Cellular stressors are known to inhibit the p53–RPA70 (replication protein A, 70 kDa subunit) complex, and RPA70 increases cellular DNA repair in cancer cells. We hypothesized that regulation of RPA70-mediated DNA repair might be responsible for the inhibition of apoptosis in hypoxic tumours. We have shown that, in cancer cells, hypoxia disrupts the p53–RPA70 complex, thereby enhancing RPA70-mediated NER (nucleotide excision repair)/NHEJ (non-homologous end-joining) repair. In normal cells, RPA70 binds to the p53-NTD (N-terminal domain), whereas this binding is disrupted in hypoxia. Phosphorylation of p53-NTD is a crucial event in dissociating both NTD–RPA70 and p53–RPA70 complexes. Serial mutations at serine and threonine residues in the NTD confirm that p53Ser15 phosphorylation induces dissociation of the p53–RPA70 complex in hypoxia. DNA-PK (DNA-dependent protein kinase) is shown to induce p53Ser15 phosphorylation, thus enhancing RPA70-mediated NER/NHEJ repair. Furthermore, RPA70 gene silencing induces significant increases in cellular apoptosis in the resistant hypoxic cancer cells. We have thus elucidated a novel pathway showing how DNA-PK-mediated p53Ser15 phosphorylation dissociates the p53–RPA70 complex, thus enhancing NER/NHEJ repair, which causes resistance to apoptosis in hypoxic cancer cells. This novel finding may open new strategies in developing cancer therapeutics on the basis of the regulation of RPA70-mediated NER/NHEJ repair.

INTRODUCTION

RPA (replication protein A), the eukaryote ssDNA (single-stranded DNA)-binding protein, is a heterotrimeric protein composed of three subunits, RPA70, RPA32 and RPA14 [1]. Along with ssDNA formation, RPA is required for cellular DNA metabolism processes, such as replication, recombination, checkpoints and repair. RPA70 interacts with genomic DNA and numerous other proteins involved in these processes [2,3], and is responsible for NER (nucleotide excision repair) and NHEJ (non-homologous end-joining) pathways of DNA repair [4,5]. It also interacts with BRCA1 (breast cancer early-onset 1) and BRCA2, two probable recombination mediators, as well as with p53 [6,7].

The p53 tumour suppressor co-ordinates a cellular response by transcriptional regulation of genes involved in cell-cycle arrest and apoptosis upon sensing DNA damage [8]. p53 is central to an extensive network of DNA-damage sensing, protein–protein and protein–nucleic acid interactions; RPA70 has been shown to interact with p53 under in vitro and in vivo conditions [9]. The interaction of p53 with RPA70 mediates suppression of homologous recombination [10], and in co-ordinating DNA repair through p53-dependent checkpoint control, by sensing UV damage [11]. The interaction between p53 and RPA70 inhibits p53's ability to bind sequence-specific DNA [12,13] and UV exposure greatly reduces the ability of RPA70 to bind to p53. An NMR study confirmed that p53-NTD (amino acids 37–57; NTD is N-terminal domain) binds to residues 1–120 of RPA70 [6].

Hypoxia-mediated dysregulation of critical DNA repair pathways contributes to genetic instability and tumour progression in cancer cells [14] and is a critical factor limiting the efficacy of anticancer strategies. Within solid tumours, hypoxia-induced chemo-resistance is originally attributed to poor drug distributions and to the contention that hypoxic tumour cells are predominantly quiescent [15]; however, the underlying molecular mechanism of hypoxia-induced drug resistance remains unclear. Recent studies have shown the association of hypoxic tumours with increased forms of DNA damage, including DNA strand breaks and oxidative base damage, such as 8-oxoguanine and thymine glycols [16,17]. Hypoxia has been recognized to induce several DNA-damage-response factors, such as DNA-PK (DNA-dependent protein kinase), ATM (ataxia telangiectasia mutated)/ATR (ATM- and Rad3-related), CHKl (checkpoint kinase 1)/CHK2 and BRCA1, and regulates the DNA repair [18]. NER and NHEJ are important DNA repair pathways that are responsible for the removal of helix-distorting DNA adducts, including UV-induced cyclobutane pyrimidine dimers and DNA DSBs (double-strand breaks). The hypoxia effector HIF (hypoxia-inducible factor)-1α transcriptionally regulates the expression of two NER proteins, XPC (xeroderma pigmentosum complementation group C) and XPD, after binding to HREs (hypoxia-responsive elements) [19], and the DNA repair pathway is also involved in DNA DSB repair through the targeting of HIF-1α [20]. Although hypoxia-induced DNA repair is linked to chemoresistance, little is known about the role of hypoxia on NER and NHEJ; however, both of these pathways have been implicated in hypoxia-mediated chemoresistance [21].

In the present study we have demonstrated that hypoxic cells exhibit high RPA70-mediated DNA repair through NER and NHEJ pathways. Hypoxia increased RPA70 expression and induced dissociation of the p53–RPA70 complex, which is due to DNA-PK-mediated phosphorylation of p53Ser15. Furthermore, RPA70 silencing as well as inhibition of p53Ser15 phosphorylation induced apoptosis in the resistant hypoxic cancer cells.

MATERIALS AND METHODS

Chemicals and antibodies

DMEM (Dulbecco's modified Eagle's medium), penicillin, streptomycin, FBS (fetal bovine serum), trypsin/EDTA and other chemicals of culture grade were purchased from Gibco Life Sciences. Cisplatin was purchased from Sigma. Annexin-V FITC detection kit was obtained from Becton and Dickinson. Antibodies against RPA70, p53, XPC, RAD23B, XPA, XRCC5 (X-ray cross-complementation group 5) (Ku86), XPB [TFIIH (transcription factor IIH)], CDK7 (cyclin-dependent kinase 7), Ku80, PRKDC (protein kinase, DNA-activated, catalytic polypeptide) (DNA-PKCS), NHEJ1 (XLF) and β-actin (anti-mouse IgG) were procured from Santa Cruz Biotechnology. Rabbit primary antibodies against CETN2 (centrin EF-hand protein 2) and TFIIH were obtained from Santa Cruz Biotechnology. A mouse polyclonal primary antibody against XRCC4 and a rabbit primary antibody against Ku70 was procured from Abcam.

Cell lines and culture conditions

MCF-7 and H1299 cells were obtained from the National Centre for Cell Sciences (Pune, India). The cells were cultured as monolayers in DMEM supplemented with 10% (v/v) heat-inactivated FBS and antibiotics (antimycotic mix: 10000 units/ml penicillin G, 10000 μg/ml streptomycin sulfate and 25 μg/ml amphoterecin B), and incubated at 37°C in a humidified atmosphere of 95% air and 5% CO2.

Transfections

Cells were split 2 days before transfection at a density of 5×105 cells per plate. Plasmids or siRNA (small interfering RNA) were transfected in cells using the ESCORT IV kit (Sigma). The cell line containing single rearranged copies of the reporter cassettes [22] were transfected in MCF-7 cells to study the NHEJ repair efficiency.

NHEJ and NER assay

Construction of cell reporter lines containing cassettes for analysis of the efficiency of NHEJ

The linearized reporter cassettes (0.5 μg) were transfected into cancer cells. At 1 day after transfection, cells were placed on selection medium with G418 at 1 mg/ml. G418-resistant colonies were picked 7–10 days later. The clones were screened by Southern blotting for intact reporter cassettes and the number of integrated copies was analysed. Single-copy integrants were selected for further study.

NER assay

As described previously [23], NER was measured in damaged plasmid DNA using fractionated mammalian cell extracts. The method involves use of the NER property whereby it creates a single-stranded gap of approximately 25–30 nt; filling of this gap by repair synthesis can be monitored by the incorporation of radioactive nucleotides.

