HIGHLIGHT Arsenite Methylation by Methylvitamin B 12 and Glutathione Does Not Require an Enzyme 1,2

Transcription

1 Toxicology and Applied Pharmacology 154, (1999) Article ID taap , available online at on Arsenite Methylation by Methylvitamin B 12 and Glutathione Does Not Require an Enzyme 1,2 Robert A. Zakharyan and H. Vasken Aposhian Department of Molecular and Cellular Biology, The University of Arizona, Tucson, Arizona Received August 25, 1998; accepted October 28, 1998 Arsenite Methylation by Methylvitamin B 12 and Glutathione Does Not Require an Enzyme. Zakharyan, R. A., and Aposhian, H. V. (1999). Toxicol. Appl. Pharmacol. 154, Although inorganic arsenic is methylated enzymatically by arsenic methyltransferases, which have been found in many mammalian livers, the detection of such enzymes has not been successful in surgically removed human livers. Results of the present experiments demonstrated that methylvitamin B 12 (methylcobalamin, CH 3 B 12 ) in the presence of thiols and inorganic arsenite can produce, in vitro, substantial amounts of monomethylarsonic acid (MMA) and small amounts of dimethylarsinic acid (DMA) in the absence of enzymes. Furthermore, this nonenzymatic methylation of inorganic arsenite by CH 3 B 12 was increased substantially by the presence of dimercaptopropanesulfonate (DMPS) and/or sodium selenite. The actions of DMPS and selenite together were additive. The methylation by CH 3 B 12 was neither inhibited nor stimulated by human liver cytosol. Since the amount of MMA produced by the in vitro system described in this study was not small, these results emphasize the need for a properly designed nutritional study in humans exposed to inorganic arsenic as to the relationship between vitamin B 12, selenium, and the metabolism of carcinogenic inorganic arsenic Academic Press Key Words: MMA; arsenic biotransformation; methylvitamin B 12 ; methylcobalamin; DMPS; GSH; selenium. Although the role of S-adenosylmethionine (SAM 3 in the methylation of inorganic arsenite in both prokaryotic and eukaryotic cells has been accepted for many years (Challenger, 1951; Cullen and Reimer, 1989), this has not been the case for 1 Supported in part by the Superfund Basic Research Program NIEHS Grant Number ES from the National Institute of Environmental Health Sciences and the Southwest Environmental Health Sciences Center P30-ES This is the seventh in a series of papers dealing with methylation of inorganic arsenite. The preceding paper is by Wildfang et al., DMA, dimethylarsinic acid; DMPS, Sodium 2,3-dimercapto-1-propane sulfonate; MMA, monomethylarsonic acid; SAM, S-adenosylmethionine; vitamin B 12, cobalamin B 12 ; methylvitamin B 12, methylcobalamin; CH 3 B 12, methyl-b 12. vitamin B 12 or its derivatives. Ten vitamin B 12 -dependent enzymes or enzyme systems have been isolated from bacteria, but only two have been found in the human (Glusker, 1995). The two B 12 derivatives in human metabolism are methylcobalamin and 5-deoxyadenosylcobalamin. Methylcobalamin is the coenzyme for methionine synthase and 5-deoxyadenosylcobalamin is for L-methylmalonyl CoA mutase. Methionine synthase transfers the methyl group of methyltetrahydrofolate to homocysteine with the formation of methionine and tetrahydrofolate. L-methylmalonyl-CoA is converted to succinyl- CoA by L-methylmalonyl-CoA mutase. For humans, the daily nutritional requirement of vitamin B 12 is3to5 g, which must be obtained from animal byproducts in the diet (Linnel, 1974; Dorscherholmen and Hagen, 1962). The human body has a vitamin B 12 content of about 5 mg and a daily loss of it of about 2 to 5 g (Friedrich, 1975; Chanarin, 1969). But vitamin B 12 is not found in plants. Therefore, diets that contain very little food derived from animals, as is the case in many economically poor populations, may result in a vitamin B 12 deficiency. The majority of populations that appear to have chronic inorganic arsenic exposure with its many signs and symptoms have been economically poor. However, the development of a vitamin B 12 deficiency for adults usually has been considered to be connected with intestinal diseases or defects (Nilsson et al., 1984), affecting gastrointestinal absorption. There have been a number of reports that suggest a connection between vitamin B 12 and arsenite methylation. McBride and Wolfe (1971) found that CH 3 B 12 functioned as the methyl donor in the biosynthesis of dimethylarsine from arsenate or arsenite in cell extracts of methanobacillus. Buchet and Lauwerys (1985) indicated that CH 3 B 12 can methylate arsenite at a low rate in the presence or absence of rat liver extracts. Although GSH was included in the assay, its essentiality was not shown. Styblo et al. (1995, et al. 1996) added 100 g CH 3 B 12 /ml in their arsenite methyltransferase assays, although experimental evidence for its requirement in the enzyme assay was not presented. Neither group reported that selenite and/or X/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.

