Báo cáo khoa học: Poly(ADP-ribose) polymerase-1 protects excessive DNA strand breaks from deterioration during repair in human cell extracts pot

Poly(ADP-ribose) polymerase-1 protects excessive DNA strand breaks from deterioration during repair in human cell extracts Jason L. Parsons, Irina I. Dianova, Sarah L. Allinson* and Grigory L. Dianov MRC Radiation and Genome Stability Unit, Harwell, Oxfordshire, UK Spontaneously derived DNA lesions, such as base modifications and abasic (AP) sites, and products of base oxidation and alkylation are removed by the base excision repair (BER) pathway [1]. In human cells, BER is initiated by the removal of the damaged base by a DNA glycosylase to generate an AP site that is a substrate for an AP endonuclease (APE1) which cleaves the phosphodiester backbone 5¢ to the lesion, creating a 3¢-OH and 5¢-deoxyribose phosphate (dRP) terminus [2,3]. The majority of the incised AP site pro- ceeds via so-called ‘short-patch’ repair [4] whereby DNA polymerase b (Pol b) catalyses the removal of the 5¢-dRP lesion through a b-elimination mechanism and also inserts the correct nucleotide to fill the gap [5,6]. The remaining nick in the DNA backbone is then sealed by X-ray cross-complementing gene 1 (XRCC1)–DNA ligase IIIa complex [7–9]. BER proteins also participate in processing of DNA single-strand breaks (SSB) and one-nucleotide gaps: Pol b is involved in gap filling, XRCC1–DNA ligase IIIa complex is involved in ligation, and APE1 partici- pates in processing of blocked 3¢-ends [10,11]. Further- more, other proteins, such as polynucleotide kinase and poly(ADP-ribose) polymerase-1 (PARP-1) have been implicated in the repair of SSBs [11–13]. PARP-1 is thought to be involved in BER and strand-break processing, as it has a high binding affinity for SSBs [14]. On binding to the strand break, PARP-1 catalyti- cally synthesizes poly(ADP-ribose) polymers from NAD + that covalently modify proteins, of which PARP-1 itself is a target. Subsequently PARP-1 disso- ciates from the strand break [15]. The importance of PARP-1 in repair is revealed by the fact that PARP-1 knockout mice are hypersensitive to alkylating agents Keywords base excision repair; DNA polymerase b; DNA repair; poly(ADP-ribose) polymerase-1 (PARP-1); XRCC1 Correspondence G. L. Dianov, Radiation and Genome Stability Unit, Medical Research Council, Harwell, Oxfordshire OX11 0RD, UK Fax: +44 1235 841 200 Tel: +44 1235 841 134 E-mail: g.dianov@har.mrc.ac.uk *Present address Department of Biological Sciences, Lancas- ter University, Lancaster LA1 4YQ, UK (Received 13 December 2004, revised 19 January 2005, accepted 24 February 2005) doi:10.1111/j.1742-4658.2005.04628.x Base excision repair (BER), a major pathway for the removal of simple lesions in DNA, requires the co-ordinated action of several repair and ancillary proteins, the impairment of which can lead to genetic instability. We here address the role of poly(ADP-ribose) polymerase-1 (PARP-1) in BER. Using an in vitro cross-linking assay, we reveal that PARP-1 is always involved in repair of a uracil-containing oligonucleotide and that it binds to the damaged DNA during the early stages of repair. Inhibition of PARP-1 poly(ADP-ribosyl)ation by 3-aminobenzamide blocks dissociation of PARP-1 from damaged DNA and prevents further repair. We find that excessive poly(ADP-ribosyl)ation occurs when repair intermediates contain- ing single-strand breaks are in excess of the repair capacity of the cell extract, suggesting that repeated binding of PARP-1 to the nicked DNA occurs. We also find increased sensitivity of repair intermediates to nuclease cleavage in PARP-deficient mouse fibroblasts and after depletion of PARP- 1 from HeLa whole cell extracts. Our data support the model in which PARP-1 binding to DNA single-strand breaks or repair intermediates plays a protective role when repair is limited. Abbreviations APE1, apurinic ⁄ apyrimidinic endonuclease 1; BER, base excision repair; PARP-1, poly(ADP-ribose) polymerase 1; Pol b, DNA polymerase b; PVDF, poly(vinylidene difluoride); SSB, DNA single-strand break; WCE, whole cell extract; XRCC1, X-ray cross-complementing gene 1. 