Initial Transformations in the Biodegradation of Benzothiazoles by Rhodococcus Isolates

Abstract

Benzothiazole-2-sulfonate (BTSO₃) is one of the side products occurring in 2-mercaptobenzothiazole (MBT) production wastewater. We are the first to isolate an axenic culture capable of BTSO₃ degradation. The isolate was identified as a Rhodococcus erythropolis strain and also degraded 2-hydroxybenzothiazole (OBT) and benzothiazole (BT), but not MBT, which was found to inhibit the biodegradation of OBT, BT, and BTSO₃. In anaerobic resting cell assays, BTSO₃ was transformed into OBT in stoichiometric amounts. Under aerobic conditions, OBT was observed as an intermediate in BT breakdown and an unknown compound transiently accumulated in several assays. This product was identified as a dihydroxybenzothiazole. Benzothiazole degradation pathways seem to converge into OBT, which is then transformed further into the dihydroxy derivative.

2-Mercaptobenzothiazole (MBT) and its derivatives are mainly used as rubber vulcanization accelerators. As a result of MBT manufacture, process waters originate which contain 2-hydroxybenzothiazole (OBT), benzothiazole (BT), and benzothiazole-2-sulfonate (BTSO₃), apart from MBT itself. A biological purification of such wastewaters is often considered to be problematic (1, 12, 13), probably due to the toxic effects ascribed to MBT (5, 6, 13–15). This might explain why only a few benzothiazole-degrading organisms are reported in the literature and why information on benzothiazole degradation pathways is scarce. Our own research led to the isolation of Rhodococcus rhodochrous OBT18, which is capable of OBT and BT degradation (4).

Recently, Gaja and Knapp (8) described a Rhodococcus strain PA, growing on BT as the sole source of carbon, nitrogen, and energy, and strain TA, growing on 2-aminobenzothiazole. For MBT, no degrading organisms have been isolated in pure culture so far, although many axenic bacterial soil and water isolates can transform MBT into the relatively stable methylthiobenzothiazole (7). As for BTSO₃, only mixed cultures which at high inoculum densities degraded this compound as the sole source of carbon and nitrogen have been obtained (11).

In this study we report for the first time on a pure culture, isolated on BTSO₃, and we propose the initial transformation steps in benzothiazole biodegradation.

MATERIALS AND METHODS

Materials.

Benzothiazoles were kindly provided by Bayer Antwerpen N.V.

Media.

Mineral medium MM1% was prepared as described earlier (4). In order to obtain a sulfate-free medium, sulfate salts were replaced with the respective chloride salts. For solid media, 18 g of agar was added per liter.

Isolation and identification of BTSO₃-degrading organisms.

Activated sludge samples were obtained from a full-scale purification plant treating benzothiazole-containing wastewater or from laboratory fed-batch reactors treating MBT-BT substrate combinations. A 10-fold dilution series of the sludge samples in 1% (wt/vol) NaCl was inoculated in MM1% amended with 50 mg of BTSO₃ liter⁻¹. The same procedure was repeated several times for the lowest dilution still showing BTSO₃ degradation. Stable BTSO₃-degrading cultures were obtained; these cultures were regularly transferred to fresh medium, using a 10% (vol/vol) inoculum. Finally, culture samples were plated on solid MM1% with 50 mg of BTSO₃ liter⁻¹. Colonies were selected, purified on the same medium, and after growth on nutrient agar, transferred to liquid MM1% with 50 mg of BTSO₃ liter⁻¹ to check for their BTSO₃ degradation capacity. Several BTSO₃-degrading isolates were then identified by standard Gram staining, catalase, and oxidase tests, by their BIOLOG metabolic profiles (Biolog, Inc., Hayward, California), and by their fatty acid methyl ester (FAME) profiles (Microbial ID, Inc., Newark, Delaware) (16).

Incubations were always done at 28°C for at least 1 week.

Maintenance of isolates.

Since the BTSO₃ degradation capacity was not lost after growth on nutrient agar, the isolates were routinely maintained on this medium. For the experiments with isolate BTSO₃1, cells were transferred to nutrient broth (NB) test tubes and subsequently to Erlenmeyer flasks. They were grown for 36 h at 28°C, centrifuged (30 min, 10,000 × g, 4°C), washed three times with salt solution (1% [wt/vol] NaCl), and diluted to an optical density at 595 nm (OD₅₉₅) of between 0.2 and 0.3. From this suspension, 2-ml amounts were used to inoculate 100 ml of appropriate experimental media. Strain OBT18, degrading OBT and BT (4), was maintained and cultured under the same conditions.

