Involvement of DNA-dependent Protein Kinase in UV-induced Replication Arrest*

Cells exposed to UV irradiation are predominantly arrested at S-phase as well as at the G1/S boundary while repair occurs. It is not known how UV irradiation induces S-phase arrest and yet permits DNA repair; however, UV-induced inhibition of replication is efficiently reversed by the addition of replication protein A (RPA), suggesting a role for RPA in this regulatory event. Here, we show evidence that DNA-dependent protein kinase (DNA-PK), plays a role in UV-induced replication arrest. DNA synthesis of M059K (DNA-PK catalytic subunit-positive (DNA-PKcs+)), as measured by [3H]thymidine incorporation, was significantly arrested by 4 h following UV irradiation, whereas M059J (DNA-PKcs−) cells were much less affected. Similar results were obtained with the in vitro replication reactions where immediate replication arrest occurred in DNA-PKcs+ cells following UV irradiation, and only a gradual decrease in replication activity was observed in DNA-PKcs− cells. Reversal of replication arrest was observed at 8 h following UV irradiation in DNA-PKcs+cells but not in DNA-PKcs− cells. Reversal of UV-induced replication arrest was also observed in vitro by the addition of a DNA-PK inhibitor, wortmannin, or by immunodepletion of DNA-PKcs, supporting a positive role for DNA-PK in damage-induced replication arrest. The RPA-containing fraction from UV-irradiated DNA-PKcs+ cells poorly supported DNA replication, whereas the replication activity of the RPA-containing fraction from DNA-PKcs− cells was not affected by UV, suggesting that DNA-PKcs may be involved in UV-induced replication arrest through modulation of RPA activity. Together, our results strongly suggest a role for DNA-PK in S-phase (replication) arrest in response to UV irradiation.

Cells exposed to UV irradiation are predominantly arrested in S-phase rather than at the G 1 /S boundary while repair occurs (1). The molecular mechanism of damage-induced Sphase arrest is not known; however, the effects of UV irradiation during S-phase on subsequent cell cycles are magnified in repair-deficient cells (2), indicating that these effects may be initiated by DNA damage itself. In contrast, in vitro replication experiments with cytosolic extracts from UV-damaged cells strongly indicate that UV-induced inhibition of replication is not due to a blockade of replication by DNA damage itself; rather, irradiation probably induces a mechanism that inhibits DNA replication (3,4). It is not known how DNA damage induces the inhibition of DNA replication and yet permits DNA repair; however, proteins such as replication protein A (RPA 1 ; also known as human single-stranded DNA-binding protein) and proliferating cell nuclear antigen (PCNA) are involved in both processes (5)(6)(7)(8)(9)(10) and may play a role in differential regulation. Earlier in vitro studies suggested that PCNA interacts with UV-induced protein, p21 Cip1/Waf1 , which inhibits PCNA's function in DNA replication but not in repair (11)(12)(13). PCNA also interacts with GADD45 and MyD118, which are induced upon growth arrest and DNA damage, supporting a role for PCNA in damage-induced cell cycle arrest (14,15).
RPA is a heterotrimeric single-stranded DNA-binding protein (70-, 34-, and 11-kDa subunits) originally identified as an essential factor for the replication of SV40 DNA (6,9,10). In addition to its role in replication, RPA is also required for DNA repair (5,16,17) and genetic recombination (18 -20), suggesting a possible role in regulation. In replication, RPA interacts with SV40 T-antigen and DNA polymerase ␣-primase complex, which probably mediates unwinding of SV40 origin-containing DNA (21)(22)(23)(24)(25)(26)(27)(28)(29). In addition, RPA stimulates polymerase ␣, ␦, and ⑀, which suggests its potential role in the elongation stage (30,31). The middle subunit of RPA is phosphorylated in a cell cycle-dependent manner (32) and also by UV and ionizing radiation (3,33). DNA-PK is responsible for the hyperphosphorylation of the 34-kDa subunit of RPA (34,35); however, the in vivo observations with yeast and mammalian systems suggest additional involvement of other kinases, such as ataxia-telangiectasia mutant (ATM) (36,37). The observation that damageinduced RPA phosphorylation interferes with its interaction with p53 and DNA-PK suggests a positive role for RPA in regulating the p53-dependent damage checkpoint pathway (38,39). The recent in vivo finding that human RPA is phosphorylated in ATM cells defective in IR-induced S-phase arrest suggested that damage-induced RPA phosphorylation may not be coupled to the S-phase checkpoint (40). Nonetheless, it is not clear whether RPA phosphorylation plays a role in replication or repair (41,42).