2012 FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS and irradiation and that PARP-1 null cell lines display symptoms of genomic instability after treatment with DNA-damaging agents [16–18]. Furthermore, the involvement of PARP-1 in BER is also supported by the ability of PARP-1 to interact with XRCC1 [7,19] and Pol b [20] and the observation that PARP-1 is required for the assembly of XRCC1 foci after oxida- tive DNA damage [21,22]. Although it has been sug- gested that PARP-1 plays a role in BER, the reports on PARP-1 involvement in BER are controversial [14,23,24] and its function in BER is unclear. In this study, we addressed the role of PARP-1 in BER and SSB repair and demonstrate that PARP-1 prevents excessive SSBs arising during BER or as a result of direct DNA damage. Results Cross-linking of BER proteins during repair of damaged DNA Although it has previously been shown that PARP-1 binds to nicked DNA and interferes with the repair reac- tion, it is not clear whether this binding is an integral part of BER and what the physiological significance is. A cross-linking protocol was used to examine the role of PARP-1 during repair of damaged DNA [25]. This pro- tocol uses oligonucleotides containing a 3¢-biotinylated end, which are used to form a uracil-containing duplex oligonucleotide complete with a hairpin loop (Fig. 1). The oligonucleotide is subsequently bound to streptavidin magnetic beads and incubated with HeLa whole cell extract (WCE) before the addition of formaldehyde to cross-link proteins to DNA. The beads are subsequently washed, the cross-links reversed, and released pro- teins separated by gel electrophoresis and identified by immunoblotting with the corresponding antibodies. During incubation of the uracil-containing substrate with WCE, uracil-DNA glycosylase removes uracil from the substrate DNA, and APE1 incises the AP site, generating a strand break containing a 5¢-sugar phos- phate. This is further processed by Pol b and XRCC1– DNA ligase IIIa heterodimer. Cross-linking during incubation of the uracil-containing substrate with HeLa WCE revealed that PARP-1 was the first protein to be present at the substrate and it gradually dissociated from the substrate within 4 min of incubation, while Pol b and XRCC1–DNA ligase IIIa, which are required for strand break processing, are cross-linked more efficiently after PARP dissociation (Fig. 2A; left panel). However, most PARP-1 cross-linking observed during the first 30 s was most probably due to its DNA-damage-independent binding to the 5¢ end, or the hairpin loop structure of the substrate oligonucleo- tide, as similar PARP-1 cross-linking was also observed in the case of the control substrate (Fig. 2A; right panel), although damage-specific PARP-1 binding was observed from 1 min onwards. In contrast, cross-link- ing of Pol b and XRCC1 is highly damage-specific, as, using a control undamaged substrate, we were unable to significantly cross-link any of these proteins (Fig. 2A; right panel). The same filters were analyzed with antibodies raised against poly(ADP-ribose) poly- mers, and we found that bound proteins undergo sub- stantial poly(ADP-ribosyl)ation. Interestingly, the peak of poly(ADP-ribosyl)ation at 1 min of incubation using the uracil-containing oligonucleotide (Fig. 2B, line 4) correlates well with the damage-specific binding of PARP-1 (Fig. 2A, compare lines 4 and 10). Substan- tially less poly(ADP-ribosyl)ation was observed using the control oligonucleotide (Fig. 2B; right panel) and occurred before 1 min of incubation of the substrate with WCE and was associated with damage-unspecific PARP-1 binding. In comparison, poly(ADP-ribo- syl)ation using the uracil-containing oligonucleotide was still evident after up to 8 min of incubation with WCE. These results suggest the involvement of PARP-1 and PARP-1 poly(ADP-ribosyl)ation during BER of a uracil-containing oligonucleotide. Inhibition of PARP-1 poly(ADP-ribosyl)ation prevents BER in cell extracts To examine whether PARP-1 is involved in every sin- gle BER event on damaged DNA, or just simply Fig. 1. Structures of oligonucleotides used to construct 3¢-biotinylated hairpin substrates. Oligonucleotides were designed to contain the complementary sequence with a TTTT hairpin loop and a 3¢-biotinylated moiety (designated with an asterisk). Substrates (1) and (2) contain uracil and cytosine, respectively, which are base- paired with guanine. Substrates (3) and (4) contain a nick with 3¢-OH, 5¢-phosphate and 3¢-phosphate, 5¢-OH ends, respectively. J. L. Parsons et al. Role of PARP-1 in base excision repair FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS 2013 randomly competes with BER enzymes for the sub- strate, we incubated the uracil-containing oligonucleo- tide with HeLa WCE in the presence of the poly(ADP- ribosyl)ation inhibitor 3-aminobenzamide. Preventing PARP-1 poly(ADP-ribosyl)ation should block PARP-1 dissociation from nicked DNA and thus inhibit further binding of BER proteins. We found that 3-aminobenz- amide stimulated cross-linking of PARP-1 throughout the course of the reaction and completely blocked poly(ADP-ribosyl)ation (data