2 288 DMPS stimulated this nonenzymatic methylation of arsenite by CH 3 B 12. It appears that epidemiologists studying the effects of chronic exposure to inorganic arsenic may have overlooked the possible relationship between vitamin B 12 deficiency and inorganic arsenic toxicity. Therefore, we decided that there was a need to better characterize this relationship. More knowledge about it might encourage epidemiologists to study the relationship between nutrition and inorganic arsenic toxicity. The present paper presents evidence that (a) inorganic arsenite was methylated by CH 3 B 12 in a simple nonenzymatic system to produce mainly MMA and small amounts of DMA; (b) in such a system, a reducing environment such as GSH, DMPS, or sodium selenite was required; and (c) selenite and the vicinal dithiol DMPS together stimulated this methylation in an additive manner. This proposed nonenzymatic pathway for the methylation of inorganic arsenite does not exclude enzymatic methylation of arsenite by SAM (for review, see Aposhian, 1997) or by vitamin B 12 derivatives. At this time, however, no one has detected an enzymatic methylation involving B 12 derivatives in mammalian liver. MATERIALS AND METHODS Reagents. Carrier-free [ 73 As]arsenate (5.85 Ci/ l) was obtained from Los Alamos National Laboratory (Los Alamos, NM). It was reduced to arsenite by the method of Reay and Asher (1977). Glutathione (GSH), S- methyl-adenosyl-l-methionine (SAM), cobalamin (vitamin B 12 ), and methylcobalamin (CH 3 B 12 ) were purchased from Sigma (St. Louis, MO). [ 14 C] Methylarsonic acid (4.5 mci/mmol) was purchased from American Radiolabeled Chemicals (St. Louis, MO). [ 14 C]Dimethylarsinic acid (11.2 mci/mmol) was a generous gift from Management Technology (Research Triangle Park, NC). Sodium selenite pentahydrate (Na 2 O 3 Se 5H 2 O [MW ]) was purchased from Fluka (Milwaukee, WI). DMPS was obtained from Heyl (Berlin, Germany). All other chemicals were analytical reagent grade or of the highest quality obtainable. Water was deionized and doubly distilled. Preparation of human liver cytosol. Biopsied human liver (8 g), immediately frozen in dry ice after removal, was obtained from The International Institute for the Advancement of Medicine (a Division of the Pennsylvania Regional Tissue Bank, Scranton, PA). Procedures for cytosol preparation were carried out at 4 C essentially as described by Zakharyan et al. (1995), except homogenization was performed using a Polytron homogenizer with homogenization buffer (1:2 w/v). Cytosol was stored at 70 C. The protein content of cytosol was determined by the method of Bradford (1976) using bovine serum albumin as the standard. Assays. The assay for arsenite methylation activity using CH 3 B 12 was as follows: 0.05 M Tris HCl buffer (ph 7.8), 10 mm GSH, 1 mm MgCl 2,1mM S-adenosyl-L-methionine, 24 g vitamin B 12,24 g CH 3 B 12, 6.6 g 73 As III / 186,153 cpm; and, when used, selenite (Na 2 SeO 3 5H 2 O), 4.9 to 39.2 g per assay; 5.1 mm DMPS, and 4.6 mg human liver cytosol protein in a total volume of 325 l (final ph 7.8). The tube was tightly capped at this point as well as during subsequent steps in this procedure. The reasons for tightly capping the tubes are the addition of H 2 O 2 to cytosol results in foaming and to prevent any as yet unknown volatile intermediates from escaping. Incubation was for 90 min at 37 C. The reaction mixtures were then made to 10% H 2 O 2 (Styblo et al., 1995) and incubated for 1hatroom temperature, boiled for 5 min, chilled, and centrifuged for 10 min in an Eppendorf centrifuge at 4 C. The supernatant was placed on a PRP-x100 anion-exchange column (Hamilton, TABLE 1 Nonenzymatic Methylation of Arsenite by CH 3 B 12 and GSH MMA formed (ng) Complete Minus Mg Minus SAM Minus B Minus CH 3 B 12 Not detectable 2 Minus GSH Not