RESULTS AND DISCUSSION

Hypoxic cancer cells show increased DNA repair and inhibition of apoptosis

DNA repair protects cancer cells from apoptosis by reducing the accumulation of genomic insult [24]. Hypoxic cancer cells are resistant to drug-mediated apoptosis [25] and NER/NHEJ are known to play important roles in the repair of hypoxia-induced DNA anomalies. DNA repair pathways are shown to regulate hypoxia-induced genetic instability within the tumour [26,27], thus making tumours resistant, aggressive and metastatic [28]. We have recently shown through EPR spectroscopy that the oxygen concentration is 1.8% in the hypoxic core of the MCF-7 tumour [29]. To establish the relationship between hypoxia-mediated DNA repair and inhibition of apoptosis, we analysed the status of cellular NER and NHEJ in MCF-7 cells that were maintained at 1.8% oxygen. Cisplatin and UV, which are known to induce NER [30,31] and NHEJ [32,33] were used as the positive controls. NER was analysed by evaluation of the removal of UV-induced lesions from UV-irradiated plasmid (substrate) as described previously [23]. The relative percentage of NER repair was calculated for normoxic and hypoxic MCF-7 cells that were treated with UV (25 J/cm2) and cisplatin (20 μM) (see the Supplementary Materials and methods section at http://www.BiochemJ.org/bj/443/bj4430811add.htm) (Figure 1A). In normoxic MCF-7 cells, NER was 20% (baseline); UV and cisplatin increased NER to 49% and 41% respectively. Hypoxia exposure alone increased NER to 48%, whereas UV and cisplatin significantly increased NER to 58% and 54% respectively in hypoxic cells. Antibodies against XPA were used as a negative control to block NER [34]. NHEJ was analysed using an assay based on fluorescent detection of repaired products [22] (Figure 1B). In comparison with normoxic MCF-7 cells, UV- and cisplatin-treated cells showed 2.89- and 2.36-fold increases in NHEJ respectively. Interestingly, hypoxia induced 2.1-fold higher NHEJ than in normoxic MCF-7 cells. Furthermore, both UV and cisplatin treatment induced >3-fold increase in NHEJ in hypoxic cells in comparison with the UV- and cisplatin-treated normoxic cells. The anti-Ku70 antibody was used as a negative control to block NHEJ [35]. These results established high levels of NER and NHEJ in hypoxic cancer cells. Since hypoxia up-regulated NER and NHEJ, the cellular level of proteins involved in regulation of these DNA repair pathways were further analysed. Western blots were conducted to study the expression of the proteins involved in NER (XPC, RAD23B, CETN2, XPA, XRCC5, XPB, XPB, GF2H1 and CDK7) and NHEJ [ERCC4 (excision repair cross-complementing rodent repair deficiency complementation group 4), Ku70, Ku80, PRKDC and NHEJ1] in normoxic and hypoxic MCF-7 cells (Figure 1C). The results showed a significant increase in expression of these proteins under hypoxic conditions. The cDNA expression vectors of these proteins were used as negative and positive controls respectively. Although the expression of key proteins such as Ku70, Ku80 and TFIIH were shown to be increased in hypoxic cells previously [36,37], the expression of other DNA repair proteins was reported to be unaltered in hypoxic cancer cells [38].

Hypoxic cancer cells show high NER/NHEJ repair and chemoresistance

Figure 1
Hypoxic cancer cells show high NER/NHEJ repair and chemoresistance

(A) Nucleotide excision repair (NER) was analysed in UV (25 J/cm2)- and cisplatin (20 μM)-treated normoxic and hypoxic (1.8% O2) MCF-7 cells. Results show that NER-mediated DNA repair is 2.4-fold higher in hypoxic MCF-7 cells (dark grey bar) in comparison with normoxic cells (light grey bar). UV- and cisplatin-treated hypoxic MCF-7 cells also show a significant increase in NER. The antibody against XPA was used as a negative control to block NER [31]. (B) DNA repair through NHEJ was observed in UV (25 J/cm2)- and cisplatin (20 μM)- treated normoxic and hypoxic (1.8% O2) MCF-7 cells, as described previously [22]. NHEJ was 2.1-fold higher in hypoxic (dark grey bar) than in normoxic (light grey bar) MCF-7 cells. The anti-Ku70 antibody was used as a negative control to block NHEJ. For (A) and (B), n=10, error bars represent S.D. and significance was measured using ANOVA (A, *P<0.038; B, *P<0.036). (C) The effect of hypoxia on the cellular expression of key proteins involved in NER and NHEJ machinery was observed. Western blots show a significant increase in the expression of NER/NHEJ proteins in hypoxic MCF-7 cells (cDNA of these proteins was used as a positive control) (n=5, significance was measured using ANOVA). (D) Hypoxia-mediated chemoresistance was observed in UV- and cisplatin-treated hypoxic MCF-7 cells using annexin V staining and flow cytometry. UV and cisplatin induced 36% and 60% apoptosis, which was reduced to 12% and 16% respectively under hypoxia (grey bars). (E) The role of NER/NHEJ-mediated DNA repair in hypoxia-mediated chemoresistance was analysed. NER and NHEJ are inhibited in hypoxic MCF-7 cells by incubating the cells with XPA/XPB and Ku70/Ku80 antibodies. NER inhibition induced 48% and NHEJ inhibition induced 56% apoptosis in cisplatin-treated hypoxic MCF-7 cells (grey bars) (cisplatin induced 16% apoptosis in hypoxic MCF-7 cells). For (D) and (E), n=10, error bars indicate S.D., significance was measured using ANOVA (D, *P<0.029; E, *P<0.032).

Figure 1
Hypoxic cancer cells show high NER/NHEJ repair and chemoresistance

(A) Nucleotide excision repair (NER) was analysed in UV (25 J/cm2)- and cisplatin (20 μM)-treated normoxic and hypoxic (1.8% O2) MCF-7 cells. Results show that NER-mediated DNA repair is 2.4-fold higher in hypoxic MCF-7 cells (dark grey bar) in comparison with normoxic cells (light grey bar). UV- and cisplatin-treated hypoxic MCF-7 cells also show a significant increase in NER. The antibody against XPA was used as a negative control to block NER [31]. (B) DNA repair through NHEJ was observed in UV (25 J/cm2)- and cisplatin (20 μM)- treated normoxic and hypoxic (1.8% O2) MCF-7 cells, as described previously [22]. NHEJ was 2.1-fold higher in hypoxic (dark grey bar) than in normoxic (light grey bar) MCF-7 cells. The anti-Ku70 antibody was used as a negative control to block NHEJ. For (A) and (B), n=10, error bars represent S.D. and significance was measured using ANOVA (A, *P<0.038; B, *P<0.036). (C) The effect of hypoxia on the cellular expression of key proteins involved in NER and NHEJ machinery was observed. Western blots show a significant increase in the expression of NER/NHEJ proteins in hypoxic MCF-7 cells (cDNA of these proteins was used as a positive control) (n=5, significance was measured using ANOVA). (D) Hypoxia-mediated chemoresistance was observed in UV- and cisplatin-treated hypoxic MCF-7 cells using annexin V staining and flow cytometry. UV and cisplatin induced 36% and 60% apoptosis, which was reduced to 12% and 16% respectively under hypoxia (grey bars). (E) The role of NER/NHEJ-mediated DNA repair in hypoxia-mediated chemoresistance was analysed. NER and NHEJ are inhibited in hypoxic MCF-7 cells by incubating the cells with XPA/XPB and Ku70/Ku80 antibodies. NER inhibition induced 48% and NHEJ inhibition induced 56% apoptosis in cisplatin-treated hypoxic MCF-7 cells (grey bars) (cisplatin induced 16% apoptosis in hypoxic MCF-7 cells). For (D) and (E), n=10, error bars indicate S.D., significance was measured using ANOVA (D, *P<0.029; E, *P<0.032).

We then determined the role of NER and NHEJ in hypoxia-mediated chemoresistance in UV- and cisplatin-treated hypoxic MCF-7 cells using annexin V staining and flow cytometry (Figure 1D). In normoxic MCF-7 cells, UV and cisplatin induced 36% and 60% apoptosis in MCF-7 cells, whereas in hypoxic cells it was 12% and 16% respectively, suggesting that hypoxic MCF-7 cells are resistant to UV- and cisplatin-induced apoptosis. In order to analyse the role of DNA repair in hypoxia-induced resistance to apoptosis, cellular NER and NHEJ were inhibited by incubating hypoxic cells with antibodies against the proteins that are involved in DNA repair pathways. NER was inhibited using antibodies against XPC, XPA, ERCC5, XPB and XPD; NHEJ was inhibited using antibodies against Ku70 and Ku80. Annexin V staining showed that inhibition of NER and NHEJ led to a significant increase in UV- and cisplatin-induced apoptosis in hypoxic MCF-7 cells (Figure 1E). Besides, simultaneous inhibition of NER and NHEJ showed a more than 5-fold increase in UV- and cisplatin-induced apoptosis in hypoxic MCF-7 cells (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). These results established that NER and NHEJ are increased during hypoxia and their inhibition induces apoptosis in hypoxic/resistant cells.

