Article pubs.acs.org/est
Monitoring Biodegradation of Ethene and Bioremediation of Chlorinated Ethenes at a Contaminated Site Using Compound-Specific Isotope Analysis (CSIA) Scott O. C. Mundle,† Tiffany Johnson,† Georges Lacrampe-Couloume,† Alfredo Pérez-de-Mora,‡,⊥ Melanie Duhamel,‡ Elizabeth A. Edwards,‡ Michaye L. McMaster,§ Evan Cox,§ Kinga Révész,∥ and Barbara Sherwood Lollar*,† Departments of †Geology and ‡Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario, Canada § Geosyntec Consultants, Guelph, Ontario, Canada ∥ U.S. Geological Survey, Reston, Virginia, United States ⊥ Research Unit of Biogeochemistry and Analytics, Helmholtz Center Munich, Germany S Supporting Information *
ABSTRACT: Chlorinated ethenes are commonly found in contaminated groundwater. Remediation strategies focus on transformation processes that will ultimately lead to nontoxic products. A major concern with these strategies is the possibility of incomplete dechlorination and accumulation of toxic daughter products (cis-1,2-dichloroethene (cDCE), vinyl chloride (VC)). Ethene mass balance can be used as a direct indicator to assess the effectiveness of dechlorination. However, the microbial processes that affect ethene are not well characterized and poor mass balance may reflect biotransformation of ethene rather than incomplete dechlorination. Microbial degradation of ethene is commonly observed in aerobic systems but fewer cases have been reported in anaerobic systems. Limited information is available on the isotope enrichment factors associated with these processes. Using compound-specific isotope analysis (CSIA) we determined the enrichment factors associated with microbial degradation of ethene in anaerobic microcosms (ε = −6.7‰ ± 0.4‰, and −4.0‰ ± 0.8‰) from cultures collected from the Twin Lakes wetland area at the Savannah River site in Georgia (United States), and in aerobic microcosms (ε = −3.0‰ ± 0.3‰) from Mycobacterium sp. strain JS60. Under anaerobic and aerobic conditions, CSIA can be used to determine whether biotransformation of ethene is occurring in addition to biodegradation of the chlorinated ethenes. Using δ13C values determined for ethene and for chlorinated ethenes at a contaminated field site undergoing bioremediation, this study demonstrates how CSIA of ethene can be used to reduce uncertainty and risk at a site by distinguishing between actual mass balance deficits during reductive dechlorination and apparent lack of mass balance that is related to biotransformation of ethene.
■
INTRODUCTION Trichloroethene (TCE) is a toxic industrial solvent commonly found in contaminated groundwater in North America. Remediation strategies are primarily focused on chemical or microbial transformation processes that replace the chlorine atoms of the contaminant with hydrogen atoms in a process that can ultimately lead to a nontoxic product (ethene).1−3 In bioremediation-based strategies, TCE degradation is catalyzed by anaerobic microorganisms that convert TCE to ethene via reductive dechlorination.4−7 These biocatalyzed approaches have improved the efficiency of remediation; however, the chemical transformation of TCE to ethene is stepwise and involves successive replacement of a chlorine atom with a hydrogen atom that must initially form cis-1,2-dichloroethene (cDCE), followed by vinyl chloride (VC), which is subsequently degraded to ethene (Scheme 1).8,9 A major concern with these remediation strategies is the possibility of incomplete degradation and accumulation of cDCE and VC, the latter in particular due to its toxicity and carcinogenic properties.10−12 © 2011 American Chemical Society
A straightforward approach to evaluate the progress of remediation is to determine the mass balance of the reaction.13,14 If mass balance is achieved between TCE and ethene it implies that the biocatalyzed transformation proceeds through to the nontoxic product (ethene) without accumulation of daughter products along the pathway. However, achieving mass balance assumes that ethene is the final product of the degradation pathway. Quantifying the success of remediation by this mass balance approach is impossible if ethene is being degraded. Degradation of ethene itself can be difficult to definitively identify, as the products of ethene biodegradationlikely ethane, methane, and carbon dioxide are produced by a multitude of natural sources and cannot be used as direct indicators for biodegradation of ethene.15,16 Received: Revised: Accepted: Published: 1731
August 10, 2011 December 13, 2011 December 22, 2011 December 22, 2011 dx.doi.org/10.1021/es202792x | Environ. Sci. Technol. 2012, 46, 1731−1738
Environmental Science & Technology
Article
Scheme 1. Stepwise Pathway in Bio-Catalyzed Degradation of TCE to Ethene or Further Products
determined the isotopic fractionation associated with microbial transformations of ethene by anaerobic cultures enriched from groundwater and soil collected from the Twin Lakes wetland area at the Savannah River site in Georgia, and by aerobic cultures of Mycobacterium sp. strain JS60 originally isolated by Coleman et al.26 The second objective of this work was to apply this approach to a contaminated field site in southwestern Ontario (Canada) that is undergoing bioremediation to determine whether or not ethene biodegradation is occurring in addition to reductive dechlorination.