detectable 2 Note. Values are means SE. Complete system consisted of 0.05 M Tris HCl buffer (ph 7.8); 10 mm GSH; 1 mm MgCl 2 ;1mMS-adenosyl-Lmethionine; 24 g B 12 ;24 g CH 3 B 12 ; 73 As III, 6.6 g/186,153 cpm; in total volume of 325 l. Incubation was for 90 min at 37 C. This and other experiments in this study have been repeated and the results confirmed. The limits of detection of MMA in these experiments are 5 to 7 ng. n number of assays.. CH 3 B 12 Reaction: 73 As III O GSH Reno, NV); 25 cm 4.1 mm internal diameter. The column consists of spherical 10 m particles of a styrene-divinylbenzene copolymer with trimethylammonium exchange sites to detect putative products, such as [ 73 As]- MMA V and [ 73 As]-DMA. Detection was by a Beckman 171 Radioisotope Detector. The column was eluted with 50 mm sodium phosphate buffer (ph 6.0) (Gailer and Irgolic, 1994). Retention times for MMA and DMA were 2.86 min and 2.16 min, respectively, as were the standards. Separate experiments have shown that incubation with 10% H 2 O 2 at room temperature for 1 h followed by boiling for 5 min converted arsenite (As III ) to arsenate (As V ), and 95% of radioactivity was in the supernatant. Less than 5% of radioactivity was found in the precipitate when human liver cytosol was used in the reaction mixture. Confirmation of structure of methylated compounds by ion exchange. MMA and DMA identification was confirmed by the method of Tam et al. (1978). Reaction mixtures with and without cytosol, in presence of selenite or DMPS, were applied to a cm AG50W-x4, 100 to 200 mesh resin cation-exchange column (Bio-Rad), which had been preconditioned with 50 ml of 0.5 M HCl. The column was successively eluted with 7 ml of 0.5 M HCl, 8mlofH 2 O,8mlof5%NH 4 OH, and 21 ml of 20% NH 4 OH. MMA was eluted with 8 ml of H 2 O. DMA was eluted with 20% NH 4 OH. [ 14 C]MMA and [ 14 C]DMA were used as standards. RESULTS MMA. Methylation of arsenite by CH 3 B 12 and GSH. The nonenzymatic methylation of inorganic arsenite required methylcobalamin and GSH (Table 1). The absence of either in the reaction mixture completely prevented the nonenzymatic methylation of inorganic arsenite. Neither Mg ions, the methyl donor SAM, or the non-methylated form of vitamin B 12 was required (Table 1). Sodium selenite increased methylation. The addition of 19.6 g sodium selenite increased the methylation of inorganic arsenite fivefold in the presence of CH 3 B 12 and in the absence n

3 289 FIG. 1. Methylation of arsenite in absence or presence of human liver cytosol. Incubations and assays were performed using conditions described in Materials and Methods. For each column, the SEM and the number of assays, n, were as follows: 1, 1.96, n 3; 2, 12, n 2; 3, 10, n 2; 4, n 1; 5, 3.0, n 2; 6, 25.5, n 3. of human liver cytosol (Fig. 1, column 1 vs 4). In addition, the presence of human liver cytosol did not inhibit the methylation by methylcobalamin and GSH (Fig. 1, column 2 vs 5). In the nonenzymatic reaction at ph 7.8, CH 3 B 12 in the presence of GSH methylated 2.1% of the inorganic arsenite in the reaction mixture after incubation for 90 min at 37 C. When 39.2 g sodium selenite was added, 12% of the arsenite was methylated (data not shown). In addition, in the presence of sodium selenite (19.6 to 39.2 g) DMA (28 to 33 ng) was formed. DMPS and sodium selenite action were synergistic. DMPS, 5.1 mm, in the absence of both human liver cytosol and sodium selenite, stimulated methylation of inorganic arsenite by CH 3 B 12 approximately fourfold (Fig. 2, column 1 vs 2). The addition of 4.9 g Na selenite to the reaction mixture in the presence of DMPS approximately doubled the amount of MMA produced as compared to selenite alone (Fig. 2, column 3 vs 4). That both MMA and DMA were formed became obvious when a nonenzymatic reaction mixture