Environ. Sci. Technol. 2001, 35, 682-689
Signature Metabolites Attesting to the In Situ Attenuation of Alkylbenzenes in Anaerobic Environments MOSTAFA S. ELSHAHED, LISA M. GIEG,* MICHAEL J. MCINERNEY, AND JOSEPH M. SUFLITA Department of Botany and Microbiology, The University of Oklahoma, Norman, Oklahoma 73019
Accurate assessment of the fate of hydrocarbons spilt in aquifers is essential for gauging associated health and ecological risks. Regulatory pressure to actively remediate such contaminated ecosystems can be substantially diminished if solid evidence for in situ microbial destruction of pollutants is obtained. In laboratory incubations, sedimentassociated microorganisms from a gas condensatecontaminated aquifer anaerobically biodegraded toluene, ethylbenzene, xylene, and toluic acid isomers with stoichiometric amounts of sulfate consumed or methane produced. The activation of the alkylated aromatic contaminants involved conversion to their corresponding benzylsuccinic acid derivatives, a reaction known to occur for toluene and m-xylene decay, but one previously unrecognized for ethylbenzene, o- and p-xylene, and m-toluate metabolism. Benzylsuccinates were further biodegraded to toluates, phthalates, and benzoate. In laboratory incubations, these metabolites were transiently produced. Several of the metabolites were also detected in groundwater samples from an aquifer where alkylbenzene concentrations decreased over time, suggesting that anaerobic microbial metabolism of these contaminants also occurs in situ. Our studies confirm the utility of the aforementioned compounds as signature metabolites attesting to the natural attenuation of aromatic hydrocarbons in anaerobic environments.
Introduction Hydrocarbon mixtures are accidentally released in the environment during energy production, usage, and storage. Benzene, toluene, ethylbenzene, and the xylene isomers (BTEX) pose great environmental and regulatory concern due to their relatively high water solubility and toxicity (13). Mounting evidence shows that BTEX hydrocarbons can be degraded anaerobically under a variety of terminal electron-accepting conditions (4). The susceptibility of these substrates to microbial decay has led to greater acceptance of natural attenuation for the management of hydrocarbon pollutants instead of active intervention with technologies to increase the rate of biodegradation. However, there is growing concern that the regulatory acceptance of natural attenuation occurs too readily in the absence of credible scientific evidence (5). Reliable indicators, such as the * Corresponding author phone: (405)325-3771; fax: (405)325-7619; e-mail:
[email protected]. 682
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001
detection of unique metabolites formed in situ, would help provide a solid scientific foundation for regulatory decisions to use natural attenuation. However, the metabolic pathways for BTEX destruction by anaerobic microorganisms are still incompletely known, and thus the task of identifying signature metabolites is substantially more difficult. Laboratory studies to elucidate the biodegradation pathways and to identify prospective metabolites involved are required to substantiate the detection of the latter at contaminated sites. Early work with pure and mixed cultures suggested that methyl group oxidation or hydroxylation via benzyl alcohol and benzaldehyde led to the formation of benzoate [or benzoyl-coenzyme A (CoA-thioester)] as a central metabolic intermediate in anaerobic toluene decay (6, 7). Subsequent studies with pure nitrate- and sulfate-reducing bacteria indicated that methyl group oxidation does occur but proceeds via a novel activation reaction whereby the methyl group of the parent hydrocarbon is added to the double bond of fumarate to form benzylsuccinic acid (8-12). A similar mechanism in Azoarcus tolulyticus Tol-4 has been reported, wherein two molecules of a two-carbon fragment (presumably acetyl-CoA) are added stepwise to form the same metabolites via cinnamic acid (13). In Thauera aromatica, benzylsuccinate is subsequently converted to a CoA-thioester (14) and transformed to E-phenylitaconate (or its CoAthioester), which then undergoes modified β-oxidation yielding benzoate (or benzoyl-CoA) (15). Some toluenedegrading organisms co-metabolized xylene isomers in the presence of toluene to the corresponding toluates and to compounds tentatively identified as methylbenzylsuccinates; however, these were believed to be dead-end metabolites (8, 10, 16-19). In fact, few studies have probed the further anaerobic biodegradation of toluic acids despite their detection in aquifer environments contaminated with BTEX hydrocarbons (20-22). Recently, 3-methylbenzylsuccinate and m-toluate have been positively identified as transient intermediates in m-xylene degradation by the denitrifier, Azoarcus sp. strain T (23). Ethylbenzene metabolism by two denitrifying isolates (24-26) proceeds via a different mechanism involving the oxidation of the R-carbon to 1-phenylethanol and subsequent dehydrogenation of the latter to acetophenone. It was postulated that acetophenone is carboxylated to benzoyl acetate, which is subsequently cleaved to acetate and benzoate (or their CoA derivatives). Wilkes et al. (27) reported the partial oxidation of xylenes to methylbenzylsuccinates by a m-xylene-utilizing, sulfate-reducing bacterium and by analogy postulated the formation of ethylbenzylsuccinic or dimethylbenzylsuccinic acids from ethyltoluenes or trimethylbenzenes, respectively. However, definitive assignment of the chemical structure of the putative metabolites was not possible since authentic standards were not available. To date, a metabolite resulting from the four-carbon addition reaction to the R- or β-carbons of ethylbenzene has not been reported with either pure or mixed cultures. Many of the aforementioned laboratory studies thus suggest that addition reactions may be common for the anaerobic destruction of alkylbenzenes. Indeed, Beller et al. (28) proposed that alkylbenzene addition products, including benzylsuccinic acid and its methylated analogues, serve as unique indicators of anaerobic toluene and xylene degradation in contaminated environments. These researchers noted the need for a better understanding of the biodegradability of these metabolites and for determining whether they could be found in other hydrocarbon-contaminated aquifers in order to determine the widespread applicability of using such 10.1021/es001571u CCC: $20.00
2001 American Chemical Society Published on Web 01/11/2001