Compound-specific isotope analysis (CSIA) has been used in both laboratory and field studies to identify biodegradation of chlorinated ethenes.17−20 The δ13C isotope value of a contaminant is highly sensitive to processes that involve breaking chemical bonds. This sensitivity arises from the different reaction rates associated with cleavage of bonds containing heavy or light isotopes.21 The heavier isotope forms a stronger bond relative to the light isotope and requires more energy to break. This increases the kinetic energy barrier for the heavier atom and subtly slows the reaction (kinetic isotope effect) relative to the reaction of the lighter isotope.22 Because the heavier isotope reacts more slowly during the course of biochemical transformation, the residual starting material becomes enriched in the heavy isotope as the lighter isotope is incorporated into the product of the reaction. Initially the δ13C value of the product is preferentially rich in the lighter (12C) isotope and has a more negative δ13C value than the starting material. As the daughter product itself undergoes biodegradation, its δ13C value also undergoes a progressive enrichment trend due to the preferred rate of reaction of 12C containing compounds. The final product of a multistep process will ultimately have the original δ13C value of the starting material if two conditions are met. First, there must be complete transformation of the original contaminant to the daughter product, and second, there must be mass balance with respect to both concentrations and isotope signatures. This study demonstrates that the isotopic composition of the final daughter product of dechlorination, ethene, can be used to assess the effectiveness of remediation of contaminated sites particularly in the case where there is a lack of VOC mass balance. In such cases it can be challenging to distinguish between lack of VOC mass balance due to incomplete dechlorination versus lack of mass balance due to degradation of ethene occurring simultaneously with dechlorination. Both processes will decrease the measured concentrations of ethene, but only the latter (degradation of ethene) is likely to involve isotopic fractionation due to cleavage of the carbon−carbon double bond in ethene.20 The overall goal of this study was to test the hypothesis that carbon isotope fractionation during degradation of ethene can be used to evaluate mass balance during dechlorination and the effectiveness of bioremediation at contaminated field sites. Microbial degradation of ethene is commonly observed in aerobic systems.15 Fewer cases have been reported in anaerobic systems.16,23,24 Aerobic degradation of ethene involves processes that lead to methane and/or carbon dioxide.15 In contrast, anaerobic processes convert ethene to ethane.23 As both of these biodegradation processes are likely to involve isotopic fractionation, we hypothesized that the isotopic value of ethene may be used as an indicator to determine if ethene degradation is taking place, analogous to approaches that have been developed for chlorinated ethenes.25 The first objective of this study was to investigate the carbon isotopic fractionation during microbial degradation of ethene in both anaerobic and aerobic microcosms using CSIA. We
■
MATERIALS AND METHODS Anaerobic Microcosms. As part of a study of chlorinated ethene reductive dechlorination, anaerobic microbial microcosms were prepared by Dr. David Freedman at Clemson University from soil and groundwater samples collected in 2003 and in 2007 from two wells in the Twin Lakes wetland area at the Savannah River site in Georgia (United States) downgradient of a former TCE disposal site. The cultures were maintained with injections of cDCE (6−11 μmol) and VC (1− 2 μmol). These microcosms dechlorinated cDCE and VC to ethene with lactate as an electron donor, and in some cases, ethane production was also observed. From the microcosms established in 2003, four experimental bottles (Microcosm Experiment 1) were prepared containing 35 mL of soil and contaminated groundwater and maintained with VC or ethene (0.1−0.2% in headspace) and 3.7 g/100 mL of sodium lactate. Ethene degradation experiments were initiated with 16−25 μmol of ethene and 1.8 μmol of sodium lactate. From the microcosms established in 2007, two experimental bottles (Microcosm Experiment 2) were prepared containing 35 mL of soil and groundwater and maintained on VC (0.025% in headspace) and 3.7 g/100 mL of sodium lactate. Ethene degradation experiments were initiated with 13−22 μmol of ethene, and 3.6 μmol of sodium lactate. For all microcosm experiments, concentrations