RPA70 mediates DNA repair in hypoxic cancer cells

Hypoxia induces a variety of genetic alterations, including activation of oncogenes and inactivation of tumour-suppressor genes. RPA70 plays a major role in the NER and NHEJ [39] pathways of DNA repair, but the cellular mechanism of these pathways in hypoxic cancer cells is unknown. Since levels of NER and NHEJ were increased in hypoxic MCF-7 cells, RPA70 mRNA and protein levels were analysed. A 6-fold increase in RPA70 mRNA (Figure 2A) and a 4.8-fold increase in RPA70 protein level were observed in hypoxic MCF-7 cells (Figure 2B); RPA70 siRNA and RPA70 cDNA were used as negative and positive controls. Furthermore, the role of RPA70 in NER/NHEJ was established by analysis of NER/NHEJ-mediated DNA repair in hypoxic MCF-7 cells after RPA70 gene silencing (Figure 2C, panels i and ii, grey bar). RPA70 silencing reduced NER by 69% (compare with Figure 1A) and NHEJ by 73% (compare with Figure 1B). To establish whether hypoxia- and RPA70-induced DNA repair is linked to chemoresistance, RPA70 gene silencing was performed in UV- and cisplatin-treated hypoxic MCF-7 cells (Figure 2D). Annexin V staining showed 52% and 84% apoptosis in UV- and cisplatin-treated hypoxic MCF-7 cells. The apoptosis induced by RPA70 silencing and by inhibition of NER/NHEJ (compare with Figure 1F) was similar, suggesting that a hypoxia-mediated increase in DNA repair may be responsible for chemoresistance in hypoxic cancer cells. To study the role of RPA70, we studied the expression of proteins involved in NER/NHEJ by Western blotting. RPA70 siRNA significantly decreased the expression of proteins involved in NER/NHEJ (Figure 2E). Irradiation with UV led to a substantial increase in the expression of NER/NHEJ member proteins, but RPA70 siRNA abolished the UV-induced increase in protein expression. The results confirmed the role of RPA70 in hypoxia-mediated DNA repair.

Hypoxia-mediated increase in NER/NHEJ is RPA70-dependent

Figure 2
Hypoxia-mediated increase in NER/NHEJ is RPA70-dependent

(A) RPA70 mRNA level was analysed in UV- and cisplatin-treated normoxic/hypoxic MCF-7 cells, using RT (reverse transcription)–PCR (see the Supplementary Materials and methods section and Supplementary Table S1 at http://www.BiochemJ.org/bj/443/bj4430811add.htm for more details). Hypoxia induced a 6-fold increase in RPA-70 mRNA level; UV and cisplatin treatment also significantly increased RPA70 mRNA expression. RPA70 siRNA was used as a negative control. n=5. (B) RPA70 protein level was analysed using Western blotting. The results show an increase in RPA70 protein in hypoxic MCF-7 cells, and UV and cisplatin increase the expression further. n=5. (C) NER (i) and NHEJ (ii) were observed in hypoxic MCF-7 cells when the RPA70 gene was silenced using RPA70 siRNA. RPA70 silencing reduced NER to 16% from 48% (compare with Figure 1A); NHEJ was reduced 2.85-fold (grey bars). n=10, error bars indicate S.D., significance was measured using ANOVA (*P<0.039), suggesting that hypoxia-mediated NER/NHEJ are RPA-70 dependent. (D) The effect of RPA-70 on hypoxia-mediated chemoresistance was observed by silencing RPA-70 in UV/cisplatin-treated hypoxic MCF-7 cells. Annexin V staining shows a significant increase in the apoptotic fraction from 12% and 16% (in hypoxic cells) to 52% and 84% (in hypoxic cells with RPA silencing) respectively in RPA-70 silenced MCF-7 cells (n=7). *P<0.042. (E) The effect of RPA70 on cellular expression of key NER and NHEJ protein was analysed in hypoxic MCF-7 cells. RPA70 siRNA reduced the expression of NER/NHEJ proteins. UV-induced overexpression of NER/NHEJ proteins was reversed upon RPA70 silencing; cDNA of the NER/NHEJ proteins was used as positive controls (n=4). *P<0.046.

Figure 2
Hypoxia-mediated increase in NER/NHEJ is RPA70-dependent

(A) RPA70 mRNA level was analysed in UV- and cisplatin-treated normoxic/hypoxic MCF-7 cells, using RT (reverse transcription)–PCR (see the Supplementary Materials and methods section and Supplementary Table S1 at http://www.BiochemJ.org/bj/443/bj4430811add.htm for more details). Hypoxia induced a 6-fold increase in RPA-70 mRNA level; UV and cisplatin treatment also significantly increased RPA70 mRNA expression. RPA70 siRNA was used as a negative control. n=5. (B) RPA70 protein level was analysed using Western blotting. The results show an increase in RPA70 protein in hypoxic MCF-7 cells, and UV and cisplatin increase the expression further. n=5. (C) NER (i) and NHEJ (ii) were observed in hypoxic MCF-7 cells when the RPA70 gene was silenced using RPA70 siRNA. RPA70 silencing reduced NER to 16% from 48% (compare with Figure 1A); NHEJ was reduced 2.85-fold (grey bars). n=10, error bars indicate S.D., significance was measured using ANOVA (*P<0.039), suggesting that hypoxia-mediated NER/NHEJ are RPA-70 dependent. (D) The effect of RPA-70 on hypoxia-mediated chemoresistance was observed by silencing RPA-70 in UV/cisplatin-treated hypoxic MCF-7 cells. Annexin V staining shows a significant increase in the apoptotic fraction from 12% and 16% (in hypoxic cells) to 52% and 84% (in hypoxic cells with RPA silencing) respectively in RPA-70 silenced MCF-7 cells (n=7). *P<0.042. (E) The effect of RPA70 on cellular expression of key NER and NHEJ protein was analysed in hypoxic MCF-7 cells. RPA70 siRNA reduced the expression of NER/NHEJ proteins. UV-induced overexpression of NER/NHEJ proteins was reversed upon RPA70 silencing; cDNA of the NER/NHEJ proteins was used as positive controls (n=4). *P<0.046.

Hypoxia dissociates the p53–RPA70 complex

DNA-damage-induced activation of p53 is always accompanied by the release of the heterotrimeric RPA protein [9]. However, the mechanism of hypoxia-mediated RPA70 activation is unknown. As the p53–RPA70 complex is shown to exist in cancer cells [6], and this complex dissociates upon UV exposure [9], we asked if such a complex is present in hypoxic cancer cells. IP (immunoprecipitation) using anti-p53 and anti-RPA70 antibodies showed that the p53–RPA70 complex is intact in normoxic MCF-7 cells (Figure 3A); H1299 cells (p53−/−) were used as a negative control. Furthermore, RPA70 is known to bind to p53-NTD under in vitro conditions, as it mimics the ssDNA strand [12], which has also been confirmed by NMR studies [6]. To check whether such interactions occur in cells, p53 and p53 NTD cDNA along with RPA70 cDNA constructs were co-transfected in both MCF-7 and H1299 cells, and co-IP confirmed their interaction. Exogenous addition of the NTD led to disruption of the p53–RPA70 complex and formation of the NTD–RPA70 complex, suggesting selective binding of NTD towards RPA70 (Figure 3B). IP with an anti-HA (haemagglutinin) antibody showed that HA-tagged NTD binds to both p53 and RPA70 (Figure 3B). Interaction of the NTD with RPA70 was also confirmed in H1299 cells that were transfected with NTD cDNA (Figure 3). These results confirmed that p53 interacts with RPA70 through its N-terminus in cancer cells.

