Y265C DNA Polymerase Beta Knockin Mice Survive Past Birth and Accumulate Base Excision Repair Intermediate Substrates

Y265C DNA polymerase beta knockin mice survive past birth and accumulate base excision repair intermediate substrates

Alireza G. Senejani, Shibani Dalal, Yanfeng Liu, Timothy P. Nottoli, James M. McGrath, Connor S. Clairmont, and Joann B. Sweasy1

Departments of Therapeutic Radiology and Genetics, Yale University School of Medicine, New Haven, CT 06520

Edited by Philip C. Hanawalt, Stanford University, Stanford, CA, and approved March 15, 2012 (received for review January 18, 2012) DNA is susceptible to damage by a wide variety of chemical agents frequency, and the types of mutations induced include point that are generated either as byproducts of cellular metabolism or mutations and small deletions (13–16). exposure to man-made and harmful environments. Therefore, to Mice completely deficient in pol β die immediately after birth maintain genomic integrity, having reliable DNA repair systems is (17), although SV-40 immortalized mouse embryo fibroblasts important. DNA polymerase β is known to be a key player in the (MEFs) lacking pol β are viable (18). The inviability of the pol β– base excision repair pathway, and mice devoid of DNA polymerase deficient mice has been ascribed to the presence of high levels of beta do not live beyond a few hours after birth. In this study, we apoptotic cell death in the newly generated postmitotic neurons characterized mice harboring an impaired pol β variant. This Y265C (17). In this study, we have constructed the Y265C “knockin” pol β variant exhibits slow DNA polymerase activity but WT lyase mouse model, given the slow polymerase and mutator activity of activity and has been shown to be a mutator polymerase. Mice the Y265C protein. Approximately 40% of homozygous mutant expressing Y265C pol β are born at normal Mendelian ratios. How- mice (c/c) survive past birth, and these mice are significantly ever, they are small, and 60% die within a few hours after birth. smaller than their WT (wt) and heterozygous (c/+) counterparts. Slow proliferation and significantly increased levels of cell death Homozygous mutant embryos have fewer proliferative cells and are observed in many organs of the E14 homozygous embryos increased levels of apoptosis in various organs. Primary MEFs compared with WT littermates. Mouse embryo fibroblasts pre- isolated from these mice exhibit elevated levels of chromosomal pared from the Y265C pol β embryos proliferate at a rate slower abberations, show sensitivity to MMS, and accumulate BER than WT cells and exhibit a gap-filling deficiency during base ex- intermediates, including single- and double-strand breaks (SSBs cision repair. As a result of this, chromosomal aberrations and and DSBs). In combination, our results suggest that pol β is single- and double-strand breaks are present at significantly critical for the repair of endogenous DNA damage and for higher levels in the homozygous mutant versus WT mouse embryo genome maintenance. fibroblasts. This is study in mice is unique in that two enzymatic activities of pol β have been separated; the data clearly demon- Results strate that the DNA polymerase activity of pol β is essential for Homozygous Pol BY265C/Y265C (c/c) Mice Are Small but Born at Expected survival and genome stability. Mendelian Ratios. We used a