Appl Microbiol Biotechnol (2004) 65: 739–746 DOI 10.1007/s00253-004-1625-3
ENVI RON MENTA L BIOTECHNOLO GY
Jason T. Popesku . Ajay Singh . Jian-Shen Zhao . Jalal Hawari . Owen P. Ward
Metabolite production during transformation of 2,4,6-trinitrotoluene (TNT) by a mixed culture acclimated and maintained on crude oil-containing media Received: 29 October 2004 / Revised: 29 March 2004 / Accepted: 7 April 2004 / Published online: 5 May 2004 # Springer-Verlag 2004
Abstract Metabolites formed during 2,4,6-trinitrotoluene (TNT) removal by a mixed bacterial culture (acclimated and maintained on crude oil-containing medium and capable of high rates of TNT removal) were characterized. In resting cell experiments in the absence of glucose, 46.2 mg/l TNT were removed in 171 h (87.5% removal), with a combined total formation of 7.7 mg/l amino-4,6dinitrotoluene (ADNT) and 0.3 mg/l 4,4′-azoxytetranitrotoluene and 2′,4-azoxytetranitrotoluene, leaving 70% of the initial TNT unaccounted for. In the presence of glucose, resting cells removed 45.4 mg/l TNT in 49 h (95.5% removal), with 9.1 mg/l ADNT and 2.4 mg/l azoxy compounds being produced, leaving 70.3% of the TNT unaccounted for. Growing cells (glucose present) were capable of removing 44.2 mg/l TNT within 21 h (97.9% removal), with the concomitant formation of 1.8 mg/l ADNTs and 2.2 mg/l azoxy compounds. Denitrated TNT in the form of 2,6-dinitrotoluene was also produced in growing cells with a maximum amount of 1.31 mg/l after 28 h, followed by a slight decrease with time, leaving 88.5% of the initial TNT unaccounted for after 171 h. Radiolabeled 14C-TNT studies revealed 4.14% mineralization after an incubation period of 163 days with growing cells.
Introduction Thus far, a biological pathway for the complete mineralization of 2,4,6-trinitrotoluene (TNT) has not been described. In most cases, aerobic bacteria transform the TNT molecule by reducing one or two nitro groups to J. T. Popesku . A. Singh . O. P. Ward (*) Department of Biology, University of Waterloo, Waterloo, N2L 3G1, Ontario, Canada e-mail:
[email protected] Tel.: +1-519-8884567 Fax: +1-519-7460614 J.-S. Zhao . J. Hawari Biotechnology Research Institute, National Research Council, Montreal, H4P 2R2, Quebec, Canada
hydroxylamino or amino groups and produce different isomers of aminonitroaromatic compounds, which in turn usually accumulate in the culture medium without further metabolism (Montpas et al. 1997; Duque et al. 1993; French et al. 1998). Some bacterial strains have been found to grow on TNT as a sole nitrogen source (French et al. 1998), but attempts to culture bacterial strains on TNT as the sole carbon source were not successful. There are reports of the oxidative removal of the nitro group from p-nitrophenol as nitrite catalyzed by a monooxygenase enzyme (Spain et al. 1979) and the dioxygenase-catalyzed elimination of the nitro groups from an aromatic ring with the formation of the corresponding catechol and nitrite (Spanggord et al. 1991). In the presence of oxygen, several microorganisms are able to initiate the degradation of nitroaromatic compounds with the oxygenolytic removal of the nitro group, yielding nitrite and a hydroxylated aromatic compound (Meulenberg and de Bont 1995). The major metabolites produced by the aerobic biodegradation of TNT are 4- and 2-amino-4,6-dinitrotoluene (4- and 2ADNT; Hawari et al. 2000; Snellinx et al. 2002). We previously reported that a mixed culture (maintained by subculture in a basal salts medium supplemented with crude oil and other hydrocarbon sources for several years in cyclone fermenters; Ward et al. 2003) had the ability to rapidly transform TNT (Popesku et al. 2003). The rates of TNT transformation observed with this culture were higher than the highest rates obtained from cultures originating from and acclimated in TNT-containing environments (Montpas et al. 1997; Tharakan and Gordon 1999; Radtke et al. 2000; Wikström et al. 2000). Here, we explore patterns of TNT removal and metabolite formation by resting and growing cells to gain an insight into the metabolic processes involved.