Growth of the isolates on benzothiazoles.

MM1% was amended with 25 mg of BT liter⁻¹ and 50 mg of MBT, OBT, or BTSO₃ liter⁻¹ alone or in combinations. After inoculation with strain BTSO₃1, samples were taken at regular intervals and scored for OD₅₉₅ and BT concentrations. Noninoculated blanks were provided, and all incubations were done on a linear shaker at 28°C.

Resting cell experiments.

Strain OBT18 or BTSO₃1 was grown in a final volume of 800 ml of NB in 2-liter Erlenmeyer flasks. After 24 h of growth, cells were harvested (30 min, 10,000 × g, 4°C) and resuspended to an OD₅₉₅ of 5 in MM1% with 1 mM concentrations of different benzothiazole substrates. As controls, a cell suspension in MM1% without any benzothiazoles and noninoculated blanks were included. In aerobic conditions, incubations were in shaken Erlenmeyer flasks. For anaerobic incubations, vials were closed with rubber stoppers and made anaerobic by flushing with N₂ for 20 min. At regular intervals, 1-ml samples were withdrawn, centrifuged for 5 min at 10,000 × g, and analyzed for remaining benzothiazole concentrations.

Isolation of an unknown intermediate compound.

For the isolation of the unknown metabolite, a resting cell experiment was performed in a similar way as described above but on a larger scale. Strain OBT18 cells, harvested from 3 liters of NB, were resuspended in 600 ml of MM1% amended with 1 mM OBT. After 4 h of aerobic incubation, cells were removed by centrifugation (30 min, 10,000 × g) and the supernatant was analyzed by high-pressure liquid chromatography (HPLC) to check for the disappearance of OBT and the presence of the intermediate compound. The supernatant was then acidified to pH 1 with concentrated HCl, and after the addition of 10 g of NaCl, a 50-ml aliquot was extracted four times with 25 ml of diethyl ether. The organic phases were combined and concentrated by evaporation in vacuo. Thin-layer chromatography (TLC) on silica gel plates (20 by 20 cm; granule diameter, 200 μm; Merck, Overijse, Belgium) with a 70:30 (vol/vol) ethyl acetate-chloroform solvent system showed the presence of four spots under UV light. After preparative TLC under similar conditions, the four separated bands were removed from the TLC plate. The adsorbed compounds were eluted with a 70:30 (vol/vol) ethyl acetate-acetonitrile mixture, and each fraction was then evaporated to dryness in vacuo.

Aliquots were dissolved in ethyl acetate prior to mass spectroscopy (Hewlett Packard 5989A mass spectrometer with a solid probe; chemical ionization in methane) or in methanol prior to reversed-phase HPLC analysis (Kromasil column, 10 cm long; stationary phase, C₁₈; mobile phase, water-methanol, 50:50; flow rate, 1 ml/min; Waters photodiode array detector 990). An aliquot of fraction 3 was also acetylated with 1 drop of acetic acid anhydride and 1 drop of pyridine. After incubation for 3 h at room temperature, the mass spectrum was recorded.

Analytical methods.

Benzothiazoles were determined by HPLC analysis by the method of De Vos et al. (3). Alternatively, HPLC analyses were performed in similar conditions on a Lichrosorb RP18 column with a mobile phase of 35:65 (vol/vol) acetonitrile-water to which the commercially available ion pair reagent Pic A (Fluka) was added.

Nitrate, nitrite, sulfate, sulfite, and sulfide were analyzed by ion chromatography (Dionex 2000i; HPIC-As3 column; eluent, 35 mM carbonate-bicarbonate solution).

RESULTS

Isolation and characterization of a BTSO₃-degrading isolate.

The starting material for the isolation of benzothiazole-degrading organisms, was activated sludge from a full-scale wastewater treatment plant and from lab-scale reactors fed with benzothiazoles. After several unsuccessful isolation attempts, dilution to extinction in mineral medium with 50 mg of BTSO₃ liter⁻¹ proved to be a suitable method for the isolation of BTSO₃-degrading organisms. Several axenic cultures were obtained, using BTSO₃ as the sole source of carbon, nitrogen, and energy. They were all gram-positive, oxidase-negative, and catalase-positive bacteria, and all had similar morphological characteristics. By using BIOLOG data (similarity index of >0.86) and FAME analysis (similarity index of >0.50), they were tentatively identified as R. erythropolis strains.