standard gene targeting approach (19) to replace a WT (wt) Pol B allele with the Y265C (Pol Bc) variant mutagenesis | oxidative DNA damage allele in mouse embryonic stem (ES) cells (20) (S1 Text and Fig. S1). Homozygous Pol Bc/c E11 and E14 embryos and newborn mice NA damage occurs at a rate of 20,000 lesions per cell per day. are approximately 33% ± 2% smaller (P < 0.0001) than their Pol DThus, having a reliable DNA repair system is crucial to B+/+ WT and heterozygous Pol B+/Y265C (c/+) littermates, as maintain the integrity of cellular DNA (1–3). DNA polymerase shown in Fig. 1 A and B. Pol Bc/c mice were born at the expected beta (pol β) is a 39-kDa protein (4) that is a member of the X family Mendelian ratios (Table S1); however approximately 60% of the group of DNA polymerases. pol β is a key player in base excision homozygotes died within 24 h after birth. Postmortem analysis did repair (BER) (4–6). During BER, damaged bases are recognized not identify gross abnormalities in any organs including the brain, and excised by lesion-specific DNA glycosylases (7). Subsequently, heart, lung, liver, stomach, and kidneys. The total body weight of the DNA backbone is cleaved by AP endonuclease (APE1), cre- the homozygous mice that survived remained significantly (P < ating 3′-OH and 5′-deoxyribose phosphate (5′-dRP) ends. pol β, 0.0001) smaller during the first 3 wk of life relative to those of age- B the major polymerase during BER, fills in the gap by adding nu- matched WT and heterozygous littermates (Fig. 1 ). cleotide/s onto the 3′-OH using polymerase activity, and removes Pol Bc/c the 5′-dRP group using its lyase activity. The nick is then sealed by Organs of Embryos Exhibit High Levels of Cell Death and Low XRCC1-DNA ligase IIIα (8). For oxidative damage, bifunctional Levels of Proliferation. Small mice can result from a decreased cellular proliferation or increased levels of cell death or a combi- glycosylases play a major role and cleave the backbone. The DNA nation of both. First, we characterized cellular proliferation in the ends are remodeled by phosphatases or phosphodiesterases (9) organs of the embryos, using Ki67, and found that there was before pol β fills in the gap. Therefore, pol β functions in the gap- significantly less localization of Ki67 to tissues from the c/c mice filling step of all short-patch BER subpathways. Results from small-scale studies have shown that about one third of all human tumors express pol β variant proteins (10). We Author contributions: A.G.S., S.D., T.P.N., J.M.M., and J.B.S. designed research; A.G.S., S.D., have shown that some of these tumor-associated variant forms of Y.L., T.P.N., and C.S.C. performed research; A.G.S. and J.B.S. analyzed data; and A.G.S. and pol β can induce a mutator phenotype and chromosomal aber- J.B.S. wrote the paper. rations by either error-prone or slow gap filling, respectively (10– The authors declare no conflict of interest. 12). The Y265C pol β variant was identified in a gastric carci- This article is a PNAS Direct Submission. noma and has been shown to be a strong mutator polymerase, 1To whom correspondence should be addressed. E-mail: [email protected]. both in vitro and in vivo (13, 14). In murine LN12 cells, ex- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. pression of Y265C leads to an eightfold increase in mutation 1073/pnas.1200800109/-/DCSupplemental.

