Distinct roles of XRCC1 in genome integrity in Xenopus egg extracts

Abstract

Oxidative DNA damage represents one of the most abundant DNA lesions. It remains unclear how DNA repair and DNA damage response (DDR) pathways are coordinated and regulated following oxidative stress. While XRCC1 has been implicated in DNA repair, it remains unknown how exactly oxidative DNA damage is repaired and sensed by XRCC1. In this communication, we have demonstrated evidence that XRCC1 is dispensable for ATR-Chk1 DDR pathway following oxidative stress in Xenopus egg extracts. Whereas APE2 is essential for SSB repair, XRCC1 is not required for the repair of defined SSB and gapped plasmids with a 5’-OH or 5’-P terminus, suggesting that XRCC1 and APE2 may contribute to SSB repair via different mechanisms. Neither Polymerase beta nor Polymerase alpha is important for the repair of defined SSB structure. Nonetheless, XRCC1 is important for the repair of DNA damage following oxidative stress. Our observations suggest distinct roles of XRCC1 for genome integrity in oxidative stress in Xenopus egg extracts.

Introduction

Cells of all organisms are constantly exposed to threats, such as oxidative stress, from endogenous sources or environmental agents [1-3]. Oxidative stress-induced DNA damage includes oxidized base damage or sugar moiety damage, apurinic/apyrimidinic (AP) sites, single-strand breaks (SSBs), double-strand breaks (DSBs), interstrand crosslinks (ICLs), and oxidatively-generated clustered DNA lesions [4-6]. Oxidative DNA damage is repaired primarily by base excision repair (BER) while other DNA repair pathways such as nucleotide excision repair (NER), mismatch repair (MMR), and nucleotide incision repair (NIR) are backup options [5, 7-10]. The molecular mechanism of BER pathway includes damaged base removal by DNA glycosylase to generate AP site, SSB generation by APE1 or bi-functional glycosylase, and subsequent gap filling and ligation reactions [11]. BER pathway is composed of short-patch and long-patch sub-pathways, which has been reconstituted with purified human proteins in vitro [7, 12].

Representing about 10 percent of all DNA lesions, SSBs are generated from oxidative stress, intermediate products of DNA repair, or aborted activity of cellular enzymes such as DNA topoisomerase 1 [11, 13, 14]. A four-step mechanism of SSB repair including SSB detection, DNA end processing, DNA gap filling, and DNA ligation has been proposed previously [15]. In addition, recent studies suggest that SSBs can also be resolved by homologous recombination or alternative homologue-mediated SSB repair [16-18]. Unrepaired SSBs hinder proper DNA transcription and accurate DNA replication of the genome, leading to cancer and neurodegenerative disorders [15, 19]. For example, defective SSB repair is responsible for senescence and neoplastic escape of epithelial cells [20]. We, and others, have demonstrated that oxidative stress triggers both ATM-Chk2 and ATR-Chk1 DNA damage response (DDR) pathways. ATM can be activated through a disulfide bond formation and conformation change in a DNA-independent manner following oxidative stress in mammalian cells [21-24]. ATR DDR pathway can be activated by oxidative stress-damaged chromatin DNA and defined SSB structure [25, 26]. Although ATM is proposed to be activated by SSBs, it is not known how exactly SSBs activate ATM [24].

X-ray Repair Cross Complementing Protein 1 (XRCC1) has been implicated in different types of DNA repair pathways, including BER, NER, SSB repair, non-homologous end joining (NHEJ), and microhomology-mediated end joining (MMEJ) pathways [27-31]. XRCC1 acts as a scaffolding protein to recruit a multitude of factors to the site of DNA damage [27, 32, 33]. Furthermore, XRCC1 interacts with PCNA and DNA polymerase alpha, participating in DNA replication [34-36]. XRCC1-deficient mice are embryonically lethal, suggesting its physiological significance for development [37]. While over-expression or under-expression of XRCC1 has been linked to cancer, XRCC1 variants with Arg194Trp, Arg280His, or Arg399Gln mutant have been studied via epidemiological analysis and meta-analysis [27, 38-40]. DNA Polymerase beta (Pol beta) complexes with XRCC1 and acts as the main repair DNA polymerase in the BER pathway [41, 42]. Pol beta has the unique ability to repair DNA gaps smaller than 6 nucleotides but is incredibly error prone [43]. More recently, numerous XRCC1 case studies have come out uncovering specific mutations in these proteins that are believed to have a correlation to cancer progression [44, 45]. These mutations are believed to play a role in genome instability by compromising some aspect of XRCC1 functions. However, exact roles and mechanisms of XRCC1 in DNA repair and DDR pathways in response to oxidative DNA damage and SSBs remain unclear.

