Previous PageTable Of ContentsNext Page

3 Results and Discussion

In the initial studies using inoculum from the Holston AAP, approximately 200 _M TNT was rapidly depleted in the serum bottles and was no longer detected after 13 days. While there was no apparent lag in methane production in TNT unamended controls, methane production in TNT amended bottles was inhibited until TNT and the subsequent diaminonitrotoluene isomers had been depleted. Because of the large amount of background methane production, it was difficult to evaluate differences in methane production between the unamended controls and experimental bottles. The bottles were allowed to incubate until methane production had slowed significantly. The headspace of the bottles was flushed and replaced with N2:CO2 before respiking with 100 _M TNT. TNT was rapidly depleted and not detected after the first day of incubation (Figure 1)._ More than 70 percent of the beginning TNT concentration was accounted for as the monoaminodinitrotoluene and diaminonitrotoluene isomers (Figure 1). The ortho nitro group of TNT was preferentially reduced relative to the para nitro group. For example, 2-amino-4,6-dinitrotoluene was preferentially formed relative to the 4-amino-2,6-dinitrotoluene isomer. Throughout the incubation period, 2A46DNT accounted for more than 80 percent of the two monoaminodinitrotoluene isomers (Figure 1). This is in contrast to reports observing a preferential reduction of TNT at the para nitro group (Pasti-Grigsby et al. 1996; McCormick, Feeherry, and Levinson 1976; Fiorella and Spain 1997). Further reduction of the monoaminonitrotoluene isomers led to the formation of 2,4-diamino-6-nitrotoluene and 2,6-diamino-4-nitrotoluene (Figure 1). Interest-ingly, we observed the formation of 2,6-diamino-4-nitrotoluene. Observations of this diaminonitrotoluene isomer are infrequent (Price, Brannon, and Hayes 1995), however, we consistently observed its formation in numerous test runs with sludge obtained from both the municipal and industrial wastewater treatment plants. The diaminonitrotoluenes were not persistent. Their fate was unknown, but we suspected triaminotoluene as a likely product based on previous studies demonstrating its formation (Funk et al. 1993; McCormick, Feeherry, and Levinson 1976; and Preuss, Fimpel, and Diekert 1993).

A transitory peak was observed in HPLC chromatograms, which we tentatively identified as triaminotoluene. Later analyses of archived samples by HPLC equipped with a photodiode array detector demonstrated it was not triaminotoluene. Analyses of the serum bottles, however, did demonstrate the presence of triaminotoluene in low concentrations. This trace amount of TAT could not possibly account for the approximately 350 _M TNT added to these bottles. We thought the TNT reduction products may have become bound to the organic matter. The samples were hydrolyzed using an acid-base hydrolysis procedure (Thorne and Leggett 1997). No TNT or other reduction products were observed after hydrolysis (data not shown), suggesting that binding to organic matter was not responsible for their disappearance.

TNT was only partially transformed in sterile controls. At the end of 24 days, all the TNT was accounted for as the following (expressed as a percentage of the initial TNT concentration): TNT, 26 percent; 2A46DNT, 52 percent; 4A26DNT, 14 percent; 24DA6NT, 4 percent; 26DA4NT, 4 percent. The ortho nitro group of TNT was preferentially reduced rather than the para group, similar to that observed in live incubations. We were slightly surprised at the formation of the diaminonitrotoluene isomers in the sterile controls, but this has been observed by others as well (Krumholz et al. 1997).

Although the fate of TNT was unknown, we suspected the diaminonitrotoluene isomers converged to triaminotoluene, based in part because it was detected at the end of the experiment in low concentrations. TAT is known to be unstable (Preuss, Fimpel, and Diekert 1993; Krumholz et al. 1997; Ederer, Lewis, and Crawford 1997) and its detection can be problematic because of this (Preuss, Fimpel, and Diekert 1993; Krumholz et al. 1997). To determine if TAT was a major intermediate during the anaerobic transformation of TNT, we repeated the experiment, but analyzed for TAT within 8 hours of taking the liquid samples. The biological activity (i.e., methane production rates) in the bottles had decreased substantially; therefore, glucose was added as a cosubstrate to ensure active methane production. Soon after the addition of TNT, a peak appeared in the HPLC chromatogram of the samples with a retention time identical to that of a TAT standard. The spectrum index plot of a TAT standard (Figure 2) was identical to that of the TNT degradation intermediate (Figure 3). The purity angle of the peak indicated it was a very pure peak (less than 1) with no other contaminants (Table 1). TAT increased to near stoichiometric concentrations (Figure 4). After 10 days of incubation, nearly 75 percent of the TNT could be accounted for as TAT. As TAT concentrations were decreasing near the end of the incubation, the spectrum index plot looked less like the authentic TAT standard (Figures 5 and 6). The purity angle also increased, indicating other compounds were coeluting with the TAT (Table 1). This suggested that TAT was being degraded further. No triaminotoluene was observed in cosubstrate unamended controls or sterile controls (data not shown).

