Purification and Characterization of Streptomyces griseus Catechol O-Methyltransferase

Abstract

A soluble (100,000 × g supernatant) methyltransferase catalyzing the transfer of the methyl group of S-adenosyl-l-methionine to catechols was present in cell extracts of Streptomyces griseus. A simple, general, and rapid catechol-based assay method was devised for enzyme purification and characterization. The enzyme was purified 141-fold by precipitation with ammonium sulfate and successive chromatography over columns of DEAE-cellulose, DEAE-Sepharose, and Sephacryl S-200. The purified cytoplasmic enzyme required 10 mM magnesium for maximal activity and was catalytically optimal at pH 7.5 and 35°C. The methyltransferase had an apparent molecular mass of 36 kDa for both the native and denatured protein, with a pI of 4.4. Novel N-terminal and internal amino acid sequences were determined as DFVLDNEGNPLENNGGYXYI and RPDFXLEPPYTGPXKARIIRYFY, respectively. For this enzyme, the K_m for 6,7-dihydroxycoumarin was 500 ± 21.5 μM, and that for S-adenosyl-l-methionine was 600 ± 32.5 μM. Catechol, caffeic acid, and 4-nitrocatechol were methyltransferase substrates. Homocysteine was a competitive inhibitor of S-adenosyl-l-methionine, with a K_i of 224 ± 20.6 μM. Sinefungin and S-adenosylhomocysteine inhibited methylation, and the enzyme was inactivated by Hg²⁺, p-chloromercuribenzoic acid, and N-ethylmaleimide.

A variety of bacterial and fungal methyltransferases are involved in antibiotic (3, 8–10, 15, 18, 29) and aflatoxin (37, 38) biosynthesis, as well as in the methylation of other compounds (7, 12, 19, 24). Methyltransferases isolated and characterized from streptomycetes usually have defined roles in antibiotic biosynthesis. Biosynthetically related enzymes associated with secondary metabolism often display kinetic properties, including very high substrate specificities, relatively low K_m and V_max values rendering them generally impractical for broad-scale biocatalytic use.

We previously showed that cultures of Streptomyces griseus sequentially converted 7-methoxycoumarin to a mixture of 6-hydroxy-7-methoxycoumarin and 7-hydroxy-6-methoxycoumarin (27) by a pathway involving O deethylation, aromatic hydroxylation to a 6,7-catechol, and subsequent methylation. The intermediacy of 6,7-dihydroxycoumarin was established by incubating growing cultures of S. griseus with the catechol as a substrate to accumulate both monomethyl-ether isomers. The formation of 7-hydroxy-6-methoxycoumarin and 6-hydroxy-7-methoxycoumarin indicated the presence of a methyltransferase enzyme system in S. griseus. The incorporation of the ¹⁴C-labeled methyl group from ¹⁴C-methylmethionine demonstrated that an S-adenosyl methionine (SAM)-dependent transmethylating system was involved in this reaction. We now describe the development of a simple catechol-based assay method and its use in the purification and characterization of a new and highly active catechol O-methyltransferase (COMT) from S. griseus.

MATERIALS AND METHODS

Materials and reagents.

6,7-Dihydroxycoumarin, catechol, 4-nitrocatechol, fisetin, quercetin, dopa, dopamine, epinephrine, norepinephrine, S-adenosyl-l-methionine (SAM), dithiothreitol (DTT), phenylmethylsulfonyl fluoride (PMSF), adenosine triphosphate, methyltetrahydrofolate, methionine, methylmethionine, methylcobalamin, p-chloromercuribenzoate, N-ethylmaleimide, protocatechuic acid, DEAE-Sepharose, and Sephacryl S-200 were purchased from Sigma Chemical Co. (St. Louis, Mo.). Caffeic acid was from Lancaster Synthesis, Inc. (Windham, N.H.). DEAE-cellulose (DE-52) was from Whatman, Ltd. (Maidstone, Kent, England). Polyvinylidene difluoride membranes were from Millipore (Bedford, Mass.). Reagents for protein assay and electrophoresis were from Bio-Rad (Hercules, Calif.) and Pierce (Rockford, Ill.). High-performance liquid chromatography (HPLC) analysis and metabolite isolation were carried out with a Shimadzu instrument (Columbia, Md.). Reversed-phase (C₁₈) columns were purchased from Alltech (Deerfield, Ohio).

