Normal immunoglobulin gene somatic hypermutation in Polκ–Polι double-deficient mice

Immunology Letters 98 (2005) 259–264

Normal immunoglobulin gene somatic hypermutation in Pol␬–Pol␫ double-deficient mice Takeyuki Shimizua,1 , Takachika Azumaa,∗ , Mariko Ishigurob , Naoko Kanjob , Shuichi Yamadac , Haruo Ohmorib a

b

Division of Biosignaling, Research Institute for Biological Sciences (RIBS), Tokyo University of Science, Yamazaki 2669, Noda, Chiba 278-0022, Japan Department of Gene Information Analysis, Institute for Virus Research, Kyoto University, Shogoin Kawara-cho 53, Sakyo-ku, Kyoto 606-8507, Japan c Department of Signal Transduction, Institute for Virus Research, Kyoto University, Shogoin Kawara-cho 53, Sakyo-ku, Kyoto 606-8507, Japan Received 15 October 2004; received in revised form 24 November 2004; accepted 24 November 2004 Available online 15 December 2004

Abstract Somatic hypermutation (SHM) occurs in the variable region of immunoglobulin genes in germinal center B cells where it plays an important role in affinity maturation of the T cell-dependent immune response. Although the precise mechanism of SHM is still unknown, it has been suggested that error-prone DNA polymerases (Pol) are involved in SHM. Pol␫ is a member of the error-prone Y-family of DNA polymerases which exhibit translesion synthesis activity in vitro and are highly mutagenic when replicating on non-damaged DNA templates. In BL2 cell line stimulated to induce SHM, the induction is Pol␫-dependent. However, in 129-derived strains of mice deficient in Pol␫, SHM is normal. One possible explanation for this discrepancy is that a Pol␫ deficiency in mice might be compensated for by another error-prone DNA polymerase, such as Pol␬, which also belongs to the Y-family of DNA polymerases. Although SHM in Pol␬-deficient mice is normal, their deficiency might be compensated for by Pol␫. In this study, we generated Pol␬–Pol␫ double-deficient mice and examined them for SHM. We found that the double-deficient mice had the normal SHM frequency and profile, rendering them indistinguishable from Pol␬-deficient mice and thus conclude that Pol␫ and Pol␬ are dispensable for SHM in mice. © 2004 Elsevier B.V. All rights reserved. Keywords: Somatic hypermutation; Immunoglobulin; Affinity maturation; DNA polymerase

1. Introduction In order to create diversity among antibody genes, B cells use two kinds of genetic modification of immunoglobulin variable region genes. In precursor B cells, V(D)J recombination establishes the primary antibody repertoire. When na¨ıve B cells encounter T cell-dependent antigens, some of ∗

Corresponding author. Tel.: +81 4 7121 4082; fax: +81 4 7121 4089. E-mail address: [email protected] (T. Azuma). 1 Present address: Department of Biochemistry, Sapporo Medical University School of Medicine, South-1 West-17, Chuo-ku, Sapporo 060-8556, Japan. 0165-2478/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.imlet.2004.11.022

them migrate into B cell follicles in secondary lymphoid organs and form a germinal center (GC). Immunoglobulin genes in GC centroblasts mutate so as to increase affinity to antigens. This event is called somatic hypermutation (SHM) [1]. B cells which have acquired a higher affinity to antigen are selected and differentiate into plasma cells or memory B cells [2]. SHM mainly occurs in the variable region within 1–2 kb of the immunoglobulin promoter, although few mutations are found in the constant region. The mutation rate is 10−4 –10−3 per base pair per generation, which is a millionfold higher than the rate of spontaneous mutation. Mutations are introduced with high frequency in the SHM hot spots, motifs RGYW and TAA (hotspot positions are italicized).

