432
Chem. Res. Toxicol. 2010, 23, 432–442
Genotoxicity of Soluble and Particulate Cadmium Compounds: Impact on Oxidative DNA Damage and Nucleotide Excision Repair Tanja Schwerdtle,*,† Franziska Ebert,† Christina Thuy,‡ Constanze Richter,‡ Leon H. F. Mullenders,§ and Andrea Hartwig‡ Institut fu¨r Lebensmittelchemie, Westfa¨lische Wilhelms-UniVersita¨t Mu¨nster, Corrensstrasse 45, 48149 Mu¨nster, Germany, Institut fu¨r Lebensmittelchemie und Lebensmitteltechnologie, Fachgebiet Lebensmittelchemie und Toxikologie, Technische UniVersita¨t Berlin, GustaV-Meyer-Allee 25, 13355 Berlin, Germany, Department of Toxicogenetics, Leiden UniVersity Medical Center, EinthoVenweg 20, 2333 ZC, Leiden, The Netherlands ReceiVed December 14, 2009
Water-soluble and particulate cadmium compounds are carcinogenic to humans. While direct interactions with DNA are unlikely to account for carcinogenicity, induction of oxidative DNA damage and interference with DNA repair processes might be more relevant underlying modes of action (recently summarized, for example, in Joseph, P. (2009) Tox. Appl. Pharmacol. 238, 271-279). The present study aimed to compare genotoxic effects of particulate CdO and soluble CdCl2 in cultured human cells (A549, VH10hTert). Both cadmium compounds increased the baseline level of oxidative DNA damage. Even more pronounced, both cadmium compounds inhibited the nucleotide excision repair (NER) of BPDEinduced bulky DNA adducts and UVC-induced photolesions in a dose-dependent manner at noncytotoxic concentrations. Thereby, the uptake of cadmium in the nuclei strongly correlated with the repair inhibition of bulky DNA adducts, indicating that independent of the cadmium compound applied Cd2+ is the common species responsible for the observed repair inhibition. Regarding the underlying molecular mechanisms in human cells, CdCl2 (as shown before by Meplan, C., Mann, K., and Hainaut, P. (1999) J. Biol. Chem. 274, 31663-31670) and CdO altered the conformation of the zinc binding domain of the tumor suppressor protein p53. In further studies applying only CdCl2, cadmium decreased the total nuclear protein level of XPC, which is believed to be the principle initiator of global genome NER. This led to diminished association of XPC to sites of local UVC damage, resulting in decreased recruitment of further NER proteins. Additionally, CdCl2 strongly disturbed the disassembly of XPC and XPA. In summary, our data indicate a general nucleotide excision repair inhibition by cadmium compounds, which is most likely caused by a diminished assembly and disassembly of the NER machinery. These data reveal new insights into the mechanisms involved in cadmium carcinogenesis and provide further evidence that DNA repair inhibition may be one predominant mechanism in cadmium induced carcinogenicity. Introduction Cadmium is a ubiquitous environmental pollutant that has been classified as a human carcinogen by the International Agency for Research on Cancer (1-3). Major routes of exposure are inhalation and ingestion. For nonsmoking individuals of the general population, food is the most important source of cadmium exposure; smoking of one pack of cigarettes per day doubles cadmium uptake (4-6). With respect to occupational exposure, the primary route of uptake is through the inhalation of dust and fumes and to some extent incidental ingestion of dust from contaminated hands or food (6, 7). Cadmium oral exposure occurs toward soluble cadmium salts and in the case of inhalative exposure toward particulate cadmium such as CdO as the predominant form of airborne cadmium. One question of particular concern is whether water-soluble and waterinsoluble cadmium compounds exert similar effects, i.e., whether biological effects and adverse responses are caused by a common species or whether completely different mechanisms * To whom correspondence should be addressed. Phone: +49-25183-33874. Fax: +49-251-83-33396. E-mail: Tanja.Schwerdtle@ uni-muenster.de. † Westfa¨lische Wilhelms-Universita¨t Mu¨nster. ‡ Technische Universita¨t Berlin. § Leiden University Medical Center.
