Involvement of Vertebrate Polκ in Translesion DNA Synthesis across DNA Monoalkylation Damage*

DNA lesions that escape excision repair pathways can cause arrested DNA replication. This replication block can be processed by translesion DNA synthesis (TLS), which is carried out by a number of specialized DNA polymerases. A sequential lesion bypass model has been proposed; one of the lesion-specific polymerases inserts nucleotide(s) opposite the damaged template, followed by extension from the inserted nucleotide by the same or another polymerase. Polζ and Polκ have been proposed as candidates for executing the extension step in eukaryotic cells. We previously disrupted separately Rev3, the catalytic subunit of Polζ, and Polκ in chicken B lymphocyte DT40 cells. We found that each cell line showed significant UV sensitivity, implying that both contribute to UV radiation damage repair. In the present studies we generated REV3–/–POLK/– double knock-out cells to determine whether they participate in the same or different pathways. The double mutant was viable and proliferated with the same kinetics as parental REV3–/– cells. The cells showed the same sensitivity as REV3–/– cells to UV, ionizing radiation, and chemical cross-linking agents. In contrast, they were more sensitive than REV3–/– cells to monofunctional alkylating agents, even though POLK/– cells barely exhibited increased sensitivity to those. Moreover Polk-deficient mouse embryonic stem and fibroblast cells, both of which have previously been shown to be sensitive to UV radiation, also showed moderate sensitivity to methyl methanesulfonate, a monofunctional alkylating agent. These data imply that Polκ has a function in TLS past alkylated base adducts as well as UV radiation DNA damage in vertebrates.

Chromosomal DNA in living organisms is continually exposed to a variety of genotoxic agents from exogenous and endogenous sources. Unrepaired DNA damage can lead to replication fork arrest and the formation of gaps and breaks in sister chromatids. Such DNA lesions are processed by two major postreplicational repair pathways: homologous recombination repair and translesion DNA synthesis (TLS) 2 (1). While homologous recombination repair promotes DNA synthesis by facilitating recombination between damaged sister chromatids with the other intact ones, TLS functions by synthesizing DNA past lesions using a number of specialized DNA polymerases (2)(3)(4).
A number of TLS polymerases have been identified in yeasts and mammals. Pol, Pol, and Pol are conserved between species, while Pol is lacking in budding yeast. Biochemical studies have suggested that lesion bypass is effected by two sequential nucleotide incorporation events (4). The first step is insertion of nucleotides opposite the damaged template nucleotide(s), and the second step is extension from the inserted nucleotide(s). While mammalian and yeast Pol efficiently promotes replication through major UV-induced DNA lesions such as cyclobutane pyrimidine dimer both as an inserter and as an extender (5)(6)(7), Pol appears to act more efficiently at extending from the inserted nucleotide(s).
Yeast and mammalian Pol is comprised of the Rev3 catalytic subunit and the Rev7 subunit (8 -10). Rev1, a member of the Y family of DNA polymerases, has limited deoxycytidyltransferase activity (8,11) and can physically interact with Rev7 as well as with other Y family polymerases, such as Pol, Pol, and Pol (12)(13)(14). We previously characterized REV3 Ϫ/Ϫ cells from the chicken B lymphocyte DT40 cell line, and observed that Rev3 is involved in the maintenance of chromosomal DNA as well as in the tolerance of various types of DNA damage (15). We also observed that Rev1, Rev3, and Rev7 may act as a functional unit in cellular tolerance of a variety of genotoxic stresses, as do the orthologous yeast proteins (16).
Pol is a member of the Y family of DNA polymerases together with Pol, Pol, and Rev1 (17). Purified human Pol can bypass adducts of N-2-acetylaminofluorene and benzo[a]pyrene, and templates containing 8-oxoguanine, abasic sites, and 1,N 6 -ethenodeoxyadenosine (18 -22). In addition, Pol efficiently extends various nucleotides incorporated opposite O 6 -methyl guanine and 8-oxoguanine (23). Therefore, it has been suggested that the role of Pol in TLS is as an extender, but at present in vivo evidence in support of this suggestion is lacking.
