Degradation of 1,2,3,4-Tetrachlorobenzene by Pseudomonas chlororaphis RW71

Abstract

Pseudomonas chlororaphis RW71 mineralized 1,2,3,4-tetrachlorobenzene, a highly recalcitrant pollutant hitherto not known to be degraded by pure cultures, as a sole source of carbon and energy, thereby releasing stoichiometric amounts of chloride. The transient excretion of tetrachlorocatechol in the early growth phase suggests an initial attack by a dioxygenase to form the corresponding dihydrodiol which rearomatizes to the catechol. The activity of chlorocatechol 1,2-dioxygenase in crude cell extracts was found to be extraordinarily high towards 3-chlorocatechol (ratio of 2.6 compared to catechol) and other chlorocatechols, including tetrachlorocatechol, which was transformed at a low but significant rate. Further identification of tetrachloromuconic acid, 2,3,5-trichlorodienelactone, 2,3,5-trichloromaleyl acetic acid, and 2,4-dichloro-3-oxoadipic acid as their methyl esters, together with high specific enzyme activities for chlorinated substrates, implicated a functioning chlorocatechol pathway to be induced during growth.

In recent years numerous studies on the aerobic bacterial degradation of chlorobenzenes used as environmentally critical pesticides, solvents, odorizers, and building blocks in chemical syntheses have been performed (7, 21, 24, 32, 38, 41, 49, 52, 53, 60). Public and scientific concern about their release (2) has increased efforts to study detoxification processes through microbial degradation. These investigations identified new bacterial strains capable of degrading such persistent and recalcitrant compounds as carbon and energy sources, or at least of attacking by cometabolism (1, 50, 59), though many pollutants still resist microbial degradation and detoxification (12, 20). Most of the key enzymes of the degradative pathways as well as the corresponding genes and their regulatory elements have been investigated in detail (3, 41, 56–58, 61). These studies clearly demonstrated that degradation of chlorobenzenes is initiated by a constitutively expressed (chloro-) benzene dioxygenase and a dihydrodiol dehydrogenase forming chlorocatechols which then are degraded by an inducible chlorocatechol pathway (41) to Krebs cycle intermediates (14). Studies on the degradation of all haloaromatics mineralized via chlorocatechols revealed that effective degradation proceeded only through those chlorocatechols with a free carbon at position 4 or 5 of the aromatic ring. Chlorocatechols with substitutions at both positions are strong inhibitors of (chloro-) catechol 1,2-dioxygenase ([Cl]C12O) (5) and probably are responsible for the observation that 1,2,3-trichlorobenzene (1,2,3-TCB) is only metabolized by cooxidation (11, 15, 24). Other isomeric chlorobenzenes, such as 1,3,5-TCB, 1,2,3,5-tetrachlorobenzene (1,2,3,5-TeCB), and penta- and hexachlorobenzene, resist attack by bacterial dioxygenases (31), presumably due to steric hindrance. To our knowledge this is the first time that aerobic mineralization of 1,2,3,4-TeCB by a pure culture has been demonstrated.

MATERIALS AND METHODS

Isolation and growth of bacteria.

A previously enriched mixed culture from samples of contaminated subsurface soil of a decommissioned pesticide production plant (11) and capable of growth on 1,2,3,4-TeCB as a sole source of carbon and energy served as the starting material for the isolation of a pure culture. Colonies grown on plates were isolated and reinoculated into mineral salts medium. Single colonies and defined consortia were tested for their ability to utilize 1,2,3,4-TeCB. The mineral salts medium used in this study consisted of 4.1 g of Na₂HPO₄ · 2H₂O, 0.4 g of KH₂PO₄, 0.5 g of (NH₄)₂SO₄, 0.1 g of MgCl₂ · 6 H₂O, 50 mg of Ca(NO₃)₂ · 4H₂O, 20 mg of Fe(NH₄)₂-citrate, and 0.1 ml of a trace element solution per liter without EDTA (36). The initial pH was 7.7. Solid medium contained 15 g of Agar No. 1 (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) per liter. Growth curves and end point determinations were performed in chloride-free medium in which chloride salts had been replaced by sulfate salts. The phosphate concentration provided sufficient buffering capacity to offset the release of hydrochloric acid during growth on 1,2,3,4-TeCB, although the concentration of phosphate had been halved to avoid depression of the growth rate. Liquid cultures were grown at 30°C in Erlenmeyer flasks with baffles and fitted with Teflon-sealed screw caps on a rotary shaker at 150 rpm. Flasks (100 ml) were incubated on an overhead rotating mixer at 120 rpm. 1,2,3,4-TeCB and other solid substrates corresponding to a 5 mM concentration were directly supplied as crystals to liquid media, whereas volatile compounds were fed through the gas phase from a test tube inserted through the seal of the screw cap of the Erlenmeyer flask. For growth on agar plates, solid substrates were either supplied in the lids of petri dishes or, in the case of liquid volatile substrates, provided in the headspace. The substrate spectrum of the organism was tested on gradient plates in order to minimize toxic effects. Growth on and dechlorination of substrate were monitored as increasing turbidity and decreasing pH. The optical density at 600 nm (OD₆₀₀) was determined on a spectrophotometer (model DU-70; Beckman Instruments Inc., Berkeley, Calif.). For quantitative determination of bacteria, the cells were counted by direct microscopic inspection. For enumeration of living bacterial cells (CFU), 10-μl aliquots were plated on solid Luria-Bertani (LB) medium after appropriate dilution and counted.

Identification of the bacterium.