260

T. Shimizu et al. / Immunology Letters 98 (2005) 259–264

While the precise mechanism for SHM is still unclear, it is widely believed that introduction of a DNA lesion followed by error-prone DNA repair may be responsible [3–5]. A critical factor in the induction of SHM and class switch recombination is activation-induced cytidine deaminase (AID), which is now known as the only essential lymphocyte-specific factor needed for SHM to occur [6,7]. Interestingly, mutations found in switch regions have the same characteristics as SHM in the variable region [8,9]. This suggests that mechanisms of SHM and class switch recombination have a common AIDmediated step. The exact function of AID in SHM is still under investigation, but it is suggested that this deaminase can cause DNA lesions by converting dC residues in singlestranded DNA to dU [10]. dU and abasic sites generated by the action of uracil DNA N-glycosylase (UNG) at dU are presumed to be repaired by base excision or mismatch repair pathways. These repair pathways are usually error-free, although errors may occur in SHM, possibly due to the participation of error-prone DNA polymerase(s) (Pol). Recently, a number of error-prone DNA polymerases have been identified in mammals. The Y-family of DNA polymerases (Pol␬, Pol␩, Pol␫, REV1) show very high rates of nucleotide misincorporation and exhibit translesion DNA synthesis (TLS) activity in vitro [11]. Pol␨ belongs to another family (B) of DNA polymerases. While mammalian Pol␨ enzymes have not been purified as yet, they are believed from analogy with the yeast homologue to have TLS activity. Experiments have been carried out to test the possibility that some of these polymerases may act as mutators in SHM [12]. Pol␬-deficient mice show a normal SHM frequency and profile [13,14]. In other studies, introduction of the antisense RNA of Pol␨ reduces the rate of SHM [15,16]. XP-V patients, who lack active Pol␩ enzyme, show a normal SHM frequency, but exhibit an atypical base-exchange pattern [17]. Targeted destruction of the REV1 gene in a chicken B cell line (DT40) decreases the SHM frequency [18]. These results suggest that Pol␨, Pol␩, and REV1 may participate in SHM. In each of the above cases, however, SHM does not disappear, suggesting that when one of these polymerases is not available, its function may be taken over by other polymerases. In the case of Pol␫, two apparently contradictory observations have been reported. In a human Burkitt’s lymphoma cell line (BL2) stimulated to induce SHM, targeted disruption of the POLI gene coding for Pol␫ reduces the frequency of mutations to the background level [19]. On the other hand, 129-derived strains of mice with a natural nonsense mutation in their Poli gene show a normal SHM frequency and profile [20]. One possible explanation for this discrepancy is that other polymerases can act as replacements for Pol␫ in 129-derived mice in vivo, whereas in the BL2 cell line, the amount or recruitment of other polymerases may not be sufficient to compensate for the Pol␫ deficiency. We expected that the outcome of a Pol␫ deficiency on SHM might be exaggerated under conditions where one of the other error-prone polymerases is not available. We have generated Pol␬–Pol␫ double-deficient mice by mating 129 SvJ mice with Polk knockout mice. After immu-

nization of mice with T cell-dependent antigen, SHM in the immunoglobulin heavy chain gene was analyzed.

2. Materials and methods 2.1. Mice and genotyping Polk knockout mice were described previously [21]. 129 SvJ mice were purchased from Clea Japan, Inc. and crossed with Polk knockout mice. The genotype of the Polk locus was determined by PCR using the primers GCTACTTCGAATTACCATGCAAGG and CTCCTTACTCACAGCTCTATATTTGTCAAA for wild type, and GGGCCAGCTCATTCCTCCACTCATGATC and CAGGCTGCAGGGTTGGAAACAGCCACAC for the knockout loci. Genotyping of the Poli locus was described previously [20]. Briefly, the Poli locus was amplified with the primers TTCGACCTGGGCATAAAAGCAATCC and GGGCAGTTTGCAGTCAAGGGCCACC. The amplified product (92 bp) was analyzed by electrophoresis after digestion by the restriction enzyme TaqI whose site is destroyed in the 129 SvJ mice by a nonsense mutation in codon 27 of the Poli gene. 2.2. Immunization and cell sorting (4-Hydroxy-3-nitrophenyl)acetyl (NP) conjugates of chicken gamma globulin (NP-CGG) were prepared for used as antigen. NP-CGG was precipitated with alum and 100 ␮g of antigen was intraperitoneally injected. Two weeks after immunization, mice were sacrificed and spleens were taken. A single-cell suspension was prepared by a conventional method. Cells were incubated with biotin-conjugated antimouse IgD and anti-CD43 antibodies, followed by exposure to streptavidin-conjugated MACS beads and separated using MACS LS columns (Miltenyi Biotec, Germany). Cells in the flow-through fraction, containing enriched GC B cells, were stained by anti-B220-PE and anti-GL7-FITC antibodies. B220+ GL7+ cells were purified as a source of GC B cells using a FACS Vantage cell sorter (BD Biosciences, CA). Cells binding to the column were eluted and stained with antiB220-PE and anti-IgD-FITC antibodies. B220+ IgD+ cells were purified as a source of na¨ıve B cells. All antibodies were purchased from BD Biosciences or e-Bioscience (CA). Purities of GC B cells and na¨ıve B cells were 79–88 and 95–100%, respectively. 2.3. PCR and sequencing Cells were lysed in 10 mM Tris–HCl, pH 8.5, and 0.5 mg/ml proteinase K (105 cells/50 ␮l) at 50 ◦ C for 2.5 h. Proteinase K was inactivated by heating at 95 ◦ C for 10 min. A 10 ␮l aliquot of cell lysate, corresponding to 2 × 104 cells, was used in PCR amplification of the immunoglobulin heavy chain lo-

