Molecular Evidence of the Involvement of the Nucleotide Excision Repair

Mar 24, 2007 - Molecular Evidence of the Involvement of the Nucleotide Excision Repair (NER) System in the Repair of the Mono(ADP-Ribosyl)ated DNA ...
0 downloads 0 Views 277KB Size
Download PDF
694

Chem. Res. Toxicol. 2007, 20, 694-700

Molecular Evidence of the Involvement of the Nucleotide Excision Repair (NER) System in the Repair of the Mono(ADP-Ribosyl)ated DNA Adduct Produced by Pierisin-1, an Apoptosis-Inducing Protein from the Cabbage Butterfly Masanobu Kawanishi,† Kazuki Matsukawa,† Isao Kuraoka,‡,⊥ Takeji Takamura-Enya,§,# Yukari Totsuka,§ Yasuko Matsumoto,§ Masahiko Watanabe,§,∇ Yue Zou,| Kiyoji Tanaka,‡ Takashi Sugimura,§ Keiji Wakabayashi,§ and Takashi Yagi*,† EnVironmental Genetics Laboratory, Frontier Science InnoVation Center, Osaka Prefecture UniVersity, 1-2 Gakuen-cho, Sakai, Osaka 599-8570, Japan, Graduate School of Frontier Bioscience, Osaka UniVersity, 1-3 Yamada-oka, Suita, Osaka 565-0871, Japan, Cancer PreVention Basic Research Project, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan, and Department of Biochemistry and Molecular Biology, Quillen College of Medicine, East Tennessee State UniVersity, Johnson City, Tennessee 37614 ReceiVed December 21, 2006

Pierisin-1 is a potent apoptosis-inducing protein found in the pupal extract of the cabbage white butterfly. Pierisin-1 catalyzes the mono(ADP-ribosyl)ation of the 2′-deoxyguanosine residue and produces a bulky adduct, N2-(ADP-ribos-1-yl)-2′-deoxyguanosine (N2-ADPR-dG) in DNA. Here, we examined the involvement of the nucleotide excision repair (NER) system in the removal of N2-ADPR-dG in Escherichia coli (E. coli) and human cells. The results of mobility shift gel electrophoresis assays using a 50-mer oligodeoxynucleotide containing a single N2-ADPR-dG showed that E. coli UvrAB proteins bound to the N2-ADPR-dG in Vitro. Incubation of the adducted oligodeoxynucleotides with UvrABC resulted in the incision of the oligonucleotides in Vitro. The results of filter binding and gel mobility shift assays using human XPA protein showed that XPA bound to DNA containing N2-ADPR-dGs in Vitro. Finally, we introduced plasmids containing N2-ADPR-dGs into E. coli and human cells. N2-ADPR-adducted plasmids replicated l0 times and 20 times less efficiently in NER-deficient E. coli and human cells than in their wild-type counterparts, respectively. More mutations were induced in the plasmid propagated in NER-deficient cells than that in wild-type human cells. These results indicate the involvement of the NER system in the repair of N2-ADPR-dG in both E. coli and human cells. Introduction Pierisin-1 is a 98-kDa protein, abundant in the fifth instar larvae and the early phase pupae of the cabbage white butterfly, Pieris rapae (1-4). This protein induces apoptotic cell death in various human tumor cell lines (1, 2, 5, 6). Molecular cloning of the complementary DNA (cDNA) of pierisin-1 from Pieris rapae revealed that the protein consists of 850 amino acids and that its 27-kDa N-terminal portion shares sequence homology with the enzymatic units of ADP-ribosylating toxins, including the A-subunit of the cholera toxin (7). Remarkably, the target molecule for ADP-ribosylation by pierisin-1 is DNA, not proteins, thus providing a sharp contrast to bacteria-derived ADP-ribosylating toxins such as the cholera toxin, pertussis toxin, and diphtheria toxin (8). Pierisin-1 efficiently catalyzes the mono(ADP-ribosyl)ation of double-stranded DNA with β-NAD (8). The ADP-ribose moiety of β-NAD is transferred * To whom correspondence should be addressed. Phone: +81-(0)72254-9862. Fax: +81-(0)72-254-9938. E-mail: [email protected]. † Osaka Prefecture University. ‡ Osaka University. § National Cancer Center Research Institute. | East Tennessee State University. ⊥ Present address: Cancer Biochemistry Laboratory, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 811-1395, Japan. # Present address: Department of Applied Chemistry, Kanagawa Institute of Technology, 1030 Shimo-ogino, Atsugi 243-0292, Japan. ∇ Present address: School of Pharmacy, Shujitsu University, 1-6-1 Nishigawara, Okayama, Okayama 703-8516, Japan.

by pierisin-1 to the amino group at the N2 position of deoxyguanosine (dG), and a bulky adduct, N2-(ADP-ribos-1yl)-2′-deoxyguanosine (N2-ADPR-dG1; shown in Figure 1), is produced (8). Using a 32P-postlabeling method, N2-ADPR-dG was detected in DNA from pierisin-1-treated HeLa cells (8). Pierisin-1 is, therefore, a unique ADP-ribosyltransferase targeting the exocyclic amino group of dG residues of DNA, and the resulting product, N2-ADPR-dG, is a novel DNA adduct. However, because pierisin-1 is found only in Pieris butterflies (1), the presence of N2-ADPR-dG in other organisms is under investigation now. DNA lesions frequently lead to mutagenesis and carcinogenesis. Mutational analyses with the hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus in Chinese hamster lung (CHL) cells or the supF gene in the shuttle vector plasmid pMY189 as the target marker gene revealed that pierisin-1 enhanced G/C to C/G and G/C to T/A base substitutions (9). Mutations are often generated during translesion DNA synthesis with several DNA polymerases. A primer extension study in Vitro with mammalian DNA polymerases (Pol) η, ι, and κ using 1 Abbreviations: N2-ADPR-dG, N2-(ADP-ribos-1-yl)-2′-deoxyguanosine; HPRT, hypoxanthine-guanine phosphoribosyltransferase; CHL, Chinese hamster lung; NER, nucleotide excision repair; XPA, xeroderma pigmentosum A; X-gal, 5-bromo-4-chloro-3-indoyl-β-D-galactoside; IPTG, isopropyl-β-D-thiogalactoside; GST, Glutathione-S-transferase; XP-A, group A xeroderma pigmentosum patient; CPDs, cyclobutane pyrimidine dimers; 6-4PPs, 6-4 photoproducts; CHO, Chinese hamster ovary.

