Targeted disruption of Np95 gene renders murine embryonic stem cells hypersensitive to DNA damaging agents and DNA replication blocks.

NP95, which contains a ubiquitin-like domain, a cyclin A/E-Cdk2 phosphorylation site, a retinoblastoma (Rb) binding motif, and a ring finger domain, has been shown to be colocalized as foci with proliferating cell nuclear antigen in early and mid-S phase nuclei. We established Np95 nulligous embryonic stem cells by replacing the exons 2-7 of the Np95 gene with a neo cassette and by selecting out a spontaneously occurring homologous chromosome crossing over with a higher concentration of neomycin. Np95-null cells were more sensitive to x-rays, UV light, N-methyl-N"-nitro-N-nitrosoguanidine (MNNG), and hydroxyurea than embryonic stem wild type (Np95(+/+)) or heterozygously inactivated (Np95(+/-)) cells. Expression of transfected Np95 cDNA in Np95-null cells restored the resistance to x-rays, UV, MNNG, or hydroxyurea concurrently to a level similar to that of Np95(+/-) cells, although slightly below that of wild type (Np95(+/+)) cells. These findings suggest that NP95 plays a role in the repair of DNA damage incurred by these agents. The frequency of spontaneous sister chromatid exchange was significantly higher for Np95-null cells than for Np95(+/+) cells or Np95(+/-) cells (p < 0.001). We conclude that NP95 functions as a common component in the multiple response pathways against DNA damage and replication arrest and thereby contributes to genomic stability.

have lost a checkpoint control may be as DNA damage-sensitive as cells that have lost DNA repair capability. Many genes are involved in controlling the cell cycle and determining checkpoints (1)(2)(3). Furthermore, a possible link between checkpoint failure, hypersensitivity to DNA damage or replication blockage, and genomic instability has provided an important insight into processes contributing to the cellular dysfunction that leads to cancer (4).
We previously produced a monoclonal antibody, Th-10a mAb, 1 that recognizes a 95-kDa mouse nuclear protein (NP95). NP95 was detected by the Th-10a mAb, specifically in the S phase of normal mouse thymocytes. In contrast, mouse T cell lymphoma cells showed a constantly high level for NP95 accumulation irrespective of cell stages during the cell cycle (5). By immunoscreening a gt-11 cDNA expression library with the Th-10a mAb, we isolated the cDNA encoding NP95 (6). Sequencing of the whole 3.5-kb cDNA revealed that NP95 is a novel nuclear protein with an open reading frame consisting of 782 amino acids. The open reading frame contains an unusual N-terminal domain that bears a striking resemblance to ubiqutin, a leucine zipper motif, a zinc finger motif, a potential ATP/GTP binding site, a putative cyclin A/E-cdk2 phosphorylation site, retinoblastoma protein (Rb)-binding motifs, and a ring finger domain (6). NP95 was shown to be localized in S phase nuclei as dot-like foci (7). Double immunostaining for NP95 and chromatin-bound PCNA revealed that NP95 was almost exclusively colocalized with chromatin-bound PCNA throughout the nucleus in early S phase and partly in mid-S phase. However, distinct localization of the two proteins became evident in some part in mid-S phase and late-S phase (7,8). These results suggested that NP95, rather than committing DNA replication itself as a component of replication machinery, contributes to some other DNA replication-linked events (8).
In this study, to gain further insights into the physiological role of NP95, we attempted to produce a targeted disruption of mouse Np95. We made strenuous but ultimately futile efforts to obtain Np95 null mice, probably because Np95 inactivation caused mid-gestational embryonic lethality. We, therefore, established Np95-null embryonic stem (ES) cells by treating Np95 ϩ/Ϫ ES cells with a higher concentration of G418 (2 mg/ ml) to select out a mitotic crossing over. We examined the effect of the homozygous Np95 disruption on the sensitivity to x-rays, UV light, and N-methyl-NЈЈ-nitro-N-nitrosoguanidine (MNNG) and found that Np95-null ES cells were more sensitive to treatment with these agents than ES wild type (E14.wt) or 19(Np95 ϩ/Ϫ ) cells. In addition, Np95-null ES cells turned out to be more sensitive to DNA replication inhibitor hydroxyurea (HU) than E14.wt or 19(Np95 ϩ/Ϫ ) cells. To obtain further insights into the role of NP95 in DNA recombination and chromosomal stability, we examined the frequency of spontaneous sister chromatid exchange (SCE) in NP95-deficient cells. In view of the characteristics of NP95-deficient cells, we suggest that NP95 functions as a common component in the multiple DNA damage response pathways and that it plays a role in the maintenance of genomic stability.