6632–6637 | PNAS | April 24, 2012 | vol. 109 | no. 17 www.pnas.org/cgi/doi/10.1073/pnas.1200800109 Downloaded by guest on September 30, 2021 Fig. 1. Pol BY265C/Y265C (c/c) mice are small. (A) Pol BC/C embryos were found to be 33% ± 2% smaller than WT littermates at embryo day E11 and E14. No other noticeable physiological abnormalities were evident in any of the embryos. (B) The average body weight of the Pol Bc/c 1-d-old pups is approximately 1 g (Inset), whereas the Pol B+/+ and Pol Bc/+ littermates were ∼1.4 g. The average weights of mice between days 1 and 21 shows that the Pol BC/C mice gain weight at a slower rate compared with their Pol B+/+ and Pol Bc/+ littermates. Data represent the mean weight of ﬁve to eight mice, with error bars showing the SEM.

(P < 0.0001) as shown in Fig. 2A. To explore the possibility that accumulation of c/c MEFs in G2/M suggested that the DNA organs of the c/c mice exhibit increased levels of cell death, we used damage response was activated in these cells, likely during DNA terminal deoxytransferase-mediated dUTP biotin nick end label- replication. Our results showing that CHK1 protein is phosphor- ing (TUNEL staining) in organs from the embryos. Significantly ylated to a greater extent in the c/c versus WT MEFs supports this higher levels of TUNEL staining were observed in many organs of idea (Fig. 3C). Taken together, our results both in organs and in the E14 homozygous embryos compared with WT, as shown in Fig. MEFs suggest that the small size of the mice is due to a combina- 2B. The brains of the Pol Bc/c E14 embryo exhibited almost a 40- tion of decreased proliferation and increased cell death and may fold increase in cell death compared with their c/+ and WT lit- result from replication fork stalling, collapse, or stress. In- termates, similar to what has been shown in the Pol B knockout terestingly, growth of the c/c MEFs in 5% oxygen yielded nearly embryos. Although some cell death was observed in the fetal livers equivalent proliferation to WT MEFs grown in 20% oxygen (Fig. of the WT and c/+ embryos, there was sevenfold more in the fetal S4), suggesting that reactive oxygen species are likely responsible livers of the c/c littermates. Several other organs including the for the slow growth phenotype that we observed in the c/c MEFs. lung, heart, stomach, and kidney were also found to have relatively higher levels of TUNEL-positive cells in the c/c embryos . Caspase- Pol Bc/c MEFs Display Increased Levels of Chromosomal Aberrations. 3 localization to these organs suggested an apoptotic mechanism Given that Y265C catalyzes DNA synthesis at a slower rate than for cell death (Fig. S2). In combination, these results indicate that WT (15, 21), we speculated that unfilled gaps could lead to both slow proliferation and increased cell death contribute to the increases in the numbers of chromosomal aberrations per cell. small size of the c/c mice. Therefore, we analyzed metaphase spreads from each genotype. The Pol Bc/c MEFs exhibited a significant increase in chromo- Mouse Embryonic Fibroblasts Generated from Pol Bc/c Mice Proliferate some fragments, as well as interchromosome and intrachro- Slowly and Accumulate in G2/M. We also characterized pro- mosome fusions, as shown in Fig. 4. In contrast, relatively liferationandcelldeathinMEFsisolatedfrom14-d-oldem- limited numbers of chromosomal fragments and fusions were bryos. To examine cellular proliferation, we compared the observed in the metaphase spreads of WT and c/+ MEFs. These growth of c/c to c/+ and WT MEFs using early passage (P0 or P1) results show that increased levels of genomic instability are cells. As shown in Fig. 3A, the Pol Bc/c MEFs double at about half present in primary c/c MEFs. the rate of MEFs isolated from their Pol Bc/+ and Pol B+/+ littermates. As demonstrated in Fig. S3, apoptosis was slightly in- Y265C MEFs Are Deficient in BER. Many types of chromosomal creased in MEFs isolated from the c/c versus WT mice. aberrations arise from DNA breaks. We hypothesized that cells Interestingly, we show that at day 5, approximately twofold greater expressing only Y265C protein would be unable to completely fill numbers of c/c MEFs accumulate in G2/M compared with WT all small gaps that arise in DNA during BER, leading to the MEFs (Fig. 3B).The slower proliferation rate and greater accumulation of DNA breaks. To test this hypothesis, we first

Fig. 2. Organs from Pol Bc/c embryos exhibit slow proliferation and high levels of cell death. (A)Im- munohistochemistry results using the Ki67 marker showed cells in c/c embryos have fewer proliferative cells than WT littermates. Percentages are calculated by counting number of positively stained cells over the total number of cells using 10–15 images (×40) of immunohistochemistry sections representative of the same organ. (B) Immunohistochemistry sections of E14 embryos from various organs of homozygotes show high levels of apoptosis (±STD) in represented tissues. (Inset) Immunohistochemistry sections (40×) representative of brain and fetal liver with darker cells indicating the TUNEL-positive labeled cells. The homozygous mutant brains and fetal liver had

40-fold and sevenfold, respectively, increases in TUNEL- GENETICS positive cells compared with WT. Represented P values are calculated using the unpaired t test.

Senejani et al. PNAS | April 24, 2012 | vol. 109 | no. 17 | 6633 Downloaded by guest on September 30, 2021 Fig. 3. Pol Bc/c MEFs exhibit decreased proliferation. (A) Homozygous mutant c/c MEFs were found to grow more slowly compared with WT and heterozygous littermates. Approximately 1.5 × 105 cells from passage 1–2 MEFs were plated and then counted every other day. (B) Pol Bc/c MEFs accumulate in G2/M. Cell cycle analysis was performed by flow cytometry using propidium iodide (PI) to stain the DNA. Graph shows the numbers of cells in various cell cycle phases versus fluorescence intensity. (C) Increased level of phosphorylated ChK1 in c/c versus WT MEFs. (Upper) Images of Western blot; (Lower) quantification of the results. Lane 1, +/+ MEFs; lanes 2, C/C MEFs.