and δ13C values were measured for ethene and ethane in the headspace as ethene was biodegraded. Four aqueous control bottles were prepared containing deionized water, 13−33 μmol of ethene, and 1.8− 3.6 μmol of sodium lactate, along with three groundwater controls containing an artificial groundwater solution (Middeldorp et al.27) and similar amounts of ethene and sodium lactate. All microcosms were setup in an anaerobic chamber (Coy Laboratory Products) and amended with a gas mix of 80% N2, 10% CO2, and 10% H2. All experiments were performed in 125-mL glass bottles sealed with crimped blue butyl rubber stoppers (control studies showed no sorption),28 and maintained and incubated at room temperature inside the anaerobic glovebox after the method of Sherwood Lollar et al.29 The headspace of the bottles was purged with a deoxygenated filter-sterilized mixture of 80% N2 and 20% CO2 prior to addition of ethene. Aerobic Microcosms Cultures. Aerobic ethene degradation microcosms were prepared from cultures of Mycobacterium sp. strain JS60 (originally isolated by Coleman et al. 2002) from 1732
dx.doi.org/10.1021/es202792x | Environ. Sci. Technol. 2012, 46, 1731−1738
Environmental Science & Technology
Article
Figure 1. Concentration (μmol) versus time (days) for anaerobic degradation of ethene (open circles) to ethane (solid circles) for (A) Microcosm Experiment 1, and (B) Microcosm Experiment 2. Results are shown for one bottle but are representative of results in all replicate bottles in each case. (C) Logarithmic Rayleigh plot of combined data for anaerobic degradation of ethene for Microcosm Experiment 1. (D) Logarithmic Rayleigh plot of combined data for anaerobic degradation of ethene by Microcosm Experiment 2. Error bars are based on total uncertainty of ±0.5‰ in δ13C and ±7% on concentration (see text).
stocks frozen at −80 °C. Approximately 0.13 mL of culture was added to each of three 125-mL serum bottles containing approximately 62 mL of Minimal Salts Medium (MSM) to produce a stock solution as described by Coleman et al.26 Three experimental bottles containing 5 mL of JS60 stock culture, 43 mL of MSM (23 mM phosphate buffer, 5 mM ammonium sulfate), and 2 mL of trace metals solution were prepared. Concentrations and δ13C values for ethene were measured from the headspace as ethene biodegradation occurred. Replicate experiments were run using 5-mL aliquots of culture taken from each of the three experimental bottles after all of the ethene from the first experiment was degraded. The 5-mL aliquots were added to inoculate new bottles containing 43 mL of MSM and 2 mL of trace metals solution, and 175−195 μmol of ethene. Laboratory Microcosm Analytical Methods. Ethene and ethane concentrations for anaerobic microcosms, and ethene concentrations for aerobic microcosms, were measured using headspace analysis as described by Slater et al.30 Three hundred μL of headspace was withdrawn using gastight syringes and compositional analysis was completed using a Varian 3400 gas chromatograph fitted with a 30 m × 0.53 mm ID GS-Q (J & W Scientific) column and a flame-ionization detector. For the aerobic experiments, after each headspace sample was taken from a culture or control bottle, an equivalent volume of filtersterilized oxygen (UHP 9.8, Praxair) was injected into the bottle to restore pressure and to prevent oxygen depletion. In
the anaerobic experiments the headspace volume removed was replaced with an equal volume of filter sterilized gas (80% N2, 10% CO2, and 10% H2). The GC temperature program was 35 °C held for 2 min, increased at 30 °C per minute to 70 °C, and held for 1 min. The helium carrier gas flow through the column was 7.0 ± 0.7 mL min−1. Measured concentrations were calibrated to an external standard. Reproducibility on analysis of standards was ±5% (relative error). Because each bottle was sampled at multiple time points, the fraction remaining ( f) for each sample was corrected for the total moles of ethene or ethane withdrawn over the course of the experiments. For isotope measurements the headspace was sampled using gastight syringes after the method of Slater et al.30 and δ13C values were measured on a continuous flow GC/C/IRMS consisting of a Varian 3400 gas chromatograph, microcombustion furnace, and Finnigan MAT 252 isotope ratio mass spectrometer. The GC was fitted with a 60 m × 0.32 mm ID GS-Q (J&W Scientific) column and the temperature was held isothermal at 35 °C. Total uncertainty on δ13C values is ±0.5‰, incorporating both the accuracy of the measurement with respect to international standards (V-PDB) and the reproducibility on replicate measurements of the sample.31 Control bottles in all experiments did not show changes in ethene concentrations outside of analytical uncertainty, and showed no ethane or methane production. Control bottles did 1733