Hypoxia induces dissociation of the p53–RPA70 complex

Figure 3
Hypoxia induces dissociation of the p53–RPA70 complex

(A) p53–RPA70 binding was analysed by IP in normoxic MCF-7 (p53+/+) and H1299 (p53−/−) cells with anti-p53 and anti-RPA70 antibodies showing binding between p53 and RPA70. RPA70 and p53 siRNAs were used as controls. (B) The p53 NTD and RPA70 interaction was analysed in MCF-7 cells transfected with NTD cDNA and H1299 cells transfected with p53 and NTD cDNA. IP using anti-RPA70 antibody shows that NTD displaces p53 from the p53–RPA70 complex and forms a strong NTD–RPA70 complex in MCF-7 cells. IP using anti-HA antibody shows that the NTD binds to both p53 and RPA70 as separate complexes. The NTD–RPA70 interaction was also observed in H1299 cells. (C) In the presence of cisplatin, the p53–RPA70 complex was absent in MCF-7 cells. (D) The effect of cisplatin on the stability of the NTD–RPA70 complex was analysed in MCF-7 and H1299 cells. In MCF-7 cells, IP with anti-HA, anti-p53 and anti-RPA70 antibodies did not show binding between p53 and RPA70 or NTD and RPA70. In H1299 cells, IP with anti-RPA70 and anti-HA antibodies showed no binding between NTD and RPA70. (E) The effect of hypoxia on p53 and RPA70 binding was analysed in MCF-7 cells. p53 and RPA70 show binding in normoxic MCF-7 cells; however, in hypoxic MCF-7 cells, IP with anti-p53 and anti-RPA70 antibodies showed no binding between p53 and RPA70. For all experiments, n=5. C-ter, C-terminal; N-ter, N-terminal; Nor, normoxia.

Figure 3
Hypoxia induces dissociation of the p53–RPA70 complex

(A) p53–RPA70 binding was analysed by IP in normoxic MCF-7 (p53+/+) and H1299 (p53−/−) cells with anti-p53 and anti-RPA70 antibodies showing binding between p53 and RPA70. RPA70 and p53 siRNAs were used as controls. (B) The p53 NTD and RPA70 interaction was analysed in MCF-7 cells transfected with NTD cDNA and H1299 cells transfected with p53 and NTD cDNA. IP using anti-RPA70 antibody shows that NTD displaces p53 from the p53–RPA70 complex and forms a strong NTD–RPA70 complex in MCF-7 cells. IP using anti-HA antibody shows that the NTD binds to both p53 and RPA70 as separate complexes. The NTD–RPA70 interaction was also observed in H1299 cells. (C) In the presence of cisplatin, the p53–RPA70 complex was absent in MCF-7 cells. (D) The effect of cisplatin on the stability of the NTD–RPA70 complex was analysed in MCF-7 and H1299 cells. In MCF-7 cells, IP with anti-HA, anti-p53 and anti-RPA70 antibodies did not show binding between p53 and RPA70 or NTD and RPA70. In H1299 cells, IP with anti-RPA70 and anti-HA antibodies showed no binding between NTD and RPA70. (E) The effect of hypoxia on p53 and RPA70 binding was analysed in MCF-7 cells. p53 and RPA70 show binding in normoxic MCF-7 cells; however, in hypoxic MCF-7 cells, IP with anti-p53 and anti-RPA70 antibodies showed no binding between p53 and RPA70. For all experiments, n=5. C-ter, C-terminal; N-ter, N-terminal; Nor, normoxia.

The influence of genotoxic/cellular stresses such as cisplatin or hypoxia on the stability of the p53–RPA70 complex was then analysed. IP showed that, in cisplatin-treated MCF-7 cells, the p53–RPA70 complex is also disrupted (Figure 3C). Cisplatin also disrupted the NTD–RPA70 complex both in MCF-7 and H1299 cells (Figure 3D). Interestingly, IP using anti-p53 and anti-RPA70 antibodies showed that the p53–RPA70 complex is disrupted in hypoxic MCF-7 cells (Figure 3E). In a similar manner, hypoxia disrupted the NTD–RPA70 complex in MCF-7 and H1299 cells (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/443/bj4430811add.htm), which might be due to p53-NTD post-translational modifications.

p53Ser15 phosphorylation is responsible for disruption of the p53–RPA70 complex

The dissociation of the p53–RPA70 complex in hypoxia led us to hypothesize that p53 post-translational modifications might be responsible for this. The p53 NTD (amino acids 1–126) [40] has no known site for acetylation; however, it contains nine serine and threonine residues which undergo phosphorylation [41]. The post-translational phosphorylation of p53 and NTD were analysed in UV-, cisplatin- and hypoxia-treated MCF-7 and H1299 cells. IP using anti-phospho-p53 antibodies showed that both p53 and NTD are phosphorylated upon UV, cisplatin and hypoxia treatment (Figure 4A). To establish the role of p53 phosphorylation, the p53–RPA70 and NTD–RPA70 complexes were analysed in the presence of serine/threonine kinase inhibitors. Serine/threonine kinase inhibitors abolished the disruption of both the p53–RPA70 and NTD–RPA70 complexes in UV-, cisplatin- and hypoxia-treated MCF-7 (Figure 4B, panel i) and H1299 (Figure 4B, panel ii) cells. The efficiency of serine/threonine kinase inhibitors in inhibition of p53 phosphorylation was tested as a control (Supplementary Figure S3 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). These results suggest that p53-NTD phosphorylation might regulate the dissociation of the p53–RPA70 complex. We then sought to find the residues in p53-NTD whose phosphorylation was crucial for the disruption of the p53–RPA70 complex. UV- and hypoxia-treated H1299 cells that were transfected with p53 cDNA carrying mutations at Ser6, Ser9, Ser15, Ser20, Ser33 and Ser37 and Thr18, Thr55 and Thr81 residues [42] showed that these mutations were unable to dissociate p53 from RPA70 (Figure 4C). This data suggests that phosphorylation of either one or a group of these residues were responsible for the regulation of the p53–RPA70 complex. In order to identify those key amino acid residues, p53 cDNA constructs coding for the point mutations at Ser6, Ser9, Ser15, Ser20, Ser33 and Ser37 and Thr18, Thr55 and Thr81 residues were used [43], and the p53–RPA70 complex was analysed by IP (Figure 4D). H1299 cells transfected exclusively with the p53Ser15 mutant showed stabilization of the p53–RPA70 complex in the presence of UV and hypoxia (Figure 4D, top panel). p53 cDNA constructs carrying point mutations at Ser6, Ser9, Ser20, Ser33, Ser37, Thr18, Thr55 and Thr81 were unable to inhibit the disruption of the p53–RPA70 complex. These results established that p53Ser15 was responsible for the regulation of the p53–RPA70 protein complex. Inhibition of Ser15 phosphorylation using serine/threonine kinase inhibitors or by mutating the p53Ser15 residue prevented the UV-, cisplatin- and hypoxia-mediated disruption of the p53–RPA70 complex. The UV- and hypoxia-treated H1299 cells transfected with wild-type p53 cDNA were used as a control (Figure 4D, top panel). These results convincingly established that the phosphorylation of p53 at Ser15 disrupted p53–RPA interaction.

