Rev3L encodes the catalytic subunit of DNA polymerase ζ (pol ζ) in mammalian cells. In yeast, pol ζ helps cells bypass sites of DNA damage that can block replication enzymes. Targeted disruption of the mouse Rev3L gene causes lethality midway through embryonic gestation, and Rev3L−/− mouse embryonic fibroblasts (MEFs) remain in a quiescent state in culture. This suggests that pol ζ may be necessary for tolerance of endogenous DNA damage during normal cell growth. We report the generation of mitotically active Rev3L−/− MEFs on a p53−/− genetic background. Rev3L null MEFs exhibited striking chromosomal instability, with a large increase in translocation frequency. Many complex genetic aberrations were found only in Rev3L null cells. Rev3L null cells had increased chromosome numbers, most commonly near pentaploid, and double minute chromosomes were frequently found. This chromosomal instability associated with loss of a DNA polymerase activity in mammalian cells is similar to the instability associated with loss of homologous recombination capacity. Rev3L null MEFs were also moderately sensitive to mitomycin C, methyl methanesulfonate, and UV and γ-radiation, indicating that mammalian pol ζ helps cells tolerate diverse types of DNA damage. The increased occurrence of chromosomal translocations in Rev3L−/− MEFs suggests that loss of Rev3L expression could contribute to genome instability during neoplastic transformation and progression. (Cancer Res 2006; 66(1): 134-42)

Sites of DNA damage can block synthesis by DNA polymerases δ and ε, the enzymes responsible for rapid, high-fidelity replication of most of the eukaryotic genome (1). Stalled replication forks can be reactivated by a specialized group of DNA damage bypass polymerases that function to insert bases across from damaged bases, so that the DNA replication apparatus can proceed (2, 3). Many types of DNA adducts may be bypassed by one or more of these polymerases. The switch from a stalled DNA polymerase to one capable of bypass is controlled at least in part by the ubiquitination state of the proliferating cell nuclear antigen (4, 5). Some types of DNA damage are especially challenging to bypass, such as apurinic sites where a base has been lost, or bulky chemical adducts derived from exogenous sources. A DNA polymerase with relaxed template specificity can function to bypass the damaged base at a cost of increased mutation frequency. The most remarkable example of such an enzyme is the DNA polymerase ζ (pol ζ). In the yeast Saccharomyces cerevisiae, pol ζ is composed of catalytic Rev3 and accessory Rev7 subunits (6, 7). The ability to extend mispaired termini may be the most unusual and biologically relevant catalytic activity of this polymerase. Yeast pol ζ can efficiently extend termini left by various DNA polymerases opposite a variety of types of different DNA damage (814).

The absence of REV3 in yeast causes a large decrease in the number of UV light–, γ-ray-, and methyl methanesulfonate (MMS)–induced base pair and frameshift mutations (7). A substantial number of spontaneous mutations are also dependent on pol ζ and presumably result from participation of the enzyme in bypass of endogenous oxidative and hydrolytic damage (15, 16). Pol ζ is also sometimes involved in recombinational repair of strand breaks, in a manner yet to be defined. During recombinational repair of a directed double-strand break in budding yeast, point mutations can arise in the vicinity of the recombination event, and these depend on functional Rev3 (17).

Human and mouse REV3L are 353-kDa proteins with DNA polymerase domains and a sequence required for interaction with REV7 but have a region of ∼1,400 amino acids not found in yeast REV3 (1820). REV3L antisense experiments in HeLa cells show that mammalian DNA pol ζ participates in DNA damage–induced mutagenesis (21). The frequency of UV-induced Hprt mutants is also much reduced in fibroblasts from a Rev3L antisense mouse model, and it seems that most UV-induced mutations in mammals are dependent on pol ζ (22, 23). The misinsertion and/or extension activities of pol ζ may also function during hypermutation of antibody V genes (22, 24) and in the mutator phenotype induced by hepatitis C virus (25).

In these cell and animal antisense suppression models, some detectable expression of REV3L remains. For a definitive investigation of pol ζ function, we and others generated mouse knockout models. Inactivation of Rev3L causes embryonic lethality during midgestation (2628). Tissues in many areas of homozygous null embryos are disorganized with significantly reduced cell density, suggesting a requirement for bypass of endogenous DNA damage by pol ζ. It may be relevant that oxidative damage to DNA could occur with increasing frequency during early midgestation due to the metabolic switch to oxidative phosphorylation that occurs at this point in development (29, 30).

We report here that viable Rev3L−/− mouse embryonic fibroblasts (MEFs) could be generated by inactivating p53 function. A chromosome analysis and study of resistance to DNA damage show that pol ζ plays an important role in chromosome stability in mammalian cells.

Generation of MEFs and cell growth. The Rev3L null allele inactivates REV3L by replacing two exons encoding the conserved DNA polymerase family B motif V and part of motif I with a promoterless IRES cassette containing a fusion of the lacZ and neomycin resistance genes (28). The p53 null allele originates from the disruption generated in the Donehower laboratory (31). After timed matings between Rev3L+/−;p53+/− and Rev3L+/−;p53+/− or Rev3L+/−;p53−/− mice of B6;129 background, embryos were isolated on days E10.5-12.5. DNA from embryo yolk sacs and cultured cells was prepared using the Qiagen (Valencia, CA) DNeasy Tissue kit and genotyped by PCR (28). Individual embryos were dissociated and placed into culture as tissue explants in a humidied 5% CO2 incubator. Initial culture medium was a 1:1 mixture of DMEM with 4.5 g/L glucose and Ham's F12 (both from Life Technologies, Gaithersburg, MD), supplemented with 1% nonessential amino acids, 1% glutamine, and 10% fetal bovine serum (FBS; Myclone, Logan, UT).

MEFs from a single Rev3L−/−;p53−/− embryo began to proliferate following 3 months of occasional partial media changes and were then passaged without a notable growth defect. Rev3L+/+ and Rev3L+/− control MEF lines with p53−/− backgrounds were derived from the same litter of embryos as the Rev3L null line. Immortalized cell lines were established from spontaneously transformed clones following continuous passaging, and cell genotypes were confirmed by PCR (28, 32). Following immortalization, cells were cultured in DMEM (Mediatech, Herndon, VA; with 4.5 g/L glucose) with 10% FBS, 1% glutamine, 100 units/mL penicillin, and 100 μg/mL streptomycin in 75-cm2 flasks in 10% CO2.

The growth rate of immortalized Rev3L−/− embryonic fibroblasts and a Rev3L+/+ control was measured by plating 4 × 104 cells in six-well plates and counting cell numbers over a period of 5 days. The experiment was repeated thrice with cells from passages 12 to 17, and for each experiment, triplicate wells were washed, trypsinized, and counted individually using a hemacytometer. Cell viability was determined by trypan blue exclusion, and cell population doubling times were calculated using counts from days 2 and 4, during the period of exponential cell growth.

