Environ. Sci. Technol. 2001, 35, 4449-4456
Further Biogeochemical Characterization of a Trichloroethene-Contaminated Fractured Dolomite Aquifer: Electron Source and Microbial Communities Involved in Reductive Dechlorination A. M. HOHNSTOCK-ASHE,† S. M. PLUMMER,‡ R. M. YAGER,| P . B A V E Y E , § A N D E . L . M A D S E N * ,† Department of Microbiology, Wing Hall Cornell University, Ithaca, New York 14853, Department of Crop and Soil Sciences, Bradfield Hall, Cornell University, Ithaca, New York 14853, and USGS, Ithaca, New York 14850
A recent article presented geochemical and microbial evidence establishing metabolic adaptation to and in-situ reductive dechlorination of trichloroethene (TCE) in a fractured dolomite aquifer. This study was designed to further explore site conditions and microbial populations and to explain previously reported enhancement of reductive dechlorination by the addition of pulverized dolomite to laboratory microcosms. A survey of groundwater geochemical parameters (chlorinated ethenes, ethene, H2, CH4, DIC, DOC, and δ13C values for CH4, DIC, and DOC) indicated that in situ reductive dechlorination was ongoing and that an unidentified pool of organic carbon was contributing, likely via microbial respiration, to the large and relatively light onsite DIC pool. Petroleum hydrocarbons associated with the dolomite rock were analyzed by GC/MS and featured a characteristically low δ13C value. Straight chain hydrocarbons were extracted from the dolomite previously found to stimulate reductive dechlorination; these were particularly depleted in hexadecane (HD). Thus, we hypothesized that HD and related hydrocarbons might be anaerobically respired and serve both as the source of onsite DIC and support reductive dechlorination of TCE. Microcosms amended with pulverized dolomite demonstrated reductive dechlorination, whereas a combusted dolomite amendment did not. HD-amended microcosms were also inactive. Therefore, the stimulatory factor in the pulverized dolomite was heat labile, but that component was not HD. Amplified Ribosomal DNA Restriction Analysis (ARDRA) of the microbial populations in well waters indicated that a relatively low diversity, sulfur-transforming community outside the plume was shifted toward a high diversity community including Dehalococcoides ethenogenes-type microorganisms inside the zone of contamination. These observations illustrate biogeochemical intricacies of in situ reductive dechlorination reactions.
Introduction The industrial solvent trichloroethylene (TCE) is among the most ubiquitous of groundwater contaminants (1). Under anaerobic conditions, both TCE and the related solvent 10.1021/es0110067 CCC: $20.00 Published on Web 10/06/2001
2001 American Chemical Society
tetrachloroethylene (PCE) can undergo a variety of reactions (2, 3) that include sequential microbial reductive dechlorination to lesser chlorinated ethenes, dichloroethylene (cisDCE), and vinyl chloride (VC) and to nontoxic ethene (4, 5). These chemical transformations can be carried out cometabolically by sulfate-reducing, methanogenic, and acetogenic microorganisms that produce reactive transition metal coenzymes (4, 6). More importantly, PCE and TCE can be used as physiological electron acceptors for dehalorespiring bacteria (7, 8). Because dehalorespiration leads to the growth of microorganisms, this process is robust and provides a foundation for potential intrinsic bioremediation of sites contaminated with chlorinated solvents (2, 3, 9, 10). Dehalorespiration has become the focus of considerable scientific inquiry (11-13). Many pure cultures capable of dehalorespiration have been isolated and include microorganisms from the phylogenetically diverse genera Desulfitobacterium, Desulfuromonas, Dehalospirillum, Dehalobacter, and Dehalococcoides (14-20). A single representative of the latter genus, Dehalococcoides ethenogenes 195, can completely dechlorinate PCE to ethene (4, 8). Understanding the microbial ecology of these organisms, particularly D. ethenogenes in contaminated sites, contributes to accruing fundamental knowledge of bioremediation technology (21). Reductive dechlorination reactions can only occur in geochemical settings conducive to key electron donor/ acceptor reactions. In such sites, tangible geochemical and microbiological “footprints” of dehalorespiration are evident (2). Clear evidence for complete reductive dechlorination of TCE in situ in a fractured dolomite aquifer near Niagara Falls, NY was obtained in a previous study (22). Yager et al. (22) reported the presence of cis-DCE and VC on site, although neither of these compounds was among the pollutants released. Ethene, the product of complete reductive dechlorination of TCE, was also detected. In addition, Yager et al. (22) demonstrated that microorganisms native to site waters inside, but not outside, the contaminant plume were adapted to catalyze reductive dechlorination reactions. Furthermore, stimulation of reductive dechlorination activity was demonstrated in laboratory microcosms containing contaminated groundwater amended with samples of the geologic aquifer matrix (pulverized fractured dolomite). Yet questions about the in situ source of electrons driving reductive dechlorination and the mechanism of stimulation (surface area effect versus nutrients versus carbon source) remain unanswered. The objectives of the present investigation were to advance case-specific biogeochemical knowledge of reductive dechlorination processes by (i) expanding the study site’s geochemical database; (ii) investigating the in-situ source of carbon that might support microbial reductive dechlorination reactions, (iii) attempting to explain the previously reported stimulation of reductive dechlorination by the pulverized dolomite, and (iv) analyzing the composition of the microbial community both inside and outside of the contaminant plume using 16S rDNA procedures.