indicators for intrinsic bioremediation (28). Aside from one other study conducted by the same research group (29), there have been few reports on the finding of these anaerobic signature metabolites in contaminated groundwater environments. At a highly reduced gas condensate-impacted aquifer, we previously detected numerous putative bacterial metabolites of BTEX hydrocarbons in the groundwater, including toluic acids and a proposed in situ indicator of anaerobic alkylbenzene metabolism, namely, 3-methylbenzylsuccinic acid (22). In the contaminated portion of the aquifer where these metabolites were found, BTEX concentrations in the sediments decreased over the course of 4 yr at rates ranging from approximately 50 to 300 mg kg-1 yr-1 at different sampling locations. These BTEX losses correlated with the consumption of oxygen and sulfate and the production of Fe(II), sulfide, and methane relative to an uncontaminated area. Hydrogen measurements indicated that sulfate reduction and methanogenesis were the predominant electron-accepting processes in the contaminated region, which correlated with higher numbers of these microbial populations in the contaminated sediments. Furthermore, the anaerobic biodegradation of toluene, ethylbenzene, and each of the xylene isomers was observed in laboratory incubations using inoculum from this site. Collectively, these data argued that anaerobic biodegradation of alkylbenzenes was occurring in situ (22). The goal of the present study was to further investigate the pathways by which alkylbenzenes were biodegraded in this gas condensate-impacted aquifer. Toluic acids were included in this study given that little is known about the fate of these putative xylene metabolites under anaerobic conditions and their prevalence in the contaminated groundwater at the site (22). Our approach was to conduct laboratory biodegradability studies in order to identify metabolites of alkylbenzene decay by anaerobic microbial populations derived from the aquifer and to determine whether the same metabolites were present in the anaerobic, contaminated groundwater. We found that the microbial populations had the potential to anaerobically biodegrade such compounds and produced a variety of signature metabolites that minimally included alkylbenzylsuccinic acids that were, in turn, detected in the field. This integration of laboratory biodegradability studies with chemical analysis of field samples provided simple yet conclusive evidence for in situ anaerobic biodegradation of alkylated aromatic contaminants.
Materials and Methods Alkylbenzene- and toluate-degrading slurries were established under sulfate-reducing and methanogenic conditions using sediments collected from a gas condensate-contaminated aquifer where previous evidence (22) indicated that intrinsic bioremediation of BTEX hydrocarbons was occurring. Concentrations of BTEX hydrocarbons in the groundwater at this site averaged 20 mg/L but have been detected as high as 50 mg/L (22). In an anaerobic glovebox, 25 g of sediment and 55 mL of a defined mineral medium (30; containing 2 mM sulfate for sulfate-amended incubations) or contaminated groundwater from the site were added to 120-mL serum bottles, which were closed with butyl rubber stoppers or composite stoppers (comprised of a butyl rubber stopper fused with a Teflon-coated stopper) and crimped with aluminum seals. After preparation, headspaces were exchanged with 20% CO2 in N2. Hydrocarbons (2-5 µL, resulting in concentrations of approximately 200-500 µM) were added as pure liquids to the slurries, while toluates were added to initial concentrations of 300-750 µM from an anoxic stock solution. Triplicate incubations were prepared for each treatment, except for the heat-inactivated controls, which were prepared in duplicate. Controls in these experi-
ments included heat-inactivated incubations for each compound and substrate-unamended incubations for each electron-accepting condition. The experiments were conducted at room temperature (22-24 °C). Groundwater was sampled from five monitoring well locations within the contaminated aquifer (22) on eight occasions over the course of 4 yr and preserved at pH < 2 prior to extraction for putative metabolites. Extracted samples from uncontaminated wells downgradient and upgradient of the BTEX plume were used for comparison. Alkylbenzenes were quantified by headspace analysis on a gas chromatograph (GC) equipped with a flame ionization detector (FID) and a Carbograph VOC capillary column (30 m × 0.25 mm i.d., Alltech, Deerfield, IL) held at 150 °C. Toluate concentrations were determined by high-performance liquid chromatography using a reversed-phase C18 column (250 mm × 4.6 mm, 5 µm particle size; Alltech, Deerfield, IL), a mobile phase (1 mL/min) of methanol:water:phosphoric acid (60: 40:0.5), and a UV detector operated at 254 nm. Sulfate concentrations were measured by ion chromatograhy using an AS4A-SC column (4-mm particle size) and a bicarbonatebuffered mobile phase (Dionex, Sunnyvale, CA). Methane was analyzed by GC-FID equipped with a packed column heated to 60 °C (Poropak Q, 80/100; Supelco, Bellefonte, PA). Detection and quantification of metabolites in laboratory incubations or groundwater was done by GC-mass spectrometry (MS). Subsamples from cultures were usually acidified to pH < 2 with H2SO4 and extracted with ethyl acetate. In some cases, cultures were first treated with NaOH (pH > 12) for 30 min to cleave prospective CoA-thioester bonds in the metabolites, but this was found to be unnecessary since no new metabolites were detected with this treament. Concentrated organic extracts were reacted with N,O-bis(trimethylsilyl)trifluoroacetamide (Pierce Chemical Co., Rockford, IL) to form trimethylsilyl (TMS) derivatives. Components in the extracts were separated on a DB-5 capillary column (30 m × 0.25 mm i.d., J&W Scientific, Folsom, CA) using a GC equipped with a mass selective detector operated in the scan mode. The oven temperature was held at 40 °C for 2 min, raised at a rate of 4 °C/min to 240 °C, and then held at 240 °C for 8 min. Metabolite identifications were made by comparing GC retention times and mass spectral profiles with authentic standards. Because methyl- and m-carboxybenzylsuccinic acids were not commercially available, these putative addition metabolites were synthesized according to published methods (31-33). To quantify metabolites in cultures and groundwater samples, calibration curves were prepared from known quantities of authentic standards and analyzed by the GC-MS method described above.