p53Ser15 phosphorylation controls dissociation of the p53–RPA70 complex

Figure 4
p53Ser15 phosphorylation controls dissociation of the p53–RPA70 complex

(A) IP with anti-p53 and anti-phospho-p53 (phos-p53) antibodies shows the absence of p53 phosphorylation in normoxic MCF-7 cells. UV and cisplatin significantly induced p53 phosphorylation. Hypoxia exposure induces minimal phosphorylation of p53 protein; UV and cisplatin treatment induce p53 phosphorylation in hypoxic cells. Hypoxic (1.8% O2) H1299 cells were transfected with NTD cDNA; IP confirms that hypoxia induces minimal phosphorylation of NTD in the presence of UV and cisplatin (n=7). (B) The effect of serine/threonine kinase inhibitors on the stability of p53–RPA70 and NTD–RPA70 complexes was analysed in (i) MCF-7 and (ii) H1299 cells (transfected with NTD cDNA). p53 and RPA-70 were present in a complex in untreated cells. UV and hypoxia disrupted the p53–RPA70 complex. Addition of serine/threonine kinase inhibitors to UV, cisplatin and hypoxia-treated cells stabilized the p53–RPA70 complex. Similar results were observed for NTD–RPA70 binding in NTD-transfected H1299 cells (ii) (n=6). (C) H1299 cells were transfected with p53 cDNA construct carrying mutations at serine and threonine residues of p53 NTD; p53 mutated at Ser6, Ser9, Ser15, Ser20, Ser33, Ser37 and Thr18, Thr55 and Thr81 residues are unable to disrupt the p53–RPA70 complex (n=8). (D) To identify the phosphorylated residues, H1299 cells were transfected with a series of mutant p53 cDNA constructs which code for S6A, S9A, S15A, S20A, S33A, S37A, T18A, T55A and T81A. p53–RPA70 complex formation was analysed in each case. UV and hypoxia were used as signals to induce dissociation of the p53–RPA70 complex. The results show that transfection of the p53-Ser15 mutant abolished the hypoxia- and UV-induced dissociation of the p53–RPA70 complex. No other p53 mutant was able to show any effect on hypoxia- or UV-induced disruption of the p53–RPA70 complex, suggesting that p53Ser15 phosphorylation was mandatory for hypoxia- or UV-induced dissociation of the p53–RPA70 complex (n=5). *P<0.02. WT, wild-type; MT, mutant.

Figure 4
p53Ser15 phosphorylation controls dissociation of the p53–RPA70 complex

(A) IP with anti-p53 and anti-phospho-p53 (phos-p53) antibodies shows the absence of p53 phosphorylation in normoxic MCF-7 cells. UV and cisplatin significantly induced p53 phosphorylation. Hypoxia exposure induces minimal phosphorylation of p53 protein; UV and cisplatin treatment induce p53 phosphorylation in hypoxic cells. Hypoxic (1.8% O2) H1299 cells were transfected with NTD cDNA; IP confirms that hypoxia induces minimal phosphorylation of NTD in the presence of UV and cisplatin (n=7). (B) The effect of serine/threonine kinase inhibitors on the stability of p53–RPA70 and NTD–RPA70 complexes was analysed in (i) MCF-7 and (ii) H1299 cells (transfected with NTD cDNA). p53 and RPA-70 were present in a complex in untreated cells. UV and hypoxia disrupted the p53–RPA70 complex. Addition of serine/threonine kinase inhibitors to UV, cisplatin and hypoxia-treated cells stabilized the p53–RPA70 complex. Similar results were observed for NTD–RPA70 binding in NTD-transfected H1299 cells (ii) (n=6). (C) H1299 cells were transfected with p53 cDNA construct carrying mutations at serine and threonine residues of p53 NTD; p53 mutated at Ser6, Ser9, Ser15, Ser20, Ser33, Ser37 and Thr18, Thr55 and Thr81 residues are unable to disrupt the p53–RPA70 complex (n=8). (D) To identify the phosphorylated residues, H1299 cells were transfected with a series of mutant p53 cDNA constructs which code for S6A, S9A, S15A, S20A, S33A, S37A, T18A, T55A and T81A. p53–RPA70 complex formation was analysed in each case. UV and hypoxia were used as signals to induce dissociation of the p53–RPA70 complex. The results show that transfection of the p53-Ser15 mutant abolished the hypoxia- and UV-induced dissociation of the p53–RPA70 complex. No other p53 mutant was able to show any effect on hypoxia- or UV-induced disruption of the p53–RPA70 complex, suggesting that p53Ser15 phosphorylation was mandatory for hypoxia- or UV-induced dissociation of the p53–RPA70 complex (n=5). *P<0.02. WT, wild-type; MT, mutant.

DNA-PK phosphorylates p53Ser15 in hypoxia

DNA-PK is known to induce p53Ser15 phosphorylation under in vitro conditions [44,45], and ATR kinase is also activated in an RPA-dependent manner [46]. We thus asked whether DNA-PK could phosphorylate p53 under hypoxia. The DNA-PK activity was increased in hypoxic MCF-7 cells (Figure 5A), and IP (using anti-DNA-PK antibodies) showed that there was cellular interaction between p53 and DNA-PK (Figure 5B). The observed interaction was further disrupted by DNA-PK siRNA with a decrease in total p53 phosphorylation as well as p53Ser15 phosphorylation in UV-treated normoxic MCF-7 cells (Figure 5C). The fraction of total phosphorylated p53 as well as p53 phosphorylated at the Ser15 residue was low in hypoxic MCF-7 cells (Figure 5C), whereas DNA-PK silencing abolished the total p53 phosphorylation and p53Ser15 phosphorylation in hypoxic MCF-7 cells (Figure 5C). In addition, co-IP showed that silencing of DNA-PK abolished the disruption of the p53–RPA70 complex in hypoxic MCF-7 cells (Figure 5D) and caused an increase in the apoptotic fraction of cisplatin-treated hypoxic cells (69%) (Figure 5E). An interaction between BRCA2 and RAD51 was shown to be disrupted due to CDK-mediated phosphorylation of BRCA2 at its Ser3291 residue [47].

DNA-PK phosphorylates p53 at Ser15 in hypoxic cancer cells

Figure 5
DNA-PK phosphorylates p53 at Ser15 in hypoxic cancer cells

(A) The kinase activity of DNA-PK was analysed in hypoxic MCF-7 cells. Results showed a 3-fold increase in the DNA-PK activity in hypoxic MCF-7 cells (grey bar). DNA-PK siRNA was used as a control. Error bars indicate S.D., significance was calculated using ANOVA (*P<0.018), n=10. (B) The binding between DNA-PK and p53 was observed in UV-, cisplatin- and hypoxia-treated MCF-7 cells. IP using anti-DNA-PK antibody shows the p53–DNA-PK interaction in all three cases. Hypoxia induced a marked increase in this interaction (n=4). (C) The role of the p53–DNA-PK interaction upon total p53 as well as p53Ser15 phosphorylation was determined in UV- and hypoxia-treated MCF-7 cells; IP performed using anti-phospho-p53 and anti-phospho-p53Ser15 antibodies show that the level of the total phosphorylated p53 and p53 phosphorylated at the Ser15 residue was low in hypoxic MCF-7 cells. DNA-PK siRNA abolished total p53 phosphorylation in hypoxic MCF-7 cells, suggesting the role of DNA-PK in hypoxia-mediated p53 phosphorylation (n=4). (D) Effect of DNA-PK silencing on dissociation of the p53–RPA70 complex was analysed. IP shows inhibition of hypoxia-mediated dissociation of p53 and RPA70 binding (n=3). (E) The effect of DNA-PK gene silencing on the cell viability and apoptosis in the hypoxic MCF-7 cells was observed. Annexin V staining showed 12% and 69% apoptosis upon treatment with cisplatin (n=10). IB, immunoblot; phos p53, phosphorylated p53.

Figure 5
DNA-PK phosphorylates p53 at Ser15 in hypoxic cancer cells

(A) The kinase activity of DNA-PK was analysed in hypoxic MCF-7 cells. Results showed a 3-fold increase in the DNA-PK activity in hypoxic MCF-7 cells (grey bar). DNA-PK siRNA was used as a control. Error bars indicate S.D., significance was calculated using ANOVA (*P<0.018), n=10. (B) The binding between DNA-PK and p53 was observed in UV-, cisplatin- and hypoxia-treated MCF-7 cells. IP using anti-DNA-PK antibody shows the p53–DNA-PK interaction in all three cases. Hypoxia induced a marked increase in this interaction (n=4). (C) The role of the p53–DNA-PK interaction upon total p53 as well as p53Ser15 phosphorylation was determined in UV- and hypoxia-treated MCF-7 cells; IP performed using anti-phospho-p53 and anti-phospho-p53Ser15 antibodies show that the level of the total phosphorylated p53 and p53 phosphorylated at the Ser15 residue was low in hypoxic MCF-7 cells. DNA-PK siRNA abolished total p53 phosphorylation in hypoxic MCF-7 cells, suggesting the role of DNA-PK in hypoxia-mediated p53 phosphorylation (n=4). (D) Effect of DNA-PK silencing on dissociation of the p53–RPA70 complex was analysed. IP shows inhibition of hypoxia-mediated dissociation of p53 and RPA70 binding (n=3). (E) The effect of DNA-PK gene silencing on the cell viability and apoptosis in the hypoxic MCF-7 cells was observed. Annexin V staining showed 12% and 69% apoptosis upon treatment with cisplatin (n=10). IB, immunoblot; phos p53, phosphorylated p53.

