at all.

Fig. 3.
NES AID mutant-induced S-region DNA breaks. (A) AID-ER mutants were introduced into CH12F3–2 cells with a retrovirus vector. Protein expression levels of AID-ER mutants were evaluated by Western blot analysis with an anti-ER antibody. As an internal ...

We next examined DNA breakage in the Sμ region by various mutants of AID-ER 3 h hours after adding 4-OHT (Fig. 3C). Unexpectedly, the NES mutants (JP8Bdel-ER and L196A-ER) induced DNA breaks in both the Sμ and V regions even more strongly than WT AID-ER did; almost no breaks were observed in the β2 microglobulin gene. The mutants P20, G23S, and L172A/G197A, which retained their NES activity (9, 13), exhibited much lower levels of DNA cleavage compared with WT AID-ER. The catalytically inactive mutant, KSS-ER, did not cause DNA cleavage (Fig. 3C). To exclude any possibility that endogenous AID was induced by 4-OHT, we monitored the endogenous AID transcript by real-time PCR and found no increase in endogenous AID expression within 3 h (Fig. S1). Thus, these results clearly indicate that the DNA cleavage and CSR activities of the AID mutants are not always correlated.

It is believed that SHM can be initiated from single-strand nicks, whereas CSR requires DSB. We deemed it possible that NES mutants would preferentially induce single-strand nicks in the S region in addition to the V region, because the NES mutants exhibited enhanced SHM activities. Although the labeling efficiency of nicked ends by TdT is low (19), we could not exclude the possibility that most of the labeled DNA ends induced by NES mutants were nicked ends. To further explore this possibility, we specifically labeled DSB ends with a biotinylated double-strand DNA linker by blunt-end DNA ligation. We found that JP8Bdel-ER induced DSB in the Sμ region to a level comparable with WT AID-ER, whereas KSS-ER showed no DNA breakage activity above the background (Fig. 3D). These results clearly demonstrate that C-terminal mutants of AID (JP8Bdel and L196A) are strongly active in DNA cleavage in the V and S regions, although they are defective in CSR.

CSR Defect of the ΔC Mutant Is Not Due To Excess DNA Breakage.

NES mutants showed higher breakage activities in both the S and V regions, suggesting that excess DNA breaks might interfere with recombination. To explore this possibility, we titrated the JP8Bdel-ER activity by changing the 4-OHT concentration. Increasing doses of 4-OHT gradually enhanced DNA cleavage in the Sμ and Sα regions by both AID-ER and JP8Bdel-ER without affecting the target specificity; the Sγ1 region and β2 microglobulin gene were not cleaved (Fig. 4A). Although DNA cleavage activity was reduced at lower 4-OHT concentrations, the CSR defect of JP8Bdel-ER was not rescued (Fig. 4B); in fact, we detected a very weak CSR activity of JP8Bdel-ER at higher, rather than lower, 4-OHT concentrations. These results clearly indicate that JP8Bdel is defective in CSR at a step that follows DNA breakage in the S region.

Fig. 4.
Effect of 4-OHT titration on CSR and DNA breakage. (A) CH12 AID-ER or JP8Bdel-ER cells were treated with the indicated concentrations of 4-OHT for 3 h. Then DNA breakage assays were performed for Sμ, Sγ1, Sα, and β2m. ( ...

IgM Loss Is Associated With CSR Defects by NES Mutants.

We explored the possibility that the C-terminal region of AID is involved in the repair phase that follows DNA breakage. If DNA cleavage in the Sμ region is not repaired appropriately, then B cells should lose their surface IgM expression. Consequently, we examined the surface expression of IgM in addition to IgA 2 days after the addition of 4-OHT. We found that IgMIgA double-negative (IgM/IgA-negative) cells were more abundant in the total population of JP8Bdel-ER–expressing CH12F3–2 cells than in that of AID-ER–expressing cells (Fig. 5A). Although the percentage of IgA-positive cells was much lower in the JP8Bdel-ER– and L196A-ER–expressing populations than in the WT AID-ER–expressing population, the total percentage of IgA-positive plus IgM/IgA-negative cells was comparable to that of the AID-ER–expressing population (Fig. 5A and B). CH12F3–2 cells expressing P20-ER or G23S-ER showed a slightly increased prevalence of IgA+ cells but not of IgM/IgA-negative cells, in agreement with its weak DNA cleavage activity. The relative abundance of IgA plus IgM/IgA-negative cells was almost completely coincident with the DNA breakage activity of the AID mutants.