The pure culture was characterized by means of substrate specificities by using the BIOLOG (Hayward, Calif.) test system and by 16S rRNA gene analysis (13, 66). Additionally, the organism was classified on the basis of Bergey’s Manual of Systematic Bacteriology (33) and by standard laboratory procedures (39).

Chloride release.

Chloride ion concentrations were determined at 20°C with a chloride-sensitive electrode (type U 402-M6-S7/100; Ingold AG, Urdorf, Switzerland). Standards were made up in the mineral salts medium described above in which chloride salts had been replaced by sulfate.

Oxygen uptake determinations.

The determination of oxygen uptake rates was performed polarographically at 25°C by using a Clark-type electrode (37). Assays were performed with washed cell suspensions suspended in 50 mM phosphate buffer, pH 7.4, to an OD₆₀₀ of 0.75 in a volume of 1 ml. Stock solutions of substrates were made up in dimethyl sulfoxide (100 mM). Final concentrations in the assay were always 1 mM except for (halo-) phenolic compounds such as catechols, which were used at 0.1 mM concentrations. Uptake rates were corrected for endogenous oxygen consumption. The protein concentrations of cell suspensions were determined by the method of Schmidt et al. (47).

Preparation of washed cell suspensions and cell extracts.

Growing cells were freed of residual TeCB crystals by washing and filtration, harvested by centrifugation at 10,000 × g for 20 min at 4°C, washed twice with 50 mM Tris-HCl buffer (pH 7.5), and resuspended in 2 ml of 33 mM Tris-HCl buffer (pH 8.0). Crude cell extracts were prepared by two passages through a chilled French pressure cell (Aminco, Silver Spring, Md.) at 10,000 lb/in². The exudate was centrifuged at 100,000 × g for 30 min at 4°C, and the supernatant fluid was used for assays. Soluble protein concentrations were determined by the method of Bradford with bovine serum albumin as a standard (4).

Enzyme assays.

All enzyme assays were performed at 25°C with a Shimadzu spectrophotometer (model UV 2100; Shimadzu Corp., Kyoto, Japan). Specific activities are expressed as units per milligram of protein. One unit was defined as the amount of enzyme required for the conversion of 1 μmol of substrate per min. C120 (EC 1.13.11.1) activity was measured as described by Dorn and Knackmuss (8, 9). Catechol 2,3-dioxygenase (C23O) (EC 1.13.11.2) activity was assayed as described by Nozaki (30). Interfering C23O activity was eliminated by incubation in the presence of 0.01% (vol/vol) H₂O₂ for 10 min as described by Nakazawa and Yokota (28). Extinction coefficients for chloromuconates were reported by Dorn and Knackmuss (8, 9) and Sander et al. (41). Chloromuconate cycloisomerase (EC 5.5.1.7) was measured by modifications of published procedures (45, 46) determining the decrease in substrate absorbance at 260 nm. Reaction mixtures contained 33 mM Tris-HCl buffer (pH 8), 2 mM MnCl₂, and a 0.1 mM substrate concentration. The cycloisomerase activity of RW71 was measured after total conversion of the corresponding catechol. Dienelactone hydrolase (EC 3.1.1.45) activity was assayed as described by Schlömann et al. (44). The extinction coefficient used for the cis isomer was 17,000 M⁻¹ cm⁻¹. For the estimation of 2,3,5-trichlorodienelactone, the ɛ₂₈₀ was determined to be 10,200 M⁻¹ cm⁻¹. The level of conversion of 2,3,5-trichlorodienelactone in the absence of crude cell extract was subtracted from the enzyme reaction. Maleylacetate reductase (EC 1.3.1.32) was measured spectrophotometrically by the decrease of the cofactor NADH at 340 nm. The standard assay mixture contained 50 mM Tris-HCl buffer (pH 7.5), 0.2 mM 2,3,5-trichloromaleylacetate, 20 μl of crude extract (65 μg of protein), and 0.2 mM NADH. The degree of nonspecific NADH oxidation by crude extract was subtracted. The reaction was started by addition of NADH after total conversion of 2,3,5-trichlorodienelactone and monitored at 280 nm as well as by high-performance liquid chromatography (HPLC) analysis. The activities represent the means of at least two independently performed experiments.

For turnover experiments, cells of RW71 were pregrown in LB medium or 1,2,3,4-TeCB, harvested in the linear growth phase, washed and pelleted twice (10 min, 4°C, 10,000 × g), resuspended in phosphate buffer (54 mM, pH 7.4), adjusted to an OD₆₀₀ of 1.0 to 5.0, and incubated with 1,2,3,4-TeCB corresponding to a 3 mM concentration. Conversion experiments with crude extracts of bacteria grown on 1,2,3,4-TeCB involved (i) 4 ml of Tris-HCl buffer (33 mM; pH 8.0) containing 0.1 mM tetrachlorocatechol, 2 to 5 mM EDTA, and 750 μl of crude extract (665 μg of protein); or (ii) 0.95 ml of Tris-HCl buffer (33 mM; pH 8.0) or histidine-HCl buffer (10 mM; pH 6.5) buffer containing 2,3,5-trichlorodienelactone (0.1 mM) and 50 μl of crude extract (28.5 μg of protein). Overlay spectra were monitored with a spectrophotometer (model UV 2100; Shimadzu Corp.) in the range from 200 to 400 nm.

Analytical methods.

