Molecular Immunology 53 (2013) 214–217
Contents lists available at SciVerse ScienceDirect
Molecular Immunology
journal homepage: www.elsevier.com/locate/molimm
Short communication
Analysis of Ig gene hypermutation in Ung−/− Polh−/− mice suggests that UNG and
A:T mutagenesis pathway target different U:G lesions
Shuyin Li a,b , Yaofeng Zhao a , Ji-Yang Wang b,∗
a
b
State Key Laboratory of AgroBiotechnology, College of Biological Sciences, China Agricultural University, Beijing 100094, China
Laboratory for Immune Diversity, Research Center for Allergy and Immunology, RIKEN Yokohama Institute, Yokohama, Yokohama 230-0045, Japan
a r t i c l e
i n f o
Article history:
Received 14 July 2012
Accepted 7 August 2012
Available online 4 September 2012
Keywords:
Immunoglobulin gene hypermutation
Activation-induced cytidine deaminase
Uracil DNA glycosylase
DNA polymerase ␩
Cell cycle
a b s t r a c t
The activation-induced cytidine deaminase (AID) initiates Ig gene hypermutation by converting cytosine
to uracil (U) and generating a U:G lesion. Genetic and biochemical studies suggest that the AID-triggered
U:G lesions are processed by three mutagenic pathways to induce mutations at both C:G and A:T pairs.
First, direct replication of the U:G lesion leads to C to T and G to A transitions. Second, U can be excised
by the uracil DNA glycosylase (UNG) and the replication/processing of the resulting abasic site leads to
transversions and transitions at C:G pairs. Third, the U:G lesion is recognized by an atypical mismatch
repair (MMR) pathway which generates mutations at A:T pairs in a DNA polymerase ␩ (POLH)-dependent
manner. To further explore whether these three mutagenic pathways function competitively or independently, we have analyzed Ig gene hypermutation in mice deficient in both UNG and POLH. Compared with
WT mice, UNG deficiency caused elevated frequency of C:G mutations, suggesting that UNG-mediated
U excision led to error-free as well as error-prone repair. In contrast, UNG deficiency did not affect the
frequency and patterns of A:T mutations, suggesting that the MMR did not target U:G lesions normally
recognized and processed by UNG. In addition, POLH deficiency did not affect the frequency and patterns
of C:G mutations and UNG POLH double deficiency showed an additive effect of single deficiency. Based
on these observations and previous results, along with the recent finding that UNG excises AID-triggered
U predominantly during G1 phase of the cell cycle, it appears that UNG and MMR targets U:G lesions
generated during G1 and S phases of the cell cycle, respectively.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
During an immune response against protein antigen, B cells
are activated and undergo rapid expansion in the germinal centers (GC) of the periphery lymphoid tissues. GC B cells undergo
Ig gene somatic hypermutation (SHM), which introduces random
point mutations into Ig genes and is essential for the generation
of high-affinity antibodies. SHM is initiated by the activationinduced cytidine deaminase (AID) (Muramatsu et al., 2000), which
is thought to catalyze the deamination of cytosine to uracil and
generate a U:G lesion on DNA (Chaudhuri et al., 2003).
Based on the genetic and biochemical data, Neuberger et al. have
provided an excellent model explaining how various mutations are
Abbreviations: AID, activation-induced cytidine deaminase; GC, germinal center;
MMR, mismatch repair; POLH, DNA polymerase ␩; SHM, somatic hypermutation;
Ts, transition; Tv, transversion; UNG, uracil DNA glycosylase.
∗ Corresponding author at: Laboratory for Immune Diversity, RIKEN Research Center for Allergy and Immunology, 1-7-22 Suehiro-cho, Tsurumi, Yokohama 230-0045,
Japan. Tel.: +81 45 503 7041; fax: +81 45 503 7040.
E-mail address: oh@rcai.riken.jp (J.-Y. Wang).
0161-5890/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.molimm.2012.08.009
generated (Di Noia and Neuberger, 2007; Maul and Gearhart, 2010;
Rada et al., 1998, 2004; Reynaud et al., 2009). Accordingly, mutations are induced during replication and repair of the AID-triggered
U:G lesion by three major pathways. First, direct replication of U:G
can lead to C to T and G to A transitions (Ts) as U is structurally similar to thymine (T) and normally pairs with adenine (A). Second, U
can be excised by the uracil DNA glycosylase (UNG) and the replication of the resulting abasic site is thought to generate transversions
(Tv) as well as Ts at C:G pairs. Third, the U:G lesion is recognized by
an atypical mismatch repair (MMR) pathway, resulting in the generation of A:T mutations in a DNA polymerase ␩ (POLH)-dependent
manner (Bardwell et al., 2004; Delbos et al., 2005, 2007; Faili et al.,
2004; Frey et al., 1998; Langerak et al., 2007; Martomo et al., 2004,
2005; Rada et al., 1998; Zeng et al., 2001).