Materials and methods A mixed microbial population isolated from petroleum hydrocarbon-contaminated soil and maintained on crude
740
oil, motor oil, or diesel in cyclone fermenters was used in the present investigation. About 50% of the volume of the reactor (1 l) was replaced with fresh medium and 5,000 mg/l of the appropriate petroleum source every week. For inoculum preparation, 5 ml of culture from each of the cyclones were combined and the supernatant after centrifugation was discarded at 4,000 rpm for 10 min. The pellet was washed twice and re-suspended in 0.1 M phosphate buffer (pH 7) to a volume of 15 ml. TNT (98% pure) was provided by Defense Research and Development Canada, Valcartier, Quebec. TNT was added as a 20,000 mg/l stock solution in acetone. Growing cell experiments were carried out with 5 ml of modified mineral salts medium (Van Hamme et al. 2000) containing 100 mg/l TNT in 25-ml Erlenmeyer flasks. The mineral salts medium contained per liter: 1.4 g KH2PO4, 2.7 g Na2HPO4·7H2O, 0.2 g MgSO4·7H2O, 0.1 g Na2CO3, 0.05 g CaCl2·2H2O, 0.005 g FeSO4, 1.0 g NH4Cl, and Fig. 1 Time-courses for TNT removal and metabolite formation
3.0 ml of trace metals solution. The trace metals solution contained per liter: 0.162 g FeCl3·6H2O, 0.0144 g ZnCl3·4H2O, 0.012 g CoCl2·6H2O, 0.012 g Na2MoO4·2H2O, 0.006 g CaCl2·2H2O, 1.9 g CuSO4·5H2O, 0.05 g H3BO4, and 35 ml HCl. Culture flasks were inoculated with a 2% inoculum and incubated at 30°C on an orbital shaker (Labline, Melrose Park, Ill.) set at 200 rpm. A HgCl2-poisoned control was also included in all the experiments to observe any abiotic transformation of TNT. For resting cell experiments, the mineral salts medium in the culture flasks was replaced with either 0.1 M sodium phosphate (pH 7.0), 0.1 M 2-(N-morpholino)ethane sulfonic acid (MES; pH 6.0), or tris(hydroxymethyl) aminomethane (pH 8.0) and inoculated with washed cells to achieve a final cell optical density at 600 nm (OD600) of 3.0. All experiments were carried out in duplicate, the variability between replicate samples within experiments
741
was less than 5%, and the variability in reproducibility between experiments was less than 10%. For TNT mineralization studies, uniformly ring-labeled [U-14C]-TNT [160,000 disintegrations/min (dpm)] was used in sealed 120-ml serum bottles under the same incubation conditions as were used for non-labeled TNT degradation. The [U-14C]-TNT (specific activity of 124 μCi mmol−1 1 Ci=37 Gbq) had a chemical purity of 99%, as described by Ampleman et al. (1995). A glass tube containing 1 ml of KOH solution was used to trap CO2, as described by Hodgson et al. (2000). A Tri-Carb 2100TR liquid scintillation analyzer (Packard Instrumental Company, Meriden, Conn.) was used to count radioactivity. The serum bottles were regularly flushed with air. The experiments were run in duplicate. The methods for TNT analysis and microbial growth determination were described by Popesku et al. (2003). The identification of metabolites was performed by comparing the UV spectrum produced from a 996photodiode array detector on the HPLC with authentic azoxy isomers provided by R.J. Spanggord (SRI International, Calif.): 4,4′-azoxy showed two UV peaks (238.2 nm, 323.3 nm), 2′,4-azoxy showed two UV peaks (243 nm, 323 nm), and 2,2′-azoxy showed one peak (276 nm).
Results Time-courses for TNT removal and metabolite formation, monitored for resting cells with and without glucose and for cells growing in glucose-supplemented mineral salts medium, are presented in Fig. 1 and Table 1. For resting cells without glucose, TNT removal was 53.0% and 87.5% after 21 h and 171 h, respectively. Concomitant with TNT removal, there was a gradual appearance of the 2-ADNT and 4-ADNT isomers (7.7 mg/ l total at 171 h), representing approximately 16.8% molar equivalent of TNT transformed. Nitrite concentration increased to a maximum of 2 mg/l after 49 h, which represents 19.3% molar equivalents of TNT, based on the removal of one nitrogen atom from the ring. However, no denitrated metabolite was detected. For resting cells with glucose, TNT removal was more rapid, with 95.5% removal after 49 h. ADNT isomer production at 171 h was 8.4 mg/l or 17.7% molar equivalent of TNT removed. An additional three peaks were observed, with retention times (RTs) of 11.0, 12.3, and 14.8 min. Their UV spectra were very similar, with a maximal peak at 233.5, 238.2 and 238.2 nm, respectively. The last one matches 4-N-acetamido-2,6-DNT perfectly. The compound with an RT of 12.3 min may be 2-Nacetamino-2,4-DNT, because the 2,2′-azoxy is produced in very small amounts. The identity of the compound with a wide peak and a RT of 11.0 min is unknown. These compounds were produced at low levels, all of which decreased after 49 h of incubation. Nitrite concentrations were lower (maximum 1.5 mg/l, TNT molar equivalent 15.5% at 21 h) than under no-glucose resting cell
conditions. No denitrated product was observed in glucose-supplemented resting cell experiments. A gradual depletion of glucose was observed (Table 1). Even though initial cell concentrations for growing cells were considerably lower than for resting cells, 98% of the TNT was removed within 21 h. The active period of growth was between 0 h and 28 h, with a concomitant depletion of glucose. Production of ADNT isomers at 171 h was 4.4% molar equivalents of TNT, much lower than was observed in resting cells. Nitrite accumulation under growing cell conditions (maximum 0.67 mg/l, TNT molar equivalent 7.3% at 5 h) was less than that observed with resting cells. The sole denitrated product from TNT degradation, 2,6dinitrotoluene (DNT), was observed in growing cells with a maximal concentration of 1.3 mg/l (TNT molar equivalent: 3.6%) observed after 24 h of incubation. The compound had a retention time of 22.1 min and was identified by its UV spectrum against a standard sample of 2,6-DNT. In addition to the ADNT metabolites, azoxytetranitrotoluenes were detected based on UV spectral analysis compared with standard spectral data. Concentrations of these compounds gradually increased and appeared to be dead-end products. Equal amounts of 2′,4-azoxy and 4,4′axozy were produced in resting cells without glucose, with a combined total of