10.1021/tx600360b CCC: $37.00 © 2007 American Chemical Society Published on Web 03/24/2007

Repair of an ADP-Ribosylated DNA Adduct

Chem. Res. Toxicol., Vol. 20, No. 4, 2007 695

Figure 1. Pierisin-1-catalyzed mono(ADP-ribosyl)ation of 2′-deoxyguanosine and the resulting adduct N2-(ADP-ribos-1-yl)-2′-deoxyguanosine (N2-ADPR-dG).

a site-specifically modified oligodeoxynucleotide containing a single N2-ADPR-dG revealed that Pol κ bypassed N2-ADPRdG by preferentially inserting dCMP opposite the lesion and extended DNA synthesis from a mismatched primer terminus where dG, dA, or dT had been placed opposite the lesion. Pol ι incorporated dCMP and dTMP opposite N2-ADPR-dG but could not extend the DNA strand past the lesion. Pol η inserted dCMP and extended the DNA strand poorly past the lesion (10). An in Vitro study also indicated that N2-ADPR-dG inhibited DNA replication by Pol R and Pol β (11). Nucleotide excision repair (NER) is a pathway by which cells remove bulky DNA adducts induced by UV light and chemicals. It consists of a simple “cut and patch” system that operates by excision of a short single-stranded DNA fragment, followed by restoration of the duplex DNA structure through repair synthesis (12). In Escherichia coli (E. coli), UvrA, B, and C proteins play an important role in DNA damage recognition and excision in NER (13). In mammalian cells, xeroderma pigmentosum A (XPA) protein, which preferentially binds to distorted DNA molecules, acts as a bridging factor between the recognition intermediate and the ultimate excision complex (12). We report here that E. coli repair proteins UvrAB recognize and UvrABC excise N2-ADPR-dG. The results of filter binding and gel mobility shift assays showed that XPA protein binds to N2-ADPR-dG. Furthermore, we observed lower replication efficiency and higher mutation frequency of N2-ADPR-adducted plasmids in NER-deficient human cells (XP-A cells) and E. coli cells than in their wild-type counterparts.

Experimental Procedures Materials. Ampicillin, 5-bromo-4-chloro-3-indoyl-β-D-galactoside (X-gal), chloramphenicol, isopropyl-β-D-thiogalactoside (IPTG), nalidixic acid, and β-NAD were obtained from Sigma-Aldrich Japan (Tokyo, Japan). Restriction endonuclease DpnI, T4 DNA ligase, and calf intestine alkaline phosphatase were purchased from New England Biolabs (Beverly, MA). T4 polynucleotide kinase and restriction endonuclease BssHI were obtained from Toyobo Co., Ltd. (Osaka, Japan). Pronase was purchased from Kaken Pharmaceutical Co., Ltd. (Tokyo, Japan). T7 DNA polymerase was obtained from United States Biochemical Corp. (Cleveland, OH). [γ-32P]ATP (3,000 Ci/mmol) and [R-32P]dCTP (3,000 Ci/mmol) were obtained from Amersham Biosciences, Ltd. (Buckinghamshire, U.K.). Micrococcal nuclease and phosphodiesterase II (bovine spleen) were obtained from Worthington Biochemical Co. (Lakewood, NJ), and calf intestine alkaline phosphatase was from Takara Bio Co. (Kyoto, Japan). All oligonucleotides used in this study were of HPLC grade and purchased from Sigma Genosys Japan (Tokyo, Japan). UvrA, UvrB, and UvrC proteins were purified as described previously (13). XPA protein and Glutathione-S-transferase (GST)XPA fusion protein were prepared as described (14). Anti-GST antibody was obtained from GE Healthcare Bio-Sciences Co. (Piscataway, NJ). Human Cells. SV40-transformed normal human fibroblast cells, WI38-VA13 (15), obtained from the American Type Culture Collection (Rockville, MD), and DNA repair-deficient cells,

