Methylation of Halogenated Phenols and Thiophenols by Cell Extracts of Gram-Positive and Gram-Negative Bacteria ALASDAIR H

Home , Dimethyl sulfate

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1988, p. 524-530 Vol. 54, No. 2 0099-2240/88/020524-07$02.00/0 Copyright C) 1988, American Society for Microbiology Methylation of Halogenated Phenols and Thiophenols by Cell Extracts of Gram-Positive and Gram-Negative Bacteria ALASDAIR H. NEILSON,* CARIN LINDGREN, PER-AKE HYNNING, AND MIKAEL REMBERGER Swedish Environmental Research Institute, Box 21060, S-100 31 Stockholm, Sweden Received 29 September 1987/Accepted 20 November 1987

0-methylation of 2,6-dibromophenol was studied in cell extracts prepared from Rhodococcus sp. strain 1395. 0-methylation activity was enhanced by the addition of S-adenosyl-L-methionine but was not affected by the addition of 5-methyltetrahydrofolate nor by up to 10 mM MgCl2 or EDTA. By using 2,6-dibromophenol, 4,5,6-trichloroguaiacol, and pentachlorothiophenol as the substrates, 0-methylation activity was also demonstrated in extracts from two other Rhodococcus sp. strains, an Acinetobacter sp. strain, and a Pseudomonas sp. strain. A diverse range of chloro- and bromophenols, chlorothiophenols, chloro- and bromoguaiacols, and chloro- and bromocatechols were assayed as the substrates by using extracts prepared from strain 1395; all of the compounds were methylated to the corresponding anisoles, veratroles, or guaiacols, which have been identified previously from experiments using whole cells. The specific activity of the enzyme towards the thiophenols was significantly higher than it was towards all the other substrates-high activity was found with pentafluorothiophenol, although the activity with pentafluorophenol was undetectable with the incubation times used. For the chlorophenols, the position of the substituents was of cardinal importance. The enzyme had higher activity towards the halogenated catechols than towards the corresponding guaiacols, and selective 0-methylation of the 3,4,5-trihalogenocatechols yielded predominantly the 3,4,5-trihalogenoguaiacols. As in experiments with whole cells, neither 2,4-dinitrophenol, hexachlorophene, nor 5-chloro- or 5-bromovanillin was 0-methylated. The results showed conclusively that the methylation reactions were enzymatic and confirmed the conclusion from extensive studies using whole cells that methylation of halogenated phenols may be a significant alternative to biodegradation.

