Environ. Sci. Technol. 2007, 41, 2261-2269
Comparative Analysis of Three Tetrachloroethene to Ethene Halorespiring Consortia Suggests Functional Redundancy R E B E C C A C . D A P R A T O , †,‡,⊥ FRANK E. LO ¨ F F L E R , †,§ A N D J O S E P H B . H U G H E S †,|,* School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, Department of Civil and Environmental Engineering, Rice University, Houston, Texas 77005, School of Biology, Georgia Institute of Technology, Atlanta, Georgia 30332, and School of Material Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332
Three anaerobic, dechlorinating consortia were enriched from different sites using methanol and tetrachloroethene (PCE) and maintained for approximately 3 years. These consortia were evaluated using chemical species analysis including distribution of dechlorination products, production of organic acids and methane, and using qualitative and quantitative PCR (qPCR), terminal restriction fragment length polymorphism (TRFLP), and denaturing gradient gel electrophoresis (DGGE) with primers specific to Dehalococcoides 16S rRNA gene sequences. TRFLP and analysis of organic acids revealed differing fermentative populations in each consortium, which were dominated by acetogens. Monitoring methane production combined with qPCR for archaea showed that complete dechlorination of PCE-toethene occurred in the presence and absence of methanogens. The 16S rRNA gene-based analyses demonstrated that enrichment with PCE resulted in dechlorinating communities dominated by Dehalococcoides and Dehalobacter, and that up to four different PCEdechlorinating organisms coexisted in one consortium. Further, the DGGE analysis suggested that at least one consortium contained multiple Dehalococcoides strains. The combined analysis of 16S rRNA and reductive dehalogenase genes suggested that one consortium contained a member of the Dehalococcoides “Cornell” group with the ability to respire VC.
Introduction Tetrachloroethene (PCE) and trichloroethene (TCE) are among the most abundant groundwater contaminants requiring remediation (1). Under anaerobic conditions, PCE * Corresponding author phone: (404) 894-2201; fax: (404) 8948266; e-mail:
[email protected]. † School of Civil and Environmental Engineering, Georgia Institute of Technology. ‡ Rice University. § School of Biology, Georgia Institute of Technology. | School of Material Science and Engineering, Georgia Institute of Technology. ⊥ Current address: Geosyntec Consultants, 6770 South Washington Ave., Titusville, Fl 32780. 10.1021/es061544p CCC: $37.00 Published on Web 02/27/2007
2007 American Chemical Society
can be reduced to nontoxic ethene via TCE, dichloroethenes (mainly cis-1,2-dichloroethene [cis-DCE]), and vinyl chloride (VC) by halorespiring microorganisms, and it is this terminal electron-accepting process that forms the basis of PCE and TCE bioremediation. Several halorespiring organisms dechlorinate PCE-to-cis-DCE, such as Dehalobacter restrictus (2), Sulfurospirillum multivorans (3), and Geobacter lovleyi (4), but for bioremediation to be successful, complete dechlorination to ethene must be achieved. To date, only two Dehalococcoides strains have been isolated that are capable of halorespiration of DCEs to ethene, while using hydrogen as their electron donor (4, 5). Harnessing the metabolic activity of halorespiring bacteria in contaminated aquifers is achieved through the introduction of electron donors that stimulate anaerobic organisms, which are capable of producing hydrogen. Hydrogen production is necessary because it is the only electron donor that Dehalococcoides can utilize. While direct hydrogen addition for the stimulation of Dehalococcoides and other halorespiring bacteria has been attempted (6), the injection of fermentable substrates (lactate, molasses, hydrogenreleasing compound (HRC), emulsified vegetable oil, chitin, etc.) is a more common approach (7-9). The anaerobic microbial community enriched by these fermentable substrates may include fermentative organisms that produce hydrogen and organic acids, dechlorinators that use the hydrogen (and in some cases organic acids), and other bacteria capable of utilizing hydrogen and organic acids produced (i.e., methanogens (10), sulfate reducers (11), and iron reducers (12)). Because appropriate community structure is presumed to play a critical role in the success or failure of complete dechlorination in bioremediation systems, considerable study has taken place regarding the diversity of halorespiring organisms and their metabolic disposition (13-17) and the flow of energy within anaerobic communities to the organisms capable of rapidly producing ethene (10, 18, 19). Recently, researchers have started examining the structure of anaerobic mixed cultures capable of complete TCE dechlorination and have