Fig. 5.
C-terminal mutants of AID-induced IgM/IgA-negative cells. (A) CH12 cells expressing AID-ER mutants were cultured with or without 1 μM 4-OHT for 2 days. IgA and IgM expression were analyzed by flow cytometry. IgM and IgA expression of 4-OHT–treated ...

The percentage of IgM+ cells also was reduced more drastically in Aicda/ spleen B cells expressing JP8Bdel than in the same cells expressing WT AID (Fig. S2). The loss of surface IgM without CSR to IgA was not due to the internalization of surface Ig or CSR to other isotypes, because the IgM cells expressing JP8Bdel-ER also lost intracellular IgM and kappa light chain expression (Figs. S3 and S4). These results suggest that the aberrations owing to the JP8Bdel mutant occurred after its DNA cleavage of the S region.

Next, we explored whether the expression of the Ig gene is disturbed by vigorous SHM in the VH region. We sequenced the VH region of surface Ig-negative cells after the 4-OHT treatment of CH12F3–2 cells expressing AID-ER or JP8Bdel-ER. As we reported previously (20), the mutation frequencies in the VH region in AID-ER– or JP8Bdel-ER–expressing CH12 cells were 2 orders of magnitude lower than in normal SHM (Table S2). This indicates that SHM of the V region is not the cause of surface IgM loss in these cells.

Increased Aberrant Recombination by the ΔC Mutant.

Because S region cleavage by AID is known to induce chromosomal translocations to other loci, including c-myc (21), we examined the interchromosomal translocations between c-myc and Ig. We analyzed c-myc/Ig translocation by nested PCR 24 h after 4-OHT treatment in CH12F3–2 cells transfected with AID-ER or JP8Bdel-ER. We tested 20 pools of DNA each from 7.5 × 104 cells to compare the relative frequency of c-myc/Sμ translocation. Each band was confirmed to be a c-myc/Sμ recombinant by sequencing. The translocation bands appeared only after 4-OHT was added. The frequency of translocation was 12-fold greater in JP8Bdel-ER–expressing cells than in WT AID-ER–expressing cells (12 per 1.5 × 106 vs. 1 per 1.5 × 106) (Fig. 6). These results suggest that the C-terminal region of AID is required for efficient recombination between S regions, and that it thereby suppresses aberrant recombination.

Fig. 6.
JP8Bdel-ER induces enhanced c-myc/Ig translocation. CH12 cells expressing AID-ER or JP8Bdel-ER were stimulated with or without 1 μM 4-OHT for 2 days. Translocations between the c-myc and Ig genes were detected by nested PCR. Arrowheads indicate ...


Because SHM and CSR are independent events, how the 198-residue AID can differentially introduce point mutations in the V region and recombination in the S region is unclear. It has been postulated that the C-terminal region of AID is required for targeting DNA cleavage to the S region, because C-terminal mutations of AID specifically abolish the CSR activity (7, 8). This assumption led to the proposal that the selection of SHM or CSR by AID is determined by the DNA cleavage of the specific target, the V or S region.

In this study, we found that the C-terminal NES mutants of AID induced efficient DSB in the S region despite having a severe defect in CSR. WT AID and C-terminal NES mutants also introduced DNA breaks in the V region of CH12 cells, although the frequency of SHM in CH12 cells was far less than that in germinal center B cells, as reported previously (20). These results clearly indicate that DNA cleavage target specificity is not the determinant of the SHM or CSR activity of AID; rather, one or more subsequent stages, such as repair or recombination, are the critical determinants of the DNA alteration products.