We have previously demonstrated the 0-methylation of MATERIALS AND METHODS chloroguaiacols by dense cell suspensions of a number of gram-positive and gram-negative bacteria as well as by Substrates. The fluoro-, chloro-, and bromophenols and cultures of gram-positive strains growing at the expense of the thiophenols were of the highest commercially available various carbon sources (1). In addition, we have shown that purity; each had a gas chromatographic (GC) purity of a structurally diverse range of chloro- and bromophenols are >98%, and all of them were free of anisoles. The thioan- 0-methylated to the corresponding anisoles. These results isoles were prepared from the thiophenols by methylation led us to propose that the 0-methylation reaction may be an with dimethyl sulfate in KOH solution-pentachlorothioan- environmentally important alternative to biodegradation (2). isole was purified by recrystallization from methanol, and A number of issues remain unresolved; however, these 2,6-dichloro- and pentafluorothioanisoles (which were liq- issues cannot readily be addressed by experiments with uids) were purified by chromatography on silica gel (Kie- whole cells, in which permeability barriers could account for selgel 60, 70/280 mesh; Merck). 4,5,6-Tribromoguaiacol was the failure to observe 0-methylation of some compounds. prepared by brominating guaiacol with excess bromine in We know of only one previous study (19) that has demon- carbon disulfide for 24 h at room temperature. The solvent strated bacterial 0-methylation of substituted phenols in cell was removed, and the product was crystallized from hexane extracts of a bacterial strain, in that case one classified as a and then from toluene. By analogy to the chlorination of Mycobacterium sp., although recently Drotar et al. (7) guaiacol, which produced 4,5,6-trichloroguaiacol exclu- examined the corresponding reaction in a number of aro- sively (13), we assumed that this product was the 4,5,6- matic thiols, including the heterocyclic compounds benzthia- tribromo- compound. The mass spectrum of the 0-acetate zole-2-thiol and purine-6-thiol. had a parent ion (m/z, 300) corresponding to C9H7Br3O3. Therefore, we have examined the 0-methylation activity 3,4,5-Tribromocatechol was prepared by brominating cate- of cell extracts from several strains of bacteria in which we chol with 3.5 mol of bromine in acetic acid at room temper- found activity in whole cells. The results showed without ature. The solvent was removed, and the residue was ambiguity the following facts. (i) The 0-methylation reaction crystallized repeatedly from hexane. The resulting product was enzymatic. (ii) The natural methyl donor was probably still contained small amounts of the di- and tetrabromocate- S-adenosyl-L-methionine. (iii) Extracts prepared from gram- chols but was purified by high-performance liquid chroma- positive strains generally displayed considerably higher ac- tography by using a semipreparative Nucleosil (Macherey- tivity than those prepared from gram-negative strains. (iv) Nagel, Duren, Federal Republic of Germany) C18 column The structure of the phenolic substrates was of cardinal (particle size, 5 ,um) with a mobile phase of acetonitrile- importance in determining the rates of 0-methylation. phosphate buffer (50 mM, pH 3.0) (60:40, vol/vol) with a flow rate of 1.0 ml min-1. The detector system was set at a wavelength of 284 nm. The di-O-acetate had a mass spec- * Corresponding author. trum with a parent ion (m/z, 428) corresponding to 524 VOL. 54, 1988 CELL-FREE METHYLATION OF HALOGENATED PHENOLS 525