was extracted and placed on an ion-exchange column. Both MMA and DMA were eluted (Fig. 3). It was clear that with CH 3 B 12, GSH, selenite, and DMPS present, DMA was formed, although in relatively small amounts. Human liver cytosol (2.3 mg of protein/reaction mixture) did not significantly affect formation of MMA (columns 1 and 5), in presence of sodium selenite and DMPS (columns 4 and 8), in presence of only sodium selenite (columns 3 and 7), or in presence of only DMPS (columns 2 and 6). None of these reactions was significantly inhibited or stimulated by human liver cytosol. FIG. 2. DMPS and/or sodium selenite stimulated nonenzymatic methylation of arsenite by CH 3 cobalamin. Incubations and assays were performed using conditions described in Materials and Methods. For each column, the standard error of the mean and the number of assays, n, were as follows: 1, 1.86, n 3; 2, 1.00, n 2; 3, 12.00, n 2; 4, n 1; 5, 2.85, n 3; 6, 0.33, n 3; 7, 5.48, n 4; 8, 12.0, n 2. DISCUSSION The number of humans in developing countries drinking water containing amounts of inorganic arsenate/arsenite that are known to cause cancer of the skin and internal organs is FIG. 3. Confirmation of MMA and DMA formed after nonenzymatic methylation of arsenite by CH 3 B 12. The reaction mixture was extracted and chromatographed on Dowex 50, 100 to 200 mesh cation-exchange column ( cm) by method of Tam et al. (1978), see Materials and Methods.

4 290 substantial. For example, in West Bengal, India, the number of people at risk has been estimated to be 8 million (Guha Mazumder et al., 1998). The nutritional condition of these economically poor populations and its relation to arsenic carcinogenesis remains a neglected area of investigation. Actual measurements of methionine, vitamin B 12, and/or selenium in the diet of such populations are lacking. This is so, even when it has been known that the essential amino acid methionine is a precursor of SAM, the cofactor for the methylation of inorganic arsenite and MMA (Vahter and Marafante, 1987) by the highly purified rabbit liver methyltransferase that catalyzes the reactions for the biotransformation of inorganic arsenite to methylated derivatives (Zakharyan et al., 1995; Aposhian, 1997). It should be emphasized that the detection of arsenite methyltransferase in human liver has been, as yet, unsuccessful (Buchet and Lauwerys, 1985, 1988; Zakharyan and Aposhian, unpublished). In addition, a number of mammals appear to be deficient or lacking the arsenic methyltransferase enzymes (Aposhian, 1997; Healy et al., 1997; Zakharyan et al., 1996). SAM does not nonenzymatically transfer methyl moieties to arsenite (Zakharyan et al., 1995). Our studies clearly show a relationship between CH 3 B 12, glutathione, selenite, and arsenite methylation. Chemically, methylcobalamin is highly reactive (Wood and Fanchiang, 1979). For example, it will nonenzymatically transfer the methyl group from CH 3 cobalamin to the thiolate of homocysteine yielding methionine. The rate of this reaction was increased sevenfold by using partially purified Escherichia coli methionine synthetase, which has CH 3 cobalamin as its coenzyme (Guest et al., 1964). Glutathionylcobalamin has been identified as an intermediate in the formation of cobalamin coenzymes (Pezacka et al., 1990). Chemical model studies have shown that glutathionylcobalamin, in the presence of a thiol substrate, is converted to cobalt I since the subsequent addition of SAM to the incubation medium resulted in CH 3 cobalamin formation (Pezacka et al., 1988). The identification of glutathionylcobalamin in mammalian cells and the methylation of this compound by SAM in a nonenzymatic reaction suggest that glutathionylcobalamin has physiological significance. Methylcobalamin can transfer by nonenzymatic means its methyl