Chromatography.

Thin-layer chromatography (TLC) was performed on silica gel GF₂₅₄ (Merck) layers of 0.5 mm of thickness for analysis and 1 mm of thickness for preparative layer chromatography. Layers were prepared on glass plates (20 by 20 cm) with a Quickfit Industries (London, England) spreader. Plates were air dried and activated at 120°C for 1 h before use. Plates were developed in several solvent systems, and the developed chromatograms were visualized under 254-nm UV light before being sprayed with Pauly's reagent to detect phenols. Pauly's reagent consisted of three separate solutions: 0.5% NaNO₂, 0.5% sulfanilic acid in 2% HCl, and 5% NaOH in 50% ethanol. Equal volumes of NaNO₂ and sulfanilic acid solutions were mixed immediately prior to use, and plates were sprayed with this mixture and then with NaOH before being warmed with a heat gun to give burnt-orange and red colors for phenolic compounds.

Column chromatography was carried out with silica gel (Baker 60, 40 to 140 mesh; Phillipsburg, N.J.), which was oven activated at 120°C for 30 to 60 min before use. Columns were slurry packed in benzene and eluted with a mixture of benzene-ethanol (70:30 [vol/vol]), while 8-ml fractions were collected at flow rate of 1.5 ml/min. The collected fractions were examined by TLC.

HPLC was performed with a Shimadzu SPD-M6A instrument equipped with an LC-10AD solvent delivery system, SIL-10A Shimadzu auto injector, and a photodiode array detector. A reversed-phase μ-Bondapak-C₁₈ column (Alltech; 25 by 0.46 cm) was eluted with a solvent system of methanol-water-acetic acid (25:75:0.1) at a flow rate of 1 ml/min and an operating pressure of 76 kg/cm². Under these conditions, retention volumes of 6,7-dihydroxycoumarin, 6-hydroxy-7-methoxycoumarin, and 7-hydroxy-6-methoxycoumarin were 10, 14, and 25 ml, respectively.

Organism and culture conditions.

S. griseus ATCC 13273 is maintained in the University of Iowa, College of Pharmacy, culture collection and was grown and stored on Sabouraud dextrose agar in sealed screw-cap tubes at 4°C. Organisms were cultivated in a medium composed of 5 g of soybean meal, 20 g of dextrose, 5 g of yeast extract, 5 g of NaCl, and 5 g of K₂HPO₄ in 1 liter of distilled water, which was adjusted to pH 7.0 with 6 N of HCl. Media were sterilized in an autoclave at 121°C for 20 min. Cultures were incubated by shaking at 250 rpm at 28°C on New Brunswick Scientific (Edison, N.J.) G25 Gyrotory shakers. A 10% inoculum derived from a 72-h-old, first-stage culture was used to initiate the second-stage culture, which was incubated as described before. After 24 h, 6,7-dihydroxycoumarin (2 mg/ml [11 mM]) was added to the second-stage culture to induce enzyme synthesis. The culture was harvested 48 h later, and cells were pelleted by centrifugation at 10,000 × g for 20 min and washed with 0.5% NaCl (wt/vol). Cell pellets were stored at −70°C until needed. Typical wet cell yields by this cultivation process were approximately 20 g/liter. When metabolites of 6,7-dihydroxycoumarin were isolated and purified by chromatography, cultures were pooled and extracted exhaustively with ethyl acetate, and then the extracts were dried over anhydrous sodium sulfate and evaporated to a brown oil.

COMT enzyme assay.

