- Email: [email protected]
Normal hypermutation in antibody genes from congenic mice defective for DNA polymerase ι
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/dnarepair
Brief report
Normal hypermutation in antibody genes from congenic mice defective for DNA polymerase Stella A. Martomo a , William W. Yang a , Alexandra Vaisman b , Alex Maas c , Masayuki Yokoi d , Jan H. Hoeijmakers c , Fumio Hanaoka d,e , Roger Woodgate b , Patricia J. Gearhart a,∗ a
Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, Baltimore, MD 21224, United States Laboratory of Genomic Integrity, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892, United States c MGC-Department of Cell Biology and Genetics, Center for Biomedical Genetics, Erasmus Medical Center, Rotterdam, The Netherlands d Graduate School of Frontier Biosciences, Osaka University, Osaka, Japan e Discovery Research Institute, RIKEN and Solution-Oriented Research for Science and Technology, Japan Science and Technology Agency, Wako, Japan b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Several low fidelity DNA polymerases participate in generating mutations in immunoglobu-
Received 4 November 2005
lin genes. Polymerase is clearly involved in the process by causing substitutions of A:T base
Received in revised form 9
pairs, whereas polymerase has a controversial role. Although the frequency of mutations
December 2005
was decreased in the BL2 cell line deficient for polymerase , hypermutation was normal in
Accepted 13 December 2005
the 129 strain of mice, which has a natural nonsense mutation in the Poli gene. It is possible
Available online 26 January 2006
that the mice compensated for the defect over time, or that polymerase substituted in the absence of polymerase . To examine polymerase in a genetically defined background,
Keywords:
we backcrossed the 129 nonsense mutation to the C57BL/6 strain for six generations. Class
Immunoglobulins
switch recombination and hypermutation were studied in these mice and in congenic mice
Somatic hypermutation
doubly deficient for both polymerases and . The absence of both polymerases did not
Class switch recombination
affect production of IgG1, indicating that these enzymes are not involved in switch recom-
Pol
bination. Poli−/−F6 mice had the same types of nucleotide substitutions in variable genes
Pol
as their C57BL/6 counterparts, and mice doubly deficient for polymerases and had the
Congenic mice
same mutational spectrum as Polh−/− mice. Thus, polymerase did not contribute to the mutational spectra, even in the absence of polymerase . Published by Elsevier B.V.
1.
Introduction
Immunoglobulins undergo extraordinary diversification after B lymphocytes are stimulated by foreign antigens or mitogens.
∗
Corresponding author. Tel.: +1 410 558 8561; fax: +1 410 558 8157. E-mail address: [email protected] (P.J. Gearhart).
1568-7864/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.dnarep.2005.12.006
This allows animals to produce antibodies with high affinity for the cognate antigen and to change heavy chain classes with unique effector functions. The activation-induced deaminase (AID) protein triggers a chain of events that results
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
in both nucleotide substitutions in variable genes to encode high affinity antibodies, and DNA strand breaks in switch regions to recombine heavy chain constant genes [1]. AID, which is directed to the variable and switch regions by an unknown mechanism, deaminates cytosine in DNA to uracil [2–8]. Uracils in DNA have indeed been detected in bacteria that overexpress the AID protein [9]. Such uracils are mutagenic, and are usually repaired in an error-free manner by the base excision repair pathway. However, during somatic hypermutation, the uracils are processed to produce base substitutions in the following ways [10]: (a) they could be copied by a high fidelity polymerase to produce transitions of C:G base pairs; (b) they could be removed by UNG uracil DNA glycosylase [4,11,12], and the abasic site could be copied by a low fidelity polymerase to generate transitions and transversions of C:G; or (c) they could bind to mismatch repair MSH2–MSH6 proteins and recruit a low fidelity polymerase to synthesize mutations near the uracil, including at A:T pairs [13–16]. Since DNA polymerases ultimately insert the mutations, it is of interest to identify which ones are involved in the process [17]. Ten polymerases have been examined for hypermutation:  [18], ␦ [19], [20,21], [22–24], [25,26], [27,28], [29], [29,30], Rev1 [31,32], and [33,34]. Of these, DNA polymerase (pol) has the strongest effect on hypermutation. People and mice that are deficient for this polymerase have a normal frequency of mutations, but have significantly fewer substitutions of A:T pairs, and correspondingly higher numbers of mutations of C:G pairs [22,35,36]. The residual mutations are likely generated by another polymerase(s), which may substitute in the absence of pol . Pol was proposed to be involved since the BL2 Burkitt’s lymphoma cell line, which had the POLI gene knocked out, had fewer mutations than when pol was restored [25]. However, its role was questioned when the 129/SvJ strain of mice, which has a natural nonsense mutation in the Poli gene, was found to have normal hypermutation [26]. As it is possible that the 129 strains evolved to compensate for the lack of pol , we backcrossed the mutation into C57BL/6 for six generations to create congenic mice on a defined genetic background. Hypermutation and class switch recombination were then examined in these mice and in mice doubly deficient for pols and .
procedures were reviewed and approved by the Animal Ethics Committees (National Institute on Aging, National Institutes of Health, Baltimore; Osaka University; and Erasmus Medical Center).
2.2.
Class switch recombination
Splenic B cells from three to four mice of each genotype were stimulated for 4 days with LPS plus IL-4, stained with propidium iodide and fluorescein-conjugated anti-mouse IgG1, and analyzed by flow cytometry as described [36]. DNA was then prepared from these cells, and the –␥1 switch junctions were amplified with the following primers: for the switch region, first and second forward primers were described previously [14]; for the ␥1 region, first reverse primer (nucleotides 3301–3324 and 8722–8745 of GenBank/EMBL/DDJB under accession no. M12389), 5 -CAATTAGCTCCTGCTCTTCTGTGG3 , and second reverse primer (nucleotides 3301–3319 and 8722–8740), 5 -acgaattcAGCTCCTGCTCTTCTGTGG-3 . Primers for the second nested amplification had the restriction sites XbaI (second forward) and EcoRI (in italics for second reverse). Nested PCR was done using Expand Long Template PCR System (Roche, Roche Diagnostics, IN). The first amplification was 30 cycles in which the first 10 cycles were done at an annealing temperature of 55 ◦ C for 30 s, and extension at 68 ◦ C for 1 min. This was followed by another 20 cycles using the same conditions except for an extension time of 2.5 min. The second PCR was done the same as the first except the annealing temperature was changed to 58 ◦ C. PCR products of 100–600 bp were then cloned and sequenced.
2.3.
Hypermutation
Cells from Peyer’s patches of three to nine non-immunized mice from each genotype were stained with phycoerythrinlabeled antibody to B220 and fluorescein-labeled peanut agglutinin (PNA). The cells were isolated by flow cytometry, and DNA was prepared from B220+ PNA+ cells. The 492-bp intron region downstream of JH 4 from rearranged VH J558 genes was amplified and sequenced using previously described primers [36].
2.
Materials and methods
3.
2.1.