Detection and identification of metabolites produced by growing or resting cells or cell extracts were performed by reversed-phase HPLC on a Shimadzu LC-10AD liquid chromatograph equipped with a DGU-3A degasser, an SPD-M10A diode array detector, and a FCV-10AL solvent mixer on an SC 125-by-4.6-cm Lichrospher 100 RP8, 5.0 μm column (Bischoff, Leonberg, Germany). The aqueous solvent systems (flow rate, 1.0 ml/min) contained 0.1% (wt/vol) ortho-phosphoric acid and 31 to 67% (vol/vol) methanol. Metabolites were identified by comparing their retention times and in situ UV spectra from 200 to 400 nm with those of known standards. The net elution volumes of metabolites were as follows (63% methanol): for tetrachloro-cis,cis-muconic acid, 1.4 min; for 2,3,5-trichlorodienelactone, 1.6 min; for 2,3,5-trichloromaleylacetic acid, 2.9 min; and for tetrachlorocatechol, 9.4 min.

Gas chromatography-mass spectrometry (GC-MS) studies were carried out on a GC (GC-17A; Shimadzu) equipped with an XTI-5 column from Resteck (Bellefonte, Pa.) coupled to the QP-5000 quadrupole mass spectrometer operating in the electron impact mode at 70 eV with an ion source temperature of 320°C. Helium was used as a carrier gas at a flow rate of 1.0 ml/min. The oven temperature was maintained at 60°C for 2 min and then increased to 150°C at a rate of 20°C/min within 20 min, followed by an increase to 320°C at a rate of 6°C/min. Samples (1.0 μl) were injected into the GC, which was operating in the splitless mode with an injector temperature of 270°C.

Isolation and derivatization of metabolites.

Culture supernatants or enzyme assay mixtures were extracted three times with equal volumes of ice-cold ethyl acetate at a neutral pH for the recovery of nondissociated compounds and at pH 1.5 for dissociated metabolites. The organic phase was dried with anhydrous magnesium or sodium sulfate, and after evaporation of the solvent, crystalline residues were directly derivatized with a freshly prepared solution of diazomethane in diethyl ether until the yellow color of the reagent remained stable. The derivatized extracts were directly analyzed by GC-MS without further purification.

Chemicals.

Benzene and substituted benzenes as well as tetrachlorocatechol were from Aldrich (Steinheim, Germany), Fluka (Buchs, Switzerland), Merck (Darmstadt, Germany), and Riedel-de Haën (Seelze, Germany). 3-Chloro-, 4-chloro-, 3,6-dichloro-, 4,5-dichloro-, and 3,4,6-trichlorocatechol were kindly provided by H.-A. Arfmann (Gesellschaft für Biotechnologische Forschung, Braunschweig, Germany). Chlorinated muconates were prepared in situ by incubating 3-chlorocatechol, 4-chlorocatechol, 3,5-dichlorocatechol, and 3,4,6-trichlorocatechol with crude extract from Escherichia coli BL21 (DE3) harboring either the chlorocatechol 1,2-dioxygenase gene tfdC (35) or the corresponding tetC gene from RW71 (unpublished data) in the pRSET6a (48) or peT9a (Novagen) expression system, respectively. A tetrachloromuconic acid standard was prepared by peracetic acid oxidation of tetrachlorocatechol, according to the method of Wacek and Fiedler (64). A 2,3,5-trichlorodienelactone standard was a generous gift from S. R. Kaschabek (Universität-Gesamthochschule Wuppertal, Wuppertal, Germany). 2,3,5-Trichloromaleylacetate was either prepared in situ by alkaline hydrolysis of the corresponding dienelactone or by enzymatic conversion. Biochemicals were obtained from Boehringer GmbH (Mannheim, Germany). All other chemicals were of the highest commercially available grade.

RESULTS

Isolation and identification of bacterial strain.

A mixed culture degrading 1,2,3,4-TeCB (11) maintained on solid medium and subcultured for approximately 3 years was used as the starting material for this study. After 2 further months of alternating subcultivation on liquid and solid media, a pure culture capable of growth on 1,2,3,4-TeCB as the sole source of carbon and energy was isolated. The purity was checked by microscopic examination and by streaking on LB plates. The strain was gram negative, strictly aerobic, and oxidase and catalase positive, and grew at 37°C as well as 4°C. Motility was observed neither on complex solid nor in liquid media. The organism was rod shaped on LB medium and about 1.2 to 1.5 μm by 0.6 μm in size. In the late linear growth phase in LB medium about 60% of the rod-shaped cells were connected to each other in pairs. In liquid mineral salts medium containing 1,2,3,4-TeCB, chains of bacteria of up to 10 spherical cells were formed. Colonies on solid selective medium were transparent and about 0.2 to 0.5 mm in diameter. The organism lost its ability to grow on 1,2,3,4-TeCB reproducibly after about 4 days of growth on LB broth, indicating that degradative genes were presumably encoded on a plasmid or another transmissible element and lost in the absence of selective pressure.

The organism was classified on the basis of 16S rRNA gene analysis by comparison of the obtained sequence data with known sequences in the databank of the Gesellschaft für Biotechnologische Forschung, Division of Microbiology (23, 54). DNA analysis of the 1,500 nucleotides revealed that the isolate is highly homologous (99.0%) to the type strain of Pseudomonas chlororaphis. Analysis of substrate specificity by the BIOLOG test system supported the identification of the genus. By this system, strain RW71 was identified as Pseudomonas corrugata (87%), since P. chlororaphis was not present as a reference.

Substrate spectrum and growth characteristics of P. chlororaphis RW71.