XP2OS(SV), established from a Japanese group A xeroderma pigmentosum patient (XP-A) (16), were used. All cells were cultured in Dulbecco’s modified minimum essential medium (Nikken, Kyoto, Japan) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT). Preparation of Mono(ADP-Ribosyl)ated 50-mer Oligonucleotide. Pierisin-1 was purified from the pupae of P. rapae, as described (2). For site-specific modifications, 50 µg of a 13-mer oligonucleotide (5′-CCTTCCGTCTCCC, the underlined G is a target of ADP-ribosylation) was mono(ADP-ribosyl)ated in 100 µL of buffer containing 10 µg of Pronase-treated pierisin-1 (7), 1 mM of β-NAD, 2 mM of EDTA (pH 7.5), and 50 mM of Tris-HCl (pH 7.5) at 37 °C for 16 h (8, 11). The ADP-ribosylated 13-mer oligonucleotide was separated using an Agilent 1100 series HPLC system equipped with an Agilent G1315B photodiode array detector (Agilent Technologies, Waldbronn, Germany) and a column of COSMOSIL 5C18-MS (4.6 mm × 250 mm; Nacalai Tesque, Kyoto, Japan) at a flow rate of 1 mL/min. The solvent system was as follows: a linear gradient from 7% acetonitrile to 15% acetonitrile over 145 min in 50 mM triethylamine acetate (pH 7.0, TEAA) (8, 11). Under these conditions, R- or β-isomers of the ADP-ribosylated 13-mer oligonucleotide eluted at 96 and 97 min, and the unmodified 13-mer oligonucleotide eluted at 98 min. These R- and β-isoforms are easily epimerized under aqueous conditions (8). Fractions containing the modified oligonucleotides were collected, and the nucleotides were concentrated by evaporation. In order to confirm that the adduct was formed on the oligonucleotide, 1 nmol of ADPribosylated oligonucleotide was digested to mononucleotides by 22 units of micrococcal nuclease and 0.1 unit of phosphodiesterase II for 4 h at 37 °C in 17 mM sodium succinate buffer (pH 6.0) containing 8 mM CaCl2, then to nucleosides by 10 units of calf intestine alkaline phosphatase for 4 h at 37 °C in 50 mM Tris-HCl (pH 8.0) (17). The digest was separated by HPLC with a column of COSMOSIL 5C18-MS (4.6 mm × 250 mm) at a flow rate of 1 mL/min. The solvent system was as follows: isocratic 4% acetonitrile in 50 mM TEAA (pH 7.0) for 20 min then a linear gradient from 4% to 15% acetonitrile over 15 min in 50 mM TEAA (pH 7.0). Under these conditions, normal nucleosides eluted at 16 min, and both isomers of N2-(ADP-ribos-1-yl)-2′-deoxyguanosine eluted at 37 min. The retention time and the on-line UV spectrum of the isomers were identical to the authentic standards obtained previously (8). The modified and unmodified 13-mer oligoncleotides and the 18-mer oligonucleotide (5′-CAATCAGGCCAGATCTGC) were phosphorylated with T4 polynucleotide kinase. A 19-mer oligonucleotide (5′-GACTACGTACTGAATTCTC) was 5′-terminally labeled with [γ-32P]ATP. The phosphorylated 13-mer (20 pmol) was ligated with stoichiometric quantities of 18-mer and 19-mer using 40 units of T4 DNA ligase in the presence of a 40-mer template strand (5′-TCTGGCCTGATTGGGGAGACGGAAGGGAGAATTCAGTACG) containing the complementary sequence of 40 bases in a solution containing 10 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 10 mM DTT, and 1 mM ATP. Ligation was carried out at 16 °C for 12 h. The ligated 50-mer products were purified by denaturing 12% polyacrylamide gel electrophoresis (PAGE). The modified 50-mer migrated slower than the unmodified 50-mer oligonucleotide. The purified nucleotides were annealed with a 50mer complementary strand (5′-GCAGATCTGGCCTGATTGGG-

696 Chem. Res. Toxicol., Vol. 20, No. 4, 2007

Figure 2. Binding of UvrA or UvrAB to N2-ADPR-dG-modified DNA. Indicated concentrations of UvrA/B were incubated with 20 fmol of a 50 bp double-stranded DNA fragment containing a single site-specific N2-ADPR-dG. The 5′-end of the modified strand was labeled with 32P. The reaction mixture was electrophoresed and visualized by autoradiography. DNA-UvrA2, DNA-UvrA2B, and DNA-UvrB represent the formation of UvrA2, UvrA2B, and UvrB complexes with substrate DNA, respectively. DNA denotes a substrate free of proteins. The graph below the gel represents the relative radioactivity of bound DNA in total substrate DNA.

Figure 3. Formation of the stable DNA-UvrB complex. UvrA (2.5 nM) and UvrB (100 nM) were incubated with the N2-ADPR-dGmodified DNA (20 fmol) described in Figure 2 and the unmodified control DNA for the indicated time. The reaction mixture was electrophoresed and visualized by autoradiography. DNA-UvrB represents the formation of UvrB complexes with substrate DNA. DNA denotes a substrate free of proteins. The graph below the gel represents relative radioactivity of DNA-UvrB in total substrate DNA.

GAGACGGAAGGGAGAATTCAGTAGGTAGTC) in a solution containing 10 mM Tris-HCl, 150 mM NaCl, and 1 mM EDTA by heating at 80 °C for 5 min and slowly cooling to room temperature. The annealed substrates were purified by ethanol precipitation. UvrAB Gel Mobility Shift Assays. UvrA and UvrB proteins binding to 50-mer DNA substrates were determined by gel mobility shift assay (18, 19). Typically, the DNA substrate (20 fmol) was incubated with UvrA and UvrB at varying concentrations (Figures 2 and 3) at 37 °C for 15 min or the indicated time (Figure 3) in 20 µL of UvrABC buffer (50 mM Tris-HCl, pH 7.5, 50 mM KCl, 10 mM MgCl2, and 5 mM DTT) in the presence of 1 mM ATP. The Uvr subunits were diluted and premixed into the storage buffer (30 mM Tris-HCl (pH 7.5), 1 mM EDTA, 1 mM DTT, 0.01% Triton X-100, 0.1 M NaCl, and 50% glycerol) before mixing DNA. After incubation, 2 µL of 80% (v/v) glycerol was added, and the mixture was immediately loaded onto a 6% native polyacrylamide gel in TBE running buffer and electrophoresed at room temperature for 3 h. The polyacrylamide gel and TBE running buffer contained 10 mM MgCl2 and 1 mM ATP. The DNA-protein complexes were