Mice
3.1. Class switch recombination is normal in mice deficient for pols and
C57BL/6 mice were purchased from Jackson Laboratories; Poli−/−F6 mice were produced by backcrossing mice with the pol mutation from the 129 strain to C57BL/6 for six generations; Polh−/− mice have been described [36]; and Poli−/−F6 Polh−/− mice were produced by crossing the respective strains. The mutant Poli allele was detected as previously described [26]. Briefly, the wild type gene has a TaqI restriction site in a 88-bp PCR product of exon 2 of the Poli gene; after cleavage with TaqI, two smaller fragments of 49 and 39 bp are produced. The nonsense mutation destroys the TaqI site, leaving the 88-bp product intact. Heterozygous mice have both products, and mutant homozygous mice have only the 88-bp fragment. Mice were used at 3–6 months of age. All animal
393
Results
Congenic strains are produced by transferring a mutation from one genetic background to another inbred strain through repeated backcrossing and selecting for progeny with the mutation. The congenic strain and the inbred partner are identical at all loci except for the transferred locus. To generate congenic mice carrying the 129-derived nonsense mutation in exon 2 of the Poli gene [26], 129 mice were mated to C57BL/6 mice. C57BL/6 mice were chosen since there is extensive data in the literature on hypermutation in this strain. Offspring with the Poli mutation were then backcrossed to C57BL/6 for several generations. After six generations, F6 mice would contain 98.4% of the C57BL/6 background and 1.6% of the 129
394
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
Fig. 1 – Heavy chain class switching to IgG1. Spleen cells were stimulated in vitro with LPS and IL-4 for 4 days. (A) Flow cytometry. Cells were stained with propidium iodide (PI) and anti-IgG1. The percentage of cells that switched is shown in the lower right quadrant. (B) Length of homology at –␥1 junctions. DNA was prepared from the cells and sequenced to determine microhomology between switch regions. Around 28 clones from each genotype were sequenced.
background, including the Poli mutation. The pol -deficient mice were also on the C57BL/6 background [36]. Heavy chain class switching was then compared in four genotypes of mice on the same genetic background: C57BL/6, Poli−/−F6 , Polh−/− , and Poli−/−F6 Polh−/− . Splenic B cells were stimulated in vitro with LPS and IL-4 to induce switching to IgG1, and the results are shown in Fig. 1A. After 4 days, all four strains had similar levels of cells with membrane IgG1. Furthermore, there was no difference in the length of nucleotide homology at the junctions of recombination between and ␥1 switch regions (Fig. 1B), suggesting that pols and do not fill in the staggered ends generated during double-strand break repair.
3.2. pol
Somatic hypermutation is affected by pol but not
Hypermutation was examined in a 492-bp intron region downstream of VH J558 gene segments rearranged to the JH 4 gene segment in B cells from Peyer’s patches. The number of clones with unique junctions from the joining of variable, diversity, and joining gene segments from B220+ PNA+ cells is shown in Fig. 2A. Although there were more clones with no mutations from Poli−/−F6 mice, it is difficult to draw a conclusion about overall frequencies because they reflect arbitrary exposure of
Peyer’s patch B cells to environmental antigens in the gut. Mutation frequencies were thus calculated for only the clones that had one or more mutations. C57BL/6, Poli−/−F6 , and Polh−/− clones had identical frequencies of 1.3 × 10−2 mutations per bp, and Poli−/−F6 Polh−/− clones had a slightly lower frequency of 0.9 × 10−2 mutations per bp. The location of these mutations in the 492-nucleotide sequence is shown in Fig. 3; there were no obvious differences between the four genotypes. To assess the effect of the polymerases on the overall spectra of substitutions, the types of changes from the nontranscribed strand are recorded in Fig. 2B. For Poli−/−F6 , 128 more clones were included from unsorted Peyer’s patch cells. Thus, 127 mutations were identified in C57BL/6 clones, 131 in Poli−/−F6 clones, 136 in Polh−/− clones, and 121 in Poli−/−F6 Polh−/− clones. Clones from Polh−/− and Poli−/−F6 Polh−/− mice had fewer mutations of A:T and more mutations of C:G compared to clones from C57BL/6 and Poli−/−F6 mice. There was no significant difference between C57BL/6 and Poli−/−F6 clones, and between Polh−/− and Poli−/−F6 Polh−/− clones. The specific types of substitutions are summarized in Table 1, and show a significant decrease in A to G substitutions in Polh−/− and Poli−/−F6 Polh−/− clones compared to C57BL/6 and Poli−/−F6 clones. The complementary T to C substitutions were also decreased in Polh−/− and Poli−/−F6 Polh−/− clones, although to a lesser extent.
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
395
Fig. 2 – Hypermutation in JH 4 introns. DNA from Peyer’s patch B cells was sequenced for 492 bp downstream of rearranged JH 4 genes. (A) Frequency in PNA+ cells. Total number of clones analyzed is shown in the center of each circle. Segments represent the proportion of clones that contain the indicated number of mutations. Frequencies in mutations per bp were calculated for clones with one or more mutations. (B) Total mutations are grouped for each of the four nucleotides. For Poli−/−F6 clones, an additional 128 clones from unsorted cells were included for sequence analysis. Mutations are recorded from the nontranscribed strand.
4.
Discussion
129-derived strains of mice harbor a nonsense mutation in exon 2 of the Poli gene that abrogates production of pol
[26]. Since these mice had normal mutation of immunoglobulin variable genes, we suggested that pol is not required for somatic hypermutation. However, even though we were unable to detect pol in Western blots of tissue extracts,
Fig. 3 – Location of mutations in the JH 4 intron. The intron sequence is shown with substitutions from the four genotypes above and below the sequence. C57BL/6, blue letters above the sequence; Poli−/−F6 , red letters above; Polh−/− , green letters below; and Poli−/−F6 Polh−/− , pink letters below.
396
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
Table 1 – Substitutions in JH 4 regions from pol - and pol -deficient mice Substitution
C57BL/6 (127 mut) (%)
Poli−/−F6 (131 mut) (%)
Polh−/− (136 mut) (%)
Poli−/−F6 Polh−/− (121 mut) (%)
1a , b 5 5
2c ,d 4 7
3e 1 2
2f 1 5
A to:
G T C
21 15 3
18 9 6
T to:
C A G
14 5 3
9 5 4
C to:
T A G
14 4 1
18 3 5
37 7 5
23 8 11
G to:
A T C
14 3 3
16 1 6
20 5 9
22 4 11
Mutations (mut) are recorded from the nontranscribed strand and are corrected for nucleotide composition. Data were analyzed using the Bonferroni correction [42], and were considered significant (p < 0.05/36 = 0.0014) for the following types of substitutions. a b c d e f
A to G: C57BL/6 vs. Polh−/− ; p = 2.2 × 10−8 . A to G: Poli−/−F6 vs. Polh−/− ; p = 1.3 × 10−7 . A to G: C57BL/6 vs. Poli−/−F6 Polh−/− ; p = 4.2 × 10−6 . A to G: Poli−/−F6 vs. Poli−/−F6 Polh−/− ; p = 1.9 × 10−5 . T to C: C57BL/6 vs. Polh−/− ; p = 6.2 × 10−4 . T to C: C57BL/6 vs. Poli−/−F6 Polh−/− ; p = 7.1 × 10−4 .
it is possible that the amber stop codon could somehow be suppressed, perhaps by a suppressor tRNA, to produce undetectable levels of the protein that are nonetheless sufficient for somatic mutation, or that the 129 strain has mutations in other genes that could compensate for the lack of pol . To address these issues, we backcrossed the Poli nonsense mutation into a genetically defined wild type background by producing C57BL/6 mice congenic with the Poli defect. It is unlikely that a gene encoding a suppressor tRNA would be retained after multiple backcrosses unless the gene is linked to the Poli locus. Thus, assuming independent segregation of genes, Poli−/−F6 mice should be 98% identical to C57BL/6, and would provide an excellent model to study the role of pol in hypermutation. In addition, since pol may possibly compensate in the absence of pol to produce normal hypermutation, we studied hypermutation in mice doubly deficient for both pols and . Class switch recombination between constant genes has not been studied in pol - or pol /-deficient mice. As shown here, there was no defect in switching to IgG1 in B cells from mice deficient for either or both polymerases. However, pol does synthesize mutations in switch regions [23,24], indicating that the enzyme is present at the loci, and it may participate in repairing strand breaks. For example, the staggered ends that are generated during switch recombination [37] could be processed to create blunt ends by either filling in the ends with a polymerase or cutting back with an exonuclease. If pol is involved in synthesis at the ends, there may be fewer blunt ends in its absence, and more joins with short stretches of microhomology between the and ␥ switch regions. An analysis of –␥1 switch junctions from the polymerase-deficient cells showed no difference in the length of nucleotide homology compared to wild type cells, suggesting that pols and are not involved in synthesis of these ends.