Besides 1,2,3,4-TeCB, 1,3- and 1,4-dichlorobenzene and 1,2,4-TCB served as sole carbon and energy sources, with the exception that 1,2,4-TCB was not used as a growth substrate in liquid culture when it was fed over the vapor phase or directly provided in the liquid medium. Supplied from the lid of the petri dish, 1,2,4-TCB allowed growth on mineral salts plates. The organism failed to grow with benzene, monochlorobenzene, 1,2-dichlorobenzene, 1,2,3-TCB, 1,3,5-TCB, 1,2,3,5-TeCB, 1,2,4,5-TeCB, and penta- and hexachlorobenzene. Catechol, 3-chloro, 4-chloro-, 3,6-dichloro-, 4,5-dichloro-, 3,4,6-trichloro-, tetrachloro-, and 3-fluorocatechol; salicylate; 4-chloro-, 5-chloro-, and 3,5-dichlorosalicylate; benzoate; 2-chloro-, 3-chloro-, and 4-chlorobenzoate; phenol; and all of the chlorinated phenols were not utilized even when carefully fed at a low concentration in order to avoid toxic effects. Glucose, fructose, fumarate, acetate, and succinate as well as chloroacetate and chlorosuccinate were used as sole sources of carbon and energy. The degradative potential for chlorobenzenes was completely lost during growth on nonselective (noninducing) substrates.

The ability of RW71 to grow on 1,2,3,4-TeCB is shown in Fig. 1. Chloride was released stoichiometrically, suggesting complete mineralization of 1,2,3,4-TeCB. This substrate was mineralized up to an amount corresponding to a total substrate concentration of 6 mM, with almost linear response of culture turbidity, protein content, chloride release, decrease in pH, and the number of live bacterial cells (CFU) (Fig. 2). Higher amounts of 1,2,3,4-TeCB were not completely degraded due to the too-low pH and limiting buffer strength. A pH below 6.5 was inhibitory for growth, although some 1,2,3,4-TeCB was converted further. End point determinations of growth parameters of parallel batches were performed after substrate crystals had been consumed. Cultures containing an amount corresponding to 6 to 10 mM 1,2,3,4-TeCB were worked up after a pH below 6.0 was reached. During growth on 1,2,3,4-TeCB doubling times depended on the size of the crystals. The very low solubility of 1,2,3,4-TeCB (0.4 mg/liter) in water and crystal size diversity resulted in growth rate deviations of up to 50%. These problems were overcome when the substrate was added as a fine mortar-ground suspension. The fastest generation time in batch culture was 21 h for strain RW71. The yield was 53 g of protein per mol of 1,2,3,4-TeCB.

FIG. 1.

Growth of P. chlororaphis RW71 on 1,2,3,4-TeCB. Parallel batch cultures initially contained 1,2,3,4-TeCB corresponding to a substrate concentration of 3 mM. The concentration of 1,2,3,4-TeCB was monitored by comparison to a standard curve by HPLC after addition of 4 volumes of methanol to the culture broth. Data represent mean values from two independent cultures.

FIG. 2.

End point determination of culture turbidity, protein content, chloride release, and changes of pH after growth of P. chlororaphis RW71 on increasing concentrations of 1,2,3,4-TeCB. Batches were stopped in parallel when no visible substrate was left or no increase in culture turbidity was detected because of inhibition by a pH that was too low.

Oxygen uptake rates and enzyme activities.

Further studies of the metabolic pathway for 1,2,3,4-TeCB degradation by P. chlororaphis RW71 were conducted to elucidate the degradative mechanism. Washed cell suspensions of P. chlororaphis RW71, pregrown on either 1,2,3,4-TeCB or succinate as a control, were tested for their oxidative potential towards chlorobenzenes and for chlorocatechols as possible pathway intermediates (Table 1). Significant oxygen uptake rates were found for 1,2,4-TCB and 1,2,3,4-TeCB only, but not for the other chlorinated benzenes tested. Catechol and most of its chlorinated derivatives were oxidized with high rates, with the exception of 4,5-dichlorocatechol, which was oxidized at a relatively low rate. No detectable uptake of oxygen was found with tetrachlorocatechol by 1,2,3,4-TeCB-grown cell suspensions. Specific uptake rates at least twofold higher than that for catechol were determined for 3-chloro-, 3,5-dichloro-, and 3,6-dichlorocatechol, whereas the activities for 4-chloro- and 3,4-dichlorocatechol were lower than that for catechol. Cells exhibited the same oxygen uptake rate with 3,4,6-trichlorocatechol as with catechol. Phenol and chlorinated phenols were not oxidized (data not shown). Succinate-grown cells of strain RW71 were not induced for the oxidation of chlorocatechols and chlorinated benzenes.

TABLE 1.

Specific rates of oxygen uptake with washed cells of P. chlororaphis RW71

Assay substrate	Specific oxygen uptake ratea after growth of RW71 on:
	1,2,3,4-TeCB	Succinate
Catechol	32.6	<0.1
3-Chlorocatechol	82.3	<0.1
4-Chlorocatechol	29.1	<0.1
3,4-Dichlorocatechol	23.8	<0.1
3,5-Dichlorocatechol	115.0	<0.1
3,6-Dichlorocatechol	67.5	<0.1
4,5-Dichlorocatechol	1.3	<0.1
3,4,6-Trichlorocatechol	31.9	<0.1
Tetrachlorocatechol	<0.1	<0.1
Benzene	<0.1	<0.1
Chlorobenzene	<0.1	<0.1
1,2-Dichlorobenzene	<0.1	<0.1
1,3-Dichlorobenzene	<0.1	<0.1
1,4-Dichlorobenzene	<0.1	<0.1
1,2,3-TCB	<0.1	<0.1
1,2,4-TCB	4.4	<0.1
1,3,5-TCB	<0.1	<0.1
1,2,3,4-TeCB	8.6	<0.1
1,2,4,5-TeCB	<0.1	<0.1
1,2,3,5-TeCB	<0.1	<0.1

^aResults represent means of at least two independently performed experiments. The oxygen uptake rates are expressed as specific activities (nanomoles of O₂ consumption per minute per milligram of protein) and are corrected for endogenous respiration. Chlorobenzenes and chlorocatechols were equivalent to concentrations of 1 and 0.1 mM, respectively. 