determined whether the c/c cells were sensitive to MMS. MMS overexpression of the dRP Lyase domain in SV40 immortalized leads to alkylation of the N3 of adenine and the N7 of guanine Pol B knockout MEFs restores resistance to MMS, our results in and these adducts are repaired by BER. We expected that primary MEFs suggest that the presence of dRP lyase activity and Y265C would be less able to fill gaps resulting from removal of robust polymerase activity is critical for BER and MMS resistance. DNA damage, leading to cellular sensitivity. As shown in Fig. 5A, Our results indicate that the coupling of polymerase and lyase c/c MEFs are more sensitive to MMS than c/+ or WT MEFs, activities is important for BER. Overexpression of nonphysi- indicating that BER is deficient in these cells. Note that the c/c ological levels of lyase activity could function together with other MEFs do not exhibit increased sensitivity to UV light (Fig. S5). non-BER polymerases during BER to resolve gaps in DNA. In Next, we characterized BER in cellular extracts of c/c, c/+ and addition, immortalized MEFs are likely to exhibit different bi- WT MEFs to determine whether Y265C could support BER. We ological responses to DNA-damaging agents. assessed BER of uracil and hypoxanthine, two base lesions that arise from endogenous deamination of cytosine and adenine, re- Intermediate BER Substrates Accumulate in MEFs. The BER de- fi spectively (3). As shown in Fig. 5 B and C, extracts prepared from c/c ciency of the c/c MEFs is expected to lead to the accumulation MEFs do not support BER of these lesions during the gap-filling of BER reaction intermediates, i.e., single-strand breaks (SSBs) step, even though Y265C is expressed in these extracts (Fig. 5D). and double-strand breaks (DSBs). To determine whether this is Note that n+1 synthesis is significantly delayed in the c/c extract, the case, we assessed the levels of SSBs and DSBs in cells treated having a slower rate than what is observed in the WT and c/+ with MMS. MMS is expected to increase the levels of DNA extracts (Fig. S6). As shown in Fig. 5E, Y265C protein possesses damage, leading to increases in the numbers of single nucleotide fi β dRP lyase activity. Therefore, the gap-filling deficiency of Y265C gaps needed to be lled in by pol Y265C. Results from the Pol Bc/c likely leads to the accumulation of BER intermediates. Although alkaline comet assay indicate the MEFs treated with MMS carry significantly higher levels of SSBs, as the c/c MEFs exhibited larger DNA tail moments versus WT cells after treatment with MMS (Fig. 6A). c/c To determine whether the Pol B MEFs exhibit increased levels of DSBs compared with WT MEFs we counted the numbers of of γH2AX foci in MEFs treated with MMS. The levels of γH2AX foci were also significantly higher in MEFs carrying the mutant allele, even after they were given time to recover (Fig. 6B). As shown in Fig. 6C, the levels of γH2AX 7 h after MMS treatment remain higher in the c/c MEFS versus the WT cells. In fact, a decrease in the levels of γH2AX protein in the WT MEFs but not the c/c MEFs is observed and the levels of this protein increase in the c/c MEFs during this time. We also assessed the levels of γH2AX foci in nontreated cells. Remarkably, the Pol Bc/c MEFs exhibited significantly increased levels of endogenous DSBs compared with MEFs isolated from their Pol B+/+ littermates (Fig. 6D). In combination, these data show that cells expressing only pol β Y265C accumulate BER intermediates at increased levels compared with WT cells.

Homozygous Mutant MEFs Treated with MMS Exhibit Apoptosis. The Fig. 4. Increased levels of chromosomal aberrations of c/c primary MEFs. Di- accumulation of SSBs and DSBs could lead to cell death by verse types of irregular Giemsa-stained chromosomes were counted from the apoptosis. To measure apoptotic events, we costained the MEFs metaphase spreads of more than 100 cells of c/c, c/+, and WT MEFs. As shown, with 7-AAD and annexin V (surface-exposed phosphatidylserine ﬁ < homozygous c/c MEFs contain signi cantly (P 0.05) elevated numbers of marker on apoptotic cells) and measured the signal for 10,000– chromosomal aberrations compared with WT and c/+ MEFs. GraphPad prism. ﬂ Two-way ANOVA was used to determine the P value. (Inset) Representative 20,000 cells by ow cytometry. Upon exposure to MMS, the metaphase spreads from Pol Bc/c MEFs with arrows indicating two chromo- homozygous mutant MEFs display a higher level of apoptosis somal irregularities: a chromosome fragment and a single chromatid break. N, compared with WT cells (47% versus 22% 24 h after treatment; number of metaphase spreads examined for each genotype. Fig. 7). These data suggest that the accumulation of BER