ClOH7Br3O4. Careful partial methylation of the tribromo- ,u, 2 mg* ml-'), 30 to 60 nmol of the substrate (20 ,ul of a catechol in dichloromethane with dimethyl sulfate (added in solution in water or acetone, 500 ,ug ml-'), and the appro- portions of ca. 0.3 mol/mol of catechol) and tetramethylam- priate buffer to a total volume of 1.0 ml. It was shown that, monium hydroxide at room temperature yielded a product at this concentration, acetone had no effect on the enzyme containing predominantly the desired 3,4,5-tribromo- activity. In experiments in which the thiophenols were the guaiacol. This product was purified by semi-preparative substrates, it was essential to use freshly prepared solutions high-performance liquid chromatography as described above in acetone to avoid serious interference from chemically by using a mobile phase of methanol-phosphate buffer produced oxidation products in the analysis of the thioani- (80:20, vol/vol); the product was finally purified by dissolv- soles. 5-Methyltetrahydrofolate (Sigma) (final concentration, ing the 3,4,5-tribromoguaiacol in hexane and removing the 40 ,uM), EDTA (final concentrations, 1 and 10 mM), and small amount of insoluble tetramethylammonium carbonate magnesium chloride (1 and 10 mM) were added as necessary. by filtration. The product had a GC purity of >96%, and the The assay mixture was incubated in darkness with shaking at mass spectrum of the 0-acetate was identical to that of the 22°C, and the reaction was stopped after 60 min by addition 4,5,6-tribromoguaiacol prepared as described above. The of 1 drop of 6 M HCI. The effect of pH was examined in GC retention time of the 3,4,5-tribromoguaiacol, however, phosphate buffer at pH 6.0 and 7.0, in Tris buffer at pH 7.0, was shorter than that of the 4,5,6-isomer; this retention time 8.0, and 9.0, and in glycine buffer at pH 9.0 and 10.0; the is consistent with the corresponding values for the trichlo- buffer concentration was 50 mM. Appropriate control tubes roguaiacols. Therefore, we are convinced that the structural were incubated without the cell extract or without S-ade- assignment of the tribromoguaiacols is correct. 3,4,5-Tribro- nosyl-L-methionine, or the extracts were boiled for 10 min. moveratrole was prepared by methylating 4,5,6-tribromo- All measurements were carried out in duplicate, which guaiacol with dimethyl sulfate in KOH solution and was differed by <10%; specific activities are given as picomoles recrystallized from methanol. The mass spectrum had a of product formed per minute per milligram of protein. parent ion (m/z, 372) corresponding to C8H7Br3O2. Analysis of the anisoles and veratroles. Anisoles and vera- The relative GC retention times of compounds not hitherto troles were analyzed as described previously (2). The assay described by us are given in Table 1. The instrumentation mixture was acidified with a few drops of 6 M HCI, extracted used was that described previously (16), except that the with 1.5 ml of a mixture of t-butyl methyl ether-hexane (1:2, following temperature program was used: isothermal for 1 vol/vol) containing the appropriate internal standard (pen- min at 150°C, increasing by 2°C. min-' to 235°C, which was tachlorobenzene or hexachlorobenzene), and then further then maintained for 5 min. extracted with the same solvent mixture but lacking the Bacterial strains. All of the bacterial strains used, together internal standard. The combined organic phase was washed with their biochemical characteristics, have been described with 2 ml of 1 M KOH to remove excess phenol and assayed previously (1). We examined three gram-positive strains by using anisole and veratrole standards. tentatively assigned to the genus Rhodococcus (strains 1395, Analysis of the halogenated guaiacols. Analysis of the 1623, and 1632) and two gram-negative strains belonging to halogenated guaiacols necessitated their conversion into the the genera Pseudomonas (strain 1631) and Acinetobacter O-acetates, and it was found that acetylation was also (strain 1678). advantageous for analysis of the thioanisoles; interference Growth of cells. Cells were grown aerobically at 23°C as from chemically produced oxidation products of the unre- described previously (15) in VV2 medium supplemented acted thiophenols was thereby reduced to a minimum. with 4-hydroxybenzoate (strains 1395, 1623, and 1678), Therefore, the following minor modifications of the proce- 3-hydroxybenzoate (strains 1631 and 1632), or succinate dure described above were introduced. (i) Solid ascorbic (strain 1395) or in nutrient broth (Oxoid CM1) (strain 1395). acid (ca. 20 mg) was added to the acidified sample. (ii) The Cells were harvested after growth for 40 (Nutrient broth), 60 combined extracts were dried (Na2SO4) before acetylation, (other substrates), or 120 h (strain 1632) by centrifugation and after acetylation, excess acetic anhydride was removed (2,000 x g, 10 min, 4C), washed three times with 50 mM by shaking the extracts for 5 min with 0.8 M K2CO3 (4 ml). phosphate buffer (pH 7.0), and then centrifuged at 15,000 x Identification of the products. Most of the metabolites, g (20 min). The cell pellets were frozen at -20°C. which were neutral compounds, have already been charac- Preparation of cell extracts. The frozen cells were thawed terized in previous studies with whole cells. Therefore, in slowly at room temperature, suspended in 50 mM phosphate these cases the products were identified solely on the basis buffer (pH 7.0) (2 ml/g [wet weight] of cells), and disrupted of comparison of the relative GC retention times with those by sonication (Branson model B12) with cooling in an ice of known standards. The thioanisoles and the halogenated bath for 30-s periods (10 times for gram-positive cells, which guaiacols formed from the catechols were identified addi- were more resistant to rupture, and 7 times for the more tionally on the basis of electron impact mass spectrometric readily broken gram-negative cells). Cell debris was removed by centrifugation (15,000 x g, 30 min, 4°C), the supernatant was centrifuged (100,000 x g, 2 h), and the clear TABLE 1. GC retention times relative to that of brownish-yellow supernatant was dialyzed overnight at 4°C tetrachloroguaiacol 0-acetate against the phosphate buffer. Extracts were kept at 4°C and Compound Relative showed no appreciable loss of activity after storage for 1 retention time month. Pentafluorothioanisole ...... 0.15 Assay of enzyme activity. Enzyme activity was assayed on 2,6-Dichlorothioanisole ...... 0.52 the basis of the rate of formation of the appropriate anisoles, Pentachlorothioanisole ...... 1.22 veratroles, and guaiacols, which were quantified by GC 3,4,5-Tribromoguaiacol ...... 1.36 analysis. The reactions were carried out in 10-ml tubes with 4,5,6-Tribromoguaiacol ...... 1.41 ...... 1.86 Teflon-lined screw caps. The tubes contained cell extract Tetrabromoguaiacol 3,4,5-Tribromoveratrole ...... 1.18 to 200 ca. 1.0 mg of 40 nmol of (100 ,ul, protein), S-adenosyl- Tetrabromoveratrole ...... 1.58 L-methionine p-toluene sulfonate (Sigma Chemical Co.) (20 526 NEILSON ET AL. APPL. ENVIRON. MICROBIOL. comparison with known compounds. A VG Masslab TRIO-2 mass spectrometer was used under the operating conditions previously described (16). Protein content in the cell extracts was determined by the method of Lowry et al. (14) by using bovine serum albumin 0A4 (fraction V, Sigma) as the standard. 0.3-