DNA-PK is a nuclear serine/threonine protein kinase consisting of a 460-kDa catalytic subunit (DNA-PKcs) and the Ku heterodimer (Ku70 and Ku80). DNA-PK is activated by ionizing radiation and UV irradiation (43). The Ku heterodimer regulates DNA-PK's kinase activity upon binding to DNA (44 -46). Mouse and human cells deficient in DNA-PKcs are hypersensitive to ionizing radiation and defective in V(D)J recombination (43,47), suggesting a role for the kinase in doublestrand break repair and recombination. DNA-PK associates with the RNA polymerase I and II transcription complexes and may negatively regulate them (48 -51). Recent observations also suggest a possible role for DNA-PK in controlling apoptosis and the length of telomeric chromosomal ends (43,52,53). DNA-PKcs is a member of the phosphatidylinositol-3 kinase (PI-3 kinase) family and shares amino acid sequence homology in its carboxyl-terminal kinase domain with other family members, including the ATM gene, the ATM-related gene, and p110 PI-3 kinase (54,55). All members of the PI 3-kinase family are activated by stress; PI-3 kinase is regulated by heat shock and DNA-damage, and ATM and DNA-PK are activated by DNA damage (45, 56 -58). Recent observations indicate that DNA-PK mutant cells exhibit sensitivity to UV irradiation and cisplatin and are associated with lower nucleotide excision repair activity, suggesting a positive role for DNA-PK in DNA repair (59). Also, studies with cisplatin-resistant and -sensitive cells indicate that higher levels of DNA-PKcs expression promote cell resistance to DNA-damaging drugs, whereas the low DNA-PK activity is associated with cells with a drug-sensitive phenotype (60,61). These results suggest that DNA-PK not only senses DNA damage but also functions as a transmitter of signals that allow repair of damaged DNA and protects cells from apoptosis.
Previous studies with UV-irradiated HeLa cells suggested a role for RPA in UV-induced inhibition of replication because this event was reversed by the addition of RPA (3). In this report, we investigated a role for DNA-PK in replication arrest following UV irradiation. We present evidence that DNA-PK plays an essential role in UV-induced replication arrest, such that DNA-PK, upon UV irradiation, acts to induce replication arrest without affecting DNA repair activity.

EXPERIMENTAL PROCEDURES
Preparation of Plasmids, Antibodies, and Proteins-SV40 replication origin-containing plasmid, pSV01⌬EP, and SV40 T-antigen were prepared as described previously (62). Antipolymerase ␣ monoclonal antibody (SJK237) and anti-RPA (p34 and p70) polyclonal antibodies (from rabbits) were described previously (62). An anti-DNA-PKcs antibody was a kind gift from Dr. C. Anderson (Brookhaven National Laboratory), and an anti-PCNA antibody was purchased from Calbiochem. DNA-PK holoenzyme was purified from HeLa cells according to the procedure described previously (44).