dx.doi.org/10.1021/es202792x | Environ. Sci. Technol. 2012, 46, 1731−1738
Environmental Science & Technology
Article
Table 1. Summary of Enrichment Values Calculated in Anaerobic and Aerobic Microcosms Degrading Ethene anaerobic degradation of ethene microcosm experiment 1
aerobic degradation of ethene
microcosm experiment 2
JS60 ( f < 20% omitted)
JS60 (all data)
bottle
ε (‰)
CI (95%)
r
ε (‰)
CI (95%)
r
ε (‰)
CI (95%)
r
ε (‰)
CI (95%)
r2
1 2 3 4 5 6 all
−7.5 −6.4 −6.8 −6.3
1.6 0.6 0.6 0.6
0.95 0.99 0.99 0.98
−3.6 −4.3
1.1 0.9
0.87 0.92
−6.7
0.4
0.97
−4.0
0.8
0.86
−4.3 −3.2 −5.5 −4.8 −4.5 −3.3 −4.5
0.6 0.6 1.3 1.0 0.6 0.6 0.4
0.96 0.93 0.87 0.92 0.96 0.93 0.87
−3.3 −2.8 −3.4 −2.9 −3.4 −2.8 −3.0
0.9 0.6 1.1 0.6 0.5 0.5 0.3
0.89 0.92 0.84 0.94 0.97 0.95 0.88
2
2
not show changes in ethene δ13C values outside of analytical uncertainty (±0.5‰ of the starting value). Data Analysis. Enrichment factors (per mil) for ethene degradation were calculated using the Rayleigh model for closed systems (eq 1):
⎛ δ + 1⎞ ln f × (α − 1) = ln⎜ t ⎟ ⎝ δ0 + 1 ⎠
2
an Agilent 5890 and a Finnigan Delta XP mass spectrometer via a combustion interface. The GC was fitted with a 60 m × 0.25 mm GSQ column with a temperature program of 35 °C for 3 min, increasing at 10 °C/min to 90 °C, then increasing at 25 °C/min to 210 °C, and holding for 5 min. Depending on the concentration of the sample, 5−80 mL of groundwater was injected into the purge and trap concentrator using a gastight syringe. The sample was purged for 11 min with a helium flow of 40 mL/min. The trap was then purged with dry helium for 1 min, followed by heating the trap to 250 °C. The desorbed compounds were injected over the course of 4 min into the GC through a heated line (260 °C) with a flow rate of 50 mL/min. The split varied from 1:1 to 20:1 as a function of the concentration of compounds in each sample.
(1)
where f is the fraction of ethene remaining, δt is the δ13C value of ethene at time t, δ0 is the initial δ13C value, and α is the fractionation factor.32 Fractionation factors were converted to enrichment factors (ε) using the equation ε = (α − 1). Field Site Description. At an industrial site in southwestern Ontario, Canada (ISSO) TCE (used as a degreasing agent) was released to the subsurface through historical facility operations (Figure S1). The site geology consists of 3−6 m (m) of clayey-silt overburden overlying a fractured carbonate bedrock aquifer system that includes cherty, fossiliferous, and argillaceous limestones and dolostones of the Bois Blanc Formation; the dolostones and shales of the Bertie Formation; and the underlying shales, evaporites, and dolostones of the Salina Formation.33 The water table is located within the overburden at a depth of 1−2 m below the ground surface. Groundwater flow in the bedrock aquifer occurs primarily through subparallel bedding plane fractures, and also via secondary porosity features resulting from dissolution of calcite and gypsum. Maximum concentrations of TCE (>140 mg/L) have been detected in the bedrock groundwater. Significant concentrations of cDCE and VC also occur suggesting some degree of dechlorination by native microorganisms at the site. In 2008, an enhanced in situ bioremediation (EISB) system was installed at the site. The EISB system consists of a recirculation loop, where groundwater is extracted from three wells located downgradient of the suspected source areas. The extracted groundwater is amended with electron donor (ethanol) and injected into the bedrock aquifer through recharge wells located upgradient of the extraction wells in the vicinity of the source area(s). In October 2009, select recharge wells were bioaugmented with KB-1 microbial culture to enhance the reductive dechlorination of TCE and its degradation products (cDCE and VC) to ethene. Samples were collected in February 2010 for CSIA analysis. Field Site Analytical Methods. Groundwater field samples were collected in 40-mL volatile organic compound (VOA) vials treated with 1 mL of 6 N hydrochloric acid to avoid further biodegradation after sampling.25 The samples were shipped and stored at 4 °C. The samples were analyzed for VOCs at a commercial laboratory. CSIA was completed using a purge and trap Teledyne-Tekmar XPT concentrator coupled to
■