RESULTS AND DISCUSSION 0.2- The data summarized in Table 2 showed that the 0- 0.1 methylation reaction was mediated by a soluble enzyme which probably used S-adenosyl-L-methionine as the natural 0 0.5 1.0 0 20 40 60 6 7 8 9 10 methyl donor and which was not adversely affected by the mg PROTEIN TiME (MIN.) pH presence of up to 10 mM magnesium chloride or EDTA. The FIG. 1. Rates of synthesis of 2,6-dibromoanisole from 2,6-dibro- enzymatic reaction was linear up to protein concentrations mophenol by cell extracts from strain 1395. of ca. 1 mg. ml-', was essentially linear with time up to 60 min, and had a pH optimum of ca. 7.0 (Fig. 1). All of the products formed were identical to those synthe- between 3,4,5-trichlorophenol (specific activity, 0.7) and sized by the whole-cell suspensions which we have exam- 2,6-dichlorothiophenol (specific activity, 105). The signifi- ined. The products from the thiophenols were conclusively cance of a number of structural effects clearly emerged: (i) identified by mass spectrometric comparison with known the pattern of substitution, e.g., 2,4,6- versus 3,4,5-trichlo- compounds (Fig. 2). Mono-O-methylation of the chloro- and rophenol; (ii) the markedly greater activity of the haloge- bromocatechols to the guaiacols was demonstrated, and the nated thiophenols compared with the corresponding phe- predominant products (>80%) from the trihalogenated com- nols, which is consistent with the fact that the former pounds were the 3,4,5- isomers (Fig. 3). This finding is compounds are much more effective nucleophiles than the consistent with that of a previous experiment using cultures latter; (iii) the greater activity of the halogenated catechols of the same strain (1395) growing at the expense of 4- than that of the corresponding guaiacols; and (iv) the appar- hydroxybenzoate (1). Consistent with the studies using ent lack of a consistent effect of substituting bromine for whole cells, the activity of extracts from the gram-positive chlorine. strains (1395 and 1632) was significantly greater than that of For the gram-positive strain 1395, there was only a mod- extracts from the gram-negative strains (1631 and 1678) erately significant correlation (P = 0.08) between the specific (Table 3). These results reinforce our contention that 0- activity observed for whole cells and that found in the cell methylation may be accomplished effectively, particularly extracts. by gram-positive organisms. It should be noted particularly that whereas all of the The enzyme was clearly able to methylate a structurally simple halogenated phenols and thiophenols (with the excep- diverse range of halogenated phenolic compounds. Nonethe- tion of hexachlorophene) were methylated either by cell less, there were significant differences in enzymatic activity extracts or by whole cells, there were some significant towards the various substrates (Table 3 and Table 4), and exceptions. 5-Chlorovanillin and 2,4-dinitrophenol were not these results were entirely consistent with the data from methylated by whole cells of strain 1395 and were apparently experiments which used whole cells (1, 2). Although 0- degraded without forming detectable intermediates (unpub- methylation of pentafluorophenol had been demonstrated lished results); neither of these compounds nor 5-bromova- with whole-cell suspensions (2), the rate of pentafluoro- nillin was methylated by cell extracts. Therefore, permeabil- phenol 0-methylation by cell extracts was too low to be ity barriers were not responsible for the resistance of these measured after incubation for 1 h. Specific activities for the compounds to 0-methylation, and it is probable that the other substrates ranged over at least two powers of ten presence of the strongly electron-attracting aldehyde and nitro groups renders the electron density on the phenolic groups too low for reaction with the methyl TABLE 2. Specific activity of cell extracts from strain 1395 group of towards S-adenosyl-L-methionine. Therefore, these results serve to 2,6-dibromophenol (40 ,uM) underline our contention (2) that 0-methylation of halogen- Sp act (pmol - ated phenolic compounds should be regarded as an impor- Assay conditionsa min-' * mg of protein -) tant alternative to biodegradation. The specific activities were of the same order of magnitude Crude extract alone ...... 1.0 as those found for 0- and S-methylation of substituted Crude extract + SAM (4 ,uM) ...... 5.2 phenolic compounds in investigations in other laboratories Crude extract + SAM (10 iM) ...... 6.9 using bacterial cell extracts. Our results were to Crude extract + SAM (20 p.M) ...... 7.7 comparable those of a Crude extract + SAM (30 uM) ...... 8.6 Japanese study (19) using extracts from a Myco- Crude extract + SAM (40 FiM) ...... 8.3 bacterium sp. assayed against a range of chlorophenols, Crude extract + THF (40 ,iM) ...... 1.0 tetrachloroguaiacol, tetrachlorocatechol, and two polyhalogenated thiophenols, although our activity was more strin- SAM minus extract ...... 0 gently dependent on the structure of the substrate. The THF minus extract ...... 0 enzymatic activities found in the present investigation were Boiled extract + SAM (40 jiM) ...... 0 broadly similar to the thiol Final + SAM methyltransferase activities pellet (40 FiM) ...... 0.9 found in a recent study (7) using extracts from a Corynebac- a Abbreviations: SAM, S-adenosyl-L-methionine; THF, 5-methyltetrahy- terium sp. and from three Pseudomonas sp. strains assayed drofolate. against 2-nitrothiophenol and benzthiazole-2-thiol. In the VOL. 54, 1988 CELL-FREE METHYLATION OF HALOGENATED PHENOLS 527