Somatic hypermutation in the JH 4 intron region adjacent to rearranged variable genes was normal in Poli−/−F6 mice in terms of mutation frequency in the mutated clones and the types of nucleotide substitutions, similar to previous results for the parental 129 strain [26]. As shown recently, Polh−/− mice had significantly fewer mutations of A:T pairs [35,36], and Poli−/−F6 Polh−/− mice also had fewer mutations of A:T. Thus, pol did not contribute to the mutational spectra, even in the absence of pol . Our observations are therefore consistent with other studies showing that pol did not influence hypermutation in mice doubly deficient for pol (129/Ola strain) and pol [35], and in mice deficient for both pols (129/SvJ strain) and [38]. Perhaps the decrease in mutations reported in pol -deficient BL2 cells [25] is due to inherent differences between cell lines and mice. For example, mutations in human B-cell lines, including BL2, consist mostly of substitutions of C:G base pairs, whereas mutations in B cells isolated from humans and mice have equal frequencies of mutation of C:G and A:T pairs. These dissimilar results suggest that different proteins are involved in cell lines versus animals. The dominant role of pol in hypermutation was further confirmed by the virtual absence of A to G substitutions in pol -deficient mice. These are in accord with the in vitro specificity of pol , which misinserts G opposite template T most frequently [39]. The results suggest that pol preferentially synthesizes a repair patch on the nontranscribed strand using template T on the transcribed strand, since the complementary T to C category of mutations was less affected. It thus appears that no other polymerase can substitute for pol to generate A to G mutations in vivo, and supports a role for pol in producing the bias of mutations of A greater than T in immunoglobulin genes [40,41]. It remains to be determined which low fidelity DNA polymerases are responsible for the mutations of C:G and residual mutations of A:T seen in the Polh−/− clones. Since pol
dna repair
5
plays such a dominant role in the hypermutation process, experiments would be most informative using double gene disruptions of pol and polymerases which have been shown to affect mutations of C:G in variable genes, such as pol [33,34] and Rev1 [31]. It is also possible that another enzyme may substitute when a polymerase is physically absent. One way to avoid this problem is to generate mice with a knockin mutation that allows expression of a catalytically inactive polymerase, rather than total deletion of the enzyme.
Acknowledgements We thank I. Rogozin for statistical analyses; J. Chrest, C. Morris, and R. Wersto for flow cytometry; R. Hendriks for mouse tissue; H. Roest for mouse breeding; and D. Wilson and R. Brosh for comments. This work was supported, in part, by the NIH Intramural Research program.
( 2 0 0 6 ) 392–398
[13]
[14]
[15]
[16]
[17]
references [18] [1] T. Honjo, H. Nagaoka, R. Shinkura, M. Muramatsu, AID to overcome the limitations of genomic information, Nat. Immunol. 6 (2005) 655–661. [2] S.K. Petersen-Mahrt, R.S. Harris, M.S. Neuberger, AID mutates E. coli suggesting a DNA deamination mechanism for antibody diversification, Nature 418 (2002) 99–104. [3] J. Di Noia, M.S. Neuberger, Altering the pathway of immunoglobulin hypermutation by inhibiting uracil-DNA glycosylase, Nature 419 (2002) 43–48. [4] C. Rada, G.T. Williams, H. Nilsen, D.E. Barnes, T. Lindahl, M.S. Neuberger, Immunoglobulin isotype switching is inhibited and somatic hypermutation perturbed in UNG-deficient mice, Curr. Biol. 12 (2002) 1748–1755. [5] P. Pham, R. Bransteitter, J. Petruska, M.F. Goodman, Processive AID-catalysed cytosine deamination on single-stranded DNA simulates somatic hypermutation, Nature 424 (2003) 103–107. [6] A.R. Ramiro, P. Stravrapoulos, M. Jankovic, M.C. Nussenzweig, Transcription enhances AID-mediated cytidine deamination by exposing single-stranded DNA on the nontemplate strand, Nat. Immunol. 4 (2003) 452–456. [7] K. Yu, F.-T. Huang, M.R. Lieber, DNA substrate length and surrounding sequence affect the activation-induced deaminase activity at cytidine, J. Biol. Chem. 279 (2004) 6496–6500. [8] U. Basa, J. Chaudhuri, C. Alpert, S. Dutt, S. Ranganath, G. Li, J.P. Schrum, J.P. Manis, F.W. Alt, The AID antibody diversification enzyme is regulated by protein kinase A phosphorylation, Nature 438 (2005) 508–511. [9] S.A. Martomo, D. Fu, W.W. Yang, N.S. Joshi, P.J. Gearhart, Deoxyuridine is generated preferentially in the nontranscribed strand of DNA from cells expressing activation-induced cytidine deaminase, J. Immunol. 174 (2005) 7787–7791. [10] N. Maizels, Immunoglobulin gene diversification, Annu. Rev. Genet. 39 (2005) 23–46. [11] C. Rada, J.M. Di Noia, M.S. Neuberger, Mismatch recognition and uracil excision provide complementary paths to both Ig switching and the A/T-focused phase of somatic mutation, Mol. Cell 16 (2004) 163–171. [12] K. Imai, G. Slupphaug, W.I. Lee, P. Revy, S. Nonoyama, N. Catalan, L. Yel, M. Forveille, B. Kavli, H.E. Krokan, H.D. Ochs, A. Fischer, A. Durandy, Human uracil-DNA
