Methylated Trivalent Arsenic Species Are Genotoxic - Chemical

Feb 28, 2001 - AS3MT, GSTO, and PNP polymorphisms: Impact on arsenic methylation and implications for disease susceptibility. Ray Antonelli , Kan Shao...
2 downloads 12 Views 209KB Size
Chem. Res. Toxicol. 2001, 14, 355-361

355

Methylated Trivalent Arsenic Species Are Genotoxic† Marc J. Mass,*,‡ Alan Tennant,‡ Barbara C. Roop,‡ William R. Cullen,§ Miroslav Styblo,| David J. Thomas,⊥ and Andrew D. Kligerman‡ Environmental Carcinogenesis Division (MD-68), National Health and Environmental Effects Research Laboratory/Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, Department of Chemistry, University of British Columbia, Vancouver, British Columbia, Canada, Department of Pediatrics, University of North Carolina, Chapel Hill, North Carolina 27599, and Environmental Toxicology Division, National Health and Environmental Effects Research Laboratory/Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711 Received December 14, 2000

The reactivities of methyloxoarsine (MAsIII) and iododimethylarsine (DMAsIII), two methylated trivalent arsenicals, toward supercoiled φX174 RFI DNA were assessed using a DNA nicking assay. The induction of DNA damage by these compounds in vitro in human peripheral lymphocytes was assessed using a single-cell gel (SCG, “comet”) assay. Both methylated trivalent arsenicals were able to nick and/or completely degrade φX174 DNA in vitro in 2 h incubations at 37 °C (pH 7.4) depending on concentration. MAsIII was effective at nicking φX174 DNA at 30 mM; however, at 150 µM DMAsIII, nicking could be observed. Exposure of φX174 DNA to sodium arsenite (iAsIII; from 1 nM up to 300 mM), sodium arsenate (from 1 µM to 1 M), and the pentavalent arsenicals, monomethylarsonic acid (from 1 µM to 3 M) and dimethylarsinic acid (from 0.1 to 300 mM), did not nick or degrade φX174 DNA under these conditions. In the SCG assay in human lymphocytes, methylated trivalent arsenicals were much more potent than any other arsenicals that were tested. On the basis of the slopes of the concentration-response curve for the tail moment in the SCG assay, MAsIII and DMAsIII were 77 and 386 times more potent than iAsIII, respectively. Because methylated trivalent arsenicals were the only arsenic compounds that were observed to damage naked DNA and required no exogenously added enzymatic or chemical activation systems, they are considered here to be direct-acting forms of arsenic that are genotoxic, though they are not, necessarily, the only genotoxic species of arsenic that could exist.

Introduction Arsenic is a natural contaminant of the geosphere (1) associated with the occurrence of adverse health effects and cancer throughout the world (2-7). Arsenic compounds leach from omnipresent arsenic-containing ores into water sources that ultimately are consumed as drinking water. The majority of disease linked to arsenic exposure is associated with chronic ingestion of contaminated drinking water. Arsenic is also an occupational carcinogen that is associated with lung cancer through inhalation of arsenic trioxide liberated after smelting of precious and semiprecious metals. Copper smelting, in particular, is an occupation that has been documented † This manuscript has been reviewed by the National Health and Environmental Effects Research Laboratory at the U.S. Environmental Protection Agency and approved for publication. The views expressed in this paper are those of the authors and do not necessarily represent the views or policy of the U.S. Environmental Protection Agency. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. * To whom requests for reprints and editorial correspondence should be addressed. E-mail: [email protected]. ‡ Environmental Carcinogenesis Division (MD-68), National Health and Environmental Effects Research Laboratory/Office of Research and Development, U.S. Environmental Protection Agency. § University of British Columbia. | University of North Carolina. ⊥ Environmental Toxicology Division, National Health and Environmental Effects Research Laboratory/Office of Research and Development, U.S. Environmental Protection Agency.