la S. M. A 142 159 7'. N. N1 a. U1 I11 177 23. Is. elS I1I21 1..] II..1 I133 1.1 II ,-I,,, ,,,,,19 ,*,,,,a,,I,',,,,*,,I'III,P :' I,,,,, ,, l 1l.1.1-1 i,q |||.1.,, , , 'F- ,- r2| L-219 N1 K11 U 123 N1IU 1K 2K 22Z

11 9 3~~~~~~~~~~~~~~~~~~~~~~~~~~~~~3 I 71. I 142 ~~~~~~~~~~~~~~~~~~~~~159 I 5so3a2 ss 1 12 41N ?l?|l@SlB9 iie t , 11,1l, 711 1 tZ21 3 a I N K 1 1a 14 li 1K 2K 22

214 K B 71 155 U 191 23 636a ; 3N 1111 143 11 x 5 n r i In~~12II 1 leI3I l K 2K 22 *| 7 1 11111I* -8.c .1I '. '1.' * TI. n'IS' ' '-. 14 I",' * JI'.".'oi IIIT IIII,i'.""VIII',

I 214

15 21|

11 63 757p M III 1l1 175J It |3 1Jr 115 46tC IV7155iIU104j I16 In M 1 133 12.1. A11,1t.. Ila

S 5'I 1 , 2671 193 3* N III~1117 263 ill29

7 I i e 1 n 231N

N1 1K I M 256

246

23 ,- iL1

FIG. 2. Mass spectrometric comparison of products from cell-free methylation of thiophenols with those of known thioanisoles. The upper panel refers to the metabolites; the lower one refers to the standards. (A) 2,6-Dichlorothioanisole. (B) Pentafluorothioanisole. (C) Pentachlorothioanisole. 528 NEILSON ET AL. APPL. ENVIRON. MICROBIOL.

In 36 A 71

56. 77 46 345

a316

4 33 317 331 I1971111|-~~~~~~JLlo IL A 672? JL :W-k1 L. 56 16 156 M6 25 3 356 46

11[K 361 I I M~~~~~~~~~

I 5 4 3316 2 I70 26 M w,n,7l,,.sl,,.113Inlog 2I2I342 56 166 156 266 256 3 356 466 IK ~~~~~~~~~~~~~~~~~~~~~~~226 96.* B

56 211

a ? P .. 147 6 6 to 71 77 04 99 111 9 133 L. i1813 i M 1In 126 4 1if 166 26 226 241 266 266

76 66 211

04 II 63 ~~~~~~147163 27

166 12 1N 1n 1n M a M 2M FIG. 3. Mass spectrometric comparison of products from cell-free methylation of trihalogenocatechols with those of known 3,4,5- trihalogenoguaiacols. The upper panel refers to the metabolites; the lower one refers to the standards. (A) 3,4,5-Tribromoguaiacol. (B) 3,4,5-Trichloroguaiacol. latter investigation, substituted phenols were apparently not ylation activity was constitutive. This is consistent with our examined, although in a previous one (10) on the specificity previous finding 0-methylation was elicited effectively dur- of the enzyme from Tetrahymena thermophila activity was ing growth of the cells with betaine, gluconate, or succinate apparently restricted to substituted thiophenols, and chloro- (1). These results are environmentally significant since they phenols did not serve as effective substrates. On the basis of the present evidence, we cannot determine whether the same enzyme is involved in all of these methylation reactions, nor can we assess its possible relationship to the TABLE 3. Specific activities of cell extracts from various strains catechol 0-methyltransferase isolated from Candida tropi- supplemented with S-adenosyl-L-methionine (40 ,uM) calis (20). Sp act (pmol - min- * mg of protein-') For strain 1395, enzymatic activity was not influenced by of extract from: the nature of the growth substrate since the specific activities (in picomoles per minute per milligram of protein) after Substrate Gram-positiveGram-positive strain: Gram-negative growth with 4-hydroxybenzoate (8.2), succinate (9.2), and ~~~strain: nutrient broth (14.3) were not considered to be significant. In 1395 1623 1632 1631 1678 addition, no enhanced activity (8.2 pmol. min-' mg of protein-') was found when 4,5,6-trichloroguaiacol (100 ,g. 2,6-Dibromophenol 10.0 0.65 18.1 1.4 1.7 6.8 0.3 13.0 0.7 1.0 was present the last 10 h of with 4,5,6-Trichloroguaiacol liter-') during growth Pentachlorothiophenol 81 3.9 199 21 10 4-hydroxybenzoate. Therefore, we suggest that the 0-meth- VOL. 54, 1988 CELL-FREE METHYLATION OF HALOGENATED PHENOLS 529