Kawanishi et al. visualized, and the radioactivity of the corresponding bands was quantified by autoradiography using a Bio imaging analyzer (Fuji Photo Film Co., Ltd., Tokyo, Japan). UvrABC Incision Assays. UvrABC incision reactions were performed essentially as described previously (18, 19). The 50mer DNA substrates (20 fmol) were incised by UvrABC (15 nM UvrA, 250 nM UvrB, and 100 nM UvrC) in UvrABC buffer with 1 mM ATP at 37 °C for 0, 5, 10, 30, and 60 min. The Uvr subunits were diluted and premixed into the storage buffer before mixing DNA. The reactions were terminated by adding EDTA (20 mM). The samples were denatured with formamide and analyzed by electrophoresis on a 12% polyacrylamide sequencing gel under denaturing conditions with TBE buffer. The gel was subjected to autoradiography as described above. XPA Filter-Binding Assays. Nitrocellulose filter-binding assays were performed as described by Robins et al. (20) with some modifications. A DNA fragment (2.72 kb) from BssHII-digested pBluescript II, labeled with [R-32P]dCTP by T7 DNA polymerase, was used as a DNA probe. DNA was extracted with phenol, precipitated with ethanol and treated with Pronase-treated pierisin-1 (10 ng/µL) in 50 mM Tris-HCl buffer (pH 7.5) containing 1 mM β-NAD and 2 mM of EDTA (pH 7.5), at 37 °C for 1 h. It was then extracted again with phenol, precipitated with ethanol, and resuspended in 10 mM Tris-HCl (pH 8.5). The binding reaction mixture contained 10 ng of the DNA fragment and the indicated amount of XPA protein in 50 µL of binding buffer (25 mM Tris-HCl (pH 7.7), 50 mM NaCl, 1 mM MgCl2, 20 µM Zn(CH3COO) 2, 1 mM DTT, 100 µg/mL bovine serum albumin (BSA), and 5% glycerol). After the mixture was incubated for 30 min at 4 °C, a 50 µL aliquot was applied to alkali-pretreated nitrocellulose filters (Millipore HAWP 02500, 0.45 mm pore size) and rinsed twice with 0.5 mL of binding buffer without BSA. The residual radioactivity of the filters was counted using a liquid scintillation counter. XPA Gel Mobility Shift Assays. The other DNA fragment (180 bp) from BssHII-digested pBluescript II, labeled with [R-32P]dCTP by T7 DNA polymerase, was ADP-ribosylated with Pronase-treated pierisin-1 as described above. One nanogram of the ADP-ribosylated DNA fragment or control DNA fragment without ADPribosylation was incubated with the indicated amount of XPA proteins in binding buffer at 4 °C for 0.5 h. The reaction mixtures were then loaded onto 6% nondenaturing polyacrylamide gel, and electrophoresis was carried out at room temperature for 1 h. In supershift assays (mobility shift with antibody), the DNA fragment (1 ng) and GST-XPA (400 ng) were preincubated for 25 min, and anti-GST monoclonal antibody (100 ng) was added to the reaction mixture followed by an additional 5 min of incubation. After the reaction, the samples were loaded onto polyacrylamide gel. In competition assays, the DNA fragment (0.2 ng) and XPA (400 ng) were incubated with 0.2-0.5 ng of the non-labeled DNA fragment, and then the reaction mixtures were subjected to gel electrophoresis (Figure 6C). The DNA-protein complexes were visualized by autoradiography as described above. Replication Efficiency and the supF Gene Mutation Assay. The shuttle vector plasmid pMY189 (30 µg), derived from plasmid pZ189 (21, 22), was incubated with various doses of Pronase-treated pierisin-1, 0.1 mM β-NAD, and 1 mM EDTA in 100 µL of 50 mM Tris-HCl at pH 7.5 (8). The reaction was allowed to proceed for 1 h at 37 °C followed by phenol/chloroform extraction and ethanol precipitation. The plasmids were redissolved in 100 µL of Dulbecco’s PBS solution (pH 7.5). The human cells WI38-VA13 or XP2OS(SV) (5 × 106 cells) and pierisin-1-treated pMY189 in PBS solution (200 µL) were placed in an electroporation chamber (electrodes 0.3 cm apart) (PDS, Inc., Madison, WI), and the cells were transfected with the plasmids by electric pulses (600 V, 4 times). The cells were incubated at 37 °C for 72 h in a CO2 incubator, and then the plasmids were extracted using a QIAprepspin miniprep kit (QIAGEN GmbH, Hilden, Germany) and digested with restriction endonuclease DpnI to eliminate the nonreplicated plasmids retaining the bacterial methylation pattern. The plasmid DNA replicated in human cells was introduced into the indicator bacterium KS40/pKY241, a nalidixic acid-resistant derivative of

Repair of an ADP-Ribosylated DNA Adduct

Chem. Res. Toxicol., Vol. 20, No. 4, 2007 697 Micro Pulsar electroporation apparatus. Samples of the bacteria were plated on LB agar containing ampicillin and incubated for 24 h at 37 °C. Colonies were counted, and their replication capability was calculated as the ratio of the number of colonies with the pierisintreated plasmid against those with the untreated plasmid.

Results

Figure 4. Incision of 5′-end-labeled N2-ADPR-dG-modified 50 bp duplexes by UvrABC. The DNA substrates described in Figure 3 (20 fmol each) were incubated with UvrABC (15 nM UvrA, 250 nM UvrB, and 100 nM UvrC) for the indicated time, separated with polyacrylamide sequence gel, and visualized by autoradiography. The DNA substrate with N2-ADPR-dG resulted in two bands in the sequencing gel because of R- and β-isoforms: R-N2-ADPR-dG and β-N2-ADPRdG. The graph below the gel represents the relative radioactivity of the incised product in total DNA.

Figure 5. Binding of the XPA protein to the ADP-ribosylated DNA fragment in a filter-binding assay. ADP-ribosylated DNA (10 ng; represented by b) or a control DNA fragment without ADP-ribosylation (10 ng; represented by O) was incubated with the indicated amount of XPA proteins. The DNA bound to protein was collected on nitrocellulose filters.