DNA damage sensitivity. Cytotoxic and cytostatic effects of DNA damage were assayed by measuring ATP levels. ATP is present in all metabolically active cells and decreases rapidly upon necrotic or apoptotic cell death. Cells were grown in phenol-free DMEM plus 10% FBS in a 10% CO2 atmosphere. Cells used for these experiments were from passages 13 to 23. The ATPlite1step (Perkin-Elmer, Boston, MA) luminescence system and a Perkin-Elmer TopCount luminometer were used to determine cell viability. To ascertain the suitability of this assay for DNA damage sensitivity studies using MEFs, a range of 125 to 50,000 cells per well were seeded in quadruplicate, and ATP levels were quantified 5 hours later. The relationship between cell number and luminescent signal was linear from 125 to 20,000 cells per well for both Rev3L+/+ and Rev3L−/− MEF cell lines, with low variance and a <1% backgound signal. This confirmed the precision of this method and guided the optimal cell seeding density and ATP quantification time to remain in linear range. The cell sensitivity assays were done as follows: 1,250 cells per well were seeded (100 μL) in quadruplicate in black 96-well plates (Perkin-Elmer) in DMEM + 10% FBS. Twenty-four hours later, cells were treated with the indicated concentrations of mitomycin C (MMC; Sigma, St. Louis, MO) or MMS (Sigma) for 18 hours. Cells were UV and γ-irradiated in complete medium at a dose rate of 0.7 J/m2/s from a UV-C source and at 19.66 Gy/h from a 137Cs source, respectively. Cells were washed with complete medium following treatment, fresh medium was added, and the cells were then grown for 2 days at which time untreated controls were 80% to 90% confluent. Results were expressed as the number of cells in treated wells relative to untreated control wells (% control). IC50 values were defined as the concentration or dose resulting in 50% growth inhibition.

Cytogenetics and micronucleus formation. Cells were treated with 0.5 μg/mL colcemid for 2 hours followed by harvesting and trypsin-Giemsa banding (33). Five individual metaphase cells from each coded cell line were captured digitally, and composite karyotypes were prepared using the CytoVision Ultra System (Applied Imaging, Santa Clara, CA). For spectral karyotyping analysis, new coded slides were prepared using cells from an earlier passage than for G-band analysis. Chromosomes were hybridized to 24-color, combinatorially labeled SkyPaint probes (Applied Spectral Imaging, Carlsbad, CA) as described (34). Chromosome and probe denaturation, hybridization, washing, and staining procedures were done as specified by the manufacturer's protocol. Fields of well-separated metaphase chromosomes were captured and analyzed with the SkyVision I system (Applied Spectral Imaging). The aberration analysis was scored before decoding the genotype and was done in two parts as described in Results.

To quantify micronuclei, 5 × 105 cells per MEF line (passages 13-18) were seeded onto 15-mm coverslips in six-well dishes. Cells were fixed in 4% paraformaldehyde (pH 7.0) and permeabilized with 1% Triton 100-X for 10 minutes at room temperature. Slides were washed with PBS and mounted using Vectashield with 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA) to stain the DNA. DNA from the indicated number of cells was analyzed using the ×100 objective of an Olympus Provis AX70 microscope with reflected light fluorescence. DAPI-stained micronuclei were scored only if completely separated from the nucleus, as distinct from nuclear blebs. Results are expressed as the frequency of micronuclei (%) relative to the total number of interphase nuclei scored.

Generation of proliferating MEFs lacking functional pol ζ. Disruption of the mouse Rev3L gene causes lethality during midgestation days E9.5-E12.5. Rev3L−/− embryos from this developmental stage were isolated and placed into culture to derive MEFs. In repeated attempts, we and others found that MEFs that are Rev3L−/− but have an otherwise normal genetic background, divide very poorly, if at all, in vitro (2628, 35). Altering embryo age, culture medium, and MEF culture method did not result in cells that could be passaged. In each case, Rev3L−/− fibroblasts entered a quiescent state with cell death occurring over weeks to months.

Developmental and cellular proliferation defects due to loss of DNA repair genes have been “rescued” to different extents by simultaneously removing p53 (3638). In contrast, we did not detect any effect of removal of p53 for development of Rev3L−/− embryos, a finding also reported by other groups (35, 39). Rev3L+/−;p53+/− mice were intercrossed and embryos from days E10.5-12.5 were examined. Of the Rev3L−/−;p53−/− embryos identified, four were resorbing (inviable) and four were viable. The latter embryos did not develop further than their Rev3L−/−;p53+/− or Rev3L−/−;p53+/+ littermates and were developmentally impaired relative to Rev3L+/+ and Rev3L+/− controls of any p53 genotype.

To determine if loss of p53 would assist the ability of Rev3L null cells to divide in vitro, 39 embryos from crosses between Rev3L+/−;p53+/− and Rev3L+/−;p53−/− mice were isolated and cultured. Rev3L+/+ and Rev3L+/− fibroblasts proliferated immediately and similarly. Embryonic fibroblasts from a single E10.5 Rev3L−/−;p53−/− mouse embryo placed in culture began to proliferate after 3 months. Cell division was without delay from the time of first passage, and the Rev3L null MEFs had a characteristic spindle-shaped, fibroblast-like morphology. The Rev3L null genotype was confirmed (Fig. 1A). Cells in one of four Rev3L−/−;p53+/− embryos placed into culture showed some mitotic activity during 8 months of partial medium changes, but this eventually ceased. Rev3L+/− and Rev3L+/+ MEFs that were p53−/− were also obtained from the same litter of embryos and these were passaged along with the Rev3L−/− cells until spontaneous immortalization occurred.

Figure 1.

Establishment and growth of a Rev3L−/− MEF cell line. A, PCR strategy used to genotype Rev3L+ and Rev3L alleles. Two essential DNA polymerase active site exons are removed in the Rev3L allele and replaced with a lacZ-neor (β-geo) cassette. Common primer F1 together with (+) allele R1 and (−) allele R2 primers generates 457-bp wild-type and 734-bp targeted mutant PCR products, respectively. For Rev3L−/− cells, PCR results of passages 2 and 7 are shown. Genotypes of Rev3L+/+ and Rev3L+/− cells from the same litter of embryos are flanking the Rev3L−/− cells. A single passage for each is shown. B, growth curves for mouse embryonic fibroblasts, either Rev3L+/+;p53−/− (○) or Rev3L−/−;p53−/− (•). Equal cell numbers (4 × 104 per well) were plated in six-well plates, and for each experiment, triplicate wells were counted over a period of 5 days. Points, mean of three independent experiments; bars, SE.

Figure 1.

Establishment and growth of a Rev3L−/− MEF cell line. A, PCR strategy used to genotype Rev3L+ and Rev3L alleles. Two essential DNA polymerase active site exons are removed in the Rev3L allele and replaced with a lacZ-neor (β-geo) cassette. Common primer F1 together with (+) allele R1 and (−) allele R2 primers generates 457-bp wild-type and 734-bp targeted mutant PCR products, respectively. For Rev3L−/− cells, PCR results of passages 2 and 7 are shown. Genotypes of Rev3L+/+ and Rev3L+/− cells from the same litter of embryos are flanking the Rev3L−/− cells. A single passage for each is shown. B, growth curves for mouse embryonic fibroblasts, either Rev3L+/+;p53−/− (○) or Rev3L−/−;p53−/− (•). Equal cell numbers (4 × 104 per well) were plated in six-well plates, and for each experiment, triplicate wells were counted over a period of 5 days. Points, mean of three independent experiments; bars, SE.

Close modal

We hypothesized that Rev3L−/− MEFs might have an increased cell doubling time due to impaired or delayed completion of DNA replication. Cell growth of Rev3L−/− and Rev3L+/+ embryonic fibroblasts was measured (Fig. 1B). During exponential growth, the population doubling time of Rev3L−/− embryonic fibroblasts was slightly increased relative to a Rev3L+/+ fibroblast line (19.1 versus 17.2 hours). The Rev3L null cells routinely had a higher plating efficiency and/or lower lag time before entering exponential growth. Likewise, the cell density at confluence of Rev3L−/−;p53−/− MEFs was always higher than control lines (Fig. 1B; data not shown). Viability of the cells during the growth and drug sensitivity experiments shown below ranged from 86% to 97% for the Rev3L null cells and 91% to 95% for the control cells.

