500
Chem. Res. Toxicol. 2001, 14, 500-510
Chromium(VI) Reduction by Catechol(amine)s Results in DNA Cleavage in Vitro: Relevance to Chromium Genotoxicity David I. Pattison,† Michael J. Davies,‡ Aviva Levina,† Nicholas E. Dixon,§ and Peter A. Lay*,† Centre for Heavy Metals Research, School of Chemistry, University of Sydney, NSW 2006, Australia, The Heart Research Institute, 145-147 Missenden Road, Camperdown, NSW 2050, Australia, and Research School of Chemistry, Australian National University, Canberra, ACT 0200, Australia Received October 31, 2000
Catechols are found extensively in nature both as essential biomolecules and as the byproducts of normal oxidative damage of amino acids and proteins. They are also present in cigarette smoke and other atmospheric pollutants. Here, the interactions of reactive species generated in Cr(VI)/catechol(amine) mixtures with plasmid DNA have been investigated to model a potential route to Cr(VI)-induced genotoxicity. Reduction of Cr(VI) by 3,4-dihydroxyphenylalanine (DOPA) (1), dopamine (2), or adrenaline (3) produces species that cause extensive DNA damage, but the products of similar reactions with catechol (4) or 4-tert-butylcatechol (5) do not damage DNA. The Cr(VI)/catechol(amine) reactions have been studied at low added H2O2 concentrations, which lead to enhanced DNA cleavage with 1 and induce DNA cleavage with 4. The Cr(V) and organic intermediates generated by the reactions of Cr(VI) with 1 or 4 in the presence of H2O2 were characterized by EPR spectroscopy. The detected signals were assigned to Cr(V)-catechol, Cr(V)-peroxo, and mixed Cr(V)-catechol-peroxo complexes. Oxygen consumption during the reactions of Cr(VI) with 1, 2, 4, and 5 was studied, and H2O2 production was quantified. Reactions of Cr(VI) with 1 and 2, but not 4 and 5, consume considerable amounts of dissolved O2, and give extensive H2O2 production. Extents of oxygen consumption and H2O2 production during the reaction of Cr(VI) with enzymatically generated 1 and N-acetyl-DOPA (from the reaction of Tyr and N-acetyl-Tyr with tyrosinase, respectively) were correlated with the DNA cleaving abilities of the products of these reactions. The reaction of Cr(VI) with enzymatically generated 1 produced significant amounts of H2O2 and caused significant DNA damage, but the N-acetyl-DOPA did not. The extent of in vitro DNA damage is reduced considerably by treatment of the Cr(VI)/catechol(amine) mixtures with catalase, which shows that the DNA damage is H2O2-dependent and that the major reactive intermediates are likely to be Cr(V)-peroxo and mixed Cr(V)-catechol-peroxo complexes, rather than Cr(V)-catechol intermediates.
Introduction Chromium(VI) is known to be a major industrial carcinogen (1), but it does not cause appreciable DNA damage in vitro (2, 3). An uptake-reduction mechanism is believed to be the method by which Cr(VI) carcinogenesis occurs (4). This involves the reduction of Cr(VI) by intracellular reductants to generate Cr(V) and Cr(IV) intermediates and finally Cr(III), but Cr(V/IV) intermediates are proposed to cause DNA damage that leads ultimately to tumor development (4). In vitro studies indicate that DNA damage [e.g., single-strand breaks, protein-DNA cross-linking, and formation of Cr(III)DNA complexes] occurs when Cr(VI) is reduced by ascorbate and glutathione (5, 6, and references therein). Moreover, relatively stable Cr(V) and Cr(IV) complexes cause DNA damage in vitro and mutations in bacterial and mammalian cells (7-11). Other studies have shown † ‡ §
University of Sydney. The Heart Research Institute. Australian National University.