Results Laboratory Incubations with Alkylbenzenes and Toluic Acids. In initial laboratory incubations, toluene was the only alkylbenzene degraded under methanogenic conditions, whereas toluene, ethylbenzene, and each of the xylene isomers were degraded under sulfate-reducing conditions by the aquifer microflora. Each of these hydrocarbons was removed in less than 100 d as compared with sterile controls. All toluic acid isomers were also degraded under sulfatereducing and methanogenic conditions within a similar time frame. After the initial observation of alkylbenzene or toluate loss, the slurries were repeatedly amended with substrate (and sulfate in sulfate-reducing incubations), which was further consumed by the populations. After 150 d of incubation, the stoichiometric ratio of substrate degraded to sulfate consumed or methane produced was evaluated (Table 1). In the absence of growth, stoichiometric predictions indicate that 4.5 mol of sulfate are consumed or that 4.5 mol of methane are produced per mol of toluene or toluic acid, VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
683
TABLE 1. Stoichiometric Ratios for Degradation of Alkylbenzenes and Toluic Acids under Sulfate-Reducing and Methanogenic Conditions
enrichmenta
µmol of substrate consumedb
toluene, SR o-xylene, SR m-xylene, SR p-xylene, SR ethylbenzene, SR toluene, M o-toluate, SR m-toluate, SR p-toluate, SR o-toluate, M m-toluate, M p-toluate, M
91 38 40 77 35 31 45 24 41 42 52 40
µmol of electron expected sulfate acceptor/ ratio of reduced or substrate acceptor/ methane ratio substrate producedc (mol/mol) (mol/mol)d 392 267 395 138 267 128 183 113 162 169 172 131
4.3 5.1 5.1 3.9 7.0 4.1 4.1 4.7 3.9 4.0 3.3 3.3
4.5 5.3 5.3 5.3 5.3 4.5 4.5 4.5 4.5 4.5 4.5 4.5
a Values are averages of triplicates. Sulfate-reducing incubations are marked SR while methanogenic incubations are marked M. b Values shown are corrected for the amount of substrate consumed relative to heat-inactivated controls. c Values shown are corrected for the amount of sulfate reduced or methane produced in substrate-unamended controls. d Values shown are based on theoretical equations describing complete substrate metabolism to CO2 or CH4, excluding biomass.