Sensitivity of DNA pol ζ null cells to DNA-damaging agents. The Rev3L null MEF line was used to determine how a lack of DNA pol ζ affects growth and survival after treatment with DNA-damaging agents. Relative cell survival was determined by measuring total ATP levels, which is a direct measure of viable cell number. The assay has a wide dynamic range and low background and was validated for MEFs as described in Materials and Methods.

The Rev3L−/− MEFs were tested for their sensitivity to UV light, γ-irradiation, MMS, and MMC (Fig. 2). The mutagenicity and cytotoxicity of the former three DNA-damaging agents has been extensively studied with yeast rev3Δ strains (7). MMC was tested because it can introduce interstrand cross-links into DNA. rev3-mutant yeast cells are hypersensitive to photoactivated psoralen, which can produce such cross-links (40). The ratio of the IC50 values for these DNA-damaging agents provides a measure of the relative sensitivities of the Rev3L null cells compared with Rev3L wild-type cells as follows: MMC, 5.0-fold; UV, 3.7-fold; γ irradiation, 2.3-fold; MMS, 1.8-fold. Similar absolute and relative sensitivity of the Rev3L null MEFs and a second wild-type control line to MMC, MMS, and γ irradiation was observed when experiments were done in six-well plates, and cell numbers were determined using a hemacytometer. Microscopic examination revealed that both Rev3L null and control cells displayed a classic increased size response following γ-irradiation. These results indicate that mammalian pol ζ functions in the tolerance of several diverse types of DNA damage.

Figure 2.

Sensitivity of DNA pol ζ null cells to DNA-damaging agents. Equal numbers of Rev3L−/− (•) and Rev3L+/+ (○) were plated in quadruplicate in 96-well plates. Cells were exposed to (A) MMC, (B) MMS, (C) UV-C, and (D) γ-irradiation as described in Materials and Methods. After 2 days, cytotoxicity was determined by quantification of ATP levels in treated and untreated cells. Points, mean of three independent experiments (% untreated control); bars, SE.

Figure 2.

Sensitivity of DNA pol ζ null cells to DNA-damaging agents. Equal numbers of Rev3L−/− (•) and Rev3L+/+ (○) were plated in quadruplicate in 96-well plates. Cells were exposed to (A) MMC, (B) MMS, (C) UV-C, and (D) γ-irradiation as described in Materials and Methods. After 2 days, cytotoxicity was determined by quantification of ATP levels in treated and untreated cells. Points, mean of three independent experiments (% untreated control); bars, SE.

Close modal

Increased chromosome number, fusion, and fragmentation in Rev3L−/− MEFs. We next looked for evidence of chromosomal instability associated with Rev3L deletion. Chromosome spreads were stained with Giemsa, and individual metaphases were counted for total chromosome numbers (Table 1). All cell lines had undergone polyploidization, an event commonly observed during establishment of MEF cell lines (41). The numbers of chromosomes were increased in the Rev3L−/−;p53−/− cells (range, 78-189) relative to the Rev3L+/−;p53−/− and Rev3L+/−;p53+/+ control lines (ranges, 56-129 and 61-121). The majority of Rev3L null cells were near pentaploid (90-107 chromosomes), whereas the majority of control MEFs were hypotetraploid (69-76 chromosomes). Hypotetraploidy is the usual state of spontaneously immortalized MEFs.

Table 1.

Chromosome numbers in Rev3L mutant MEFs

MEF cell linePassageRangeMeanMedianModal karyotypes
Rev3L+/−;p53+/+ 14 56-129 77 73 73 
Rev3L+/−;p53−/− 12 61-121 75 72 69, 74 
Rev3L−/−;p53−/− 12 78-189 110 101 91, 98, 103, 115 
MEF cell linePassageRangeMeanMedianModal karyotypes
Rev3L+/−;p53+/+ 14 56-129 77 73 73 
Rev3L+/−;p53−/− 12 61-121 75 72 69, 74 
Rev3L−/−;p53−/− 12 78-189 110 101 91, 98, 103, 115 

NOTE: Chromosome spreads were stained with Giemsa, and total chromosome numbers were counted for 25 individual metaphases of each cell line. Statistical averages are shown. Based on the distribution of ploidies for all three cell lines, it is likely that the two to three metaphases at the high end of the range for each line resulted from failed cytokinesis of cells with chromosome numbers at the low end of the range.

Karyotyping revealed significantly increased spontaneous chromosomal instability in Rev3L−/−;p53−/− cells (Table 2; Fig. 3A). There were increased numbers of fused/translocated chromosomes, chromosomes with terminal deletions, and rearranged “marker” chromosomes not identifiable by G-banding. For all three categories, the increase was 4- to 5-fold over control cell lines (Table 2). The additional chromosomes found in the Rev3L null line were divided approximately equally between whole, intact chromosomes and those that were rearranged or had terminal deletions (9.2 and 11.2 average events per metaphase, respectively).

Table 2.

Summary of G-band karyotype analysis

MEF cell lineMean chromosome no., karyotyped cellsLossGainMarker chromosomesTerminal deletionsAdditions/translocations
Rev3L+/−;P53+/+ 77 7.4 4.2 0.2 0.2 
Rev3L+/−;p53−/− 75 9.2 1.8 1.8 0.6 
Rev3L−/−;p53−/− 92 8.6 9.2 7.8 2.4 
MEF cell lineMean chromosome no., karyotyped cellsLossGainMarker chromosomesTerminal deletionsAdditions/translocations
Rev3L+/−;P53+/+ 77 7.4 4.2 0.2 0.2 
Rev3L+/−;p53−/− 75 9.2 1.8 1.8 0.6 
Rev3L−/−;p53−/− 92 8.6 9.2 7.8 2.4 

NOTE: Five G-banded metaphases were completely karyotyped. All results are expressed as the mean number of each category per metaphase. The numbers of intact chromosomes for each autosome and sex chromosome were counted for each karyotype. The “loss” and “gain” columns refer to average net variations in whole chromosome numbers from a 4N chromosome number. Marker chromosomes are defined as chromosome rearrangements not identifiable by G-banding. Both of the control lines lost more chromosomes than they gained. This, together with the low total numbers of rearranged chromosomes and those with deletions (last three columns), accounts for their net hypotetraploid mean numbers [80 − loss + (gain + markers + deletions + fusions)]. The Rev3L−/−;p53−/− cells gained slightly more intact chromosomes than they lost. This, and the greater number of chromosomes with rearrangements and deletions, comprises the near-pentaploid mean number for the karyotyped Rev3L−/−;p53−/− cells.

Figure 3.

Chromosomal instability in Rev3L−/− MEFs. A, chromosomal instability in a representative G-banded karyotype of a Rev3L−/−;p53−/− metaphase. A wide range of whole chromosome gains and losses are seen (e.g., chromosomes 9 and 12). Two translocations involving chromosomes 7 and 17 (open arrows). A chromosome 19 carrying a deletion (closed arrow). Both large and small supernumerary marker chromosomes, as well as double-minute chromosomes (dmin), are placed between mouse chromosome 19 and the sex chromosomes. One marker contains a chromatid break (chtb). The increase in both structurally normal and abnormal chromosomes leads to a total chromosome number of 95 in this metaphase. The Rev3L−/−;p53−/− cell line is derived from a male embryo. B, micronucleus formation in Rev3L+/+;p53−/− and Rev3L−/−;p53−/− cells. The frequency of spontaneous micronucleus formation in Rev3L+/+ and Rev3L−/− interphase nuclei was determined by DAPI staining. Sample number was >1,400 cells for each line.

Figure 3.