A general procedure for measuring catechols was based upon Arnow's method (2). The COMT assay buffer was 50 mM Tris-HCl (pH 7.5) containing 1 mM DTT, 10 mM MgCl₂, 2 mM SAM, 1 mM 6,7-dihydroxycoumarin, and 20 to 200 μg of enzyme in a final volume of 1 ml. Enzyme reaction mixtures were incubated at 35°C for 30 min before being terminated by adding 1 ml of 0.5 N HCl, 1 ml each of 10% NaNO₂ and 10% NaMo0₄, and finally 1 ml of 1 N NaOH. The A₅₁₀ of this solution was immediately measured to determine the amounts of catechol substrates consumed in the reaction. Units of COMT activity were defined as the amount of enzyme that catalyzed the methylation of 1 μmol of 6,7-dihydroxycoumarin to either 7-hydroxy-6-methoxycoumarin or 6-hydroxy-7-methoxycoumarin per min under standard assay conditions.

Protein was measured by the Bradford protein microassay (5) with bovine serum albumin as the standard.

Induction of COMT activity.

The time course for appearance of COMT activity in S. griseus was measured in cultures from 0 to 120 h after addition of 6,7-dihydroxycoumarin as an inducer. To establish that enzyme synthesis was attributable to the presence of inducer, tetracycline (200 μg/ml) and choramphenicol (350 μg/ml) were added 18 h after induction was started. At each time point, 12 g of mycelium obtained from 600 ml of culture was washed with 0.5% NaCl, pelleted by centrifugation at 20,000 × g for 10 min, and resuspended in 100 ml of 50 mM Tris-HCl (pH 7.5), containing 10 mM MgCl₂, 2 mM PMSF, 2 mM DTT, and 10% glycerol. Cells were disrupted by French press homogenization at 18,000 lb/in², and the resulting homogenate was centrifuged for 20 min at 20,000 × g. The supernatant was then centrifuged at 100,000 × g for 45 min to give a soluble preparation suitable for COMT and protein assays.

COMT purification.

A 60-g sample of cell pellet was suspended in 160 ml of cold 50 mM Tris-HCl buffer (pH 7.5) containing 2 mM PMSF, 2 mM DTT, 10 mM MgCl₂, 0.1 mM SAM, and 10% (vol/vol) glycerol (standard buffer) and passed twice through a French press at 15,000 lb/in². The cell homogenate was centrifuged at 20,000 × g for 30 min, and the resulting supernatant was centrifuged again at 100,000 × g for 1 h 15 min at 4°C. The 100,000 × g supernatant was used directly for subsequent enzyme purification, all of which was conducted at 4°C. The 100,000 × g supernatant was brought to 40% ammonium sulfate saturation and stirred for 30 min. The cloudy suspension was centrifuged at 20,000 × g for 20 min, and the supernatant was brought to 85% ammonium sulfate saturation. After being stirred for 30 min, the pellet obtained by centrifugation at 20,000 × g for 20 min was dissolved in standard buffer and dialyzed against 10 mM Tris-HCl (pH 7.5) containing 1 mM DTT, 10 mM MgCl₂, 0.1 mM SAM, and 10% glycerol. The supernatant obtained after centrifugation of the dialysate at 100,000 × g (30 ml, 695 mg of protein) was loaded onto a DE-52 column (50 by 2 cm), which was previously equilibrated with standard buffer. The column was washed with 1 bed volume consisting of 75 ml of the same buffer at a flow rate of 30 ml/h. Then enzyme was eluted by a linear gradient of 0 to 0.5M KCl in the same buffer, while 2-ml fractions were collected and assayed for COMT activity.

Fractions (80 to 100) were pooled, lyophilized, resuspended in 10 ml of standard buffer, and then dialyzed for 4 h against standard buffer. The dialyzed preparation (41 mg of protein) was then applied to a DEAE-Sepharose column (30 by 2 cm) previously equilibrated with the same buffer. The enzyme was eluted with a 0 to 0.5 M KCl linear gradient at a flow rate of 30 ml/h while 2-ml fractions were collected. Active fractions (55 to 66) were combined, concentrated by lyophilization, resuspended in 1 ml of standard buffer, and dialyzed against standard buffer. This preparation was then loaded onto a Sephacryl S-200 column (90 by 1 cm), equilibrated and eluted with the same buffer, and 0.5-ml fractions were collected. Active fractions 35 to 40 containing pure COMT were combined for subsequent analysis.