In this study, we generated REV3 Ϫ/Ϫ POLK /Ϫ double knock-out cells and analyzed their sensitivity to various DNA-damaging agents. The double mutant cells displayed the same sensitivity to UV radiation, ionizing radiation (IR), and cross-linking agents as did parental REV3 Ϫ/Ϫ cells. However, they showed increased sensitivity to monofunctional alkylating agents. Additionally, Polk-deficient mouse embryonic stem (ES) cells and embryonic fibroblasts (MEF) showed a moderate increase in sensitivity to killing following exposure to the monofunctional alkylating agents methyl methanesulfonate (MMS). Collectively, these observations suggest that Pol might have a conserved in vivo function in TLS past monoalkylated bases in vertebrates. * This work was supported by a grant from the Core Research for Evolution Science and Technology (CREST) of Japan Science and Technology Agency and by a Center of Excellence (COE) grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed.

MATERIALS AND METHODS
Cell Culture and Gene Targeting-The conditions for cell culture, selection, and DNA transfections of DT40 cells have been described previously. The REV3 Ϫ/Ϫ POLK /Ϫ double knock-out cells were produced from REV3 Ϫ/Ϫ cells (15) by disrupting the single POLK gene in the sex chromosome in DT40 cells with a chicken POLK disruption construct containing a puromycin-resistant selection marker cassette (24). REV1 Ϫ/Ϫ POLK /Ϫ double knock-out cells were generated from REV1 Ϫ/Ϫ cells (25) with a POLK disruption construct containing a histidinol-D-resistant selection marker cassette (24). To express chicken POLK, a full-length chicken POLK was inserted into the expression vector p176 (26), linearized with PvuI, and transfected into the cells. Generation of Polk-targeted mouse ES cells and MEFs have been described previously (27,28).
Analyses of Chromosome Aberrations and Sister Chromatid Exchange Events-Measurement of chromosome aberrations and sister chromatid exchanges was performed as described previously (24).
Colony Formation Assay-Colony formation assay of DT40 cells with media containing methylcellulose was performed as previously described for measuring sensitivities to UV, IR, cisplatin (cis-platinum (II) diaminodichloride), and MMS (24). For exposure of cells to mitomycin C (MMC; Kyowa Hakko Kogyo, Tokyo, Japan) and N-methyl-NЈnitro-N-nitrosoguanidine (MNNG; Nacalai Tesque, Kyoto, Japan), cells were treated at 39.5°C in 1 ml of complete medium containing MMC or MNNG for 1 h. Exposure of cells to ethyl methanesulfonate (Nacalai Tesque) was performed in medium containing no serum for 1 h, followed by addition of serum to recover the cells that are bound to the bottom of the dishes. The sensitivity of Polk-targeted mouse ES cells to MMS was determined by measuring the colony-forming efficiency of cells treated over a range of doses as described (27). Wild-type TT2 and Polk-targeted ES cells were grown on feeder plates supplied with ES medium; ES medium contains 15% fetal calf serum and 1,000 units/ml Leukemia inhibitory factor in DMEM. Before MMS treatment, cells were harvested, replated on feeder plates, and allowed to adhere to plates for 6 h. After 16-h MMS treatment in ES medium, cells were washed with phosphate-buffered saline once, and cultured for 10 days in ES medium, then fixed, stained, and counted. For exposure of wild-type and Polk-deficient MEFs to MMS, harvested cells were plated in DMEM supplemented with 15% fetal calf serum and allowed to adhere for 2 h. MMS treatment was performed in DMEM for 16 h. Viable colonies were counted after 2 weeks culture.
observation suggests that although both Rev3 and Pol play important roles in maintaining the integrity of chromosomal DNA during the cell cycle, Pol cannot fulfill this function in the absence of Rev3. REV3 Ϫ/Ϫ POLK /Ϫ and REV3 Ϫ/Ϫ Cells Have the Same Sensitivity to UV Radiation, IR, and Cross-linking Agents-To analyze the DNA repair capacity of REV3 Ϫ/Ϫ POLK /Ϫ cells, we examined viability of cells after various genotoxic treatments using colony formation assays. POLK /Ϫ cells are sensitive to UV radiation, but we did not observe a significant increase in the sensitivity of REV3 Ϫ/Ϫ POLK /Ϫ cells to this agent ( Fig. 2A). Disruption of POLK in wild-type or in REV3 Ϫ/Ϫ cells did not increase sensitivity to IR exposure and cross-linking agents (Fig. 2,  B-D). These observations suggest that Pol and Pol operate in the same pathway in the response to UV radiation-induced adducts and that Pol does not function in TLS across IR-induced base damage or adducts generated by cross-linking agents.