Chromosomal instability in Rev3L−/− MEFs. A, chromosomal instability in a representative G-banded karyotype of a Rev3L−/−;p53−/− metaphase. A wide range of whole chromosome gains and losses are seen (e.g., chromosomes 9 and 12). Two translocations involving chromosomes 7 and 17 (open arrows). A chromosome 19 carrying a deletion (closed arrow). Both large and small supernumerary marker chromosomes, as well as double-minute chromosomes (dmin), are placed between mouse chromosome 19 and the sex chromosomes. One marker contains a chromatid break (chtb). The increase in both structurally normal and abnormal chromosomes leads to a total chromosome number of 95 in this metaphase. The Rev3L−/−;p53−/− cell line is derived from a male embryo. B, micronucleus formation in Rev3L+/+;p53−/− and Rev3L−/−;p53−/− cells. The frequency of spontaneous micronucleus formation in Rev3L+/+ and Rev3L−/− interphase nuclei was determined by DAPI staining. Sample number was >1,400 cells for each line.

Close modal

Marker chromosomes were loosely divided into those larger than the smallest mouse chromosome (chromosome 19) and those smaller than chromosome 19. The small marker chromosomes are similar to human small supernumerary marker chromosomes, which usually originate from the five human acrocentric chromosomes (42). One third of the total marker chromosomes for both Rev3L−/−;p53−/− (13 total) and Rev3L+/−;p53−/− (3 total) cells appeared in this category and were mostly heterochromatic (darkly staining bands). These may arise following the creation of dicentrics between telocentric chromosomes, which subsequently break within the q arm close to one centromere. This confers stability to both the large chromosome and the small marker. Ten of the 25 Rev3L−/−;p53−/− cells had very small chromosome fragments with the characteristic dumbbell or bilobed shape of double minutes (Fig. 3A). These were not included in the chromosome count and were not seen for either of the control MEFs. Ring or multiradial chromosomes were not observed in the Rev3L null cells.

The chromosome instability of Rev3L−/−;p53−/− cells was also examined by a noncytogenetic method. Micronuclei can form from DNA fragments that are not incorporated into daughter nuclei because they lack a functional centromere or by fragmentation of anaphase bridges (43). The Rev3L−/−;p53−/− cells had a >2-fold increase in the number of micronuclei relative to control Rev3L+/+;p53−/− cells (Fig. 3B). Compared with Rev3L+/+ cells, more Rev3L−/− cells had multiple micronuclei, and both nuclei and micronuclei from the Rev3L−/−;p53−/− MEFs showed greater size heterogeneity. Futher, three anaphase bridges were observed in the Rev3L null cells, whereas none were seen in the control.

Increased translocation events in the absence of DNA pol ζ. Spectral karyotyping was used to characterize and quantify the rearrangements occurring in earlier passage cells in more detail. Spectral karyotyping enables unequivocal definition of rearrangements on a chromosome-by-chromosome basis, and our analysis concentrated on the detection and quantification of aberrations rather than cellular karyotyping. Robertsonian-like translocations (or centric fusions), isochromosomes (see Fig. 4 legend for definition), and mouse long arm translocations were found in both Rev3L−/− and control Rev3L+/− MEFs (Fig. 4A and B). Three different classes of aberration were found only in the Rev3L null cells. These were dicentrics, insertions, and compound isochromosome/translocation events (Fig. 4B).

Figure 4.

Spectral karyotyping analysis of DNA pol ζ null cells reveals details of genome instability. A, types of structural chromosome aberration found in control Rev3L+/− MEFs: isochromosome (i), Robertsonian-like translocations (rob), and translocations (der). Cells analyzed were passages 8 and 9. B, these three types of events were also found in Rev3L−/− MEFs (top). Examples of aberrations found only in Rev3L−/− MEFs (bottom): dicentrics (dic), insertions (ins), and combined isochromosome/translocation events (i and t). The der(19)ins(3;19) and i(3)+t(3;19) abnormalities are possible products of a reciprocal translocation event. Rev3L−/− cells analyzed were passages 7 and 9. Mouse chromosomes are telocentric containing essentially only a long arm, with a short arm composed solely of centromeric and telomeric repeat sequences. Two types of defect can occur to generate metacentric (biarmed) chromosomes. Isochromosomes may result from DNA breakage within the short arm or centromere followed by post-replication reunion of the sister chromatids to generate a chromosome with two copies of the long arm flanking the now metacentric centromere (72). Interchromosome fusions involving the mouse short arm (from the telomere to the centromere) produce metacentric chromosomes (Robertsonian-like translocations). The exchange usually involves the short arm of both participants, but a Robertsonian-like configuration may also be created between one short and one long arm.

Figure 4.

Spectral karyotyping analysis of DNA pol ζ null cells reveals details of genome instability. A, types of structural chromosome aberration found in control Rev3L+/− MEFs: isochromosome (i), Robertsonian-like translocations (rob), and translocations (der). Cells analyzed were passages 8 and 9. B, these three types of events were also found in Rev3L−/− MEFs (top). Examples of aberrations found only in Rev3L−/− MEFs (bottom): dicentrics (dic), insertions (ins), and combined isochromosome/translocation events (i and t). The der(19)ins(3;19) and i(3)+t(3;19) abnormalities are possible products of a reciprocal translocation event. Rev3L−/− cells analyzed were passages 7 and 9. Mouse chromosomes are telocentric containing essentially only a long arm, with a short arm composed solely of centromeric and telomeric repeat sequences. Two types of defect can occur to generate metacentric (biarmed) chromosomes. Isochromosomes may result from DNA breakage within the short arm or centromere followed by post-replication reunion of the sister chromatids to generate a chromosome with two copies of the long arm flanking the now metacentric centromere (72). Interchromosome fusions involving the mouse short arm (from the telomere to the centromere) produce metacentric chromosomes (Robertsonian-like translocations). The exchange usually involves the short arm of both participants, but a Robertsonian-like configuration may also be created between one short and one long arm.

Close modal

Both the total number of all aberrations and the total number of different (nonclonal) aberrations were increased 4- to 5-fold in Rev3L−/− MEFs (Table 3). This is similar to the increase in numbers of aberrant events as determined by G-banding. Certain types of aberrations were specifically enhanced in Rev3L null cells. Aberrations occurring along the chromosome long arm were increased 15-fold in Rev3L null cells (the sum of translocations, isochromosomes plus translocations, insertions, and dicentrics). The largest element of this increase is the number of translocation events. Numbers of events involving the short arm and centromere (isochromosomes and Robertsonian-like translocations) were equivalent in Rev3L−/− and Rev3L+/− MEFs.

Table 3.

Loss of DNA pol ζ causes elevated translocation frequency

Rev3L+/−;p53−/−Rev3L−/−;p53−/−
Total no. chromosomes analyzed 1,081 1,110 
Total no. aberrations 10 56 
Total no. different (nonclonal) aberrations* 9 (0.8%) 37 (3.3%) 
Robertsonian-like translocations 
Isochromosomes (iso) 
iso + der (complex translocations) 
der (translocations) 25 
Insertions 
Dicentrics 
Rev3L+/−;p53−/−Rev3L−/−;p53−/−
Total no. chromosomes analyzed 1,081 1,110 
Total no. aberrations 10 56 
Total no. different (nonclonal) aberrations* 9 (0.8%) 37 (3.3%) 
Robertsonian-like translocations 
Isochromosomes (iso) 
iso + der (complex translocations) 
der (translocations) 25 
Insertions 
Dicentrics 
*

The numbers of nonclonal (unique) aberrations for each line is the sum of the aberration types shown below this total. The analysis was done in two series using passages 8 and 9 for Rev3L+/−;p53−/− cells and passages 7 and 9 for Rev3L−/−;p53−/− cells. For the Rev3L null cells, the majority of aberrations in both cases were translocations (70% and 85%).