Construction of the Np95 Targeting Vector
To construct the targeting vector of Np95, genomic DNA segments corresponding to the Np95 were obtained by screening of the strain 129 genomic library, 129 (9), with 1.8 kb of N-terminal cDNA of Np95 cDNA. Selected genomic fragments containing the Np95 gene were subcloned into pBluescriptII SK (Ϫ). Sequencing of each clone was done using dye terminator beginning with the T3 and T7 ends (GenBank TM accession number AB066245). The exon-intron boundary predictions were based upon comparison of the genomic sequence with Np95 cDNA (6). The targeting vector, pNPKO1, was constructed using two genomic clones, pNP1 and pNP7. The upstream 1.5-kb BamHI fragment from pNP1 and downstream 6.2-kb SalI/EcoRI fragment from pNP7 were inserted into sites of restriction enzymes, ClaI and XbaI, in pgk-Neo cassette. This insertion resulted in deletion from exon 2 to exon 7 (see Fig. 1A).

Generation of NP95-deficient ES Cells
E14 cells were grown on embryonic fibroblast feeder cells, electroporated in the presence of 30 g of NotI linearized targeting vector, pNPKO1, and selected on neomycin-resistant feeder cells with G418 at 125 g/ml for 7 days. Homologous recombinants were identified by PCR using primers NC (ϩ) (AGAGGAGAGTGCGGATCGCCGCCGTGAG-AG) in exon 1 and NEO (Ϫ) (GATTCGCAGCGCATCGCCTTCTATCG) in the poly(A) signal of the pgk-Neo cassette. PCR-positive clones were confirmed with Southern blots using probe A (300 bp) and probe B (EcoRI/KpnI fragment outside of pNPKO1). Heterozygous ES cell lines for the Np95 gene were expanded for several passages and plated onto 10 gelatin-coated 100-mm dishes at a density of 1.0 ϫ 10 6 cell/dish, cultured for 24 h, and grown in the presence of 2 mg/ml of G418. Homozygous clones that survived for 10 days were identified by Southern blots using probe A. Inactivation of the Np95 gene was confirmed by Western blot with Th-10a mAb.

PCR Analysis and Southern Blot Analysis for Identification
ES cell colonies surviving for 7 days were trypsinized and grown for 1.5 days in 96-well plates. The cells were trypsinized and duplicated. Half of these cells were centrifuged and boiled in 100 l of distilled water. A portion of cell lysate (10 l) was subjected to 30-cycle PCR amplification of homologous recombined allele using the primer set, exon 1 primer, NC (ϩ), and pgk poly(A) signal primer, NEO (Ϫ). Amplification was performed with LA Taq polymerase (TaKaRa Bio, Inc., Kyoto, Japan) as follows: initial 4-min incubation at 94°C, 30 cycles consisting of 95°C for 30 s, 60°C for 30 s, and 72°C for 90 s with a final 5-min elongation step at 72°C. Genomic DNA isolated from ES cell lines was digested with BamHI, separated by electrophoresis through 0.6% agarose gel, and transferred to nylon membrane in 2ϫ SSC. The probe utilized is described in Fig. 1A and was 32 P-labeled with random primer labeling kit (TaKaRa Bio, Inc.). The membrane was washed twice in 2ϫ SSC and analyzed with Bio-image analyzer BAS2000.