The degradative potential was further investigated by measuring enzyme activities in crude extracts from cells pregrown on the same substrates as for oxygen uptake rate determinations (Table 2). C23O activity was detected neither in cells grown on selective medium nor on succinate medium used as the control. (Cl)C12O activities were not present in extracts of cells grown with succinate, showing that this enzyme was not induced by this substrate. The ClC12O activity in strain RW71 grown on 1,2,3,4-TeCB displayed a ratio of 1:2.6 for catechol and 3-chlorocatechol. The activities for 4-chlorocatechol and 3,5-dichlorocatechol were determined to be 1.5- and 1.7-fold higher than those for catechol, respectively. The activity for ortho cleavage of 4,5-dichlorocatechol was 13% of that for catechol, and that for the oxidation of tetrachlorocatechol was still detectable. The specific activities varied in different batches depending on the growth phase, not affecting the ratio of specific activity between catechol and chlorinated catechols. The K_m values were 5.0 μM for catechol, 0.97 μM for 3-chlorocatechol, and 1.17 μM for 4-chlorocatechol, thus showing a significantly higher apparent affinity for chlorocatechols. Preliminary investigations on the stability of the ClC120 in crude cell extracts revealed a decrease of 50% activity in 28 h at 0°C. The cycloisomerase-specific activity of RW71 after growth on 1,2,3,4-TeCB was determined to be 0.025 μmol/min/mg of protein for 2,4-dichloro-cis,cis-muconate and 0.058 μmol/min/mg of protein for 2,3,5-trichloro-cis,cis-muconate. Activities for the cycloisomerization of cis,cis-muconate, 2-chloro-cis,cis-muconate, and 3-chloro-cis,cis-muconate could not be detected. Maleylacetate reductase activity from RW71 was determined to be 0.076 μmol/min/mg of protein for 2,3,5-trichloromaleylacetic acid.

TABLE 2.

Specific activity of ClC12O in cell extracts of P. chlororaphis RW71

Assay substrate	Sp acta after growth of RW71 on:
	1,2,3,4-TeCB	Succinate
Catechol	0.22 (7.66)	<0.01
3-Chlorocatechol	0.58	<0.01
4-Chlorocatechol	0.35	<0.01
3,4-Dichlorocatechol	0.12	<0.01
3,5-Dichlorocatechol	0.38	<0.01
3,6-Dichlorocatechol	0.20	<0.01
4,5-Dichlorocatechol	0.029b	<0.01
3,4,6-Trichlorocatechol	0.13	<0.01
Tetrachlorocatechol	—c (0.19)	<0.01

^aResults represent means of at least two independently performed experiments. Specific activities are expressed in units per milligram of protein. 

^bThe molar extinction coefficient of 3,4-dichloromuconic acid was not determined but was assumed to be similar to values of the other dichlorinated muconic acids. Activities in parentheses represent the ΔE per minute per milligram of protein. 

^cThe molar extinction coefficient (ɛ) of tetrachloromuconic acid is not known.  

Preparation and characterization of metabolites.

Approximately every fourth to fifth culture developed a transient violet color upon inoculation. The same color could also be obtained when Fe³⁺ ions were added to an aqueous solution of tetrachlorocatechol. Culture supernatants were analyzed by HPLC and, after extraction, by GC-MS. These analyses confirmed the presence of tetrachlorocatechol, with maximum concentrations of 7.5 μM. Tetrachloro-1,2-dihydro-1,2-dihydroxycyclohexa-3,5-diene (TeCB dihydrodiol) was not detected. Another metabolite was present in small amounts in the derivatized acidic extract. Comparison of its methyl ester with an authentic standard allowed identification of this compound as the cycloisomerization product of the corresponding tetrachloromuconic acid, 2,3,5-trichloro-4-carboxymethylene-but-2-en-4-olide (2,3,5-trichlorodienelactone methyl ester [Fig. 3B]). The molecular peak at m/z = 256 features the isotopic pattern of a trichlorinated compound. The ion at m/z = 225 can be explained by loss of one methoxy group showing the same isotopic pattern. The base peak at m/z = 59 indicates the carboxymethoxy group.

FIG. 3.

70-eV mass spectra showing the relative (Rel.) abundances of dimethyl 3,4,5,6-tetrachloromuconate (A) and methyl 2,3,5-trichloro-4-carboxymethylenebut-2-en-4-olide (2,3,5-trichlorodienelactone monomethyl ester) (B).