that Cr(V) complexes can be detected by EPR spectroscopy in animals that have been exposed to Cr(VI) (12, 13). Reactive oxygen species (ROS)1 have also been implicated in these reactions by studies of the O2 and catalase dependencies of in vitro DNA damage (14-18, and references therein). However, recent studies have shown that the DNA damage is different from that induced by • OH (5, 19), and there is strong evidence that in vitro DNA damage originates from the interaction of Cr(V)peroxo species with reductants such as ascorbate (15, 20). EPR studies of reactions of Cr(VI) with ascorbic acid and/ or H2O2 showed that transient Cr(V)-peroxo and mixed Cr(V)-ascorbato-peroxo complexes are generated (2123), while assessments of O2 consumption during Cr(VI) reduction by ascorbate are consistent with mixed Cr(V)peroxo-ascorbato complexes causing in vitro DNA damage (23). However, O2 consumption also occurs in the 1 Abbreviations: DOPA, 3,4-dihydroxyphenylalanine; ROS, reactive oxygen species; TBC, 4-tert-butylcatechol.
10.1021/tx000229s CCC: $20.00 © 2001 American Chemical Society Published on Web 04/26/2001
Chromium(VI) Reduction by Catechol(amine)s
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 501
Chart 1. Structures of the Catechol(amine)s Used in This Study, and Two of the Primary DOPA Oxidation Products, Dopaquinone and Dopachrome
subsequent oxidation of organic radical intermediates generated by interaction of Cr(VI/V/IV) with biological reductants, and these radicals may also be capable of cleaving DNA (23). To gain a better understanding of the initial molecular events that ultimately lead to Cr(VI) carcinogenicity, it is necessary to consider all possible intracellular reductants (24). Catechols (1,2-benzenediols) constitute another redox active group found in mammalian cells (e.g., hormones, neurotransmitters, and in oxidized proteins) and are relevant to some lung cancers (e.g., via the metabolism of aromatic pollutants), but their reactivity with Cr(VI) has only recently been studied in detail (25). The reactions of Cr(VI) with a variety of catechol(amine)s have been studied by EPR spectroscopy and give rise to several Cr(V) complexes and semiquinone radicals under certain conditions (25). Preliminary studies of the DNA cleaving ability of these reaction products have been reported (26). In addition, Liu et al. (27) showed that treatment of DNA with Cr(VI) and cigarette smoke solutions (which contain catecholic compounds) causes DNA strand breaks at levels greater than those of solutions of Cr(VI) or cigarette smoke alone. They proposed that the active species are •OH produced from Fenton-like reactions of Cr(VI) with H2O2 generated by a variety of mechanisms, including catechol autoxidation. It has been shown that fractionated cigarette smoke solutions cause DNA damage, and the catechol-containing fractions are the most damaging (28). The ability of catechol-quinone compounds to oxidize DNA in the presence of Cu(II/I) was demonstrated previously, both with metabolized aromatic air pollutants (29) and with free and protein-bound 3,4-dihydroxyphenylalanine (DOPA) residues (30, 31). DNA damage was proposed to be •OH-mediated, but the role of semiquinone radicals was not discounted (29-31). Reported here are studies of the in vitro reactions of Cr(VI)/catechol(amine) mixtures with plasmid DNA. The
effect of added H2O2 was studied in conjunction with assessments of O2 consumption and H2O2 production during the Cr(VI) reduction reactions. Further EPR experiments show that additional Cr(V) complexes are generated when H2O2 is added to the Cr(VI)/catechol(amine) reactions.