and 5.25 mol of sulfate are consumed per mol of ethylbenzene or xylene. Taking into account background levels of sulfate consumption in substrate-unamended controls and nonbiological losses in autoclaved controls, we found reasonably good agreement between theoretical and actual sulfate consumption or methane production given the amount of hydrocarbon or toluic acid consumed (Table 1). Thus, these alkylated aromatic degrading enrichments were extracted, derivatized, and analyzed for putative metabolites. Metabolites in Laboratory Incubations. A metabolite that had the same GC retention time and mass spectral profile as the TMS derivative of benzylsuccinic acid (Table 2) was detected in toluene-degrading enrichments under sulfatereducing and methanogenic conditions. Benzoate was also detected in several incubations but not E-phenylitaconate, cinnamate, phenylacetate, or other putative toluene intermediates (13). In analogous fashion, methylbenzylsuccinates were postulated to be metabolites in anaerobic xylene decay (8, 10, 19). In xylene-amended, sulfate-reducing enrichments, we detected metabolites eluting at different GC retention times that were characterized by similar mass spectral profiles suggestive of TMS-derivatized methylbenzylsuccinic acids, all having fragment ions at m/z 73, 105, 145, 159, 204, 235, 248, 351, and 366 (molecular ion) (Table 2). To verify this, we chemically synthesized each isomer and the GC retention times and mass spectral profiles of the biologically produced metabolites matched those of the authentic standards (Table 2). Thus, 2-, 3-, and 4-methylbenzylsuccinates were positively identified to be metabolites of o-, m-, and p-xylene degradation, respectively. Furthermore, toluic acid isomers were detected as metabolites in xylene-amended incubations, and their GC retention times and mass spectral profiles also matched those of authentic standards (Table 2). Benzoate, toluates, and addition products were detected in concentrations ranging from 0.8 to 3 µM when approximately 75-90% of the parent substrate had been consumed. In ethylbenzene-amended sulfate-reducing incubations, we detected benzoate (up to 1 µM) but were unable to detect 1-phenylethanol and acetophenone, which have been identified as ethylbenzene metabolites under denitrifying conditions (24-26). However, we found a metabolite that had essentially the same mass spectral features observed for the 684
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001
methylbenzylsuccinic acids but differed in the relative abundance of the various fragment ions (Figure 1, Table 2) and GC retention time (38.2 min). We hypothesized that an analogous addition reaction had occurred during the anaerobic decay of ethylbenzene in these enrichments. The GC retention time and mass spectral characteristics of the putative ethylbenzene metabolite matched those of the chemically synthesized standard of TMS-derivatized 3-phenyl1,2-butanedicarboxylic acid (Figure 1A,B). As was also observed for the methylbenzylsuccinates, this metabolite was only transiently produced (up to 1 µM) in that it was not detected at the beginning of the experiment nor after ethylbenzene was consumed to below detection limits. In m-toluate-degrading sulfidogenic enrichments and all toluate-degrading methanogenic enrichments, we detected a metabolite (0.7-1.4 µM) with a GC retention time and mass spectral profile corresponding to TMS-derivatized benzoic acid (Table 2). Metabolites matching the GC retention times and mass spectral profiles of TMS-derivatized chemical standards of phthalate, isophthalate, and terephthalate were detected in o-, m-, and p-toluate-degrading methanogenic enrichments, respectively. Phthalates were detected in the cultures in concentrations ranging from 0.8 to 2 µM when at least 75% of the parent toluate had been consumed. o-Carboxybenzaldehyde was detected in the o-toluateamended methanogenic incubation (Table 2). Isophthalate was also detected in m-toluate-degrading, sulfate-reducing enrichments along with a metabolite having the same GC retention time and mass spectral profile as a chemically synthesized standard of 3-carboxybenzylsuccinic acid (TMS derivative, Figure 1C,D). This latter compound is likely the product of the addition of the aryl methyl carbon of m-toluate over the double bond of fumarate. The addition products, toluic acids, and phthalic acids were not detected in enrichments without substrate nor were these compounds present in culture extracts from enrichments where the substrate had been consumed to below the level of detection. This suggested that these compounds were transient intermediates in anaerobic alkylbenzene and toluate decay. However, it has been postulated that benzylsuccinate derivatives may be co-metabolically produced as dead-end products (8, 10, 19). To test whether the addition products were susceptible to anaerobic decay, time-course experiments were performed with cultures incubated with m-xylene or m-toluate as representative compounds. As m-xylene was degraded, m-methylbenzylsuccinate was produced and subsequently consumed (Figure 2A). The concentration of m-methylbenzylsuccinate was found to transiently accumulate to a maximum concentration of 6 µM, representing about 2% of the initial substrate. m-Toluate was also detected as a transient intermediate of m-xylene decay, reaching concentrations as high as 40 µM before subsequent consumption (Figure 2A). Similarly, m-carboxybenzylsuccinate, isophthalate, and benzoate transiently accumulated during the degradation of m-toluate (Figure 2B). m-Carboxybenzylsuccinic acid accumulated to about 5 µM when approximately 80% of the m-toluate was degraded. Thus, the appearance and subsequent consumption of m-methylbenzylsuccinic acid, m-carboxybenzylsuccinic acid, m-toluate, and isophthalate suggest that these compounds are intermediates rather than dead-end metabolites. Metabolites in Groundwater. Previously, we positively identified several putative metabolites from alkylbenzene decay in the groundwater at the gas condensate-contaminated aquifer from which the laboratory enrichments were derived. These metabolites included all of the toluic acid isomers, but 3-methylbenzylsuccinic acid was only tentatively identified due to the lack of an authentic standard (22). Given the identification of alkylbenzene and toluic acid metabolites in the laboratory incubations and the synthesis of authentic