6634 | www.pnas.org/cgi/doi/10.1073/pnas.1200800109 Senejani et al. Downloaded by guest on September 30, 2021 Fig. 5. The Pol Bc/c MEFs show sensitivity to MMS and are BER deﬁcient. (A) Pol Bc/c MEFs are sensitive to MMS. Ap- proximately 1.5 × 105 cells from passage 1–2 MEFs were plated. On the following day, the cells were treated for 1 h with various concentrations of MMS and counted 5 d later. No differences were exhibited to treatment with UV light (Fig. S3). Data represent the mean of three independent experiments with error bars showing SEM. (B and C) Base excision repair assays of MEF cell extracts using either a Uracil (B) or hypoxanthine (C) substrate as described previously (12). BER assays were carried out as described in Experimental Procedures.(Upper) Images of the denaturing gel; (Lower) quantiﬁcation of the results. Lane 1, annealed oligo substrate; lanes 2 and 3, substrate incubated with +/+ extract and in absence and in presence of dNTPs, respectively; lanes 4 and 5, substrate incubated with c/+ extract in absence and presence of dNTPs, respectively; lanes 6 and 7, substrate incubated with c/c extract and in absence and presence of dNTPs, respectively. (D) Western blot analysis shows expression of DNA polymerase β protein in WT (+/+) and homozygous (c/c) mutant MEF cell extracts. (E) Y265C has dRP lyase activity. The 5′-dRP–containing DNA substrate (5 nM) was incubated in buffer R at 37 °C for 20 min with 100 nM of polymerase β (WT or Y265C), followed by the addition of 340 mM NaBH4 and stabilization of the reaction product as indicated in Experimental Procedures. Products were analyzed by a denaturing 20% poly- acryamide gel and visualized by autoradiography.

intermediates leads to apoptotic cell death in addition to polymerase (15). Here we show that homozygous Pol Bc/c mice chromosomal aberrations. are born at normal Mendelian ratios and are small, and that 60% of these mice die 1–2 d after birth. Many of the cells within the Discussion organs of the Pol Bc/c mice exhibit reduced proliferation and The pol β polypeptide has two enzymatic activities: dRP lyase increased levels of cell death compared with WT. Primary MEFs and DNA polymerase. The dRP lyase activity is used to remove isolated from the Pol Bc/c mice also exhibit reduced proliferation, the 5′dRP group that results from incision of the DNA backbone accumulate in G2/M, and have increased levels of phosphory- by APE1 after DNA damage is removed by a monofunctional lated CHK1, indicative of an activated DNA damage response. DNA glycosylase. Using its polymerase activity, pol β ﬁlls small These MEFs also accumulate BER intermediate substrates in- gaps in DNA that result from the removal of DNA damage by cluding SSBs and DSBs, which lead to cell death or genomic both mono- and bifunctional DNA glycosylases. The Y265C instability. These results are consistent with the interpretation polymerase possesses normal dRP lyase activity but is a very slow that normal DNA polymerase activity of pol β is critical for the

Fig. 6. Pol Bc/c MEFs accumulate intermediate BER substrates. (A) Pol Bc/c MEFs show increased levels of SSB. DNA tail moment of 100–125 cells of each genotype was quantified using CometScore software (Tritek). Pol Bc/c MEFs have significantly higher levels of SSBs than WT upon 1 and 2 h of exposure to 1.5 mM MMS. Represented P values are calculated using the unpaired t test. (B)In- duced levels of γH2AX foci in the Pol Bc/c MEFs treated with MMS. Results are the average numbers of foci in 30–50 cells for each time point. Star and arrow show the time point where the MMS was removed (after 60 min exposure) and the cells were washed and allowed to recover for the time denoted. P values are calculated using GraphPad prism, unpaired t test. (C) Treated Pol Bc/c MEFs show increased levels of γH2AX compared with WT. (Up- per) Images of a Western blot; (Lower) Quantification of the results. Lane 1, untreated +/+ MEFs; lanes 2 and 3, +/+ MEFs treated with 1.5 mM MMS for 1 h and recovered for 3 and 7 h, respectively; lane 4, untreated C/C MEFs; lanes 5 and 6, C/C MEFs treated with 1.5 mM MMS for 1 h and recovered for 3 and 7 h, respectively. c/c (D) Pol B MEFs show significantly in- GENETICS creased levels of γH2AX foci compared with WT.