Our findings also indicate that AID cannot distinguish the S region from the V region in selecting a cleavage target. The final outcome of DNA SHM versus CSR is determined by AID, but this occurs after DNA cleavage, which also is induced by AID. This finding is generally consistent with previous reports that the targeting specificity of DNA breakage by AID is not very strict (22, 23) and that another NES mutant introduces mutations in the Sμ region of B cells (8). Thus, we conclude that AID has at least two functions—DNA cleavage and recombination— and that SHM versus CSR is not determined by cleavage targeting.

The C-terminal region of AID is responsible for at least 3 AID functions: (i) nuclear export of AID, (ii) recombination after S region cleavage, and (iii) association with poly(A)+ RNA (15). In view of CSR's dependence on new protein synthesis, we can speculate that AID forms an mRNA protein complex through its C-terminal region and edits the mRNA in the complex, after which the mRNA is exported to the cytoplasm for translation. The translated protein may be involved in the recombination reaction between the cleaved S regions. The C-terminal residues involved in nuclear export and association with mRNA appear to overlap, at least partially; however, the fact that some mutations in the C-terminal region of AID (P20 and L172A/G197A) cause reduction of CSR without changing cytoplasmic localization suggests that these residues may be specifically involved in mRNA association.

We found fewer DNA breaks in the V region than in the S region, regardless of AID or its mutants. This conclusion is based on the comparison of target signal versus input ratios for Sμ and the V region (Figs. 3C and S5), and is consistent with a previous report showing that the V region of the light chain gene is not associated with γH2AX focus formation in GC B cells (24). Furthermore, DNA cleavage by AID is dependent on transcription of the target. The S region can be replaced by DNA sequences with palindromic sequences (25), which can easily form a secondary structure when the R-loop is formed during transcription. Such a structure could be recognized as a target of single-strand cleavage by an endonuclease. Computer analysis predicts that the S-region sequences form secondary structures much more easily than the V-region sequences do (1). Nicking of the S region may occur more frequently at different positions on both DNA strands, giving rise to DSB with staggered ends. In contrast, nicking of the V region likely is less frequent, and most of the cleaved ends may remain as nicks.

G23S was originally reported to be specifically defective in SHM. Although the overexpression of G23S by a retroviral vector shows comparable CSR activity to that of WT AID in Aicda/ B cells (9), G23S-ER exhibited reduced CSR activity compared with WT AID in our CH12 cells. G23S-ER also exhibited lower DNA breakage activity than WT AID in both the V and S regions. These results suggest that SHM-specific mutants, such as G23S, are hypomorphic for DNA cleavage activity, and that reduced DNA cleavage affects SHM more strongly because the V region is a weaker substrate. Because SHM is not induced in CH12 cells or in LPS-stimulated spleen cells, which express a large amount of AID (20, 26), it is unlikely that AID itself has an additional function as an SHM regulator.

We found that C-terminal NES mutants of AID generate a large proportion of cells without surface Ig and cause aberrant recombination, such as c-myc/Igh translocation, instead of the correct S-S recombination. Translocation between c-myc and Sμ alone cannot explain the frequent loss of IgM. Because AID cleaves many specific genes (22, 23), it is not surprising that various other types of translocation occur, resulting in frequent loss of IgM. In fact, most of the JP8Bdel-expressing CH12F3–2 cells died within 3 days of the 4-OHT addition (Fig. S6).