TABLE 2. Comparison of GC-MS Characteristics of TMS-Derivatized Metabolites Detected in Culture Supernatants and TMS-Derivatized Authentic Standardsa,b metabolite
GC RT (min)c
benzylsuccinic acid
37.4
2-methylbenzylsuccinic acid
39.2
3-methylbenzylsuccinic acid
39.3
4-methylbenzylsuccinic acid
39.5
o-toluic acid
21.0
m-toluic acid
21.9
p-toluic acid
22.5
phthalic acid
32.6
isophthalic acid
34.3
terephthalic acid
35.1
benzoic acid
18.3
o-carboxybenzaldehyde
26.3
characteristic mass spectrum m/z (% of base ion) in culture supernatant standard 73(100), 131(41), 145(49), 147(50), 190(31), 205(34), 221(42), 234(21), 337(23), 352(5) 73(100), 105(33), 145(36), 159(34), 204(16), 235(25), 248(15), 351(5), 366(3) 73(100), 105(18), 145(30), 159(37), 204(16), 235(17), 248(11), 351(5), 366(3) 73(100), 105(32), 145(38), 159(30), 204(17), 235(23), 248(15), 351(7), 366(3) 91(53), 119(95), 149(45), 193(100), 208(31) 91(43), 119(63), 149(57), 193(100), 208(14) 91(37), 119(65), 149(66), 193(100), 208(13) 45(14), 73(38), 147(100), 221(12), 295(23), 310(3) 73(26), 135(6), 177(8), 205(13), 221(16), 279(18), 295(100), 310(12) 73(41), 103(19), 221(30), 251(15), 295(100), 310(10) 45(13), 51(16), 77(51), 105(76), 135(6), 179(100), 194(10)
73(100), 131(36), 145(49), 147(41), 190(29), 205(32), 221(42), 234(25), 337(25), 352(8) 73(100), 105(39), 145(43), 159(33), 204(17), 235(22), 248(19), 351(5), 366(3) 73(100), 105(17), 145(32), 159(36), 204(16), 235(20), 248(19), 351(8), 366(4) 73(100), 105(34), 145(36), 159(36), 204(17), 235(27), 248(12), 351(6), 366(3) 91(51), 119(100), 149(44), 193(92), 208(29) 91(43), 119(66), 149(60), 193(100), 208(15) 91(38), 119(67), 149(65), 193(100), 208(12) 45(14), 73(38), 147(100), 221(11), 295(27), 310(3) 73(26), 135(3), 177(8), 205(13), 221(16), 279(18), 295(100), 310(12) 73(41), 103(19), 221(27), 251(16), 295(100), 310(7) 45(15), 51(21), 77(52), 105(76), 135(9), 179(100), 194(10)
75(52), 77(51), 105(79), 133(98), 179(100), 207(90), 222(2)
75(21), 77(50), 105(75), 133(95), 179(100), 207(87), 222(3)
parent enrichmentd toluene (M, SR)
o-xylene (SR) m-xylene (SR) p-xylene (SR) o-xylene (SR) m-xylene (SR) p-xylene (SR) o-toluate (M) m-toluate (M, SR) p-toluate (M) toluene (M, SR), ethylbenzene (SR), all toluates (M), m-toluate (SR) o-toluate (M)
a Authentic standards, other than the methylbenzylsuccinic acids, were commercially available. 2-Methylbenzylsuccinic acid retained impurities; therefore, a melting point could not be determined. 3-Methylbenzylsuccinic acid mp 128-129 °C, literature mp 128-129 °C (32); 4-methylbenzylsuccinic acid mp 118-119 °C, literature mp 114-115 °C (31). b Metabolites were not detected in substrate-unamended incubations, heat-killed incubations, or when the substrate was depleted. c GC retention times of TMS-derivatized authentic standards and culture metabolites according to the temperature program described in Materials and Methods. d Methanogenic enrichments are designated (M) while sulfate-reducing enrichments are designated (SR).
addition product metabolites, the contaminated groundwater from the aquifer was further examined for the presence of 3-phenyl-1, 2-butanedicarboxylic acid, methylbenzylsuccinic acids, carboxybenzylsuccinic and phthalic acids, and carboxybenzaldehydes. Of these, isophthalic, terephthalic, 3-phenyl-1,2-butanedicarboxylic, and 4-methylbenzylsuccinic acids were positively identified in several of the groundwater extracts, and the identification of 3-methylbenzylsuccinic acid was confirmed (Figure 3). The GC retention times and mass spectral profiles of these acids in groundwater matched those of authentic standards (Table 2). Furthermore, a peak eluting at the same GC retention time as o-carboxybenzaldehyde was detected, but the mass spectral profile was more suggestive of m- or p-carboxybenzaldehyde, thus identification could not be made with certainty. Of these metabolites, the toluates were the most abundant in the groundwater, with o-, m-, and p-toluate concentrations ranging from 60 to 185, from 30 to 600, and from 50 to 400 nM, respectively. The benzylsuccinic acid analogues were detected at relatively lower concentrations, ranging from 30 to 100 nM, and the phthalates were the least abundant, ranging from 10 to 50 nM. These concentrations are approximately 3-4 orders of magnitude less that those of the BTEX hydrocarbons detected in the groundwater (2050 mg/L or roughly 200-600 µM). It is of interest to note that these metabolites were not always detected in the contaminated groundwater at every sampling event. For example, 3-methylbenzylsuccinic acid was detected in March 1997, June 1997, June 1998, March 1999, and August 2000 but not in October 1997, March 1998, or October 1998. While isophthalate was only detected in the contaminated groundwater on four of the eight sampling events, m-toluic acid was always detected in at least one of the sampling wells. Furthermore, there was no discernible pattern as to where within the contaminated plume these metabolites were
found; at times they were detected in sampling wells located near the source of contamination, and at other times they were present further downgradient within the plume. As with the laboratory incubations, the sporadic finding of these metabolites in the field may reflect their transient accumulation in situ. Metabolites were never detected in sampling wells located beyond the plume of BTEX contamination nor in groundwater upgradient from the source of contamination.