TABLE 4. Specific activities of cell extracts of strain 1395 sistence of halogenated phenols and thiophenols discharged supplemented with S-adenosyl-L-methionine (40 ,uM) into the aquatic environment should take into account all of towards various substrates these additional factors. Sp act (pmol - Substrate min-' * mg of protein- 1) ACKNOWLEDGMENTS 2,6-Dichlorophenol ...... 4.4 We thank the Knut and Alice Wallenberg Fund for partial 2,6-Dibromophenol ...... 7.8, loa financial support towards the purchase of the mass spectrometer and 2,6-Dichlorothiophenol ...... 83, 129a the high-performance liquid chromatography equipment and Bo 2,4,6-Trichlorophenol ...... 12, 17a Lambert, Karolinska Institutet Department of Clinical Genetics, for 2,4,6-Tribromophenol...... 5.0, 7.6a providing extended access to the sonicator. 3,4,5-Trichlorophenol ...... 0.7, 1.0a Pentachlorophenol ...... 1.0, 1.4a Pentachlorothiophenol...... 81, 110a LITERATURE CITED Pentafluorothiophenol ...... 33, 64a ...... 3.8 3,4,5,-Trichloroguaiacol 1. Allard, A.-S., M. Remberger, and A. H. Neilson. 1985. Bacterial ...... 6.8'a 4,5,6-Trichloroguaiacol 4.4, 0-methylation of chloroguaiacols: effect of substrate concentra- Tetrachloroguaiacol ...... 0.9, 1.5a tion, cell density, and growth conditions. Appl. Environ. Mi- ...... 1.6 3,4,5-Tribromoguaiacol crobiol. 49:279-288...... 7.5' 4,5,6-Tribromoguaiacol 4.4, 2. Allard, A.-S., M. Remberger, and A. H. Neilson. 1987. Bacterial ...... 1.9 Tetrabromoguaiacol 0-methylation of halogen-substituted phenols. Appl. Environ. 3,4,5-Trichlorocatechol...... job Microbiol. 53:839-845. Tetrachlorocatechol ...... 4.4C 3,4,5-Tribromocatechol ...... 27, 3. Atlas, E., K. Sullivan, and C. S. Giam. 1986. Widespread 28a,b occurrence of aromatic ethers in Tetrabromocatechol ...... polyhalogenated the marine 7.9c atmosphere. Atmos. Environ. 20:1217-1220. a On the basis of determinations from two separate extracts. 4. Buser, H.-R., and M. D. Muller. 1986. Methylthio metabolites of b Calculated for synthesis of the 3,4,5-trihalogenoguaiacols. polychlorobiphenyls identified in sediment samples from two c Calculated for synthesis of the guaiacols. lakes in Switzerland. Environ. Sci. & Technol. 20:730-735. 5. Cairns, T., E. G. Siegmund, and F. Kirk. 1987. Identification of several new metabolites from pentachlorobenzene by gas chro- imply that cells do not require previous exposure to a matography/mass spectrometry. J. Agric. Food Chem. 35:433- xenobiotic compound to accomplish its 0-methylation. 439. These results confirm and extend our hypothesis that the 6. Cserjesi, A. J., and E. L. Johnson. 1972. Methylation of pen- 0-methylation of halogenated phenols, guaiacols, catechols, tachlorophenol by Trichoderma virgatum. Can. J. Microbiol. and thiophenols may be a significant alternative to biodeg- 18:45-49. radation. In view of the known toxicity and bioconcentration 7. Drotar, A., G. A. Burton, Jr., J. E. Tavernier, and R. Fall. 1987. Widespread occurrence of bacterial thiol methyltransferases potential of these neutral metabolites (18), this reaction may and the biogenic emission of methylated sulfur gases. Appl. be of particular environmental significance and may provide Environ. Microbiol. 53:1626-1631. at the same time a plausible rationalization of the occurrence 8. Drotar, A. M., and R. Fall. 1985. Microbial methylation of in the environment of halogenated anisoles having no estab- benzenethiols and release of methylthiobenzenes. Experientia lished anthropogenic origin (see the references in reference 41:762-764. 1). The environmental occurrence of thioanisoles, however, 9. Drotar, A. M., and R. R. Fall. 1985. Methylation of xenobiotic is more complex. Their presence in environmental samples thiols by Euglena gracilis: characterization of a cytoplasmic (4, 5) must be interpreted with caution since it is mechanis- thiol methyltransferase. Plant Cell Physiol. 