to be associated with increases in lung cancer attributed to arsenic (8). In recent times, the most extreme cases of arsenic-related poisoning involving hundreds of thousands of individuals have been reported in Bangladesh and West Bengal, India, due to ingestion of arseniccontaminated drinking water pumped from tube wells (9). Several modes of action have been proposed for arseniccontaining compounds as genotoxicants and as carcinogens (10). It has been asserted commonly that arsenicals act without direct interaction with DNA, although dimethylarsinic acid (DMAsV)1 is thought to undergo a peroxyl radical-mediated reaction with DNA at high concentrations (11). Inhibition of DNA ligase (12), disruption of polymerization of tubulin in the mitotic spindle (13, 14), interference with DNA methylation (15, 16), interference with signal transduction pathways (17-19), chronic stimulation of growth factors (20), and induction of oxidative damage (21, 22) have all been proposed as modes of action for the carcinogenicity of arsenicals. Although each of these modes of action could contribute 1 Abbreviations: iAs, generic term for inorganic forms of arsenic, including arsenite and arsenate; iAsIII, sodium arsenite; iAsV, sodium arsenate; MAsIII, methyloxoarsine (CH3AsIIIO); DMAsIII, iododimethylarsine [(CH3)2AsIIII]; MAsV, monomethylarsonic acid; DMAsV, dimethylarsinic acid; SCG, single-cell gel; MMS, methylmethanesulfonate; LMA, low-melting point agarose; DD water, sterile deionized distilled water.

10.1021/tx000251l CCC: $20.00 © 2001 American Chemical Society Published on Web 02/28/2001

356

Chem. Res. Toxicol., Vol. 14, No. 4, 2001

to an indirect mode of action for the carcinogenicity or genotoxicity of arsenicals, no single mode of action is widely accepted. Unlike the more stable and less genotoxic methylated metabolites of arsenic in the pentavalent state (MAsV and DMAsV ) (23, 24), methylated trivalent arsenic metabolites have been shown to be highly reactive (25, 26), and they are at least as toxic to mammalian cells in culture as iAsIII (27-29). Methylated trivalent arsenicals, both monomethylarsonous acid and dimethyarsinous acid, have been reported to be among the metabolites of iAs found in human urine (30, 31). In this preliminary communication, we report that MAsIII and DMAsIII nick and break DNA in vitro in a DNA nicking assay without the need for exogenously added enzymatic or chemical activation. These compounds are also effective DNAdamaging agents in the single-cell gel assay in human lymphocytes. The methylated trivalent arsenicals we assessed are approximately 77-386-fold more potent as DNA-damaging agents in the SCG assay than is iAsIII. Methylated pentavalent arsenicals are not active in the SCG assay. Among the trivalent and pentavalent arsenicals we assessed, only methylated trivalent arsenicals could nick or break DNA in the φX174 DNA nicking assay.

Materials and Methods Note: Arsenic compounds are toxic and potentially carcinogenic. Handle these compounds with appropriate safety precautions. Arsenic Compounds. iAsIII (>98% pure) and iAsV (>98% pure, as the heptahydrate) were obtained from Sigma Chemical Co. (St Louis, MO). DMAsV (99.6% pure, sodium salt) was obtained from Ansul Chemical Co. (Weslaco, TX). MAsV (99% pure, sodium salt) was obtained from Chem Services (West Chester, PA). MAsIII [99% pure;2 an X-ray crystallographic structure for AsIIICH3O is consistent with a tetramer (32)] and DMAsIII (99% pure2) were synthesized at the University of British Columbia. Briefly, MAsIII was synthesized from an aqueous solution of MAsV by reaction with SO2. The solution was evaporated to dryness and the oxide extracted with benzene (26). DMAsIII was prepared beginning with DMAsV dissolved in a solution of potassium iodide in sulfuric acid. Reduction with SO2 resulted in the separation of DMAsIII as a dense yellow liquid (33). MAsIII is presumed to form CH3AsIII(OH)2 (monomethylarsonous acid) per Cullen et al. (26), and by extension of similar chemical considerations, DMAsIII is presumed to form (CH3)2AsIIIOH (dimethylarsinous acid) in dilute aqueous solution. DNA Nicking Assay. φX174 RF I DNA was obtained from Amersham/Pharmacia Biotech (Piscataway, NJ). The arsenicals were dissolved in DD water and incubated in 15 µL reaction volumes each containing 50 ng of φX174 DNA buffered with 10 mM Tris-HCl and 1 mM EDTA (pH 7.4). The arsenicals, or the highest volume of DD water used as a control (0 µM), and DNA were incubated for 2 h at 37 °C. The entire 15 µL reaction mixtures were loaded onto 0.8% agarose gels that contained 50 µg of ethidium bromide/100 mL of agarose. Electrophoresis was performed in 0.5× Tris-borate EDTA buffer (pH 8) at 120 V for 90-120 min. The gel was photographed on a UV light box using a Kodak DC120 camera. The preparation of φX174 RF I DNA as certified by the manufacturer contained 85% RFI (closed supercoiled form) and ∼15% nicked (relaxed) form, each migrating as a distinct band. The determination of nicking was the comparison of the faster-migrating RFI form of φX174 DNA with the nicked form with lower electrophoretic mobility. When the 2W.