Monitoring of the time-dependent transformation of tetrachlorocatechol with crude cell extract by in situ UV spectroscopy (Fig. 4) showed on-column cycloisomerization in the acid HPLC solvent (pH 1.5) of tetrachloromuconic acid, the first product of the enzymatic reaction. The resulting product of this chemical cyclization, 2,3,5-trichlorodienelactone, became visible as a peak with extreme tailing. Acidification to a pH of 1.5 prior to injection of the sample containing tetrachloromuconic acid resulted in one distinct peak of the lactone. Its in situ UV spectrum and retention volume were identical with that of the authentic 2,3,5-trichlorodienelactone standard. Rapid acidification to a pH below 1.0 and immediate extraction with ice-cold ethylacetate followed by direct methylation with diazomethane of the 100-fold scaled-up catechol 1,2-dioxygenase assay with tetrachlorocatechol as the substrate resulted in one single product. Structure elucidation by GC-MS provided indications for a dimethylated tetrachloromuconic acid; the mass spectrum of the diester is given in Fig. 3A. The molecular ion (m/z = 306) is of very low intensity (0.1%). Loss of one chlorine forms the base peak (m/z = 271); further characteristic peaks are observed at m/z = 275 and m/z = 249, due to loss of one methoxy and one carboxymethoxy group from the molecular ion. The high intensity of the carboxymethoxy group is in accordance with the expected intensity for polychlorinated dimethylated muconic acids (65). The isotopic pattern of the peak at m/z = 249 shows the ratio (78:100:48:10:0.8) of a tetrachlorinated compound; that of m/z = 271 shows the theoretical ratio (100:96.5:31.3:3.5) of a trichlorinated substance.

FIG. 4.

Conversion of tetrachlorocatechol by cell extract of P. chlororaphis RW71 to the corresponding tetrachloromuconic acid, which is detectable only as the cyclic product 2,3,5-trichlorodienelactone, due to the acidic conditions used for separation by HPLC.

Enzymatic conversion of chemically synthesized 2,3,5-trichlorodienelactone by dienelactone hydrolase activity from crude cell extract resulted in the stoichiometric formation of the corresponding 2,3,5-trichloromaleylacetic acid (monitored by HPLC) (data not shown). Spectral changes monitored in the UV (Fig. 5) were identical to those obtained during hydrolytical cleavage of the authentic standard of chlorinated dienelactone in the presence of 0.1 N NaOH. The specific conversion rate by the (chloro-) dienelactone hydrolase in cell extracts of strain RW71 grown on 1,2,3,4-TeCB was 388 μmol/min/mg of protein for 2,3,5-trichlorodienelactone. The specific activity against unchlorinated cis-dienelactone was significantly lower at 6.4 μmol/min/mg of protein. Chlorodienelactone hydrolase activity was not detected when cells were grown with succinate.

FIG. 5.

Spectral changes during metabolism of 2,3,5-trichlorodienelactone by cell extract of P. chlororaphis RW71. Spectra were monitored until no further changes were detectable. Chemical transformation of 2,3,5-trichlorodienelactone without cell extract did not result in the formation of 2,3,5-trichloromaleylacetic acid. The numbers 0 to 5 correspond to spectra taken in intervals of 1 min.

GC-MS analysis performed after acidic extraction and methylation of the 2,3,5-trichlorodienelactone biotransformation assay revealed a peak whose retention time was identical with that also present in the total ion chromatogram obtained from respective preparations of the above-mentioned supernatants of violet-colored cultures. The spectrum was interpreted as having been derived from the trimethylated 2,3,5-trichloro-4-hydroxymuconic acid (Fig. 6A). The base peak at m/z = 267 results from the loss of one chlorine atom (M⁺ = 302; <0.1%). The isotopic patterns of the fragment at m/z = 243 is in full accordance with the theoretical pattern of a trichlorinated compound and originates through loss of one carboxymethoxy group at m/z = 59, showing an intense signal. Fragment m/z = 212 indicates the loss of both one methoxy and one carboxymethoxy group from the molecular ion. Other metabolites were present in low amounts in the assay mixture: a mass spectrum with fragments at m/z = 252 indicates the loss of one chlorine; a signal at m/z = 217 is indicative of the elimination of two chlorines and at m/z = 165 for the split off of a methoxy and a carboxymethoxy group from the fragment at m/z = 252. The signal at m/z = 137 represents the loss of two carboxymethoxy groups. These features are indicative of trichloromaleylacetic acid dimethyl ester, the tautomeric form of the above 2,3,5-trichloro-4-hydroxymuconic acid (Fig. 6B). The intensity of the molecular ion peak was again <0.1%. Acidification of the above-mentioned reaction mixtures led to the known decarboxylation of 3-oxo acids through α-cleavage of the 2,3,5-trichloro-4-methoxymuconic acid, yielding another compound, the mass spectrum of which was interpreted as 2,3-dichloro-4-methoxybut-2-enoic acid methyl ester (Fig. 6C). The fragment at m/z = 196 is probably due to the release of chloromethylacetate upon α-cleavage. The base peak at m/z = 181 can be explained by an additional loss of one methyl group, and the signal at m/z = 165 can be explained by the elimination of the methoxy group from m/z = 196.

FIG. 6.

70-eV mass spectra showing the relative (Rel.) abundances of 2,3,5-trichloro-4-methoxymuconic acid dimethyl ester (A), the corresponding 2,3,5-trichloromaleylacetic acid dimethyl ester (B), 2,3-dichloro-4-methoxybut-2-enoic acid methyl ester (C), 3,5-dichloro-4-oxopentanoic acid methyl ester (D), and 2,4-dichloro-3-methoxyhexanedioic acid dimethyl ester (E).