Discussion In this study, we sought to determine the mechanisms whereby sediment-associated anaerobic populations biodegraded toluene, ethylbenzene, and the xylene and toluate isomers in laboratory incubations and to interpret these results in light of observations of putative metabolites in the field in order to provide evidence for in situ anaerobic biotransformation of such contaminants. As has been observed in this study, toluene and xylene degradation under sulfate-reducing conditions (and methanogenic conditions for toluene) have been reported to occur by mixed laboratory cultures derived from a variety of hydrocarbon-contaminated sites (34). In contrast, sulfidogenic ethylbenzene has rarely been noted, although one study showed partial consumption of ethylbenzene under sulfate-reducing conditions in a BTEX mixture by thermophilic consortia (35). Furthermore, aside from one study examining o-toluate decay by methanogenic sewage sludge populations (36), little has been reported on anaerobic toluate degradation. Much work has focused on the anaerobic oxidation of xylenes to toluates (16, 17, 23), but here we show that toluates themselves are further susceptible to decay under highly reduced conditions. The finding of benzylsuccinic acid in toluene-degrading sulfate-reducing and methanogenic laboratory incubations VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
685
FIGURE 1. Mass spectral profile of (A) TMS derivative of a metabolite detected in ethylbenzene-degrading, sulfate-reducing culture extracts. (B) TMS derivative of an authentic chemical standard of 3-phenyl-1,2-butanedicarboxylic acid (free acid mp 145-147 °C, literature mp 146-149 °C (31)). (C) TMS derivative of a metabolite detected in m-toluate-degrading sulfate-reducing culture extracts. (D) TMS derivative of an authentic chemical standard of m-carboxybenzylsuccinic acid (free acid retained impurities therefore melting point could not be determined). supports the novel fumarate addition pathway for anaerobic toluene metabolism elucidated in pure cultures (4). In cometabolic studies with toluene-degrading populations (pure and mixed cultures) in the presence of toluene and a xylene isomer, toluates and/or methylbenzylsuccinates have been tentatively identified as dead-end metabolites based on mass spectral analogy to benzylsuccinate (8, 10, 16-19). Positive identifications of the methylbenzylsuccinates have generally remained elusive because of the lack of authentic standards. However, Krieger et al. (23) recently identified 3-methylbenzylsuccinate as a transient metabolite of m-xylene degradation under nitrate-reducing conditions, and here we offer positive identification of 2-, 3-, and 4-methylbenzylsuccinates as transient metabolites in sulfidogenic o-, m-, and p-xylene degradation, respectively. We also detected the respective toluic acid isomers, suggesting that the sulfatereducing populations biodegrade the xylene isomers to toluic acids via initial activation to the respective methylbenzylsuccinates (Figure 3). The identification of 3-carboxybenzylsuccinate in mtoluate-degrading sulfate-reducing incubations indicates that activation by an addition reaction is not limited solely to hydrocarbons. Indeed, a similar addition metabolite has also recently been reported to occur in the metabolism of m-cresol by a sulfate reducer (37). To our knowledge, phthalates have not been previously identified as metabolites in the anaerobic decay of toluates, although one aerobic microorganism is known to convert p-toluate to terephthalate (38). The detection of phthalates and benzoate in toluate-amended cultures provides a plausible mechanism for the anaerobic destruction of xylenes, via methylbenzylsuccinic acid, toluic, 686
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001
FIGURE 2. Transient detection of (A) 3-methylbenzylsuccinate and m-toluate in m-xylene-degrading enrichments and (B) benzoate, isophthalate, and 3-carboxybenzylsuccinate in m-toluate-degrading enrichments incubated under sulfate-reducing conditions.
FIGURE 3. Proposed pathway for the degradation of alkylbenzenes in contaminated ecosystems. Compounds detected in groundwater are marked G while compounds detected in laboratory incubations are marked L: (I) toluene; (II) o-xylene; (III) m-xylene; (IV) p-xylene; (V) ethylbenzene; (VI) benzylsuccinic acid; (VII) 2-methylbenzylsuccinic acid; (VIII) 3-methylbenzylsuccinic acid; (IX) 4-methylbenzylsuccinic acid; (X) 3-phenyl-1,2-butanedicarboxylic acid; (XI) o-toluic acid; (XII) m-toluic acid; (XIII) p-toluic acid; (XIV) 3-carboxybenzylsuccinic acid; (XV) phthalic acid; (XVI) isophthalic acid; (XVII) terephthalic acid; (XVIII) benzoic acid. and phthalic acid isomers, which are likely decarboxylated to form benzoic acid (Figure 3). Ethylbenzene degradation by nitrate-reducing organisms involves the production of benzoate via 1-phenylethanol and
acetophenone (24, 26). We were unable to detect these metabolites in the culture supernatants of our sulfidogenic ethylbenzene-degrading enrichments by GC-MS but did detect an addition product analogous to that observed in the VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
687