MBM7070 (23, 24), using a Micro Pulsar electroporation apparatus (Bio-Rad Laboratories, Hercules, CA). Bacterial cells were plated on Luria-Bertani (LB) agar containing 50 µg/mL nalidixic acid, 150 µg/mL ampicillin, 30 µg/mL chloramphenicol, IPTG, and X-gal to select plasmids containing the mutated supF genes. Samples of the bacteria were plated on LB agar containing ampicillin and chloramphenicol to measure the total number of transformants. Plasmid recovery (replication capability) was calculated as the ratio of the number of transformants with the pierisin-treated plasmid against those with the untreated plasmid. After the plates were incubated for 24 h at 37 °C, the colonies were counted, and mutation frequencies were calculated. After E. coli cells with the mutated plasmids were cultured overnight in 2 mL of LB medium, the plasmids were extracted and purified with the QIAprep-spin miniprep kit, and the size of the mutated plasmid was checked by agarose gel electrophoresis. The base sequences of the supF gene of plasmids that were not aberrant in size were determined with an ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer Corporation, Foster City, CA) and the M13-21 primer, using an ABI377 automatic DNA sequencer (21). To measure the replication capability in E. coli, ADP-ribosylated pMY189 (3 ng each) was introduced into E. coli strains, WP2 (trpE65, malB15, lon-11, and sulA1), WP2uVrA (as WP2, but uVrA) WP100 (as WP2, but uVrA and recA), or ZA60 (recA), using a

Binding and Incision of N2-ADPR-dG by UvrABC Nuclease. To understand the structural basis of recognition and incision of N2-ADPR-dG by UvrABC, a gel mobility shift assay was performed to examine DNA adduct binding with increasing concentrations of UvrA and UvrB. As shown in Figure 2, the incubation of a single N2-ADPR-dG-modified 50 bp DNA with UvrA produced a shifted complex consistent with the previously reported formation of a DNA-UvrA2 complex (25-27). The unshifted band represents a substrate free of proteins. UvrA also bound to unmodified DNA with lower efficiency (data not shown). Incubation of UvrA (10 nM) in the presence of excess UvrB (200 nM), with N2-ADPR-dG-modified DNA, resulted in the formation of an additional complex that migrated faster than the DNA-UvrA2 complex. This was the DNA-UvrB complex resulting from the dissociation of UvrA2 from the DNAUvrA2B complex (13, 26, 27). No DNA-UvrB complex was observed in the reaction with unmodified DNA (data not shown). The DNA-UvrA2B complex had a slightly slower mobility than the DNA-UvrA2 complex. We then attempted to detect a stable DNA-UvrB complex in the time-course experiment. In Figure 3, the amount of DNA-UvrB complex increased with increasing incubation time of UvrA (2.5 nM) and UvrB (100 nM) with modified DNA. Nonmodified DNA did not support the formation of the DNA-UvrB complex. As shown in Figure 4, the 50 bp DNA substrate containing a single N2-ADPR-dG adduct was incised by UvrABC nuclease. The DNA substrate with N2-ADPR-dG resulted in the formation of two bands in the sequencing gel from the R- and β-isoforms. This substrate was labeled at the 5′-end of the modified strand. Therefore, an incised DNA product with the 5′-end labeled does not have N2-ADPR-dG and could form a single band. The amount of the incised product increased with the incubation time of UvrABC with modified DNA. No incision product was observed when UvrABC was incubated with an intact DNA substrate. These results indicated that E. coli repair protein UvrABC recognized N2-ADPR-dG -containing DNA and cleaved a fragment containing the damage. Binding of XPA Protein to Pierisin-1-Treated DNA. Numerous studies have provided evidence that XPA displays a binding preference for damaged DNA (28, 29). Here, we tested the DNA binding of XPA to an ADP-ribosylated DNA fragment. 32P-labeled DNA treated with pierisin-1 was incubated with various amounts of XPA protein, and then the reaction mixture was flowed through a filter. If DNA binds to XPA protein, then radioactivity will remain on the filter. As indicated in Figure 5, the binding of the XPA protein to the pierisin-1-treated DNA was dependent on the amount of XPA protein. When the XPA protein was mixed with pierisin-1-treated DNA, the radioactivity retained by the filter was 28, 36, and 46% for 100, 200, and 400 ng of XPA protein, respectively. When the XPA protein was mixed with intact DNA, the retained radioactivity was independent of the amount of XPA protein added, that is around 10% of radioactivity was retained as a result of the damageindependent DNA binding activity of the XPA protein. We also performed electrophoretic mobility shift analysis to examine the binding of pierisin-1-treated DNA to GST-fusion

698 Chem. Res. Toxicol., Vol. 20, No. 4, 2007

Kawanishi et al.

XPA protein (Figure 6A). Because ADP-ribosylated DNA has a higher molecular weight than intact DNA, a retardation of the modified DNA was observed even in the absence of the protein. When GST-XPA was added to the ADP-ribosylated DNA substrate, the bands that appeared were shifted further. The intensities of these bands increased, and those of the free DNA bands decreased when the amount of added GST-XPA increased. When anti-GST antibody was added to the binding mixture, supershifted bands were observed (Figure 6B). This result reinforces that the fraction shifted from free ADPribosylated DNA contains GST-XPA. Furthermore, the intensity of the shifted band decreased when non-32P-labeled ADPribosylated DNA was added as a competitor (Figure 6C). These results show that XPA protein preferentially binds to pierisin1-treated DNA. Replication and Mutation of Pierisin-1-Treated Plasmids. We introduced ADP-ribosylated plasmids into cells to allow replication. In NER-deficient E. coli cells (uVrA strain), replication capability diminished to 0.02% at 8 ng/µL pierisin-1 (Figure 7A). In the NER and recombination repair double-deficient (uVrArecA) strain, it diminished to 0.07% at the same concentration. In contrast, the replication capabilities of the wild-type and the recombination-repair-deficient (recA) strains were 0.20.4% at the same concentration. As shown in Figure 7B, the extent of plasmid recovery, which is the capacity of the plasmid to replicate in mammalian cells, decreased with increasing treatment concentrations of pierisin1. In human NER-deficient XP-A cells, the extent of plasmid recovery diminished to 3.3% at 2 ng/µL pierisin-1. In contrast, the extent of plasmid recovery was 71% at the same concentration in DNA-repair-proficient WI38-VA13. We pursued mutations of the plasmids replicated in human cells. As shown in Figure 7C, mutation frequency increased in both human cells with increasing concentrations of pierisin-1. The background mutation frequency was 4.8 × 10-4, and the frequency was increased 18 times in normal cells by treating the plasmid with 8 ng/µL pierisin-1. In XP cells, the background mutation frequency was 1.6 × 10-4, and the mutation frequency increased 35 times in XP cells when the plasmid was treated with 2 ng/ µL pierisin-1. The difference in the mutation frequencies between two cell lines was smaller than that between the extents of plasmid recovery. Base sequence analysis revealed that the majority (>94%) of N2-ADPR-dG-induced mutations were base substitutions in both cell lines (Table 1). Frameshift mutations were minor. The majority of base substitutions (97-98%) occurred at guanine or cytosine in both cells. The most frequent mutations were G/C to T/A transversions, and the next most frequent mutations were G/C to C/G transversions in either cell line, and also G/C to A/T transitions in normal cells. Figure 6. Binding of the XPA protein to the ADP-ribosylated DNA fragment in the gel mobility shift assay. (A) The ADP-ribosylated DNA fragment (1 ng) or control DNA fragment without ADP-ribosylation (1 ng) was incubated with the indicated amount of XPA proteins. The reaction mixture was electrophoresed and visualized by autoradiography. The graph below the gel represents the relative radioactivity of XPADNA in total substrate DNA. (B) The DNA fragment (1 ng) and GSTXPA (400 ng) were preincubated for 25 min, and anti-GST antibody (100 ng) was added to the reaction mixture. After 5 min of incubation, the samples were loaded onto polyacrylamide gel. (C) DNA fragment (0.2 ng) and XPA (400 ng) were incubated with the indicated amount of non-labeled DNA fragment, and then the reaction mixtures were subjected to gel electrophoresis. The graph below the gel represents the relative radioactivity of XPA-DNA in total substrate DNA. XPADNA and antibody-XPA-DNA represent the formation of XPA complexes with substrate DNA and antibody complexes with XPA and DNA, respectively. ADP-ribosyrated DNA and intact DNA denote substrates free of proteins. ADP-ribosylated DNA has a higher molecular weight than intact DNA.