focused mainly on the dechlorinating communities (20-22). Less information is available on the variations of fermentative community structure on dechlorination activity and the ability to rapidly achieve complete dechlorination of PCE (23). In the studies presented herein, we examine the community structure of three dechlorinating consortia originating from different sources and enriched identically over a period of 3 years. The common feature for each consortium was exposure to PCE/TCE prior to enrichment without significant ethene production. One consortium had been studied in our laboratory previously and produced VC as its primary dechlorination endpoint (24). The other two consortia were enriched from separate contaminated groundwater plumes where no ethene was detected. The initial objective was to determine if ethene formation could be established in all three consortia following laboratory enrichment. Subsequently, studies focused on correlations between activity and community structure in each consortium. Each consortium was enriched with PCE as the electron acceptor and methanol as a source of carbon and reducing equivalents. Maintenance under these conditions established rapid and complete dechlorination, indicating that the native microbiology was capable of complete detoxification. The examination of community structure and function was conducted through chemical analyses (product distribution for dechlorination, fermentation, and methanogenesis), comparing dechloriVOL. 41, NO. 7, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2261
nation activities using methanol and hydrogen as the electron donors, and molecular biology techniques including quantitative real-time PCR (qPCR), terminal restriction fragment length polymorphism (TRFLP), and denaturing gradient gel electrophoresis (DGGE). These comparisons suggest functional redundancy and that different populations contributed to fermentation, methanogenesis, and PCE-to-cis-DCE dechlorination, though all three consortia contained varying Dehalococcoides strains.
Materials and Methods Chemicals. The following chemicals were obtained in neat liquid form: PCE (99+% Sigma-Aldrich, St. Louis, MO), TCE (99.5%, Sigma-Aldrich), cis-DCE (97% Acros Organics, Morris Plains, NJ), and methanol (MeOH) (HPLC grade, Fisher Scientific). Sodium hydroxide (NaOH) (1 M) was obtained from Fisher Scientific. Gaseous chemicals obtained from Matheson Tri-gas (Parsippany, NJ) included VC (8% vol/vol, balance N2), ethene (99.5%), and methane (99%). Nitrogen (99.9%) and 80% H2/20% CO2 were obtained from Airgas (Atlanta, GA). Nutrient Medium. Reagent-grade chemicals and deionized water were used in the preparation of nutrient medium. The medium was prepared using concentrated stock solutions (10 mL each/L of medium) of Trace Element Solution I (50 mg/L ZnCl2, 50 mg/L MnCl2‚4H2O, 50 mg/L H3BO3, 250 mg/L CoCl2‚6H2O, 50 mg/L NiCl2‚6H2O, 50 mg/L Na2MoO4‚ 2H2O); Trace Element Solution II (1,000 mg/L (NaPO3)16, 250 mg/L KI, 50 mg/L NH4VO3), and Basal Salts (40 g/L KCl, 40 g/L MgCl2‚6H2O, 40 g/L NH4Cl, 14 g/L KH2PO4, 2.5 g/L CaCl2‚ 2H2O). Sodium bicarbonate (7-10 g/L) was added as a buffer, and 300 mg/L Na2S‚9H2O and 40 mg/L FeCl2‚4H2O were added to reduce the medium. Analytical Methods. Gas chromatography (GC) was used to determine the concentrations of all chlorinated ethenes, ethene and methane, using headspace analysis as described previously (24). Standards were prepared by adding known volumes of PCE, TCE, and cis-DCE dissolved in MeOH, and VC, ethene, and methane gases at known volumes, to a serum bottle (70 mL) containing deionized water (50 mL). Organic acids (citrate, malate, pyruvate, succinate, lactate, fumarate, formate, acetate, propionate, and butyrate) were quantified using high performance liquid chromatography (HPLC) as described previously (25). Aqueous samples (1.5 mL) were centrifuged; the supernatant was filtered (0.2 µm) and frozen at -20 °C. Samples (475 µL) were acidified by adding 25 µL of 1 M H2SO4. Calibration curves for the organic acids were prepared by a dilution series spanning a concentration range from 0.1 to 10 mM, except for fumarate, which was 0.01 to 1 mM. Bioreactor Design. Three draw-and-fill bioreactors were constructed using Nalgene low-density polyethylene carboys (two 25 L and one 30 L, Fisher Scientific). These reactors were fitted with polypropylene closures, which allowed for headspace sampling, liquid sampling, and addition of nutrient medium, PCE, MeOH, and NaOH. Bioreactors were filled with nutrient medium (10 L) and inoculated. One bioreactor (reactor “Owls”, OW) received an inoculum from a PCE-to-VC (80%) and ethene (20%)dechlorinating continuous flow reactor that had been established with sludge from a wastewater treatment facility (26). The other two bioreactors were seeded with contaminated groundwater from Nebraska (reactor “Cornhuskers”, CH) and Texas (reactor “Longhorns”, LN), respectively, and were obtained with the assistance of Groundwater Services Inc. (Houston, TX). The groundwater (15 L for CH and 10 L for LN) was pumped into the reactors using a peristaltic pump. The reactors were operated in a draw and fill mode with fluid taken out each day and replaced with an equal volume (250 mL/d for the 20-L reactors and 313 mL/d for the 2262