Western Blotting
3 ϫ 10 6 ES cells were seeded in 10-cm dishes, and after 12 h, they were washed twice with PBS(ϩϩ) and immediately lysed in lysis buffer (8) plus SDS sample buffer (New England BioLabs, Beverly, MA). Equal amounts of whole cell lysates were separated by 4 -20% gradient SDS-PAGE (Dai-ichi Pure Chemicals, Tokyo, Japan) and transferred to Immobilon-P (polyvinylidene difluoride) membranes (Millipore Co., Bedford, MA). Membranes blocked in 5% skim milk (DIFCO Laboratories, Detroit, MI) in PBS for 1 h were rinsed and incubated with anti-NP95 (Th-10a) or anti-PCNA (PC10, BD PharMingen) antibody in 0.5% skim milk in PBS for 1 h at room temperature, washed three times with PBS-T (0.1% Tween in PBS). Binding of primary antibody was revealed using horseradish peroxidase-coupled anti-rat IgG antibody P0450 (DAKO, Kyoto, Japan) diluted at 1:5000 in 0.5% skim milk in PBS for 1 h at room temperature and washed three times with PBS-T. Signals were developed by enhanced chemiluminescence (Lumi-Light Plus; Roche Molecular Biochemicals).

Flow Cytometric Analysis
Cells were fixed with cold 70% ethanol for 30 min at 0°C, washed twice with PBS supplemented with 5% fetal calf serum (FPBS), resuspended with 30 l of anti-Np95 monoclonal antibody (Th-10a) in FPBS, and incubated for 1 h at room temperature. The cells were then washed twice with FPBS and resuspended in fluorescein isothiocyanate-conjugated anti-rat IgG (F234, Dako Japan, Kyoto, Japan). After incubation for 1 h at room temperature, the cells were washed twice with FPBS and subjected to flow cytometric analysis (FACscan, BD PharMingen).

Transfection of Np95 cDNA into Np95-null Cells
Np95 cDNA was inserted into pcDNA 3.1 vector containing a V5 epitope and His tag (Invitrogen). To construct pIRES.hyg.Np95.V5HA expression vector, an Np95 cDNA fragment containing the Kozack sequence in the N terminus and connected with a V5 epitope and His tag in the C terminus was inserted into the BamHI site of the pIRES.hyg vector (CLONTECH). pIRES.hyg.Np95.VA.HA plasmid or empty vector (pIRES.hyg) were linearized by digestion with FspI, and 19.4(Np95 Ϫ/Ϫ ) cells were transfected with these plasmid DNAs using LipofectAMINE 2000 according to instructions provided by the manufacturer (Invitrogen). The cells were subsequently cultured for 10 days in the presence of Hygromycin B (70 -100 g/ml, Invitrogen). Thirty colonies resistant to Hygromycin B were isolated, and DNA from these colonies was isolated and analyzed to determine whether the full-length Np95 cDNA was detected by PCR. Finally, these clones were confirmed to produce NP95 by flow cytometric analysis following immunostaining of the cells with anti-NP95 mAb(Th-10a) and anti-V5 mAb.
In Vitro Treatment X-irradiation-ES clones were trypsinized and plated on gelatincoated 100-mm dishes (Falcon, BD PharMingen) at a density ranging from 1.5 ϫ 10 3 to 6 ϫ 10 3 per dish. Following incubation at 37°C for 16 h, ES cells were exposed to various doses of x-rays. Irradiation was performed with a Shinai-III x-ray generator (Shimadzu Seisakusho Ltd., Kyoto, Japan) at 0.72 Gy/min (200 kVp (kilovolts at peak), 20 mA, with filters of 0.5 mm copper and 0.5 mm aluminum).
UV Irradiation-ES clones were trypsinized, plated on gelatincoated 100-mm dishes, and cultured at 37°C for 16 h. ES cells were washed twice and then overlaid with 5 ml of Dulbecco's PBS(Ϫ) (Nissui Pharmaceutical, Co. Ltd., Tokyo, Japan). The cells were irradiated with UV light emitted from germicidal lamps (15 watts, GL-15, Toshiba, Tokyo, Japan) at room temperature. The fluence rate was 0.285 J/m 2 /s as measured with a UV intensity meter (Topcon UVR-254, Tokyo Kogaku, Tokyo, Japan).