Discussion Pierisin-1 catalyzes mono(ADP-ribosyl)ation of the 2′-dG residue and produces a bulky adduct, N2-ADPR-dG in DNA (Figure 1). The highly conserved and versatile NER pathway mainly removes helix-distorting lesions, including bulky chemical adducts as well as UV-induced cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts (6-4PPs). Shiotani et al. previously reported that NER-deficient strains of Chinese hamster ovary (CHO) cells showed higher pierisin-1-induced cytotoxicity and apoptosis than wild-type AA8 cells (30). Moreover, in wild-type AA8 cells, the adduct level increased to about 3/105 nucleotides and then decreased after the cells were treated with 1000 ng/mL pierisin-1 for 6 h, whereas in NER- mutants, the adduct levels continuously increased during culture and reached about 5/105 nucleotides by 24 h (30).

Repair of an ADP-Ribosylated DNA Adduct

Chem. Res. Toxicol., Vol. 20, No. 4, 2007 699 Table 1. Types of Pierisin-1-Induced Base Substitutions in the supF Gene of pMY189 Propagated in Human WI38VA13 (Wild-Type) and XP2OS(SV) (XP-A) Cells number of base changes (%) base substitution

Figure 7. Replication efficiency and mutation frequency of ADPribosylated plasmids. Data are the mean ( SD of at least three independent experiments. (A) The pMY189 plasmid containing an ampicillin-resistant gene was treated with the indicated concentration of pierisin-1. The plasmid was introduced into E. coli strains of the indicated genotype. E. coli strains were plated on LB agar containing ampicillin and incubated for 24 h. Colonies were counted, and replication capability was calculated as the ratio of the number of colonies with pierisin-treated plasmids against those with nontreated plasmids. * denotes that the mean is statistically different (P < 0.05). (B) The pierisin-1-treated shuttle vector pMY189 was introduced into NER-proficient human WI38-VA13 or NER-deficient human XP2OS(SV) cells. After incubation for 3 days, replicated plasmids were recovered and introduced into indicator E. coli. Plasmid recovery was calculated as described in Experimental Procedures. (C) Mutation frequencies of the supF gene on the pierisin-1-treated pMY189 replicated in human cells.

WI38VA13

XP2OS(SV)

G/C to T/A G/C to C/G G/C to A/T T/A to C/G T/A to A/T T/A to G/C

45 (38) 34 (29) 34 (29) 2 (2) 2 (2) 1 (1)

53 (41) 41 (32) 31 (24) 2 (2)

total

118 (100)

128 (100)

1 (1)

In E. coli, UvrABC proteins mediate DNA damage recognition and excision (13). In the E. coli NER model (31), the UvrA dimer loads UvrB onto the damaged DNA site. The interaction of UvrA2B with the lesion causes unwinding, denaturing, and opening of the local DNA duplex at the adduct. This conformational change allows direct UvrB access to the adducted base, resulting in the formation of a stable DNA-UvrB complex. UvrC interacts with UvrB in the DNA-UvrB intermediate, and then this UvrBC complex acts as an endonuclease. Here, we showed that UvrAB preferentially bound to the site-specific N2ADPR-dG-modified 50-mer DNA (Figure 2). The incubation of UvrAB with modified DNA led to the accumulation of the DNA-UvrB complex (Figure 3), whereas the incubation of all three UvrABC components resulted in the incised product of the modified substrate (Figure 4). Furthermore, NER-deficient E. coli showed less replication efficiency of the ADP-ribosylated plasmid (Figure 7A). Taken together, NER is involved in the repair of N2-ADPR-dG in E. coli. In mammalian cells, XPA protein binds to distorted DNA and acts as a bridging factor between the recognition intermediate and the ultimate excision complex (12). In this study, the results of the filter-binding assay and gel mobility shift assay indicated that the XPA protein preferentially binds to pierisin1-treated DNA fragments in Vitro (Figures 5 and 6). The mammalian NER mechanism has not been elucidated well. Initial studies suggested that DNA damage was recognized by XPA (32), and recent studies have also implied that XPC takes part in damage recognition and initiation of NER (33, 34). In in Vitro model reactions, excision occurred faster when XPA and RPA were first present as damage-recognition factors (34). In this study, XP-A cells showed less replication efficiency of the ADP-ribosylated plasmid than NER-proficient cells (Figure 7B). This is consistent with the fact that NER-deficient strains of CHO cells show higher sensitivity to pierisin-1 (30). Thus, NER is involved in the repair of N2-ADPR-dG in mammalian cells. Most DNA lesions may be removed by DNA repair mechanisms, but some may escape from repair mechanisms and lead to mutagenesis. Our previous study showed that, in mammalian cells, N2-ADPR-dG generates G to C and T transversion mutations in HPRT and supF genes (9). Here, in XP-A cells, the mutation frequency of the supF gene was higher than that in wild-type cells (Figure 7C); however, the characteristics of base substitutions were not significantly different (Table 1). Moreover, mutational hotspots appeared at the same positions (103, 104, 159, 160, 168, and 169 of the supF gene) in XP-A and wild-type cells (data not shown). An in Vitro study with human Y-family polymerases indicated that Pol κ may play an important role in both error-free and error-prone bypasses of N2-ADPR-dG because Pol κ could extend DNA synthesis from the mismatch terminus (10). In this study, we assessed the efficiency of E. coli and human NER systems for repairing N2-ADPR-dG produced by pierisin-