The 53BP1 and Mre11-Rad50-Nbs1 (MRN) complex is thought to be important for holding the 2 DNA ends at the DSB site, because 53BP1-null and hypomorphic mutations of genes encoding the MRN complex proteins show aberrant chromosomal translocations. 53BP1-, H2AX-, and ATM-deficient B cells and Nbs1 conditional knockout B cells demonstrate reduced CSR and more frequent chromosomal aberrations after the induction of AID (27). Nbs1- and ATM-deficient B cells also demonstrate enhanced c-myc/IgH translocation in response to switch stimulation. These B cells exhibit defective anchoring of DNA break ends and a loss of chromosomal ends after AID activation. Similarly, the C-terminus of AID may be required for holding AID-induced DNA break ends, to facilitate correct CSR and simultaneously suppress aberrant interchromosomal recombination after DNA cleavage. The involvement of AID in a repair step or in S–S synapsis formation has been proposed (8, 28). The recombination function of AID is dependent on its nuclear export (7, 8), its translation (16), and formation of a complex with poly(A)+ RNA (15). It was reported that DNA breaks in the Sμ and Sγ1 regions induced by yeast endonuclease I-Sce-I are sufficient for CSR even without AID, at a low frequency (29). In fact, JP8Bdel exhibited significant levels of CSR, albeit much lower than in WT AID (Fig. 4B). However, a normal CSR level requires an additional supportive function of AID that depends on AID's C-terminal region.

All of the NES consensus amino acid mutants, which localized predominantly to the nucleus, exhibited lower CSR activity than WT AID. CSR was inhibited by the addition of LMB 1 h before AID activation. In contrast to their CSR activity, all of the NES mutants demonstrated higher SHM activity than WT AID. AID's C-terminal region, which is involved in mRNA association and nuclear export of AID, is not required for DNA cleavage of the S and V regions. This finding does not necessarily contradict the RNA-editing model, because AID may edit non- poly(A)+ RNA, such as micro RNA that does not require nuclear export for translation in the cytoplasm. In theory, C-terminally truncated AID can cleave the V and S regions by DNA deamination; however, the recent finding that AID mutant N51A without DNA deamination activity retains CSR activity makes the involvement of DNA deamination in CSR unlikely (18). In addition, contradictory to the DNA deamination hypothesis, U removal activity of UNG is not required for CSR (10).

In summary, AID has at least 2 different functions: DNA cleavage, which occurs without the C-terminal region, and a recombination-specific function that requires the C-terminal region responsible for the nuclear export of AID by the NES motif and the association with polyA+ RNA.

Materials and Methods

Construction of Vectors and Cell Lines.

To make AID-GFP fusion constructs, human AID and its mutants were amplified by PCR and cloned into the EcoRI/BamHI sites of the pFB-GFP vector, as described previously (13). To make pFB-IRES-CD8, the NcoI/NotI fragment of the mouse CD8α coding sequence was amplified by PCR and cloned into the NcoI/NotI sites of pFB-IRES-GFP instead of GFP (16). The SalI/BamHI fragment of hAID and its mutants were cloned into the SalI/BamHI sites of pFB-IRES-CD8. pFB AID-ERpuro and pFB JP8Bdel-ERpuro were described previously (4). For pFB L196A-ERpuro and P20-ERpuro, the AID portion of pFB AID-ERpuro was replaced with mutants carrying the L196A and P20 sequence changes. To make the CH12AID-ER and AID-ER mutants, the pFB AID-ERpuro and mutant AID-ERpuro were introduced into CH12F3–2 cells through retroviral gene transfer and selected with puromycin for 1 week.

CSR and SHM.

WT and mutant AID-IRES-CD8 were introduced into Aicda/ B cells by retroviral infection. After being infected, the B cells were cultured with 25 μg/mL of LPS and 15 ng/mL of IL-4 for 3 days, and then analyzed for IgG1 expression by flow cytometry. WT and mutant pFB AID-IRES CD8 were introduced into NTZ-PI cells. After 3 days, the GFP-expressing cells were quantified by flow cytometry. The EGFP sequence was amplified from the NTZ-PI cells' genomic DNA and cloned into the pGEM-T Easy vector (Promega). The sequence of each construct was confirmed by sequencing.


DC-PCR was performed as described previously (16). C-myc/IgH translocations were detected by nested PCR as described previously (21).

Biotin-Labeling DNA Break Assay.