Strain BTSO₃1 was selected for further studies. To optimize the conditions for BTSO₃ degradation, the effects of several environmental parameters were investigated. It was found that changing the substrate concentrations between 50 and 600 mg of BTSO₃ liter⁻¹ did not influence the lag phase (typically 30 h) or the rate of BTSO₃ degradation. In addition, the growth rate on BTSO₃ and the final cell density were not significantly different at pH values between 6 and 8 or at salt concentrations between 1 and 3%. To determine the inorganic end products of BTSO₃ mineralization, strain BTSO₃1 was grown on mineral medium MM1%, amended with 600 mg of BTSO₃ liter⁻¹ but without any inorganic sulfur or nitrogen source. As can be seen from Fig. 1, growth and BTSO₃ disappearance were strongly correlated. Nitrogen and sulfur were found as ammonia, sulfite, and sulfate. Nitrate or nitrite was not detected.

FIG. 1.

Biodegradation of 600 mg of BTSO₃ liter⁻¹ (2.8 mM) by R. erythropolis BTSO₃1 in a sulfate- and ammonium-free mineral medium. Growth was measured as OD₅₉₅. Symbols: ⧫, BTSO₃; ▄, sulfite; +, sulfate; ∗, ammonium; □, growth expressed as OD₅₉₅.

Biotransformation of other benzothiazole substrates by R. erythropolis BTSO₃1.

Strain BTSO₃1 was inoculated in MM1% with 50-mg/liter concentrations of various benzothiazoles as the sole sources of carbon, nitrogen, and energy. Growth and concomitant substrate disappearance were observed on OBT, BT, and BTSO₃. However, the lag phase for BT degradation was generally longer than for OBT or BTSO₃ degradation, and in some experiments, no BT disappearance was observed at all. During incubations on MBT, the medium turned yellow, and later, a reddish precipitate was formed, although no MBT turnover could be seen by HPLC analysis.

When strain BTSO₃1 was grown on BT, OBT, and BTSO₃ together, the substrates were simultaneously degraded (Fig. 2). In contrast, the addition of MBT almost completely inhibited the degradation of the three other compounds.

FIG. 2.

Growth of strain BTSO₃1 on a mixture of benzothiazole substrates. The initial OBT, BT, and BTSO₃ concentrations were 0.33, 0.19, and 0.23 mM, respectively. Growth was measured as OD₅₉₅. Symbols: ▴, BT; ■, OBT; ⧫, BTSO₃; □, growth expressed as OD₅₉₅.

Benzothiazole biodegradation pathways.

The newly isolated strain BTSO₃1 degrades OBT, BT, and BTSO₃. Earlier we characterized the isolate OBT18, capable of BT and OBT degradation only (4). We now wanted to study and compare the biodegradation pathways of the different benzothiazole substrates in these two strains. To this end, either strain BTSO₃1 or OBT18 was grown on NB and used for resting cell experiments. For incubations with strain BTSO₃1 under aerobic conditions, the following observations were made. MBT concentrations did not change, although a yellow compound was formed in the medium. Incubations with BT led to the transient accumulation of OBT (determined in two different chromatographic systems with coelution of OBT standards), which reached a maximum concentration right before all BT had disappeared from the medium (Fig. 3). During incubations with OBT or BT, an unknown compound reached a maximal concentration when the original substrate was almost exhausted and was subsequently degraded as well (Fig. 4). Based on its retention time in the chromatographic system, this intermediate must be of a rather high polarity, like BTSO₃. In anaerobic resting cell assays, BT, MBT, and OBT concentrations remained constant throughout the test period, whereas BTSO₃ was transformed into OBT in nearly equimolar amounts (Fig. 5).

FIG. 3.

Aerobic transformation of 1 mM BT by resting cells of strain BTSO₃1. Similar results were obtained in three independent experiments.

FIG. 4.

Transient accumulation of an unknown intermediate during the aerobic transformation of 1 mM OBT by resting cells of strain BTSO₃1. The concentration of the intermediate was calculated by using the response factor of BTSO₃ in the chromatographic system. Similar results were obtained in three independent experiments.

FIG. 5.

Anaerobic transformation of 1 mM BTSO₃ by resting cells of strain BTSO₃1. Similar results were obtained in three independent experiments.

For strain OBT18, similar phenomena were observed, except that BTSO₃ was not degraded.