Xenopus egg extracts have been widely used as a biochemical system in studies of DNA metabolism, and findings from Xenopus system can be verified in mammalian system [46-48]. Low-speed supernatant (LSS), high-speed supernatant (HSS), and nucleoplasmic extracts (NPE) are three different types of Xenopus egg extracts [49-51]. We have demonstrated that the ATR-Chk1 DDR pathway is activated by hydrogen peroxide-induced oxidative stress in Xenopus LSS system, and that a BER protein APE2 (AP endonuclease 2, also known as Apn2 or Apex2) plays an essential role for ssDNA generation and assembly of checkpoint protein complex including ATR, ATRP, and TopBP1 to activate the ATR-Chk1 DDR pathway in oxidative stress [5, 25]. Furthermore, APE2’s conserved Zf-GRF motif in its extreme C-terminus is required for binding to ssDNA and its 3’-5’ exonuclease activity in the activation of ATR-Chk1 DDR following oxidative stress [52]. In addition, ATR-Chk1 DDR is activated by a defined SSB plasmid in Xenopus HSS system [26]. However, it remains unknown whether APE2 plays a direct role in the repair of oxidative DNA damage and SSBs.

In this communication, we demonstrate evidence that XRCC1 is dispensable for ATR-Chk1 DDR pathway following oxidative stress, and that XRCC1 depletion enhances DDR pathway activation. Surprisingly, XRCC1 is dispensable for the repair of defined SSB and gapped plasmids with a 5’-OH or 5’-P terminus in Xenopus HSS system. The repair of defined SSB plasmid requires APE2, but not Pol beta nor Pol alpha, in the HSS system. Lastly, we have shown that XRCC1 is important of the repair of DNA damage following oxidative stress in Xenopus LSS system.

Materials and methods

Experimental procedures for Xenopus laevis egg extracts and sperm chromatin

Experiments of Xenopus laevis were carried out at UNC Charlotte, Charlotte, NC 28223, USA. Xenopus laevis care and use was approved by UNC Charlotte’s Institutional Animal Care and Use Committee (IACUC). The preparation of Xenopus LSS, HSS, and sperm chromatin was described previously [48, 49, 53]. Immunodepletion of APE2 from HSS was performed as previously described [25, 26, 52]. Immunodepletion of XRCC1, Pol beta, or Pol alpha from HSS was performed using a similar approach as APE2 depletion in HSS. To deplete XRCC1 from LSS, 200 μL of LSS was incubated with ~40 μL of rProtein A Sepharose beads (GE Healthcare) pre-coupled with anti-XRCC1 antibodies, for ~40 min at 4°C with constant mixing. Typically, 3-round depletion is needed to get ~150 μL of XRCC1-depleted LSS from 200 μL of LSS.

For experiments in Figure 1, hydrogen peroxide (100mM) was added to mock- or XRCC1 depleted LSS or HSS, which was supplemented with sperm chromatin (4,000/μL). After incubation of different time as indicated at room temperature, 5 μL of reaction mixture was added with 45 μL of sample buffer for immunoblotting analysis. The remaining reaction mixture was diluted with 150 μL of Buffer XB (50 mM sucrose, 100 mM KCl, 100 μM CaCl₂, 2 mM MgCl₂, 10 mM Hepes, pH 7.7) and layered on a 400 μL of sucrose cushion (1.1 M sucrose in Buffer XB), and spun (11,000 rpm, 30 min, 4°C) with a swinging bucket. After centrifugation, the supernatants were removed, and the chromatin-bound protein factions were resuspended with sample buffer and examined via immunoblotting analysis.

Figure 1.

XRCC1 is not required for ATR-Chk1 DNA damage response pathway in Xenopus LSS system.

(A) Hydrogen peroxide (final concentration 100 mM) was added to mock- or XRCC-depleted LSS, which was supplemented with sperm chromatin and incubated for 45 minutes. Extracts were examined via immunoblotting analysis for Chk1 phosphorylation (i.e., Chk1 P-Ser344) and total Chk1. (B) Quantification of Chk1 P-S344 (normalized to total Chk1) from Panel (A). a.u., arbitrary units. (C) Chromatin fractions from Experiments in Panel (A) were isolated and examined via immunoblotting as indicated. Histone 3 serves as loading control. (D) ATR inhibitor VE-822 (final concentration 10 μM) or ATM inhibitor KU55933 (final concentration 100 μM) was added to XRCC1-depleted LSS, then supplemented with hydrogen peroxide (final concentration 100 mM) and sperm chromatin. After a 45-minute incubation, total egg extracts were examined via immunoblotting as indicated. (E) Quantification of Chk1 P-S344 (normalized to total Chk1) from Panel (D). a.u., arbitrary units. (F) Hydrogen peroxide (final concentration 100 mM) was added to mock- or XRCC-depleted HSS supplemented with sperm chromatin, followed by a 45-minute incubation. Extracts were examined via immunoblotting analysis for Chk1 and Chk1 P-Ser344. (G) Quantification of Chk1 P-S344 (normalized to total Chk1) from Panel (F). a.u., arbitrary units. (H) Chromatin fractions from Experiments in Panel (F) were isolated and examined via immunoblotting as indicated. (A,C,D,F,H) shows representative results from two independent experiments.