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
397
glycosylase deficiency associated with profoundly impaired immunoglobulin class-switch recombination, Nat. Immunol. 4 (2003) 1023–1028. Z. Li, S.J. Scherer, D. Ronai, M.D. Iglesias-Ussel, J.U. Peled, P.D. Bardwell, M. Zhuang, K. Lee, A. Martin, W. Edelmann, M.D. Scharff, Examination of Msh6- and Msh3-deficient mice in class switching reveals overlapping and distinct roles of MutS homologues in antibody diversification, J. Exp. Med. 200 (2004) 47–59. S.A. Martomo, W.W. Yang, P.J. Gearhart, A role for Msh6 but not Msh3 in somatic hypermutation and class switch recombination, J. Exp. Med. 200 (2004) 61–68. T.M. Wilson, A. Vaisman, S.A. Martomo, P. Sullivan, L. Lan, F. Hanaoka, A. Yasui, R. Woodgate, P.J. Gearhart, MSH2–MSH6 stimulates DNA polymerase , suggesting a role for A:T mutations in antibody genes, J. Exp. Med. 201 (2005) 637–645. E.D. Larson, M.L. Duquette, W.J. Cummings, R.J. Streiff, N. Maizels, MutS␣ binds to and promotes synapsis of transcriptionally activated immunoglobulin switch regions, Curr. Biol. 15 (2005) 470–474. M. Seki, P.J. Gearhart, R.D. Wood, DNA polymerases and somatic hypermutation of immunoglobulin genes, EMBO Rep. 6 (2005) 1143–1148. G. Esposito, G. Texido, U.A. Betz, H. Gu, W. Muller, U. Klein, K. Rajewsky, Mice reconstituted with DNA polymerase -deficient fetal liver cells are able to mount a T cell-dependent immune response and mutate their Ig genes normally, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 1166–1171. A. Longacre, T. Sun, R.E. Goldsby, B.D. Preston, U. Storb, Ig gene somatic hypermutation in mice defective for DNA polymerase ␦ proofreading, Int. Immunol. 15 (2003) 477–481. H. Zan, A. Komori, Z. Li, A. Cerutti, A. Schaffer, M.F. Flajnik, M. Diaz, P. Casali, The translesion DNA polymerase plays a major role in Ig and bcl-6 somatic hypermutation, Immunity 14 (2001) 643–653. M. Diaz, L.K. Verkoczy, M.F. Flajnik, N.R. Klinman, Decreased frequency of somatic hypermutation and impaired affinity maturation but intact germinal center formation in mice expressing antisense RNA to DNA polymerase , J. Immunol. 167 (2001) 327–335. X. Zeng, D.B. Winter, C. Kasmer, K.H. Kraemer, A.R. Lehmann, P.J. Gearhart, DNA polymerase is an A–T mutator in somatic hypermutation of immunoglobulin variable genes, Nat. Immunol. 2 (2001) 537–541. A. Faili, S. Aoufouchi, S. Weller, F. Vuillier, A. Stary, A. Sarasin, C.A. Reynaud, J.C. Weill, DNA polymerase is involved in hypermutation occurring during immunoglobulin class switch recombination, J. Exp. Med. 199 (2004) 265–270. X. Zeng, G.A. Negrete, C. Kasmer, W.W. Yang, P.J. Gearhart, Absence of DNA polymerase reveals targeting of C mutations on the nontranscribed strand in immunoglobulin switch regions, J. Exp. Med. 199 (2004) 917–924. ´ A. Faili, S. Aoufouchi, E. Flatter, Q. Gueranger, C.A. Reynaud, J.C. Weill, Induction of somatic hypermutation in immunoglobulin genes is dependent on DNA polymerase iota, Nature 419 (2002) 944–947. J.P. McDonald, E.G. Frank, B.S. Plosky, I.B. Rogozin, C. Masutani, F. Hanaoka, R. Woodgate, P.J. Gearhart, 129-derived strains of mice are deficient in DNA polymerase and have normal immunoglobulin hypermutation, J. Exp. Med. 198 (2003) 635–643. D. Schenten, V.L. Gerlach, C. Guo, S. Velasco-Miguel, C.L. Hladik, C.L. White, E.C. Friedberg, K. Rajewsky, G. Esposito, DNA polymerase deficiency does not affect somatic
398
[28]
[29]
[30]
[31]
[32]
[33]
[34]
dna repair
hypermutation in mice, Eur. J. Immunol. 32 (2002) 3152–3160. T. Shimizu, Y. Shinkai, T. Ogi, H. Ohmori, T. Azuma, The absence of DNA polymerase does not affect somatic hypermutation of the mouse immunoglobulin heavy chain gene, Immunol. Lett. 86 (2003) 265–270. B. Bertocci, A. De Smet, E. Flatter, A. Dahan, J.C. Bories, C. Landreau, J.C. Weill, C.A. Reynaud, Cutting edge: DNA polymerases and are dispensable for Ig gene hypermutation, J. Immunol. 168 (2002) 3702–3706. J.F. Ruiz, D. Lucas, E. Garcia-Palomero, A.I. Saez, M.A. Gonzalez, M.A. Piris, A. Bernad, L. Blanco, Overexpression of human DNA polymerase (Pol ) in a Burkitt’s lymphoma cell line affects the somatic hypermutation rate, Nucleic Acids Res. 32 (2004) 5861–5863. L.J. Simpson, J.E. Sale, Rev1 is essential for DNA damage tolerance and non-templated immunoglobulin gene mutation in a vertebrate cell line, EMBO J. 22 (2003) 1654–1664. J.G. Jansen, A. Tsaalbi-Shtylik, P. Langerak, F. Calleja, C.M. Meijers, H. Jacobs, N. de Wind, The BRCT domain of mammalian Rev1 is involved in regulating DNA translesion synthesis, Nucleic Acids Res. 33 (2005) 356–365. K. Masuda, R. Ouchida, A. Takeuchi, T. Saito, H. Koseki, K. Kawamura, M. Tagawa, T. Tokuhisa, T. Azuma, J. O-Wang, DNA polymerase contributes to the generation of C/G mutations during somatic hypermutation of Ig genes, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 13986–13991. H. Zan, N. Shima, A. Xu, A. Al-Qahtani, A.J. Evinger III, Y. Zhong, J.C. Schimenti, P. Casali, The translesion DNA polymerase plays a dominant role in immunoglobulin gene somatic hypermutation, EMBO J. 24 (2005) 3757–3769.
