In VitroMethylation of Inorganic Arsenic in Mouse Intestinal Cecum

TOXICOLOGY AND APPLIED PHARMACOLOGY ARTICLE NO.

147, 101–109 (1997)

TO978269

In Vitro Methylation of Inorganic Arsenic in Mouse Intestinal Cecum Larry L. Hall,* S. Elizabeth George,† Michael J. Kohan,† Miroslav Styblo,‡,1 and David J. Thomas* *Pharmacokinetics Branch, Experimental Toxicology Division, †Genetic and Cellular Toxicology Branch, Environmental Carcinogenesis Division, National Health and Environmental Effects Research Laboratory, United States Environmental Protection Agency, Research Triangle Park, North Carolina 27711; and ‡Curriculum in Toxicology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599 Received March 31, 1997; accepted July 21, 1997

In vitro methylation of inorganic arsenic in mouse intestinal cecum. Hall, L. L., George, S. E., Kohan, M. J., Styblo, M., and Thomas, D. J. (1997). Toxicol. Appl. Pharmacol. 147, 101–109. The capacity of mouse intestinal cecal microflora to methylate inorganic arsenicals (iAs) was examined in vitro under conditions of restricted bacterial growth. Cecal contents incubated under anaerobic conditions at 377C for 21 hr methylated up to 40% of either 0.1 mM arsenite (iAsIII) or 0.1 mM arsenate (iAsV). Methylarsenic (MAs) was the predominant metabolite; however, about 3% of either substrate was converted to dimethylarsenic (DMAs). Over the first 6 hr, the rate of methylation was several times greater for iAsIII than for iAsV. There was a 3-hr delay in the production of methylated metabolites from iAsV, suggesting that reduction of iAsV to iAsIII before methylation could be rate limiting. Over the concentration range of 0.1 to 10 mM of iAsIII or iAsV, there was an approximately linear increase in the production of MAs and DMAs. There was evidence of saturation or inhibition of methylation at 100 mM of either substrate. Substrate concentration had little effect on MAs/DMAs ratio. Incubation of cecal contents at 07C abolished methylation of either arsenical. Under aerobic or anaerobic conditions, cecal tissue homogenates produced little MAs or DMAs from either arsenical. Addition of potential methyl group donors, L-methionine and methylcobalamin, into cecal contents significantly increased the rate of methylation, especially for iAsV. Addition of glutathione, but not L-cysteine, had a similar effect. Selenite, a recognized inhibitor of iAs methylation in mammalian tissues, inhibited methylation of either substrate by cecal contents. These data suggest that cecal microflora are a high capacity methylation system that might contribute significantly to methylation of iAs in intact animals.

Human populations worldwide are exposed to inorganic arsenate (iAsV) and inorganic arsenite (iAsIII) by consumption of contaminated water and food (Thornton and Farago, 1996). Chronic exposure to iAsV or iAsIII from these sources has been associated with increased incidences of cancer of 1 To whom correspondence should be addressed at Frank Porter Graham Child Development Center, The University of North Carolina at Chapel Hill, CB#8180, Chapel Hill, NC 27599-8185. Fax: (919) 966-7532. E-mail: [email protected].

skin, bladder, and lung (Guo et al., 1994). In many species, iAsV is rapidly reduced to iAsIII and methylated to yield mono-, di-, and trimethylated metabolites (Styblo et al., 1995a). Glutathione (GSH) can reduce pentavalent arsenicals (including iAsV) to trivalency both in aqueous systems (Scott et al., 1993; Delnomdedieu et al., 1994a) and in intact erythrocytes (Delnomdedieu et al., 1994b). The oxidative methylation of iAs is catalyzed by arsenic methyltransferase(s) that use S-adenosyl-L-methionine (AdoMet) as a methyl group donor (Buchet and Lauwerys, 1985; Zakharyan et al., 1995; Styblo et al., 1996a). Both methylarsonate (MAsV) and dimethylarsinate (DMAsV) are found in the urine of many mammalian species following exposure to iAs; DMAsV is commonly the major urinary metabolite (Styblo et al., 1995a). The acute toxicities of MAsV and DMAsV in laboratory species are manyfold lower than that of either iAsIII or iAsV, suggesting that methylation is a detoxification process (Yamauchi and Fowler, 1994). However, little is known about the toxicity of trivalent methylated species (MAsIII, DMAsIII), the putative intermediates in arsenic biomethylation (Cullen et al., 1984; Styblo et al., 1995b). Buchet and Lauwerys (1985) reported that liver cytosol was the main site for the methylation of iAs in rat. Using in vitro techniques, arsenic methylation in rat liver cytosol has been further characterized with respect to the substrate specificity and effects of essential cofactors (GSH, AdoMet) and selected inhibitors (Buchet and Lauwerys, 1985, 1988; Styblo et al., 1995b, 1996a). An enzyme that catalyzes both monoand dimethylation of iAs has been purified from rabbit liver cytosol (Zakharyan et al., 1995). In addition to biomethylation of iAs in mammalian tissues, the role of metabolism by the microflora of the gastrointestinal (GI) tract warrants consideration. A variety of microorganisms methylate iAs to mono-, di-, or trimethylated derivatives (Gadd, 1993); however, the extent of metabolism in GI tract has not been well characterized. Rowland and Davies (1981) investigated the in vitro metabolism of arsenic at different sites in the GI tract of the rat. The contents of cecum or small intestine reduced iAsV to iAsIII. Small amounts of MAs and DMAs were formed during a 120hr incubation of iAsV with cecal contents, but not during

101

AID

TAP 8269

/

6h21$$$321

10-10-97 10:37:45

0041-008X/97

toxc

AP: TAP

102

HALL ET AL.