not shown). It subse- quently blocked access of the substrate to BER pro- teins, as we were unable to cross-link XRCC1 or Pol b under these conditions (Fig. 2C). These data suggest that PARP-1 binding to repair intermediates is always involved in BER and that poly(ADP-ribosyl)ation- dependent dissociation of PARP-1 is required for further BER progression. PARP-1 is not essential for binding of XRCC1 and Polb to damaged DNA As PARP-1 is the first protein to interact with a nicked AP site during BER, it was interesting to test whether XRCC1–DNA ligase IIIa and Pol b binding to DNA is specifically associated ⁄ affected by PARP-1 and poly(ADP-ribosyl)ation. To test this, we generated HeLa WCE depleted of PARP-1 by using 3-amino- benzamide-Affigel beads. Using immunoblotting analy- sis, we showed that, compared with the original HeLa extract (Fig. 3A, WCE), the PARP-1-depleted extracts were  95% devoid of PARP-1 (Fig. 3A, I.D.) without significantly affecting the concentrations of other BER proteins, such as Pol b and XRCC1 (data not shown). On incubation of the uracil substrate with PARP-1- depleted extracts, Pol b and XRCC1 are still com- plexed to DNA (Fig. 3B), although the concentrations peak at earlier time points than in the presence of PARP-1 and the amounts of protein subsequently plat- eau, as observed from 30 s up to 2 min. As a result of faster binding, both Pol b and XRCC1 dissociate ear- lier from DNA. There is also no significant formation of poly(ADP-ribose) polymers in the absence of PARP-1. We thus conclude that the critical event in an in vitro BER reaction is poly(ADP-ribosyl)ation and dissociation of PARP-1, rather than interaction of BER proteins with PARP-1. PARP-1 binding protects repair intermediates from deterioration It was previously proposed that PARP-1 binding may protect DNA strand breaks from nuclease attack when BER enzymes are a limiting factor in repair [24]. To test this hypothesis, we compared PARP-1 involvement in the repair of a substrate containing either a 3¢-OH and 5¢-phosphate strand break or a 3¢-phosphate and 5¢-OH strand break. The latter substrate requires poly- nucleotide kinase to dephosphorylate the 3¢ end and to phosphorylate the 5¢ end before ligation and is A B C Fig. 2. Involvement of PARP-1 in BER in human cell extracts. (A) A uracil-containing (left panel), or the corresponding control (right panel) biotinylated hairpin substrate was bound to magnetic strept- avidin beads before incubation with 100 lg HeLa WCE. After incu- bation, proteins were cross-linked to DNA using 0.5% (v ⁄ v) formaldehyde, and the beads subsequently washed. Cross-links were reversed, proteins separated by SDS ⁄ PAGE (10% gel), trans- ferred to PVDF membranes and analysed by immunoblotting with the indicated antibodies. (B) The membrane in (A) was stripped and reprobed with antibodies raised against poly(ADP-ribose) polymers (PAR). (C) A uracil-containing biotinylated hairpin substrate was bound to magnetic streptavidin beads before incubation with 100 lg HeLa WCE in the presence of the poly(ADP-ribosyl)ation inhibitor 3-aminobenzamide (1 m M). After the times indicated, pro- teins were cross-linked to DNA and the beads subsequently washed. Cross-links were reversed and proteins separated by SDS ⁄ PAGE (10% gel), transferred to PVDF membranes and ana- lysed by immunoblotting with the indicated antibodies. Time zero equates to cross-linking immediately after extract addition. Role of PARP-1 in base excision repair J. L. Parsons et al. 2014 FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS expected to be repaired much more slowly than a frank DNA nick. Subsequently using 5¢-end-labelled substrates, we showed that repair of the 3¢-OH and 5¢-phosphate strand break is accomplished within 4 min, whereas only a small fraction of the 3¢-phosphate and 5¢-OH strand break was processed after 8 min (Fig. 4A,B). Correspondingly, repair of the latter sub- strate involves more PARP-1 binding and more poly- (ADP-ribosyl)ation, as observed in the formaldehyde cross-linking assay (compare Fig. 4C,D). Therefore, we propose that persisting SSBs can cause several cycles of PARP-1 binding and dissociation. We further spe- culate that this repetitive binding of PARP-1 is required only when the number of unrepaired strand breaks exceeds the amount of rate-limiting repair enzyme, and it may play a protective role masking the strand break from nuclease attack. To further demon- strate the protective role of PARP-1, we blocked repair of a uracil-containing oligonucleotide substrate by removing deoxyribonucleotide triphosphates from the reaction mixture. Under these conditions, repair is