The addition of TNT to the serum bottles was inhibitory to methane formation (Figure 7). Before respiking with TNT, methane production rates in the experimental bottles and unamended controls were similar. Methane formation stopped immediately upon respiking with TNT, while there was no decrease in methane production rates in the TNT unamended controls (Figure 7). In the bottles respiked with TNT, methane production resumed when the newly formed diaminonitrotoluenes were almost depleted. After 60 days of incubation, methane production in bottles amended with TNT was less than the unamended controls, suggesting there was no conversion of the ring carbons to CH4 and CO2. The inhibition of methane production by TNT was not surprising since nitroaromatic compounds are known to lyse methanogenic bacteria and inhibit methane formation in anaerobic sewage sludge (Gorontzy, Kuver, and Blotevogel 1993). TNT is also known to be mutagenic and toxic to microorganisms (Roberts, Ahmad, and Pendharkar 1996). We observed a similar phenomena with a RDX-degrading methanogenic enrichment culture growing on ethanol (Adrian and Sutherland, DRAFT). Methane production was inhibited upon the addition of RDX, but resumed when RDX was no longer detected in the liquid phase. The mechanism of methane inhibition appears to involve the competition of RDX-degrading microorganisms and methanogens for H2 produced during the metabolism of ethanol. There is only one other report we are aware of suggesting such a phenomenon. Boopathy and Kulpa reported TNT may substitute for sulfate as catabolic electron acceptors for Desulfovibrio spp. (Boopathy and Kulpa 1992).

Evaluation of the effect of substrates on TNT degradation showed that some cosubstrates enhanced TNT degradation rates relative to unamended controls (Table 2). For example, the addition of glucose resulted in a TNT degradation rate of 6.3 _M day-1, the highest rate observed (Table 2). Ethanol also enhanced the TNT degradation rate, resulting in a degradation rate of 4.2 _M day-1, a 1.9-fold increase over the cosubstrate unamended control. TNT degradation rates in bottles amended with acetate were similar to those in the unamended control, indicating acetate had no effect (Table 2). The influence of cosubstrates on degradation rates may be related to their ability to serve as H2 donors. Complex substrates are fermented by other members of a bacterial consortium, producing H2 as one of the products, which becomes available to the TNT degrading bacteria. TNT degradation rates are enhanced by the addition of cosubstrates, such as glucose and ethanol, which result in the production of H2 (Gibson and Sewell 1992; Gottschalk 1986). The metabolism of acetate does not result in the production of H2, therefore it does not enhance TNT degradation rates. Other researchers have reported this also, suggesting that H2 was the electron donor responsible for stimulating reductive dechlorination reactions (Gibson and Sewell 1992; Fennel and Gossett 1997).

These studies did not give any insight into the subsequent reduction of 2A46DNT. Reduction of the second nitro group could occur either at the ortho or para-nitro group, resulting in the formation of either 26DA4NT or 24DA6NT, respectively. To determine the fate of 2A46DNT, we used it to spike bottles. In glucose unamended bottles, 2A46DNT was reduced at both the ortho and para positions (Table 3). After 3 days, 94 percent of the starting substrate was recovered as a mixture of 2A46DNT, 24DA6NT, and 26DA4NT. 26DA4NT accounted for more than 65 percent of the two diaminonitrotoluene isomers, indicating the ortho nitro group was preferentially reduced. Interestingly, in glucose amended bottles, 24DA6NT accounted for more than 70 percent of the two diaminonitrotoluene isomers, indicating the para nitro group was preferentially reduced, a finding opposite that observed with the glucose unamended bottles (Table 3). In the sterile control, 100 percent of the starting material was recovered as a mixture of 2A46DNT, 24DA6NT, and 26DA4NT (Table 3). The DANT isomers were formed in almost identical amounts. The literature search preliminary to this study revealed no other reports demonstrating the formation of both diaminonitrotoluene isomers from the subsequent reduction of 2A46DNT.

TNT biodegradation proceeded in a stepwise reduction of the nitro groups to triaminotoluene (Figure 8). This biodegradation pathway has several notable differences from pathways published by other researchers. For example, we observed the formation of the 2,6-diamino-4-nitrotoluene isomer, a reduction product not typically reported by others (McCormick, Feeherry, and Levinson 1976; Krumholtz et al. 1997; Preuss, Fimpel, and Diekert 1993; Funk et al. 1993). We are aware of only one other report that noted small amounts of 26DA4NT being formed (Price, Brannon, and Hayes 1995). We also observed the formation of both diaminonitrotoluene isomers before complete reduction of the monoaminonitrotoluene isomers had occurred. Some studies have suggested the aminonitrotoluene must be completely reduced before reduction of the second nitro group occurs (Fiorella and Spain 1997; Haidour and Ramos 1996).

Figure 1. TNT biodegradation in serum bottles incubated under methanogenic conditions.

Figure 4. Formation of triaminotoluene during the anaerobic biodegradation of TNT in liquid cultures in serum bottles incubated under methanogenic conditions.
Figure 5. Spectrum index plot of TNT biodegradation intermediate after 17 days incubation. The spectrum index is from the apex and two inflection points at half peak height.
Figure 6. Spectrum index plot of TNT biodegradation intermediate after 30 days incubation. The spectrum index is from the apex and two inflection points at half peak height.
Figure 7. Inhibition of methane production by TNT.
Figure 8. Anaerobic biodegradation pathway for TNT transformation to triamino-toluene. The relative thickness of the arrows indicate the predominant routes of nitro group reduction observed.


Table 1. Purity angles for authentic triaminotoluene and a TNT biodegradation intermediate observed in serum bottles amended with TNT and glucose.

Table 2. Degradation rates for TNT in the presence of different cosubstrates.

Table 3. Substrate concentrations (percent) in serum bottles held under methanogenic conditions after spiking with 2A46DNT. The concentrations are normalized to the initial concentration of 2A46DNT added to the serum bottles. The final concentration was determined after 3 days incubation.

Previous PageTable Of ContentsNext Page