All chromosomes except for 6, 11, and Y were observed in rearrangements; 10 different chromosomes were involved in three to seven events each, whereas six chromosomes were involved in one to two events (Table 4). The spectral karyotyping analyses were completed in two equally sized series with cells from different passages. For the Rev3L null cells, only one aberration identified in the first series [der(2)t(1;2)] was found in the second group, indicating that the chromosomal instability is a continuous process occurring cell by cell and dispersed throughout the mouse genome.

Table 4.

Chromosome aberrations found in Rev3L(+/−) and Rev3L(−/−) MEFs

irobtdicinsi and t
Rev3L+/− MEF aberrations      
    i(1) rob(6;15) der(2)t(2;7)    
    i(7) rob(6;7) der(5)t(5;16)    
    i(18) rob(7;15), rob(12;19)     
Rev3L−/− MEF aberrations      
    i(1) rob(2;5) der(1)t(1;2)9) dic(2;5) der(10)ins(16;10) i(3)+t(3;19) 
    i(13) rob(4;14) der(1)t(1;5) dic(17) der(19)ins(3;19) i(13)t(13;14) 
 rob(4;5) der(1)t(1;5;16)    
 rob(11;15) der(2)t(1;2)    
  der(2)t(2;10)    
  der(2)t(2;15)    
  der(2)t(2;17)    
  der(3)t(2;3)    
  der(3)t(3;10)    
  der(4)t(X;4)    
  der(5)t(2;5)    
  der(8)t(8;10)    
  der(9)t(9;18)    
  der(10)t(2;10)    
  der(10)t(10;17)    
  der(12)t(12;14)    
  der(14)t(7;14)    
  der(15)t(2;15)    
  der(15)t(15;19)    
  der(16)t(10;16)    
  der(16)t(14;16)    
  der(17)t(2;17)    
  der(19)t(8;19)    
  der(X)t(X;15)    
  der(X)t(X;1    
irobtdicinsi and t
Rev3L+/− MEF aberrations      
    i(1) rob(6;15) der(2)t(2;7)    
    i(7) rob(6;7) der(5)t(5;16)    
    i(18) rob(7;15), rob(12;19)     
Rev3L−/− MEF aberrations      
    i(1) rob(2;5) der(1)t(1;2)9) dic(2;5) der(10)ins(16;10) i(3)+t(3;19) 
    i(13) rob(4;14) der(1)t(1;5) dic(17) der(19)ins(3;19) i(13)t(13;14) 
 rob(4;5) der(1)t(1;5;16)    
 rob(11;15) der(2)t(1;2)    
  der(2)t(2;10)    
  der(2)t(2;15)    
  der(2)t(2;17)    
  der(3)t(2;3)    
  der(3)t(3;10)    
  der(4)t(X;4)    
  der(5)t(2;5)    
  der(8)t(8;10)    
  der(9)t(9;18)    
  der(10)t(2;10)    
  der(10)t(10;17)    
  der(12)t(12;14)    
  der(14)t(7;14)    
  der(15)t(2;15)    
  der(15)t(15;19)    
  der(16)t(10;16)    
  der(16)t(14;16)    
  der(17)t(2;17)    
  der(19)t(8;19)    
  der(X)t(X;15)    
  der(X)t(X;1    

NOTE: The events summarized in Table 3 are listed. Telomere sequence may or may not be present at fusion point. Breakpoints are unknown. Due to the nature of this analysis, reciprocal events may only be inferred.

Abbreviations: i, isochromosome; rob, Robertsonian-like translocation. t, translocation; der, derivative events; dic, dicentric; ins, insertion; i and t, combined isochromosome and translocation events.

The distribution of chromosome aberrations is not entirely random. Chromosome 2 was involved in 11 long arm aberrations (Table 4), a number larger than expected even taking chromosome size into account. This may indicate the presence of one or more breakage or recombination hotspots on mouse chromosome 2. For each of the four der(2) translocations (involving chromosomes 1, 10, 15, and 17), a derivative translocation involving each of the partners and chromosome 2 was detected, suggesting that reciprocal exchanges occur. A possible reciprocal event involving chromosomes 3 and 19 was more complex (Fig. 4B). Based on chromosome length, it was unambigous that seven of the translocations (involving six different chromosomes) in Rev3L null cells involved only a partial chromosome as one of the exchange partners. This shows that rearrangements occur within the chromosome long arm in Rev3L null cells and correlate with the terminal chromosome deletions found by G-banding.

Mammalian cells can proliferate rapidly in the absence of Rev3L. The absence of Rev3L in the mouse causes impaired embryonic and extraembryonic tissue development, leading to lethality during midgestation. This suggests that pol ζ may be required to help cells tolerate endogenous DNA damage during mammalian development. Similarly, cells from Rev3L null embryos do not proliferate in vitro but remain quiescent with cell death occurring over time (2628, 35). The null Rev3L cells from midgestation embryos may have accumulated levels of unrepaired endogenous DNA damage that activate DNA damage checkpoints and preclude cell division in vitro.

As reported here, cells from a Rev3L−/−;p53−/− embryo began dividing rapidly after 3 months in culture. Dividing cells from one of four Rev3L−/−;p53+/− embryos were observed after an even longer period (6-8 months) but ceased division before being passaged. Mitotic activity of Rev3L−/− cells with a p53 wild-type background was never observed, even after extended culture periods. Most likely, a p53-dependent checkpoint is only partly responsible for the arrested division observed in cultured Rev3L−/− cells. A similar conclusion was reached by Zander and Bemark, who also generated proliferating Rev3L null MEFs in a p53 mutant background (44).

The population doubling time of our Rev3L−/− embryonic fibroblasts was ∼11% longer than a Rev3L+/+ control, whereas the plating efficiency and saturation density were higher for the Rev3L null cells. Vertebrate Rev3-null cells capable of cell division were first reported for the gene-targeted chicken lymphocyte DT40 cell line, and a comparable ∼12% increase in cell cycle time was found (45). p53 expression is not detectable in DT40 cells (46), emphasizing that checkpoint and apoptosis functions must be compromised in vertebrate cells for in vitro cell division to occur in the absence of pol ζ. The isolation of mitotically active Rev3L null MEFs confirms that pol ζ is not essential for chromosome replication in vertebrates. As discussed below, it is however, required for genome stability in mammalian cells.

Generalized sensitivity of pol ζ–disrupted cells to DNA-damaging agents. The modest sensitivity of Rev3L−/− MEFs to several different DNA-damaging agents indicates that pol ζ function increases tolerance to a broad range of DNA adducts. Rev3L null MEFs are relatively more sensitive to mitomycin C and UV-C irradiation than γ-irradiation and MMS. This suggests that in MEFs, there are more redundant mechanisms of tolerance to ubiquitous endogenous types of DNA damage (oxidative base lesions and base methylations) than to less frequent DNA lesions originating from environmental sources. Other specialized DNA polymerases are capable of synthesis at sites of simple base damage or AP sites (47). DNA pol ζ may be more uniquely able to insert and/or extend bases opposite the larger, more helix-distorting lesions introduced by mitomycin C and UV-C radiation.

These results are generally consistent with data on sensitivity to DNA damage damaging agents in other mutant organisms and cells lacking pol ζ. Yeast rev3 mutant cells are moderately sensitive to cross-linking agents and UV and mildly sensitive to γ-irradiation and MMS (12, 4850). Yeast REV3 null cells are not hypersensitive to the oxidizing agents hydrogen peroxide and menadione (51). Neurospora crassa and Arabidopsis thaliana cells disrupted for Rev3 also follow this sensitivity pattern to these same agents (52, 53). DT40 cells deleted for Rev3 are sensitive to these four types of DNA damage, with a notable difference being the pronounced sensitivity of these cells to cisplatin (45). Independently isolated Rev3L−/−;p53−/− MEFs are modestly sensitive to both cisplatin and UV radiation (44). A REV3L antisense knockdown cell line, which retains a low level of REV3L expression, is mildly sensitive to cisplatin and has little or no UV sensitivity (21, 54).