apply for toxicity and/or carcinogenicity. So far, only few experimental carcinogenicity studies have dealt with different cadmium species. For oral exposure, most animal studies have applied soluble cadmium compounds, which exist as the Cd2+ ion regardless of the initial salt. Experimental inhalation studies in laboratory animals yielded carcinogenic effects when applying CdCl2, CdSO4, CdO, or CdS (8-10). With respect to mechanistic studies related to cadmiuminduced genotoxicity which were conducted exclusively with water-soluble cadmium compounds, direct genotoxic and mutagenic effects were restricted to mostly high concentrations; however, comutagenic effects in combination with other DNA damaging agents were observed at lower, noncytotoxic concentrations (11). Therefore, rather indirect mechanisms of genotoxicity are discussed, such as interference with enzymes of cellular antioxidant systems, modulation of gene expression, and signal transduction and inhibition of DNA methylation. Furthermore, one of the most relevant mechanisms at low concentrations appears to be the interference of cadmium with DNA repair processes, leading to a diminished removal of DNA lesions induced by endogenous and exogenous sources (5, 11-15). Thus, cadmium has been shown to inhibit mismatch repair as well as base excision repair (BER) and nucleotide excision repair
10.1021/tx900444w 2010 American Chemical Society Published on Web 01/21/2010
Cd Disturbs Nucleotide Excision Repair Machinery
(NER1), which are responsible for the removal of mispaired bases, base modifications produced by alkylation, deamination or oxidation, and bulky lesions, mainly produced by environmental mutagens (12, 16). Global genome NER (GG-NER) is a rather complex mechanism; in total, over 30 proteins are involved. After DNA damage recognition, incisions at sites flanking the lesion occur, and an oligonucleotide (24-32 nucleotides) containing the lesion is excised, followed by subsequent restoration of the original DNA sequence by polymerization/ligation using the nondamaged strand as a template. Detection and incision of the lesion are known to be the crucial steps in GG-NER initiation. Here, XPC is believed to be the principle initiator of this DNA repair pathway, XPA is an absolutely essential part of the core preincision complex, and XPG has a structural role in the assembly of the preincision complex and a catalytic role in making the incision 3′ to the damaged site in DNA (recently summarized in refs 17-19). With respect to the inhibition of NER in cultured cells, cadmium has been shown to inhibit the repair of B[a]P-induced lesions, as measured by the comet assay (20), and the removal of UVC-induced DNA lesions (21). Interactions with DNA repair processes appear to be due to interactions with DNA repair proteins; potential binding sites are cysteines, e.g., of zinc binding proteins, leading to the substitution of zinc and conformational changes and in consequence possibly to altered protein function (22-27). However, up to now the exact mechanisms are not fully understood, especially as effects of cadmium on isolated key NER proteins have only been demonstrated in subcellular systems. Interestingly, the tumor suppressor protein p53 also harbors a zinc binding domain, which is essential for DNA binding and p53 function as transcription factor. In this context, Meplan et al. demonstrated that cadmium chloride alters p53 conformation in MCF7 cells, inhibits its DNA binding, and down regulates transcriptional activation of a reporter gene (28). As p53 has recently been shown to serve as a transcription factor for two important NER genes, XPC and p48 (29, 30), a cadmium induced p53 conformational change may also result in altered p53 NER downstream effects. In the present study, we compared the genotoxic effects of soluble CdCl2 and particulate CdO and correlated the observed effects with cellular bioavailability of Cd2+. For the first time, the impact of cadmium on the removal of several NER substrates and the assembly and disassembly of key proteins of the NER machinery were investigated in one study.