Elevated Sensitivity of REV3 Ϫ/Ϫ POLK /Ϫ Cells to Monoalkylating Agents-We previously showed that POLK /Ϫ DT40 cells do not display notably increased sensitivity to monofunctional alkylating agents (24). However, in the present studies we observed that disruption of POLK significantly sensitized REV3 Ϫ/Ϫ cells to these agents (Fig. 3, A-C). To confirm that the elevated sensitivities were caused by the disruption of the POLK gene in REV3 Ϫ/Ϫ cells, REV3 Ϫ/Ϫ POLK /Ϫ cells were reconstituted with chicken POLK cDNA. Expression of the chicken POLK gene restored the tolerance of REV3 Ϫ/Ϫ POLK /Ϫ cells efficiently (Fig. 3, A-C).
These results indicate that Pol functions in TLS past monoalkylated adducts.
To determine whether Pol has a role in TLS past monoalkylated adducts in mammalian cells, we examined Polk-deficient mouse ES cell lines. Wild-type, Polk ϩ/Ϫ , and two independent Polk Ϫ/Ϫ cells were exposed to MMS. Polk ϩ/Ϫ cells exhibited the same sensitivity as wildtype cells. However, Polk Ϫ/Ϫ cells showed moderately increased sensitivity to killing by MMS (Fig. 4A). Increased MMS sensitivity was observed also in Polk-deficient MEF cells (Fig. 4B). Thus, Pol may have a role in bypassing DNA lesions induced by monofunctional alkylating agents. Furthermore, the data suggest complex functional interactions between Pol and Rev3, since the two genes are epistatic for UV radiation sensitivity, whereas Pol can act as a back up for Rev3 in cellular tolerance to MMS. This complex interaction led us to examine functional interactions between Pol and Rev1 by generating REV1 Ϫ/Ϫ -POLK /Ϫ double mutant cells.
Phenotypic Comparison between REV1 and REV3 Disruption in a POLK /Ϫ Background-It has been reported that the Rev1 protein binds to both Rev7 and Pol, which implies that these genes may have functional relationships (12)(13)(14). Despite this biochemical result, we recently reported that REV1 Ϫ/Ϫ , REV3 Ϫ/Ϫ , and REV7 Ϫ/Ϫ DT40 cells and triple mutant cells show the same sensitivity to MMS, IR, UV radiation, and cross-linking agents (16). We therefore concluded that Rev1 may act cooperatively with Rev3 but perhaps not with other DNA polymerases. To determine whether or not Rev1 and Pol function inde-FIGURE 2. The same sensitivities of REV3 ؊/؊ and REV3 ؊/؊ POLK /؊ cells to UV, IR, and cross-linkers. The indicated genotypes of cells were exposed to UV (A), ␥-rays (B), cisplatin (cis-platinum (II) diaminodichloride) (CDDP) (C), and MMC (D), and the sensitivities were measured by colony formation assay. Note that cells were exposed to the crosslinkers for 1 h. The doses and concentrations of genotoxic agents are displayed on the x axis on a linear scale, while the fractions of surviving colonies are displayed on the y axis on a logarithmic scale. Error bars show the S.E. for at least three independent experiments. WT, wild-type. FIGURE 3. REV3 ؊/؊ POLK /؊ cells showed higher sensitivities than REV3 ؊/؊ cells to monoalkylating agents. The indicated genotypes were exposed to ethyl methanesulfonate (EMS) (A), MMS (B), and MNNG (C) for 1 h, and the sensitivities were measured by colony formation assay as in Fig. 2. Expression of the chicken POLK in REV3 Ϫ/Ϫ POLK /Ϫ cells restored the tolerances. WT, wild-type. pendently of each other as do Rev3 and Pol, we generated REV1 Ϫ/Ϫ -POLK /Ϫ mutant cells. Disruption of both REV1 and POLK was not lethal to the cells, and four independent REV1 Ϫ/Ϫ POLK /Ϫ mutant clones were obtained. Targeting events were verified by the appearance of a smaller band and the disappearance of the band that appears in wild-type cells (Fig. 5A). All the REV1 Ϫ/Ϫ POLK /Ϫ clones proliferated with slower kinetics compared with wild-type and POLK /Ϫ cells and with the same kinetics as did the REV1 Ϫ/Ϫ mutant (Fig. 5B). The MMS sensitivity of REV1 Ϫ/Ϫ POLK /Ϫ cells was significantly more severe than that of REV1 Ϫ/Ϫ cells (Fig. 5C). Furthermore, REV1 Ϫ/Ϫ POLK /Ϫ and REV3 Ϫ/Ϫ -POLK /Ϫ cells exhibited the same sensitivity as did REV1 Ϫ/Ϫ and REV3 Ϫ/Ϫ cells to MMS (compare Figs. 3B and 5C). Similarly increased sensitivity of REV1 Ϫ/Ϫ POLK /Ϫ cells was observed with other monofunctional alkylating agents (data not shown). These results suggest that Pol has some functions that are independent of both Rev1 and Rev3.