5
( 2 0 0 6 ) 392–398
[35] F. Delbos, A. De Smet, A. Faili, S. Aoufouchi, J.C. Weill, C.A. Reynaud, Contribution of DNA polymerase to immunoglobulin gene hypermutation in the mouse, J. Exp. Med. 201 (2005) 1191–1196. [36] S.A. Martomo, W.W. Yang, R.P. Wersto, T. Ohkumo, Y. Kondo, M. Yokoi, C. Masutani, F. Hanaoka, P.J. Gearhart, Different mutation signatures in DNA polymerase - and MSH6-deficient mice suggest separate roles in antibody diversification, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 8656–8661. [37] C.E. Schrader, E.K. Linehan, S.N. Mochegova, R.T. Woodland, J. Stavnezer, Inducible DNA breaks in Ig S regions are dependent on AID and UNG, J. Exp. Med. 202 (2005) 561–568. [38] T. Shimizu, T. Azuma, M. Ishiguro, N. Kanjo, S. Yamada, H. Ohmori, Normal immunoglobulin gene somatic hypermutation in Pol–Pol double-deficient mice, Immunol. Lett. 98 (2005) 259–264. [39] T. Matsuda, K. Bebenek, C. Masutani, I.B. Rogozin, F. Hanaoka, T.A. Kunkel, Error rate and specificity of human and murine DNA polymerase , J. Mol. Biol. 312 (2001) 335–346. [40] V.I. Mayorov, I.B. Rogozin, L.R. Adkison, P.J. Gearhart, DNA polymerase contributes to strand bias of mutations of A versus T in immunoglobulin genes, J. Immunol. 174 (2005) 7781–7786. [41] J. Spencer, D.K. Dunn-Walters, Hypermutation at A–T base pairs: the A nucleotide replacement spectrum is affected by adjacent nucleotides and there is no reverse complementarity of sequences flanking mutated A and T nucleotides, J. Immunol. 175 (2005) 5170–5177. [42] E. Lander, L. Kruglyak, Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results, Nat. Genet. 11 (1995) 241–247.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/dnarepair
Brief report
Normal hypermutation in antibody genes from congenic mice defective for DNA polymerase Stella A. Martomo a , William W. Yang a , Alexandra Vaisman b , Alex Maas c , Masayuki Yokoi d , Jan H. Hoeijmakers c , Fumio Hanaoka d,e , Roger Woodgate b , Patricia J. Gearhart a,∗ a
Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, Baltimore, MD 21224, United States Laboratory of Genomic Integrity, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892, United States c MGC-Department of Cell Biology and Genetics, Center for Biomedical Genetics, Erasmus Medical Center, Rotterdam, The Netherlands d Graduate School of Frontier Biosciences, Osaka University, Osaka, Japan e Discovery Research Institute, RIKEN and Solution-Oriented Research for Science and Technology, Japan Science and Technology Agency, Wako, Japan b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Several low fidelity DNA polymerases participate in generating mutations in immunoglobu-
Received 4 November 2005
lin genes. Polymerase is clearly involved in the process by causing substitutions of A:T base
Received in revised form 9
pairs, whereas polymerase has a controversial role. Although the frequency of mutations
December 2005
was decreased in the BL2 cell line deficient for polymerase , hypermutation was normal in
Accepted 13 December 2005
the 129 strain of mice, which has a natural nonsense mutation in the Poli gene. It is possible
Available online 26 January 2006
that the mice compensated for the defect over time, or that polymerase substituted in the absence of polymerase . To examine polymerase in a genetically defined background,
Keywords:
we backcrossed the 129 nonsense mutation to the C57BL/6 strain for six generations. Class
Immunoglobulins
switch recombination and hypermutation were studied in these mice and in congenic mice
Somatic hypermutation
doubly deficient for both polymerases and . The absence of both polymerases did not
Class switch recombination
affect production of IgG1, indicating that these enzymes are not involved in switch recom-
Pol
bination. Poli−/−F6 mice had the same types of nucleotide substitutions in variable genes
Pol
as their C57BL/6 counterparts, and mice doubly deficient for polymerases and had the
Congenic mice
same mutational spectrum as Polh−/− mice. Thus, polymerase did not contribute to the mutational spectra, even in the absence of polymerase . Published by Elsevier B.V.
1.
Introduction
Immunoglobulins undergo extraordinary diversification after B lymphocytes are stimulated by foreign antigens or mitogens.
∗
Corresponding author. Tel.: +1 410 558 8561; fax: +1 410 558 8157. E-mail address: [email protected] (P.J. Gearhart).
1568-7864/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.dnarep.2005.12.006
This allows animals to produce antibodies with high affinity for the cognate antigen and to change heavy chain classes with unique effector functions. The activation-induced deaminase (AID) protein triggers a chain of events that results
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
in both nucleotide substitutions in variable genes to encode high affinity antibodies, and DNA strand breaks in switch regions to recombine heavy chain constant genes [1]. AID, which is directed to the variable and switch regions by an unknown mechanism, deaminates cytosine in DNA to uracil [2–8]. Uracils in DNA have indeed been detected in bacteria that overexpress the AID protein [9]. Such uracils are mutagenic, and are usually repaired in an error-free manner by the base excision repair pathway. However, during somatic hypermutation, the uracils are processed to produce base substitutions in the following ways [10]: (a) they could be copied by a high fidelity polymerase to produce transitions of C:G base pairs; (b) they could be removed by UNG uracil DNA glycosylase [4,11,12], and the abasic site could be copied by a low fidelity polymerase to generate transitions and transversions of C:G; or (c) they could bind to mismatch repair MSH2–MSH6 proteins and recruit a low fidelity polymerase to synthesize mutations near the uracil, including at A:T pairs [13–16]. Since DNA polymerases ultimately insert the mutations, it is of interest to identify which ones are involved in the process [17]. Ten polymerases have been examined for hypermutation:  [18], ␦ [19], [20,21], [22–24], [25,26], [27,28], [29], [29,30], Rev1 [31,32], and [33,34]. Of these, DNA polymerase (pol) has the strongest effect on hypermutation. People and mice that are deficient for this polymerase have a normal frequency of mutations, but have significantly fewer substitutions of A:T pairs, and correspondingly higher numbers of mutations of C:G pairs [22,35,36]. The residual mutations are likely generated by another polymerase(s), which may substitute in the absence of pol . Pol was proposed to be involved since the BL2 Burkitt’s lymphoma cell line, which had the POLI gene knocked out, had fewer mutations than when pol was restored [25]. However, its role was questioned when the 129/SvJ strain of mice, which has a natural nonsense mutation in the Poli gene, was found to have normal hypermutation [26]. As it is possible that the 129 strains evolved to compensate for the lack of pol , we backcrossed the mutation into C57BL/6 for six generations to create congenic mice on a defined genetic background. Hypermutation and class switch recombination were then examined in these mice and in mice doubly deficient for pols and .
procedures were reviewed and approved by the Animal Ethics Committees (National Institute on Aging, National Institutes of Health, Baltimore; Osaka University; and Erasmus Medical Center).
2.2.
Class switch recombination
Splenic B cells from three to four mice of each genotype were stimulated for 4 days with LPS plus IL-4, stained with propidium iodide and fluorescein-conjugated anti-mouse IgG1, and analyzed by flow cytometry as described [36]. DNA was then prepared from these cells, and the –␥1 switch junctions were amplified with the following primers: for the switch region, first and second forward primers were described previously [14]; for the ␥1 region, first reverse primer (nucleotides 3301–3324 and 8722–8745 of GenBank/EMBL/DDJB under accession no. M12389), 5 -CAATTAGCTCCTGCTCTTCTGTGG3 , and second reverse primer (nucleotides 3301–3319 and 8722–8740), 5 -acgaattcAGCTCCTGCTCTTCTGTGG-3 . Primers for the second nested amplification had the restriction sites XbaI (second forward) and EcoRI (in italics for second reverse). Nested PCR was done using Expand Long Template PCR System (Roche, Roche Diagnostics, IN). The first amplification was 30 cycles in which the first 10 cycles were done at an annealing temperature of 55 ◦ C for 30 s, and extension at 68 ◦ C for 1 min. This was followed by another 20 cycles using the same conditions except for an extension time of 2.5 min. The second PCR was done the same as the first except the annealing temperature was changed to 58 ◦ C. PCR products of 100–600 bp were then cloned and sequenced.
2.3.
Hypermutation
Cells from Peyer’s patches of three to nine non-immunized mice from each genotype were stained with phycoerythrinlabeled antibody to B220 and fluorescein-labeled peanut agglutinin (PNA). The cells were isolated by flow cytometry, and DNA was prepared from B220+ PNA+ cells. The 492-bp intron region downstream of JH 4 from rearranged VH J558 genes was amplified and sequenced using previously described primers [36].
2.
Materials and methods
3.
2.1.