Experimental Procedures Caution: The following chemicals are hazardous and should be handled carefully: (+)-anti-BPDE, CdCl2, and CdO. Materials. Dulbecco’s modified Eagle’s medium (DMEM), fetal calf serum, penicillin-streptomycin solutions, trypsin, leupeptin, aprotinin, pepstatin, PMSF, formaldehyde solution (37%), CdCl2 (>99,99%), and CdO (>99,99%) were products of Sigma (Deisenhofen, Germany). Genetecin (G418) was purchased from Invitrogen (Karlsruhe, Germany). The culture dishes were supplied by Biochrom (Berlin, Germany). Triton X-100 was obtained from Pierce (Oud-Beijerland, The Netherlands), hydroxyapatite (high resolution) from Calbiochem (Bad Soden, Germany), and Giemsa stain and 1 Abbreviations: (+)-anti-BPDE, (+)-anti-benzo[a]pyrene-7,8-diol 9,10epoxide; CPD, cyclobutane pyrimidine dimers; GG-NER, global genome nucleotide excision repair; 6-4PP, 6-4 photoproducts; XPC, Xeroderma pigmentosum complementation group C protein; XPA, Xeroderma pigmentosum complementation group A protein; XPG, Xeroderma pigmentosum complementation group G protein.
Chem. Res. Toxicol., Vol. 23, No. 2, 2010 433 ProteinaseK from Merck (Darmstadt, Germany). Calf thymus DNA and RNase A were purchased from Roche (Mannheim, Germany), and phenol/chloroform/isoamyl alcohol (25:24:1), Tween 20, TEMED, Tris (>99%), and acrylamide (37.5:1) were purchased from Roth (Karlsruhe, Germany). Enhanced Chemiluminescence (ECL) reagent and PVDF membranes were from GE Healthcare (Mu¨nchen, Germany), and the protein assay reagent solution was supplied by Bio-Rad (Mu¨nchen, Germany). The antibodies PAb1620 (specific for the wild type, folded p53 conformation), PAb240 (specific for the conformational mutant unfolded p53), and PAb416 (negative control) and protein G Plus/Protein A agarose suspension were obtained from Calbiochem (Darmstadt, Germany). The primary p53 antibody NCL-p53-CM1 was obtained from Novocastra (Newcastle, UK), the polyclonal rabbit antibody against XPC (H-300), polyclonal rabbit antibody against XPA (Fl-273), monoclonal mouse antibody against XPG (847), and the rabbit polyclonal antibody against actin were obtained from Santa Cruz Biotechnology (Santa Cruz, USA), and the monoclonal mouse antibodies against 6-4 photoproducts and cyclobutan pyrimidine dimers were from Medical & Biological Laboratories Co. (Japan). The secondary HRP-conjugated antibodies were from Santa Cruz Biotechnology (Santa Cruz, USA), the secondary Cy3-conjugated antibody from Jackson Immunoresearch Laboratories (Westgrove, USA), the secondary Alexa Fluor 488-conjugated antibody from Invitrogen (Paisley, UK), and Vectashield mounting medium containing DAPI (1 µg/mL) from Vector Laboratories (Burlingame, USA). All other chemicals were of p.a. grade and were from Merck (Darmstadt Germany) or Fluka Chemie (Buchs, Germany). r-7,t-8,t-9,c-10tetrahydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene (tetrol I-1) was obtained from the NCI Chemical Carcinogen Reference Standard Repository (Kansas City, USA) and (+)-anti-benzo[a]pyrene-7,8diol 9,10-epoxide ((+)-anti-BPDE) from Professor Dr. G. GrimmerStiftung (Grossharnsdorf, Germany), Biochemisches Institut fu¨r Umweltcarcinogene. S. Boiteux (Fontenay, aux Roses, France) kindly provided the Fpg protein. Cell Culture and Incubation. A549 cells were grown as monolayers in DMEM containing 10% fetal calf serum, 100 U/mL penicillin, and 100 µg/mL streptomycin. The cultures were incubated at 37 °C with 5% CO2 in air and 100% humidification. Normal human diploid skin fibroblasts derived from a healthy individual were immortalized by telomerase transfection (VH10hTert) (31). VH10hTert cells were grown in tissue culture dishes as monolayers in DMEM supplemented with 10% FCS, penicillin (100 U/mL), streptomycin (100 µg/mL), and G418 (25 µg/mL). The cells were incubated at 37 °C with 5% CO2 in air and 100% humidity. For repair experiments, confluent VH10hTert cells were used to prevent post-treatment replication. Cells