R. Cullen, personal communication.

Communications intensity of the nicked bands had become equal to that of the supercoiled form as determined visually, or the intensity of the band containing nicked DNA was greater than that of the supercoiled form, the DNA was considered to have sustained nicks. (This is a conservative estimate of nicking since progressive increases in the intensity of the upper band compared with the same band in control incubations can be seen with increasing concentrations of the DNA-damaging arsenicals.) All reaction preparations were performed under HEPAfiltered, laminar-flow, 100% exhaust, biological safety cabinets. Pipet tips and tubes used in reactions were certified by the manufacturer to be nuclease-free, and all preparations for the assays described above were routinely performed wearing nitrile or latex gloves as standard operating procedures to prevent contamination from nucleases that might be contained in fingerprints. In addition, we tested the DD water and methylated trivalent arsenicals for the presence of DNA-degrading activities before and after they had been passed through centrifugal filtration membranes (Millipore, Boston, MA), and no evidence for nuclease-associated DNA-degrading activities was detected. Further, the 1 mM EDTA contained in the TrisHCl buffer used above was found to completely inhibit the nicking activity of 0.1 unit (capable of degrading 100 ng of DNA in 15 min) of exogenously added DNAse I (Promega) in a 2 h incubation with 50 ng of φX174 RFI DNA at 37 °C. All arsenicals used in these studies have been used in various culture systems (phage, bacterial, and mammalian cell cultures) in other studies in our laboratories for incubation periods of up to 6 weeks at two physically separate institutions. In no instance was bacterial or fungal growth observed, nor was there any evidence of contamination of arsenicals by microorganisms. Single-Cell Gel Assay. The single-cell gel assay was performed on human peripheral blood lymphocytes from volunteers in accordance with an approved human subjects protocol. There were five male subjects and one female subject who were 3050 years in age. All were in good health, and none smoked cigarettes. Blood was collected by sterile venipuncture into Vacutainer tubes. Lymphocytes were isolated using a density gradient and then washed three times with phosphate-buffered saline. One million isolated lymphocytes per milliliter were then treated with either culture medium (vehicle control), 75 µM MMS (positive control), or each concentration of each arsenical in RPMI 1640 culture medium (GIBCO-BRL). The culture medium was the only vehicle used to dissolve the arsenicals used in these studies, and all were completely soluble at the concentrations that were employed. Cells were incubated for 2 h at 37 °C in a humidified incubator in an atmosphere of 5% CO2 in air. The cells were then washed twice in phosphatebuffered saline. Slides for the single-cell gel assay were prepared by dipping glass slides into 1% agarose (prepared in distilled water) and drying at 60 °C. Cells were added to the slides by mixing approximately 10 000 cells in 10 µL with 190 µL of 0.5% LMA held at 42 °C. Ninety microliters of this mixture was pipetted onto each of two slides and covered with a 24 mm × 50 mm glass coverslip. The slides were placed on ice for 5 min, after which the coverslip was removed, 90 µL of LMA was applied with a pipet, and the coverslip was replaced. After 5 min on ice, the coverslip was removed, and the slides were placed in a lysis buffer (pH 11.0) containing 2.5 M NaCl, 100 mM disodium EDTA, 10 mM Tris, 1% sarcosyl, 1% Triton X-100, and a 10% aqueous solution of freshly opened dimethyl sulfoxide. Slides were stored at 4 °C overnight in the lysis solution. Following lysis, the slides were placed in a denaturing electrophoresis buffer [300 mM NaOH and 1 mM EDTA (pH 13)] for 30 min Electrophoresis was performed for 20 min at 25 V (1.0 V/cm). The slides were neutralized with 0.4 M Tris-HCl (pH 7.5) for 15 min, placed in 95% ethanol for 5 min, and allowed to airdry overnight. Slides were stained with 35 µL of SYBR Green I (10× in distilled water; Molecular Probes, Eugene, OR) under a coverslip. Cells were viewed with a Leitz Orthoplan microscope equipped with a 100 W Hg fluorescent source and a Leitz I3 filter cube. Images of “comets” were captured using a Dage CCD