xylene-amended enrichments. The identified product, 3-phenyl-1,2-butanedicarboxylic acid, suggests that an addition reaction, analogous to that observed for toluene, occurs at the carbon R to the aromatic nucleus. Under nitratereducing conditions, the methylene carbon is also the site of initial attack by ethylbenzene dehydrogenase in Azoarcus sp. strain EB1 (25). It was considered that an addition reaction could also occur at the carbon β to the aromatic nucleus, yielding 2-phenylethylsuccinate. However, this seems less plausible given the increased reactivity of the R carbon due to its proximity to the aromatic nucleus providing a site of relative chemical instability. The lack of detection of 1-phenylethanol and acetophenone does not necessarily preclude their formation, given the mixed nature of the enrichments. However, the identification of 3-phenyl-1,2-butanedicarboxylic acid argues that ethylbenzene is activated in a different manner under sulfate-reducing conditions. Thus, laboratory incubations pointed to a unifying route for anaerobic alkylbenzene decay under sulfate-reducing conditions wherein each alkyl substituent is activated by a fourcarbon addition reaction in a stepwise manner (Figure 3). Many of the metabolites of anaerobic alkylbenzene and toluate decay identified in the laboratory incubations were also detected in the contaminated aquifer from which the microbial populations were derived (Figure 3) and wherein multiple lines of evidence argued that the observed BTEX loss was attributed to microbiological processes (22). Concentrations of metabolites in laboratory incubations were generally 1-3 orders of magnitude less than those of added substrates (0.5-40 µM versus 200-750 µM) whereas metabolites in the groundwater were generally 3-4 orders of magnitude less than dissolved BTEX concentrations measured in the groundwater (0.01-0.6 µM versus 200-600 µM). These differences may be due to losses by dilution or dispersion in the aquifer, which would not be a factor in laboratory incubations. On the basis of comparison with authentic standards, the alkylbenzylsuccinic acids that were positively identified in laboratory incubations were also detected in groundwater samples. The finding of these uniquely anaerobic metabolites in our field and laboratory studies supports the proposal by Beller et al. (28) that these compounds may be useful for assessing in situ anaerobic biodegradation of alkylbenzenes. At the time that this proposal was made, these addition products were thought to be stable, “dead-end” metabolites of alkylbenzene decay, although laboratory studies, including the one conducted here, have since shown that these are transient intermediates (4, 23). The fact that we were still able to detect these signature metabolites in the contaminated groundwater despite their transient nature further validates their usefulness as in situ indicators of a contaminated environment undergoing active anaerobic biodegradation. Because toluates and phthalates may be metabolites of aerobic or anaerobic xylene or toluate metabolism, respectively (4, 38, 39, this study), they cannot be used as unique indicators of anaerobic metabolism in situ. However, their transient accumulation in anaerobic laboratory incubations and sporadic detection in sulfiderich contaminated groundwater suggests that they were formed under anaerobic conditions at this gas condensateimpacted site. They may still serve as useful indicators of in situ biotransformations of alkylbenzenes under a variety of redox conditions. Thus, the integration of laboratory studies with groundwater analysis demonstrated that the potential for anaerobic hydrocarbon decay observed in the laboratory was realized in situ. The combination of both approaches alleviated the disadvantages of each method. For example, the presence of a putative metabolite in groundwater does not indicate whether the metabolite is transiently produced nor if it tends to persist. Further, it is often difficult to postulate a pathway 688
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001
for the metabolism of a compound of interest from groundwater analysis alone. Both of these disadvantages can be overcome by laboratory biodegradation studies. However, a major criticism of laboratory incubations is that they only demonstrate the potential for biodegradation. This can be overcome by analysis of groundwater samples for metabolites detected in laboratory incubations. Analysis of groundwater samples by GC-MS in the selected ion-monitoring mode for the presence of signature metabolites identified in laboratory incubations can provide a direct assessment of the environmental relevance of laboratory findings, especially in field sites where BTEX hydrocarbons appear to be attenuating. We believe that this combined approach will be of specific value in assessing intrinsic bioremediation of fuel-contaminated sites. Previous approaches require more extensive analyses, which tends to undermine the low cost advantage of intrinsic bioremediation processes. We propose that an integration of laboratory studies and groundwater analyses to determine signature microbial metabolites be used as an initial survey for determining whether intrinsic bioremediation of BTEX hydrocarbons is feasible at a given site. This approach meets the last two requirements of the National Research Council’s guidelines for documenting in situ metabolism at far less effort and cost than previous approaches (5, 40). This would be especially useful for small, contaminated sites, which comprise roughly 90% of the reported hydrocarbon release incidents (5, 40).
Acknowledgments This work was supported in part with Grants DE-FG03-96ER-62287 and DE-FG03-96-ER20214/A003 from the Department of Energy, R-827015 from the Environmental Protection Agency and N000149910076 from the Office of Naval Research.