SDS-PAGE.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to the method of Laemmli (21) with 3% stacking and 12% separating gels containing 0.1% SDS. The protein standards (Sigma) used for estimation of subunit molecular masses were bovine serum albumin (66 kDa), ovalbumin (45 kDa), glyceraldehyde-3-phosphodehydrogenase (36 kDa), carbonic anhydrase (29 kDa), bovine pancreas trypsinogen (24 kDa), soybean trypsin inhibitor 20.1 kDa), and bovine milk α-lactalbumin (14.2 kDa).

Native M_r determination.

The native molecular mass (M_r) of the COMT was obtained by chromatography over a column (90 by 1 cm) of Sephacryl S-100, eluted with 20 mM Tris-HCl (pH 7.5) containing 1 mM DTT, 10 mM MgCl₂, and 0.1 mM SAM at a flow rate of 0.2 ml/min. Molecular standards run simultaneously were alcohol dehydrogenase (150 kDa), bovine serum albumin (66 kDa), and carbonic anhydrase (29 kDa).

pI determination.

The pI of the purified enzyme was estimated by chromatofocusing on a PBE 94 (Pharmacia) column. The sample to be examined was dialyzed in 20 mM imidazole-HCl buffer (pH 7.4) containing 1 mM DTT, 10 mM MgCl₂, and 0.1 mM SAM. The column was eluted with Polybuffer 74 (Pharmacia) covering a pH range of 4 to 7, while 1-ml fractions were collected. The pH of each fraction was immediately measured, and fractions were assayed for enzyme activity.

N-terminal and internal amino acid sequence determinations.

Amino acid sequences were determined by Edman degradation and analysis in an automated sequencer at the Protein Structure Facility, College of Medicine, University of Iowa, Iowa City.

Protein cleavage for peptide mapping was carried out at 25°C for 1 h with 100 ng of α-chymotrypsin (catalog no. 7762; Sigma) to digest 20 μg of purified enzyme in 50 μl of 50 mM (NH₄)₂CO₃ (pH 8.5). The resulting peptide fragments were separated by SDS-PAGE (15% polyacrylamide). The separated peptides were transferred to a polyvinylidene difluoride membrane by electroblotting at 50 V for 1 h in 0.3 M Tris-HCl (pH 10.4) containing 10% methanol. Peptide bands were visualized by 0.1% Coomassie blue R-250 staining in 40% methanol. A 12-kDa peptide fragment was selected for N-terminal amino acid sequence analysis.

Kinetics.

A highly purified COMT preparation was used for all enzyme kinetic studies. Kinetic parameters (V_max and K_m) were obtained for 6,7-dihydroxycoumarin and SAM from Lineweaver-Burk double-reciprocal plots under initial velocity conditions. Values for the variable substrate 6,7-dihydroxycoumarin (0.1 to 1.0 mM) were determined at one saturating concentration of SAM (2 mM), and similarly the values for variable SAM (0.1 to 1.0 mM) were measured at a fixed concentration (1 mM) of 6,7-dihydroxycoumarin. The K_i of purified COMT for homocysteine was determined in incubation mixtures containing 1.0 mM 6,7-dihydroxycoumarin, 0.2 to 1.0 mM SAM, and 0.2 to 1.0 mM homocysteine.

Kinetic constants were determined in duplicate at two different protein concentrations by using Lineweaver-Burk and the EZ-fit program developed by Perrella (25).

Reaction stoichiometry.

The mole ratio of product formed versus substrate consumed was determined by sampling a 0.2-ml incubation containing 20 μg of pure COMT at 10, 30, and 50 min. Samples were diluted with equal volumes of methanol, and 50 μl of these solutions was injected for HPLC analysis.

Unless otherwise stated, all experiments were carried out at least in triplicate, and the results were expressed as means of those data. Standard deviations were determined and are reported. All purification steps were performed at least three times with similar results.