Chromosomes in Rev3L null embryonic fibroblasts are structurally unstable. Tetraploidy arises when aberrant cytokinesis doubles chromosome content. This lessens restrictions on the types of chromosome abnormalities that can accumulate in cells. The Rev3L null cells were near pentaploid due to gain of whole chromosomes and aberrant chromosomes. The gain of complete chromosomes may be caused by mis-segregation of chromatids that are not fully replicated and separated. The incidence of chromosomes with rearrangements or terminal deletions was 4- to 5-fold greater in Rev3L null cells than in the control lines. Rev3−/− DT40 cells had a 3-fold increase in the frequency of spontaneous aberrations was found (45). These were breaks and gaps rather than exchanges. MEFs may be relatively more proficient at DSB repair by nonhomologous end-joining (NHEJ) than are DT40 cells. Small chromosome fragments resembling double minutes were found in ∼40% of the Rev3L null cells and not in control cells. Different models for the origin of these extrachromosomal acentric species all postulate DNA breaks as an essential step (55).

A form of multicolor fluorescence in situ hybridization known as COBRA was used to examine metaphases from Rev3L−/− embryos, and the incidence of translocations seemed to be increased several-fold (35). The total numbers of rearrangements found were low, and different categories of translocations were not reported. We did spectral karyotyping analysis to identify and quantify the different types of chromosome rearrangements occuring in Rev3L null MEFs. The most remarkable finding is the order of magnitude increase in the number of translocations in Rev3L null cells relative to control Rev3L+/− cells. The well-dispersed distribution of translocations among the mouse autosomes indicates that the increase in translocation frequency is not due to recurrent breakage fusion bridge cycles but is instead due to random genomic instability. Other indications for increased chromosomal breakage in the null cells are the appearance of insertions, which may involve multiple DNA breaks for their formation, and compound translocation/isochromosome events. These compound events most likely arise from sequential rearrangements.

Chromosome breakage in Rev3L null cells. The high frequency of DNA rearrangements in Rev3L null cells suggests that pol ζ is required to prevent multiple double-strand breaks from occurring in the mammalian genome. In the absence of this enzyme, replication forks stalled at sites of DNA damage may actively or passively collapse, leading to DNA double-strand breaks. Homologous recombination and NHEJ repair pathways are functional in these cells and will rapidly repair DNA breaks, both correctly and incorrectly (56). Frequent reciprocal translocations occur in mouse embryonic stem cells when there are two double-strand breaks on different chromosomes (57).

Although translocations were significantly elevated, chromosome exchanges involving telomeric and centromeric repeats in Rev3L−/− and Rev3L+/− MEFs were equivalent. In contrast, chromosomes from telomerase null cells with shortened telomeres form increased numbers of p-p arm fusions (58). MEFs from NHEJ null mutants (Ku70, Ku80, DNA-PK, and Lig4) have a high incidence of chromosome and chromatid fragmention, and the increased numbers of Robertsonian-like translocations in all these mutants indicate that NHEJ functions to repair breaks occurring within the short arm of mouse chromosomes (5961). Fusions involving short arms comprise about half of the total translocations found in these NHEJ null cells, which is clearly different than the pattern in pol ζ null cells.

Cytogenetic studies of fibroblasts from homologous recombination-deficient mice show more similarity to the results presented here for Rev3L null MEFs. Chromosome rearrangements involving the chromosome long arm are the most common aberration in Rad51d- and Xrcc2-deficient cells. Loss of RAD51D protein results in a 24-fold increase of chromosome exchanges with a 6-fold increase in end-to-end fusions (62). Immortalized Xrcc2−/− MEFs have a 100-fold increase in chromosomal rearrangements compared with Xrcc2+/+ MEFs but no difference in the number of end-to-end fusions (63). A possible explanation for the relative difference is that RAD51D associates with telomeres, whereas Xrcc2 does not (62, 64). The similar distribution of aberrations in immortalized Rad51d, Xrcc2, and Rev3L null MEFs may reflect the fact that both homologous recombination and translesion synthesis are pathways that help cells tolerate DNA damage encountered by replication forks. In the absence of either pathway, broken DNA ends may form with increased frequency and be joined to nonhomologous chromosomes.

The equal frequency of Robertsonian-like translocations and isochromosomes in Rev3L−/− and Rev3L+/− cells indicates that loss of pol ζ does not significantly affect exchanges at telomeres and centromeres. Either DNA pol ζ does not function during replication of telomeric and centromeric repeat sequences, or other DNA translesion bypass polymerases and DNA repair pathways are able to perform compensatory functions when pol ζ is absent. Any such backup mechanisms are not sufficient to prevent chromosomal instability, however, when DNA replication stalls at other genome sequences.

Implications for tumorigenesis. Human REV3L maps to chromosome 6q21. This region of 6q has long been postulated to contain multiple tumor suppressor genes. Chromosome deletions of 6q21 have most often been reported for hematopoietic neoplasms, including acute lymphoblastic leukemia, T-cell lymphoma, and gastric high-grade large B-cell lymphoma (6567). Some lymphomas derive from aberrant somatic hypermutation. As pol ζ may normally function in this process, and there is an increased incidence of translocations in the absence of Rev3L, it is possible that loss or dysregulation of this polymerase may sometimes be a formative event for such cancers. Some cases of chronic lymphocytic leukemia, non-Hodgkin's lymphoma, and diffuse large B cell lymphoma may originate from hypermutating B cells, and all three have been found to have frequent deletions of 6q21 (6870). Furthermore, the REV3L gene is entirely within the FRA6F fragile site, and the 3′ region of human REV3L overlaps one of two breakage hotspots (71). Deletion breakpoints within this 1.2-Mbp fragile site occur in many blood and solid tissue cancers.

Compromising the function of pol ζ could have significant consequences for carcinogenesis. Human chromosome instability syndromes and the transgenic mice generated to model them are often prone to tumorigenesis. As chromosome translocations can be associated with oncogene activation, their increased occurrence in Rev3L−/− MEFs implies that Rev3L null cells could have increased oncogenic potential. Loss of Rev3L expression, particularly after loss of p53 function, might occur during neoplastic progression and contribute to the genome instability inherent in many tumors.

Note: The cytogenetic studies were carried out in the University of Pittsburgh Cancer Institute Cytogenetics Facility.

Grant support: NIH grants CA098675 (R.D. Wood) and P30 CA47904 (R.B. Herberman).

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.

We thank Christine Saunders and the Cancer Research UK Cell Production team for expert assistance with embryo culture, members of our laboratory and Drs. Laura Niedernhofer, Robert Sobol, Karen Vasquez, and Christopher Bakkenist for discussion.