incubation with contents of small intestine (Rowland and Davies, 1981). The present work provides new data on the in vitro methylation of iAs by cecal contents from the mouse. These data were obtained under conditions of limited bacterial growth that permitted better characterization of the kinetics of methylation of iAs. Over a wide range of substrate concentrations, extensive methylation of iAsV or iAsIII by cecal contents was observed. For either iAs species, MAs was the predominant metabolite formed by cecal contents. iAsIII was methylated more rapidly than was iAsV. Addition of some methyl group donors or thiols increased the rate of methylation, especially for iAsV. The results presented here suggest that methylation by microflora in GI tract could be an important factor in the whole body metabolism of iAs. METHODS Radiolabeled compounds. Pentavalent radiolabeled arsenicals used in these experiments included [73As]arsenic acid (iAsV, sp act Ç1.5 1 106 mCi/mmol, Meson Production Facility, Los Alamos National Laboratory, NM), [14C]methylarsonic acid, disodium salt (MAsV, 10 mCi/mmol, ICN Radiochemicals, Irvine, CA), and [14C]dimethylarsinic acid, sodium salt (DMAsV, 10 mCi/mmol, ICN Radiochemicals). Radiolabeled trivalent arsenicals, iAsIII, MAsIII, and DMAsIII, were prepared from corresponding pentavalent species by reduction with a metabisulfite–thiosulfate reagent (Reay and Asher, 1977; Styblo et al., 1995b). Identities of the putative products of the reduction were confirmed using a previously described TLC technique (Styblo et al., 1995b). Solutions of radiolabeled arsenicals were prepared in deionized distilled water. Sodium arsenate (Sigma) or sodium arsenite (Sigma) were added to obtain desired concentrations of 73As-labeled iAsV or iAsIII in incubation mixtures; 14C-labeled methylated arsenicals were used without addition of carrier. Preparation of cecal contents. Adult CD-1 male mice (41–47 g) (Charles River Laboratories, Raleigh, NC) were killed by CO2 asphyxiation and placed into an anaerobic chamber (Coy Laboratory Products, Grass Lake, MI) containing an atmosphere of 10% H2 , 5% CO2 , and 85% N2 . The intestinal cecum was removed under sterile conditions and a small incision was made in cecal tissue. Each cecum was placed into 5 ml of deoxygenated modified VPI buffer (pH 5) that contained 0.1 g CaCl2 , 0.2 g MgSO4 , 0.5 g K2HPO4 , 0.5 g KH2PO4 , 5.0 g NaHCO3 , and 1.0 g NaCl per liter. Ceca were then vortexed to release contents and cecal tissue was removed. After the weight of cecal contents was determined, the cecal contents (each in 5 ml of the buffer) were combined. A homogeneous suspension of the contents of five ceca was used in each series of experiments. Replicate aliquots of this homogeneous suspension (0.5 ml for 14Clabeled samples and 0.1 ml for 73As-labeled samples) were pipetted into the reaction vials in an anaerobic chamber. Amber borosilicate vials (StepVial Snap-Crimp Vial, 12 1 32 mm, SRI, Inc., Eatontown, NJ) were used for 0.5-ml samples. For 0.1-ml samples, a 0.2-ml conic insert (StepVial, SRI) was placed into the each reaction vial to reduce the volume of atmosphere above the suspension. Reaction vials were capped (Snap cap GC 11 mm Silicone/PTFE, StepVial, SRI) before removal from chamber to preserve the anaerobic microenvironment. Only freshly prepared samples were used in the experiments. Vials containing the suspension of cecal contents were held on ice before addition of arsenicals and incubation under the conditions described below. Preliminary studies found that maintenance of cecal microflora under these culture conditions results in less than a doubling of the number of anaerobic organisms for incubation periods of up to 45 hr. Preparation of cecal tissue. Following removal of cecal contents, cecal tissue was washed with 5 ml of the sterile VPI buffer. The tissue from five

AID

TAP 8269

/

6h21$$$322

10-10-97 10:37:45

ceca were combined and homogenized in 8.3 ml of the buffer (0.24 g/ml) under either anaerobic or aerobic conditions. Aliquots of 0.1 ml of homogenate were pipetted into the reaction vials with conic inserts as described above. Incubation of cecal contents and cecal tissue homogenates with arsenicals. Capped reaction vials were placed into a N2-purged plastic glove bag. A radiolabeled arsenical (500 nCi of 73As or 10 nCi of 14C) was injected into each vial under the N2-atmosphere using a 10-ml Hamilton syringe (Hamilton, Reno, NV). Other additives (AdoMet, GSH, methylcobalamin (MetCob), L-cysteine (Cys), L-methionine (Met), sodium selenite) were obtained from Sigma and were injected as required. To minimize exposure of samples to O2 , the injected volume was limited to 1 ml of each compound (2 ml for GSH). Samples were incubated at 377C, at room temperature or on ice for up to 45 hr. Enumeration of cecal microbiota after incubation. To determine the effect of incubation and treatment on the microbial population, dilutions of the cecal contents prepared as described above were used as inocula for selective media. Dilutions (two-, four-, or six-fold) were made in VPI buffer. Total lactobacilli were enumerated on Rogosa SL agar (Difco, Detroit, MI) and total enterococci were enumerated on KF medium (Unipath Co., Oxoid Division, Ogdensburg, NY). Total anaerobes were determined on Brucella laked blood agar which contained 42.0 g of Brucella agar/liter (BBB Microbiology Systems, Cockeysville, MD), 5% (v/v) laked defibrinated sheep blood (Environmental Diagnostics, Burlington, NC), 1 mg of vitamin K1/ ml (Sigma), and 5 mg of hemin/ml (Sigma) (Holdeman et al., 1977; Summanen et al., 1993). Predominantly anaerobic Gram-negative rods were enumerated on Brucella laked blood agar supplemented with 7.5 mg of vancomycin/ml and 75 mg of kanamycin/ml (Summanen et al., 1993). The lactose fermenting and nonfermenting enteric bacilli were enumerated on MacConkey agar (Difco). All plates were incubated anaerobically for 48 hr except MacConkey plates that were incubated aerobically. Analysis of radiolabeled arsenic metabolites. After incubation, reaction vials were placed on ice. To liberate arsenicals bound to the insoluble fraction of the suspension of cecal contents and/or to proteins, samples were mixed (1:1) with 0.2 M CuCl in 2 M HCl (Styblo et al., 1996b) and transferred into plastic tubes. The capped plastic tubes were held at 1007C for 5 min. CuCl-treated samples were cooled on ice and ultrafiltered by centrifugation at 12,000g for 10 min using Microcon microconcentrators (Amicon, Beverly, MA) with 10-kDa molecular weight cutoff. Retentates on filters were washed with 100 ml of the CuCl reagent and centrifuged as above. The first and second ultrafiltrates were combined for further analysis. Radioactivities in retentates and ultrafiltrates were measured using a Minaxi g 5000 counter or a Tri-Carb 2200CA liquid scintillation analyzer (Packard, Downers Grove, IL). After CuCl treatment and ultrafiltration, 90 to 95% of the total radioactivity originally in each sample before processing was present in the combined ultrafiltrate. Ultrafiltrates were oxidized with 10% H2O2 to convert trivalent arsenicals to pentavalency (Styblo et al., 1995b). The oxidized ultrafiltrates that contained the 73As label were analyzed by TLC (Styblo et al., 1995b, 1996b) on Baker-flex PEI-F cellulose plates (J. T. Baker, Inc., Phillipsburg, NJ) that were developed with an acetone:acetic acid:water (2:1:1) mobile phase. Radiolabeled standards were prepared in oxidized ultrafiltrates from CuCl-treated cecal contents and were separated in parallel with samples. An AMBIS 4000 imaging detector (Scanalytics, Billerica, MA) was used to detect and quantify the distribution of radioactive species on TLC plates. Oxidized ultrafiltrates that contained 14 C-labeled arsenicals were analyzed by ion-exchange chromatography (Tam et al., 1978). Here, an aliquot of the oxidized ultrafiltrates from CuCltreated cecal contents (0.4 ml) was applied to a column of Dowex 50W resin with bed volume of 2 ml (0.8 1 4 cm) that was eluted by stepwise addition of 0.5 M HCl, distilled deionized water, 5% NH4OH, and 30% NH4OH. Notably, the amount of Cu present in analyzed samples did not exceed the capacity of the resin and did not affect the chromatographic separation of MAsV and DMAsV. Authentic 14C-labeled MAsV and DMAsV standards were prepared by addition of the labeled compounds to oxidized