Methods Study Site and Sampling. The study site is adjacent to a manufacturing facility 8 km east of Niagara Falls, NY. * Corresponding author phone: (607)255-3086; fax: (607)255-3904; e-mail:
[email protected]. † Department of Microbiology, Cornell University. ‡ Present address: Colorado School of Mines, Golden, CO. | USGS. § Department of Crop and Soil Sciences, Cornell University. VOL. 35, NO. 22, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4449
Characteristics of this site and both groundwater sampling and fixation procedures have been previously published (22-27). Duplicate samples for analysis of ethene, methane, dissolved organic carbon (DOC), and dissolved inorganic carbon (DIC) were gathered and fixed. For measurements of DIC, the method of Hunkeler et al. (28) was followed. Forty milliliter I Chem vials were filled to the brim. Within 30 min of sampling, 2 mL was removed and replaced with 0.4 mL of 2 M NaOH and 1.6 mL of 1.2 M BaCl2. The vials were then closed, sealed with electrical tape, and kept at 4 °C. Two wells were selected for microbial community characterization: well 87-12, located at the center of the downgradient plume (well screen 7-10 m below the surface) and well 89-17, located 0.8 km east of 87-12 outside of the contaminated zone [well screen 9-11 m below ground surface (22)]. Cells for DNA extraction were concentrated from ∼5 L of groundwater onto 142 mm diameter Millipore Corp. (Bedford, MA) Durapore membranes (0.22 µm pore size) on site. Filters were placed in sterile Whirlpak bags and frozen immediately by immersion in a dry ice-ethanol bath. Samples were subsequently transferred to a -80 °C freezer in the laboratory. Rock cores were obtained [wireline coring with air and a diamond bit (22)] from uncontaminated [October 1999] and contaminated [March 2000] areas. Rock core was submerged in water pumped from the borehole and kept on ice or 4 °C until being dried, pulverized, and autoclaved for use in microcosms or chemically analyzed. Aseptically chipped samples of the March 2000 rock core were used as the source of microbial inocula or for microscopic analysis, the latter after being fixed on site by being placed in 5% formaldehyde. Chemical Analyses of Groundwater, Rock, and Microcosm Headspace Gases. TCE, chlorinated metabolites, H2, and methane were analyzed on-site during sampling by R. W. Lee, U.S. Geological Survey, Austin, TX. Samples were collected from a 250 mL gas sampling bulb (29). Headspace samples of the gas bubble were taken and injected into three separate gas chromatographs. Volatile chlorinated compounds were analyzed on a Photovac 10S50 with a photoionization detector. Hydrogen was analyzed on a Trace Analytical TGA3 with a reducing gas detector, and methane was determined with a Micro Sensor Technology unit fitted with a thermal conductivity detector. Dissolved organic carbon and inorganic carbon were measured using standard U.S. Geological Survey techniques (22, 30). Ethene and methane in headspaces of fixed water samples were analyzed as previously described (22), as were chlorinated ethenes in microcosm headspaces. Initial geochemical analyses of dolomite rock samples were conducted by Baseline Resolution, Inc. (The Woodlands, TX), which completed a Soxhlet extraction, asphaltene removal by n-pentane treatment, medium-pressure liquid chromatographic separation (of saturates, aromatics, and nitrogen-, sulfur-, and oxygencontaining compounds), GC/MS analysis of saturates and aromatics, and bulk total organic carbon analysis. For subsequent GC/MS analysis of the dolomite rock, 3:1 ethyl acetate extracts (vol: rock wt, 4 days at 22 °C) were used. Each extract was evaporated to dryness, redissolved in 20 µL ethyl acetate, and analyzed by GC/MS. Solvent blanks were prepared concomitantly with the dolomite samples. One microliter was injected into a HP5890 series II gas chromatograph, equipped with a HP 5971A mass selective detector and a HP-5 column (30 m, 0.25 mm ID, 0.25 µm film thickness). The temperature program was as follows: 60 °C (1 min), 10 °C per minute to 260 °C (3 min), 20 °C per minute to 300 °C (15 min). The automatically controlled pressure was set for 12 psi (at 150 °C), and the carrier gas was helium. The mass spectrometer was operated at a pressure of 5.2 × 10-6 Torr in scan mode for masses between 10 and 550 at an EM voltage of approximately 1700 V. Injector and detector 4450