Mice
3.1. Class switch recombination is normal in mice deficient for pols and
C57BL/6 mice were purchased from Jackson Laboratories; Poli−/−F6 mice were produced by backcrossing mice with the pol mutation from the 129 strain to C57BL/6 for six generations; Polh−/− mice have been described [36]; and Poli−/−F6 Polh−/− mice were produced by crossing the respective strains. The mutant Poli allele was detected as previously described [26]. Briefly, the wild type gene has a TaqI restriction site in a 88-bp PCR product of exon 2 of the Poli gene; after cleavage with TaqI, two smaller fragments of 49 and 39 bp are produced. The nonsense mutation destroys the TaqI site, leaving the 88-bp product intact. Heterozygous mice have both products, and mutant homozygous mice have only the 88-bp fragment. Mice were used at 3–6 months of age. All animal
393
Results
Congenic strains are produced by transferring a mutation from one genetic background to another inbred strain through repeated backcrossing and selecting for progeny with the mutation. The congenic strain and the inbred partner are identical at all loci except for the transferred locus. To generate congenic mice carrying the 129-derived nonsense mutation in exon 2 of the Poli gene [26], 129 mice were mated to C57BL/6 mice. C57BL/6 mice were chosen since there is extensive data in the literature on hypermutation in this strain. Offspring with the Poli mutation were then backcrossed to C57BL/6 for several generations. After six generations, F6 mice would contain 98.4% of the C57BL/6 background and 1.6% of the 129
394
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
Fig. 1 – Heavy chain class switching to IgG1. Spleen cells were stimulated in vitro with LPS and IL-4 for 4 days. (A) Flow cytometry. Cells were stained with propidium iodide (PI) and anti-IgG1. The percentage of cells that switched is shown in the lower right quadrant. (B) Length of homology at –␥1 junctions. DNA was prepared from the cells and sequenced to determine microhomology between switch regions. Around 28 clones from each genotype were sequenced.
background, including the Poli mutation. The pol -deficient mice were also on the C57BL/6 background [36]. Heavy chain class switching was then compared in four genotypes of mice on the same genetic background: C57BL/6, Poli−/−F6 , Polh−/− , and Poli−/−F6 Polh−/− . Splenic B cells were stimulated in vitro with LPS and IL-4 to induce switching to IgG1, and the results are shown in Fig. 1A. After 4 days, all four strains had similar levels of cells with membrane IgG1. Furthermore, there was no difference in the length of nucleotide homology at the junctions of recombination between and ␥1 switch regions (Fig. 1B), suggesting that pols and do not fill in the staggered ends generated during double-strand break repair.
3.2. pol
Somatic hypermutation is affected by pol but not
Hypermutation was examined in a 492-bp intron region downstream of VH J558 gene segments rearranged to the JH 4 gene segment in B cells from Peyer’s patches. The number of clones with unique junctions from the joining of variable, diversity, and joining gene segments from B220+ PNA+ cells is shown in Fig. 2A. Although there were more clones with no mutations from Poli−/−F6 mice, it is difficult to draw a conclusion about overall frequencies because they reflect arbitrary exposure of
Peyer’s patch B cells to environmental antigens in the gut. Mutation frequencies were thus calculated for only the clones that had one or more mutations. C57BL/6, Poli−/−F6 , and Polh−/− clones had identical frequencies of 1.3 × 10−2 mutations per bp, and Poli−/−F6 Polh−/− clones had a slightly lower frequency of 0.9 × 10−2 mutations per bp. The location of these mutations in the 492-nucleotide sequence is shown in Fig. 3; there were no obvious differences between the four genotypes. To assess the effect of the polymerases on the overall spectra of substitutions, the types of changes from the nontranscribed strand are recorded in Fig. 2B. For Poli−/−F6 , 128 more clones were included from unsorted Peyer’s patch cells. Thus, 127 mutations were identified in C57BL/6 clones, 131 in Poli−/−F6 clones, 136 in Polh−/− clones, and 121 in Poli−/−F6 Polh−/− clones. Clones from Polh−/− and Poli−/−F6 Polh−/− mice had fewer mutations of A:T and more mutations of C:G compared to clones from C57BL/6 and Poli−/−F6 mice. There was no significant difference between C57BL/6 and Poli−/−F6 clones, and between Polh−/− and Poli−/−F6 Polh−/− clones. The specific types of substitutions are summarized in Table 1, and show a significant decrease in A to G substitutions in Polh−/− and Poli−/−F6 Polh−/− clones compared to C57BL/6 and Poli−/−F6 clones. The complementary T to C substitutions were also decreased in Polh−/− and Poli−/−F6 Polh−/− clones, although to a lesser extent.
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
395
Fig. 2 – Hypermutation in JH 4 introns. DNA from Peyer’s patch B cells was sequenced for 492 bp downstream of rearranged JH 4 genes. (A) Frequency in PNA+ cells. Total number of clones analyzed is shown in the center of each circle. Segments represent the proportion of clones that contain the indicated number of mutations. Frequencies in mutations per bp were calculated for clones with one or more mutations. (B) Total mutations are grouped for each of the four nucleotides. For Poli−/−F6 clones, an additional 128 clones from unsorted cells were included for sequence analysis. Mutations are recorded from the nontranscribed strand.
4.
Discussion
129-derived strains of mice harbor a nonsense mutation in exon 2 of the Poli gene that abrogates production of pol
[26]. Since these mice had normal mutation of immunoglobulin variable genes, we suggested that pol is not required for somatic hypermutation. However, even though we were unable to detect pol in Western blots of tissue extracts,
Fig. 3 – Location of mutations in the JH 4 intron. The intron sequence is shown with substitutions from the four genotypes above and below the sequence. C57BL/6, blue letters above the sequence; Poli−/−F6 , red letters above; Polh−/− , green letters below; and Poli−/−F6 Polh−/− , pink letters below.
396
d n a r e p a i r 5 ( 2 0 0 6 ) 392–398
Table 1 – Substitutions in JH 4 regions from pol - and pol -deficient mice Substitution
C57BL/6 (127 mut) (%)
Poli−/−F6 (131 mut) (%)
Polh−/− (136 mut) (%)
Poli−/−F6 Polh−/− (121 mut) (%)
1a , b 5 5
2c ,d 4 7
3e 1 2
2f 1 5
A to:
G T C
21 15 3
18 9 6
T to:
C A G
14 5 3
9 5 4
C to:
T A G
14 4 1
18 3 5
37 7 5
23 8 11
G to:
A T C
14 3 3
16 1 6
20 5 9
22 4 11
Mutations (mut) are recorded from the nontranscribed strand and are corrected for nucleotide composition. Data were analyzed using the Bonferroni correction [42], and were considered significant (p < 0.05/36 = 0.0014) for the following types of substitutions. a b c d e f
A to G: C57BL/6 vs. Polh−/− ; p = 2.2 × 10−8 . A to G: Poli−/−F6 vs. Polh−/− ; p = 1.3 × 10−7 . A to G: C57BL/6 vs. Poli−/−F6 Polh−/− ; p = 4.2 × 10−6 . A to G: Poli−/−F6 vs. Poli−/−F6 Polh−/− ; p = 1.9 × 10−5 . T to C: C57BL/6 vs. Polh−/− ; p = 6.2 × 10−4 . T to C: C57BL/6 vs. Poli−/−F6 Polh−/− ; p = 7.1 × 10−4 .
it is possible that the amber stop codon could somehow be suppressed, perhaps by a suppressor tRNA, to produce undetectable levels of the protein that are nonetheless sufficient for somatic mutation, or that the 129 strain has mutations in other genes that could compensate for the lack of pol . To address these issues, we backcrossed the Poli nonsense mutation into a genetically defined wild type background by producing C57BL/6 mice congenic with the Poli defect. It is unlikely that a gene encoding a suppressor tRNA would be retained after multiple backcrosses unless the gene is linked to the Poli locus. Thus, assuming independent segregation of genes, Poli−/−F6 mice should be 98% identical to C57BL/6, and would provide an excellent model to study the role of pol in hypermutation. In addition, since pol may possibly compensate in the absence of pol to produce normal hypermutation, we studied hypermutation in mice doubly deficient for both pols and . Class switch recombination between constant genes has not been studied in pol - or pol /-deficient mice. As shown here, there was no defect in switching to IgG1 in B cells from mice deficient for either or both polymerases. However, pol does synthesize mutations in switch regions [23,24], indicating that the enzyme is present at the loci, and it may participate in repairing strand breaks. For example, the staggered ends that are generated during switch recombination [37] could be processed to create blunt ends by either filling in the ends with a polymerase or cutting back with an exonuclease. If pol is involved in synthesis at the ends, there may be fewer blunt ends in its absence, and more joins with short stretches of microhomology between the and ␥ switch regions. An analysis of –␥1 switch junctions from the polymerase-deficient cells showed no difference in the length of nucleotide homology compared to wild type cells, suggesting that pols and are not involved in synthesis of these ends.