toxc

AP: TAP

IN VITRO ARSENIC METHYLATION BY MOUSE CECAL MICROFLORA

103

ultrafiltrates from CuCl-treated cecal contents. In this sample matrix, MAsV was eluted with both HCl and water; DMAsV was eluted in 30% NH4OH. Chromatographic fractions were mixed with UltimaGold scintilation cocktail (Packard Instrument Co., Inc., Meriden, CT) and radioassayed with a TriCarb 2200CA liquid scintillation analyzer. To determine whether volatile 14C-labeled metabolites were generated by the demethylation of 14C-DMAs species by cecal microflora, the atmosphere of the reaction vials incubated with 14C-labeled DMAs for 45 hr was purged with Ç50 ml of air and the purge air was bubbled in series through a liquid scintillation vial containing an aqueous suspension of activated charcoal and a liquid scintillation vial containing Ultima Gold scintillation cocktail. The 14C contents of these traps were determined by liquid scintillation spectrometry, as described above. Statistical analysis. Student’s two-sided t test was used to evaluate differences in the production of arsenic metabolites by different experimental groups. To analyze changes in microbiotic populations under different experimental conditions, one-way or two-way ANOVA (SigmaStat, San Rafael, CA) was performed on logarithmic-transformed data.

RESULTS

Time Course of the Methylation of iAsIII and iAsV by Cecal Contents To examine the methylation of inorganic arsenicals by cecal microflora, cecal contents were incubated at 377C under anaerobic conditions with 73As-labeled iAsV or iAsIII (0.1 mM each, 10 pmol of iAs per sample) for up to 45 hr. TLC analysis of the oxidized ultrafiltrates from CuCl-treated cecal contents found that after a 6-hr incubation about 33% of iAsIII was methylated (Fig. 1a). Over the same time interval, only about 8% of iAsV was methylated (Fig. 1b). Further production of both methylated metabolites occurred between 6 and 21 hr of incubation. After 21 hr, there was a statistically significant difference (p õ 0.025) in the total amount of MAs produced from iAsIII or iAsV (36 vs 29% of the applied dose). In contrast, over the same time period, DMAs accounted for only about 3% of the applied dose. Notably, the amount of DMAs approximately doubled between 21 and 45 hr (Fig. 1). The yields of methylated metabolites produced from 0.1 mM iAsIII during a 21-hr incubation at room temperature (Ç257C) were similar to those found at 377C (Fig. 1a). Production of methylated metabolites from iAsV was lower at room temperature than at 377C (Fig. 1b). Based on the limits of detection of the radioassay, less than 10 fmol of methylated arsenicals was produced by cecal contents incubated with either 0.1 mM iAsIII or iAsV at 07C for 21 hr (data not shown). Effect of Substrate Concentration on Methylation of iAsIII and iAsV by Cecal Contents To examine the metabolic capacity of the cecal methylation system, cecal contents were incubated with 0.1 to 100 mM iAsIII or iAsV (10 to 10,000 pmol of iAs per sample) for 21 hr at 377C. Regardless of valence, the production of methylated metabolites increased in proportion to the con-

AID

TAP 8269

/

6h21$$$322

10-10-97 10:37:45

FIG. 1. Time course of the methylation of 0.1 mM iAsIII (a) and 0.1 mM iAsV (b) by cecal contents under anaerobic conditions. MAs (s) and DMAs (h) produced at 377C. MAs (l) and DMAs (j) produced at 257C (mean { standard deviation, n Å 4).