blocked at the stage of a one-nucleotide gap created by the sequential action of uracil-DNA glycosylase, APE1 and Pol b, and PARP-1 is continuously poly (ADP-ribosyl)ated (Figs 5A and 6A). As the repair gap is not filled, the 18-mer repair intermediate is accu- mulated (Figs 5B and 6B). This repair intermediate can be attacked by cellular nucleases if not protected by PARP-1. Indeed, we observed an increased rate of degradation of the 18-mer fragment in both PARP- depleted cell extract (Fig. 5B,C) and cell extract prepared from PARP-knockout cells (Fig. 6B,C) com- pared with the original HeLa cell extract or cell extract prepared from wild-type mouse cells, respectively. Furthermore, addition of purified PARP-1 protein to PARP-1-deficient cell extracts restored the stability of the 18-mer repair intermediate. Discussion We have used an in vitro cross-linking assay to study the role of PARP-1 protein during BER. We find that PARP-1 binds to the incised AP site at the very early stages of BER, as has been previously observed by using photoaffinity cross-linking during repair in cell extract [26]. We also find that inhibition of poly(ADP- ribosyl)ation by 3-aminobenzamide blocks PARP-1 dissociation and completely prevents further repair. These data suggest that PARP-1 is always an integral part of the BER process and that processing of uracil in human cell extracts may be divided into two major steps: removal of the damaged base and incision of the generated AP site as a first step and processing of the incised AP site as a second step. These steps are clearly marked by PARP-1 binding (Fig. 7). This is in agree- ment with previously published data [23,24] that suggest that BER is NAD + -dependent and that poly(ADP-ribosyl)ation is required for PARP-1 disso- ciation and subsequent BER progression. It should be noted that PARP-1 antibodies are very specific and would not recognize poly(ADP-ribosyl)ated PARP-1. Therefore, the decreased binding of PARP-1 observed in Fig. 2A may be interpreted as either PARP-1 disso- ciation or poly(ADP-ribosyl)ation. However, cross- linking dynamics in the same reaction support the model in which PARP is being poly(ADP-ribosyl)ated first and then dissociates from DNA. The level of damage-specific PARP-1 cross-linking was maximal at 0.5 min (binding) and then decreased after 2 min of the repair reaction [poly(ADP-ribosyl)ation and disso- ciation from DNA]. In contrast, the presence of other A B Fig. 3. PARP-1 is not required for assembly of BER proteins on damaged DNA in human cell extracts. (A) HeLa WCE was depleted of PARP-1 with 3-aminobenzamide-Affigel and confirmed by analy- sing 30 lg of the corresponding extracts by SDS ⁄ PAGE (10% gel) and immunoblotting with PARP-1 antibodies. (B) A uracil-containing biotinylated hairpin substrate was bound to streptavidin magnetic beadsbeforeincubationwith100lg HeLa WCE depleted of PARP-1. After the times indicated, proteins were cross-linked to DNA and the beads subsequently washed. Cross-links were reversed and proteins separated by SDS ⁄ PAGE (10% gel), transferred to PVDF membranes and analysed by immunoblotting with the indicated antibodies. The membrane was subsequently stripped and reprobed with antibodies raised against poly(ADP-ribose) polymers (PAR). J. L. Parsons et al. Role of PARP-1 in base excision repair FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS 2015 BER proteins was maximal at 4 min (Fig. 2A, lane 6). This suggests that when there are sufficient amounts of repair enzymes present, they efficiently remove PARP-1 from nicked DNA. In agreement with the previous finding that PARP-1 deficiency is not decreasing the rate of short-patch BER [20,23,27,28], we also did not find any effect of PARP-1 depletion on the cross-link- ing efficiency of BER proteins during repair of a ura- cil-containing substrate. Instead, removal of PARP-1 accelerated the loading and subsequent dissociation of XRCC1–DNA ligase IIIa and Pol b on the nicked DNA (compare Figs 2A and 3B). This finding is in a good agreement with previous observations of acceler- ated BER in PARP-1-deficient cell extracts [23]. However, the dispensability of PARP-1 from an in vitro BER assay may not correctly reflect the role of PARP-1 in cellular BER. Numerous studies have indi- cated that PARP-1 may play an important role in living cells, as PARP-1-deficient cells are genetically unstable and sensitive to DNA-damaging agents [16– 18]. A mechanism by which PARP-1 actively recruits repair proteins to the site of the strand break may be applicable to an in vivo situation in which the task of strand-break identification and accumulation of repair enzymes