were treated with the respective cadmium compound, (+)anti-BPDE and/or UVC as described for the respective experiments. CdO particles were suspended directly before each experiment by sonification for 30 min in bidistilled water; soluble CdCl2 was dissolved in bidistilled water. (+)-anti-BPDE was dissolved in water free THF/5% triethylamine (1 mg/mL) and stored at -80 °C. To avoid hydrolyses of the epoxide, dilutions from stock solutions were always prepared with fresh solvent (water free THF/1% triethylamine) immediately before incubation. The final concentration of the solvent in medium was always 0.1%. In the case of repair studies after incubation with (+)-anti-BPDE, cells were washed twice with fresh culture medium. Global and Local UVC-Irradiation. Before UVC-irradiation, the culture medium was removed, and cells were washed once with PBS. UVC-irradiation at a wavelength of 254 nm was achieved applying a General Electric germicidal lamp (Bioblock Scientific, Illkirch, France). Doses were verified before each experiment by a dosimeter (UV-radiometer211, PRC Krochmann GmbH, Berlin, Germany). For local UVC-irradiation, cells on coverslips were covered with an UVC-blocking isopore polycarbonate filter with a pore size of 8 µm and UVC-irradiated. Cells were locally irradiated with a noncytotoxic dose of 30 J/m2. After irradiation, the filter was removed, and in the case of repair studies, the culture medium removed prior to UVC-irradiation was added for postincubation.
434
Chem. Res. Toxicol., Vol. 23, No. 2, 2010
Cell Number and Colony Forming Ability. Cells were incubated as indicated for the respective experiments, washed with PBS, trypsinized, and counted, and 300 (A549) or 500 (VH10hTert) cells/ dish were seeded for the determination of colony forming ability. After 7 days of incubation, colonies were fixed with ethanol, stained with Giemsa, counted, and calculated as percent of control. Untreated controls exhibited colony forming abilities of about 75% (A549 cells) and 33% (VH10hTert cells). Induction of Oxidative DNA Damage in Cultured Human Cells. DNA strand breaks and Fpg-sensitive sites were determined by the alkaline unwinding technique in combination with bacterial formamidopyrimidine-DNA-glycosylase (32). Fpg recognizes 7,8dihydro-8-oxoguanine (8-oxoguanine), 2,6-diamino-4-hydroxy-5formamido-pyrimidine (Fapy-Gua), 4,6-diamino-5-formamidopyrimidine (Fapy-Ade), and to a smaller extent 7,8-dihydro-8oxoadenine (8-oxoadenine) as well as apurinic/apyrimidinic sites (AP sites) and converts them into DNA strand breaks by its DNA endonuclease activity (33, 34). Briefly, 1 × 105 cells were seeded and allowed to attach for at least 24 h before treatment with the respective cadmium compounds. At the end of the treatment, the medium was removed, cells were washed with cold PBS, and lesions were quantified and calculated as described before (32). Detection of BPDE-Induced DNA Adducts. When investigating the effects of the cadmium compounds on B[a]P-induced adduct formation and repair, the ultimate carcinogenic B[a]P metabolite (+)-anti-BPDE was applied to avoid the interference of cadmium with phase I metabolism of B[a]P. BPDE-DNA adduct levels were measured by an improved, highly sensitive HPLC/fluorescence assay based on the release of the corresponding tetrols as described previously (35, 36). Briefly, 2-4 × 106 logarithmically growing cells were trypsinized after incubation, washed three times with ice-cold Tris-buffered saline (0.0027 M KCl, 0.137 M NaCl, and 0.025 M Tris-base, pH 7.4), and collected by centrifugation. DNA was isolated and its concentration was determined spectrophotometrically by measuring the UV absorbance at 260 nm. Absorbance ratios at 260/230 and 260/280, reflecting the purity of DNA, were always ∼2.3 and >1.85, respectively. After the hydrolysis of adducted DNA (10-100 µg) and subsequent neutralization, tetrol I-1 was quantified by HPLC with fluorescence detection, and adduct levels