Preparation of SSB and gapped plasmids

The SSB plasmid with a 5’-OH was prepared as recently described [26, 54]. Briefly, the plasmid pS was treated with Nt.BstNBI to generate a site-specific nick between C435 and T436, followed by Calf Intestinal Alkaline Phosphatase (CIP) treatment to generate a 5’-OH. Furthermore, SSB plasmid with 5’-P was prepared by treating pS with Nt.BstNBI but without subsequent CIP. The SSB plasmid with a 5’-OH or 5’-P was further purified from agarose gel with QIAquick gel extraction kit.

To prepare gapped plasmids with 5’-OH or 5’-P, the SSB plasmid with a 5’-OH or 5’-P was treated with recombinant GST-APE1, respectively, in an exonuclease buffer (20 mM KCl, 10 mM MCl₂, 2 mM DTT, 50 mM HEPES, pH 7.5) at 55°C for 20 min to generate 1-3nt gap in the 3’-5’ direction, followed by phenol-chloroform extraction and purification. As shown in our recent study, this APE1-pretreatment can generate ~1-3 nt gap at the plasmid’s nick in the 3’ to 5’ direction [26]. The gapped plasmids with 5’-OH or 5’-P was also further purified from agarose gel with QIAquick gel extraction kit.

Analysis of DNA repair products of SSB or gapped plasmids in Xenopus HSS system

The SSB or gapped plasmid with 5’-OH or 5’-P was added to mock-, XRCC1-, or APE2-depleted HSS (final concentration 75 ng/μL). After incubation at room temperature for different times, nuclease-free water was added to each reaction, followed by DNA repair product isolation procedure including phenol-chloroform extractions, as described previously in our recent studies [26, 54]. The purified DNA repair products were then examined via agarose electrophoresis.

Recombinant DNA and proteins

Recombinant pGEX-4T1-XRCC1 was generated by cloning the coding region (nt 164-2119) of Xenopus laevis XRCC1 (GenBank: BC045032, Xenopus Gene Collection IMAGE ID: 5543195) into EcoRI- and XhoI-digested pGEX-4T1, as previously described [26]. Recombinant pGEX-4T1-Pol beta was made by cloning the coding region (nt 245-1249) of Xenopus laevis Pol beta (GenBank: BC106329, Xenopus Gene Collection IMAGE ID: 7203966) into pGEX-4T1 with EcoRI- and XhoI sites. GST-tagged recombinant proteins were expressed and purified in E. coli DE3/BL21 according to standard protocol. Purified recombinant proteins were confirmed on SDS-PAGE gels with coomassie staining.

Immunoblotting analysis and antibodies

For immunoblotting analysis, samples were denatured in the presence of reducing Laemmli buffer for 10 min at 95 °C and run on 5–15% polyacrylamide gels at a constant 25 mA per gel. Separated proteins were transferred to PVDF membranes (Immobilon-P 0.45 μm, Millipore) using wet transfer at 110 V for 80 min. PVDF membranes will be washed, blocked, and incubated with appropriate primary and secondary antibodies. Membranes were washed again 3 times with TBST and incubated with WesternBright ECL or Sirius substrate for 10 minutes (Advansta, USA). Fluorescence was observed using X-ray film or a ChemiDoc MP Imaging System (BIO-RAD, USA).

Anti-XRCC1 and anti-Pol beta antibodies were raised in rabbits against GST-XRCC1 and GST-Pol beta, respectively (Cocalico Biologicals). Anti-Xenopus APE2 antibodies was described previously [25]. Antibodies against ATM were provided by Dr. Zhongsheng You [55]. Antibodies against ATRIP were provided by Dr. Howard Lindsay [56]. Antibodies against RPA32 and Pol alpha were provided by Dr. Matthew Michael [57]. Antibodies against Chk1 phosphorylation at Ser345 were purchased from Cell Signaling Technology. Anti-ATM phosphorylation at Ser1981 was purchased from Rockland. Antibodies against Histone 3 were purchased from Abcam. Antibodies against Chk1, GST, and Myc were purchased from Santa Cruz Biotechnology.

COMET Assays

COMET assays were preformed using the OxiSelect Comet Assay Kit from Cell BioLabs, Inc. The procedure was modified and tailored to Xenopus laevis cell lysates as opposed to mammalian cell samples. LSS reactions were performed as described above until the sample buffer would be added. From that point the entire ~50μL reaction was diluted with 1mL of cold PBS and spin in a swinging bucket tabletop centrifuge at 2,000rpm for 5 min at 4°C. After centrifugation, the supernatant was aspirated and the pellet was resuspended with 0.2mL of cold PBS. From this point the procedure follows the OxiSelect standard procedure for both the alkaline (pH>13) and neutral conditions (pH=~7.0) with the only exception being SYBR Gold was used, and diluted 1:10,000 in TE buffer pH 7.5. After the slides were allowed to dry the nuclei were observed with fluorescent microscope using a FITC filter. Images were taken using DP Controller software (Olympus Corporation, JPN) and analyzed using Comet Assay IV Lite software (Instem, UK).