needed for repair is extremely difficult consid- ering the size of the human genome. Indeed, poly- (ADP-ribosyl)ation-dependent accumulation of XRCC1 in repair foci after oxidative DNA damage and during SSB repair has recently been reported [21,22]. There- fore, most probably the purpose of poly(ADP-ribo- syl)ation of PARP-1 is not only to allow PARP-1 to dissociate from the repair site but also to attract BER proteins. In support of this idea, it has been shown that DNA ligase III and XRCC1 proteins have poly- (ADP-ribose)-binding motifs and that they and Pol b have also been shown to preferentially interact with poly(ADP-ribosyl)ated PARP [19,29,30]. We also find that, when repair is inefficient, unmodified PARP-1 can rebind to the substrate con- taining the strand break and probably accomplish sev- eral cycles if enough NAD + is provided. We further speculate that this rebinding plays a role in protecting AB CD Fig. 4. Repair of 5¢ or 3¢-phosphorylated nick-containing oligonucleotides and analysis of PARP-1 binding and poly(ADP-ribosyl)ation by cross- linking. A 5¢-phosphorylated (A) or a 3¢-phosphorylated (B) nick-containing hairpin substrate was 5¢-end-labelled with [ 32 P]ATP[cP], bound to magnetic streptavidin beads, and incubated with 100 lg HeLa cell extract for the times indicated. DNA–beads were subsequently purified and washed, and formamide loading dye added. DNA was separated by SDS ⁄ PAGE (10% gel). The phosphorimage of the corresponding gels is shown. The 5¢-phosphorylated (C) or the 3¢-phosphorylated (D) nick-containing hairpin substrates were also bound to streptavidin mag- netic beads before incubation with HeLa WCE for the times indicated and subsequent cross-linking with 0.5% formaldehyde. The beads were subsequently washed, the cross-links reversed, and the proteins separated by SDS ⁄ PAGE (10% gel), transferred to PVDF membranes, and analysed by immunoblotting with the indicated antibodies. Time zero equates to cross-linking immediately after extract addition. Role of PARP-1 in base excision repair J. L. Parsons et al. 2016 FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS strand-break-containing DNA from nuclease attack. Indeed we find that when repair is blocked, the degra- dation of repair intermediates is more intensive in PARP-1-deficient cell extracts, although we find that human cell extracts are more responsive to PARP-1 deficiency than mouse cells. In conclusion, our study suggests that PARP-1 is always involved in BER of DNA base lesions, and SSB and is important for pre- venting degradation of excessive unrepaired DNA strand breaks by cellular nucleases. However, the end- protecting function may be just one of the multiple functions of PARP-1 in DNA metabolism. Experimental procedures Materials Synthetic oligodeoxyribonucleotides were purchased from MWG-Biotech (Ebersberg, Germany) and gel purified on a 20% polyacrylamide gel. Streptavidin magnetic beads and magnetic separation rack were purchased from New England Biolabs (Beverly, MA, USA). Recombinant human PARP-1 was obtained from Alexis Biochemicals (Notting- ham, UK). Antibodies XRCC1 (ab144) and DNA ligase III (ab587) antibodies were purchased from Abcam Ltd (Cambridge, UK) PARP-1 antibodies (C2-10) were purchased from Alexis Corpora- tion Ltd, and antibodies raised against poly(ADP-ribose) polymers were purchased from Trevigen (Gaithersburg, MD, USA). Antibodies against rat Pol b were raised in rabbit and affinity purified as described [31]. Cell extracts Mouse embryonic fibroblasts derived from normal and PARP-1 knockout mice were kindly provided by G. de Murcia (ESBS-CNRS, Strasbourg, France). Cells were maintained in Dulbecco’s modified Eagle’s medium supple- mented with 10% (v ⁄ v) fetal bovine serum and antibiotics. A B C Fig. 5. Increased degradation of repair inter- mediates in PARP-depleted cell extracts. (A) HeLa WCE was incubated with a FAM-label- led oligonucleotide duplex in the absence of deoxyribonucleotide triphosphates for the times indicated, and aliquots of the reaction mixture were separated by SDS ⁄ PAGE (10% gel), transferred to PVDF membranes, and analysed by immunoblotting with poly(ADP-ribose) polymer (PAR) antibodies. (B) HeLa and PARP-1-depleted HeLa WCE was incubated with a FAM-labelled oligonucleotide duplex in the absence of deoxyribonucleotide triphosphates for the times indicated before the addition of formamide loading dye. PARP-1-depleted HeLa WCE was also complemented with 100 ng PARP-1 protein and incubated for 6 min before separation of the DNA by SDS ⁄ PAGE (20% gel). (C) The oligonucleotide fragments were analysed using Quantity One software, which indicates