Further investigations of the lower pathway with respect to the loss of halide from the preceeding 2,3,5-trichloromaleylacetate was performed by a 90-fold-scaled-up maleylacetate reductase experiment. After acidic extraction and derivatization three major peaks were detected by GC-MS. The spectrum of one of these compounds (Fig. 6D) was interpreted as the acid-catalyzed decarboxylation product of 2,4-dichloro-3-oxoadipate dimethyl ester, 3,5-dichloro-4-oxopentanoic acid methyl ester. Peaks at m/z = 167 and m/z = 140 resulted from loss of one methoxy and one carboxymethoxy group, respectively, from the molecular ion peak (M⁺ = 199); both isotopic patterns indicate a dichlorinated compound. The ion at m/z = 149 resulted from loss of a ·ClCH₂ group (m/z = 49). Further characteristic ions are m/z = 122 and m/z = 77, derived from cleavage of the molecular ion peak both revealing a ratio of 3:1, indicating monochlorinated compounds. The spectrum of the second compound (Fig. 6E) was interpreted as being derived from the trimethylated enol (M⁺ = 271) of 2,4-dichlorooxoadipate. The ion at m/z = 199 originated from loss of methyl acetate through α-cleavage of the molecular ion. Elimination of one methoxy and one carboxymethoxy group from the molecular ion explains the peak at m/z = 181, and the recorded relative intensities are those of a dichlorinated substance. Loss of both carboxymethoxy groups and the methoxy group resulted in a peak at m/z = 122. The peak at m/z = 87 resulted from loss of chlorine and the additional loss of the carboxymethoxy group from m/z = 181; the intensities are features of a monochlorinated compound. The mass spectrum of a third metabolite was identical to that of 2,3,5-trichloromaleylacetic acid dimethyl ester (Fig. 6B), due to residual 2,3,5-trichloromaleylacetic acid in the assay.

DISCUSSION

Aerobic degradation of the recalcitrant compound 1,2,3,4-TeCB by a pure culture is described here for the first time. The newly isolated P. chlororaphis strain RW71 exhibits a chlorocatechol pathway, revealing specific activities of the ClC12O superior to those previously described. Earlier publications showed ratios of specific activities of ClC12Os between 0.4:1 and 1.6:1 for 3-chlorocatechol towards catechol (5, 8, 9, 16, 26, 32, 41, 43, 49, 53). On the other hand, the narrow substrate spectrum of strain RW71 indicates a very high selectivity for only few chlorobenzenes. An oxidative potential for unchlorinated benzene is not present as was reported for other chlorobenzene degraders (41). Spiess et al. (53) have isolated a strain capable of mineralizing 1,4-dichlorobenzene but not benzene and monochlorobenzene or other dichloro- or trichlorobenzenes. Though the initial (chloro-) benzene dioxygenase of Xanthobacter flavus 14p1 attacked 1,3-dichlorobenzene and the enzymes of the chlorocatechol pathway were able to convert 3,5-dichlorocatechol efficiently, the strain could not grow on 1,3-dichlorobenzene (51). This feature is presumably caused by the restricted effector specificity of the transcriptional activator of the lower chlorocatechol pathway, but which, however, could not be triggered by 2,4-dichloromuconic acid. The inducer, in analogy to that described by McFall et al. (25), who identified 2-chloromuconic acid as one of the two possible inducers of the chlorocatechol pathway, here should be 2,5-dichloromuconic acid. Similar characteristics in terms of substrate consumption were exhibited by strain RW71, which was not able to grow on benzene or monochloro- and 1,2-dichlorobenzene but was able to grow on 1,3-dichloro- and 1,4-dichlorobenzene, and 1,2,4-TCB (only on solid medium) and with 1,2,3,4-TeCB in liquid culture and on solid medium.

The initial oxidative attack by the chlorobenzene dioxygenase is assumed to proceed via the dihydrodiol of 1,2,3,4-TeCB, which should yield tetrachlorocatechol upon enzymatic dehydrogenation and rearomatization. The bacterium Acidovorax sp. strain PS14 (previously Pseudomonas sp. strain PS14) was proposed to attack 1,2,4,5-TeCB by a dioxygenolytic reaction proceeding directly to 3,4,6-trichlorocatechol (41). The accumulation of chlorocatechols during growth on chlorobenzenes has been described before by other authors (49, 53, 62). Final proof of the dioxygenolytic dehalogenation by the broad-spectrum chlorobenzene dioxygenase from Burkholderia sp. strain PS12, a strain also able to mineralize 1,2,4,5-TeCB, was demonstrated by incorporation of both molecules of ¹⁸O₂ into the aromatic ring (3). Initial reductive dechlorination or dehydrodehalogenation by strain RW71, yielding 1,2,4-TCB or either 2,3,4- or 2,3,6-trichlorophenol as the possible intermediate, respectively, cannot be fully excluded. Neither tri- nor tetrachlorophenol was oxidized or detected as an intermediate, nor did strain RW71 dehalogenate 1,2,3,4-TeCB anaerobically. Theoretically, an initial monooxygenase reaction could lead to 2,3,4,5-tetrachlorophenol, which, followed by chlorophenol hydroxylase activity, would furnish tetrachlorocatechol.

Sander et al. (41) were able to produce dihydrodiols and chlorocatechols from the corresponding chlorobenzenes by the constitutively expressed initial dioxygenases and dihydrodiol dehydrogenases of cells pregrown on complex medium. However, our analogously performed experiments were not successful. We assume that the corresponding genes of strain RW71 are instable, since the capability for chlorobenzene degradation was repeatedly lost during growth in LB medium.

cis,cis-Muconic acid was found to be the inducer of the catBCA operon (34) and 2-chloromuconic acid was recently identified as the inducer for ClcR-mediated transcriptional activation of the clcABD(E) operon (25). In analogy, tetrachloromuconic acid probably functions as an inducer of the chlorocatechol pathway in strain RW71. But too-low intracellular concentrations could be responsible for the accumulation of tetrachlorocatechol. The latter compound could also chelate the central Fe³⁺ ion of ortho-cleaving enzymes (5), thereby diminishing the turnover rate.