Senejani et al. PNAS | April 24, 2012 | vol. 109 | no. 17 | 6635 Downloaded by guest on September 30, 2021 developmental programming, including active demethylation that involves TDG (23, 24). We suggest that many, if not all, of the dRP moieties are removed by the Y265C pol β during BER, but that only a subset of the small gaps that arise after damage excision by either mono- or bifunctional DNA glycosylases are filled in by Y265C. It is possible that other nonprocessive polymerases, including pol λ, could function in place of pol β Y265C and fill in small gaps in the DNA. The death of 100% of the mice in the complete absence of pol β dRP lyase activity suggests that the dRP activity of pol β during development cannot be performed by another polymerase. Studies have shown that the polymerase activity of the 31-kDa fragment of pol β, lacking the dRP lyase activity, is very low, and that pol β is likely to initially bind to DNA via its 8 kDa domain (25, 26). Thus, we suggest that gaps filled by the Y265C polymerase during development might be the ones at which Y265C binds to and removes the dRP group. It is also possible that pol β is specifically required to fill gaps that occur during development, and that a subset of gaps remain unfilled or partially filled in the mice, leading to their death just after birth.

Accumulation of BER Intermediates in Pol Bc/c MEFs. We have shown that the cells of many organs of Pol Bc/c embryos exhibit reduced proliferation and cell death. Our studies with MEFs isolated from these embryos show that they have DSBs, increased levels Fig. 7. Induced apoptosis in the Pol Bc/c MEFs exposed to methyl meth- of phosphorylated CHK1, and increased numbers of chromo- anesulfonate (MMS). (A) Fold change of early and late apoptotic cells after somal aberrations per cell in the absence of exogenous DNA MMS treatment with 1.5 mM MMS for 1 h followed by recovery for the damage. This suggested to us that BER intermediates, in the represented time. Relative counts of Annexin V + 7-AAD–positive cells were form of single nucleotide gaps and SSBs, accumulate in the cells measured using flow cytometry. (B and C). Effects of MMS treatment on due to incomplete BER. Our tissue culture studies of the early (PE Annexin V staining; lower right) and apoptosis (PE Annexin V and primary MEFs show that they are sensitive to MMS. Because 7-AAD staining; upper right) was measured by flow cytometry. (B) FACs +/+ Y265C possesses dRP lyase activity, our studies suggest that the analysis of Pol B MEFs after 24 h recovery following 1 h treatment with MMS sensitivity results from slow DNA polymerase activity by 1.5 mM MMS. (C) FACs analysis of Pol Bc/c MEFs after 24 h following 1 h treatment with 1.5 mM MMS. Representative examples of two to three Y265C. In support of this suggestion, extracts prepared from the Pol Bc/c separate experiments are shown. MEFs cannot support BER of either uracil or hypoxanthine at the DNA synthesis step. We used the alkaline comet assay to demonstrate that the level of SSBs is higher in the MEFs survival of a majority of mice past birth.. Our results also strongly after treatment with MMS, and we also showed that the Pol Bc/c suggest that BER is a major cellular process during development. MEFs continue to accumulate many more DSBs 3 h after re- covering from MMS treatment. These results suggest that, once Polymerase Activity of pol β Is Critical for Survival of Mice Past Birth. the damaged base is excised, Y265C attempts to fill in the gaps Two versions of pol β knockout mice have been reported (17, 22) but is unable to completely fill in all of the gaps that are pro- and in both cases the mice are small and die shortly after birth. duced by removal of damaged bases. These gaps accumulate, and In each of these studies, the mice were completely deficient of some of them are converted into DSBs, likely during DNA both pol β dRP lyase and polymerase activity, so it was not replication. We suggest that the consequences of DSB accumulation include chromosomal aberrations and cell death. possible to determine whether the dRP lyase or polymerase ac- β tivities, or both, were necessary for survival. The Y265C knockin In summary, we have shown that the pol polymerase activity is critical to prevent the accumulation of BER intermediates and mice that we produced are also small, but 40% of them survive activation of the DNA damage response during development. past birth. Because Y265C exhibits normal dRP lyase activity, β β The consequences of slow pol activity at the organismal level our data suggest that the dRP lyase activity of Pol is critical for include slow proliferation, cell death, and genomic instability. survival of the mice beyond 1 day after birth. However, the fact Our data also support the conclusion that BER is a major cel- that the majority of mice die soon after birth indicates that the lular process during development. DNA polymerase of pol β is also critical for survival. The lyase activity of pol β is necessary for the removal of the 5′ Experimental Procedures dRP group that results from the action of a monofunctional Construction and Genotyping of pol β Y265C Mutant Mice. A standard gene DNA glycosylase followed by incision by APE1. In the sub- targeting approach (19) was used to construct the knockin pol β Y265C mice pathway of BER initiated by monofunctional glycosylases, in- (SI Text and Fig. S1). Animals were genotyped by PCR analysis of tail DNA (DirectPCR, VIAGEN) using the forward primer Y265C-F1: 5′AGAAAAG- cluding UNG1, SMUG, AAG, and TDG, both the polymerase CAGCTTCCAGCAG and reverse primer Y265C-R4: 5′CAGACTTTCCAAGTG- and dRP lyase activities of pol β are required for completion of CAGGAT. The following amplification conditions were used: 95 °C for 3 min BER. In the subpathway of BER initiated by the bifunctional followed by 30–40 cycles of 95 °C denaturation for 30 s, 60 °C annealing for glycosylases including NTH1, NEIL 1,2, and 3, only the poly- 15 s, and 72 °C extension for 1 min; DNA fragments of 440 bp (WT) and/or merase activity of pol β is required for completion of BER. 490 bp (mutant) were expected to be produced. Damage removed by both of these subpathways likely arises Single-Cell Gel Electrophoresis Assay (Alkaline Comet Assay). Equal numbers of during development. Some of the base damage may result from cells (4 × 105) from early passages (P0 or P1) were plated onto a 6-cm plate in the presence of reactive oxygen and nitrogen species or the in- the presence of 5 mL DMEM containing 10% (vol/vol) FBS and 1× penicillin- corporation of uracil into DNA, and some of it may be due to streptomycin. After 1 d, the cells were incubated with media containing