the relative density of the fragments produced. J. L. Parsons et al. Role of PARP-1 in base excision repair FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS 2017 HeLa cell pellets were purchased from Paragon (Aspen, CO, USA). WCEs were prepared by the method of Manley et al. [32] and dialysed overnight against buffer containing 25 mm Hepes ⁄ KOH, pH 7.9, 100 mm KCl, 12 mm MgCl 2 , 0.1 mm EDTA, 17% glycerol and 2 mm dithiothreitol. Extracts were divided into aliquots and stored at )80 °C. Cross-linking assay Streptavidin magnetic beads were blocked in 5% (v ⁄ v) non- fat milk and subsequently washed with Binding buffer (20 mm Tris ⁄ HCl, pH 7.5, 0.5 m NaCl, 1 mm EDTA) using the magnetic separator rack. Beads were then incubated with hairpin substrates containing a 3¢-biotinylated moiety, at room temperature with agitation for 30 min in Binding buffer. The DNA–beads were subsequently washed with Wash buffer [25 mm Hepes, pH 7.9, 100 mm KCl, 12 mm MgCl 2 ,1mm EDTA, 5% (v ⁄ v) glycerol and 2 mm dithio- threitol]. The DNA–beads (250 fmol DNA per reaction) were then included in a reaction containing 100 lg HeLa cell extract in 50 lL buffer containing 50 m m Hepes ⁄ KOH, pH 7.8, 50 mm KCl, 10 mm MgCl 2 , 0.5 mm EDTA, 1.5 mm dithiothreitol, 2.5% (v ⁄ v) glycerol, 20 lm dCTP, 20 lm dATP, 20 lm dGTP, 20 lm dTTP, 2 mm ATP, 25 mm phosphocreatine (diTris salt; Sigma), 2.5 lg creatine phosphokinase (type I; Sigma), 0.25 mm NAD + and 1 lg carrier DNA (single-stranded 30-mer oligonucleotide). Reactions were incubated for the time indicated at 30 °C before cross-linking with formaldehyde (0.5%, final concen- tration) for a further 10 min at 30 °C (time zero equates to cross-linking immediately after extract addition). The beads were washed twice with 50 lL Wash buffer and resuspend- ed in 20 lL SDS ⁄ PAGE sample buffer [25 mm Tris ⁄ HCl, pH 6.8, 2.5% (v ⁄ v) 2-mercaptoethanol, 1% (w ⁄ v) SDS, 5% (v ⁄ v) glycerol, 1 mm EDTA, 0.15 mgÆmL )1 bromophenol blue]. Cross-links were reversed by heating for at least 2 h at 65 °C, and proteins were separated on a SDS ⁄ 10% poly- acrylamide gel followed by transfer to a poly(vinylidene difluoride) (PVDF) membrane and immunoblot analysis with the indicated affinity-purified antibodies. For direct comparison, proteins cross-linked from different substrates were analysed on the same immunoblot. Extracts were also preincubated with 3-aminobenzamide (1 mm) for 20 min before the addition of DNA–beads and subsequent cross- linking for studies investigating PARP-1 poly(ADP-ribo- syl)ation inhibition. Formaldehyde cross-linking is essential A B C Fig. 6. Increased degradation of repair inter- mediates in PARP-deficient cell extracts. (A) PARP-1 + ⁄ + WCE was incubated with a FAM-labelled oligonucleotide duplex in the absence of deoxyribonucleotide triphosphates for the times indicated. Aliquots of the reac- tion mixture were separated by SDS ⁄ PAGE (10% gel), transferred to PVDF membranes, and analysed by immunoblotting with PAR antibodies. (B) PARP-1 + ⁄ + and PARP-1 – ⁄ – WCE were incubated with a FAM-labelled oligonucleotide duplex in the absence of deoxyribonucleotide triphosphates for the times indicated before the addition of forma- mide loading dye. PARP-1 – ⁄ – WCE was also complemented with 100 ng PARP-1 protein and incubated for 6 min before separation of the DNA by SDS ⁄ PAGE (20% gel). (C) The oligonucleotide fragments were analysed using Quantity One software, which indicates the relative density of the fragments produced. Role of PARP-1 in base excision repair J. L. Parsons et al. 2018 FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS for detecting protein interactions with DNA-bound strept- avidin magnetic beads, as no proteins were pulled down without formaldehyde treatment. Depletion of PARP-1 from HeLa WCE with 3-aminobenzamide-Affigel 3-Aminobenzamide was covalently attached to Affigel (Bio-Rad) by the method of Ushiro et al. [33]. Then 0.5 mL 3-aminobenzamide-Affigel (50% slurry) was washed three times with 5 mL buffer containing 50 mm Hepes ⁄ KOH, pH 7.8, 50 mm KCl, 10 mm MgCl 2 , 0.5 mm EDTA, 1 mm dithiothreitol, 17% (v ⁄ v) glycerol, 100 lgÆmL )1 BSA and protease inhibitors (1 lgÆmL )1 each of chymostatin, pepstatin and leupeptin and 1 mm phenyl- methanesulfonyl fluoride). Then 1 mL WCE (17 mgÆmL )1 ) was mixed with 0.5 mL 3-aminobenzamide-Affigel and incubated overnight at 4 °C. Affigel beads were removed by centrifugation, and the extract was divided into aliquots and stored at )80 °C. Repair assays on streptavidin beads Oligonucleotides were 5¢-end-labelled with [ 32 P]ATP[cP] using T4-polynucleotide kinase, and unincorporated label was