group from cobalt to heavy metal ions. To do this, cleavage of the Co C bond is required. This bond can be severed under different conditions to yield a carbanion (CH 3 ), a radical (CH 3 ), or carbonium ion (CH 3 ). The methylation by methylcobalamin of mercuric ion (Hg 2 ), lead (Pb 2 ), and palladium (Pd 2 ) are examples of an electrophilic attack, with the carbanion (CH 3 ) reacting with positively charged metal ions. However, this does not appear to be the case with methylation of arsenite. Hogenkamp et al. (1985, 1987) demonstrated heterolytic cleavage of the carbon cobalt bond of methylcobalamin by the nucleophilic thiolate anion under relatively mild conditions. In our ph 7.8 reaction mixture with its highly reducing conditions, methylation of arsenite by CH 3 B 12 may occur by nucleophilic attack on the Co C bond by an arsenite GSH complex. Schrauzer et al. s proposal (1973) for a thiol-promoted electrophilic attack on the Co C bond and transfer of CH 3 group as a crypto carbanion to arsenic salt is less likely in the highly reductive conditions of our experiments. With selenite present, arsenite methylation by methylcobalamin may occur as follows: selenite in presence of GSH will form selenide, which will replace glutathione in the (GS) 3 As complex. As III in this compound will be methylated by CH 3 B 12 with nucleophilic attack on Co C bond by arsenite of lower oxidation state. Formation of dimethylselenide is not excluded. Folate also appears to be related to this overall problem. One of folate s important metabolic roles is to be the coenzyme that transfers single carbon units to a number of different acceptors. This coenzyme has a role in amino acid and nucleotide metabolism. Vitamin B 12 and folate are involved in methionine synthesis. If there is a vitamin B 12 deficiency, folate will become metabolically unavailable for other reactions. For this reason, the clinical symptoms of B 12 deficiency are indistinguishable from a deficiency of folate. Since methionine is an essential amino acid and a precursor of SAM, the cofactor for purified rabbit liver arsenic methyltransferase (Aposhian, 1997), and since methyl B 12, GSH, and selenite will methylate inorganic arsenite nonenzymatically (Figs. 1 and 2), it is important to carefully analyze the vitamin B 12, selenium, and methionine content in the diet of populations chronically exposed to inorganic arsenic. REFERENCES Aposhian, H. V. (1997). Enzymatic methylation of arsenic species and other new approaches to arsenic toxicity. Annu. Rev. Pharmacol. Toxicol. 37, Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, Buchet, J. P., and Lauwerys, R. (1985). Study of inorganic arsenic methylation by rat liver in vitro: Relevance for the interpretation of observations in man. Arch. Toxicol. 57, Buchet, J. P., and Lauwerys, R. (1988). Role of thiols in the in vitro methylation of inorganic arsenic by rat liver cytosol. Biochem. Pharmacol. 37, Challenger, F. (1951). Biological methylation. Adv. Enzymol. 12, Chanarin, I. (1969). The Megaloblastic Anaemias. Blackwell Scientific, Oxford. Cullen, W. R., and Reimer, K. J. (1989). Arsenic speciation in the environment. Chem. Rev. 89, Dorscherholmen, A., and Hagen, P. S. (1962). Kinetics of ingested radiocyanocobalamin. In Man, Vitamin B 12 and Intrinsic Factor (K. C. Heinrich, and F. Enke, Eds.), pp , Verlag, Stuttgart. Friedrich, W. (1975). Vitamin B 12 and verwandte corrinoide. In Fermente- Hormone-Vitamine (R. Ammon, and G. T. Dirschirl, Eds.), Vol. III/2, pp , Verlag, Stuttgart. Gailer, J., and Irgolic, K. J. (1994). The ion-chromatography behavior of