centration of substrate over the range of 0.1 to 10 mM (Fig. 2). MAs was always the predominant metabolite. It accounted for 36 to 46.4% of the amount of substrate in samples incubated with iAsIII and for 29.2 to 41.5% in samples incubated with iAsV (Fig. 2, insets). The percentage of substrate converted to DMAs increased with increasing concentration either of iAsIII (from 3.2 to 9.8%) or iAsV (from 2.9 to 10.1%). In samples incubated with 100 mM iAsIII, the relative yields of MAs and DMAs dropped to 17.3 and 0.6%, respectively. Lower relative yields of both methylated metabolites were also found in samples incubated with 100 mM iAsV. Here, MAs accounted for 16.0% and DMAs for 1.6% of the substrate (Fig. 2).

toxc

AP: TAP

104

HALL ET AL.

menting enteric bacilli which were composed mainly of Escherichia coli increased in all samples (iAsIII, iAsV, no arsenicals) after 21 hr of incubation. The size of most microbial populations was unaffected by incubation with either iAsIII or iAsV. However, the number of nonlactose fermenting enteric bacilli was reduced by the presence of either iAsIII or iAsV after 21 hr of incubation. Effects of Methyl Group Donors on Methylation of iAsIII and iAsV by Cecal Contents

FIG. 2. Effect of substrate concentration on methylation of iAsIII (a) and iAsV (b) by cecal contents. Production of MAs (s) and DMAs (h) at 377C over 21 hr under anaerobic conditions (mean { standard deviation, n Å 4).

Effect of Incubation with iAsIII and iAsV on Microbiota Populations of Cecal Contents The possible relation between the kinetics of iAs methylation and changes in the size and species composition of the cecal microbiota during incubation was examined. Here, cecal contents were incubated at 377C with 100 mM iAsIII or iAsV (nonradioactive arsenicals were used in these experiments) or without arsenicals for up to 21 hr. The microbial species in samples were enumerated after 3, 6, or 21 hr of incubation. Statistical analysis of log-transformed data is shown in Table 1. Although total anaerobes and facultatives (enterococci, lactobacilli, and obligately anaerobic Gramnegative rods) were unchanged, the number of lactose fer-

AID

TAP 8269

/

6h21$$$323

10-10-97 10:37:45

The effects of AdoMet, MetCob, and Met on the methylation of iAs by cecal microflora were examined (Table 2). Cecal contents were incubated at 377C with 0.1 mM iAsIII or iAsV (10 pmol per sample) with the addition of a single methyl group donor. Incubations were carried out for 6 hr, a time interval during which the size of populations of most bacterial species did not change significantly. In samples containing cecal contents without an added methyl group donor, the amount of MAs produced from iAsIII or iAsV was 2.53 or 0.75 pmol, respectively (Table 2). DMAs was not detected in these samples. Addition of 1 mM AdoMet did not significantly change methylation patterns for either iAs species. Addition of 1 mM Met significantly increased the production of MAs from either iAsIII (3.34 pmol) or iAsV (1.85 pmol). However, DMAs (0.11 pmol) was detected only in iAsVcontaining samples. Addition of 10 mM Met increased the production of both methylated metabolites from either inorganic arsenical. Addition of 100 mg of MetCob/ml (0.74 mM) did not affect the production of MAs from iAsIII but did increase significantly the production of MAs from iAsV. Addition of MetCob increased production of MAs from iAsIII and of both MAs and DMAs from iAsV. To determine whether chemical (nonenzymatic) methylation of iAs occurred, samples that contained VPI buffer, an arsenical, and a methyl group donor were prepared and incubated in parallel with samples that contained these ingredients and cecal contents. Based on the limits of detection of the radioassay, less than 10 fmol of methylated arsenicals was produced in any sample that contained a potential methyl group donor but did not contain cecal contents (data not shown). Effect of Thiols and Selenite on Methylation of iAsIII and iAsV by Cecal Contents To examine the effects of thiols, known cofactors for iAs methylation in tissues (Buchet and Lauwerys, 1988; Styblo et al., 1996a), on the methylation of iAsIII and iAsV by cecal microflora, cecal contents were incubated at 377C for 6 hr with either arsenical and with 10 mM GSH or 10 mM Cys (Table 2). Compared with controls, addition of 10 mM GSH significantly increased production of MAs from both iAsIII and iAsV. However, no DMAs was detected in these GSHsupplemented samples. Addition of 10 mM Cys had no effect on methylation patterns for either iAsIII or iAsV. No methyl-

toxc

AP: TAP

105

IN VITRO ARSENIC METHYLATION BY MOUSE CECAL MICROFLORA

TABLE 1 Effects of iAsIII or iAsV on Microbiota Populations of Intestinal Cecal Contents a Bacterial countsb in cecal content incubated with Bacterial population

Incubation time (hr)

Total anaerobes and facultatives

3 6 21 3 6 21 3 6 21 3 6 21 3 6 21 3 6 21

Obligately anaerobic Gram-negative rods Enterococci

Lactobacilli

Lactose-fermenting enteric bacilli Nonlactose-fermenting enteric bacilli

100 mM iAsIII

100 mM iAsV

10.3 { 10.1 { 10.2 { 9.6 { 9.5 { 9.7 { 8.8 { 8.6 { 8.8 { 8.9 { 8.6 { 8.7 { 6.0 { 6.1 { 7.9 { 6.0 { 6.0 { õ4.2c,d

10.1 { 10.2 { 10.3 { 9.5 { 9.5 { 9.7 { 8.5 { 8.8 { 8.8 { 8.6 { 8.9 { 8.7 { 5.7 { 6.4 { 8.0 { 5.7 { 6.3 { õ4.2c,d

0.02 0.16 0.06 0.04 0.07 0.01c 0.06 0.14 0.06 0.03 0.12 0.03 0.03 0.05c 0.05c 0.11 0.05

0.10 0.04 0.03 0.05 0.03 0.02c 0.07 0.07 0.10 0.07 0.05 0.07 0.05 0.05c 0.01c 0.05 0.08

No iAs 10.1 10.1 10.5 9.5 9.7 9.5 8.6 8.8 8.9 8.7 8.9 8.6 5.7 6.1 8.0 5.7 6.0 4.9

{ { { { { { { { { { { { { { { { { {

0.14 0.07 0.05 0.05 0.01 0.09c 0.01 0.01 0.03 0.06 0.07 0.05 0.02 0.20c 0.03c 0.02 0.06 0.24c

Cecal contents were incubated at 377C for up to 21 hr with 100 mM iAsIII or iAsV. Each value represents the logarithm of the average colony forming units/g of cecal contents from three replicates { SE. c Significant effect (p õ 0.05) of the time of incubation on microbial count. d Significant effect (p õ 0.05) of the presence of iAsIII or iAsV on microbial count by two-way ANOVA. a b

ated species were found in samples that contained VPI buffer and corresponding combinations of arsenicals and thiols but did not contain cecal contents (data not shown).