removed on a Sephadex G-25 spin column. To prepare the hairpin substrates, the oligonucleotides were incubated at 90 °C for 3–5 min before slow cooling to room tempera- ture. Substrates were then bound to streptavidin beads as described above before determination of repair capacity of HeLa WCE. Streptavidin beads bound with radiolabelled hairpin substrates (250 fmol DNA per reaction) were incu- bated with 100 lg HeLa WCE in 50 lL Reaction buffer at 37 °C for the time indicated before addition of 12.5 lL 500 mm EDTA to stop the reaction. Beads were washed twice with 50 lL10mm Tris ⁄ HCl (pH 8.0), 100 mm EDTA and resuspended in 20 lL formamide loading dye. DNA was separated on a 10% denaturing polyacrylamide gel, and the gel exposed to intensifying screens at 4 °C before analysis by phosphorimaging. Oligonucleotide degradation in WCE Reactions were reconstituted in a reaction mixture (10 lL) that contained 45 mm Hepes, pH 7.8, 70 mm KCl, 7.5 mm MgCl 2 , 0.5 mm EDTA, 1 mm dithiothreitol, 2 mm ATP, 2mgÆmL )1 BSA and a FAM (6-carboxyfluorescein)-labelled 30-mer oligonucleotide duplex containing a uracil residue at position 19 ( 5 ng, 500 fmol), in the absence of dNTPs to prevent repair of incised AP sites. The reactions were initi- ated by addition of 20 lg cell extract and incubated for the indicated time at 37 °C. When PARP-1-depleted or PARP- deficient WCEs were complemented with purified recombin- ant human PARP-1 protein, 100 ng (890 fmol) PARP-1 protein was added before incubation. The reactions were stopped by the addition of 10 lL gel loading buffer. After incubation at 90 °C for 3 min, the reaction products were separated by electrophoresis on a 20% denaturing poly- acrylamide gel. Acknowledgements We are grateful to Gilbert de Murcia for providing PARP-1-knockout cells. References 1 Barnes DE & Lindahl T (2004) Repair and genetic consequences of endogenous DNA base damage in mammalian cells. Annu Rev Genet 38, 445–476. 2 Robson CN & Hickson ID (1991) Isolation of cDNA clones encoding a human apurinic ⁄ apyrimidinic Fig. 7. Involvement of PARP-1 in BER. After processing of the DNA lesion by a DNA glycosylase and APE, the strand break thus revealed is recognized by PARP-1 dimer. After NAD + -dependent poly(ADP-ribosyl)ation of PARP-1, automodified PARP-1 then disso- ciates from DNA. The release of automodified PARP-1 exposes the intermediate to Pol b and XRCC1–DNA ligase IIIa, which perform the remainder of the repair process. If not enough repair enzymes are available, the PARP-binding cycle is repeated. J. L. Parsons et al. Role of PARP-1 in base excision repair FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS 2019 endonuclease that corrects DNA repair and mutagenesis defects in E. colixth (exonuclease III) mutants. Nucleic Acids Res 19, 5519–5523. 3 Demple B, Herman T & Chem DS (1991) Cloning and expression of APE1, the cDNA encoding the major human apurinic endonuclease: definition of a family of DNA repair enzymes. Proc Natl Acad Sci USA 88, 11450–11454. 4 Dianov G, Price A & Lindahl T (1992) Generation of single-nucleotide repair patches following excision of uracil residues from DNA. Mol Cell Biol 12, 1605– 1612. 5 Matsumoto Y & Kim K (1995) Excision of deoxyribose phosphate residues by DNA polymerase b during DNA repair. Science 269, 699–702. 6 Prasad R, Beard WA, Strauss PR & Wilson SH (1998) Human DNA polymerase beta deoxyribose phosphate lyase: substrate specificity and catalytic mechanism. J Biol Chem 273, 15263–15270. 7 Caldecott KW, Aoufouchi S, Johnson P & Shall S (1996) XRCC1 polypeptide interacts with DNA poly- merase beta and possibly poly(ADP-ribose) polymerase, and DNA ligase III is a novel molecular ‘nick-sensor’ in vitro. Nucleic Acids Res 24, 4387–4394. 8 Nash RA, Caldecott KW, Barnes DE & Lindahl T (1997) XRCC1 protein interacts with one of two dis- tinct forms of DNA ligase III. Biochemistry 36, 5207– 5211. 9 Cappelli E, Taylor R, Cevasco M, Abbondandolo A, Caldecott K & Frosina G (1997) Involvement of XRCC1 and DNA ligase III gene products in DNA base excision repair. J Biol Chem 272, 23970–23975. 10 Parsons JL, Dianova II & Dianov GL (2004) APE1 is the major 3¢-phosphoglycolate activity in human cell extracts. Nucleic Acids Res 32, 3531–3536. 11 Caldecott KW (2003) XRCC1 and DNA strand break repair. DNA Repair (Amst) 2, 955–969. 12 Whitehouse CJ, Taylor RM, Thistlethwaite A, Zhang H, Karimi-Busheri F, Lasko DD, Weinfeld M & Caldecott KW (2001) XRCC1 stimulates human polynucleotide kinase activity at damaged DNA termini and accelerates DNA single-strand break repair. Cell 104, 107–117. 