700 Chem. Res. Toxicol., Vol. 20, No. 4, 2007

1. UvrAB and XPA proteins specifically bound to DNA containing N2-ADPR-dG, and UvrABC nuclease incised the DNA. The replication capability of plasmids treated with pierisin-1 in NER-deficient E. coli and human XP-A cells was lower than that in the proficient cells, and more mutations were induced in the plasmids propagated in NER-deficient cells than in the proficient cells. Taken together, we concluded that NER is involved in the repair of N2-ADPR-dG.

Kawanishi et al.

(15)

(16)

(17)

Acknowledgment. This work was supported in part by a Grant-in-Aid for Cancer Research from the Ministry of Health, Labor, and Welfare of Japan and Grants-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology.

(18)

(19)

References (1) Koyama, K., Wakabayashi, K., Masutani, M., Koiwai, K., Watanabe, M., Yamazaki, S., Kono, T., Miki, K., and Sugimura, T. (1996) Presence in Pieris rapae of cytotoxic activity against human carcinoma cells. Jpn. J. Cancer Res. 87, 1259-1262. (2) Watanabe, M., Kono, T., Koyama, K., Sugimura, T., and Wakabayashi, K. (1998) Purification of pierisin, an inducer of apoptosis in human gastric carcinoma cells, from cabbage butterfly, Pieris rapae. Jpn. J. Cancer Res. 89, 556-561. (3) Watanabe, M., Nakano, T., Shiotani, B., Matsushima-Hibiya, Y., Kiuchi, M., Yukuhiro, F., Kanazawa, T., Koyama, K., Sugimura, T., and Wakabayashi, K. (2004) Developmental stage-specific expression and tissue distribution of pierisin-1, a guanine-specific ADP-ribosylating toxin, in Pieris rapae. Comp. Biochem. Physiol., A 139, 125131. (4) Sugimura, T. (2002) Serendipitous discoveries from sudden inspirations and the joy of being a scientist. Biochem. Biophys. Res. Commun. 296, 1037-1038. (5) Kono, T., Watanabe, M., Koyama, K., Kishimoto, T., Fukushima, S., Sugimura, T., and Wakabayashi, K. (1999) Cytotoxic activity of pierisin, from the cabbage butterfly, Pieris rapae, in various human cancer cell lines, Cancer Lett. 137, 75-81. (6) Kanazawa, T., Kono, T., Watanabe, M., Matsushima-Hibiya, Y., Nakano, T., Koyama, K., Tanaka, N., Sugimura, T., and Wakabayashi, K. (2002) Bcl-2 blocks apoptosis caused by pierisin-1, a guaninespecific ADP-ribosylating toxin from the cabbage butterfly. Biochem. Biophys. Res. Commun. 296, 20-25. (7) Watanabe, M., Kono, T., Matsushima-Hibiya, Y., Kanazawa, T., Nishisaka, N., Kishimoto, T., Koyama, K., Sugimura, T., and Wakabayashi, K. (1999) Molecular cloning of an apoptosis-inducing protein, pierisin, from cabbage butterfly: possible involvement of ADP-ribosylation in its activity. Proc. Natl. Acad. Sci. U.S.A. 96, 10608-10613. (8) Takamura-Enya, T., Watanabe, M., Totsuka, Y., Kanazawa, T., Matsushima-Hibiya, Y., Koyama, K., Sugimura, T., and Wakabayashi, K. (2001) Mono(ADP-ribosyl)ation of 2′-deoxyguanosine residue in DNA by an apoptosis-inducing protein, pierisin-1, from cabbage butterfly. Proc. Natl. Acad. Sci. U.S.A. 98, 12414-12419. (9) Totsuka, Y., Kawanishi, M., Nishigaki, R., Matsukawa, K., Yagi, T., Takamura-Enya, T., Watanabe, M., Sugimura, T., and Wakabayashi, K. (2003) Analysis of HPRT and supF mutations caused by pierisin1, a guanine specific ADP-ribosylating toxin derived from the cabbage butterfly. Chem. Res. Toxicol. 16, 945-952. (10) Kawanishi, M., Matsukawa, K., Ohashi, E., Takamura, T., Totsuka, Y., Watanabe, M., Sugimura, T., Wakabayashi, K., Hanaoka, F., Ohmori, H., and Yagi, T. (2006) Translesion DNA Synthesis across Mono ADP-Ribosylated Deoxyguanosine by Y-Family DNA Polymerases. In New DeVelopments in Mutation Research, (Valon, C., Eds.), pp 133-148, Nova Science Publishers, New York. (11) Shiotani, B., Kobayashi, M., Watanabe, M., Yamamoto, K., Sugimura, T., and Wakabayashi, K. (2006) Involvement of the ATR- and ATMdependent checkpoint responses in cell cycle arrest evoked by pierisin1. Mol. Cancer Res. 4, 125-133. (12) Dip, R., Camenisch, U., and Naegeli, H. (2004) Mechanisms of DNA damage recognition and strand discrimination in human nucleotide excision repair. DNA Repair 3, 1409-1423. (13) Zou, Y., Liu, T. M., Geacintov, N. E., and Van Houten, B. (1995) Interaction of the UvrABC nuclease system with a DNA duplex containing a single stereoisomer of dG-(+)- or dG-(-)-anti-BPDE. Biochemistry 34, 13582-13593. (14) Nagai, A., Saijo, M., Kuraoka, I., Matsuda, T., Kodo, N., Nakatsu, Y., Mimaki, T., Mino, M., Biggerstaff, M., Wood, R. D., Sijbers, A., Hoeijmakers, J. H. J., and Tanaka, K. (1995) Enhancement of damage-