Quantification and statistical analysis

ImageJ was utilized to quantify bands of SSB/gapped plasmid and repair products from DNA repair assays in Figures 2 and 3. Intensity of Chk1 P-S344 bands were quantified using ImageJ and normalized with Chk1 bands from Figure 1. Statistical analysis was performed using GraphPad Prism8. t-test was performed for statistical analysis in Figures 2 and 3. Ordinary one-way ANOVA (Tukey’s multiple comparisons test) was chosen for statistical analysis in Figure 4B and 4D.

Figure 2.

XRCC1 is dispensable for the repair of SSB or gapped plasmids in Xenopus HSS system.

(A) SSB plasmid with a 5’-OH was incubated in mock- or XRCC1-depleted HSS. After different timepoints (0, 5, 30, 90 min), DNA repair products were isolated and examined on agarose gel. * indicates partially contaminated DSB in our prep. (B) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (A) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=4). (C) SSB plasmid with a 5’-P was added to mock- or XRCC1-depleted HSS. After different times, DNA repair products were isolated and examined on agarose gel. (D) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (C) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=3). (E) Gapped plasmid with a 5’-OH was added to mock- or XRCC1-depleted HSS. After different times, DNA repair products were isolated and examined on agarose gel. (F) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and gapped plasmid) from Panel (E) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=3). (G) Gapped plasmid with a 5’-P was added to mock- or XRCC1-depleted HSS. After different times, DNA repair products were isolated and examined on agarose gel. (H) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and gapped plasmid) from Panel (G) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=3).

Figure 3.

APE2, but not Pol beta nor Pol alpha, is important for SSB repair in Xenopus HSS system.

(A) Recombinant Myc-APE2 was added back to APE2-depleted HSS. Then, SSB plasmid with a 5’-OH was added to mock- or APE2-depleted HSS. After different timepoints (1, 5, 30 min), DNA repair products were isolated and examined on agarose gel. (B) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (A) was analyzed using Image J. ** indicates p<0.01; * indicates p<0.05 (n=3). (C) SSB plasmid with a 5’-OH was incubated in mock- or Pol beta-depleted HSS. After different timepoints (0, 5, 30 min), DNA repair products were isolated and examined on agarose gel. (D) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (C) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=3). (E) SSB plasmid with a 5’-OH was incubated in the presence of DMSO or Aphidicolin (295 μM) in HSS. After different timepoints (0, 5, 30 min), DNA repair products were isolated and examined on agarose gel. (F) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (E) was analyzed using Image J. ** indicates p<0.01; * indicates p<0.05 (n=3). (G) SSB plasmid with a 5’-OH was incubated in mock- or Pol alpha-depleted HSS. After different timepoints (0, 5, 30 min), DNA repair products were isolated and examined on agarose gel. (H) DNA repair capacity (%, i.e., intensity of DNA repair products / intensity of DNA repair products and SSB plasmid) from Panel (G) was analyzed using Image J. “n.s.” represents no significance (p>0.05, n=3). (A,C,G) * indicates partially contaminated DSB in our prep.

Figure 4.

XRCC1 is important to repair DNA damage following oxidative stress in Xenopus egg extracts.

(A) Hydrogen peroxide and sperm chromatin were added to mock- or XRCC1-depleted LSS. After a 30-minute incubation, reaction mixture was further examined with COMET assays under alkaline condition. Representative images are shown. (B) Quantification of DNA damage from four reactions shown in panel (A). *** indicates p<0.0001; ** indicates p<0.001. (C) Hydrogen peroxide and sperm chromatin were added to mock- or XRCC1-depleted LSS. After a 30-minute incubation, reaction mixture was further analyzed using COMET assays under neutral condition. Representative images are shown. (D) Quantification of DNA damage from four reactions shown in panel (C). *** indicates p<0.0001; * indicates p<0.01. “n.s.” shows no significance.

Results

XRCC1 is not required for ATR-Chk1 DDR pathway activation following oxidative stress in Xenopus LSS and HSS systems

Xenopus XRCC1 contains an N-terminal domain (NTD), a nuclear localization signaling (NLS) in the middle, and two BRCT domains (i.e., BRCT1 and BRCT2) in the C-terminus (Figure S1A), which is very similar to its homologue in humans [58]. Clustral Omega analysis of XRCC1 shows high identity or similarity in the amino acid sequence in Xenopus, human, and mouse (Figure S1B), suggesting that XRCC1 may have similar or conserved functions during evolution. In particular, BLASTP analysis shows that there are 54% (366/677) identities and 67% (458/677) positives between Xenopus XRCC1 (GenBank#: AAH45032) and human XRCC1 (GenBank#: NP_006288).