RESULTS

For enzyme purification and characterization, we devised a simple and general spectrophotometric assay method based upon the reaction of catechols with sodium nitrite and sodium molybdate under alkaline conditions. In general, catechols form deep purple complexes with A₅₁₀. Standard curves of 6,7-dihydroxycoumarin gave linear absorbances over the range of 10 to 60 μg/ml. Under the assay conditions, neither of the monomethyl-ether products obtained by 6,7-dihydroxycoumarin methylation (Fig. 1) reacted with Arnow's reagent, and they showed no interfering A₅₁₀. Caffeic acid, 4-nitrocatechol, dopa, dopamine, norepinephrine, epinephrine, and protocatechuic acid all reacted with sodium nitrite and sodium molybdate in alkali to give purple complexes with A₅₁₀. Unlike other catechols, the flavonoid fisetin gave an absorbance maximum at 410 nm under standard assay conditions. Highly reproducible assays were obtained when as little as 0.015 U of COMT activity was used. The assay afforded a rapid, sensitive, and reproducible means of measuring the amount of catechol consumed during COMT methylation by S. griseus and provided the basis for enzyme purification.

FIG. 1.

S. griseus COMT conversions of 6,7-dihydroxycoumarin to 6-hydroxy-7-methoxycoumarin and 6-methoxy-7-hydroxycoumarin.

Mycelial cell growth and the time for detection of maximum COMT levels in cell extracts were established. COMT activity was barely detectable before 12 h, and it reached a maximum level 48 h after the addition of substrate to stage II cultures (Fig. 2). To establish that apparent enzyme induction was a function of de novo protein synthesis, cultures with tetracycline (200 μg/ml) or chloramphenicol (350 μg/ml) added 18 h after addition of inducer gave cell extracts containing about one-third of the COMT activity measured in normal induced cultures (Fig. 2). Based on these experiments, cell extracts were routinely prepared from stage II cultures harvested 48 h after induction with 2 mg of 6,7-dihydroxycoumarin per ml.

FIG. 2.

Induction of COMT by S. griseus. 6,7-Dihydroxycoumarin (2 mg/ml [11 mM]) was added as an inducer to 24-h, second-stage cultures. Tetracycline at 200 μg/ml (▴) or chloramphenicol at 350 μg/ml (○) was added 18 h later, and COMT-specific activities in induced controls (□) were compared to those in antibiotic-containing and induced cultures. Mean values are shown for triplicate samples with standard deviations of not more than ±4%.

A four-step purification procedure was used to obtain pure COMT from S. griseus cell extracts centrifuged at 100,000 × g (Table 1). During purification, it was necessary to stabilize cell extracts by the inclusion of DTT, SAM, and glycerol in buffers. Ammonium sulfate precipitation removed nearly half of the protein from the crude cell extracts, while retaining 80.7% of the activity. A 15-fold increase in enzyme specific activity was achieved by DEAE-cellulose column chromatography, while DEAE-Sepharose and Sephacryl S-200 column chromatographies yielded COMT with a final specific activity of 19.7 U/mg of protein. The procedure was highly reproducible, giving an excellent yield of 4.1 mg of pure protein, with 46% recovery of activity and 141-fold purification from the crude cell extract. By this purification, soluble COMT represented approximately 0.3% of total protein in the crude extract.

TABLE 1.

Purification of S. griseus catechol O-methyltransferase

Purification stepa	Total protein (mg)b	Total activity (U)	Sp act (U/mg)	Yield (%)	Fold (sp act)
Crude extract	1,205	176	0.14	100
NH4SO4 (40–85%)	695	142	0.2	81	1.42
DE-52	41	125	3.0	71	21.7
DEAE-Sepharose	11	101	9.1	57	65
Sephacryl S-200	4.1	81	19.7	46	141

^aCrude extract included 2 mM PMSF, 10% glycerol, and 2 mM DTT in extraction buffer. 

^bFractions 80 to 100, 55 to 65, and 41 to 46 are represented by DE-52, DEAE-Sepharose, and Sephacryl S-200, respectively. 

COMT properties.