were calculated as described elsewhere (35). Intracellular Distribution of Cadmium in Cultured Human Cells. For the assessment of intracellular distribution, logarithmically growing A549 cells were treated for 24 h with CdCl2 or CdO, and after cell fractionation, cadmium was determined by atomic absorption spectroscopy (Perkin-Elmer 4110 ZL, AS-72) in cytoplasm and nuclear protein extracts, which were prepared as described before (37). Briefly, 1-2 × 106 logarithmically growing cells were trypsinized after incubation, collected by centrifugation, washed two times with ice-cold PBS, resuspended by lysis buffer I (0.01 M HEPES, pH 7.9, 0.01 M KCl, 0.0015 M MgCl2, 0.3 M saccharose, 0.0005 M DTT, 0.0006 M PMSF, and 0.00047 M leupeptin) and incubated for 15 min on ice. After the addition of 0.5% NP-40 and centrifugation (4 °C, 1500g, 15 min), the cytoplasm was removed and stored at -80 °C until the measurement of cadmium by atomic absorption spectroscopy. Nuclei were washed three times with lysis buffer I, thereafter incubated with lysis buffer II (0.01 M HEPES, pH 7.9, 0.4 M KCl, 0.0015 M MgCl2, 25% glycerine, 0.0005 M DTT, 0.0006 M PMSF, and 0.00047 M leupeptin) for 30 min on ice, centrifugated (4 °C, 10000g, 15 min), and the nuclear extract was removed and stored at -80 °C until measurement of cadmium by atomic absorption spectroscopy. For the calculation of cadmium concentration, volumes of cells and nuclei were measured by an automatic cell counter (Casy-1, Roche Innovatis AG) in each sample; these measurements are based on noninvasive (dye-free) electrical current exclusion with signal evaluation via pulse area analysis. In the observed concentration ranges, both cadmium compounds showed no significant effects, neither on cell volumes
Schwerdtle et al. nor on nuclei volumes. Mean ((SD) volumes of cells and nuclei were 2.69 ((0.16) × 10-12 L and 3.87 ((0.27) × 10-13 L, respectively. p53 Immunoprecipitation with Conformation Specific Antibodies. p53 immunoprecipitation was carried out as described by Meplan et al. (28) with modifications as indicated below. After the incubation of exponentially growing A549 cells, 12 × 106 cells were washed with PBS, lysed on ice in 1000 µL of immunoprecipitation buffer (0.01 M Tris, pH 7.6, 0.14 M NaCl, 0.5% Nonidet P-40, 0.5 mM PMSF, 0.5 µg/mL leupeptin, 2 µg/mL aprotinin, and 0.7 µg/mL pepstatin), scraped by a rubber policeman, and gently shaken on ice for 45 min prior to 5 min of centrifugation at 15000g at 4 °C. Supernatants were precleared by incubation with 1 µg of PAb 416 (a nonanti-p53 antibody specific for large T antigen of SV40) and 20 µL of protein G Plus/Protein A agarose suspension by shaking for 45 min at 4 °C. After 2 min of centrifugation (10000g, 4 °C) and protein quantification in the supernatants (Bradford analysis), aliquots of supernatants were taken for immunoprecipitation with 1 µg of monoclonal antibodies PAb1620 (specific for the wild type, folded p53 conformation), PAb240 (specific for the conformational mutant unfolded p53), and PAb416 (negative control). Fifteen microliters of protein G Plus/Protein A Agarose Suspension was added, samples were gently shaken for 18 h at 4 °C, and immune complexes were collected by centrifugation (12000g, 1 min, 4 °C) and washed twice with immunoprecipitation buffer and once with PBS. Finally, precipitates were denatured in Laemmli buffer and analyzed by Western blot experiments as described above using the rabbit polyclonal antip53 antibody CM-1 (1:1000 dilution) and a HRP-conjugated goat antirabbit immunoglobulin G as secondary antibody. Detection was carried out by a chemiluminescence imaging system (LAS 3000, raytest, Straubenhardt, Germany). Detection of UVC-Induced Photolesions. UVC-induced cyclobutane pyrimidine dimers (CPD) and 6-4 photoproducts (6-4PP) were quantified after local UVC-irradiation of fully confluent VH10hTert cells by fluorescent labeling as described before (38) with minor