MNNG Treatment-ES clones were trypsinized and plated on gelatin-coated 100-mm dishes at 1.0 -1.5 ϫ 10 3 cells per dish. After incubation at 37°C for 16 h, ES cells were exposed to various concentrations of MNNG in ES medium for 60 min at 37°C, washed twice with PBS(Ϫ), and then cultured further in ES medium.
HU Treatment-ES clones were trypsinized and plated on gelatincoated 100-mm dishes at 1.5 ϫ 10 3 cells. Following incubation for 16 h at 37°C, ES medium was removed, and various concentrations of HU (Sigma) were added to the cultures. After the treatment for 4 h at 37°C, ES cells were washed with 15 ml of Dulbecco's modified Eagle's medium and cultured further in ES medium.

Colony Formation
Following these treatments, ES cells were incubated with ES medium for 7 days, and colonies were scored. Plating efficiencies ranged from 30 to 50%. Relative survivals in terms of the loss of clonogenicity were based on at least two or three experiments carried out in triplicate.

Radioresistant DNA Synthesis (RDS) Assay
Twenty thousand ES cells were plated into labeling ES medium containing 0.02 Ci/ml, [2-14 C]thymidine (7 mCi/mmol; PerkinElmer Life Sciences) and were grown for 24 h. After removal of the radioactive medium, each culture was further incubated in fresh ES medium for at least another 2 h to deplete residual 14 C-labeled thymidine from endogenous DNA precursor pools. Cultures were then irradiated with 4 Gy of x-rays. At specific times during subsequent incubation, the corresponding x-ray-and sham-treated cell monolayers were pulselabeled for 1 h with 5 Ci/ml [methyl-3 H]thymidine (88.2 Ci/mmol, PerkinElmer Life Sciences). The paired cultures were rinsed with ice-cold 1ϫ SSC and lysed with 1 ml of lysis solution (0.05% SDS, 10 mM EDTA). The lysates were precipitated on wet Whatman GF/C glass filters followed by addition of 1 ml of 8% perchrolic acid. The filters were rinsed with ice-cold 4% perchrolic acid, rinsed twice with 70 and 100% ethanol, dried, and counted in a liquid scintillation spectrometer (Beckman Instruments). The rate of semiconservative DNA synthesis in irradiated cultures was calculated as follows and expressed as a percentage of incorporations for sham-treated controls, as shown in Equation 1,

SCE Analysis
ES clones were trypsinized and plated on gelatin-coated 25-cm 2 flasks (for x-rays and MNNG treatment) and 10-cm 2 plates (for UV treatment) at a 1.0 ϫ 10 6 cell density. After overnight culture, ES medium was changed, and the cells were treated with x-rays, UV, and MNNG as described above. The cells were then cultured for approximately two cell cycle periods (24 h) with ES medium containing 5 M bromodeoxyuridine and pulsed with 0.04 g/ml Colcemid for the final 1 h. The cells were collected by centrifugation and exposed to 0.075 M KCl hypotonic solution for 12 min at 37°C. The cells were washed twice with the fixative (methanol:acetic acid ϭ 3:1) and suspended in a small volume of the fixative. The cell suspension was dropped onto ice-cold wet glass slides and air-dried at 60°C for 2 h. Sister chromatid differential staining was obtained by a method described earlier (12,13). Briefly, the cells on the 2-day slides were incubated with 0.5 g/ml Hoechst 33258 in 0.067 M phosphate buffer, pH 7.0, for 10 min, rinsed with 0.067 M PBS (pH 7.0) and then overlaid with coverslips. The coverslipped slides were exposed to black light ( ϭ 352 nm) at a distance of 9 cm for 2 h. After removal of coverslips, the slides were incubated in 2ϫ SSC (0.3 M NaCl plus 0.03 M sodium citrate) solution at 60°C for 1 h and then stained with 5% Giemsa solution in 13 mM phosphate buffer, pH 6.8, for 10 min, rinsed in water, and air-dried. Student's t test was employed for statistical analyses.