Since ATR was previously shown to regulate p53 Ser15 phosphorylation during anoxia [48], we asked whether ATR kinase has any role in controlling p53 Ser15 phosphorylation during hypoxia (1.8% O2). ATR kinase is not linked to p53 Ser15 phosphorylation at 1.8% O2 concentration (see Supplementary Figure S4 at http://www.BiochemJ.org/bj/443/bj4430811add.htm), although it is responsible for inducing the phosphorylation of p53 at the Ser15 residue in anoxic (0% O2) cancer cells. ATR gene silencing did not affect the p53–RPA70 interaction in hypoxic MCF-7 cells (see Supplementary Figure S5 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). Furthermore, HIF-1 was found to increase RPA70 protein expression in hypoxic MCF-7 cells (see Supplementary Figure S6 at http://www.BiochemJ.org/bj/443/bj4430811add.htm) and HIF-1 gene silencing reduced RPA70 protein expression (Supplementary Figure S6).

Hypoxia itself is insufficient to induce DNA damage; however, it can induce genetic instability through resistance to apoptosis and decreased DNA repair in tumour tissue [16]. Once DNA damage signalling pathways are initiated in response to hypoxia, a number of kinases, including ATR and DNA-PK, are activated, leading to phosphorylation of p53 [48]. ATM, ATR and DNA-PK belong to the phosphoinositide 3-kinase-related kinase family and phosphorylate substrates, which are essential to transduce checkpoint signals to downstream effectors, including the p53 and the BRCA1 breast cancer tumour-susceptible protein [49].

Previously, the role of ATR in inducing the phosphorylation of H2AX in hypoxic (2% O2) and anoxic (0.02% O2) RKO cells was observed by Hammond et al. [48]. It was reported that histone H2AX was clearly phosphorylated by ATR in response to extreme hypoxia/anoxia, but remained unaffected at 2% oxygen. These finding suggest that, similar to p53, the histone H2AX might also be phosphorylated by another stress-activated kinase pathway in hypoxic cancer cells (2% O2). ATM and DNA-PK matched cell lines showed that both p53Ser15 and H2AX were phosphorylated in response to hypoxia; however, a deficiency in ATR had no effect on H2AX phosphorylation. These results support our observation in the present study that both DNA-PK and ATR are active and induce p53Ser15 phosphorylation in anoxic cancer cells; however, only DNA-PK is responsible for p53Ser15 phosphorylation in cells exposed to physiological levels of hypoxia (1.8%).

The increase in the DNA-PK kinase activity was shown further to be HIF-1-dependent, and HIF-1 siRNA significantly (50%) reduced the hypoxia-induced increase in DNK-PK kinase activity (see Supplementary Figure S7 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). HIF-1 also regulated the cellular expression of the DNA-PK protein level (see Supplementary Figure S8 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). The DNA-PK protein level increased in hypoxic cancer cells (Supplementary Figure S8), and this increase was HIF-1-dependent; HIF-1 gene silencing reduced the hypoxia-induced increase in the DNA-PK protein level (Supplementary Figure S8). HIF-1 gene silencing also reduced p53 Ser15 phosphorylation in hypoxic MCF-7 cells (see Supplementary Figure S9 at http://www.BiochemJ.org/bj/443/bj4430811add.htm). These results collectively suggest that the pathway of p53- and RPA70-mediated DNA repair in hypoxic cancer cells is HIF-dependent.

HIF-1 is the key regulator of the cellular response to oxygen deprivation. Under hypoxia, HIF-1 is stabilized, enters the nucleus and binds to HREs to transactivate a variety of hypoxia-responsive genes [50], therefore contributing to the adaptive response to hypoxic conditions. Um et al. [51] previously showed that hypoxia-induced accumulation of the transcription factor HIF-1 is reduced in DNA-PK-deficient cells. However, the authors did not show whether DNA-PK is activated under this stress condition. The results of the present study show that hypoxia induces DNA-PK kinase activity and HIF-1 regulates the cellular level of DNA-PK protein.

In general, high DNA repair corrects the spontaneous damage of genomic DNA and inhibits the generation of apoptotic signals [24]. Regulation of DNA repair pathways has been shown to regulate hypoxia-induced genetic instability within tumours [26,27]. In addition to making tumours resistant, aggressive and metastatic [28], hypoxia up-regulates DNA repair machinery [36,37]. The expression of key proteins such as Ku70, Ku80 and TFIIH are increased in hypoxic cells [36,37]; however, a contradictory report shows that expression of DNA repair proteins remains unaltered in hypoxic cells [38]. Thus the present study shows that hypoxia-induced chemoresistance might be due to high RPA70-mediated NER/NHEJ DNA repair. We propose a model to explain the mechanism of hypoxia-induced dissociation of the p53–RPA70 protein complex and its probable effect upon hypoxia-induced RPA70-mediated NER/NHEJ and chemoresistance (Figure 6). In conclusion, we have demonstrated a novel mechanism that shows how DNA-PK phosphorylates p53–NTD, thus disrupting the p53–RPA70 complex. This finding might explain the reason for induction of NER/NHEJ repair and chemoresistance in hypoxic cancer cells. Since RPA70 gene silencing induces significant apoptosis in the resistant hypoxic cancer cells, this study may be of importance in developing new cancer therapeutics based upon regulation of RPA70-mediated NER and NHEJ repair pathways.

A model of the dissociation of the p53–RPA70 complex in hypoxia, which may lead to chemoresistance

Figure 6
A model of the dissociation of the p53–RPA70 complex in hypoxia, which may lead to chemoresistance
Figure 6
A model of the dissociation of the p53–RPA70 complex in hypoxia, which may lead to chemoresistance

Abbreviations

     
  • ATM

    ataxia telangiectasia mutated

  •  
  • ATR

    ATM- and Rad3-related

  •  
  • BRCA

    breast cancer early-onset

  •  
  • CDK

    cyclin-dependent kinase

  •  
  • CETN2

    centrin EF-hand protein 2

  •  
  • CHK

    checkpoint kinase

  •  
  • DMEM

    Dulbecco's modified Eagle's medium

  •  
  • DNA-PK

    DNA-dependent protein kinase

  •  
  • DSB

    double-strand break

  •  
  • ERCC

    excision repair cross-complementing rodent repair deficiency complementation group

  •  
  • FBS

    fetal bovine serum

  •  
  • HA

    haemagglutinin

  •  
  • HIF

    hypoxia-inducible factor

  •  
  • HRE

    hypoxia-responsive elements

  •  
  • IP

    immunoprecipitation

  •  
  • NER

    nucleotide excision repair

  •  
  • NHEJ

    non-homologous end-joining

  •  
  • NTD

    N-terminal domain

  •  
  • PRKDC

    protein kinase, DNA-activated, catalytic polypeptide

  •  
  • RPA

    replication protein A

  •  
  • siRNA

    small interfering RNA

  •  
  • ssDNA

    single-stranded DNA

  •  
  • TFIIH

    transcription factor IIH

  •  
  • XP

    xeroderma pigmentosum complementation group

  •  
  • XRCC

    X-ray cross-complementation group

AUTHOR CONTRIBUTION

Esha Madan designed the study, performed the experiments and wrote the paper. Rajan Gogna designed the study and performed the experiments. Uttam Pati designed the study, analysed the results, provided research material and wrote the paper.

We thank Professor Abbas Ali Mahdi (Chhatrapati Shahuji Maharaj Medical University, Lucknow, India) for the cDNA clones and the antibodies used in Figure 1(C).

FUNDING

We thank Jawaharlal Nehru University LRE and UGC [University Grants Commission (India)] for funding support to U.P.