Literature Cited (1) Gerin, M.; Siemiatycki, J.; Desy, M.; Krewski, D. Am. J. Ind. Med. 1998, 34, 144. (2) Maltoni, C.; Ciliberti, A.; Pinto, C.; Soffritti, M.; Belpoggi, F.; Menarini, L. Ann. N.Y. Acad. Sci. 1997, 837, 15. (3) Huff,J. E.; Eastin, W.; Roycroft, J.; Eustis, S. L.; Haseman, J. K. Ann. N.Y. Acad. Sci. 1998, 534, 427. (4) Heider, J.; Spormann, A. M.; Beller, H. R.; Widdel, F. FEMS Microbiol. Rev. 1999, 22, 459. (5) National Academy of Sciences. Natural Attenuation for Groundwater Remediation; National Academy Press: Washington, DC, 2000. (6) Colberg, P. J. S.; Young, L. Y. In Microbial Transformation and Degradation of Toxic Organic Chemicals; Young, L. Y., Cerniglia, C., Eds.; Wiley-Liss: New York, 1995; pp 307-330. (7) Altenschmidt, U.; Fuchs, G. J. Bacteriol. 1992, 174, 4860. (8) Evans, P. J.; Ling, W.; Goldschmidt, B.; Ritter, E. R.; Young, L. Y. Appl. Environ. Microbiol. 1992, 58, 496. (9) Biegert, T.; Fuchs, G.; Heider, J. Eur. J. Biochem. 1996, 238, 661. (10) Beller, H. R.; Spormann, A. M. J. Bacteriol. 1997, 179, 670. (11) Beller, H. R.; Spormann, A. M. Appl. Environ. Microbiol. 1997, 63, 3729. (12) Rabus, R.; Heider, J. Arch. Microbiol. 1998, 170, 377. (13) Chee-Sanford, J. C.; Frost, J. W.; Fries, M. R.; Zhou, J.; Tiedje, J. M. Appl. Environ. Microbiol. 1996, 62, 964. (14) Leutwein, C.; Heider, J. Microbiology 1999, 145, 3265. (15) Leuthner, B.; Heider, J. J. Bacteriol. 2000, 182, 272. (16) Arcangeli, J.-P.; Arvin, E. Biodegradation 1995, 6, 19. (17) Jørgensen, C.; Nielsen, B.; Jensen, B. K.; Mortensen, E. Biodegradation 1995, 6, 141. (18) Rabus, R.; Widdel, F. Arch. Microbiol. 1995, 164, 448. (19) Beller, H. R.; Spormann, A. M.; Sharma, P. K.; Cole, J. R.; Reinhard, M. Appl. Environ. Microbiol. 1996, 62, 1188. (20) Wilson, B. H.; Wilson, J. T.; Kampbell, D. H.; Bledsoe, B. E.; Armstrong, J. M. Geomicrobiol. J. 1990, 8, 225. (21) Cozzarelli, I. M.; Herman, J. S.; Baedecker, M. J. Environ. Sci. Technol. 1995, 29, 458. (22) Gieg, L. M.; Kolhatkar, R. V.; McInerney, M. J.; Tanner, R. S.; Harris, S. H., Jr.; Sublette, K. L.; Suflita, J. M. Environ. Sci. Technol. 1999, 33, 2560.
(23) Krieger, C. J.; Beller, H. R.; Reinhard, M.; Spormann, A. M. J. Bacteriol. 1999, 181, 6403. (24) Ball, H. A.; Johnson, H. A.; Reinhard, M.; Spormann, A. M. J. Bacteriol. 1996, 178, 5755. (25) Johnson, H. A.; Spormann, A. M. J. Bacteriol. 1999, 181, 5662. (26) Rabus, R.; Widdel, F. Arch. Microbiol. 1995, 163, 96. (27) Wilkes, H.; Boreham, C.; Harms, G.; Zengler, K.; Rabus, R. Org. Geochem. 2000, 31, 101. (28) Beller, H. R.; Ding, W.-H.; Reinhard, M. Environ. Sci. Technol. 1995, 29, 2864. (29) Reinhard, M.; Shang, S.; Kitanidis, P. K.; Orwin, E.; Hopkins, G. D.; Lebron, C. A. Environ. Sci. Technol. 1997, 31, 28. (30) McInerney, M. J.; Bryant, M. J.; Pfennig, N. Arch. Microbiol. 1979, 122, 129. (31) Bickford, W. G.; Fisher, G. S.; Dollear, F. G.; Swift, C. E. J. Am. Oil Chem. Soc. 1948, July, 251. (32) Shechter, H.; Barker, H. C. J. Org. Chem. 1956, 21, 1473. (33) Kousaka, T.; Tatsuya, S.; Ryohei, O. Kogyo Kagaku Zasshi 1968, 71, 1738.
(34) Krumholz, L.; Caldwell, M. E.; Suflita, J. M. In Bioremediation: Principles and Applications; Crawford, R., Crawford, D., Eds.; Cambridge University Press: Cambridge, 1996; pp 61-99. (35) Chen, C.-I.; Taylor, R. T. Appl. Microbiol. Biotechnol. 1997, 48, 121. (36) Londry, K. L.; Fedorak, P. M. Can. J. Microbiol. 1993, 39, 812. (37) Mu ¨ ller, J. A.; Galushko, A. S.; Kappler, A.; Schink, B. Arch. Microbiol. 1999, 172, 287. (38) Locher, H. H.; Malli, C.; Hooper, S. W.; Vorherr, T.; Leisinger, T.; Cook, A. M. J. Gen. Microbiol. 1991, 137, 2201. (39) Singer, J. T.; Finnerty, W. R. In Petroleum Microbiology; Atlas, R. M., Ed.; Macmillan Publishing: New York, 1984; pp 299354. (40) National Research Council. In-situ bioremediation: when does it work? National Academy Press: Washington, DC, 1993.
Received for review August 9, 2000. Revised manuscript received November 16, 2000. Accepted November 21, 2000. ES001571U
VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
689