26:847-850. a tically more likely that are formed from non-sulfur- 10. Drotar, A. M., and R. R. Fall. 1986. Characterization of they xenobiotic methyltransferase and its role in detoxification in containing precursors by nucleophilic displacements medi- Tetrahymena thermophila. Pestic. Biochem. Physiol. 25:396- ated by a glutathione-dependent reaction; for example, this 406. process has been demonstrated unambiguously with a num- 11. Gee, J. M., and J. L. Peel. 1974. Metabolism of 2,3,4,6- ber of microorganisms by using 1-chloro-2,4-dinitrobenzene tetrachlorophenol by microorganisms from broiler house litter. as a substrate (12) and in more detail with T. thermophila by J. Gen. Microbiol. 85:237-243. using pentachloronitrobenzene as a substrate (15). 12. Lau, E. P., L. Niswander, D. Watson, and R. R. Fall. 1980. To put these observations on cell-free 0- and S-methyla- Glutathione-S-transferase is present in a variety of microor- tion into environmental perspective, one must take into ganisms. Chemosphere 9:565-569. and F. 1980. struc- consideration a number of additional facts. Neither reac- 13. Lindstrom, K., Osterberg. Synthesis, X-ray (i) ture determination, and formation of 3,4,5-trichloroguaiacol tion is confined to procaryotic organisms, since 0-methyla- occurring in kraft pulp spent bleach liquors. Can. J. Chem. 58: tion is well documented in a number of fungi (6, 11) and 815-822. S-methylation has been demonstrated in two protozoa, Eu- 14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. glena gracilis (9) and T. thermophila (10), and in the yeast 1951. Protein measurement with the Folin phenol reagent. J. Saccharomycopsis lipolytica (8). (ii) Although the methyla- Biol. Chem. 193:265-275. tion reaction has hitherto been restricted to aerobic organ- 15. Murphy, S. E., A. Drotar, and R. Fall. 1982. Biotransformation isms, the demethylation of chloroanisoles and chlorovera- of the fungicide pentachloronitrobenzene by Tetrahymena ther- troles which has been demonstrated under anaerobic mophila. Chemosphere 11:33-39. is in anoxic sediments. 16. Neilson, A. H., A.-S. Allard, P.-A. Hynning, M. Remberger, and conditions (17) probably widespread L. Landner. 1983. Bacterial methylation of chlorinated phenols (iii) The occurrence of halogenated anisoles in air samples and guaiacols: formation of veratroles from guaiacols and high- from isolated regions (3) is consistent with the relatively high molecular-weight chlorinated lignin. Appl. Environ. Microbiol. vapor pressures of these compounds and illustrates nicely 45:774-783. the complexity of distribution patterns which may emerge. 17. Neilson, A. H., A.-S. Allard, C. Lindgren, and M. Remberger. Therefore, a satisfactory assessment of the fate and per- 1987. Transformations of chloroguaiacols, chloroveratroles, and 530 NEILSON ET AL. APPL. ENVIRON. MICROBIOL.

chlorocatechols by stable consortia of anaerobic bacteria. Appl. tion. Can. J. Fish. Aquat. Sci. 41:1502-1512. Etiviron. Microbiol. 53:2511-2519. 19. Suzuki, T. 1978. Enzymatic methylation of pentachlorophenol 18. Neilson, A. H., A.-S. Allard, S. Reiland, M. Remberger, A. and its related compounds by cell-free extracts of Mycobacte- Tarnholm, T. Viktor, and L. Landner. 1984. Tri- and tetra- rium sp. J. Pestic. Sci. 3:441-443. chloroveratrole, metabolites produced by bacterial 0-methyla- 20. Veser, J. 1987. Kinetics and inhibition studies of catechol tion of tri- and tetra-chloroguaiacol: an assessment of their O-methyltransferase from the yeast Candida tropicalis. J. Bac- bioconcentration potential and their effects on fish reproduc- teriol. 169:3696-3700.