We recently reported that XRCC1 is not required for the activation of ATR-Chk1 DDR pathway in response to a defined SSB plasmid in Xenopus HSS system [26]. As briefly mentioned previously [26], we constructed recombinant GST-XRCC1, which was expressed in E. coli DE3 cells with IPTG induction and purified following vendor’s protocol. SDS-PAGE analysis verified the purified recombinant GST-XRCC1, which was shown at ~100 kD position on gel, as expected (Figure S2A). The purified recombinant GST-XRCC1 was utilized for custom antibodies production in rabbits from Cocalico Biologicals Inc. Anti-XRCC1 antibodies were used to immunodeplete endogenous XRCC1 successfully from LSS and HSS, respectively (Figure S3A,S3B). To test whether XRCC1 is important for ATR-Chk1 DDR pathway following oxidative stress, we added hydrogen peroxide and sperm chromatin into mock- or XRCC1-depleted LSS. After a 45-min incubation, total egg extracts examination via immunoblotting (IB) analysis demonstrated that hydrogen peroxide triggered Chk1 phosphorylation and ATM phosphorylation in mock-depleted LSS (Figure 1A,1D), consistent with previous studies [25, 52]. A previous report using mammalian cells has demonstrated that oxidative stress activates ATM DDR pathway via a cysteine residue-mediated dimerization of ATM [23]. We noticed that hydrogen peroxide did not trigger ATM phosphorylation nor Chk1 phosphorylation in the absence of chromatin DNA in LSS (Lane 4 to Lane 2, Figure S2C), suggesting that hydrogen peroxide-induced oxidative stress may not activate ATM in the absence of DNA in Xenopus egg extracts.

Notably, hydrogen peroxide-induced Chk1 phosphorylation and ATM phosphorylation were slightly increased when XRCC1 was depleted in LSS (Lane 4 vs. Lane 2, Figure 1A; Lane 8 vs. Lane 2, Figure 1D). Interestingly, XRCC1-depletion also triggered mild Chk1 phosphorylation without the treatment of hydrogen peroxide (Lane 3 vs. Lane 1, Figure 1A). It seems that quantification of Chk1 P-S344 bands (normalized to Chk1 bands, Figure 1B) is consistent with the observations of IB analysis from Figure 1A. Chromatin fraction analysis shows that the recruitment of ATRIP and RPA32 to chromatin with the presence and absence of hydrogen peroxide was slightly increased when XRCC1 was depleted from LSS (Figure 1C), which is consistent with the ATR activation.

A recent study demonstrates that ATM is activated by DNA damage induced by XRCC1-deficiency to prevent DSB formation in mammalian cells [24]. To determine whether the slightly enhanced Chk1 phosphorylation in XRCC1-depelted LSS is due to ATM activation, we added ATM specific inhibitor KU55933 in mock- and XRCC1-depleted LSS. The addition of KU55933 reversed the increased Chk1 phosphorylation with the presence and absence of hydrogen peroxide in XRCC1-depleted LSS (Lanes 7&8, 11&12, Figure 1D,1E), suggesting a role of ATM in response to XRCC1 deficiency-induced DNA damage. ATM phosphorylation induced by hydrogen peroxide was compromised by the addition of KU55933 in both mock- and XRCC1-depleted LSS (Figure 1D). Furthermore, ATR specific inhibitor VE-822 impaired Chk1 phosphorylation in both mock- and XRCC1-depeleted LSS, regardless of hydrogen peroxide (Lanes 3&4, Lane 9&10, Figure 1D,1E). These observations suggest that XRCC1 is not required for hydrogen peroxide-induced ATR-Chk1 DDR activation and that XRCC1 depletion may trigger both ATR and ATM activation under non-perturbed conditions.

There are several different types of Xenopus egg extracts: LSS, HSS, and NPE [48, 49]. Generally speaking, chromatin DNA can’t form nuclear membrane for DNA synthesis in HSS, due to the lack of membrane fractions and CDKs and DDKs [48]. Taking advantage of DNA replication-deficiency in HSS, we sought to determine whether XRCC1 deficiency induces Chk1 phosphorylation in HSS. As shown in Figure 1F,1G, XRCC1 depletion in HSS did not result in noticeable Chk1 phosphorylation under normal condition. Chromatin fraction analysis also shows that neither ATRIP nor RPA32 was noticeably increased on chromatin under normal conditions in XRCC1-depleted HSS (Figure 1H). These observations suggest that DNA replication may be needed to generate necessary DNA damage for DDR pathway activation when XRCC1 is absent. Similar to the observations in LSS system (Figure 1A), hydrogen peroxide-induced Chk1 phosphorylation was not decreased in XRCC1-depleted HSS (Figure 1F,1G).

Overall, XRCC1 is dispensable for ATR-Chk1 DDR pathway activation following oxidative stress in Xenopus LSS and HSS systems.