T. Shimizu et al. / Immunology Letters 98 (2005) 259–264

cus using primers VHJ558FR3: GGAGAATTCAGCCTGACATCTGAGGACTCTGC and JCint: GGAGCGGCCGCTCCACCAGACCTCTCTAGACAGC. A high-fidelity enzyme, KOD plus DNA polymerase (TOYOBO, Japan), was used for PCR amplification. Amplified products were digested with NotI and EcoRI, and fragments containing VH J558-JH 4 rearrangements were purified by agarose gel electrophoresis and cloned into pBluescriptIIKS+ plasmid vector (Stratagene, CA). The inserts were subjected to DNA sequence analysis using the DNA sequencer, ACE-2000XLE (Beckman Coulter, CA).

3. Results Pol␬–Pol␫ double-deficient mice were obtained by crossing 129 SvJ mice with Polk knockout mice. The doubledeficient mice were generated with normal Mendelian segregation, looked normal, and were fertile. When B and T cell populations in bone marrow and spleen cells were analyzed by flow cytometry, there was no significant difference between Polk single- and Polk–Poli double-mutant mice (data not shown). Ten-week-old mice, all of which had the targeted Polk locus and a different Poli locus, were immunized by an intraperitoneal injection of NP-CGG with alum. Two weeks after the immunization, mice were sacrificed. Anti-NP antibody in the serum was analyzed by ELISA and there was no significant difference between Polk single- and Polk–Poli double-deficient mice (data not shown). Spleen cells of 2–4 mice of the same genotype were combined for cell sorting. From four Polk−/− Poli+/+ and three Polk−/− Poli+/− mice, cells of the same genotypes were combined. Five Polk−/− Poli−/− were immunized and cells of two or three mice were pooled. GC B cells and na¨ıve B cells were purified using a cell sorter as described in Section 2. For the analysis of SHM, the immunoglobulin heavy chain gene was amplified by PCR. Primers used for amplification were specific to framework region 3 of the J558 VH gene family and to the intron between JH 4 and C␮. Amplified DNA containing VH -JH 4 rearrangements was ligated into a plasmid vector and transfected into bacteria. Colonies were randomly picked out and plasmid DNAs were purified and sequenced. Sequences of non-functional genes (out-of-frame rearrangements or stop codons in VDJ junctions) were ignored. Most of the sequences were unique, although there were a few identical or very similar sequences which were eliminated. In order to avoid the possibility of antigen-derived selection, a 537-bp intron region, which began just downstream of JH 4 and ended at the JCint primer-annealed site, was analyzed for SHM. The sequence registered in the GenBank database (accession numbers: J00440 and J00480) was used as the germ-line sequence for comparison. From all genotypes, 20 sequences from na¨ıve B cells were compared to the germ line sequence and no mutation was noted except for five nucleotide variations which might have been due to polymorphism between strains. This

261

Fig. 1. Frequency of mutations. Numbers of mutations within a 537-bp region are shown. Total numbers of clones analyzed were 49 and 47 for Polk−/− Poli−/− and Polk−/− mice, respectively. Average numbers of mutations per clone with standard deviations are shown in the box. Total numbers of mutations found in the 537-bp region were 309 (49 clones) in Polk−/− Poli−/− mice and 234 (47 clones) in Polk−/− mice.