It remains less clear as to whether these three mutagenic
pathways function competitively or independently. To further
understand the relationship of these three pathways, we have
established mice deficient in both UNG and POLH and compared
their mutation frequency and patterns of Ig genes with those in WT
and singly deficient mice. We found that absence of UNG increased
the frequency and changed the patterns of C:G mutations but had
no affect on A:T mutations. Conversely, absence of POLH did not
S. Li et al. / Molecular Immunology 53 (2013) 214–217
(A)
(B)
8
6
4
2
0
4
% B220+PNA
A+
To
otal B cells (x
x107)
Tota
al splenocytess (x107)
(C)
4
10
3
2
Ung-/-
Polh-/-
Ung-/Polh-/-/
3
2
1
1
0
WT
215
0
WT
Ung-/-
Polh-/-
Ung-/Polh-/-/
WT
Ung-/-
Polh-/-
Ung-/Polh-/-/
Fig. 1. Normal numbers of splenocytes, B cells and the frequency of GC B cells in WT, Ung−/− , Polh−/− and Ung−/− Polh−/− mice at ages of 11–13-wk. (A) Total numbers of
splenocytes. Single cell suspension of spleen from each mouse was treated with ACK lysis buffer (0.15 M NH4 Cl, 1 mM KHCO3 and 0.1 mM Na2 EDTA, pH 7.4) to eliminate red
blood cells and then counted with a hemocytometer in the presence of trypan blue. The total numbers of live cells in the spleen of each mouse are shown. (B) Total numbers
of B cells in the spleen of each mouse after purification with an IMAG B Lymphocyte Enrichment Set. (C) Frequency of the B220+ PNA+ GC B cells in each mouse before sorting.
A bar indicates the average.
affect C:G mutations and UNG POLH double deficiency showed an
additive effect of single deficiency. These observations suggested
that UNG-mediated C:G and POLH-mediated A:T mutagenesis did
not interfere with each other. Along with the findings that UNG
excises AID-triggered U predominantly during G1 phase (Sharbeen
et al., 2012) and that rapid DNA synthesis is important for induction
of A:T mutations (Kano et al., 2011), it appears that UNG and A:T
mutagenesis targets U:G lesions generated during G1 and S phase
of the cell cycle, respectively.
Lymphocyte Enrichment Set (BD Biosciences) and then stained with
APC-B220 (BD Biosciences) and FITC-PNA (Vector Laboratories).
B220+ PNA+ GC B cells were then sorted using an Aria cell sorter and
>104 GC B cells were collected. Genomic DNA was extracted and the
JH 4 intronic region was amplified with forward primer J558Fr3 (5 CAGCCTGACATCTGAGGACTCTGC-3 ) and reverse primer JHCHint
(5 -CTCCACCAGACCTCTCTAGACAGC-3 ) as described (Kano et al.,
2012). The PCR products were cloned into the pCR2.1 vector for
sequencing. Only unique sequences were analyzed in each mouse.
2. Materials and methods
3. Results and discussion
2.1. Mice
3.1. Ig gene hypermutation in WT, Ung−/− , Polh−/− and
Ung−/− Polh−/− mice
Ung−/− (Nilsen et al., 2003) and Polh−/− mice (Ohkumo et al.,
2006) were kindly provided by Dr. Deborah Barnes and Dr. Fumio
Hanaoka, respectively. Ung−/− mice had a mixed genetic background of 129 and C57BL/6 and were backcrossed to C57BL/6 for
three generations before crossing with Polh−/− mice. Polh−/− mice
had been backcrossed with C57BL/6 for 12 generations (Kano et al.,
2012). Ung+/− Polh+/− were crossed to obtain WT, Ung−/− , Polh−/−
and Ung−/− Polh−/− mice. The mice were maintained under specific
pathogen-free conditions and all experiments were approved by
the Animal Facility Committee of RIKEN Yokohama Institute (Permission number 20-025).