XRCC1 is dispensable for the repair of plasmid-based SSB or gapped structures with defined clean ends in Xenopus egg extracts

We recently demonstrated that a defined SSB plasmid with a 5’-OH can be repaired in the Xenopus HSS system and that SSB-induced ATR activation is required for SSB repair [26]. To test whether XRCC1 is required for repairing the defined SSB plasmid, we added SSB plasmid with a 5’-OH in mock- or XRCC1-depleted HSS. After different timepoints, DNA repair products were isolated and examined via agarose gel electrophoresis. The SSB repair in the absence of XRCC1 is similar to that in the presence of XRCC1 (Figure 2A). Quantification of DNA repair capacity at 5 min and 30 min shows no significance between the mock- and XRCC1-depletion HSS (Figure 2B). This observation suggests that XRCC1 is dispensable for repairing the defined SSB plasmid with a 5’-OH in the Xenopus HSS system. Next, we tested whether the presence of a 5’-P at the SSB site may affect the potential role of XRCC1 in DNA repair. The DNA repair capacity of the SSB plasmid with a 5’-P in XRCC1-depleted HSS was similar to that of mock-depleted HSS (Figure 2C,2D), suggesting that XRCC1 is not required for the repair of SSB plasmid with a 5’-P. Then, we sought to determine whether a gapped plasmid with 5’-OH or 5’-P can be repaired when XRCC1 is depleted in HSS. Notably, the repair of the gapped plasmid with a 5’-OH in XRCC1-depleted HSS is not noticeably affected in comparison to that in mock-depleted HSS (Figure 2E,2F), suggesting that XRCC1 is dispensable for repairing the gapped structure with a 5’-OH. We found similar results using a gapped plasmid with a 5’-P (Figure 2G,2H). Together, these observations suggest that XRCC1 is dispensable for the repair of SSB or gapped structures with defined clean ends such as 5’-OH or 5’-P terminus in Xenopus egg extracts.

The repair of SSB plasmid with defined clean ends requires APE2, but not Pol beta or Pol alpha

We recently reported that APE2 is required for SSB end resection in the 3’-5’ direction and SSB-induced ATR-Chk1 DDR pathway in the HSS system [26]. To test whether APE2 is important for SSB repair, we removed APE2 from HSS (Figure S3C), and found that the repair of SSB plasmid with a 5’-OH was significantly compromised in APE2-depleted HSS (Figure 3A,3B). To determine whether SSB repair deficiency in APE2-depleted HSS is due to APE2 absence, we added recombinant Myc-APE2 back to APE2-depleted HSS (Figure S3C), and found that adding back Myc-APE2 rescued the SSB repair deficiency in APE2-depleted HSS (Figure 3A,3B). These observations suggest that APE2 is required for the repair of SSB plasmid with defined clean ends such as a 5’-OH in the HSS system.

To determine the potential role of Pol beta for SSB repair in Xenopus egg extracts, we constructed pGEX-4T1-Pol beta plasmid, and purified recombinant GST-Pol beta protein was verified via SDS-PAGE gel (Figure S2B). GST-Pol beta runs to a position around 70 kD, as expected (Figure S2B). Next, GST-Pol beta protein was injected to rabbits for custom antibodies production. Anti-Pol beta antibodies were able to remove endogenous Pol beta from Xenopus HSS successfully (Figure S3D). As shown in Figure 3C, repair capacity of the SSB plasmid with a 5’-OH in Pol beta-depleted HSS was similar to that in mock-depleted HSS. Quantification of DNA repair capacity of SSB plasmid from three independent experiments shows no significance between mock- and Pol beta-depleted HSS (Figure 3D). These observations suggest that Pol beta is not required for the repair of the SSB plasmid with defined clean ends in HSS system.

Aphidicolin has been demonstrated to inhibit Pol alpha, delta, and epsilon in eukaryotes including Xenopus [59, 60]. To determine whether these replicative DNA polymerases are important for SSB repair, we sought to test whether the addition of aphidicolin affects SSB repair in HSS system. DNA repair of the defined SSB plasmid with a 5’-OH was significantly impaired with the addition of aphidicolin in HSS (Figure 3E,3F). To further examine the role of Pol alpha in SSB repair, we removed Pol alpha from HSS via immunodepletion (Figure S3E), and found that the difference of DNA repair of the SSB plasmid in mock-depleted HSS and Pol alpha-depleted HSS is not significant at both 5 and 30 minutes (Figure 3G,3H). These observations suggest that the potential role of Pol alpha for SSB repair in HSS is very minimal, if there is any. From these findings, we speculate that after 3’-5’ end resection by APE2, Pol delta and/or Pol epsilon, but not Pol beta nor Pol alpha, may fill the gap for SSB repair.