DISCUSSION
In this study we performed a phenotypic analysis of REV3 Ϫ/Ϫ POLK /Ϫ double mutant DT40 cells. The cells grew with the same kinetics as parental REV3 Ϫ/Ϫ cells. The slightly slower growth rates of REV3 Ϫ/Ϫ -POLK /Ϫ and the parental REV3 Ϫ/Ϫ cells than wild-type cells correlate with the increased levels of spontaneous chromosomal aberrations. However, they showed a distinct difference in their response to UV radiation and to alkylating agents. POLK /Ϫ single mutants are sensitive to UV radiation, but the REV3 Ϫ/Ϫ POLK /Ϫ double mutants are no more sensitive than REV3 Ϫ/Ϫ single mutants. In contrast POLK /Ϫ single mutants are not notably sensitive to alkylating agents, but surprisingly the double mutant is severely more sensitive than the REV3 Ϫ/Ϫ single mutant. Similar results were obtained using a REV1 Ϫ/Ϫ instead of REV3 Ϫ/Ϫ mutant background. A function of Pol in bypassing DNA lesions by monofunctional alkylating agents is also suggested by the increased sensitivity of mouse ES and MEF cell lines to monofunctional alkylating agents. We therefore infer that Pol plays a role(s) in TLS past alkylation damage, but when Rev3 and Rev1 are present this role is of minor significance in DT40 cell lines (but may be more important in mouse cell lines, probably due to the difference of the preference of polymerase usage between cell lines). However, when the Rev1-Rev3mediated pathway is absent, the Pol-dependent pathway is called into play. Thus loss of both Pol and either Rev3 or Rev1 results in hypersensitivity.
In contrast the results with UV radiation suggest that when dealing with photoproducts (PPs) Pol and Pol act in the same pathway. Cur-rently available data suggest that, at least in mammalian cells, Pol carries out both the insertion and extension steps of TLS past cyclobutane pyrimidine dimers, and there is no obvious requirement for other specialized polymerases (5,6). However, Pol cannot carry out TLS past 6-4 PPs, so Pol and Pol may be involved in TLS past this lesion. Biochemical studies have revealed that purified Pol is unable to bypass UV radiation-induced photoproducts on its own in vitro (29), but it is capable of extending from a G opposite the 3Ј-T of a T-T dimer (30). It is possible that it can also extend from a nucleotide inserted opposite the 3Јbase of a 6-4 PP, although there are no data available on this. If this is indeed the case, our present results and the biochemical studies just mentioned can be explained by a model in which Pol functions as an extender during TLS past 6-4 PP and Pol acts as both inserter and extender. This model can reconcile the discrepancy that while Pol cannot bypass UV photoproducts in vitro, POLK-deficient DT40 and mouse ES cell lines are sensitive to UV radiation. Clearly, further work is required to decide between this and other hypotheses for the function of Pol in response to different DNA-damaging agents.
In conclusion, we took advantage of the high targeting efficiency of DT40 and generated REV3 Ϫ/Ϫ POLK /Ϫ and REV1 Ϫ/Ϫ POLK /Ϫ double knock-out cells. By analyzing the mutant cells, we have identified an in vivo function of Pol in bypassing DNA lesions produced by monofunctional alkylating agents. Relationships with other TLS polymerases, such as Pol encoded by RAD30/XPV gene, will be analyzed in the future.  3-6). The digested DNA was hybridized with probe DNA as described in the legend to Fig. 1A. B, growth kinetics corresponding to the indicated cells cultures. C, increased MMS sensitivity of REV1 Ϫ/Ϫ POLK /Ϫ cells than REV1 Ϫ/Ϫ cells. The indicated genotype of cells was exposed to MMS, and the sensitivities were measured by colony formation assay. REV1 and REV3 disruption exhibited the same sensitivities both in a wild-type and POLK /Ϫ background. The concentrations of MMS are displayed on the x axis on a linear scale, while the fractions of surviving colonies are displayed on the y axis on a logarithmic scale. Error bars show the S.E. for at least three independent experiments.