RESULTS AND DISCUSSION Anaerobic Ethene Degradation and Isotopic Profiles. Anaerobic microbial degradation of ethene typically leads to stoichiometric conversion to ethane over 40 to 166 days.16 Consistent with this, we observed quantitative conversion of ethene to ethane in approximately 60 days (Microcosm Experiment 1, Figure 1A) and 100 days (Microcosm Experiment 2, Figure 1B) by the microbial cultures in this study. The δ13C of ethene increased to values between −22.5‰ and −19.4‰ (δ13C0 = −30.2 ± 0.5‰) in Microcosm Experiment 1, and −26.2‰ and −24.5‰ (δ13C0 = −30.9 ± 0.7‰) in Microcosm Experiment 2, consistent with preferential rate of biotransformation of 12C containing ethene versus ethene with a 13C. Data from each individual experimental bottle fit a Rayleigh model with r2 = 0.87−0.99 (Table 1). Calculated Rayleigh enrichment factors ranged from −7.5‰ to −6.3‰ (r2 = 0.95−0.99) for individual bottles in Microcosm Experiment 1 (Table 1). Because these agree within 95% confidence intervals (CI), a combined enrichment factor for all data of −6.7‰ ± 0.4‰ (r2 = 0.97) was calculated (Figure 1C). The enrichment factors for Microcosm Experiment 2 were −4.3‰ and −3.6‰, with a combined enrichment factor of −4.0‰ ± 0.8‰ (r2 = 0.86) (Table 1, Figure 1D). The slightly different enrichment factors between Microcosm Experiments 1 and 2 may reflect differences resulting from a shift in the microbial community between the sampling dates (2003 and 2007, respectively) of the microcosms. For both Microcosm Experiments 1 and 2, the final δ13C of ethane (Figure S2) was identical to the starting isotope value of the ethene. The quantitative conversion of ethene to ethane combined with identical δ13C values for the starting material and final product indicates complete compositional and isotopic mass balance was achieved in all of these microcosm experiments. If microbial degradation had led to any product(s) other than ethane, isotopic mass balance would not 1734
dx.doi.org/10.1021/es202792x | Environ. Sci. Technol. 2012, 46, 1731−1738
Environmental Science & Technology
Article
Figure 2. (A) Concentration (μmol) versus time (hours) for the aerobic degradation of ethene by JS60 (symbols denote different bottles, open symbols represent initial experiments, closed symbols represent repeat experiments using aliquots of culture from the initial experiments). (B) Logarithmic Rayleigh plot of combined data for aerobic degradation of ethene with data points up to 20% fraction remaining ( f) (see text). Error bars are based on total uncertainty of ±0.5‰ in δ13C and ±7% on concentration.
the degradation pathway. A major concern in this area involves accumulation of VC.17 Different approaches have attempted to address this concern, including the use of mass balance. If concentration data can be obtained for all species along the degradation pathway (including ethene), and mass balance is achieved, there is strong evidence that VC is being degraded. Isotopic mass balance has also been used to assess the accumulation of VC, where the isotope value of VC can be used as an indicator for dechlorination.38,39 The next logical step is to examine the fate of ethene directly using CSIA. This study has demonstrated that microbial degradation of ethene under both anaerobic and aerobic conditions is an isotope-fractionating process. Figure 3 illustrates a model for
have been achieved. Enzyme-catalyzed transformations of alkenes to alkanes is thought to proceed via a mechanism involving simultaneous trans-addition of hydride and a proton, which is consistent with the small ε-values (associated with concerted reactions) observed in this study.34−36 Aerobic Ethene Degradation and Isotopic Profiles. Aerobic degradation of ethene is more common than anaerobic degradation and can occur through a variety of oxidative pathways.15 In this study, aerobic degradation did not produce ethane, and occurred over a much shorter time course (8 days, Figure 2A) relative to anaerobic degradation (Figure 1A and B). The δ13C isotope values for ethene during degradation by Mycobacterium sp. strain JS60 increased to values between −23.1‰ and −10.8‰ at the end of the experiments (δ13C0 = −30.0 ± 0.5‰). For all six experiments, the data fit a Rayleigh model with r2 = 0.87−0.96 (Table 1); however, calculated epsilon values were not all within confidence intervals of each other. Particularly large confidence intervals (1.0−1.3) were found for the two experimental bottles (calculated ε values = −4.8 and −5.5, respectively) for which the final data points were close to the detection limit for δ13C of ethene based on a peak amplitude of 0.5 V after Sherwood Lollar et al.31 For all bottles there seems to be greater uncertainty in the δ13C values calculated close to the detection limit (typically at