2.2. Immunization and analysis of Ig gene hypermutation
Two WT, 3 Ung−/− , 2 Polh−/− and 2 Ung−/− Polh−/− mice at ages
of 11–13-wk-old were immunized i.p. with 100 ␮g of 4-hydroxy3-nitrophenyl-acetyl coupled to chicken ␥ globulin (NP-CGG) in
Imject Alum Adjuvant (Thermo Fisher Scientific). Two weeks later,
Spleen B cells were purified using negative sorting with the IMag B
Ung−/− Polh−/− mice developed normally with no obvious
abnormalities by appearance and had normal numbers of total
splenocytes (Fig. 1A and B) and B cells (Fig. 1B). In addition, the
frequency of the B220+ PNA+ GC B cells was not significantly different from that in WT and singly deficient mice (Fig. 1C), suggesting
that B cell activation and expansion in vivo in response to antigen
stimulation were grossly normal in these mice. We sorted GC B
cells from 2 WT, 3 Ung−/− , 2 Polh−/− and 2 Ung−/− Polh−/− mice and
analyzed mutations in the intronic region 3 of JH 4. We chose JH 4
intronic region since mutations in this region do not affect antibody
affinity and therefore represent unbiased mutations. The results
are summarized in Table 1. The overall mutation frequency in WT
mice was 0.761 × 10−2 /bp. This value was slightly lower compared
with that in our previous studies (Masuda et al., 2009; Kano et al.,
2012), which could be due to the use of a different alum conjugate
in the present study. Nevertheless, the same NP-CGG precipitated
with alum was used for immunization of all genotypes and should
not affect the comparison of Ig gene hypermutation among these
Table 1
Mutation frequency in the JH 4 intronic region of WT, Ung−/− , Polh−/− and Ung−/− Polh−/− mice.
Number of sequences
Mutated sequences (%)
Total length of mutated sequences
Total number of mutations
Overall mutation frequency (×10−2 /bp)
Mutation frequency at C:G (×10−2 /bp)
Mutation frequency at A:T (×10−2 /bp)
% mutation at C:G vs. A:T
a
b
WT (2 mice)
Ung−/− (3 mice)
Polh−/− (2 mice)
Ung−/− Polh−/− (2 mice)
270
202 (74.8%)
102,818
782
0.761
0.373
0.388
49.0:51.0
343
260 (75.8%)
132,340
1190
0.899
0.528a
0.371
58.7:41.3
263
171 (65.0%)
87,039
402
0.462
0.401
0.061b
86.8:13.2
375
264 (70.4%)
134,376
1004
0.747
0.662a
0.085b
88.6:11.4
p < 0.05 compared with mutation frequency at C:G of WT mice (unpaired t-test).
p < 0.01 compared with mutation frequency at A:T of WT mice (unpaired t-test).
216
S. Li et al. / Molecular Immunology 53 (2013) 214–217
WT
C
G
T
%
A
*
6.41
13.4
6.17
26.0
C
1 04
1.04
*
0 15
0.15
28 4
28.4
29 6
29.6
25.5
G
27.2
1.20
*
0.75
29.1
16.8
T
3.79
6.50
5.01
*
15.3
G
T
%
A
*
5.95
16.8
11.4
34.2
C
4 30
4.30
*
3 14
3.14
16.1
16 1
23 5
23.5
G
11.0
9.86
*
4.67
T
4.42
8.64
3.71
*
A
C
G
T
%
A
*
5.30
2.54
2.54
10.4
C
2.20
*
8.36
30.8
41.4
G
25.6
15.3
*
4.5
T
0.60
0.20
2.00
*
To
To
Fro
om
C
A
C
G
T
%
A
*
5.46
1.02
1.79
8.27
C
0
*
0.71
50.9
51.6
45.4
G
36.6
0.27
*
0.09
37.0
2.80
T
0.64
0.48
2.01
*
3.13
To
Fro
om
Fro
om
A
A
To
Fro
om
Ung
g-/-
Polh-/-/
Ung-/-/ Polh-/-/
Fig. 2. Relative representation of each type of base substitution in the JH 4 intronic region of WT, Ung−/− , Polh−/− and Ung−/− Polh−/− mice. The data are corrected for base
composition (A = 26.91%, T = 31.04%, C = 14.14%, G = 27.89%).
mice. Mutation frequency at C:G and A:T in WT mice was 0.373
and 0.388, respectively, so the ratio of mutations at C:G vs. A:T was
roughly 1:1. Deficiency of UNG resulted in a statistically significant
increase in the frequency of C:G mutations as compared with WT
mice (Table 1; 0.528 vs. 0.373 in WT mice) and these C:G mutations
were predominantly Ts (Fig. 2), consistent with previous studies
(Krijger et al., 2009; Rada et al., 2002, 2004; Storb et al., 2009).