Somatic hypermutation in the JH 4 intron region adjacent to rearranged variable genes was normal in Poli−/−F6 mice in terms of mutation frequency in the mutated clones and the types of nucleotide substitutions, similar to previous results for the parental 129 strain [26]. As shown recently, Polh−/− mice had significantly fewer mutations of A:T pairs [35,36], and Poli−/−F6 Polh−/− mice also had fewer mutations of A:T. Thus, pol did not contribute to the mutational spectra, even in the absence of pol . Our observations are therefore consistent with other studies showing that pol did not influence hypermutation in mice doubly deficient for pol (129/Ola strain) and pol [35], and in mice deficient for both pols (129/SvJ strain) and [38]. Perhaps the decrease in mutations reported in pol -deficient BL2 cells [25] is due to inherent differences between cell lines and mice. For example, mutations in human B-cell lines, including BL2, consist mostly of substitutions of C:G base pairs, whereas mutations in B cells isolated from humans and mice have equal frequencies of mutation of C:G and A:T pairs. These dissimilar results suggest that different proteins are involved in cell lines versus animals. The dominant role of pol in hypermutation was further confirmed by the virtual absence of A to G substitutions in pol -deficient mice. These are in accord with the in vitro specificity of pol , which misinserts G opposite template T most frequently [39]. The results suggest that pol preferentially synthesizes a repair patch on the nontranscribed strand using template T on the transcribed strand, since the complementary T to C category of mutations was less affected. It thus appears that no other polymerase can substitute for pol to generate A to G mutations in vivo, and supports a role for pol in producing the bias of mutations of A greater than T in immunoglobulin genes [40,41]. It remains to be determined which low fidelity DNA polymerases are responsible for the mutations of C:G and residual mutations of A:T seen in the Polh−/− clones. Since pol
dna repair
5
plays such a dominant role in the hypermutation process, experiments would be most informative using double gene disruptions of pol and polymerases which have been shown to affect mutations of C:G in variable genes, such as pol [33,34] and Rev1 [31]. It is also possible that another enzyme may substitute when a polymerase is physically absent. One way to avoid this problem is to generate mice with a knockin mutation that allows expression of a catalytically inactive polymerase, rather than total deletion of the enzyme.
Acknowledgements We thank I. Rogozin for statistical analyses; J. Chrest, C. Morris, and R. Wersto for flow cytometry; R. Hendriks for mouse tissue; H. Roest for mouse breeding; and D. Wilson and R. Brosh for comments. This work was supported, in part, by the NIH Intramural Research program.
( 2 0 0 6 ) 392–398
[13]
[14]
[15]
[16]
[17]
references [18] [1] T. Honjo, H. Nagaoka, R. Shinkura, M. Muramatsu, AID to overcome the limitations of genomic information, Nat. Immunol. 6 (2005) 655–661. [2] S.K. Petersen-Mahrt, R.S. Harris, M.S. Neuberger, AID mutates E. coli suggesting a DNA deamination mechanism for antibody diversification, Nature 418 (2002) 99–104. [3] J. Di Noia, M.S. Neuberger, Altering the pathway of immunoglobulin hypermutation by inhibiting uracil-DNA glycosylase, Nature 419 (2002) 43–48. [4] C. Rada, G.T. Williams, H. Nilsen, D.E. Barnes, T. Lindahl, M.S. Neuberger, Immunoglobulin isotype switching is inhibited and somatic hypermutation perturbed in UNG-deficient mice, Curr. Biol. 12 (2002) 1748–1755. [5] P. Pham, R. Bransteitter, J. Petruska, M.F. Goodman, Processive AID-catalysed cytosine deamination on single-stranded DNA simulates somatic hypermutation, Nature 424 (2003) 103–107. [6] A.R. Ramiro, P. Stravrapoulos, M. Jankovic, M.C. Nussenzweig, Transcription enhances AID-mediated cytidine deamination by exposing single-stranded DNA on the nontemplate strand, Nat. Immunol. 4 (2003) 452–456. [7] K. Yu, F.-T. Huang, M.R. Lieber, DNA substrate length and surrounding sequence affect the activation-induced deaminase activity at cytidine, J. Biol. Chem. 279 (2004) 6496–6500. [8] U. Basa, J. Chaudhuri, C. Alpert, S. Dutt, S. Ranganath, G. Li, J.P. Schrum, J.P. Manis, F.W. Alt, The AID antibody diversification enzyme is regulated by protein kinase A phosphorylation, Nature 438 (2005) 508–511. [9] S.A. Martomo, D. Fu, W.W. Yang, N.S. Joshi, P.J. Gearhart, Deoxyuridine is generated preferentially in the nontranscribed strand of DNA from cells expressing activation-induced cytidine deaminase, J. Immunol. 174 (2005) 7787–7791. [10] N. Maizels, Immunoglobulin gene diversification, Annu. Rev. Genet. 39 (2005) 23–46. [11] C. Rada, J.M. Di Noia, M.S. Neuberger, Mismatch recognition and uracil excision provide complementary paths to both Ig switching and the A/T-focused phase of somatic mutation, Mol. Cell 16 (2004) 163–171. [12] K. Imai, G. Slupphaug, W.I. Lee, P. Revy, S. Nonoyama, N. Catalan, L. Yel, M. Forveille, B. Kavli, H.E. Krokan, H.D. Ochs, A. Fischer, A. Durandy, Human uracil-DNA
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
397