Generation of NP95-deficient ES Cell
Lines-To further elucidate the physiological function of NP95, we generated cell lines with disrupted Np95. Mouse Np95 gene cloned from the genomic library of strain 129 spans about 23 kb. We con- structed a targeting vector in which part of the Np95 genomic region from exon 2 containing initiation codon to exon 7 was replaced by a pgk-neo gene cassette (Fig. 1A). After transfection with electroporation and selection in ES medium containing 125 g/ml G418, we isolated 234 G418-resistant ES clones and two clones (No. 19 and No. 169) were homologous recombinants in Np95 genome locus (Fig. 1B). To obtain NP95-deficient ES cells disrupted in both alleles, heterozygous 19(Np95 ϩ/Ϫ ) clone was selected in medium containing highly concentrated G418 (2 mg/ml), and 17 resistant clones were isolated. Two clones (19.4 and 19.7) were shown to be mutated in both of the alleles (Fig. 1B). Reverse transcript-PCR analysis (data not shown) and Western blot analysis were performed, and the Np95 gene was not expressed in these cells (Fig. 1C). Since expression of NP95 is restricted in S phase cells and co-localized with PCNA (7,8), it was assumed that NP95 was associated with DNA replication and that NP95-deficient ES cells were lethal. However, NP95-deficient ES cell lines were grown in leukemia inhibitory factor-containing medium, and the doubling time was 12 h, similar to E14.wt and 19(Np95 ϩ/Ϫ ) cell lines (data not shown). Plating efficiency of NP95-deficient ES cells was normal. This indicated that NP95 participates in DNA replication-linked events other than DNA replication itself as a replication machinery, a finding that is consistent with a previous observation (8).
Hypersensitivity to X-rays, UV, and MNNG-PCNA is known to be co-localized with proteins involved in DNA replication, DNA repair, and cell cycle regulation (14 -21, 27). To determine whether deletion of the Np95 gene affects the DNA repair, we examined the sensitivity of NP95-deficient ES cells to various types of DNA damage. E14.wt, 19(Np95 ϩ/Ϫ ), and NP95-deficient ES cells (19.4(Np95 Ϫ/Ϫ ) and 19.7(Np95 Ϫ/Ϫ )) were exposed to various doses of x-rays, and the survival curve was obtained as shown in Fig. 2. The results indicated that NP95-deficient ES cells were more sensitive to x-rays than E14.wt cells or 19(Np95 ϩ/Ϫ ) cells. D 37 (the dose required to reduce the fraction of surviving cells to 0.37) was 1.5 Gy for Np95 Ϫ/Ϫ cells, 2.4 Gy for Np95 ϩ/Ϫ cells, and 3.1 Gy for Np95 ϩ/ϩ cells. NP95-deficient cells were also more sensitive to UV light and the alkylating agent, MNNG, than E14.wt cells. D 37 for Np95 Ϫ/Ϫ cells and E14.wt Np95 ϩ/ϩ cells was ϳ2.6 J/m 2 and 4.7 J/m 2 for UV and 4.2 M and 8.5 M for MNNG, respectively (Fig. 3, A and B). Thus, NP95-deficient ES cells exhibited an ϳ2-fold sensitization to the three different kinds of DNA damaging agents as compared with E14.wt cells. The marginal difference in the cytotoxicity between E14.wt cells and 19(Np95 ϩ/Ϫ ) cells was statistically significant for 2 and 4 Gy of x-rays but not for 10 and 20 M of MNNG.
Complementation Experiment of NP95-deficient ES Cell Line-Since NP95-deficient ES cells were shown to be more sensitive to various DNA damaging agents than ES wild type cells, we examined whether transfection with Np95 cDNA into NP95-deficient ES cells could restore the response to DNA damage to a level similar to that of E14.wt and 19(Np95 ϩ/Ϫ ) cells. We constructed NP95 expression vector, in which Np95 cDNA connected with a V5 epitope and His tag at the C terminus was inserted into pIRES.hyg expression vector (CLON-TECH). The NP95 expression construct was linearized and transfected into 19.4(Np95 Ϫ/Ϫ ) cells. After screening of hygromycin B-resistant clones based on PCR and flow cytometric analysis, we obtained a considerable number of stable transfected subclones of 19.4(Np95 Ϫ/Ϫ ) cells. Among these, two Np95-transfected subclones, N2 and N8, were confirmed to have an intact structure by PCR data and concomitant detection of NP95 protein and V5 epitope. Empty vector-transfected PV-1 cells were subjected to subsequent analyses as a negative control.