suggested that errors occurring during PCR and the cloning process could be ignored. For GC B cells, more than 20 clones of each group were sequenced. By comparing with the germ-line sequence, numbers and positions of mutations were identified. Because there was no significant difference between Polk−/− Poli+/− and Polk−/− Poli+/+ mice, these sequences were combined as Polk−/− mice. The two groups of Polk−/− Poli−/− mice did not show any significant difference, thus their sequences were combined as well. Most of the mutations were single-base substitutions. Three clones from Polk−/− Poli−/− mice and two from Polk−/− mice had deletion mutations of 1–16 bp. One clone from Polk−/− Poli−/− mice and another from the Polk−/− mice had a 1-bp insertion. The numbers of base-substitution mutations in each of the determined sequences are summarized in Fig. 1. The average numbers of mutations were 6.3 and 5.0 for Polk−/− Poli−/− and Polk−/− mice, respectively. This difference was not statistically significant, thus suggesting that the absence of Pol␫ did not affect the frequency of SHM. The positions of mutations are shown in Fig. 2. SHM hot spots were highly mutated, especially the region close to JH 4. These residues were mutated in Polk−/− Poli−/− and Polk−/− mice at a similar frequency. Base-substitutions are summarized in Table 1. The ratios of transitions to transversions were also similar in Polk−/− Poli−/− and Polk−/− mice. In Polk−/− Poli−/− mice, 47.5% of changes were at A/T bases, and 52.5% were at G/C bases. The A/T- and G/C-targeted mutations in Polk−/− Table 1 Nucleotide substitutions Polk−/− Poli−/− A A C G T

C 6.0

3.2 13.7 5.0

10.2 11.2

Polk−/− G

T

13.2 7.0

8.7 14.1 4.4

3.4

A 5.8 13.8 8.9

C

G

T

5.7

13.8 5.1

8.1 13.8 4.7

8.6 8.9

2.8

Nucleotide substitutions of mutations were analyzed. Percentages of each substitution are shown. Data were normalized with the nucleotide composition of the sequenced region. Percent transition was 52.2 for Polk−/− Poli−/− mice and 50.2 for Polk−/− mice. Percent transversion was 47.8 for Polk−/− Poli−/− mice and 49.8 for Polk−/− mice.

262

T. Shimizu et al. / Immunology Letters 98 (2005) 259–264

Fig. 2. Location of mutations. The nucleotide sequence of the germ-line JH 4-C␮ intron is shown in the middle. RGYW motifs and the complementary sequences (WRCY) are overlined and underlined, respectively. For each nucleotide, the frequency of clones with substitutional mutations is shown as a percentage. The upper and lower panels show the mutations in Polk−/− Poli−/− and Polk−/− mice, respectively.

mice were 48.2 and 51.8%, respectively. All these results suggested that there was no remarkable difference between Polk−/− Poli−/− and Polk−/− mice. Thus, we concluded that Pol␬–Pol␫ double-deficient mice showed normal SHM. 4. Discussion In this study, we analyzed SHM of the immunoglobulin heavy chain gene in Pol␬–Pol␫ double-deficient mice. The

function of Pol␫ in SHM is still unclear because two contrary results have been reported [19,20]. We expected that the absence of other Y-family polymerase, in our case Pol␬, would exaggerate the effect of Pol␫ deficiency in vivo. However, the frequency and profile of SHM in Polk−/− Poli−/− mice were similar to that of Polk single knockout mice. It has been shown that Pol␫ frequently incorporates G at template T in vitro, resulting in an A to G mutation [22]. Pol␫ inserts T and G opposite to dU that will be generated by cytidine