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

specific DNA binding of XPA by interaction with the ERCC1 DNA repair protein. Biochem. Biophys. Res. Commun. 211, 960-966. Girardi, A. J., Jensen, F. C., and Koprowski, H. (1965) SV40-induced transformation of human diploid cells, crisis and recovery. J. Cell. Physiol. 65, 69-83. Takebe, H., Nii, S., Ishii, M. I., and Utsumi, H. (1974) Comparative studies of host-cell reactivation, colony forming ability and excision repair after UV irradiation of xeroderma pigmentosum, normal human and some other mammalian cells. Mutat. Res. 25, 383-390. Dasaradhi, L., and Shibutani, S. (1997) Identification of tamoxifenDNA adducts formed by R-sulfate tamoxifen and R-acetoxytamoxifen. Chem. Res. Toxicol. 10, 189-196. Minko, I. G., Zou, Y., and Lloyd, R. S. (2002) Incision of DNAprotein crosslinks by UvrABC nuclease suggests a potential repair pathway involving nucleotide excision repair. Proc. Natl. Acad. Sci. U.S.A. 99, 1905-1909. Luo, C., Krishnasamy, R., Basu, A. K., and Zou, Y. (2000) Recognition and incision of site-specifically modified C8 guanine adducts formed by 2-aminofluorene, N-acetyl-2-aminofluorene and 1-nitropyrene by UvrABC nuclease. Nucleic Acids Res. 28, 3719-3724. Robins, P., Jones, C. J., Biggerstaff, M., Lindahl, T., and Wood, R. D. (1991) Complementation of DNA repair in xeroderma pigmentosum group A cell extracts by a protein with affinity for damaged DNA. EMBO J. 10, 3913-3921. Kawanishi, M., Enya, T., Suzuki, H., Takebe, H., Matsui, S., and Yagi, T. (1998) Mutagenic specificity of a derivative of 3-nitrobenzanthrone in the supF shuttle vector plasmids, Chem. Res. Toxicol. 11, 14681473. Matsuda, T., Yagi, T., Kawanishi, M., Matsui, S., and Takebe, H. (1995) Molecular analysis of mutations induced by 2-chloroacetaldehyde, the ultimate carcinogenic form of vinyl chloride, in human cells using shuttle vectors. Carcinogenesis 16, 2389-2394. Akasaka, S., Takimoto, K., and Yamamoto, K. (1992) G:C to T:A and G:C to C:G transversions are the predominant spontaneous mutations in the Escherichia coli supF gene; an improved lacZ(am) E.coli host designed for assaying pZ189 supF mutational specificity. Mol. Gen. Genet. 235, 173-178. Seidman, M. M., Dixon, K., Razzaque, A., Zagursky, R., and Berman, M. L. (1985) A shuttle vector plasmid for studying carcinogen-induced point mutations in mammalian cells. Gene 38, 233-237. Mazur, S. J., and Grossman, L. (1991) Dimerization of Escherichia coli UvrA and its binding to undamaged and ultraviolet light damaged DNA. Biochemistry 30, 4432-4443. Visse, R., de Ruijter, M., Moolenaar, G. F., and van de Putte, P. (1992) Analysis of UvrABC endonuclease reaction intermediates on cisplatindamaged DNA using mobility shift gel electrophoresis. J. Biol. Chem. 267, 6736-6742. Van Houten, B., and Snowden, A. (1993) Mechanism of action of the Escherichia coli UvrABC nuclease: clues to the damage recognition problem. BioEssays 15, 51-59. Thoma, B. S., and Vasquez, K. M. (2003) Critical DNA damage recognition functions of XPC-hHR23B and XPA-RPA in nucleotide excision repair. Mol. Carcinog. 38, 1-13. Liu, Y., Liu, Y., Yang, Z., Utzat, C., Wang, G., Basu, A. K., and Zou, Y. (2005) Cooperative interaction of human XPA stabilizes and enhances specific binding of XPA to DNA damage. Biochemistry 44, 7361-7368. Shiotani, B., Watanabe, M., Totsuka, Y., Sugimura, T., and Wakabayashi, K. (2005) Involvement of nucleotide excision repair (NER) system in repair of mono ADP-ribosylated dG adducts produced by pierisin-1, a cytotoxic protein from cabbage butterfly. Mutat. Res. 572, 150-155. Zou, Y., and Van Houten, B. (1999) Strand opening by the UvrA(2)B complex allows dynamic recognition of DNA damage. EMBO J. 18, 4889-4901. Jones, C. J., and Wood, R. D. (1993) Preferential binding of the xeroderma pigmentosum group A complementing protein to damaged DNA. Biochemistry 32, 12096-12104. Sugasawa, K., Ng, J. M., Masutani, C., Iwai, S., van der Spek, P. J., Eker, A. P., Hanaoka, F., Bootsma, D., and Hoeijmakers, J. H. (1998) Xeroderma pigmentosum group C protein complex is the initiator of global genome nucleotide excision repair. Mol. Cell 2, 223-232. Wakasugi, M, and Sancar, A. (1999) Order of assembly of human DNA repair excision nuclease. J. Biol. Chem. 274, 18759-18768.

TX600360B