modifications. Briefly, VH10hTert cells were washed twice with cold PBS, fixed, and lysed with PBS containing 1% formaldehyde and 0.2% Triton X-100 for 30 min on ice and washed again twice with cold PBS. To visualize CPD or 6-4PP, the cellular DNA was denatured with 0.1 M HCl for 10 min at 37 °C. After washing three times with PBS, coverslips were incubated with 3% bovine albumin in PBS for 30 min at room temperature. Primary and secondary antibodies were incubated for 1 h at 37 °C and room temperature, respectively, in washing buffer (WB/PBS, 0.5% bovine albumin, 0.05% Tween-20). To achieve low background signaling, after each antibody incubation, cells were washed 3 times for 5 min with WB. Finally, cells were postfixed with PBS with 2.6% formaldehyde and mounted in Vectashield mounting medium containing DAPI. Photolesions were measured on a Zeiss Axio Imager M1 wide field fluorescence microscope, equipped with a 63× Plan Apochromat oil immersion lens and a 100 W adjustable mercury arc lamp (Zeiss, Oberkochen, Germany). Images were recorded with a cooled CCD camera (AxioCam MRm, Zeiss). Using Axio Vision (Version 4.5) imaging software, the relative fluorescence intensity of photolesions in spots of local damage and the relative fluorescence intensity of photolesions in nuclei without spots (background) were quantified; DAPI staining was used to identify cell nuclei. At least 20 images were taken per coverslip harboring at least 55 spots. For statistical analysis, a Student’s t test was performed comparing the mean relative fluorescence intensities of at least three independent experiments with each of the three coverslips to controls. Association and Dissociation of XP Proteins to Local UVC Damage Sites. To examine the impact of cadmium on damage association and dissociation of XPC, XPA, and XPG during NER, local UVC-irradiation combined with fluorescent antibody labeling was applied in human VH10hTert fibroblasts attached to coverslips. The fluorescent labeling was performed as described above, abstaining from the denaturating step (39, 40). Using Axio Vision (Version 4.5) imaging software, the relative fluorescence intensity
Cd Disturbs Nucleotide Excision Repair Machinery
Figure 1. Scanning electron microscopy of the applied CdO particles.
of XP proteins in spots of local damage and the relative fluorescence intensity of the XP proteins in nuclei without spots were measured; DAPI staining was used to identify cell nuclei. Since the used antibodies showed hardly any unspecific binding, fluorescence was mainly caused by the respective XP proteins. Additionally, specificity of the antibodies has been proven by applying XPC (XP21RO), XPA (XP25RO), and XPG (XPCS1RO) deficient cells as well as by Western blot analysis (39). To quantify the actual fluorescence intensities of DNA damage associated XP proteins, the relative fluorescence of the respective XP protein in nuclei without spots was subtracted from the relative fluorescence intensity of the XP protein in spots of local damage. At least 20 images were taken per coverslip harboring at least 55 spots. For statistical analysis, a Student’s t test was performed comparing the mean relative fluorescence intensities of at least three independent experiments with each of the three coverslips to controls. Additionally, in every coverslip the number of spots in 1000 cells were counted. Independent of the XP protein investigated, the number of spots was always around 300 with no significant differences between control cells and cadmium-exposed cells.
Results Specification of the Applied CdO Particles. One important prerequisite to elucidate cellular effects of particulate metal compounds is phagocytosis of the particles and bioavailability of the metal ions. As known from other metal particles, such as particulate NiO, one important factor is particle size. Therefore, particle size of the applied crystalline CdO was determined by scanning electron microscopy (SEM) and found to be