References

References
1
Zou
 
Y.
Liu
 
Y.
Wu
 
X.
Shell
 
S. M.
 
Functions of human replication protein A (RPA): from DNA replication to DNA damage and stress responses
J. Cell. Physiol.
2006
, vol. 
208
 (pg. 
267
-
273
)
2
Fanning
 
E.
Klimovich
 
V.
Nager
 
A. R.
 
A dynamic model for replication protein A (RPA) function in DNA processing pathways
Nucleic Acids Res.
2006
, vol. 
34
 (pg. 
4126
-
4137
)
3
Stauffer
 
M. E.
Chazin
 
W. J.
 
Structural mechanisms of DNA replication, repair, and recombination
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
30915
-
30918
)
4
Costa
 
R. M.
Chigancas
 
V.
Galhardo Rda
 
S.
Carvalho
 
H.
Menck
 
C. F.
 
The eukaryotic nucleotide excision repair pathway
Biochimie
2003
, vol. 
85
 (pg. 
1083
-
1099
)
5
Wold
 
M. S.
 
Replication protein A: a heterotrimeric, single-stranded DNA-binding protein required for eukaryotic DNA metabolism
Annu. Rev. Biochem.
1997
, vol. 
66
 (pg. 
61
-
92
)
6
Bochkareva
 
E.
Kaustov
 
L.
Ayed
 
A.
Yi
 
G. S.
Lu
 
Y.
Pineda-Lucena
 
A.
Liao
 
J. C.
Okorokov
 
A. L.
Milner
 
J.
Arrowsmith
 
C. H.
Bochkarev
 
A.
 
Single-stranded DNA mimicry in the p53 transactivation domain interaction with replication protein A
Proc. Natl. Acad. Sci. U.S.A.
2005
, vol. 
102
 (pg. 
15412
-
15417
)
7
Dutta
 
A.
Ruppert
 
J. M.
Aster
 
J. C.
Winchester
 
E.
 
Inhibition of DNA replication factor RPA by p53
Nature
1993
, vol. 
365
 (pg. 
79
-
82
)
8
Vogelstein
 
B.
Lane
 
D.
Levine
 
A. J.
 
Surfing the p53 network
Nature
2000
, vol. 
408
 (pg. 
307
-
310
)
9
Abramova
 
N. A.
Russell
 
J.
Botchan
 
M.
Li
 
R.
 
Interaction between replication protein A and p53 is disrupted after UV damage in a DNA repair-dependent manner
Proc. Natl. Acad. Sci. U.S.A.
1997
, vol. 
94
 (pg. 
7186
-
7191
)
10
Romanova
 
L. Y.
Willers
 
H.
Blagosklonny
 
M. V.
Powell
 
S. N.
 
The interaction of p53 with replication protein A mediates suppression of homologous recombination
Oncogene
2004
, vol. 
23
 (pg. 
9025
-
9033
)
11
Sommers
 
J. A.
Sharma
 
S.
Doherty
 
K. M.
Karmakar
 
P.
Yang
 
Q.
Kenny
 
M. K.
Harris
 
C. C.
Brosh
 
R. M.
 
p53 modulates RPA-dependent and RPA-independent WRN helicase activity
Cancer Res.
2005
, vol. 
65
 (pg. 
1223
-
1233
)
12
Albrechtsen
 
N.
Dornreiter
 
I.
Grosse
 
F.
Kim
 
E.
Wiesmuller
 
L.
Deppert
 
W.
 
Maintenance of genomic integrity by p53: complementary roles for activated and non-activated p53
Oncogene
1999
, vol. 
18
 (pg. 
7706
-
7717
)
13
Janus
 
F.
Albrechtsen
 
N.
Dornreiter
 
I.
Wiesmuller
 
L.
Grosse
 
F.
Deppert
 
W.
 
The dual role model for p53 in maintaining genomic integrity
Cell. Mol. Life Sci.
1999
, vol. 
55
 (pg. 
12
-
27
)
14
Bindra
 
R. S.
Schaffer
 
P. J.
Meng
 
A.
Woo
 
J.
Maseide
 
K.
Roth
 
M. E.
Lizardi
 
P.
Hedley
 
D. W.
Bristow
 
R. G.
Glazer
 
P. M.
 
Alterations in DNA repair gene expression under hypoxia: elucidating the mechanisms of hypoxia-induced genetic instability
Ann. N.Y. Acad. Sci.
2005
, vol. 
1059
 (pg. 
184
-
195
)
15
Brown
 
L. M.
Cowen
 
R. L.
Debray
 
C.
Eustace
 
A.
Erler
 
J. T.
Sheppard
 
F. C.
Parker
 
C. A.
Stratford
 
I. J.
Williams
 
K. J.
 
Reversing hypoxic cell chemoresistance in vitro using genetic and small molecule approaches targeting hypoxia inducible factor-1
Mol. Pharmacol.
2006
, vol. 
69
 (pg. 
411
-
418
)
16
Bristow
 
R. G.
Hill
 
R. P.
 
Hypoxia and metabolism. Hypoxia, DNA repair and genetic instability
Nat. Rev. Cancer
2008
, vol. 
8
 (pg. 
180
-
192
)
17
Huang
 
L. E.
Bindra
 
R. S.
Glazer
 
P. M.
Harris
 
A. L.
 
Hypoxia-induced genetic instability–a calculated mechanism underlying tumor progression
J. Mol. Med.
2007
, vol. 
85
 (pg. 
139
-
148
)
18
Bindra
 
R. S.
Crosby
 
M. E.
Glazer
 
P. M.
 
Regulation of DNA repair in hypoxic cancer cells
Cancer Metastasis Rev.
2007
, vol. 
26
 (pg. 
249
-
260
)
19
Rezvani
 
H. R.
Mahfouf
 
W.
Ali
 
N.
Chemin
 
C.
Ged
 
C.
Kim
 
A. L.
de Verneuil
 
H.
Taieb
 
A.
Bickers
 
D. R.
Mazurier
 
F.
 
Hypoxia-inducible factor-1α regulates the expression of nucleotide excision repair proteins in keratinocytes
Nucleic Acids Res.
2010
, vol. 
38
 (pg. 
797
-
809
)
20
Unruh
 
A.
Ressel
 
A.
Mohamed
 
H. G.
Johnson
 
R. S.
Nadrowitz
 
R.
Richter
 
E.
Katschinski
 
D. M.
Wenger
 
R. H.
 
The hypoxia-inducible factor-1α is a negative factor for tumor therapy
Oncogene
2003
, vol. 
22
 (pg. 
3213
-
3220
)
21
Murray
 
D.
Vallee-Lucic
 
L.
Rosenberg
 
E.
Andersson
 
B.
 
Sensitivity of nucleotide excision repair-deficient human cells to ionizing radiation and cyclophosphamide
Anticancer Res.
2002
, vol. 
22
 (pg. 
21
-
26
)
22
Mao
 
Z.
Seluanov
 
A.
Jiang
 
Y.
Gorbunova
 
V.
 
TRF2 is required for repair of nontelomeric DNA double-strand breaks by homologous recombination
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
13068
-
13073
)
23
Biggerstaff
 
M.
Wood
 
R. D.
 
Repair synthesis assay for nucleotide excision repair activity using fractionated cell extracts and UV-damaged plasmid DNA
Methods Mol. Biol.
2006
, vol. 
314
 (pg. 
417
-
434
)
24
Li
 
G.
Ho
 
V. C.
 
p53-dependent DNA repair and apoptosis respond differently to high- and low-dose ultraviolet radiation
Br. J. Dermatol.
1998
, vol. 
139
 (pg. 
3
-
10
)
25
Sullivan
 
R.
Pare
 
G. C.
Frederiksen
 
L. J.
Semenza
 
G. L.
Graham
 
C. H.
 
Hypoxia-induced resistance to anticancer drugs is associated with decreased senescence and requires hypoxia-inducible factor-1 activity
Mol. Cancer Ther.
2008
, vol. 
7
 (pg. 
1961
-
1973
)
26
Fyles
 
A. W.
Milosevic
 
M.
Wong
 
R.
Kavanagh
 
M. C.
Pintilie
 
M.
Sun
 
A.
Chapman
 
W.
Levin
 
W.
Manchul
 
L.
Keane
 
T. J.
Hill
 
R. P.
 
Oxygenation predicts radiation response and survival in patients with cervix cancer
Radiother. Oncol.
1998
, vol. 
48
 (pg. 
149
-
156
)
27
Sasabe
 
E.
Zhou
 
X.
Li
 
D.
Oku
 
N.
Yamamoto
 
T.
Osaki
 
T.
 