1
Hubscher U, Maga G, Spadari S. Eukaryotic DNA polymerases.
Annu Rev Biochem
2002
;
71
:
133
–63.
2
Goodman MF. Error-prone repair DNA polymerases in prokaryotes and eukaryotes.
Annu Rev Biochem
2002
;
71
:
17
–50.
3
Prakash S, Johnson RE, Prakash L. Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function.
Annu Rev Biochem
2005
;
74
:
317
–53.
4
Stelter P, Ulrich HD. Control of spontaneous and damage-induced mutagenesis by SUMO and ubiquitin conjugation.
Nature
2003
;
425
:
188
–91.
5
Kannouche PL, Wing J, Lehmann AR. Interaction of human DNA polymerase eta with monoubiquitinated PCNA: a possible mechanism for the polymerase switch in response to DNA damage.
Mol Cell
2004
;
14
:
491
–500.
6
Nelson J, Lawrence C, Hinkle D. Thymine-thymine dimer bypass by yeast DNA polymerase ζ.
Science
1996
;
272
:
1646
–9.
7
Lawrence CW. Cellular roles of DNA polymerase zeta and Rev1 protein.
DNA Repair (Amst)
2002
;
1
:
425
–35.
8
Johnson RE, Washington MT, Haracska L, Prakash S, Prakash L. Eukaryotic polymerases iota and zeta act sequentially to bypass DNA lesions.
Nature
2000
;
406
:
1015
–9.
9
Haracska L, Unk I, Johnson RE, et al. Roles of yeast DNA polymerases δ and ζ and of Rev1 in the bypass of abasic sites.
Genes Dev
2001
;
15
:
945
–54.
10
Nelson JR, Lawrence CW, Hinkle DC. Deoxycytidyl transferase activity of yeast REV1 protein.
Nature
1996
;
382
:
729
–31.
11
Nelson JR, Gibbs PE, Nowicka AM, Hinkle DC, Lawrence CW. Evidence for a second function for Saccharomyces cerevisiae Rev1p.
Mol Microbiol
2000
;
37
:
549
–54.
12
Zhao B, Xie Z, Shen H, Wang Z. Role of DNA polymerase eta in the bypass of abasic sites in yeast cells.
Nucleic Acids Res
2004
;
32
:
3984
–94.
13
Johnson RE, Yu SL, Prakash S, Prakash L. Yeast DNA polymerase ζ (zeta) is essential for error-free replication past thymine glycol.
Genes Dev
2003
;
17
:
77
–87.
14
Gibbs PE, McDonald J, Woodgate R, Lawrence CW. The relative roles in vivo of Saccharomyces cerevisiae Pol eta, Pol zeta, Rev1 protein and Pol32 in the bypass and mutation induction of an abasic site, T-T (6–4) photoadduct and T-T cis-syn cyclobutane dimer.
Genetics
2005
;
169
:
575
–82.
15
Quah S-K, von Borstel RC, Hastings PJ. The origin of spontaneous mutation in Saccharomyces cerevisiae.
Genetics
1980
;
96
:
819
–39.
16
Harfe BD, Jinks-Robertson S. DNA polymerase zeta introduces multiple mutations when bypassing spontaneous DNA damage in Saccharomyces cerevisiae.
Mol Cell
2000
;
6
:
1491
–9.
17
Holbeck SL, Strathern JN. A role for Rev3 in mutagenesis during double-strand break repair in Saccharomyces cerevisiae.
Genetics
1997
;
147
:
1017
–24.
18
Gibbs PE, McGregor WG, Maher VM, Nisson P, Lawrence CW. A human homolog of the Saccharomyces cerevisiae REV3 gene, which encodes the catalytic subunit of DNA polymerase zeta.
Proc Natl Acad Sci U S A
1998
;
95
:
6876
–80.
19
Lin W, Wu X, Wang Z. A full-length cDNA of hREV3 is predicted to encode DNA polymerase zeta for damage-induced mutagenesis in humans.
Mutat Res
1999
;
433
:
89
–98.
20
van Sloun PP, Romeijn RJ, Eeken JC. Molecular cloning, expression and chromosomal localisation of the mouse Rev3l gene, encoding the catalytic subunit of polymerase z.
Mutat Res
1999
;
433
:
109
–16.
21
Li Z, Zhang H, McManus TP, McCormick JJ, Lawrence CW, Maher VM. hREV3 is essential for error-prone translesion synthesis past UV or benzo[a]pyrene diol epoxide-induced DNA lesions in human fibroblasts.
Mutat Res
2002
;
510
:
71
–80.
22
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 zeta.
J Immunol
2001
;
167
:
327
–35.
23
Diaz M, Watson NB, Turkington G, Verkoczy LK, Klinman NR, McGregor WG. Decreased frequency and highly aberrant spectrum of ultraviolet-induced mutations in the Hprt gene of mouse fibroblasts expressing antisense RNA to DNA polymerase zeta.
Mol Cancer Res
2003
;
1
:
836
–47.
24
Zan H, Komori A, Li Z, et al. The translesion DNA polymerase zeta plays a major role in Ig and bcl-6 somatic hypermutation.
Immunity
2001
;
14
:
643
–53.
25
Machida K, Cheng KT, Sung VM, et al. Hepatitis C virus induces a mutator phenotype: enhanced mutations of immunoglobulin and protooncogenes.
Proc Natl Acad Sci U S A
2004
;
101
:
4262
–7.
26
Bemark M, Khamlichi AA, Davies SL, Neuberger MS. Disruption of mouse polymerase zeta (Rev3) leads to embryonic lethality and impairs blastocyst development in vitro.
Curr Biol
2000
;
10
:
1213
–6.
27
Esposito G, Godindagger I, Klein U, Yaspo ML, Cumano A, Rajewsky K. Disruption of the Rev3l-encoded catalytic subunit of polymerase zeta in mice results in early embryonic lethality.
Curr Biol
2000
;
10
:
1221
–4.
28
Wittschieben J, Shivji MKK, Lalani E-N, et al. Disruption of the developmentally regulated Rev3l gene causes embryonic lethality.
Curr Biol
2000
;
10
:
1217
–20.
29
Shepard TH, Muffley LA, Smith LT. Mitochondrial ultrastructure in embryos after implantation.
Hum Reprod
2000
;
15
Suppl 2:
218
–28.
30
Larsson NG, Wang J, Wilhelmsson H, et al. Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice.
Nat Genet
1998
;
18
:
231
–6.
31
Harvey M, McArthur MJ, Montgomery CA, Jr., Butel JS, Bradley A, Donehower LA. Spontaneous and carcinogen-induced tumorigenesis in p53-deficient mice.
Nat Genet
1993
;
5
:
225
–9.
32
Costello PS, Cleverley SC, Galandrini R, Henning SW, Cantrell DA. The GTPase rho controls a p53-dependent survival checkpoint during thymopoiesis.
J Exp Med
2000
;
192
:
77
–85.
33
Heo DS, Snyderman C, Gollin SM, et al. Biology, cytogenetics, and sensitivity to immunological effector cells of new head and neck squamous cell carcinoma lines.
Cancer Res
1989
;
49
:
5167
–75.
34
Reshmi SC, Saunders WS, Kudla DM, Ragin CR, Gollin SM. Chromosomal instability and marker chromosome evolution in oral squamous cell carcinoma.
Genes Chromosomes Cancer
2004
;
41
:
38
–46.
35
Van Sloun PP, Varlet I, Sonneveld E, et al. Involvement of mouse Rev3 in tolerance of endogenous and exogenous DNA damage.
Mol Cell Biol
2002
;
22
:
2159
–69.
36
Ludwig T, Chapman DL, Papaioannou VE, Efstratiadis A. Targeted mutations of breast-cancer susceptibility gene homologs in mice - lethal phenotypes of Brca1, Brca2, Brca1/Brca2, Brca1/p53, and Brca2/p53 nullizygous embryos.
Genes Dev
1997
;
11
:
1226
–41.
37
Patel KJ, Yu VPCC, Lee HS, et al. Involvement of Brca2 in DNA-repair.
Mol Cell
1998
;
1
:
347
–57.
38
Lim DS, Hasty P. A mutation in mouse rad51 results in an early embryonic lethal that is suppressed by a mutation in p53.
Mol Cell Biol
1996
;
16
:
7133
–43.
39