Cell Culture and UV Irradiation of Cells-Two malignant glioblastoma cells, M059K (DNA-PK ϩ ) and M059J (DNA-PK -) were obtained from Dr. M. J. Allalunis-Turner (Cross Cancer Institute, Edmonton, Canada), mouse SCID-st cells were from Drs. J. M. Brown and C. Kirchgessner (Stanford University School of Medicine, Palo Alto, CA), and NIH3T3 was from Dr. M. Marshall (Indiana University School of Medicine, Indianapolis, IN). Monolayer culture of HeLa cells, M059K, and M059J were grown in tissue culture dishes (150 ϫ 25 mm) in Dulbecco's modified Eagle's medium/F-12 supplemented with 10% fetal bovine serum, and mouse SCID-st and NIH3T3 cells were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum at 37°C in a CO 2 incubator. Culture dishes with 80% confluence were washed twice with 10 ml of phosphate-buffered saline (PBS) and were exposed to UV-C light (GE, G30T8) in the presence of 5 ml of PBS. After adding fresh medium, UV-irradiated cells were further incubated for the indicated amount of time at 37°C in a CO 2 incubator. Nonirradiated cells were also prepared the same way without UV irradiation. To study the effect of wortmannin, cells were pretreated with wortmannin (1.0 M) for 1 h prior to UV irradiation and continued to grow in the presence of wortmannin until time of harvest.
DNA Synthesis in Vivo-Cells (5 ϫ 10 5 /60-mm dish) were incubated with 0.5 Ci/ml [ 3 H]thymidine (75 Ci/mmol) for 1 h prior to UV irradiation at 10 J/m 2 . At the indicated time points, cell metabolism was stopped by the addition of 0.1 volume of 2.3 M citric acid. After washing the cells with PBS, DNA was precipitated with 10% trichloroacetic acid at 4°C for 2 h, followed by acid-insoluble radioactivity measurement.
Cell Extracts and the Fractions-Cytosolic cell extracts were prepared according to the procedure originally described by Li and Kelly (63). Briefly, monolayer cells were washed twice with PBS and hypotonic buffer. After removing the excess amount of buffer, the swollen cells were scraped into a Dounce homogenizer (approximate volume of 0.2 ml/dish) and dounced 5-8 times on ice. Cell lysates were then centrifuged at 14,000 rpm for 30 min at 4°C to remove nuclear pellets and the insoluble materials. Ammonium sulfate (AS) fractionation of cytosolic cell extracts was done as described previously (9).
Western Blot Analysis-Immunoblotting was performed as described previously (62). Protein fractions were separated on a 12% SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose (Millipore Corp.), and immunoblotted with either monoclonal or polyclonal antibodies. After incubation with either 125 I-protein A or 125 I-protein G, proteins were visualized by autoradiography.
In Vitro SV40 DNA Replication-Replication reactions were carried out as described previously (26). Briefly, reaction mixtures (40 l) contained 40 mM creatine phosphate-di-Tris salt (pH 7.7); 1 g of creatine kinase; 7 mM MgCl 2 ; 0.5 mM dithiothreitol; 4 mM ATP; 200 M UTP, GTP, and CTP; 100 M dTTP, dGTP, and dCTP; 20 M [␣-32 P]dATP (specific activity of 30,000 cpm/pmol); 0.8 g of SV40 T-antigen; 0.3 g of SV40 origin-containing DNA (pSV01⌬EP); and the indicated amounts of RPA. The reaction mixtures were incubated for 60 min at 37°C and then stopped with 40 l of stop solution containing 20 mM EDTA, 1% SDS, and Escherichia coli tRNA (0.5 mg/ml). One-tenth of the reaction mixture was used to measure the acid-insoluble radioactivity. Replication products in the remaining reaction mixture were analyzed electrophoretically, separating the isolated DNA in a 1% agarose gel with TBE buffer. The gel was subsequently dried and exposed to x-ray film.