5 291 arsenite, arsenate, methylarsonic acid and dimethylarsinic acid on the Hamilton PRP-X100 anion exchange column. Appl. Organomet. Chem. 8, Glusker, J. P. (1995). Vitamin B 12 and the B 12 coenzymes. In Vitamins and Hormones (G. Litwack, Ed.), pp. 1 76, Academic Press, New York. Guest, J. R., Friendman, S., Dilworth, M. J., and Woods, D. D. (1964). Methylcobalamin as a source of the methyl group of methionine. Ann. NY Acad. Sci. 112, Guha Mazumder, D. N., Das Gupta, J., Santra, A., Pal, A., Ghose, A., and Sarkar, S. (1998). Chronic arsenic toxicity in West Bengal The worst calamity in the world. J. Ind. Med. Assoc. 96, 4 7, 18. Healy, S. M., Zakharyan, R. A., and Aposhian, H. V. (1997). Enzymatic methylation of arsenic compounds: IV. In vitro and in vivo deficiency of the methylation of arsenite and monomethylarsonic acid in the guinea pig. Mutat. Res. 386, [erratum appears in Mutat. Res. 387, 173]. Hogenkamp, H. P., Bratt, G. T., and Kotchevar, A. T. (1987). Reaction of alkylcobalamins with thiols. Biochemistry 26, Hogenkamp, H. P., Bratt, G. T., and Sun, S.-Z. (1985). Methyl transfer from methylcobalamin to thiols. A reinvestigation. Biochemistry 24, Linnel, T. C. (1974). The fate of cobalamin in vivo. In Cobalamin: Biochemistry and Pathology (B. M. Babior, Ed.), pp , John Willey and Sons, New York. McBride, B. C., and Wolfe, R. S. (1971). Biosynthesis of dimethylarsine by methanobacterium. Biochemistry 10, Nilsson, L. O., Myrvold, H. E., Swolin, B., and Ojerskog, B. (1984). Vitamin B 12 in plasma in patients with continent ileostomy and long observation time. Scand. J. Gastroenterol. 19, Pezacka, E., Green, R., and Jacobsen, D. W. (1990). Glutathionylcobalamin as an intermediate in the formation of cobalamin coenzymes. Biochem. Biophys. Res. Commun. 169, Pezacka, E. H., Denison, C. J., Green, R., and Jacobsen, D. W. (1988). Biosynthesis of methylcobalamin: Chemical model studies with thiol cobalamin adducts and adenosylmethionine. J. Cell Biol. 107, Reay, P. F., and Asher, C. J. (1977). Preparation and purification of 74 Aslabeled arsenate and arsenite for use in biological experiments. Anal. Biochem. 78, Schrauzer, G. N., Seck, J., Holland, R., Beckhan, T., Rubin, E., and Sibert, J. (1973). Reductive dealkylation of alkylcobaloximes, alkylcobalamins, and related compounds. Simulation of corrin dependent reductase and methylgroup transfer reactions. Bioinorg. Chem. 2, Styblo, M., Delnomdedieu, M., and Thomas, D. J. (1996). Mono- and dimethylation of arsenic in rat liver cytosol in vitro. Chem. Biol. Interact. 99, Styblo, M., Yamauchi, H., and Thomas, D. J. (1995). Comparative in vitro methylation of trivalent and pentavalent arsenicals. Toxicol. Appl. Pharmacol. 135, Tam, K. H., Charbonneau, S. M., Bryce, F., and Lacroix, G. (1978). Separation of arsenic metabolites in dog plasma and urine following intravenous injection of 74 As. Anal. Biochem. 86, Vahter, M., and Marafante, E. (1987). Effects of low dietary intake of methionine, choline or proteins on the biotransformation of arsenite in the rabbit. Toxicol. Lett. 37, Wildfang, E., Zakharyan, R. A., and Aposhian, H. V. (1998). Enzymatic methylation of arsenic compounds: VI. Characterization of hamster liver arsenite and methylarsonic acid methyltransferase activities in vitro. Toxicol. Appl. Pharmacol. 152, Wood, J. M., and Fanchiang, Y. T. (1979). Mechanism for B 12 -dependent methylation. In Vitamin B 12 (B. Zagalak, and W. Friedrich, Eds.), pp , Walter de Gruyter, Berlin, New York. Zakharyan, R. A., Wildfang, E., and Aposhian, H. V. (1996). Enzymatic methylation of arsenic compounds: III. The marmoset and tamarin, but not the rhesus, monkey are deficient in methyltransferases that methylate inorganic arsenic. Toxicol. Appl. Pharmacol. 140, Zakharyan, R., Wu, Y., Bogdan, G. M., and Aposhian, H. V. (1995). Enzymatic methylation of arsenic compounds. I: Assay, partial purification, and properties of arsenite methyltransferase and monomethylarsonic acid methyltransferase of rabbit liver. Chem. Res. Toxicol. 8,