The effect of selenite, a potent inhibitor of iAs methylation in rat liver cytosol (Buchet and Lauwerys, 1985; Styblo et al., 1996a), on the methylation of iAsIII and iAsV by cecal

TABLE 2 Effect of Methyl Group Donors, Thiols, and Sodium Selenite on Methylation of iAsIII and iAsV in Intestinal Cecal Contentsa Substrate 0.1 mM iAsIII

0.1 mM iAsV

Additives

MAs (pmol)

DMAs (pmol)

MAs (pmol)

DMAs (pmol)

Control (no additives) 1 mM AdoMet Met 1 mM 10 mM MetCob 100 mg/ml 1000 mg/ml 10 mM GSH 10 mM Cys 50 mM Selenite

2.53 { 0.58b 2.78 { 0.58

NDc ND

0.75 { 0.39 1.03 { 0.50

ND 0.06 { 0.09

3.34 { 0.39d 3.11 { 0.14d

ND 0.11 { 0.13

1.85 { 0.22d 2.76 { 0.34d

0.11 { 0.07d 0.32 { 0.15d

{ { { { {

ND 0.12 { 0.24 ND ND ND

1.47 { 0.35d 1.69 { 0.14d 1.82 { 0.2d 0.81 { 0.38 0.05 { 0.09d

0.09 { 0.05d 0.18 { 0.08d ND ND ND

2.39 4.09 3.01 2.55 1.17

0.67 0.46d 0.24d 0.36 0.32d

Cecal contents were incubated at 377C for 6 hours with 0.1 mM iAsIII or iAsV. Mean { SD for four replicates is shown. c ND, not detected. d Significantly different (p õ 0.05) from corresponding control value by Student’s two-sided t test. a b

AID

TAP 8269

/

6h21$$$323

10-10-97 10:37:45

toxc

AP: TAP

106

HALL ET AL.

TABLE 3 Effects of Additives on Microbiota Populations in Intestinal Cecal Contents Incubated with iAsIII or iAsV a Bacterial countsb in cecal contents incubated with 0.1 mM iAsIII and

10 mM Met

No additives Total anaerobes and facultatives Obligately anaerobic Gram-negative rods Enterococci Lactobacilli Lactose-fermenting enteric bacilli Nonlactose-fermenting enteric bacilli

10.1 9.9 8.9 8.9 7.1 6.6

{ { { { { {

0.05 0.03 0.04 0.03 0.18 0.27c

10.1 9.9 8.7 8.8 7.1 6.1

{ { { { { {

0.07c 0.03 0.06d 0.04 0.16 0.06

10 mM GSH 10.3 9.9 8.9 8.9 6.0 5.9

{ { { { { {

0.04 0.04 0.04 0.02 0.13d 0.01d

1000 mg MetCob/ml 10.1 9.9 8.9 8.4 7.3 6.4

{ { { { { {

0.04 0.02 0.03 0.51 0.09 0.05

Bacterial countsb in cecal contents incubated with 0.1 mM iAsV and

No additives Total anaerobes and facultatives Obligately anaerobic Gram-negative rods Enterococci Lactobacilli Lactose-fementing enteric bacilli Nonlactose-fermenting enteric bacilli

10.1 9.8 8.8 8.8 7.2 5.7

{ { { { { {

0.06 0.03 0.01 0.04 0.18 0.16c

10 mM Met 10.1 9.9 8.7 8.7 7.2 6.1

{ { { { { {

0.08 0.05 0.04 0.07 0.09 0.22

10 mM GSH 10.0 9.8 8.8 8.9 5.9 5.9

{ { { { { {

0.03 0.04 0.02 0.05 0.25d 0.09

1000 mg MetCob/ml 10.1 9.8 8.9 8.9 7.5 5.6

{ { { { { {

0.04c 0.07 0.05 0.03 0.09 0.15

Cecal contents were incubated at 377C for 6 hr with arsenicals and with or without additives. Each value represents the logarithm of the average colony forming units/g cecal content from four replicates (unless indicated otherwise) { SE. c Value is an average { SE for three replicates. d Significantly different (p õ 0.05) from control (no additives) sample by one-way ANOVA. a b

contents was also examined (Table 2). Addition of 50 mM sodium selenite reduced the production of MAs from iAsIII to less than 50% of that found in control samples and almost fully abolished production of MAs from iAsV. No DMAs was detected in any selenite-treated sample. Effect of Additives on Cecal Microbiota Populations Because additives might exert their effects on the pattern and extent of iAs methylation by an effect on the size or composition of the cecal microflora, the effects of Met, MetCob, and GSH on bacterial populations were examined. For these experiments, bacterial species in cecal contents were enumerated after incubation at 377C for 6 hr in the presence of 0.1 mM nonradioactive iAsIII or iAsV and with or without Met, MetCob, or GSH. As shown in Table 3, most additives did not affect the size of cecal bacterial populations. However, the numbers of both lactose and nonlactose fermenting enteric bacilli were significantly reduced by incubation with iAsIII and 10 mM GSH. A small but significant decrease in the number of enterococci was found in cecal contents incubated with iAsIII and 10 mM Met. The number of lactose fermenting enteric bacilli was also significantly reduced by incubation with iAsV and 10 mM GSH.