6636 | www.pnas.org/cgi/doi/10.1073/pnas.1200800109 Senejani et al. Downloaded by guest on September 30, 2021 1.5 mM MMS and treated for 1 or 2 h. After treatment, the cells were Preparation of Substrate and 5′-dRP Lyase Assay. The 5′-dRP–containing DNA prepared and analyzed immediately according to published procedures (27) substrate was prepared just before use, as describe previously (28, 29); and using Cometslides (Trevigen catalog no. 4250–200-03). Image analysis of the assay was set up according to Prakash et al. (30) with minor mod- ﬁ 100–125 cells from two to three littermates of each genotype was per- i cations (SI Text). formed using CometScore software (TriTek). Cell Extract Base Excision Repair Assay. MEF extracts were prepared as de- Apoptosis Detection of Live Cells Using Flow Cytometry. Cells were washed scribed previously, and base excision repair assays were conducted with either a UDG or hypoxanthine substrate as described previously (12). BER reactions twice with cold PBS and prepared in a round-bottomed tube (BD Falcon). were quenched by addition of an equal volume of formamide gel loading After brief centrifugation, the cell pellet was resuspended in 1× Annexin V dye [90% (vol/vol) formamide, 0.3 M EDTA]. The percent primer extension ∼ × 6 binding buffer (BD Biosciences) at a cell concentration of 1 10 cells/mL. was calculated by dividing the pixels of the n + 1 product by total pixels [n + PE Annexin V (BD Biosciences catalog no. 556422) and 7-AAD (BD Biosciences (n + 1) + top band] and multiplying by 100. catalog no. 559925) were then added, and the cells were incubated for 15 min at room temperature (in the dark according to the manufacturer’s ACKNOWLEDGMENTS. This work was supported by R01 ES019179 (to J.B.S.) protocol) before analyis by ﬂow cytometry. and an Anna Fuller postdoctoral fellowship (to A.G.S.).