The purified soluble enzyme catalyzed the regiospecific transfer of the methyl group from SAM to the 6-hydroxyl group of 6,7-dihydroxycoumarin (Fig. 1). HPLC analysis showed a 1:1 mole ratio of 6,7-dihydroxycoumarin consumed to 6-methoxy-7-hydroxycoumarin formed during enzyme reactions, with no 7-methoxy-6-hydroxycoumarin observed. The enzyme was stable in the presence of 1 mM DTT, 10% (vol/vol) glycerol, and 0.1 mM SAM. In their absence, all activity was lost over 2 to 3 days at 4°C. The native molecular mass was 37.5 kDa by gel filtration chromatography and 36 kDa by SDS-PAGE (Fig. 3). The UV-visible spectrum of the purified enzyme had one absorption maximum at 280 nm, indicating the lack of prosthetic groups such as flavin or heme. COMT had an optimum requirement of 10 mM Mg⁺², showed maximum activity at pH 7.5 and 35°C, and had a pI of 4.4. K_m values of 500 ± 21.5 and 600 ± 32.5 μM were determined for 6,7-dihydroxycoumarin and SAM, respectively. EDTA (1 mM) completely inhibited the methyltransferase reaction, confirming the requirement of the enzyme for Mg²⁺. Homocysteine was a competitive inhibitor of SAM, with a K_i of 224 ± 20.6 μM. Known methyltransferase inhibitors S-adenosylhomocysteine, dl-homocysteine, and sinefungin all inhibited the COMT reaction (Table 2). Mercuric chloride, p-chloromercuribenzoate, N-ethylmaleimide, and cadmium acetate (all at 1 mM) inhibited the COMT reaction to various degrees (Table 2), indicating the participation of an SH group in catalysis. The unique N-terminal and internal amino acid sequences of the purified enzyme were DFVLDNEGNPLENNGGYXYI and RPDFXLEPPYTGPXKARIIRYFY, respectively, where X indicates ambiguity in the identity of the amino acid.

FIG. 3.

SDS-PAGE analysis of purified S. griseus catechol O-methyltransferase (20 μg of protein) and protein markers (3.6 μg per band; lane 1) stained with Coomassie blue. Markers from top to bottom are bovine albumin (66 kDa), egg albumin (45 kDa), glyceraldehyde-3-phosphate dehydrogenase (36 kDa), carbonic anhydrase (29 kDa), bovine pancreas trypsinogen (24 kDa), soybean trypsin inhibitor (20.1 kDa) and bovine milk lactalbumin (14.2 kDa).

TABLE 2.

Inhibition of S. griseus O-methyltransferase^a

Inhibitor (1 mM)	% Inhibition
EDTA	100
S-Adenosylhomocysteine	63
Sinefungin	94
HgCl2	95
p-Chloromercuribenzoate	98
N-Ethylmaleimide	100
Cadmium acetate	50

^aThe data represent the mean of two determinations. The Control reaction (without inhibitor) had activity of 19.3 μmol · min⁻¹ · mg⁻¹. 

At substrate concentrations similar to those for 6,7-dihydroxycoumarin, purified S. griseus COMT also catalyzed methylations of catechol (32%), 4-nitrocatechol (6.2%), and caffeic acid (2.1%) (Table 3). However, fisetin, catechin, quercetin, dopa, dopamine, norepinephrine, epinephrine, and procatechuic acid were not substrates for the purified enzyme. On the other hand, crude 100,000 × g cell extracts catalyzed methylations of all these compounds (Table 3). Monophenolics like 7-hydroxycoumarin and 4-hydroxycoumarin were not substrates for either the purified or crude enzyme preparations.

TABLE 3.