Materials and Methods Caution: Cr(VI) is mutagenic and carcinogenic (1), and the intermediates generated in the reduction of Cr(VI) by catechol(amine)s are capable of cleaving DNA (26). Appropriate precautions should be taken to avoid skin contact and dust inhalation while handling these chemicals. Materials. The catechol(amine)s (Chart 1) D,L-DOPA, dopamine (hydrochloride salt), adrenaline (all from Sigma), catechol, and 4-tert-butylcatechol (TBC) (Aldrich) were used as received. Fresh catechol(amine) stock solutions were prepared daily to minimize autoxidation. K2Cr2O7 (Merck, 99.5%) was used as a source of Cr(VI) without further purification. A solution of H2O2 (DHA, 3%) was standardized by iodometric titration (32). Catalase (from bovine liver, 1880 units/mg), tyrosinase (from mushroom, 3400 units/mg), Tyr, and N-acetyl-Tyr were from Sigma. The phosphate buffer (50 mM) used for all experiments was prepared by dilution of Chelex-treated 1.0 M NaH2PO4 (Merck) with Milli-Q water, and its pH was adjusted to 7.4 by addition of NaOH (Aldrich, 99.99%). Plasmid pUC9 DNA (2.67 × 103 base pairs) was prepared and dialyzed as described previously (7). EPR Spectroscopy. A Bruker EMX spectrometer was used for recording X-band EPR spectra from solutions contained in a Wilmad quartz flat cell. Spectra were calibrated with a Bruker EMX035M NMR gaussmeter in conjunction with an EMX048T Microwave Bridge Controller and an EMX032T Field Controller. The majority of the EPR spectra were acquired with the following parameters: center field, 354 mT; sweep width, 20 mT; microwave frequency, ∼9.7 GHz; microwave power, 30 mW; and modulation frequency, 100 kHz (other instrument parameters are given in the figure legends). EPR spectra were analyzed using WinEPR (33), and simulations were performed using WinSim (version 0.96, National Institute of Environmen-
502
Chem. Res. Toxicol., Vol. 14, No. 5, 2001
tal and Health Sciences) (34). All the spectra presented were filtered with the moving average method over seven experimental data points (WinEPR) (33). Typical concentrations {stated as [Cr(VI)]:[catechol(amine)]} used for acquisition of EPR spectra were (i) 10 mM:23 mM (4: 9) for excess catechol(amine) and (ii) 80 mM:5.0 mM (16:1) for excess Cr(VI). In most experiments, the H2O2 concentration was 9.2 mM, but lower H2O2 concentrations (0.23-4.6 mM) were also used. DNA Cleavage Assays. For studies with prereacted Cr(VI)/ catechol(amine) mixtures, Cr(VI) and the catechol(amine) were mixed in phosphate buffer. The reactant ratios {[Cr(VI)]: [DOPA], both in millimolar} were 80:5.0, 10:5.0, or 10:23 (to correlate with previous EPR studies) (25). Following reaction at 22 °C (for times of 0.5 or 30 min), the mixtures (3 µL) were diluted into phosphate-buffered solutions (12 µL) containing predominantly type I (negatively supercoiled) DNA (300 ng); the final Cr(VI) concentration was 2.0 mM (except when the Cr(VI) concentration was varied). After incubation at 37 °C for 1 h, reactions were halted by cooling to 4 °C. Bromophenol blue loading buffer (0.25% in 40% sucrose solution, 5 µL) was added, and samples were kept at 4 °C until DNA products were resolved by agarose gel electrophoresis. The prereacted anoxic samples were prepared in a glovebag under an Ar atmosphere, and then used to treat DNA for 4 h at 22 °C under Ar. The reactions of Cr(VI) (5.0 mM) with dopamine, catechol, or 4-tert-butylcatechol (TBC) (0.25-20 mM) were allowed to proceed for 0.5 min prior to dilution into DNA-containing solutions to give a Cr(VI) concentration of 1.0 mM, with subsequent incubation at 37 °C for 1 h. When the reactions were studied in situ, with reduction of Cr(VI) taking place in the presence of DNA, the samples were kept at 4 °C during addition of the reagents. Unless otherwise stated, components were added in the following order: DNA, phosphate buffer, Cr(VI), and, finally, freshly prepared catechol(amine). The samples were subsequently incubated at 37 °C for 1 h, before being processed as described above. Gel electrophoresis (∼20 h at ∼1 V/cm) was carried out using a 1.25% agarose gel in TBE buffer (45 mM Tris, 45 mM borate, and 1.0 mM Na2EDTA) containing 0.50 µg/mL ethidium bromide. Analysis of Agarose Gels. Gels were photographed and analyzed as described previously (7-11), except that the positive prints were scanned into an IBM-compatible computer using a ScanMaker IIsp scanner with a resolution of 300 dpi. The results are reported as the percentage of type I DNA in the control samples that was converted to type II DNA, and are representative of results from a series of gels prepared under identical conditions. The variation between duplicate samples that had been resolved on the same gel was