AID

TAP 8269

/

6h21$$$323

10-10-97 10:37:45

Methylation of MAsIII and MAsV by Cecal Contents The ability of cecal microflora to produce DMAs from MAs was investigated using 14C-labeled MAsIII and MAsV. Cecal contents were incubated at 377C for 45 hr with 1.5 mM MAsIII or MAsV (750 pmol per sample). Aliquots of the oxidized ultrafiltrates from CuCl-treated samples were analyzed by ion-exchange chromatography on Dowex 50W. After 45 hr of incubation, 9.3% of MAsIII and 4.5% of MAsV was methylated, yielding 69 { 20.1 pmol (mean { SD, n Å 4) and 33 { 11.3 pmol of DMAs, respectively (data not shown). Because demethylation of DMAs by cecal microflora could account for the low yields of this metabolite in cecal contents incubated with inorganic arsenicals and with MAsIII or MAsV, the extent of demethylation of exogenously added DMAs was examined. Cecal contents were incubated with 14 C-labeled DMAsIII or DMAsV (1.5 mM) at 377C for 45 hr. After incubation, the atmosphere above the capped samples was purged with air that was then passed through traps that contained activated charcoal and liquid scintillation cocktail to collect liberated 14C (see Methods). The cecal contents were also processed for speciation of arsenicals by Dowex 50W chromatography. No radioactivity was found in charcoal traps; radioactivity detected in Ultima Gold traps did

toxc

AP: TAP

IN VITRO ARSENIC METHYLATION BY MOUSE CECAL MICROFLORA

not exceed 0.2% of the total radioactivity in samples. DMAs was the only radiolabeled arsenical detected in CuCl-treated ultrafiltrates analyzed by Dowex chromatography (data not shown). Methylation of iAsIII and iAsV by Cecal Tissue To determine whether cecal tissue can methylate arsenic, homogenates of cecal tissue were incubated at 377C with 0.1 mM iAsIII or iAsV (10 pmol per sample) for 21 hr under either anaerobic or aerobic conditions. For either condition, average yields of MAs and DMAs from iAsIII and from iAsV did not exceed 2% of the substrate (i.e., 0.2 pmol). Addition of 1 mM AdoMet and 10 mM GSH, cofactors for the methylation of iAs in rat liver cytosol (Buchet and Lauwerys, 1985; Styblo et al., 1996a), did not change methylation patterns for either arsenical (data not shown). DISCUSSION

Although the predominant mechanism of detoxification of iAs in bacteria is believed to include reduction of iAsV to iAsIII and extrusion of iAsIII from the bacterial cell (Rosen et al., 1991; Wu and Rosen, 1993), methylation of iAs by microorganisms has also been described. Challenger (1945, 1951) demonstrated that several species of fungus can methylate iAs to volatile trimethylarsine. Other fungi (Candida humicola, Gliocladium roseum, Penicillium species) present in waste water can methylate iAs, MAs, and DMAs to trimethylated species (Cox and Alexander, 1973). A methanobacterium that methylates iAsV to MAs and DMAs under anaerobic conditions has been described (McBride and Wolfe, 1971). In C. humicola, methylarsine oxide is methylated to DMAs and methylation of methylarsine sulfide yields trimethylarsine (Cullen et al., 1989). The GSH derivative of DMAsV can also be methylated by C. humicola to trimethylarsine. In addition to evidence of biomethylation of As by microorganisms in the environment, it is clear that enteric bacteria can also methylate As. Rowland and Davies (1981) investigated the metabolism of iAs by the GI microflora of the rat. These investigators reported that the contents of the small intestine and cecum of rat reduced iAsV to iAsIII. Methylated metabolites were formed in cecal contents incubated at 377C under a CO2 atmosphere with 12 mM iAsV. After a 120-hr incubation period, MAs was the predominant metabolite, accounting for about 25% of the total arsenic. DMAs did not appear in the reaction mixture before 25 hr of incubation. Because the medium used in that experiment contained yeast extract and D-glucose and would likely support significant growth of the bacterial population, it is difficult to estimate the contribution of alterations in the size or composition of the microbial population to the observed changes in the rate or pattern of As methylation. Furthermore, the adequacy

AID

TAP 8269

/

6h21$$$323

10-10-97 10:37:45

107

of methods used in that study to liberate protein-bound or particulate-associated arsenicals for analysis is not known. The incubation medium used in the present work was designed to maintain the viability of microbes in the cecal contents and to limit the availability of nutrients that would support the expansion of the microbial population. Notably, the total number of bacteria in the incubation system did not significantly change during the incubation at 377C under anaerobic conditions with or without the addition of arsenicals. Increased numbers of enteric bacilli present after longer incubation periods were likely due to the presence of compounds in cecal contents that supported their growth. Because these subpopulations constituted a small fraction of the total cecal anaerobes, significant changes in number of anaerobic bacteria were not observed. Under these conditions, about 40% of 0.1 mM iAsIII or iAsV was methylated over 21 hr, yielding almost exclusively MAs. Over the first 6 hr of incubation, the rate of MAs synthesis from iAsIII was about fivefold greater than the rate of MAs synthesis from iAsV (Fig. 1). Indeed, there was an apparent lag period of about 3 hr in methylation of iAsV. Although our present method for the liberation of protein-bound arsenicals precludes determination of their valency, the delay in appearance of MAs in iAsV-containing reaction mixtures was likely due to the slow rate of its reduction to iAsIII, the favored substrate for biomethylation reactions (Cullen et al., 1984; Styblo et al., 1995b). Rowland and Davies (1981) showed that production of methylated metabolites from iAsV increased markedly when cecal contents were incubated under the atmosphere of H2S that reduces iAsV to iAsIII. In the present work, yields of methylated metabolites produced from iAsV, but not from iAsIII, decreased when the incubation of cecal contents was carried out at room temperature. This suggests that reduction of iAsV may be partly temperature dependent; that is, it is catalyzed by an enzyme that has a higher temperature optimum than the bacterial arsenic methyltransferase(s). To estimate the capacity of the mouse cecal methylation system, the iAs methylation was examined over a wide range of substrate concentrations. The results showed a proportional increase in production of MAs and DMAs over the range of 0.1 to 10 mM iAsIII or iAsV. Furthermore, there was evidence of saturation of metabolism when the concentration of either substrate reached 100 mM (Fig. 2, insets). Based on a production rate of about 80 pmol of MAs and DMAs per hour by cecal contents incubated with 100 mM iAs and the presence of about 5 mg of cecal contents in the standard assay system, a rate of about 16 pmol of methylated arsenical per hour per milligram of cecal contents can be calculated. For an entire mouse cecum with estimated contents of 250 mg, about 96 nmol of methylated arsenicals could be produced per 24 hr. Thus, under chronic exposure conditions, the production of methylated arsenicals by cecal microflora

toxc

AP: TAP

108

HALL ET AL.