RESULTS
In Vivo Analysis of DNA-PKcs ϩ and DNA-PKcs Ϫ Cells for UV-induced Replication Arrest-To understand the molecular mechanism of UV-induced replication arrest, we examined whether DNA-PK plays a role in this regulatory event. For this, two malignant glioblastoma human cells (M059K (DNA-PKcs ϩ ) and M059J (DNA-PKcs Ϫ )) (65) were labeled with [ 3 H]thymidine (0.5 Ci/ml) and examined for in vivo DNA synthesis at various time points following a low dose of UV irradiation (10 J/m 2 ). The amounts of DNA synthesis in asynchronously grown M059K (DNA-PKcs ϩ ) and M059J (DNA-PKcs Ϫ ) cells were similar in the absence of UV irradiation, but DNA synthesis was significantly inhibited following UV irradiation ( Fig. 1). Most importantly, much tighter replication arrest was observed with DNA-PKcs ϩ cells compared with that with DNA-PKcs Ϫ cells up to 8 h following UV irradiation ( Fig.  1), suggesting a possible role for DNA-PKcs (or its holoenzyme) in UV-induced replication arrest. It should be pointed out, however, that the inhibition of replication following UV damage in DNA-PKcs ϩ cells could be due to the G 1 checkpoint arrest that results in fewer cells traversing the G 1 /S boundary.
DNA-PK Is Essential for Immediate Replication Arrest and Its Reversal in Response to UV Damage-To further investigate a role for DNA-PKcs in UV-induced replication arrest, we prepared cell extracts from M059K (DNA-PKcs ϩ ) and M059J (DNA-PKcs Ϫ ) cells at various times following UV irradiation (10 J/m 2 ) and examined in vitro DNA replication activity using SV40 origin-containing DNA. Replication activity of DNA-PKcs ϩ cells sharply declined within 2 h following UV irradiation, whereas DNA-PKcs Ϫ cells were only slightly affected by UV irradiation (Fig. 2A, lanes 1-3 versus lanes 6 -8). A striking difference between DNA-PKcs ϩ and DNA-PKcs Ϫ was observed 12-24 h after UV irradiation, such that the reversal of inhibition of replication was observed in DNA-PKcs ϩ cells, but not in DNA-PKcs Ϫ cells (Fig. 2A, lane 5 versus lane 10). Treatment of cells with wortmannin (1.0 M), an inhibitor of DNA-PK, abolished both rapid replication arrest and the reversal of the arrest in UV-irradiated DNA-PKcs ϩ cells but showed a gradual decrease in replication activity similar to that observed in DNA-PKcs Ϫ cells ( Fig. 2A, lanes 11-15). Reversal of replication arrest in DNA-PKcs ϩ cells occurred in a UV dose-dependent manner, which requires low UV dosage (10 J/m 2 ) (Fig. 2B). With high dose UV irradiation, DNA-PK ϩ cells may induce apoptotic signal without DNA replication. In contrast to replication arrest, UV irradiation had no effect on nucleotide excision repair activity regardless of DNA-PKcs presence or absence (Fig. 2D). Taken together, these results strongly suggest that DNA-PKcs plays a crucial role in UV-induced replication arrest, and that may also be necessary for the reversal of the arrest.
Reversal of DNA-PK-mediated Replication Arrest-If DNA-PKcs (or DNA-PK holoenzyme) is directly involved in UVinduced replication arrest, we may also see stimulation of replication with cell extracts from UV-irradiated DNA-PKcs ϩ cells by blocking DNA-PK kinase activity. To examine this, cell extracts from either nonirradiated or UV-irradiated DNA-  1, 3, 5, 7, 9, and 11) and 300 g (lanes 2, 4, 6, 8, 10, and 12) of whole cell extracts were added. The top panel indicates a fluorograph of the gel, and the bottom panel is an autoradiogram.