The biotin-labeling DNA break assay was performed as described previously (12), with a slight modification. To remove dead cells, 5 million cells were layered on 40% and 70% Percoll layers and centrifuged at 1,500 g for 20 min at room temperature. Live cells were collected from the interface between the 40% and 70% Percoll. Cells were washed in cold PBS and then fixed with 1 mL of fixation buffer (3 g of bronopol, 3 g of diazolidinyl urea, 1.2 g of zinc sulfate heptahydrate, 0.29 g of sodium citrate dihydrate and 50 mM EDTA in 100 mL) for 15 min at room temperature. Then the cells were washed again in cold PBS and sequentially resuspended into cold buffer A [0.25% TritonX-100, 10 mM EDTA, 10 mM Hepes (pH 6.5)], cold buffer B [200 mM NaCl, 1 mM EDTA, 10 mM Hepes (pH 6.5)]. Nuclei were permeabilized with buffer C [100 mM Tris-HCl (pH 7.4), 50 mM EDTA, 1% TritonX-100] for 30 min on ice. Cells were sequentially washed in cold PBS and 1× TdT buffer, then resuspended into 100 μL of TdT buffer with 3 μL of 1 mM biotin-16-dUTP (Roche) and 60 U of TdT (New England Biolabs), then incubated for 1 h at 37 °C. After that, the cells were washed with buffer D [100 mM Tris-HCl (pH 7.4), 150 mM NaCl], resuspended into lysis buffer [10 mM EDTA, 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.1% SDS, 0.2 mg/mL of proteinase K], and then incubated overnight at 56 °C. Genomic DNA was isolated by phenol/chloroform extraction. Genomic DNA (20 μg) was digested with HindIII overnight at 37 °C. Biotinylated fragments were captured with 20 μL of streptavidin magnetic beads (Streptavidin MagneSphere paramagnetic particles, Promega). The particles were washed with 300 μL of TE 3 times and then resuspended into 20 μL of 10 mM Tris-HCl (pH 8.0). Three μL of the beads was used for PCR, and 6 ng of the genomic DNA of each sample was used as input. PCR was initiated by a denaturing step of 95 °C for 5 min, followed by 32 cycles of PCR (95 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s). Primer sequences are provided in SI Materials and Methods.

Supplementary Material

Supporting Information:


This research was supported by a Grant-in Aid for Specially Promoted Research (17002015) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. We thank A. Kawamura for technical support and Y. Shiraki and T. Kanda for help with manuscript preparation.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0813253106/DCSupplemental.