XRCC1 is important for repairing DNA damage following oxidative stress in Xenopus egg extracts

Our observations suggest that XRCC1 is dispensable for the repair of the defined SSB plasmid. Next, we sought to determine whether XRCC1 is important for DNA damage repair following oxidative stress using COMET assays under alkaline and neutral conditions [61]. Using COMET assays under alkaline condition, we found that the Tail Moment was enhanced after hydrogen peroxide treatment in the LSS system, suggesting that more SSBs and AP sites are generated following oxidative stress (Figures 4A,4B). Notably, XRCC1 depletion significantly increased the Tail Moment with the absence and presence of hydrogen peroxide in the LSS system, suggesting that XRCC1 is important for repairing AP sites and SSBs with complex structures following oxidative stress (Figures 4A,4B). Using COMET assays under neutral condition, we found that the Tail Moment was increased after hydrogen peroxide treatment and that XRCC1 depletion increased the Tail Moment in the presence of hydrogen peroxide in the LSS system (Figure 4C,4D), suggesting that XRCC1 is important for repairing DSBs induced by oxidative stress. Overall, our evidence suggests that XRCC1 is important for the repair of oxidative stress-derived DNA damage in Xenopus egg extracts.

Discussion

XRCC1 and DDR pathway activation

Although we cannot exclude the possibility that XRCC1 may play differential role in DNA damage response at multiple time points under different conditions, our data at 45-min incubation in Xenopus LSS system under the conditions we tested have shown that XRCC1 is dispensable for the ATR-Chk1 DDR pathway activation in response to hydrogen peroxide (Figure 1). Consistent with this, we also recently reported that XRCC1 is not required for defined SSB-induced ATR-Chk1 DDR pathway in Xenopus HSS system [26]. Our finding of the independence of XRCC1 for ATR-Chk1 DDR activation in Xenopus system is consistent with several prior studies using mammalian cells or cell lines. The lack of the BER protein expression including XRCC1, PARP1, and Ligase III alpha in human monocytes results in more SSBs and DSBs accumulation following oxidative stress [62]. Notably, the ATR-Chk1 and ATM-Chk2 DDR pathways still can be activated upon treatment with tert-butyl hydroperoxide-derived oxidative stress in XRCC1-deficient monocytes [62]. Furthermore, XRCC1 is dispensable for the MMS-induced ATR-Chk1 DDR pathway activation in human breast cancer cell line MDA-MB-549 [63]. In addition, ATR inhibition is synthetically lethal in XRCC1 deficient cells with increased cytotoxicity and accumulation of DSBs [64]. All these studies support the notion that XRCC1 is dispensable for ATR DDR pathway activation.

What are the potential roles of DDR pathway activation for XRCC1 functions? ATM-Chk2 DDR pathway may promote BER pathway via Chk2-dependent XRCC1 phosphorylation at Thr284 residue, suggesting that DDR pathway activation is earlier event than BER pathway in response to oxidative stress [65]. Furthermore, persistent DNA damage-activated ATM can phosphorylate transcription factor Sp1 in TIG-1 fibroblasts to downregulate XRCC1 expression for cell elimination via apoptosis [66]. However, it remains unknown whether ATR DDR pathway directly regulates XRCC1 functions.

We observed that no ATM phosphorylation at Ser 1981 was triggered by the addition of hydrogen peroxide in the absence of chromatin DNA in the Xenopus LSS system (Figure S2C). A previous study has shown that hydrogen peroxide triggered ATM dimerization and ATM phosphorylation at Ser 1981 using recombinant human ATM protein or endogenous ATM immunoprecipitated from human cells exposed to hydrogen peroxide [23]. We speculate that the discrepancy may be due to the different experimental systems. We note that reducing agent such as dithiothreitol is added to the Xenopus LSS system during the preparation [48, 53]. We can’t rule out the possibility that some currently unknown proteins in Xenopus LSS system may negatively regulate potential hydrogen peroxide-induced ATM dimerization. Alternatively, it will be interesting to test whether hydrogen peroxide can trigger ATM dimerization and ATM phosphorylation at Ser 1981 using total cell-free lysates from cultured mammalian cells.

Distinct functions of XRCC1 and APE2

It is widely accepted that XRCC1 function as a scaffolding protein to interact with many repair proteins, such as APE1, NEIL1, NEIL2, OGG1, UNG1, PCNA, NTH1, Pol beta, PARP1, PNKP, and Ligase 3alpha [67]. For example, XRCC1 physically interacts with APE1 and stimulates its enzymatic activity and such XRCC1-APE1 interaction is essential for repairing DNA AP site in Chinese Hamster Ovary cell lines [68]. Strikingly, our evidence from this work together with our recent studies demonstrates that XRCC1 plays distinct functions in DNA repair and DDR pathways in comparison to APE2. We have revealed the requirement of APE2 for ATR-Chk1 DDR pathway following oxidative stress in Xenopus LSS system and defined SSB-induced ATR-Chk1 DDR pathway in Xenopus HSS system [25, 26, 52]. However, XRCC1 is dispensable for ATR-Chk1 DDR pathway following oxidative stress in Xenopus LSS and HSS system (Figure 1). In addition, APE2 is required for the repair of defined SSB plasmid in Xenopus HSS system (Figure 3A). In contrast, XRCC1 is dispensable for repairing SSB and gapped plasmids with defined clean ends such as a 5’-OH or 5’-P terminus in Xenopus HSS system (Figure 2). All these observations clearly indicate different requirements of APE2 and XRCC1 for SSB repair and ATR-Chk1 DDR pathways in maintaining genome integrity (Figure S4).