These results again suggest that UNG is involved in both error-free
repair of the U:G lesion and the induction of C:G Tv. Notably, UNG
deficiency did not alter the frequency (Table 1) and patterns of A:T
mutations (Fig. 2). These observations suggest that whether or not
U is excised by UNG essentially has no affect on the generation
of A:T mutations. A:T mutations are thought to be generated by a
MMR-like pathway, which is initiated with MSH2/MSH6 recognition of the U/G lesion. The finding that UNG deficiency did not cause
increased A:T mutations suggests that the U:G lesion normally recognized by UNG is not the target of the MMR-like pathway. In other
words, MSH2/6 and UNG do not compete for the same U:G lesion to
generate mutations. Deficiency of POLH caused a dramatic reduction of A:T mutations (Table 1; 0.061 vs. 0.388 in WT) as reported
previously (Delbos et al., 2005, 2007; Faili et al., 2004; Martomo
et al., 2005; Zeng et al., 2001), but did not affect the frequency and
patterns of C:G mutations (Table 1 and Fig. 2). These observations
are consistent with the notion that POLH functions at a late stage of
A:T mutagenesis and its absence should not affect the recognition
and processing of U:G lesions by UNG. Finally, UNG and POLH double deficient mice showed an additive effect on both C:G and A:T
mutations as compared with singly deficient mice, suggesting that
UNG-mediated C:G and POLH-mediated A:T mutations functioned
in distinct pathways.
3.2. A cell cycle-based model of Ig gene hypermutation
Our results suggest that UNG and the MMR-like pathway do
not compete for the same U:G lesion for processing and subsequent mutation induction. Very recently, it has been shown that
UNG excises AID-induced U predominantly during G1 phase of the
cell cycle (Sharbeen et al., 2012). On the other hand, we have previously found that rapid DNA synthesis is important for efficient
induction of A:T mutations in an in vitro mutagenesis system in
Ramos B cells (Kano et al., 2011), which suggests that A:T mutagenesis may normally function predominantly during S phase of
the cell cycle. Conventional “error-free” MMR functions to repair
mismatches generated during DNA replication (Hsieh and Yamane,
2008; Iyer et al., 2006; Kunkel and Erie, 2005; Li, 2008) and MMR
activity has been shown to increases during S phase and is highest
on actively replicating templates (Edelbrock et al., 2009). Several
MMR components including MSH2 and MSH6 are commonly used
by both the conventional and the error-prone MMR-like pathways.
It is thus possible that the activity of the error-prone MMR-like
pathway involved in A:T mutagenesis is also highest during S phase
of the cell cycle.
Based on the results of present and previous studies, we propose a cell cycle-based model of Ig gene hypermutation (Fig. 3). In
this model, U:G lesions generated during G1 and S phase are recognized and processed by UNG and MMR, respectively. U:G lesions at
G1 phase may be processed by the following three pathways. First,
U is excised by UNG and if the cell is still at G1 phase after U removal,
the resulting abasic site is further processed by base excision repair
(BER) to restore the original C:G pair (error-free repair). Second, if
the cell enters S phase after U removal, the abasic site is subject to
translesion DNA synthesis (TLS), resulting in induction of Tv and Ts
at C:G pairs. Third, if the U:G lesion is induced at late G1 or early
S phase right before DNA replication, it may be replicated before it
G1
BER Error-free repair
S
TLS
G
G1
Tv and Ts at C:G
U
G
S
Replication
p
Ts at C:G
C
G
U
G
MSH2/6
Exo1/PCNAub/POLH Mutations at A:T
S
Fig. 3. A cell cycle-based model of Ig gene hypermutation. U:G lesions triggered
at G1 and S phase of the cell cycle are recognized and processed by UNG and the
MMR-like pathway, respectively. The cell cycle status (G1 or S) is shown in light
blue. Type of mutations generated by each pathway is highlighted in orange. BER,
base excision repair; TLS, translesion DNA synthesis; Ts, transition; Tv, transversion;
Exo1, exonuclease 1; PCNAub , PCNA ubiquitination. (For interpretation of references
to color in this figure legend, the reader is referred to the web version of this article.)