FIG. 2-continued
PKcs ϩ cells were preincubated with wortmannin at 37°C for 30 min and examined for replication activity (Fig. 3A). Replication arrest caused by UV irradiation of DNA-PKcs ϩ cells was reversed up to 80% by incubating cell extracts with wortmannin (Fig. 3A, lanes 5-7) under conditions where replication activity of nonirradiated DNA-PKcs ϩ cells was only slightly stimulated (Fig. 3A, lanes 2-4). In contrast, replication activity of cell extracts from DNA-PKcs Ϫ cells was unaffected by wortmannin in the presence or absence of UV irradiation (Fig. 3B). The amount of wortmannin (1.0 M) used in this experiment was sufficient to inhibit more than 95% of DNA-PK kinase activity present in cell extracts (Fig. 3C). Reversal of UV-induced rep-lication arrest required preincubation of extracts with wortmannin at 37°C prior to replication reaction (Fig. 3D), suggesting that DNA-PK is involved in replication arrest through a modulation of target protein. Together, this in vitro result is consistent with the in vivo observations (Figs. 1 and 2) that DNA-PKcs (or its holoenzyme) plays a crucial role in UVinduced replication arrest.
It is still possible, however, that the effect of wortmannin on the reversal of UV-induced replication arrest may not be directly related to DNA-PK, because wortmannin also inhibits other PI-3 kinases such as ATM and ATM-related. Therefore, we immunologically depleted DNA-PKcs from cell extracts us-  A and B). A, cell extracts (350 g) from either nonirradiated (lanes 1-4) or UV-irradiated (10 J/m 2 ) (lanes 5-7). DNA-PKcs ϩ (M059K) cells were preincubated with 0 M (lanes 1, 2, and 5), 0.2 M (lanes 3 and  6), or 1.0 M (lanes 4 and 7) of wortmannin for 30 min at 37°C before adding to the replication reaction mixtures. Replication reactions were carried out as described in the legend to Fig. 2 7) concentrations of wortmannin for 30 min at 37°C before adding to the replication reaction mixtures. C, effect of wortmannin on DNA-PK activity of cell extracts. Cell extracts (5.5 g) from nonirradiated or UV-irradiated cells (DNA-PKcs ϩ and DNA-PKcs Ϫ ) were used to measure DNA-PK activity in the presence of various concentrations of wortmannin (see "Experimental Procedures" for details). D, preincubation of cell extracts with wortmannin is necessary for the reversal of UV-induced replication arrest. Cell extracts (350 g) from nonirradiated (lanes 1-4) or UV-irradiated (lanes 5-11) DNA-PKcs ϩ cells were preincubated with 1.0 M wortmannin for various times at 37°C before adding to the replication mixtures. Replication reactions were carried out as described in the legend to Fig. 2. Tag, T-antigen.
ing anti-DNA-PKcs polyclonal antibody to see whether the depletion of DNA-PKcs can also reverse UV-induced replication arrest. The 460-kDa catalytic subunit of DNA-PK was successfully depleted from cell extracts of nonirradiated or irradiated cells as determined by immunoblot analysis (Fig. 4A). Immunodepletion of DNA-PKcs from extracts of nonirradiated cells had very little effect on replication activity (Fig. 4B, compare lane 2 with lane 4), whereas immunodepleted extracts from UV-irradiated cells showed marked stimulation of replication activity (Fig. 4B, compare lane 5 with lane 7). The addition of increasing amounts of purified DNA-PK holoenzyme to the immunodepleted extracts restored replication arrest (Fig. 4B,  lanes 8 -10), suggesting that the stimulation of replication activity in immunodepleted extracts was due to the removal of DNA-PKcs (or its holoenzyme) from the extracts. Also, these results suggest that the reversal of replication arrest by wortmannin treatment (Fig. 3) was due to the inhibition of DNA-PK activity rather than blocking other PI-3 kinases.