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 7, 2007
25-L reactor) of fresh medium, resulting in an 80-day retention time. Each consortium was fed 3 mmol/L-d of MeOH. The addition of PCE was initially low (0.05 mmol/ L-day) and increased to 0.25 mmol/L-day as the dechlorination performance increased. NaOH was added as necessary to maintain the pH in the reactors between 6.5 and 7.0. To prevent cross-contamination between the reactors, each reactor had its own equipment. Bioreactor Performance Monitoring. Microcosm studies were performed periodically to more precisely monitor dechlorination performance in each reactor. Serum bottles (70 mL) were sealed with Teflon-lined butyl rubber stoppers (Supelco, Bellefonte, PA) and aluminum crimp caps (Supelco) with a stir bar enclosed. The bottles were purged with N2 gas for 12 min to remove oxygen. Reactor samples (50 mL) were added to the serum bottles by syringe and sparged with N2 for 12 min to remove any residual volatile compounds. After sparging, neat PCE and MeOH were added with a gas-tight glass syringe to final concentrations of 0.2 mM and 3 mM, respectively. The microcosms were incubated on a stirrer at room temperature for 48 h before headspace samples were analyzed by GC. Microcosm studies were also used to compare dechlorination performance with MeOH and H2 as electron donors. The same procedure as above was used, but when H2 was used as the electron donor, the consortium was sparged with 80% H2/20% CO2 for 12 min. Parallel microcosms were constructed in triplicate with one set for monitoring chlorinated ethenes, ethene, and methane and the other set for monitoring organic acid production and consumption. Microcosms were incubated at room temperature. Long-Term Organic Acid Studies. To assess the fate of organic fermentation products, long-term microcosm studies (112 days) were performed. For this study, parallel microcosms were set up in triplicate: one set for monitoring organic acid production and the other set for monitoring methane production. The microcosms were sparged with N2 (12 min), MeOH was added (3 mM), but PCE was omitted, and the microcosms were incubated at room temperature. Organic acid production and utilization was monitored periodically for the entire 112 day incubation period. Methane production was measured only through day 35 because subsequent methane measurements were inaccurate due to gas leakage through the septa from repeated piercing due to the high number of sampling events. DNA Extraction and PCR. Culture fluid collected from the reactors (30 mL) was divided in two aliquots, placed into conical plastic tubes (15 mL), and centrifuged at 3250 × g for 1 h at 4 °C. The resulting pellet was used for DNA extraction with the Ultraclean Soil DNA Isolation Kit (Mo Bio Laboratories, Inc., Solana Beach, CA). The manufacturer’s protocol was followed, except that a Mini-beadbeater (Biospec Products, Bartsville, OK) was used instead of a vortex for cell lysis. The DNA was quantified spectrophotometrically at 260 nm, and the quality was verified on 1% agarose gels. PCR was performed in 20 µL reactions with 2.5 × Eppendorf PCR Mastermix (Brinkmann, Westbury, NY), which contained 1.25 U Taq polymerase and 200 µM of each deoxynucleoside triphosphate. For each reaction, 30-50 ng of template DNA was used. The primer sequences for the target 16S rRNA and reductive dehalogenase genes (RDases) are listed in Table 1. Nested PCR was also performed using the universal bacterial primers 8F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1541 R (5′-AAGGAGGTGATCCAGCCGCA-3′) in the initial PCR reaction, followed by a second PCR round with the dechlorinator 16S rRNA gene-targeted primer pairs (27). Appropriate negative and positive control reactions accompanied all PCR analyses. Instead of template DNA, the negative controls contained nuclease-free water (Fisher Scientific, Fair Lawn, NJ). Chro-
TABLE 1. Primers Used for Identifying Organisms and Reductive Dehalogenase Genes Present in Consortia target group
primer name
amplicon length (bp)
sequence
specificity
ref
Sulfurospirillum sp. strain JPD-1 Sulfurospirillum multivorans
JPDF JPDR Fp DHSPM 576 Rp DHSPM 1210
5′-CCCCATACTCCAACTTAT C-3′ 5′-TTCTAGGTGACCAGTTTCG-3′ 5′-GCTCTCGAAACTGGTTACCTA-3 5′-GTATCGCGTCTCTTTGTCCTA-3′
457
Sulfurospirillum sp. strain JPD-1 47
634
48
Desulfuromonas spp.