Role of XRCC1 in repairing different types of DNA damage

XRCC1 has been implicated in several different types of DNA repair pathways including BER, NER, SSB repair, NHEJ, and MMEJ [27, 67]. To the best of our knowledge, it is the first time to show that XRCC1 is not required for the repair of defined SSB plasmids with simple termini such as 5’-OH or 5’-P (Figure 2). Many prior studies on the role of XRCC1 in SSB repair primarily measure DNA repair of SSBs indirectly generated from stressful conditions, such as gamma-irradiation and alkylation agent methyl methanesulfonate (MMS) [28, 33]. It has been shown that XRCC1 stimulates PNKP activity to promote SSB repair using in vitro reconstitution system with recombinant proteins and defined SSB structures [32]. The different experimental systems (i.e., reconstitution system with purified proteins vs. Xenopus HSS system) and two different SSB structures may explain this discrepancy to our result. In addition, XRCC1 interacts with Pol beta and Ligase III to serve as a scaffolding protein in a reconstituted BER system [12]. Notably, in vitro biochemical analysis indicates that XRCC1 is dispensable for BER activity of 8-OH-dG, 5-hydroxycytosine, ethanoadenine, and uracil lesions, and that XRCC1 is important for the ligation step of BER and SSB repair [69].

Oxidative stress can induce several different types of DNA damage, including but not limited to, base damage, SSBs, DSBs, and AP sites (Figure S4) [5]. We note that our result does not exclude the potential role of XRCC1 for the repair of DSBs and SSBs with complex termini that may be generated in oxidative stress. Consistent with this, our evidence from COMET assays demonstrates that XRCC1 is important for the repair of oxidative stress-derived DNA damage, such as DSBs, AP sites, and SSBs with complexed termini (Figure 4 and S4). Future studies are needed to directly determine the exact roles of XRCC1 for repairing these different types of oxidative stress-derived DNA damage.

DNA polymerase for SSB repair in Xenopus egg extracts

Pol beta is responsible for two steps in BER pathway: excision of 5’-terminal deoxyribose phosphate (dRP) residue from incised AP site and gap filling [70, 71]. Both Pol alpha and Pol beta are required for repairing UV-induced DNA damage including cyclobutane pyrimidine dimers and (6-4) photoproducts in an nuclear extract from Xenopus oocytes [72]. Our evidence suggests that the repair of SSB plasmid with 5’-OH is not affected by the absence of Pol beta (Figure 3C,3D). We interpret this independence of Pol beta for SSB repair is due to the nature of SSB plasmid with simple ends on both sides. Our observation is reminiscent of a prior study showing independence of Pol beta for SSB repair of DNA damage induced by hydrogen peroxide in mouse embryonic fibroblasts [41]. Then, what DNA Polymerase is responsible for SSB repair? We found that the repair of SSB plasmid was impaired by the addition of aphidicolin (Figure 3E,3F), an inhibitor of Pol alpha, delta and epsilon. Interestingly, Pol alpha depletion had minimal effect on the repair of defined SSB plasmid (Figure 3G,3H). Notably, our prior studies have shown that SSB with a simple 5’-OH can be resected ~18-26nt in the 3’ to 5’ direction by APE2 in Xenopus HSS system [18, 26]. We speculate that Pol delta and/or Pol epsilon may be responsible for filling the gap induced by the 3’-5’ SSB end resection during repair process. Future studies are needed to test these possibilities in the step of gap filling during SSB repair process.

Overall, our evidence presented in this short communication demonstrates that, in contrast to APE2, XRCC1 is dispensable for oxidative stress-induced ATR-Chk1 DDR pathway in Xenopus system. XRCC1 may play important roles for repairing oxidative stress-derived DSBs and AP sites but not SSBs with defined clean ends under the conditions we tested. Targeting XRCC1 deficiency in breast cancer has been proposed for personalized therapy [73]. Therefore, our better understanding of roles and mechanisms of XRCC1 in genome integrity will provide insight into how design novel avenues to cancer therapies.

Abbreviations

AP

apurinic/apyrimidinic

BER

base excision repair

DDR

DNA damage response

CIP

calf Intestinal Alkaline Phosphatase

DSB

double-strand break

HSS

high-speed supernatant

ICLs

interstrand crosslinks

LSS

low-speed supernatant

MMEJ

microhomology-mediated end joining

MMR

mismatch repair

NER

nucleotide excision repair

NIR

nucleotide incision repair

NLS

nuclear localization signaling

NPE

nucleoplasmic extracts

Pol beta

DNA Polymerase beta

SSB

Single-strand break

XRCC1

X-ray Repair Cross Complementing Protein 1