Modulation of RPA Occurs in DNA-PKcs ϩ Cells but Not in DNA-PKcs Ϫ Cells following UV Irradiation-DNA-PK phosphorylates RPA in response to damage from UV irradiation and/or ionizing radiation (3,33), although the in vivo studies suggested possible involvement of ATM in RPA phosphorylation (36,37). A tight replication arrest caused by UV irradiation of DNA-PKcs ϩ cells may be due to the modulation of RPA activity. A previous study indicated that replication arrest in UV-irradiated HeLa (DNA-PKcs ϩ ) cells was partially restored by the addition of purified RPA (3). We therefore examined whether the addition of RPA can reverse replication arrest of UV-irradiated DNA-PKcs ϩ cells. Similar to Fig. 2A, cell extracts from DNA-PKcs ϩ cells compared with those from DNA-PKcs Ϫ cells showed a tight replication arrest following UV irradiation (Fig. 5A). However, the addition of purified RPA did not reverse the replication arrest observed in UV-irradiated DNA-PKcs ϩ cells (Fig. 5A, lanes 12-14). Under these conditions, replication activity of DNA-PKcs Ϫ cells was slightly stimulated by RPA (Fig. 5A, lanes 5-7). Based on the results of Figs. 3 and 4 and the fact that DNA-PK kinase activity is required to maintain replication arrest induced by UV irradiation, the result from Fig. 5A can be interpreted to mean that DNA-PK from UV-irradiated cell extracts not only modulates endogenous RPA but also affects exogenously added RPA. Alternatively, DNA-PK may be involved in UV-induced replication arrest through modulation of other replication factor(s) rather than RPA.
To test whether DNA-PK ultimately targets RPA or other replication factor(s) following UV irradiation, we biochemically fractionated cell extracts, such that the 0 -35% (w/v) (NH 4 ) 2 SO 4 fraction (AS0/35) contained a single replication factor, RPA, and the 35-65% (w/v) (NH 4 ) 2 SO 4 fraction (AS35/65) contained all other replication factors such as PCNA, polymerase ␣-primase complex, and DNA-PKcs, as determined by Western blot analysis (Refs. 9 and 26; Fig. 5B). The RPA-containing fraction (AS0/35) from UV-irradiated DNA-PKcs ϩ cells poorly supported DNA replication in vitro ( Fig. 5C; lanes 11 and 12), whereas the fraction from UV-irradiated DNA-PKcs Ϫ cells efficiently supported replication ( Fig. 5C; lanes 5 and 6). Similar results were obtained with the RPA-containing fraction (AS0/ 35) from UV-irradiated mouse severe combined immune deficiency (SCID) cells lacking DNA-PKcs, so-called SCID-st, compared with DNA-PKcs ϩ mouse cells (NIH3T3) (Fig. 5D), although the DNA-PK activity of M059K cells (DNA-PKcs ϩ ) was 25 times higher than that of NIH3T3 cells (Fig. 5E). This result strongly suggests that DNA-PKcs may be involved in UV-induced replication arrest through modulation of RPA activity. It should be pointed out, however, that the AS0/35 fraction contains numerous other proteins in addition to RPA, and the failure of this fraction to complement the in vitro replication system may be due to the presence of inhibitor(s) targeted at components of the replication machinery other than RPA. DISCUSSION The molecular mechanism of damage-induced S-phase arrest is poorly understood; however, the effects of UV irradiation during S-phase on subsequent cell cycles are magnified in repair-deficient cells (2), indicating that these effects may be initiated by the DNA damage itself. On the other hand, cytosolic extracts from UV-damaged cells poorly supported DNA replication in vitro (3), suggesting that UV irradiation may induce a mechanism that inhibits DNA replication (3,4). Previous studies with cell extracts from either UV-or heat-treated cells strongly suggested the involvement of a trans-acting factor(s) in damage-induced replication arrest (3,66). UV-induced replication arrest can be partially reversed by the addition of purified RPA (3), suggesting that RPA may be a trans-acting factor involved in S-phase arrest. This notion is supported by a study with Saccharomyces cerevisiae that shows that RPA is required for G 1 /S and S-phase checkpoint arrest in response to UV or methyl methane sulfate (67).