could be important, affecting both the amount and the species of arsenic to which the host organism is exposed. Notably, the contribution of GI microflora to the metabolism of iAs has not been considered in recent efforts to model the pharmacokinetics behavior of iAs (Mann et al., 1996). Among the agents tested in this study, addition of Met, MetCob, or GSH to cecal contents increased the conversion of iAsIII and iAsV to methylated species. Notably, these agents increased the production of MAs but had little effect on DMAs production. Addition of either Met or MetCob could stimulate the production of AdoMet by cecal microflora. Cullen and associates (1977) have reported that these aerobic species, Scopulariopsis brevicaulis, C. humicola, and G. roseum, utilize the methyl group of deuterated Met in the synthesis of trimethylated arsenicals. In addition, MetCob could serve as a methyl group donor for either abiotic or enzymatically catalyzed methylation reactions (Gilmour and Henry, 1991). Because GSH reduces iAsV to iAsIII (Scott et al., 1993, Delnomdedieu et al., 1994a), the addition of GSH into the cecal contents had a larger effect on the methylation of iAsV than on the methylation of iAsIII (Table 2). It has been previously reported that the reducing capacity of GSH added to the incubation medium can increase the reduction of iAsV inside the cells (Delnomdedieu et al., 1994b). This transduction of the reducing potential of GSH across cell membranes by a thiol–disulfide interchange mechanism has been described by Ciriolo and associates (1993). GSH has also been shown to support methylation of iAsIII in rat liver cytosol (Buchet and Lauwerys, 1985; Styblo et al., 1996a), probably by reducing of the substrate and by maintaining of critical residues in arsenic methyltransferase(s) in reduced form. Notably, selenite, an inhibitor of iAs methylation in rat liver cytosol (Buchet and Lauwerys, 1985; Styblo et al., 1996a), inhibited methylation of iAsIII and almost fully abolished methylation of iAsV by cecal contents (Table 2). Selenite was previously shown to inhibit the in vitro reduction of iAsV to iAsIII by rat intestinal microflora (Rowland and Davies, 1981), suggesting that selenium and iAs may compete for the reducing capacity of GSH. Although selenite is not a substrate for As methyltransferase purified from rabbit liver (Zakharyan et al., 1995), it may compete with iAs as a substrate for an uncharacterized bacterial methyltransferase or there may be competition between the methyltransferases that catalyze the methylation of these metalloids. The effects of chronic changes in selenium intake on the capacity to methylate iAs in the GI tract have not been examined. The predominance of MAs as the metabolite of iAs in cecal contents suggests there is a relative deficiency of the enzyme(s) responsible for the dimethylation of iAs or that the experimental conditions used in the present study did not favor DMAs synthesis. It has been previously shown in an in vitro methylation assay that contains rat liver cytosol

AID

TAP 8269

/

6h21$$$324

10-10-97 10:37:45

that the dimethylation does not occur if MAs is present in pentavalent form (Styblo et al., 1995b); in this system, addition of 10 mM GSH was not sufficient to reduce MAsV to MAsIII. In contrast, MAsIII was fully methylated to DMAs in this in vitro system. In the present study, neither MAsIII nor MAsV was extensively converted to DMAs by cecal contents. Direct measurement of the rate of demethylation of DMAsIII and DMAsV by cecal contents suggested that cecal microflora were not effective demethylators of these compounds. Notably, Cullen and associates (1989) found both iAs and DMAs in mouse cecal contents incubated anaerobically with 73As-labeled methylarsine oxide. Quinn and McMullan (1995) have partially characterized a Gram-negative bacteria (strain ASVZ) that cleaves the C–As bond. In conclusion, the present work indicates that the microflora of the mouse cecum has a large capacity for the enzymatically catalyzed synthesis of MAs from either iAsIII or iAsV. The magnitude of MAs production by GI microflora in intact animals should be determined to estimate contribution of this tissue to the overall metabolism and the kinetic behavior of iAs in the mouse and other species. ACKNOWLEDGMENTS M.S. was a postdoctoral fellow supported by Training Grant T901915 of the U.S. Environmental Protection Agency/University of North Carolina Toxicology Research Program with the Curriculum in Toxicology, University of North Carolina at Chapel Hill. We thank Ms. Karen Herbin-Davis for excellent technical assistance. A preliminary account of this research was presented at the 35th annual Society of Toxicology meeting, March 10–14, 1996. This manuscript has been reviewed in accordance with the policy of the National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

REFERENCES Buchet, J. P., and Lauwerys, R. (1985). Study of inorganic arsenic methylation by rat in vitro: Relevance for the interpretations of observations in man. Arch. Toxicol. 57, 125–129. 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, 3149–3153. Challenger, F. (1945). Biological methylation. Chem. Rev. 36, 515–361. Challenger, F. (1951). Biological methylation. Adv. Enzymol. 12, 429–491. Ciriolo, M. R., Paci, M., Sette, M., DeMartino, A., Bozzi, A., and Rotilio, G. (1993). Transduction of reducing power across the plasma membrane by reduced glutathione: A 1H-NMR spin-echo study of intact human erythrocytes. Eur. J. Biochem. 215, 711–718. Cox, D. P., and Alexander, M. (1973). Production of trimethylarsine gas from various arsenic compounds by three sewage fungi. Bull. Environ. Contam. Toxicol. 9, 84–88. Cullen, W. R., Froese, C. L., Lui, A., McBride, B. C., Patmore, D. J., and Reimer, M. (1977). The aerobic methylation of arsenic by microorganisms in the presence of L-methionine-methyl-d3 . J. Organomet. Chem. 139, 61–69.