1. Kinoshita K, Honjo T. Linking class-switch recombination with somatic hypermutation. Nat Rev Mol Cell Biol. 2001;2:493–503. [PubMed]
2. Honjo T, Kinoshita K, Muramatsu M. Molecular mechanism of class switch recombination: linkage with somatic hypermutation. Annu Rev Immunol. 2002;20:165–196. [PubMed]
3. Petersen S, et al. AID is required to initiate Nbs1/gamma-H2AX focus formation and mutations at sites of class switching. Nature. 2001;414:660–665. [PubMed]
4. Nagaoka H, Ito S, Muramatsu M, Nakata M, Honjo T. DNA cleavage in immunoglobulin somatic hypermutation depends on de novo protein synthesis but not on uracil DNA glycosylase. Proc Natl Acad Sci U S A. 2005;102:2022–2027. [PMC free article] [PubMed]
5. Muramatsu M, et al. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA-editing enzyme. Cell. 2000;102:553–563. [PubMed]
6. Petersen-Mahrt SK, Harris RS, Neuberger MS. AID mutates E. coli, suggesting a DNA deamination mechanism for antibody diversification. Nature. 2002;418:99–103. [PubMed]
7. Ta VT, et al. AID mutant analyses indicate requirement for class-switch–specific cofactors. Nat Immunol. 2003;4:843–848. [PubMed]
8. Barreto V, Reina-San-Martin B, Ramiro AR, McBride KM, Nussenzweig MC. C-terminal deletion of AID uncouples class switch recombination from somatic hypermutation and gene conversion. Mol Cell. 2003;12:501–508. [PubMed]
9. Shinkura R, et al. Separate domains of AID are required for somatic hypermutation and class-switch recombination. Nat Immunol. 2004;5:707–712. [PubMed]
10. Begum NA, et al. Uracil DNA glycosylase activity is dispensable for immunoglobulin class switch. Science. 2004;305:1160–1163. [PubMed]
11. Shroff R, et al. Distribution and dynamics of chromatin modification induced by a defined DNA double-strand break. Curr Biol. 2004;14:1703–1711. [PubMed]
12. Ju BG, et al. A topoisomerase IIbeta-mediated dsDNA break required for regulated transcription. Science. 2006;312:1798–1802. [PubMed]
13. Ito S, et al. Activation-induced cytidine deaminase shuttles between nucleus and cytoplasm-like apolipoprotein B mRNA-editing catalytic polypeptide 1. Proc Natl Acad Sci U S A. 2004;101:1975–1980. [PMC free article] [PubMed]
14. McBride KM, Barreto V, Ramiro AR, Stavropoulos P, Nussenzweig MC. Somatic hypermutation is limited by CRM1-dependent nuclear export of activation-induced deaminase. J Exp Med. 2004;199:1235–1244. [PMC free article] [PubMed]
15. Nonaka T, et al. Requirement of the carboxy-terminal domain of AID for its mRNA complex formation in vivo. Proc Natl Acad Sci U S A. 2009 in press. [PMC free article] [PubMed]
16. Doi T, Kinoshita K, Ikegawa M, Muramatsu M, Honjo T. De novo protein synthesis is required for the activation-induced cytidine deaminase function in class-switch recombination. Proc Natl Acad Sci U S A. 2003;100:2634–2638. [PMC free article] [PubMed]
17. Yoshikawa K, et al. AID enzyme-induced hypermutation in an actively transcribed gene in fibroblasts. Science. 2002;296:2033–2036. [PubMed]
18. Shivarov V, Shinkura R, Honjo T. Dissociation of in vitro DNA deamination activity and physiological functions of AID mutants. Proc Natl Acad Sci U S A. 2008;105:15866–15871. [PMC free article] [PubMed]
19. Kodym R, Horth E. Determination of radiation-induced DNA strand breaks in individual cells by non-radioactive labelling of 3′ OH ends. Int J Radiat Biol. 1995;68:133–139. [PubMed]
20. Eto T, Kinoshita K, Yoshikawa K, Muramatsu M, Honjo T. RNA-editing cytidine deaminase Apobec-1 is unable to induce somatic hypermutation in mammalian cells. Proc Natl Acad Sci U S A. 2003;100:12895–12898. [PMC free article] [PubMed]
21. Ramiro AR, et al. AID is required for c-myc/IgH chromosome translocations in vivo. Cell. 2004;118:431–438. [PubMed]
22. Kotani A, et al. Activation-induced cytidine deaminase (AID) promotes B cell lymphomagenesis in Emu-cmyc transgenic mice. Proc Natl Acad Sci U S A. 2007;104:1616–1620. [PMC free article] [PubMed]
23. Liu M, et al. Two levels of protection for the B cell genome during somatic hypermutation. Nature. 2008;451:841–845. [PubMed]
24. Odegard VH, Kim ST, Anderson SM, Shlomchik MJ, Schatz DG. Histone modifications associated with somatic hypermutation. Immunity. 2005;23:101–110. [PubMed]
25. Tashiro J, Kinoshita K, Honjo T. Palindromic but not G-rich sequences are targets of class switch recombination. Int Immunol. 2001;13:495–505. [PubMed]
26. Manser T. Mitogen-driven B cell proliferation and differentiation are not accompanied by hypermutation of immunoglobulin variable region genes. J Immunol. 1987;139:234–238. [PubMed]
27. Stavnezer J, Guikema JE, Schrader CE. Mechanism and regulation of class switch recombination. Annu Rev Immunol. 2008;26:261–292. [PMC free article] [PubMed]
28. Wuerffel R, et al. S-S synapsis during class switch recombination is promoted by distantly located transcriptional elements and activation-induced deaminase. Immunity. 2007;27:711–722. [PubMed]
29. Zarrin AA, et al. Antibody class switching mediated by yeast endonuclease-generated DNA breaks. Science. 2007;315:377–381. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...