glycosylase deficiency associated with profoundly impaired immunoglobulin class-switch recombination, Nat. Immunol. 4 (2003) 1023–1028. Z. Li, S.J. Scherer, D. Ronai, M.D. Iglesias-Ussel, J.U. Peled, P.D. Bardwell, M. Zhuang, K. Lee, A. Martin, W. Edelmann, M.D. Scharff, Examination of Msh6- and Msh3-deficient mice in class switching reveals overlapping and distinct roles of MutS homologues in antibody diversification, J. Exp. Med. 200 (2004) 47–59. S.A. Martomo, W.W. Yang, P.J. Gearhart, A role for Msh6 but not Msh3 in somatic hypermutation and class switch recombination, J. Exp. Med. 200 (2004) 61–68. T.M. Wilson, A. Vaisman, S.A. Martomo, P. Sullivan, L. Lan, F. Hanaoka, A. Yasui, R. Woodgate, P.J. Gearhart, MSH2–MSH6 stimulates DNA polymerase , suggesting a role for A:T mutations in antibody genes, J. Exp. Med. 201 (2005) 637–645. E.D. Larson, M.L. Duquette, W.J. Cummings, R.J. Streiff, N. Maizels, MutS␣ binds to and promotes synapsis of transcriptionally activated immunoglobulin switch regions, Curr. Biol. 15 (2005) 470–474. M. Seki, P.J. Gearhart, R.D. Wood, DNA polymerases and somatic hypermutation of immunoglobulin genes, EMBO Rep. 6 (2005) 1143–1148. G. Esposito, G. Texido, U.A. Betz, H. Gu, W. Muller, U. Klein, K. Rajewsky, Mice reconstituted with DNA polymerase -deficient fetal liver cells are able to mount a T cell-dependent immune response and mutate their Ig genes normally, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 1166–1171. A. Longacre, T. Sun, R.E. Goldsby, B.D. Preston, U. Storb, Ig gene somatic hypermutation in mice defective for DNA polymerase ␦ proofreading, Int. Immunol. 15 (2003) 477–481. H. Zan, A. Komori, Z. Li, A. Cerutti, A. Schaffer, M.F. Flajnik, M. Diaz, P. Casali, The translesion DNA polymerase plays a major role in Ig and bcl-6 somatic hypermutation, Immunity 14 (2001) 643–653. M. Diaz, L.K. Verkoczy, M.F. Flajnik, N.R. Klinman, Decreased frequency of somatic hypermutation and impaired affinity maturation but intact germinal center formation in mice expressing antisense RNA to DNA polymerase , J. Immunol. 167 (2001) 327–335. X. Zeng, D.B. Winter, C. Kasmer, K.H. Kraemer, A.R. Lehmann, P.J. Gearhart, DNA polymerase is an A–T mutator in somatic hypermutation of immunoglobulin variable genes, Nat. Immunol. 2 (2001) 537–541. A. Faili, S. Aoufouchi, S. Weller, F. Vuillier, A. Stary, A. Sarasin, C.A. Reynaud, J.C. Weill, DNA polymerase is involved in hypermutation occurring during immunoglobulin class switch recombination, J. Exp. Med. 199 (2004) 265–270. X. Zeng, G.A. Negrete, C. Kasmer, W.W. Yang, P.J. Gearhart, Absence of DNA polymerase reveals targeting of C mutations on the nontranscribed strand in immunoglobulin switch regions, J. Exp. Med. 199 (2004) 917–924. ´ A. Faili, S. Aoufouchi, E. Flatter, Q. Gueranger, C.A. Reynaud, J.C. Weill, Induction of somatic hypermutation in immunoglobulin genes is dependent on DNA polymerase iota, Nature 419 (2002) 944–947. J.P. McDonald, E.G. Frank, B.S. Plosky, I.B. Rogozin, C. Masutani, F. Hanaoka, R. Woodgate, P.J. Gearhart, 129-derived strains of mice are deficient in DNA polymerase and have normal immunoglobulin hypermutation, J. Exp. Med. 198 (2003) 635–643. D. Schenten, V.L. Gerlach, C. Guo, S. Velasco-Miguel, C.L. Hladik, C.L. White, E.C. Friedberg, K. Rajewsky, G. Esposito, DNA polymerase deficiency does not affect somatic
398
[28]
[29]
[30]
[31]
[32]
[33]
[34]
dna repair
hypermutation in mice, Eur. J. Immunol. 32 (2002) 3152–3160. T. Shimizu, Y. Shinkai, T. Ogi, H. Ohmori, T. Azuma, The absence of DNA polymerase does not affect somatic hypermutation of the mouse immunoglobulin heavy chain gene, Immunol. Lett. 86 (2003) 265–270. B. Bertocci, A. De Smet, E. Flatter, A. Dahan, J.C. Bories, C. Landreau, J.C. Weill, C.A. Reynaud, Cutting edge: DNA polymerases and are dispensable for Ig gene hypermutation, J. Immunol. 168 (2002) 3702–3706. J.F. Ruiz, D. Lucas, E. Garcia-Palomero, A.I. Saez, M.A. Gonzalez, M.A. Piris, A. Bernad, L. Blanco, Overexpression of human DNA polymerase (Pol ) in a Burkitt’s lymphoma cell line affects the somatic hypermutation rate, Nucleic Acids Res. 32 (2004) 5861–5863. L.J. Simpson, J.E. Sale, Rev1 is essential for DNA damage tolerance and non-templated immunoglobulin gene mutation in a vertebrate cell line, EMBO J. 22 (2003) 1654–1664. J.G. Jansen, A. Tsaalbi-Shtylik, P. Langerak, F. Calleja, C.M. Meijers, H. Jacobs, N. de Wind, The BRCT domain of mammalian Rev1 is involved in regulating DNA translesion synthesis, Nucleic Acids Res. 33 (2005) 356–365. K. Masuda, R. Ouchida, A. Takeuchi, T. Saito, H. Koseki, K. Kawamura, M. Tagawa, T. Tokuhisa, T. Azuma, J. O-Wang, DNA polymerase contributes to the generation of C/G mutations during somatic hypermutation of Ig genes, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 13986–13991. H. Zan, N. Shima, A. Xu, A. Al-Qahtani, A.J. Evinger III, Y. Zhong, J.C. Schimenti, P. Casali, The translesion DNA polymerase plays a dominant role in immunoglobulin gene somatic hypermutation, EMBO J. 24 (2005) 3757–3769.
5
( 2 0 0 6 ) 392–398
[35] F. Delbos, A. De Smet, A. Faili, S. Aoufouchi, J.C. Weill, C.A. Reynaud, Contribution of DNA polymerase to immunoglobulin gene hypermutation in the mouse, J. Exp. Med. 201 (2005) 1191–1196. [36] S.A. Martomo, W.W. Yang, R.P. Wersto, T. Ohkumo, Y. Kondo, M. Yokoi, C. Masutani, F. Hanaoka, P.J. Gearhart, Different mutation signatures in DNA polymerase - and MSH6-deficient mice suggest separate roles in antibody diversification, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 8656–8661. [37] C.E. Schrader, E.K. Linehan, S.N. Mochegova, R.T. Woodland, J. Stavnezer, Inducible DNA breaks in Ig S regions are dependent on AID and UNG, J. Exp. Med. 202 (2005) 561–568. [38] T. Shimizu, T. Azuma, M. Ishiguro, N. Kanjo, S. Yamada, H. Ohmori, Normal immunoglobulin gene somatic hypermutation in Pol–Pol double-deficient mice, Immunol. Lett. 98 (2005) 259–264. [39] T. Matsuda, K. Bebenek, C. Masutani, I.B. Rogozin, F. Hanaoka, T.A. Kunkel, Error rate and specificity of human and murine DNA polymerase , J. Mol. Biol. 312 (2001) 335–346. [40] V.I. Mayorov, I.B. Rogozin, L.R. Adkison, P.J. Gearhart, DNA polymerase contributes to strand bias of mutations of A versus T in immunoglobulin genes, J. Immunol. 174 (2005) 7781–7786. [41] J. Spencer, D.K. Dunn-Walters, Hypermutation at A–T base pairs: the A nucleotide replacement spectrum is affected by adjacent nucleotides and there is no reverse complementarity of sequences flanking mutated A and T nucleotides, J. Immunol. 175 (2005) 5170–5177. [42] E. Lander, L. Kruglyak, Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results, Nat. Genet. 11 (1995) 241–247.