The transfection with Np95 cDNA restored the resistance of 19.4(Np95 Ϫ/Ϫ ) cells to x-rays to a considerable extent, although virtually similar to that of 19(Np95 ϩ/Ϫ ) cells (N2 and N8 in Fig.  2). Less than complete recoveries to the wild type level of x-ray sensitivity were unequivocal as the difference in the surviving fraction between N8 and E14.wt was statistically significant; p Ͻ 0.02 for 2 Gy, p Ͻ 0.001 for 4 Gy. Immunostaining of these cells against NP95 followed by the fluorescence-activated cell sorter analysis revealed that the levels of NP95 expression in both N8 and N2 cells gained only about a quarter of those of E14.wt or 19(Np95 ϩ/Ϫ ) (data not shown). These inadequate levels of NP95 expression in N8 and N2 may explain the insufficient recovery. The unexpected suppression of NP95 function by the pIRES.hyg vector itself was ruled out since no difference was observed in the extent of the hypersensitivity to x-rays between 19.4(Np95 Ϫ/Ϫ ) cells and empty vector-transected PV-1 cells in an independent experiment (data not shown). Clonogenic survivals after exposure to UV light or MNNG in Np95 cDNA-transfected cells indicated that the cellular resistance to those agents was also regained, although the extent of recovery was again incomplete (N2 cells in Fig. 3, A and B). Empty vector transfection had no effect on the sensitivity of 19.4(Np95 Ϫ/Ϫ ) cells to UV or MNNG exposure (PV-1 cells in Fig. 3, A and B). These results strongly indicate that Np95 is one of the genes bestowing cellular resistance concurrently to ionizing radiation (IR), UV, and MNNG.
Hypersensitivity to HU and Complementation Experiment-Expression of a kinase-inactive allele of ATR (ATRkd) in human fibroblasts caused an increased sensitivity to IR, methyl methanesulfonate (MMS), and UV together with an DNA replication inhibitor, HU, and displayed a loss of checkpoint control (22,23). Since NP95-deficient cells were also hypersensitive to x-rays, UV, and MNNG (Figs. 2 and 3, A and B), it seemed possible to presume that the loss of NP95 likewise results in an increased sensitivity to HU. NP95-deficient ES cells were indeed more sensitive to HU than either E14.wt cells or 19(Np95 ϩ/Ϫ ) cells (19.4 cells and 19.7 cells in Fig. 4). More- over, Np95 cDNA-transfected cells restored the resistance to HU to an extent similar to that of 19(Np95 ϩ/Ϫ ) cells (N8 cells and N2 cells in Fig. 4). Empty vector-transfected PV-1 cells and 19.4(Np95 Ϫ/Ϫ ) cells showed a similar hypersensitivity to HU exposure, as we had expected. Thus, the Np95 gene appeared to confer the cellular capability to cope with DNA replication blocks incurred by HU.
Kinetics of DNA Synthesis in ES Cells after X-irradiation-The cultured skin fibroblasts derived from the patients with ataxia-telangiectasia are sensitive to IR, but such cells show little or no increase in the sensitivity to UV, alkylating agents, or anti-metabolic inhibitors of DNA replication. These cells are also characterized by an enhanced RDS (24 -26). On the other hand, the overexpression of a kinase-inactive allele of ATR in human fibroblasts causes an increased sensitivity to IR, UV light, and MMS and abrogates the G 2 /M arrest after exposure to IR. Although these cells do not show an unequivocal increase in RDS, the induction of an overexpression of wild type ATR in ataxia-telangiectasia fibroblast cells restores the ability to inhibit DNA synthesis after exposure to IR, suggesting a possible overlap between ATM and ATR functions (22).