Desulf for Desulf rev
5′-AACCTTCGGGTCCTACTGTC-3′ 5′-CGGCAACTGACCCCTATGTT-3′
835
Desulfitobacterium spp. Desulfomonile spp. Geobacter lovleyi strain SZ pceA
Dd1 Dd2 Dt1 Dt2 Geo 169F Geo 999R SpSm1F SpSm1R
5′-AATACCGNATAAGCTTATCCC-3′ 5′-TAGCGATTCCGACTTCATGTTC-3′ 5′-CAAGTCGTACGAGAAACATATC-3′ 5′-GAAGAGGATCGTCTTTCCACGA-3′ 5′-GAATATGCTCCTGATTC-3′ 5′- ACCCTCTACTTTCATAG-3′ 5′-TCGTTGCAGGTATCGCTATG-3′ 5′-TTCAACAGCAAAGGCAACTG-3′
1199 995
Sulfurospirillum multivorans, Sulfurospirillum halorespirans strain PCE-M2, Campylobacter sp. DSM 806 Desulfuromonas michiganensis strain BB1, D. acetoxidans and D. chloroethenica Desulfitobacterium dehalogenans Desulfomonile tiedjei
830
Geobacter lovleyi strain SZ
51
194
pceA gene from Sulfurospirillum multivorans
52
mosomal DNA from Sulfurospirillum sp. strain JPD-1 (obtained from J. Ferguson, University of Washington), Sulfurospirillum multivorans, Desulfuromonas michiganensis strain BB1, Desulfitobacterium sp. strain Viet1, Desulfomonile tiedjei, and Geobacter lovleyi strain SZ were used as positive controls. The amplicons were visualized on a 1% agarose gel in Tris-acetate-EDTA buffer and stained in an ethidium bromide solution (1 µg/mL). The specificity of the primers used in this work was verified using BLAST (http://www.gl.iit.edu/frame/genbank.htm). Table 1 provides information about the primers and their targets. The Sulfurospirillum multivorans targeted primers also amplify closely related organisms (e.g., Campylobacter sp. DSM 806) that were not described to dechlorinate PCE (28). Quantitative Real-Time PCR (qPCR). The numbers of archaea, Dehalococcoides, Dehalobacter, and the gene copies of tceA, bvcA, and vcrA in each consortium were estimated using qPCR. The qPCR reactions for Dehalococcoides 16S rRNA genes and the vcrA, bvcA, and tceA genes were all performed as described by Ritalahti et al. (29). Archaeal cells were enumerated using the method developed by Da Silva et al. (30). Dehalobacter 16S rRNA genes were quantified using primers described by Smits et al. (31). PCR conditions for enumerating Dehalobacter 16S rRNA genes were modified as follows: 2 min at 50 °C, 15 min at 95 °C, followed by 40 cycles of 15 s at 94 °C and 1 min at 58 °C and 30 s at 72 °C. The qPCR reactions (30 µL) contained 1 × SYBR Green PCR Master Mix (Qiagen), forward primer and reverse primer (300 nM each), and DNA template (3 µL). qPCR was carried out in a spectrofluorimetric thermal cycler (ABI Prism 7500 sequence detection system, Applied Biosystems). Standard curves for qPCR were prepared with a dilution series of quantified genomic DNA from Dehalobacter restrictus (DSMZ 9455), Methanococcus maripaludis (ATCC 43000), and Dehalococcoides sp. strain BAV1 or from plasmids carrying a single Dehalococcoides RDase gene (i.e., bvcA, tceA, vcrA). The linear range for quantification was 102-109 gene copies/µL of DNA (r2 ) 0.992), 103-108 gene copies/µL of DNA (r2 ) 0.990), and 102-109 gene copies/µL of DNA (r2 ) 0.996) for D. restrictus, M. maripaludis, and Dehalococcoides sp. strain BAV1, respectively (29). The linear range for the target RDases was 101-108 gene copies/µL of DNA for bvcA, tceA, vcrA (r2 > 0.975). The gene copy numbers were calculated as described by Ritalahti et al. (29). Genome and genomic analyses demonstrated that the 16S rRNA and the RDase genes exist as single copy genes on Dehalococcoides genomes (29, 32, 33) (http://genome.jgi-psf.org/draft_microbes/ deh_b/
49 50 50