Communications

Chem. Res. Toxicol., Vol. 14, No. 4, 2001 357

Figure 1. Effect of iAsIII and DMAsV on the integrity of φX174 DNA. The reaction mixture consisted of 50 ng of φX174 DNA with the above arsenicals in 10 mM Tris-HCl and 1 mM EDTA at pH 7.4 and 37 °C for 2 h. The reaction mixtures were subjected to electrophoresis on a 0.8% agarose gel containing ethidium bromide: R, lanes containing various nicked (relaxed) forms of φX174 DNA; and S, lanes containing the supercoiled form of φX174 DNA. Nicking was defined as the concentration of arsenical at which the R and S forms were equal in intensity or where the R band was more intense than the S band as judged visually. The right-most lane contained a 1 kb DNA ladder as a molecular weight marker. camera interfaced with a personal computer. Slides were blind coded. Fifty cells per slide and two slides per treatment concentration were analyzed using Komet 3.0 (Kinetic Imaging) “comet” analysis software. Statistical Analysis of the SCG Assay. Using the Komet 3.0 software, the tail moment was calculated for each cell according to the formula tail moment ) tail length × % of total DNA in tail/100. Slides from each treatment were run with a concurrent vehicle control and positive control and were thus analyzed in their specific group. Simple linear regression was calculated using Statgraphics for Windows 3.0 (Manuguistics, Rockville, MD) and was used to compare the potency of each compound after subtraction of the concurrent vehicle controls in the tail moment.

Results Effect of Arsenicals on OX174 RFI DNA. We assessed the effect of iAsIII, iAsV, MAsV, DMAsV, MAsIII, and DMAsIII on the electrophoretic migration of φX174 RFI DNA after a 2 h incubation at 37 °C (pH 7.4). Neither iAsIII (ranging from 1 nM to 300 mM), iAsV (from 1 µM to 1 M), MAsV (from 1 µM to 3 M), nor DMAsV (from 0.1 to 300 mM) nicked or degraded φX174 DNA or changed its electrophoretic mobility under the conditions that were tested. Representative results seen with iAsIII and DMAsV are shown in Figure 1. We found that the only arsenic derivatives that changed the conformation of φX174 RFI DNA were MAsIII and DMAsIII. We observed two types of activities: a complete degradation of DNA at higher concentrations and a nicking activity at lower concentrations. MAsIII nicked φX174 DNA beginning at 30 mM (Figure 2). At higher MAsIII concentrations, between 30 and 60 mM, φX174 DNA was fully degraded and did not appear in the gel lanes after electrophoresis. In some cases, a light smear of ethidium bromide fluorescing material was seen at much greater electrophoretic mobility. DMAsIII was the most potent φX174 DNA nicking agent. It was effective beginning at 150 µM (Figure 2), where the band intensities are approximately equal for nicked and supercoiled forms. At concentrations of 10 mM, DMAsIII completely degraded φX174 RFI DNA, and there was no discernible band. Effect of Arsenicals in the SCG Assay in Human Lymphocytes. The SCG assay is an electrophoretic