S. Li et al. / Molecular Immunology 53 (2013) 214–217
can be processed by UNG, leading to induction of C:G Ts. It is also
conceivable that when there are excess U:G lesions at G1 phase that
exceed the capacity of UNG-mediated U removal, the unprocessed
U:G lesions will be replicated in the succeeding S phase and cause
C:G Ts. In the absence of UNG, all the U:G lesions induced during G1
phase are eventually replicated to generate C:G Ts. The frequency
of these C:G Ts is greater than the frequency of total C:G mutations
generated in the presence of UNG in which case a portion of the U:G
lesions are correctly repaired by BER. On the other hand, U:G lesions
at S phase are recognized by MSH2/6 and subsequently processed
by the “error-prone” MMR-like pathway, leading to the induction
of A:T mutations. In the absence of MSH2/6, the U:G lesions generated at S phase may be left unrepaired until the cell enters G1 phase.
In this case, due to the limited capacity of UNG, many U:G lesions
are eventually replicated in the succeeding S phase to generate C:G
Ts. This model is in good agreement with the available data and
explains very well the frequency and patterns of Ig gene hypermutation in various genetic models. Our model is in striking contrast
with a model presented previously (Reynaud et al., 2009). Further
studies are required to completely understand the role of cell cycle
progression in the recognition and processing of AID-triggered U:G
lesions.
Acknowledgements
We thank Prof. D.E. Barnes and Prof. F Hanaoka for the UNG- and
POLH-deficient mice, respectively. We also thank Hiromi Mori for
excellent technical assistance, the FACS Laboratory for cell sorting,
and the Immunogenomics group for sequencing.
References
Bardwell, P.D., Woo, C.J., Wei, K., Li, Z., Martin, A., Sack, S.Z., Parris, T., Edelmann, W.,
Scharff, M.D., 2004. Altered somatic hypermutation and reduced class-switch
recombination in exonuclease 1-mutant mice. Nature Immunology 5, 224–229.
Chaudhuri, J., Tian, M., Khuong, C., Chua, K., Pinaud, E., Alt, F.W., 2003. Transcriptiontargeted DNA deamination by the AID antibody diversification enzyme. Nature
422, 726–730.
Delbos, F., Aoufouchi, S., Faili, A., Weill, J.C., Reynaud, C.A., 2007. DNA polymerase
␩ is the sole contributor of A/T modifications during immunoglobulin gene
hypermutation in the mouse. Journal of Experimental Medicine 204, 17–23.
Delbos, F., De Smet, A., Faili, A., Aoufouchi, S., Weill, J.C., Reynaud, C.A., 2005. Contribution of DNA polymerase ␩ to immunoglobulin gene hypermutation in the
mouse. Journal of Experimental Medicine 201, 1191–1196.
Di Noia, J.M., Neuberger, M.S., 2007. Molecular mechanisms of antibody somatic
hypermutation. Annual Review of Biochemistry 76, 1–22.
Edelbrock, M.A., Kaliyaperumal, S., Williams, K.J., 2009. DNA mismatch repair efficiency and fidelity are elevated during DNA synthesis in human cells. Mutation
Research 662, 59–66.
Faili, A., Aoufouchi, S., Weller, S., Vuillier, F., Stary, A., Sarasin, A., Reynaud, C.A., Weill,
J.C., 2004. DNA polymerase eta is involved in hypermutation occurring during
immunoglobulin class switch recombination. Journal of Experimental Medicine
199, 265–270.
Frey, S., Bertocci, B., Delbos, F., Quint, L., Weill, J.C., Reynaud, C.A., 1998. Mismatch
repair deficiency interferes with the accumulation of mutations in chronically stimulated B cells and not with the hypermutation process. Immunity 9,
127–134.
217
Hsieh, P., Yamane, K., 2008. DNA mismatch repair: molecular mechanism, cancer,
and ageing. Mechanisms of Ageing and Development 129, 391–407.
Iyer, R.R., Pluciennik, A., Burdett, V., Modrich, P.L., 2006. DNA mismatch repair:
functions and mechanisms. Chemical Reviews 106, 302–323.
Kano, C., Hanaoka, F., Wang, J.-Y., 2012. Analysis of mice deficient in both REV1
catalytic activity and POLH reveals an unexpected role for POLH in the generation
of C to G and G to C transversions during Ig gene hypermutation. International
Immunology 24, 169–174.
Kano, C., Ouchida, R., Kokubo, T., Wang, J.-Y., 2011. Rapid cell division contributes to
efficient induction of A/T mutations during Ig gene hypermutation. Molecular
Immunology 48, 1993–1999.