Since NP95-deficient cells are sensitive to DNA damaging agents and DNA replication block, we analyzed whether NP95deficient ES cells possess the characteristics of an enhanced RDS. We examined the time course of alterations in post-x-ray DNA synthesis after the exposure of cells to 4 Gy and after cells are pulse-labeled with [ 3 H]thymidine at various subsequent incubation times. The outcome of a typical experiment involving NP95-deficient ES cells and control ES cells is depicted in Fig. 5. In 19.4(Np95 Ϫ/Ϫ ) and 19.7(Np95 Ϫ/Ϫ ) cells, the replication rate decreased instantly, reaching a minimal level, i.e. ϳ35-50% of that of the sham-irradiated samples, at 30 min after treatment (Fig. 5). This was followed by a 1-2-h recovery phase whereupon synthesis again declined so rapidly that by 3 h, the level was only 42-55% of that occurring in undamaged cultures. The kinetics of post-irradiation DNA synthesis in 19.4(Np95 Ϫ/Ϫ ) and 19.7(Np95 Ϫ/Ϫ ) cells were almost the same as those of E14.wt and 19(Np95 ϩ/Ϫ ). The results indicate that NP95-deficient ES cells are able to inhibit the initial DNA synthesis after exposure to x-rays, similar to the human fibroblasts, which overexpressed a kinase-inactive allele of ATR or wild type ATR.

DISCUSSION
In the present study, to further explore the function of NP95, we established NP95-deficient ES cells by introducing a neo cassette into exon 2 to exon 7 of the Np95 gene on both alleles and examined the cellular responses to different kinds of DNA damage induced by x-rays, UV, and MNNG treatment. Homozygously Np95-inactivated ES cells (Np95-null cells) were found to be more sensitive to all of these agents than E14.wt(Np95 ϩ/ϩ ) or 19(Np95 ϩ/Ϫ ) cells (Fig. 2). Transfection of Np95-null cells with Np95 cDNA revealed the ability of the Np95 gene to restore the concurrent resistance to x-rays, UV, or MNNG to almost similar levels as those of 19(Np95 ϩ/Ϫ ) cells, implying that at least part of the normal response to exposure to IR, UV, or MNNG can be ascribed to Np95 gene function (Figs. 2 and 3, A and B). The DNA repair process is itself controlled by a specific set of genes that encode the enzymes catalyzing cellular responses to various types of DNA damage. Since Np95-null cells show a similar hypersensitivity to these different kinds of DNA damage, it is plausible that NP95 plays a common critical role in different repair processes. Moreover, Np95-null ES cells were hypersensitive to a brief exposure to HU as compared with E14.wt or 19(Np95 ϩ/Ϫ ) cells (Fig. 4). HU inhibits ribonucleotide reductase and results in the rapid shrinkage of the deoxynucleotide triphosphate pools. As these substrates of the DNA polymerase become exhausted, replication is arrested, and incomplete synthesis of Okazaki fragments on the lagging strand may give rise to tracts of persistent single-stranded DNA in close proximity to the stalled replication fork. The hypersensitivity to HU in Np95-null cells together with the restoration of HU resistance by the Np95 cDNA expression in the null cells clearly suggests that NP95 confers upon cells a capability to cope with HU-induced lesions at DNA replication.
It has been shown that mutations of Saccharomyces cerevisiae mec1 or Schizosaccharomyces pombe rad3 render the yeast hypersensitive to x-rays, UV light, MMS, and HU (28,29). Rad3 and mec1 are required for checkpoint responses to x-rays, UV light, MMS, and HU (30,31). In addition to cell cycle checkpoint response, these genes have been implicated in the regulation of DNA repair (32,33). Consistent with the hypothesis that ATR is the structural and functional mammalian homolog of rad3 and mec1 (30,34), an overt expression of a kinase-inactive allele of ATR (ATRkd) in human fibroblasts has been shown to bring about an increased sensitivity to ␥-irradiation, MMS, UV, and HU and also to cause a loss of checkpoint control (22,23). Liu et al. (35) proposed a model in which ATM-Chk2 and ATR-Chk1 represent two parallel branches in the mammalian DNA damage response pathway that respond primarily to different types of DNA damage. An overexpression of wild type ATR or ATRkd was shown to correct the aberrant phenotype of RDS in cells defective in ATM function, suggesting that there may be a functional overlap between ATM and ATR (22). The initial DNA synthesis in E14.wt, 19(Np95 ϩ/Ϫ ), and Np95-null ES cells following an exposure to x-irradiation was equally inhibited regardless of Np95 status (Fig. 5)  of DNA damage and the DNA replication inhibitor, HU.