O-Wang J, Kajiwara K, Kawamura K, et al. An essential role for REV3 in mammalian cell survival: absence of REV3 induces p53-independent embryonic death.
Biochem Biophys Res Commun
2002
;
293
:
1132
–7.
40
Cassier-Chauvat C, Moustacchi E. Allelism between pso1–1 and rev3–1 mutants and between pso2-1 and snm1 mutants in Saccharomyces cerevisiae.
Curr Genet
1988
;
13
:
37
–40.
41
Borel F, Lohez OD, Lacroix FB, Margolis RL. Multiple centrosomes arise from tetraploidy checkpoint failure and mitotic centrosome clusters in p53 and RB pocket protein-compromised cells.
Proc Natl Acad Sci U S A
2002
;
99
:
9819
–24.
42
Liehr T, Claussen U, Starke H. Small supernumerary marker chromosomes (sSMC) in humans.
Cytogenet Genome Res
2004
;
107
:
55
–67.
43
Hoffelder DR, Luo L, Burke NA, Watkins SC, Gollin SM, Saunders WS. Resolution of anaphase bridges in cancer cells.
Chromosoma
2004
;
112
:
389
–97.
44
Zander L, Bemark M. Immortalized mouse cell lines that lack a functional Rev3 gene are hypersensitive to UV irradiation and cisplatin treatment.
DNA Repair (Amst)
2004
;
3
:
743
–52.
45
Sonoda E, Okada T, Zhao GY, et al. Multiple roles of Rev3, the catalytic subunit of pol zeta in maintaining genome stability in vertebrates.
EMBO J
2003
;
22
:
3188
–97.
46
Takao N, Kato H, Mori R, et al. Disruption of ATM in p53-null cells causes multiple functional abnormalities in cellular response to ionizing radiation.
Oncogene
1999
;
18
:
7002
–9.
47
Seki M, Masutani C, Yang LW, et al. High-efficiency bypass of DNA damage by human DNA polymerase Q.
EMBO J
2004
;
23
:
4484
–94.
48
Grossmann KF, Ward AM, Matkovic ME, Folias AE, Moses RE. S. cerevisiae has three pathways for DNA interstrand crosslink repair.
Mutat Res
2001
;
487
:
73
–83.
49
Pavlov YI, Shcherbakova PV, Kunkel TA. In vivo consequences of putative active site mutations in yeast DNA polymerases α, epsilon, δ, and zeta.
Genetics
2001
;
159
:
47
–64.
50
McKee RH, Lawrence CW. Genetic analysis of γ-ray mutagenesis in yeast. III. Double-mutant strains.
Mutat Res
1980
;
70
:
37
–48.
51
Swanson RL, Morey NJ, Doetsch PW, Jinks-Robertson S. Overlapping specificities of base excision repair, nucleotide excision repair, recombination, and translesion synthesis pathways for DNA base damage in Saccharomyces cerevisiae.
Mol Cell Biol
1999
;
19
:
2929
–35.
52
Sakai W, Ishii C, Inoue H. The upr-1 gene encodes a catalytic subunit of the DNA polymerase zeta which is involved in damage-induced mutagenesis in Neurospora crassa.
Mol Genet Genomics
2002
;
267
:
401
–8.
53
Sakamoto A, Lan VT, Hase Y, Shikazono N, Matsunaga T, Tanaka A. Disruption of the AtREV3 gene causes hypersensitivity to ultraviolet B light and γ-rays in Arabidopsis: implication of the presence of a translesion synthesis mechanism in plants.
Plant Cell
2003
;
15
:
2042
–57.
54
Wu F, Lin X, Okuda T, Howell SB. DNA polymerase zeta regulates cisplatin cytotoxicity, mutagenicity, and the rate of development of cisplatin resistance.
Cancer Res
2004
;
64
:
8029
–35.
55
Singer MJ, Mesner LD, Friedman CL, Trask BJ, Hamlin JL. Amplification of the human dihydrofolate reductase gene via double minutes is initiated by chromosome breaks.
Proc Natl Acad Sci U S A
2000
;
97
:
7921
–6.
56
Iliakis G, Wang H, Perrault AR, et al. Mechanisms of DNA double strand break repair and chromosome aberration formation.
Cytogenet Genome Res
2004
;
104
:
14
–20.
57
Richardson C, Jasin M. Frequent chromosomal translocations induced by DNA double-strand breaks.
Nature
2000
;
405
:
697
–700.
58
Hande MP, Samper E, Lansdorp P, Blasco MA. Telomere length dynamics and chromosomal instability in cells derived from telomerase null mice.
J Cell Biol
1999
;
144
:
589
–601.
59
Ferguson DO, Sekiguchi JM, Chang S, et al. The nonhomologous end-joining pathway of DNA repair is required for genomic stability and the suppression of translocations.
Proc Natl Acad Sci U S A
2000
;
97
:
6630
–3.
60
d'Adda di Fagagna F, Hande MP, Tong WM, et al. Effects of DNA nonhomologous end-joining factors on telomere length and chromosomal stability in mammalian cells.
Curr Biol
2001
;
11
:
1192
–6.
61
Goytisolo FA, Samper E, Edmonson S, Taccioli GE, Blasco MA. The absence of the DNA-dependent protein kinase catalytic subunit in mice results in anaphase bridges and in increased telomeric fusions with normal telomere length and G-strand overhang.
Mol Cell Biol
2001
;
21
:
3642
–51.
62
Smiraldo PG, Gruver AM, Osborn JC, Pittman DL. Extensive chromosomal instability in Rad51d-deficient mouse cells.
Cancer Res
2005
;
65
:
2089
–96.
63
Deans B, Griffin CS, O'Regan P, Jasin M, Thacker J. Homologous recombination deficiency leads to profound genetic instability in cells derived from Xrcc2-knockout mice.
Cancer Res
2003
;
63
:
8181
–7.
64
Tarsounas M, Munoz P, Claas A, et al. Telomere maintenance requires the RAD51D recombination/repair protein.
Cell
2004
;
117
:
337
–47.
65
Takeuchi S, Koike M, Seriu T, et al. Frequent loss of heterozygosity on the long arm of chromosome 6: identification of two distinct regions of deletion in childhood acute lymphoblastic leukemia.
Cancer Res
1998
;
58
:
2618
–23.
66
Schlegelberger B, Himmler A, Bartles H, Kuse R, Sterry W, Grote W. Recurrent chromosome abnormalities in peripheral T-cell lymphomas.
Cancer Genet Cytogenet
1994
;
78
:
15
–22.
67
Starostik P, Greiner A, Schultz A, et al. Genetic aberrations common in gastric high-grade large B-cell lymphoma.
Blood
2000
;
95
:
1180
–7.
68
Zhang Y, Matthiesen P, Harder S, et al. A 3-cM commonly deleted region in 6q21 in leukemias and lymphomas delineated by fluorescence in situ hybridization.
Genes Chromosomes Cancer
2000
;
27
:
52
–8.
69
Sherratt T, Morelli C, Boyle JM, Harrison CJ. Analysis of chromosome 6 deletions in lymphoid malignancies provides evidence for a region of minimal deletion within a 2-megabase segment of 6q21.
Chromosome Res
1997
;
5
:
118
–24.
70
Bea S, Zettl A, Wright G, et al. Diffuse large B-cell lymphoma subgroups have distinct genetic profiles that influence tumor biology and improve gene expression-based survival prediction.
Blood
2005
;
106
:
3183
–90.
71
Morelli C, Karayianni E, Magnanini C, et al. Cloning and characterization of the common fragile site FRA6F harboring a replicative senescence gene and frequently deleted in human tumors.
Oncogene
2002
;
21
:
7266
–76.
72
Jin Y, Jin C, Salemark L, Martins C, Wennerberg J, Mertens F. Centromere cleavage is a mechanism underlying isochromosome formation in skin and head and neck carcinomas.
Chromosoma
2000
;
109
:
476
–81.