In analogy to that described by McFall et al. (25), the expression of the tcbCDEF cluster is triggered by a LysR-type regulator, TcbR. The background activity of ClC12O in the absence of inducer was less than 5% of that detected during growth of Pseudomonas sp. strain P51 on 1,2,4-trichlorobenzene (61). This low level of expression of the ClC12O of strain RW71 was not detected in crude extracts of cells pregrown on succinate.

Conversion of tetrachlorocatechol by crude cell extracts of 1,2,4,5-TeCB-grown strain PS14 (40) revealed 2,3,5-trichloro-4-oxohex-2-enedioate (2,3,5-trichloromaleylacetate) as the dead-end metabolite. Binding to the enzyme’s active site as a dianion (5) or binding of the substrate to the active site on one deprotonated hydroxyl group (17, 42), 4,5-dichlorocatechol and higher chlorinated derivatives of this substitution pattern have been described as potent inhibitors of (chloro-) catechol 1,2-dioxygenase (5). Surprisingly, strain RW71 exhibits against 4,5-dichlorocatechol 13% of the specific activity measured for catechol, thereby showing a broader substrate spectrum than ever described for similar strains. The conversion of 0.1 mM tetrachlorocatechol with crude extract of RW71 was slow but proceeded at this relatively high substrate concentration. Dorn and Knackmuss (8) reported that the ClC12O of Pseudomonas sp. strain B13 exhibited K_i values of 0.12 and 0.3 μM for the inhibition by 3,4,5-trichlorocatechol and tetrachlorocatechol, respectively, of catechol. The V_max was 0.3% for 3,4,5-trichlorocatechol compared with catechol (100%) and was not detectable for tetrachlorocatechol. Sauret-Ignazi et al. (42) reported that tetrachlorocatechol did not irreversibly inhibit the C12O of Alcaligenes eutrophus CH34, although substrate concentrations above 4 μM should have been markedly inhibitory. With this enzyme the activity toward catechol was completely restored after removal of tetrachlorocatechol by ultrafiltration. Steric hindrance due to substitution of hydrogen atoms by chlorine and the strong lipophilic character of this substrate were expected to prevent the compound from being converted faster (42).

It was confirmed by GC-MS analysis that the product formed from tetrachlorocatechol was tetrachloromuconic acid. This compound could not be tested as a substrate in the cycloisomerase assay because of the slow turnover of tetrachlorocatechol by crude cell extracts and its high instability in chemical and enzymatic preparations. Interestingly, conversion was observed only for the di- and trichlorinated muconic acids tested, not for monochlorinated muconates or muconate itself. Slow or negligible transformation of cis,cis-muconate by chloromuconate cycloisomerases has been reported previously (22, 55, 63).

The fast conversion of 2,3,5-trichlorodienelactone by crude extract and the identification of the di- and trimethylated derivatives of trichloromaleylacetic acid as well as the corresponding tautomerized hydroxymuconic acid provided strong evidence that bacterial degradation of 1,2,3,4-TeCB follows the modified ortho cleavage pathway. The existence of chlorinated maleylacetates had already been confirmed by different authors through isolation of the corresponding chloroacetylacrylic acids (16, 40, 52, 53). Conversion of 2,3,5-trichloromaleylacetate, characterized as relatively instable (18), with purified maleylacetate reductase from Pseudomonas sp. strain B13 resulted in the consumption of only 1.4 mol of NADH per mol of substrate (18). According to published specificities of maleylacetate reductases (18), halide elimination takes place only at position 2 of low-chlorinated maleylacetic acid, with subsequent consumption of 2 mol of NADH (27). We identified methylated derivatives of 2,3,5-trichloromaleylacetate, its decarboxylation product and its enol, the latter being the form in which it is present at the physiological pH. These results provide clear evidence that 2,3,5-trichloromaleylacetate represents a metabolite of tetrachlorocatechol mineralization. Identification of the dehalogenated decarboxylation product, 3,5-dichloro-4-oxopentanoate, as well as that of the α-cleaved fragment of the trimethylated enol form of 2,4-dichlorooxoadipate, obtained after enzymatic transformation in the presence of NADH by maleylacetate reductase, provides further evidence that degradation in RW71 occurs through 2,4-dichloro-3-oxoadipate as the next intermediate in the pathway. Presumably, this intermediate will be channeled into the tricarboxylic acid pathway after thiolytic cleavage, yielding chloroacetate and chlorosuccinate (both were used as carbon sources) or their corresponding coenzyme A esters: mechanistically, the dihydrogenation-dehydrodehalogenation reactions proposed by Chapman (6) and Duxbury et al. (10) should yield, in analogy to the reaction proposed by Sander et al. (41) for the conversion of 2-chloro-4-oxohex-2-enedioate, chloride release before (6) or after (10) cleavage. 3-Oxoadipate:succinyl-coenzyme A transferase and thiolase reactions were proposed to catalyze the cleavage of 3-oxoadipate to tricarboxylic acid metabolites (19). We assume that corresponding reactions take place during further breakdown of halogenated analogues. Based on the above-mentioned results we propose the pathway shown in Fig. 7 for the degradation of 1,2,3,4-TeCB.

FIG. 7.

Proposed pathway for the degradation of 1,2,3,4-TeCB by P. chlororaphis RW71. TCA, tricarboxylic acid.

Concern about ongoing man-made pollution is stimulating efforts to investigate new productive bacterial strains and their biodegradative pathways for bioremediation purposes. The 1,2,3,4-TeCB degrader P. chlororaphis RW71 represents a novel and promising tool for the rational design of optimized microorganisms able to mineralize even highly recalcitrant haloaromatic pollutants. No productive pathway via tetrachlorocatechol is yet known, except for partial dechlorination with a consortium of anaerobic bacteria (29).