Figure 2. Effect of MAsIII and DMAsIII on the integrity of φX174 DNA. The reaction mixture consisted of 50 ng of φX174 DNA with the above arsenicals in 10 mM Tris-HCl and 1 mM EDTA at pH 7.4 and 37 °C for 2 h. The reaction mixtures were subjected to electrophoresis on a 0.8% agarose gel containing ethidium bromide for 90-120 min: R, lanes containing various nicked (relaxed) forms of φX174 DNA; and S, lanes containing the supercoiled form of φX174 DNA. Nicking was defined as the concentration of arsenical at which the R and S forms were equal in intensity or where the R band was more intense than the S band as judged visually.

technique in which cells are embedded in an agarose matrix, subjected to alkaline lysis, and then a voltage is applied to separate the nuclear DNA by size. DNA that contains breaks and/or alkali labile lesions migrates from the embedded nucleus and forms a comet-like tail that can be seen and quantified when the gel is stained with a DNA-specific fluorochrome (Figure 3). The two inorganic arsenicals, iAsIII and iAsV that were assessed from 1 to 1000 µM, produced concentrationrelated linear increases in the level of DNA damage, with slopes of 0.0060 and 0.0085 microns/µM for the tail moment, respectively (Table 1). These slopes were not significantly different from each other (Figure 4A). MAsV and DMAsV were relatively inactive in the SCG assay at concentrations of up to 875 µM for MAsV or up to 1000 µM for DMAsV (Table 1 and Figure 4B). The two methylated trivalent arsenic derivatives were assessed from 1.25 to 80 µM for MAsIII and from 1.4 to 91 µM for DMAsIII , and they produced significant concentration-related increases in the slopes of the concentration-response curve for the “comet”-tail moment: 0.46 and 2.3 microns/µM for MAsIII and DMAsIII, respectively. Using the slope data, MAsIII was 54 or 77 times more potent than iAsV or iAsIII, respectively, and similarly, DMAsIII was 270 or 386 times more potent than iAsV or iAsIII, respectively (Table 1). DMAsIII produced concentration-related increases in the level of DNA damage which reached a plateau at 23 µM (Figure 4C). To summarize, the relative order of potencies of all the compounds tested in the SCG assay is as follows: DMAsIII > MAsIII . iAsV ∼ iAsIII > MAsV ∼ DMAsV. Thus, while MAsV and DMAsV appeared to be unable to damage DNA as assessed by the SCG assay, the trivalent methylated arsenicals MAsIII and DMAsIII were the most potent DNA-damaging agents assessed in these studies.

Discussion The results described here demonstrate that methylated trivalent arsenic derivatives can damage naked DNA without an exogenously added enzymatic or chemical activation system. We have confirmed that DNA damage occurs after exposure of human lymphocytes to methylated trivalent arsenic compounds. These compounds are the most potent direct-acting DNA-damaging arsenicals of which we are presently aware.

358

Chem. Res. Toxicol., Vol. 14, No. 4, 2001

Communications

Figure 3. Human leukocytes showing different levels of DNA damage using the SCG assay following exposure in culture, processed, and analyzed under identical conditions. The comets are representative of the majority of types seen in (A) control, up to 1 mM DMAsV, or up to 1 mM MAsV; (B) 1 mM iAsIII, 1 mM iAsV, or 10 µM MAsIII; and (C) exposures as low as 23 µM DMAsIII. Table 1. Effect of Trivalent and Pentavalent Arsenicals on Migration of DNA in the Comet Assaya arsenical slope (microns/µM) relative potency based upon slope iAsIII iAsV MAsIII MAsV DMAsIII DMAsV

0.0060 0.0085b 0.46c,d 0 2.3c-e 0

1 1.4 77