Substrate specificities of purified S. griseus COMT and a crude cytosolic fraction^a

Substrate	Substrate consumption (μmol · min−1 · mg of protein−1)
	Supernatant (20,000 × g)	Purified COMT
Esculetin	0.14 ± 0.004	19.2 ± 0.80
Catechol	0.09 ± 0.002	6.3 ± 0.22
Caffeic acid	0.23 ± 0.007	0.4 ± 0.012
4-Nitrocatechol	0.08 ± 0.003	1.2 ± 0.061
Fisetin	0.12 ± 0.003	0
3,4-Dihydroxyphenylalanine	0.25 ± 0.008	0
Dopamine	0.22 ± 0.007	0
Norepinephrine	0.07 ± 0.001	0
Epinephrine	0.15 ± 0.003	0
Protocatechuic acid	0.09 ± 0.002	0

^aIncubations were conducted for 30 min at 35°C in 1.0-ml volumes (pH 7.5) of 50 mM Tris-HCl containing 1 mM DTT, 10 mM MgCl₂, 0.015 to 0.03 U of purified enzyme, 2 mM SAM, and 1 mM substrate. Reactions were stopped by adding 1.0 ml of 0.5 N HCl and assayed by adding 1 ml each of 10% NaNO₂, 10% NaMoO₄, and 2 ml of NaOH. λ_max values were measured at 510 nm for all substrates but fisetin, which was measured at 410 nm. 

Methylations were not observed when the reaction was conducted with cell extracts and several other potential methyl donors, including methylmethionine, methionine, methyltetrahydrofolate, and methylcobalamin.

DISCUSSION

SAM-dependent methyltransferases are well known in streptomycetes, other bacteria, and fungi (3, 7, 8, 10, 12, 13, 19, 38). Streptomycetes catalyze methylations of secondary alcohols (9, 19, 29) and phenols (3, 8, 10). C methylations are catalyzed by an S. griseus strain during indolmycin biosynthesis (14, 31, 32) and by Streptomyces flocculus (13, 32) and Streptomyces laurentii (11) during streptonigrin biosynthesis. Phenolic methylations also are observed during aflatoxin biosynthesis by Aspergillus parasiticus (37, 38) and in emodin methylation by Aspergillus terreus (7). In other bacteria, Pseudomonas SAM-dependent enzymes catalyze the C-20 methylation of precorrin-2 (34) and O methylation of isobutyraldehydoxime (12). A protein l-glutamate methyltransferase system is known in Salmonella enterica serovar Typhimurium (30), and multiple forms of O-methyltransferase appear to be involved in the microbial conversion of abietic acid into methylabietate in Mycobacterium sp. (24). Catechol O-methyltransferases are not well known in microorganisms.

S. griseus ATCC 13273 biosynthetically incorporates O- and C-methyl groups into four different positions of the antibiotic chromomycin A-3 (23). ¹³C-nuclear magnetic resonance spectral analysis showed specific incorporations of the four methyl groups from ¹³CH₃-SAM of 11.5 to 27% to indicate the presence of one or more methyltransferase enzyme systems in this organism. Although methyltransferases are potentially useful biocatalysts for O and/or C methylations, few such enzymes have been examined for that purpose. As with our earlier work (27), we confirmed that growing cultures of S. griseus ATCC 13273 achieved the efficient methylation of 6,7-dihydroxycoumarin to afford a mixture of 6- and 7-monomethyl-ethers. Metabolites were isolated and characterized by nuclear magnetic resonance and mass spectrometry to confirm the methylation pathway (data not shown).

The key to enabling the isolation and purification of a potentially useful catechol-O-methyltransferase enzyme resided in our abilities to establish a rapid and sensitive assay for assessing enzyme activity. Assays for most methylases involved in the biosynthesis of antibiotics and aflatoxins used radiolabeled ¹⁴CH₃-SAM or ³H-SAM and antibiotic precursors that become methylated. Such enzyme assays are laborious, require rare and often unavailable biosynthetic intermediates, and are time-consuming—all of which are detrimental to the purification of unstable enzymes. The assays were all done for 30 min, based upon optical density and HPLC analyses that indicated initial velocity linearities of at least 45 min under standard assay conditions. The linearity of the assay was confirmed by using both crude and purified COMT. The simple catechol-based enzyme assay procedure described here enabled the rapid purification of a new S. griseus COMT.