We have demonstrated that NP95 is almost completely colocalized with chromatin-bound PCNA throughout the nuclei in early S phase and also partly in mid-S phase. However, distinct localization of the two proteins in some part becomes evident in mid-S phase and late-S phase (7,8). Moreover, during meiosis, NP95 is present not only in proliferating spermatogonia but also in meiotic spermatocytes and differentiating spermatids that are not proliferating (7). PCNA has recently been identified as a member of a group of proteins that associate with BRCA1 to form a large complex, BRCA1-associated genome surveillance complex (BASC) (37). On the other hand, ATR has been found to phosphorylate BRCA1 in response to damaged DNA or stalled DNA replication, and so it is conceivable that ATR-BRCA1 complex is required for the maintenance of genomic integrity and for the control of homologous recombination in both somatic and meiotic cells (38 -42). Although the interaction between NP95 and PCNA or other repair/replicationrelated proteins remains to be clarified, we propose that NP95 functions as a component of the complex containing PCNA. SCE frequency is a commonly used index of chromosomal stability, and spontaneous chromosomal instability has been correlated with cancer predisposition (43). Because SCE occurs during and soon after DNA replication, it is conceivable that SCE takes place when the replication fork encounters unrepaired lesions and/or affected repair machineries (44). The endogenously derived lesions presumably cause SCEs by a mechanism that could also operate for lesions incurred by extrinsic DNA damaging agents. Such lesions may arise spontaneously as a consequence of the action of endogenous genotoxins including reactive oxygen species. Since Np95-null cells are considered to be defective in some common DNA repair process, unrepaired lesions are expected to accumulate more abundantly in Np95-null cells than in E14.wt or 19(Np95 ϩ/Ϫ ) cells. It was necessary, therefore, to determine whether the lesions responsible for SCE induction arose more frequently in NP95-deficient cells. The frequencies of spontaneous SCEs in NP95-null ES cells were significantly higher than those of E14.wt and 19(Np95 ϩ/Ϫ ) cells (p Ͻ 0.001) (Fig. 6). The frequencies of SCEs following irradiation with 2 Gy of x-rays were significantly higher in Np95-null cells (mean (M) Ϯ S.D. and (n) for 19.4(Np95 Ϫ/Ϫ ) and 19.7(Np95 Ϫ/Ϫ ) were 0.296 Ϯ 0.058 (10) and 0.278 Ϯ 0.082 (19), respectively) than in E14.wt cells (0.210 Ϯ 0.10 (15)) and 19(Np95 ϩ/Ϫ ) cells (0.197 Ϯ 0.090 (17)), although the enhanced SCE-induction by x-rays in Np95-null cells was not statistically significant.
It has been shown that spontaneous SCE levels are significantly reduced in chicken DT40 B cells lacking the RAD51 and RAD54 genes, suggesting that homologous recombination is one of the principal mechanisms responsible for SCE formation in vertebrate cells (45). On the other hand, spontaneous SCE levels are observed as being elevated to various extents in cells from patients with Bloom's syndrome (46), in mouse cells that lack poly(ADP-ribose) polymerase (47) or KU70 (48), and in hamster cells with deficiency in XRCC1 (49). Inactivating mutations or disruptions of genes encoding these proteins have been demonstrated to lead to genomic instability. Although the frequency of spontaneously arising chromosome aberrations in NP95-deficient cells remains to be determined, it is highly likely that NP95 is responsible for the maintenance of chromosomal stability. Taken together, we propose that NP95 functions as a common component in the multiple response pathways for DNA damage and that it plays some role in the maintenance of genomic stability.