deh_b.info.html). The numbers of 16S rRNA genes for D. restrictus and for M. maripaludis have not been determined. The closest relatives to M. maripaludis contain 1-4 copies of the 16S rRNA gene according to the Ribosomal RNA Operon Copy Number Database (http://rrndb.cme.msu.edu/) and our cell number estimates assumed two 16S rRNA gene copies per archaeal genome. Terminal Restriction Fragment Length Polymorphism (TRFLP). TRFLP with the restriction enzymes HhaI, MspI, and RsaI was used to monitor and compare the microbial community structure of the three consortia. DNA extraction was performed as described above, and the 16S rRNA genes were amplified using triplicate PCR reactions (100 µL each) with a HEX-labeled forward primer 8F and the unlabeled reverse primer 1541R (sequences given above). After amplification, the PCR products were purified using a QIAquick PCR purification kit (QIAGEN, Valencia, CA). The purified amplicons were recovered in 30 µL, and the triplicates were combined (90 µL). Enzyme digests (30 µL) were prepared with purified PCR products (200 ng), reaction buffer (3 µL), and 7.5 U of restriction enzyme. The digest was performed at 37 °C for 3 h, and then the enzymes were denatured at 68 °C for 25 min. TRFLP samples were analyzed at the University of Illinois Urbana-Champaign Core Sequencing Facility (http://www.biotech.uiuc.edu/ dnasequencing.html). Dehalococcoides sp. strain FL2 (34), Dehalobacter restrictus (DSM 9455), and Acetobacterium woodii (ATCC 29683) were examined by the above TRFLP protocol to verify the fragment size predicted from in silico digestion. To generate TRFLP profiles, each profile was normalized to the profile with the lowest total fluorescence (35). Denaturing Gradient Gel Electrophoresis (DGGE). DGGE with Dehalococcoides specific primers described by Duhamel et al. (36) was performed by Microbial Insights Inc. (Rockford, TN; http://www.microbe.com/). The 16S rRNA gene was amplified using DNA from pure cultures of Dehalococcoides sp. strain BAV1 (4), strain GT (5), and strain FL2 (34), from a highly enriched culture consisting predominantly of Dehalococcoides sp. strain VS (obtained from Alfred Spormann, Stanford University) (37)) and a plasmid containing a copy of the 16S rRNA gene from D. ethenogenes strain 195 (38). Further, 16S rRNA gene fragments were amplified using DNA from consortia OW, CH, and LN. PCR products (20 µL) were electrophoresed on an 8% polyacrylamide gel with a 4565% urea-formaldehyde gradient at 55 V and 60° C for 16 h. The fragments were excised from the gel and placed into nanopure water (50 µL). PCR was performed using the excised gel band (2 µL) with the same Dehalococcoides-specific DGGE VOL. 41, NO. 7, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2263
FIGURE 1. Culture specific dechlorination of 0.25 mM PCE with (a) 3 mM MeOH and (b) excess H2: (I) OW consortium; (II) CH consortium; (III) LN consortium. Symbols: 1 PCE, 9 TCE, 2 DCE, b VC, O ETH.
TABLE 2. Enrichment Consortia Description and Time Required during Enrichment Process for Production of cis-DCE, VC, and Ethene as Major End Products time to production as major end productb,c consortium
origin
time in operation
time to first ethene productionb
cis-DCE
VC
ethene
OW CH LN
enrichment culture contaminated site NEa contaminated site TXa
3 yrs 1 month 3 yrs 2 yrs 9 months
0 months 3 months 9 months
N/A 1 month 6 months
0 months 2 months 9 months
6 months 2 yrs 2 yrs 1 month
a Groundwater samples were used to enrich consortia. g97% of initial PCE mass.
b
Time was determined in 48 h microcosm studies. c Major end product is defined as
primers. The resulting PCR product was purified using the Ultraclean PCR cleanup kit (Mo Bio Laboratories, Inc.) and sequenced. These sequences were compared to other Dehalococcoides 16S rRNA gene sequences using BLAST.
Results The three anaerobic, dechlorinating consortia were enriched and maintained for 3 years. The time period required to establish robust dechlorination activity leading to ethene formation varied in consortia OW, CH, and LN (Table 2). Consortium OW was the first (