In order to identify the unknown intermediate accumulating during OBT and BT degradation, strain OBT18 was incubated on 600 ml of MM1% with 1 mM OBT. Extraction of the supernatant with ether yielded a yellow ether phase and a lighter brown aqueous phase. When the concentrated ether phase was analyzed by TLC in an ethylacetate-chloroform solvent system, four spots were visible with R_f values of 0.7, 0.56, 0.42, and 0. The four compounds were separated by preparative TLC and finally analyzed by mass spectroscopy. For the fractions corresponding to R_f values of 0.7 and 0, the mass spectra were not usable due to a lot of noise. Fraction 2 (R_f = 0.56) was the original substrate OBT itself. Fraction 3 (R_f = 0.42) apparently contained two substances, as could be seen from the ion abundance versus time course during mass spectroscopic analysis (not shown). In the mass spectrum of substance 1, the molecular ion MH⁺ had a mass/charge ratio of 152. Although OBT has a molecular weight (MW) of 151, product 1 cannot be OBT due to differences in chromatographic properties. OBT and the present compound had R_f values of 0.56 and 0.42, respectively, and were considered to be completely separated, as no tailing was observed. It is therefore assumed that substance 1 contained a labile bond that was immediately broken in the mass spectrometer. Consequently, only a fragment of substance 1 was detected. This fragment presumably has the structure of OBT, and the labile bond might be an S-O or N-O bond in the thiazole ring of OBT (2). The mass spectrum of substance 2 is presented in Fig. 6. The peaks with mass/charge ratios of 61 and 89 are solvent peaks and correspond to the fragment (CH₃COOH₂)⁺ and to the molecular ion of the solvent ethyl acetate (MW = 88), respectively. The peak with a mass/charge ratio of 168 corresponds to the molecular ion MH⁺ of substance 2, and the peaks with ratios of 196 and 208 correspond to (M+ethyl)⁺ and (M+allyl)⁺, where M combined with ethyl or allyl fragments, originating from the solvent ethylacetate and methane. Thus, the MW of substance 2 is 167 and differs from the MW of OBT by 16. It seems that substance 2 is a hydroxylated derivative of the substrate OBT. The additional oxygen atom is bound to C, and not to N or S, because the latter bonds are too weak to be visible in mass spectra (2). To confirm whether the structure of substance 2 is indeed 2,x-dihydroxybenzothiazole (diOHBT), an aliquot was acetylated and then subjected to mass spectroscopy. In the mass spectrum, two peaks with mass/charge ratios of 210 and 252 were present, implying that substance 2 has been acetylated either once or twice. Although the acetylation reaction was not yet complete, this proves that substance 2 contains two different hydroxyl functions.

FIG. 6.

Mass spectrum of substance 2 (see Results). Based on the mass of the molecular ion, a dihydroxybenzothiazole structure was proposed. After an acetylation reaction, substance 2 was acetylated twice, which confirms this structure.

Substance 2 and the unknown intermediate that accumulated during OBT and BTSO₃ degradation had identical retention times in HPLC runs with different solvent systems, and they had identical UV spectra. It is therefore assumed that substance 2 was the accumulated intermediate.

DISCUSSION

As opposed to BTSO₃ and MBT, OBT and BT have generally been considered to be easily biodegradable (1) and axenic cultures degrading these compounds are available. R. rhodochrous OBT18 was isolated on OBT (4) and Rhodococcus strain PA was isolated on BT (8). They both degrade BT and OBT, but not MBT. BTSO₃ breakdown was not tested for strain PA and was not observed for strain OBT18. Thus, it seemed that at least the OBT and BT degradation pathways are closely related. In view of a comprehensive study of benzothiazole-biodegradative pathways, we also wanted to include a MBT degrader and a BTSO₃ degrader in the biodegradation studies. Although MBT is degraded in full- and lab-scale activated sludge systems, mixed or pure cultures with stable degradation properties have not yet been obtained. However, after several unsuccessful attempts, we have now succeeded in the isolation of a BTSO₃ degrader. The isolate BTSO₃1 was identified as an R. erythropolis strain and is also capable of OBT and BT mineralization. These results strongly suggest that the BTSO₃ degradation pathway overlaps or converges with the OBT and BT breakdown route. Compared to strain OBT18, strain BTSO₃1 has more enzymes, enabling the transformation of BTSO₃ into an intermediate that is probably common for OBT, BT, and BTSO₃ degradation. During incubations on MBT, colored products were formed, although the MBT concentrations hardly changed. Similar observations were made for strains OBT18 (4) and PA (8), which do not degrade BTSO₃. These data probably indicate that the specific enzymes involved in OBT and BT degradation can attack MBT but that either MBT itself or a toxic intermediate product prevents its further degradation. This might also explain the observed inhibitory effect of MBT on OBT and BT degradation by strains OBT18 (4) and on OBT, BT, and BTSO₃ degradation by strain BTSO₃1.