Krijger, P.H., Langerak, P., van den Berk, P.C., Jacobs, H., 2009. Dependence of
nucleotide substitutions on Ung2 Msh2, and PCNA-Ub during somatic hypermutation. Journal of Experimental Medicine 206, 2603–2611.
Kunkel, T.A., Erie, D.A., 2005. DNA mismatch repair. Annual Review of Biochemistry
74, 681–710.
Langerak, P., Nygren, A.O., Krijger, P.H., van den Berk, P.C., Jacobs, H.,
2007. A/T mutagenesis in hypermutated immunoglobulin genes strongly
depends on PCNAK164 modification. Journal of Experimental Medicine 204,
1989–1998.
Li, G.M., 2008. Mechanisms and functions of DNA mismatch repair. Cell Research 18,
85–98.
Martomo, S.A., Yang, W.W., Gearhart, P.J., 2004. A role for Msh6 but not Msh3 in
somatic hypermutation and class switch recombination. Journal of Experimental
Medicine 200, 61–68.
Martomo, S.A., Yang, W.W., Wersto, R.P., Ohkumo, T., Kondo, Y., Yokoi, M., Masutani, C., Hanaoka, F., Gearhart, P.J., 2005. Different mutation signatures in DNA
polymerase ␩- and MSH6-deficient mice suggest separate roles in antibody
diversification. Proceedings of the National Academy of Sciences of the United
States of America 102, 8656–8661.
Masuda, K., Ouchida, R., Li, Y., Gao, X., Mori, H., Wang, J.-Y., 2009. A critical role for
REV1 in regulating the induction of C:G transitions and A:T mutations during Ig
gene hypermutation. Journal of Immunology 183, 1846–1850.
Maul, R.W., Gearhart, P.J., 2010. AID and somatic hypermutation. Advances in
Immunology 105, 159–191.
Muramatsu, M., Kinoshita, K., Fagarasan, S., Yamada, S., Shinkai, Y., Honjo, T., 2000.
Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102, 553–563.
Nilsen, H., Stamp, G., Andersen, S., Hrivnak, G., Krokan, H.E., Lindahl, T., Barnes, D.E.,
2003. Gene-targeted mice lacking the Ung uracil-DNA glycosylase develop B-cell
lymphomas. Oncogene 22, 5381–5386.
Ohkumo, T., Kondo, Y., Yokoi, M., Tsukamoto, T., Yamada, A., Sugimoto, T., Kanao,
R., Higashi, Y., Kondoh, H., Tatematsu, M., Masutani, C., Hanaoka, F., 2006. UVB radiation induces epithelial tumors in mice lacking DNA polymerase eta and
mesenchymal tumors in mice deficient for DNA polymerase iota. Molecular and
Cellular Biology 26, 7696.
Rada, C., Di Noia, J.M., Neuberger, M.S., 2004. Mismatch recognition and uracil excision provide complementary paths to both Ig switching and the A/T-focused
phase of somatic mutation. Molecular Cell 16, 163–171.
Rada, C., Ehrenstein, M.R., Neuberger, M.S., Milstein, C., 1998. Hot spot focusing of
somatic hypermutation in MSH2-deficient mice suggests two stages of mutational targeting. Immunity 9, 135–141.
Rada, C., Williams, G.T., Nilsen, H., Barnes, D.E., Lindahl, T., Neuberger, M.S., 2002.
Immunoglobulin isotype switching is inhibited and somatic hypermutation perturbed in UNG-deficient mice. Current Biology 12, 1748–1755.
Reynaud, C.A., Delbos, F., Faili, A., Guéranger, Q., Aoufouchi, S., Weill, J.C., 2009. Competitive repair pathways in immunoglobulin gene hypermutation. Philosophical
Transactions of the Royal Society of London Series B 364, 613–619.
Sharbeen, G., Yee, C.W., Smith, A.L., Jolly, C.J., 2012. Ectopic restriction of DNA repair
reveals that UNG2 excises AID-induced uracils predominantly or exclusively
during G1 phase. Journal of Experimental Medicine 209, 965–974.
Storb, U., Shen, H.M., Nicolae, D., 2009. Somatic hypermutation: processivity of the
cytosine deaminase AID and error-free repair of the resulting uracils. Cell Cycle
8, 3097–3101.
Zeng, X., Winter, D.B., Kasmer, C., Kraemer, K.H., Lehmann, A.R., Gearhart, P.J., 2001.
DNA polymerase ␩ is an A–T mutator in somatic hypermutation of immunoglobulin variable genes. Nature Immunology 2, 537–541.