Our study described in this paper provides evidence that DNA-PK, in response to UV irradiation, plays a crucial role in replication arrest but has no effect on DNA repair. Replication arrest occurred immediately following UV irradiation in DNA-PKcs ϩ cells but not in DNA-PKcscells. DNA replication in the DNA-PKcs Ϫ cells was still inhibited by UV, albeit to a lesser degree than in the DNA-PKcs ϩ cells. Galloway et al. (68) recently showed that the DNA-PKcs Ϫ cells, M059J, actually expresses the DNA-PK transcripts, although at a greatly reduced level, which raises a possibility that the residual inhibition seen in this cell line may be due to a low level of DNA-PK. Interestingly, the reversal of UV-induced replication arrest was observed in DNA-PKcsϩ cells, whereas a slow decrease in replication activity occurred in DNA-PKcs Ϫ cells. These results suggested that the immediate replication arrest might be necessary for the efficient DNA repair of damaged DNA following UV damage. In light of this, the physiological role of DNA-PK in UV-induced replication arrest may be to protect cells from DNA damage. In fact, much higher cell survival was observed with M059K (DNA-PKcs ϩ ) cells compared with that with M059J (DNA-PKcs Ϫ ) in response to UV irradiation or cisplatin treatment. 2 Although both DNA-PKcs and ATM mutants are hypersensitive to ionizing radiation and radiomimetic agents, the involvement of DNA-PK in UV-induced replication arrest is probably unique for DNA-PKcs among the PI-3 kinase superfamily, because UV-induced replication arrest was still observed in ATM cells (data not shown).
Interestingly, the effect of DNA-PK on the in vivo chromosomal replication was much less dramatic than that on the in vitro SV40 replication (Figs. 1 and 2). Furthermore, the kinetics of the DNA-PK-mediated inhibition was quite different in the two systems; maximal inhibition was reached in 4 h in the in vitro replication system, whereas it took approximately 12 h in the in vivo chromosomal replication following UV irradiation. This raises a question of whether an in vitro SV40 viral DNA replication system can be used to study the in vivo Sphase checkpoint on chromosomal DNA. In fact, a recent in vitro study strongly suggested that virus-encoded protein, SV40 T-antigen, could be one of the main targets for DNA-PK (69). Alternatively, the difference between the two systems could be due to the possibility that the factors normally restricted to a cell cycle stage outside S-phase may cause the artifactual effects on the in vitro replication when cell extracts were made from an asynchronous cell population.
At this time, it is not clear how DNA-PK is involved in RPA modulation and replication arrest following UV irradiation. Our wortmannin study described in Figs. 2 and 3 strongly indicated that DNA-PK kinase activity is necessary for this regulatory event. One possibility is that RPA may be the direct target for DNA-PK in replication arrest. Earlier observation with SCID mice cells showed that RPA phosphorylation in response to ionizing radiation was somewhat correlated with its decreased ssDNA binding activity, suggesting that direct phosphorylation of RPA may be involved in damage-induced replication arrest by reducing its binding activity to DNA (37). Phosphorylation of 34-kDa subunit of RPA may lead to a disassembly of the RPA heterotrimer (70) that would eventually affect RPA function in DNA replication. Nonetheless, the in vitro experiments with hyperphosphorylated RPA showed that replication and repair activities were not affected by RPA p34 phosphorylation (41). Recent observation has indicated that the 70-kDa subunit of RPA physically interacts with DNA-PKcs, and that leads to the phosphorylation of 34-kDa subunit following DNA damage (39). RPA phosphorylation may be indirectly involved in replication arrest through the interaction of other factors that are modulated by DNA-PK (42).
RPA is also an essential factor for nucleotide excision repair; however, unlike DNA replication, repair activity of DNA-PKcs ϩ cells was not affected by UV irradiation. This result suggests that DNA-PK may function as a master regulator of differential regulation in DNA replication and repair following UV-damage. Earlier studies suggested a role for PCNA in regulating DNA replication and repair in response to DNA damage (7,8), such that UV-induced cell cycle regulatory protein, p21 Cip1/Waf1 , interacts with PCNA, inhibiting PCNA's function in DNA replication (11)(12)(13) but not in repair (12). It will be interesting to see whether both RPA and PCNA function in differential regulation in response to UV irradiation. RPA and PCNA may have distinctive roles for S-phase arrest and/or repair in response to UV damage.