BTSO₃ degradation by R. erythropolis BTSO₃1 was not affected by changes in substrate concentration, pH, or salinity within a certain range and should therefore not pose problems in the environment. When strain BTSO₃1 was grown in a sulfate- and ammonium-free medium amended with BTSO₃, the substrate was completely mineralized. BTSO₃ can therefore be used as the sole source of carbon, nitrogen, sulfur, and energy. Nitrogen was found as ammonium and sulfur mainly as sulfate (Fig. 1). However, since sulfite was temporarily detected during growth, the sulfonate- and/or thiazole ring sulfur is presumably released as sulfite and subsequently oxidized to sulfate. In the stationary phase of growth, the molar ratio of BTSO₃ degraded/ammonium produced/sulfate produced was 1:0.5:0.6. Compared to the theoretical maximum ratio of 1:1:2, 50% of the nitrogen and 70% of the sulfur are not accounted for. This might be due to assimilation or release of nitrogen and sulfur in an unknown form. In contrast, Mainprize et al. (11) measured the expected stoichiometric amounts of ammonium and sulfate after growth of a mixed culture on BTSO₃.

The metabolism of heterocyclic compounds very often involves ring hydroxylations, followed by ring cleavage. When attached to six-membered rings, five-membered rings are usually cleaved first, with or without initial hydroxylation (9). In addition, for sulfonated aromatic compounds, at least four desulfonation mechanisms have been elucidated (reference 10 and references therein): an NADH-linked dioxygenation, a monooxygenation, a hydrolytic desulfonation, and a meta ring cleavage-associated desulfonation. All of these involve (in)direct oxygenation. By analogy with the degradative pathways mentioned, it was assumed that the hydroxylated benzothiazole OBT might be an intermediate in BT and BTSO₃ degradation. From sequential induction experiments and respirometry, no decisive evidence was obtained in (dis)agreement with this assumption (results not shown). However, in resting cell assays, OBT accumulation was observed during incubations on BT under aerobic conditions (Fig. 3). Since BT and OBT were not degraded in anaerobic assays, the transformation of BT into OBT definitely requires molecular oxygen and is probably catalyzed by an oxidase or a monooxygenase enzyme. BTSO₃ was mineralized under aerobic conditions and was transformed into equimolar amounts of OBT during anaerobic assays (Fig. 5). OBT therefore is an intermediate in BTSO₃ degradation as well, and in this case the hydroxyl group probably originates from water. Moreover, it seems that the introduction of the hydroxyl function occurs simultaneously with the elimination of the sulfonate group. Otherwise, BT, rather than OBT, would have accumulated during anaerobic incubations on BTSO₃. Both on OBT and BT and in aerobic conditions only, the transient accumulation of an unknown intermediate could be detected. It was identified as a diOHBT (Fig. 6), but the exact position of the second hydroxyl function could not be determined by mass spectroscopy. So far, the presence of a diOHBT product with the second hydroxyl group on the benzene nucleus has been clearly demonstrated. Preliminary evidence indicates that a second intermediate with a labile N-O or S-O was also present. In the future, larger amounts of both products will have to be isolated and they will have to be separated to allow for infrared and nuclear magnetic resonance spectroscopic analyses and to determine their exact chemical structure.

In summary, Figure 7 shows the initial transformations during benzothiazole biodegradation, as described above. All pathways converge into OBT. OBT is hydroxylated into diOHBT, which is in turn further degraded. This scheme is consistent with the observation that OBT degradation by strain BTSO₃1 never posed problems, whereas BT and BTSO₃ degradation sometimes did not take place. This might have been due to the malfunction of one or more initial enzymes operational in BT and BTSO₃ transformation into OBT or to a low level of expression or the loss of genes encoding for these enzymes. The biodegradation scheme also confirms our earlier assumption that R. erythropolis BTSO₃1 has more genetic information than R. rhodochrous OBT18 does. Except for the BTSO₃ branch, the whole degradation scheme proposed for strain BTSO₃1, also applies for OBT18, but it remains to be investigated whether identical enzymes are involved in both rhodococci.

FIG. 7.

Initial transformations in benzothiazole biodegradation by R. erythropolis BTSO₃1, as derived from the experimental results.

For the most problematic compound, MBT, no degradation occurs with the presented enzymatic potential. Whether this is due to high substrate specificities of the enzymes involved is hard to tell, but since OBT, BT, and BTSO₃ degradation were inhibited by MBT, it is more likely that MBT itself or a transformation product inhibits a common enzyme in the reaction sequence.