13 Wiederhold L, Leppard JB, Kedar P, Karimi-Busheri F, Rasouli-Nia A, Weinfeld M, Tomkinson AE, Izumi T, Prasad R, Wilson SH, et al. (2004) AP endonuclease- independent DNA base excision repair in human cells. Mol Cell 15, 209–220. 14 Dantzer F, Schreiber V, Niedergang C, Trucco C, Flat- ter E, De La Rubia G, Oliver J, Rolli V, Menissier-de Murcia J & de Murcia G (1999) Involvement of poly(ADP-ribose) polymerase in base excision repair. Biochimie 81, 69–75. 15 D’Amours D, Desnoyers S, D’Silva I & Poirier GG (1999) Poly(ADP-ribosyl)ation reactions in the regula- tion of nuclear functions. Biochem J 324, 249–268. 16 Me ´ nessier-de Murcia J, Niedergang C, Trucco C, Ricoul M, Dutrillaux B, Mark M, Oliver FJ, Masson M, Dierich A, LeMeur M. et al. (1997) Requirement of poly(ADP-ribose) polymerase in recovery from DNA damage in mice and in cells. Proc Nat Acad Sci USA 94, 7303–7307. 17 Simbulan-Rosenthal CM, Haddad BR, Rosenthal DS, Weaver Z, Coleman A, Luo RB, Young HM, Wang ZQ, Ried T & Smulson ME (1999) Chromosomal aber- rations in PARP (- ⁄ -) mice: genome stabilization in immortalized cells by reintroduction of poly(ADP- ribose) polymerase cDNA. Proc Natl Acad Sci USA 96, 13191–13196. 18 Wang ZQ, Stingl L, Morrison C, Jantsch M, Los M, Schulze-Osthoff K & Wagner EF (1997) PARP is important for genomic stability but dispensable in apop- tosis. Genes Dev 11, 2347–2358. 19 Masson M, Niedergang C, Schreiber V, Muller S, Men- issier-de Murcia J & de Murcia G (1998) XRCC1 is spe- cifically associated with poly(ADP-ribose) polymerase and negatively regulates its activity following DNA damage. Mol Cell Biol 18, 3563–3571. 20 Dantzer F, de la Rubia G, Murcia JMD, Hostomsky Z, de Murcia G & Schreiber V (2000) Base excision repair is impaired in mammalian cells lacking poly(ADP- ribose) polymerase-1. Biochemistry 39, 7559–7569. 21 El-Khamisy SF, Masutani M, Suzuki H & Caldecott KW (2003) A requirement for PARP-1 for the assembly or stability of XRCC1 nuclear foci at sites of oxidative DNA damage. Nucleic Acids Res 31, 5526–5533. 22 Okano S, Lan L, Caldecott KW, Mori T & Yasui A (2003) Spatial and temporal cellular responses to single- strand breaks in human cells. Mol Cell Biol 23, 3974– 3981. 23 Allinson SL, Dianova II & Dianov GL (2003) Poly(ADP-ribose) polymerase in base excision repair: always engaged, but not essential for DNA damage processing. Acta Biochim Pol 50, 169–179. 24 Satoh MS & Lindahl T (1992) Role of poly(ADP- ribose) formation in DNA repair. Nature 356, 356–358. 25 Parsons JL & Dianov GL (2004) Monitoring base exci- sion repair proteins on damaged DNA using human cell extracts. Biochem Soc Trans 32, 962–963. 26 Lavrik OI, Prasad R, Sobol RW, Horton JK, Acker- man EJ & Wilson SH (2001) Photoaffinity labeling of mouse fibroblast enzymes by a base excision repair intermediate. Evidence for the role of poly(ADP- ribose) polymerase-1 in DNA repair. J Biol Chem 276, 25541–25548. 27 Vodenicharov MD, Sallmann FR, Satoh MS & Poirier GG (2000) Base excision repair is efficient in cells lack- ing poly(ADP-ribose) polymerase 1. Nucleic Acids Res 28, 3887–3896. 28 Sanderson RJ & Lindahl T (2002) Down-regulation of DNA repair synthesis at DNA single-strand interruptions Role of PARP-1 in base excision repair J. L. Parsons et al. 2020 FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS in poly(ADP-ribose) polymerase-1 deficient murine cell extracts. DNA Repair 1, 547–558. 29 Leppard JB, Dong Z, Mackey ZB & Tomkinson AE (2003) Physical and functional interaction between DNA ligase III alpha and poly (ADP-Ribose) polymer- ase 1 in DNA single-strand break repair. Mol Cell Biol 23, 5919–5927. 30 Pleschke JM, Kleczkowska HE, Strohm M & Althaus FR (2000) Poly(ADP-ribose) binds to specific domains in DNA damage checkpoint proteins. J Biol Chem 275, 40974–40980. 31 Dianova II, Bohr VA & Dianov GL (2001) Interaction of human AP endonuclease 1 with flap endonuclease 1 and proliferating cell nuclear antigen involved in long- patch base excision repair. Biochemistry 40, 12639– 12644. 32 Manley JL, Fire A, Samuels M & Sharp PA (1983) In vitro transcription: whole-cell extract. Methods Enzymol 101, 568–582. 33 Ushiro H, Yokoyama Y & Shizuta Y (1987) Purifica- tion and characterization of poly(ADP-ribose) synthe- tase from human placenta. J Biol Chem 262, 2352–2357. J. L. Parsons et al. Role of PARP-1 in base excision repair FEBS Journal 272 (2005) 2012–2021 ª 2005 FEBS 2021 . Poly(ADP-ribose) polymerase-1 protects excessive DNA strand breaks from deterioration during repair in human cell extracts Jason L. Parsons, Irina. participate in processing of DNA single -strand breaks (SSB) and one-nucleotide gaps: Pol b is involved in gap filling, XRCC1 DNA ligase IIIa complex is involved in