The involvement of hypoxia-inducible factor-1α in the susceptibility to γ-rays and chemotherapeutic drugs of oral squamous cell carcinoma cells
Int. J. Cancer
2007
, vol. 
120
 (pg. 
268
-
277
)
28
Wood
 
R. D.
 
DNA repair in eukaryotes
Annu. Rev. Biochem.
1996
, vol. 
65
 (pg. 
135
-
167
)
29
Gogna
 
R.
Madan
 
E.
Kuppusamy
 
P.
Pati
 
U.
 
Chaperoning of mutant p53 by wild-type p53 causes hypoxic tumor regression
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
2907
-
2914
)
30
Ward
 
I. M.
Minn
 
K.
Chen
 
J.
 
UV-induced ataxia-telangiectasia-mutated and Rad3-related (ATR) activation requires replication stress
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
9677
-
9680
)
31
Yajima
 
H.
Lee
 
K. J.
Zhang
 
S.
Kobayashi
 
J.
Chen
 
B. P.
 
DNA double-strand break formation upon UV-induced replication stress activates ATM and DNA-PKcs kinases
J. Mol. Biol.
2009
, vol. 
385
 (pg. 
800
-
810
)
32
Bhana
 
S.
Hewer
 
A.
Phillips
 
D. H.
Lloyd
 
D. R.
 
p53-dependent global nucleotide excision repair of cisplatin-induced intrastrand cross links in human cells
Mutagenesis
2008
, vol. 
23
 (pg. 
131
-
136
)
33
Jones
 
J. C.
Zhen
 
W. P.
Reed
 
E.
Parker
 
R. J.
Sancar
 
A.
Bohr
 
V. A.
 
Gene-specific formation and repair of cisplatin intrastrand adducts and interstrand cross-links in Chinese hamster ovary cells
J. Biol. Chem.
1991
, vol. 
266
 (pg. 
7101
-
7107
)
34
Saijo
 
M.
Matsuda
 
T.
Kuraoka
 
I.
Tanaka
 
K.
 
Inhibition of nucleotide excision repair by anti-XPA monoclonal antibodies which interfere with binding to RPA, ERCC1, and TFIIH
Biochem. Biophys. Res. Commun.
2004
, vol. 
321
 (pg. 
815
-
822
)
35
Zhong
 
Q.
Boyer
 
T. G.
Chen
 
P. L.
Lee
 
W. H.
 
Deficient nonhomologous end-joining activity in cell-free extracts from Brca1-null fibroblasts
Cancer Res.
2002
, vol. 
62
 (pg. 
3966
-
3970
)
36
Ginis
 
I.
Faller
 
D. V.
 
Hypoxia affects tumor cell invasiveness in vitro: the role of hypoxia-activated ligand HAL1/13 (Ku86 autoantigen)
Cancer Lett.
2000
, vol. 
154
 (pg. 
163
-
174
)
37
Lynch
 
E. M.
Moreland
 
R. B.
Ginis
 
I.
Perrine
 
S. P.
Faller
 
D. V.
 
Hypoxia-activated ligand HAL-1/13 is lupus autoantigen Ku80 and mediates lymphoid cell adhesion in vitro
Am. J. Physiol. Cell Physiol.
2001
, vol. 
280
 (pg. 
C897
-
C911
)
38
Bindra
 
R. S.
Gibson
 
S. L.
Meng
 
A.
Westermark
 
U.
Jasin
 
M.
Pierce
 
A. J.
Bristow
 
R. G.
Classon
 
M. K.
Glazer
 
P. M.
 
Hypoxia-induced down-regulation of BRCA1 expression by E2Fs
Cancer Res.
2005
, vol. 
65
 (pg. 
11597
-
11604
)
39
Perrault
 
R.
Cheong
 
N.
Wang
 
H.
Iliakis
 
G.
 
RPA facilitates rejoining of DNA double-strand breaks in an in vitro assay utilizing genomic DNA as substrate
Int. J. Radiat. Biol.
2001
, vol. 
77
 (pg. 
593
-
607
)
40
Sharma
 
A. K.
Ali
 
A.
Gogna
 
R.
Singh
 
A. K.
Pati
 
U.
 
p53 amino-terminus region (1–125) stabilizes and restores heat denatured p53 wild phenotype
PLoS ONE
2009
, vol. 
4
 pg. 
e7159
 
41
Thompson
 
T.
Tovar
 
C.
Yang
 
H.
Carvajal
 
D.
Vu
 
B. T.
Xu
 
Q.
Wahl
 
G. M.
Heimbrook
 
D. C.
Vassilev
 
L. T.
 
Phosphorylation of p53 on key serines is dispensable for transcriptional activation and apoptosis
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
53015
-
53022
)
42
Ashcroft
 
M.
Kubbutat
 
M. H.
Vousden
 
K. H.
 
Regulation of p53 function and stability by phosphorylation
Mol. Cell. Biol.
1999
, vol. 
19
 (pg. 
1751
-
1758
)
43
Unger
 
T.
Sionov
 
R. V.
Moallem
 
E.
Yee
 
C. L.
Howley
 
P. M.
Oren
 
M.
Haupt
 
Y.
 
Mutations in serines 15 and 20 of human p53 impair its apoptotic activity
Oncogene
1999
, vol. 
18
 (pg. 
3205
-
3212
)
44
Soubeyrand
 
S.
Schild-Poulter
 
C.
Hache
 
R. J.
 
Structured DNA promotes phosphorylation of p53 by DNA-dependent protein kinase at serine 9 and threonine 18
Eur. J. Biochem.
2004
, vol. 
271
 (pg. 
3776
-
3784
)
45
Lees-Miller
 
S. P.
Sakaguchi
 
K.
Ullrich
 
S. J.
Appella
 
E.
Anderson
 
C. W.
 
Human DNA-activated protein kinase phosphorylates serines 15 and 37 in the amino-terminal transactivation domain of human p53
Mol. Cell. Biol.
1992
, vol. 
12
 (pg. 
5041
-
5049
)
46
Derheimer
 
F. A.
O'Hagan
 
H. M.
Krueger
 
H. M.
Hanasoge
 
S.
Paulsen
 
M. T.
Ljungman
 
M.
 
RPA and ATR link transcriptional stress to p53
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
12778
-
12783
)
47
Esashi
 
F.
Christ
 
N.
Gannon
 
J.
Liu
 
Y.
Hunt
 
T.
Jasin
 
M.
West
 
S. C.
 
CDK-dependent phosphorylation of BRCA2 as a regulatory mechanism for recombinational repair
Nature
2005
, vol. 
434
 (pg. 
598
-
604
)
48
Hammond
 
E. M.
Dorie
 
M. J.
Giaccia
 
A. J.
 
ATR/ATM targets are phosphorylated by ATR in response to hypoxia and ATM in response to reoxygenation
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
12207
-
12213
)
49
Shiloh
 
Y.
 
ATM and related protein kinases: safeguarding genome integrity
Nat. Rev. Cancer
2003
, vol. 
3
 (pg. 
155
-
168
)
50
Yee Koh
 
M.
Spivak-Kroizman
 
T. R.
Powis
 
G.
 
HIF-1 regulation: not so easy come, easy go
Trends Biochem. Sci.
2008
, vol. 
33
 (pg. 
526
-
534
)
51
Um
 
J. H.
Kang
 
C. D.
Bae
 
J. H.
Shin
 
G. G.
Kim
 
D. W.
Chung
 
B. S.
Kim
 
S. H.
 
Association of DNA-dependent protein kinase with hypoxia inducible factor-1 and its implication in resistance to anticancer drugs in hypoxic tumor cells
Exp. Mol. Med.
2004
, vol. 
36
 (pg. 
233
-
242
)

Author notes

1

These authors contributed equally to this study.

Supplementary data