toxc

AP: TAP

IN VITRO ARSENIC METHYLATION BY MOUSE CECAL MICROFLORA Cullen, W. R., McBride, B. C., and Reglinski, J. (1984). The reaction of methylarsenicals with thiols: Some biological implications. J. Inorg. Biochem. 21, 179–194. Cullen, W. R., McBride, B. C., Manji, H., Pickett, A. W., and Reglinski, J. (1989). The metabolism of methylarsine oxide and sulfide. Appl. Organomet. Chem. 3, 71–78. Delnomdedieu, M., Basti, M. M., Otvos, J. D., and Thomas, D. J. (1994a). Reduction and binding of arsenate and dimethylarsenate by glutathione. Chem.-Biol. Interact, 90, 139–155. Delnomdedieu, M., Basti, M. M., Styblo, M., Otvos, J. D., and Thomas, D. J. (1994b). Complexation of arsenic species in rabbit erythrocytes. Chem. Res. Toxicol. 7, 621–627. Gadd, G. M. (1993). Microbial formation and transformation of organometallic and organometalloid compounds. FEMS Microbiol. Rev. 11, 297–316. Gilmour, C. C., and Henry, E. A. (1991). Mercury methylation in aquatic systems affected by acid deposition. Environ. Pollut. 71, 131–169. Guo, H.-R., Chen, C.-J., and Greene, H. L. (1994). Arsenic in drinking water and cancers: A brief descriptive review of Taiwan studies. In Arsenic—Exposure and Health. (W. R. Chappell, C. O. Abernathy, and C. R. Cothern, Eds.), pp. 129–138, Science and Technology Letters, Northwood, UK. Holdeman, L. V., Cato, E. P., and Moore, W. E. C. (1977). Anaerobe Laboratory Manual, 4th ed. Virginia Polytechnical Institute and State University, Blacksburg, VA. Mann, S., Droz, P. O., and Vahter, M. (1996). A physiologically based pharmacokinetic model for arsenic exposure. Toxicol. Appl. Pharmacol. 137, 8–22. McBride, B. C., and Wolfe, R. F. (1971). Biosynthesis of dimethylarsine by methanobacterium. Biochemistry 10, 4312–4317. Quinn, J. P., and McMullan, G. (1995). Carbon-arsenic bond cleavage by a newly isolated Gram-negative bacterium, strain ASVZ. Microbiol. 141, 721–727. Reay, P. F., and Asher, C. J. (1977). Preparation and purification of 74Aslabeled arsenate and arsenite for use in biological experiments. Anal. Biochem. 78, 557–560. Rosen, B. P., Weigel, U., Monticello, R., and Edwards, B. P. F. (1991). Molecular analysis of an anion pump. Arch. Biochem. Biophys. 284, 381– 385.

AID

TAP 8269

/

6h21$$$325

10-10-97 10:37:45

109

Rowland, I. R., and Davies, M. J. (1981). In vitro metabolism of inorganic arsenic by the gastrointestinal microflora of the rat. J. Appl. Toxicol. 1, 278–283. Scott, N., Hatlelid, K. M., MacKenzie, N. E., and Carter, D. E. (1993). Reaction of arsenic(III) and arsenic(V) species with glutathione. Chem. Res. Toxicol. 6, 102–106. Styblo, M., Delnomdedieu, M., and Thomas, D. J. (1995a). Biological mechanism and toxicological consequences of the methylation of arsenic. In Toxicology of Metals—Biochemical Aspects, Handbook of Experimental Pharmacology (M. G. Cherian, and R. A. Goyer, Eds.), Vol. 115, pp. 407–433. Springer Verlag, Berlin. Styblo, M., Yamauchi, H., and Thomas, D. J. (1995b). Comparative in vitro methylation of trivalent and pentavalent arsenicals. Toxicol. Appl. Pharmacol. 135, 172–178. Styblo, M., Delnomdedieu, M., and Thomas, D. J. (1996a). Mono- and dimethylation of arsenic in rat liver cytosol in vitro. Chem.-Biol. Interact. 99, 147–161. Styblo, M., Hughes, M. F., and Thomas, D. J. (1996b). Liberation and analysis of protein-bound arsenicals. J. Chromat. B 677, 161–166. Summanen, P., Baron, E. J., Citron, D. M., Strong, C., Wexler, H. M., and Finegold, S. M. (1993). Wadsworth Anaerobic Bacteriology Manual, 5th ed. Star, Belmont, CA. Tam, G. K. H., Charbonneau, S. M., Bryce, F., and Lacroix, G. (1978). Separation of arsenic metabolites in dog plasma and urine following intravenous injection of 74As. Anal. Biochem. 86, 505–511. Thornton, I., and Farago, M. (1996). The geochemistry of arsenic. In Arsenic Exposure and Health Effects (C. O. Abernathy, R. L. Calderon, and W. R. Chappell, Eds.), Vol. 2, pp. 1–16. Chapman & Hall, London. Wu, J., and Rosen, B. P. (1993). Metaloregulated expression of the ars operon. J. Biol. Chem. 268, 52–58. Yamauchi, H., and Fowler, B. A. (1994). Toxicity and metabolism of inorganic and methylated arsenicals. In Arsenic in the Environment. II. Human Health and Ecosystem Effects (J. O. Nriagu, Ed.), pp. 35–43. Wiley, New York. Zakharyan, R., Wu, Y., Bogdan, G. M., and Aposhian, H. V. (1995). Enzymatic methylation of arsenic compounds: Assay, partial purification, and properties of arsenite methyltransferase and monomethylarsonic acid methyltransferase from rabbit liver. Chem. Res. Toxicol. 8, 1029–1038.

toxc

AP: TAP