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 22, 2001
temperatures were 270 and 290 °C, respectively. Stable Isotopic Analyses. δ13C values for DOC were measured by the Environmental Isotope Laboratory at the University of Waterloo. Carbonate precipitate in the DIC samples was collected by vacuum filtration onto 0.8 µm pore size, 47 mm diameter acetate filter membranes (Micron Separations Inc., Westboro, MA). The precipitate was rinsed three times with 100 mL of warm, sterile deionized water until the pH of the final rinsate was no longer alkaline. The filters were dried at 80 °C for 15 min. The dried filtrate was scraped from the filter with a sterile spatula and collected in sterile 2 mL glass screw capped vials. The filtrate was processed by the Environmental Isotope Laboratory at the University of Waterloo or by the Biogeochemistry Isotope Laboratory, Cornell University. Duplicate samples analyzed by both facilities gave similar values of δ13C. Aquifer rock (dolomite) was analyzed by the Soil Stable Isotope Facility, Department of Crop and Soil Science, Cornell University before and after complete acidification of the dolomite (concentrated HCl) had volatilized all the carbonate present. Water samples for methane analysis were kept inverted at 4 °C until analysis by the method of von Fischer and Hedin (31). The septa of each vial was pierced with a syringe needle, and 5 mL of headspace gas was injected into a gas manifold containing a series of chemical and cryogenic traps. Following purification, the methane was oxidized to CO2, which was cryofocused and then eluted onto a modified Europa Scientific Geo 2020 stable isotope ratio mass spectrometer (PDZ Europa, Crewe, U.K.) for 13C/12C determination. Microscopy. For Acridine Orange (AO) direct cell counts, 1 mL groundwater samples were concentrated on black Isopore polycarbonate membrane filters (0.2 µm pore size, 13 mm diameter). The filters were fixed in 2% formaldehyde in phosphate buffered saline (PBS) medium on site (32). Prior to use, all solutions were filter sterilized with 0.2 µm pore size cellulose acetate filters. Filters were stained with 0.01% AO and viewed by epifluorescent microscopy as described previously (33). Samples were counted for total cells following the statistical procedures of Trolldenier (34, 35). Epifluorescent microscopy was used to examine the fracture face of dolomite rock for the presence of a biofilm community. Chips of the dolomite fracture face were removed by sterile chiseling. The chips (fracture face up) were placed on glass slides and surrounded by a rubber coverwell (Grace Bio-Labs, Bend, OR). The rubber was sealed around the edges with vaspar (1:1 vaseline:paraffin). Dolomite chips were immersed in 0.01% AO for 3 min followed by PBS for 1 min. PBS was removed, and chips were immersed in 1.5% low melting point NuSieve GTG agarose (FMC BioProducts, Rockland, ME) in PBS. A 22 mm2 coverslip was placed on top of the agarose and sealed in place with vaspar. Samples were viewed on a Zeiss Laser Scan Microscope Model 2014 with a mercury bulb. Fields were viewed with a 60x Plan-Neofluar objective lens. AO imaging was done at an excitation wavelength of 450-490 nm with a 510 nm dichroic mirror and a long pass (>520 nm) barrier filter. Image collection and analysis were performed with a Hamamatsu color chilled 3CCD camera system. Laboratory Microcosms. Microcosms and related media were prepared using standard procedures previously described (8, 22, 36). Several treatments received pulverized dolomite (22) [60 mg/mL] as the potential electron donor. The dolomite was treated in two different ways prior to being added to the microcosms: autoclaved or autoclaved and then combusted at 500 °C for 16 h. The bottles were then sealed with aluminum crimp/Teflon-coated butyl rubber septa (Wheaton, Millville, NJ) and autoclaved at 120 °C for 20 min. The bottle headspaces were flushed with sterile 70% N2-30% CO2 (Matheson Gas Products, Inc., Secaucus, NJ) and were amended (0.1%) with sterile, anaerobic Na2S‚9H2O,
TABLE 1. Geochemical Characteristics of Well Watersa,d well
TCE (µg/L)
cis-DCE (µg/L)
VC (µg/L)
ethene (µg/L)
H2 (nM)
alkalinity as CaCO3 (mg/L)
δ 13C DIC
DOC (mg/L)
δ 13C DOC
methane (µg/L)
δ 13C methane
inside plume
87-02 87-12 89-05
310 32 4.6
310 1400 270
3.7 152