S. griseus COMT is clearly inducible. Enzyme activity increased with time, and the elimination of enzyme synthesis by tetracycline or chloramphenicol confirmed that COMT formation occurred by de novo protein synthesis rather than by liberation or activation of preformed constitutive enzyme. During initial purification attempts, enzyme activity in cell extracts was unstable. Protection of COMT inactivation by PMSF and improvement of its stability by each purification step indicated that enzyme instability in crude preparations was due to serine protease(s). Other stability problems were overcome by the incorporation of 10% glycerol or 10% ethanol and 0.1 mM SAM along with DTT into buffers as described by Bauer et al. (3). Purified enzyme remained stable for several months, with little loss of activity when stored at −70°C. Under stabilizing conditions, the methyltransferase was purified to apparent electrophoretic homogeneity by a reproducible four-step process.

Like other O-methyltransferases, Mg²⁺ was essential for enzyme activity (3, 13, 19, 28, 29). With a specific activity of 19.7 U · mg of protein⁻¹, S. griseus COMT is 70 to 50,000 times more active than constitutive mammalian or other bacterial and fungal methyltransferases (4). Only an enzyme from Aspergillus terreus has a comparable specific activity of 3.18 μmol min⁻¹ mg of protein⁻¹ (7). K_m values for SAM (0.625 mM) and for 6,7-dihydroxycoumarin (0.5 mM) for S. griseus COMT are similar to those reported for enzymes from other organisms (9, 10, 12). Standard inhibitors of SAM-dependent methylases were also classic inhibitors of S. griseus COMT.

Molecular masses of methyltransferases from streptomycetes vary. COMT from S. griseus is monomeric, with a molecular mass of approximately 36 kDa. A C-methyltransferase from S. antibioticus (10) is similar in mass, while that from an S. griseus strain involved in indolmycin biosynthesis is a 55-kDa monomer (32). Methyltransferases involved in carminomycin and macrocin biosyntheses are active only as a tetramer (166 kDa) (8) and a dimer (65 kDa) (3), respectively. Fungal O-methyltransferases involved in aflatoxin biosynthesis are 40 and 168 kDa (18). A review of the literature and BLAST database analysis (1) showed that N-terminal and internal amino acid sequences for S. griseus COMT were poorly comparable to methyltransferase sequences from Streptomyces, other bacteria, and fungi (18, 20, 22, 29, 34). While there is no apparent relationship among these proteins, S. griseus COMT shows nearly 80% homology to sequences from glutathione dehydrogenase (35) and to a soybean trypsin inhibitor (6). SAM-dependent methyltransferases are very diverse enzymes that show three important conserved sequence motifs (16). The motifs are small, while the remainder of the amino acid sequences of the enzymes may be very diverse. Thus, the inability to match the properties of our N- and internal amino acid sequences with those of other methyltransferases is not surprising. 6,7-Dihydroxycoumarin is both substrate and inducer. It is perhaps not surprising that the inducer shows the best activity of all substrates examined. Methylation only occurs with catechol substrates. Methylation is highly regiospecific with the purified enzyme, with only the 6 position of 6,7-dihydroxycoumarin being methylated. Catechol and nitrocatechol also were substrates for S. griseus COMT, but several other catechols were not. Interestingly, catechols including fisetin, quercetin, and catechin are methylated during biotransformations conducted with growing cultures of S. griseus (unpublished data). Almost all catechol substrates underwent methylation when used with crude cell extracts (Table 3), indicating that S. griseus contains other SAM-dependent methyltransferases. We considered the possibility that inactive proteins in crude extracts played a protective role for COMT. However, that possibility was ruled out when incubations with purified enzyme containing bovine serum albumin again failed to catalyze methylations of dopa, dopamine, fisetin, or epinephrine. Methylations were not observed when the reaction was conducted with cell extracts and several other potential methyl donors, including methylmethionine, methionine, methyltretrahydrofolate, and methylcobalamin. This means the enzyme is very specific to S-adenosylmethionine.

This work details the isolation and characterization of a new, bacterial COMT enzyme system. The enzyme reported here is unique based upon its functional molecular weight and measured amino acid sequences. Although a COMT enzyme system was immunocytochemically identified in the eukaryotic organism Candida tropicalis (36), there are no other reports of bacterial COMT enzyme systems. We believe that this is the first catechol O-methyltransferase isolated and characterized from bacteria.