T. Shimizu et al. / Immunology Letters 98 (2005) 259–264

deamination by AID [23]. The frequencies of these and other mutations did not decrease in Polk−/− Poli−/− mice (Table 1). Taken together, we conclude that the absence of Pol␫ did not affect SHM of the immunoglobulin heavy chain, even in the absence of Pol␬. It is thought that SHM is triggered by DNA lesions generated by AID which causes a dC to dU change in the WRC sequence at the hot spot [10]. The dU lesion, if not removed by UNG, would simply result in generation of a C:G to T:A change [24]. Alternatively, dU may be converted to an abasic lesion by UNG. An abasic lesion is usually repaired via a base excision repair (BER) pathway [25,26]. When high-fidelity replicative DNA polymerases (Pol␦ and/or Pol␧) encounter this abasic site before BER removes it, they cannot bypass the lesion. The stalled replicative DNA polymerase may be transiently replaced by TLS DNA polymerases such as one of the Y-family enzymes. REV1 is known to insert a C opposite to an abasic site and Pol␨ is thought to extend it [27], generating a C:G to G:C change. Some TLS enzymes such as Pol␩, Pol␫, and Pol␬ can preferentially insert A, and less frequently and with varying efficiencies, other bases opposite to the abasic site [28–30]. These enzymes cause a change from C:G to T:A as well as other mutations. However, such enzymes are believed to copy only a short region and to be replaced again by high-fidelity replicative DNA polymerases to avoid gratuitous mutations on non-damaged DNA templates [31], which would explain the appearance of mutations only at C:G pairs in contrast to SHM which occurs equally at C:G and A:T pairs. Therefore, one has to assume that SHM must involve an additional mechanism(s) other than that described above. In this context, it is noteworthy that mice deficient in genes involved in mismatch repair (MMR) show a decreased SHM frequency, especially at A:T pairs [32–35]. In a normal MMR pathway, a single or very short base pair mismatch generated during replication is recognized by the MSH2MSH6 heterodimer. Although the mechanism distinguishing the newly synthesized daughter strand from the parental strand in eukaryotes is still unknown, the region containing the erroneously incorporated base is removed by exonuclease 1, and this portion of the daughter strand is re-synthesized by high-fidelity DNA polymerases [36]. One may well speculate that in SHM, some mismatches between G and U or an abasic lesion might be recognized by the MSH2–MSH6 heterodimer and most parts of the gap region generated by exonuclease 1 are filled by an error-prone DNA polymerase(s), instead of by high-fidelity enzymes via a normal MMR pathway. If this is the case, mutations affecting A:T, which is not a target of DNA lesions induced by AID, can be explained by the fact that the removed region is relatively long with MMR pathways compared to that in BER. The question would then be how an error-prone DNA polymerase(s) would be recruited for participation in SHM. One recent advancement to our understanding of the switching of DNA polymerases at a DNA lesion site is the finding that proliferating cell nuclear antigen (PCNA), which acts as a sliding clamp to increase processing by DNA polymerases, undergoes ubiquitination upon DNA

263

damage [37]. This PCNA modification is considered to be important in polymerase switching because it results in increased affinity to Pol␩ [38,39]. Several reports have pointed out that A:T mutations are decreased in cells of XP-V patients which lack the active Pol␩ [17,40,41]. Therefore, it is speculated that SHM is introduced by Pol␩ which is preferentially recruited as part of the replication machinery of the MMR pathway by means of ubiqutinated PCNA. It should be of great interest to examine SHM in mice defective in Pol␩. Unfortunately, Pol␩-deficient mouse has not been available at the present. Another approach for investigating this possibility is analysis of SHM in mice lacking the RAD18 gene which is critical for ubiquitination of PCNA [39].

Acknowledgements We thank Dr. Roger Woodgate for providing us with valuable information on identifying the Poli mutation in 129 Sv mice, prior to the publication of their results. We thank Mr. Yasushi Hara for performing cell sorting and Dr. William Campbell for critical reading of the manuscript. This work was supported by grant-in-aid for scientific research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to H. Ohmori and to T. Azuma).