1. De Bont R, van Larebeke N (2004) Endogenous DNA damage in humans: A review of 16. García-Villada L, Drake JW (2010) Mutational clusters generated by non-processive quantitative data. Mutagenesis 19:169–185. polymerases: A case study using DNA polymerase beta in vitro. DNA Repair (Amst) 9: 2. Wallace SS (2002) Biological consequences of free radical-damaged DNA bases. Free 871–878. Radic Biol Med 33:1–14. 17. Sugo N, Aratani Y, Nagashima Y, Kubota Y, Koyama H (2000) Neonatal lethality with fi 3. Barnes DE, Lindahl T (2004) Repair and genetic consequences of endogenous DNA abnormal neurogenesis in mice de cient in DNA polymerase beta. EMBO J 19: – base damage in mammalian cells. Annu Rev Genet 38:445–476. 1397 1404. 4. Loeb LA, Monnat RJ, Jr. (2008) DNA polymerases and human disease. Nat Rev Genet 9: 18. Sobol RW, et al. (1996) Requirement of mammalian DNA polymerase-beta in base- excision repair. Nature 379:183–186. 594–604. 19. Capecchi MR (2005) Gene targeting in mice: Functional analysis of the mammalian 5. Moon AF, et al. (2007) The X family portrait: Structural insights into biological func- genome for the twenty-first century. Nat Rev Genet 6:507–512. tions of X family polymerases. DNA Repair (Amst) 6:1709–1725. 20. Thomas KR, Capecchi MR (1987) Site-directed mutagenesis by gene targeting in 6. Uchiyama Y, Takeuchi R, Kodera H, Sakaguchi K (2009) Distribution and roles of X- mouse embryo-derived stem cells. Cell 51:503–512. – family DNA polymerases in eukaryotes. Biochimie 91:165 170. 21. Dycaico MJ, et al. (1994) The use of shuttle vectors for mutation analysis in transgenic 7. Stivers JT, Jiang YL (2003) A mechanistic perspective on the chemistry of DNA repair mice and rats. Mutat Res 307:461–478. – glycosylases. Chem Rev 103:2729 2759. 22. Gu H, Marth JD, Orban PC, Mossmann H, Rajewsky K (1994) Deletion of a DNA 8. Izumi T, et al. (2003) Mammalian DNA base excision repair proteins: Their interactions polymerase beta gene segment in T cells using cell type-specific gene targeting. Sci- and role in repair of oxidative DNA damage. Toxicology 193:43–65. ence 265:103–106. 9. Wiederhold L, et al. (2004) AP endonuclease-independent DNA base excision repair in 23. Cortázar D, et al. (2011) Embryonic lethal phenotype reveals a function of TDG in human cells. Mol Cell 15:209–220. maintaining epigenetic stability. Nature 470:419–423. 10. Starcevic D, Dalal S, Sweasy JB (2004) Is there a link between DNA polymerase beta 24. Cortellino S, et al. (2011) Thymine DNA glycosylase is essential for active DNA de- and cancer? Cell Cycle 3:998–1001. methylation by linked deamination-base excision repair. Cell 146:67–79. 11. Dalal S, et al. (2005) Prostate-cancer-associated I260M variant of DNA polymerase 25. Maciejewski MW, Liu D, Prasad R, Wilson SH, Mullen GP (2000) Backbone dynamics fi beta is a sequence-specific mutator. Biochemistry 44:15664–15673. and re ned solution structure of the N-terminal domain of DNA polymerase beta. – 12. Lang T, Dalal S, Chikova A, DiMaio D, Sweasy JB (2007) The E295K DNA polymerase Correlation with DNA binding and dRP lyase activity. J Mol Biol 296:229 253. beta gastric cancer-associated variant interferes with base excision repair and induces 26. Sobol RW, et al. (2000) The lyase activity of the DNA repair protein beta-polymerase protects from DNA-damage-induced cytotoxicity. Nature 405:807–810. cellular transformation. Mol Cell Biol 27:5587–5596. 27. Olive PL, Banáth JP (2006) The comet assay: A method to measure DNA damage in 13. Clairmont CA, Narayanan L, Sun KW, Glazer PM, Sweasy JB (1999) The Tyr-265-to-Cys individual cells. Nat Protoc 1:23–29. mutator mutant of DNA polymerase beta induces a mutator phenotype in mouse 28. Dalal S, Chikova A, Jaeger J, Sweasy JB (2008) The Leu22Pro tumor-associated variant – LN12 cells. Proc Natl Acad Sci USA 96:9580 9585. of DNA polymerase beta is dRP lyase deficient. Nucleic Acids Res 36:411–422. 14. Opresko PL, Sweasy JB, Eckert KA (1998) The mutator form of polymerase beta with 29. Kidane D, et al. (2011) DNA polymerase beta is critical for genomic stability of sperm amino acid substitution at tyrosine 265 in the hinge region displays an increase in cells. DNA Repair (Amst) 10:390–397. both base substitution and frame shift errors. Biochemistry 37:2111–2119. 30. Haracska L, Prakash L, Prakash S (2003) A mechanism for the exclusion of low-fidelity 15. Washington SL, et al. (1997) A genetic system to identify DNA polymerase beta human Y-family DNA polymerases from base excision repair. Genes Dev 17: mutator mutants. Proc Natl Acad Sci USA 94:1321–1326. 2777–2785. GENETICS

Senejani et al. PNAS | April 24, 2012 | vol. 109 | no. 17 | 6637 Downloaded by guest on September 30, 2021