References [1] MacLennan IC. Germinal centers. Annu Rev Immunol 1994;12:117–39. [2] Rajewsky K. Clonal selection and learning in the antibody system. Nature 1996;381:751–8. [3] Jacobs H, Bross L. Towards an understanding of somatic hypermutation. Curr Opin Immunol 2001;13:208–18. [4] Papavasiliou FN, Schatz DG. Somatic hypermutation of immunoglobulin genes: merging mechanisms for genetic diversity. Cell Suppl 2002;109:S35–44. [5] Li Z, Woo CJ, Iglesias-Ussel MD, Ronai D, Scharff MD. The generation of antibody diversity through somatic hypermutation and class switch recombination. Genes Dev 2004;18:1–11. [6] Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y, Honjo T. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 2000;102:553–63. [7] Yoshikawa K, Okazaki IM, Eto T, Kinoshita K, Muramatsu M, Nagaoka H, Honjo T. AID enzyme-induced hypermutation in an actively transcribed gene in fibroblasts. Science 2002;296:2033–6. [8] Nagaoka H, Muramatsu M, Yamamura N, Kinoshita K, Honjo T. Activation-induced deaminase (AID)-directed hypermutation in the immunoglobulin S␮ region: implication of AID involvement in a common step of class switch recombination and somatic hypermutation. J Exp Med 2002;195:529–34. [9] Reina-San-Martin B, Difilippantonio S, Hanitsch L, Masilamani RF, Nussenzweig A, Nussenzweig MC. H2AX is required for recombination between immunoglobulin switch regions but not for intraswitch region recombination or somatic hypermutation. J Exp Med 2003;197:1767–78. [10] Pham P, Bransteitter R, Petruska J, Goodman MF. Processive AIDcatalysed cytosine deamination on single-stranded DNA simulates somatic hypermutation. Nature 2003;424:103–7.

264

T. Shimizu et al. / Immunology Letters 98 (2005) 259–264

[11] Ohmori H, Friedberg EC, Fuchs RP, Goodman MF, Hanaoka F, Hinkle D, Kunkel TA, Lawrence CW, Livneh Z, Nohmi T, Prakash L, Prakash S, Todo T, Walker GC, Wang Z, Woodgate R. The Yfamily of DNA polymerases. Mol Cell 2001;8:7–8. [12] Storb U. DNA polymerases in immunity: profiting from errors. Nat Immunol 2001;2:484–5. [13] Schenten D, Gerlach VL, Guo C, Velasco-Miguel S, Hladik CL, White CL, Friedberg EC, Rajewsky K, Esposito G. DNA polymerase ␬ deficiency does not affect somatic hypermutation in mice. Eur J Immunol 2002;32:3152–60. [14] Shimizu T, Shinkai Y, Ogi T, Ohmori H, Azuma T. The absence of DNA polymerase ␬ does not affect somatic hypermutation of the mouse immunoglobulin heavy chain gene. Immunol Lett 2003;86:265–70. [15] Zan H, Komori A, Li Z, Cerutti A, Schaffer A, Flajnik MF, Diaz M, Casali P. The translesion DNA polymerase ␨ play a major role in Ig and bcl-6 somatic hypermutation. Immunity 2001;14:643–53. [16] Diaz M, Verkoczy LK, Flajnik MF, Klinman NR. Decreased frequency of somatic hypermutation and impaired affinity maturation but intact germinal center formation in mice expressing antisense RNA to DNA polymerase ␨. J Immunol 2001;167:327–35. [17] Zeng X, Winter DB, Kasmer C, Kraemer KH, Lehmann AR, Gearhart PJ. DNA polymerase ␩ is an A–T mutator in somatic hypermutation of immunoglobulin variable genes. Nat Immunol 2001;2:537–41. [18] Simpson LJ, Sale JE. Rev1 is essential for DNA damage tolerance and non-templated immunoglobulin gene mutation in a vertebrate cell line. EMBO J 2003;22:1654–64. [19] Faili A, Aoufouchi S, Flatter E, Gueranger Q, Reynaud CA, Weill JC. Induction of somatic hypermutation in immunoglobulin genes is dependent on DNA polymerase iota. Nature 2002;419:944–7. [20] McDonald JP, Frank EG, Plosky BS, Rogozin IB, Masutani C, Hanaoka F, Woodgate R, Gearhart PJ. 129-derived strains of mice are deficient in DNA polymerase ␫ and have normal immunoglobulin hypermutation. J Exp Med 2003;198:635–43. [21] Ogi T, Shinkai Y, Tanaka K, Ohmori H. Pol␬ protects mammalian cells against the lethal and mutagenic effects of benzo[a]pyrene. Proc Natl Acad Sci USA 2002;99:15548–53. [22] Tissier A, McDonald JP, Frank EG. Woodgate R. pol␫, a remarkably error-prone human DNA polymerase. Genes Dev 2000;14:1642–50. [23] Vaisman A, Woodgate R. Unique misinsertion specificity of pol␫ may decrease the mutagenic potential of deaminated cytosines. EMBO J 2001;20:6520–9. [24] Rada C, Williams GT, Nilsen H, Barnes DE, Lindahl T, Neuberger MS. Immunoglobulin isotype switching in inhibited and somatic hypermutation perturbed is UNG-deficient mice. Curr Biol 2002;12:1748–55. [25] Dianov G, Price A, Lindahl T. Generation of single-nucleotide repair patches following excision of uracil residues from DNA. Mol Cell Biol 1992;12:1605–12. [26] Frosina G, Fortini P, Rossi O, Carrozzino F, Raspaglio G, Cox LS, Lane DP, Abbondandolo A, Dogliotti E. Two pathways for

[27] [28]

[29] [30]

[31]

[32]

[33]

[34]

[35]

[36] [37]

[38]

[39]

[40]

[41]

base excision repair in mammalian cells. J Biol Chem 1996;271: 9573–8. Nelson JR, Lawrence CW, Hinkle DC. Deoxycytidyl transferase activity of yeast REV1 protein. Nature 1996;382:729–31. Zhang Y, Yuan F, Wu X, Rechkoblit O, Taylor JS, Geacintov NE, Wang Z. Error-prone lesion bypass by human DNA polymerase ␩. Nucl Acids Res 2000;28:4717–24. Zhang Y, Yuan F, Wu X, Taylor JS, Wang Z. Response of human DNA polymerase ␫ to DNA lesions. Nucl Acids Res 2001;29:928–35. Ohashi E, Ogi T, Kusumoto R, Iwai S, Masutani C, Hanaoka F, Ohmori H. Error-prone bypass of certain DNA lesions by human DNA polymerase ␬. Genes Dev 2000;14:1589–94. McCulloch SD, Kokoska RJ, Masutani C, Iwai S, Hanaoka F, Kunkel TA. Preferential cis–syn thymine dimmer bypass by DNA polymerase ␩ occurs with biased fidelity. Nature 2004;428:97–100. Phung QH, Winter DB, Cranston A, Tarone RE, Bohr VA, Fishel R, Gearhart PJ. Increased hypermutation at G and C nucleotides in immunoglobulin variable genes from mice deficient in the MSH2 mismatch repair protein. J Exp Med 1998;187:1745–51. Rada C, Ehrenstein MR, Neuberger MS, Milstein C. Hot spot focusing of somatic hypermutation in MSH2-deficient mice suggests two stages of mutational targeting. Immunity 1998;9:135–41. Wiesendanger M, Kneitz B, Edelmann W, Scharff MD. Somatic hypermutation in MutS homologue (MSH)3-, MSH6-, and MSH3/MSH6-deficient mice reveals a role for the MSH2-MSH6 heterodimer in modulating the base substitution pattern. J Exp Med 2000;191:579–84. Martin A, Li Z, Lin DP, Bardwell PD, Iglesias-Ussel MD, Edelmann W, Scharff MD. Msh2 ATPase activity is essential for somatic hypermutation at A–T base pairs and for efficient class switch recombination. J Exp Med 2003;198:1171–8. Kolodner RD, Marsischky GT. Eukaryotic DNA mismatch repair. Curr Opin Genet Dev 1999;9:89–96. Hoege C, Pfander B, Moldovan GL, Pyrowolakis G, Jentsch S. RAD6-dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature 2002;419:135–41. Kannouche PL, Wing J, Lehmann AR. Interaction of human DNA polymerase ␩ with monoubiquitinated PCNA: a possible mechanism for the polymerase switch in response to DNA damage. Mol Cell 2004;14:491–500. Watanabe K, Tateishi S, Kawasuji M, Tsurimoto T, Inoue H, Yamaizumi M. Rad18 guides pol␩ to replication stalling sites through physical interaction and PCNA monoubiqutination. EMBO J 2004;23:3886–96. Faili A, Aoufouchi S, Weller S, Vuillier F, Stary A, Sarasin A, Reynaud CA, Weill JC. DNA polymerase ␩ is involved in hypermutation occurring during immunoglobulin class switch recombination. J Exp Med 2004;199:265–70. Zeng X, Negrete GA, Kasmer C, Yang WW, Gearhart PJ. Absence of DNA polymerase ␩ reveals targeting of C mutations on the nontranscribed strand in immunoglobulin switch regions. J Exp Med 2004;199:917–24.