Correspondence: Ivonne Nijenhuis, Department of Isotope Biogeochemistry, Helmholtz Centre for Environmental Research – UFZ, Permoserstrasse 15, 04318 Leipzig, Germany. Tel.: +49 341 235 1356; fax: +49 341 235 1443; e-mail: firstname.lastname@example.org
A Dehalococcoides-dominated culture coupling reductive dechlorination of tetrachloroethene (PCE) to ethene to growth was enriched from a European field site for the first time. Microcosms were set up using groundwater from a chlorinated ethene-contaminated anaerobic aquifer in Bitterfeld (Germany). Active, lactate-amended microcosms capable of PCE dechlorination to ethene without the accumulation of intermediates were used for further enrichment. After three transfers on lactate as an electron donor and PCE as an electron acceptor, the enrichment was transferred to parallel cultures with one of the chlorinated ethenes as an electron acceptor and acetate and hydrogen as the carbon and energy source, respectively. After three more transfers, a highly purified culture was derived that was capable of dechlorinating PCE with hydrogen and acetate as the electron donor and carbon source, respectively. PCR, followed by denaturing gradient gel electrophoresis, cloning and sequencing revealed that this culture was dominated by a Dehalococcoides sp. belonging to the Pinellas group. Investigation of substrate specificity in the parallel cultures suggested the presence of a novel Dehalococcoides that can couple all dechlorination steps, from PCE to ethene, to energy conservation. Quantitative real-time PCR confirmed growth with PCE, cis-dichloroethene, 1,1-dichloroethene or vinyl chloride as electron acceptors. The culture was designated BTF08 due to its origin in Bitterfeld.
Microorganisms can use chlorinated ethenes as electron acceptors and couple reductive dechlorination with energy conservation. Various electron donors, for example butyrate, propionate, lactate, acetate or hydrogen, can support growth, depending on the strains involved in dechlorination as well as on the mixed microbial community present in situ (He et al., 2002; Freeborn et al., 2005). Diverse microbial strains, belonging to Deltaproteobacteria, Gammaproteobacteria and Firmicutes, can reductively dechlorinate chlorinated ethenes. However, most of them dehalogenate PCE and TCE only to cis-DCE (Holliger et al., 1998; Löffler et al., 2003; Smidt & de Vos, 2004). The organisms capable of complete detoxification of PCE and TCE to ethene are, thus far, restricted to one phylogenetic group of bacteria, the ‘Dehalococcoides’ genus within the Chloroflexi phylum (green nonsulfur bacteria) (Maymó-Gatell et al., 1997).
Reductive dechlorination is used at chloroethene-contaminated sites in bioremediation approaches such as monitored natural attenuation, biostimulation and bioaugmentation. Nevertheless, these remediation strategies often result in the accumulation of intermediate dechlorination products including VC, which is a known human carcinogen and is thus of particular concern (Bradley, 2000; European Parliament, 2006). As Dehalococcoides play a crucial role in the complete detoxification of chlorinated ethenes, they draw the attention of not only the scientific community but also of bioremediation companies. Thus far, numerous strains have been characterized that are capable of dehalogenation of diverse chlorinated environmental pollutants. For example, Dehalococcoides sp. strains BAV1, GT and VS dechlorinate chlorinated ethenes to ethene, strain CBDB1 dehalogenates chlorinated benzenes and Dehalococcoides ethenogenes strain 195 was found to have an even broader substrate spectrum including chlorinated ethenes, benzenes, dioxins and phenols (Maymó-Gatell et al., 1997; He et al., 2003a, 2005; Fennell et al., 2004; Sung et al., 2006; Adrian et al., 2007; Müller et al., 2004).
Recently, Imfeld et al. (2008) and Nijenhuis et al. (2007) investigated a microbial community potentially involved in the reductive dechlorination in Bitterfeld and detected organisms belonging to the genera Dehalobacter, Desulfuromonas, Desulfitobacterium and Dehalococcoides. In addition, Nijenhuis et al. (2007) retrieved the Dehalococcoides sequences from groundwater and transferred cultures derived from Bitterfeld. Interestingly, the sequence analysis revealed that the Bitterfeld Dehalococcoides population was dominated by only one type of microorganism that shared 99% identity with Dehalococcoides strain BAV1. Strain BAV1 is one of very few isolates capable of fast metabolic conversion of DCEs and VC to ethene (He et al., 2003a). Therefore, Nijenhuis et al. (2007) suggested that the Dehalococcoides ribotype detected in Bitterfeld could be the key organism responsible for the complete detoxification of chloroethenes at this site.
The aim of the study presented here was to enrich the microbial strains responsible for the conversion of PCE to ethene at the contaminated site in Bitterfeld using standard liquid enrichment strategies with chlorinated ethenes as electron acceptors and lactate or hydrogen as electron donors. Active cultures were transferred several times and the changes in the microbial community in subsequent enrichments were studied using denaturing gradient gel electrophoresis (DGGE) and clone libraries. Furthermore, the substrate range of enrichment cultures and the growth of the enriched Dehalococcoides ribotype were assessed.
Materials and methods
All chemicals were purchased from Fluka (Seelze, Germany), Sigma-Aldrich Chemie (Seelze, Germany) or Merck (Darmstadt, Germany) at the highest purity available. Gases were purchased from Airproducts (Hattingen, Germany).
GC (Varian Chrompack CP-3800, Middelburg, the Netherlands) with a flame ionization detector (GC-FID) equipped with a 30 m × 0.53 mm GS-Q column (J&W Scientific, Waldbronn, Germany) was used to analyze concentrations of the chlorinated ethenes and ethene. The temperature program used was as follows: 1 min at 100 °C, 50 °C min−1 to 225 °C and hold for 2.5 min. The FID was operated at 250 °C and helium was used as a carrier gas (0.69 × 105 Pa; 11.5 mL min−1). This method allowed for the separation of ethene, VC, 1,1-, trans- and cis-DCE, TCE and PCE. The analysis was automated using an HP 7694 headspace autosampler (Hewlett Packard, Palo Alto) and 0.5-mL headspace samples were added to 10-mL autosampler vials that were flushed with helium, closed with Teflon®-coated butyl rubber septa and crimped.
Microcosm and enrichment culture preparation
Groundwater samples were collected from an aquifer contaminated with chlorinated ethenes located in Bitterfeld (Eastern Germany). Microcosms were prepared as described by Nijenhuis et al. (2007). In brief, 120-mL bottles were filled with 100 mL of groundwater, closed with Teflon-coated butyl rubber septa (Wheaton Science Products, Millville, NJ) and crimped. Lactate (3 mM added as an anoxic aqueous solution) was used as the electron donor and PCE (100 μmol L−1 added neat) was the electron acceptor. Each microcosm was additionally amended with anoxic aqueous solutions of resazurin (1 mg L−1), yeast extract (25 mg L−1) and vitamin B12 (62.5 μg L−1). Active microcosms were further transferred according to the enrichment scheme shown in Fig. 1.
In the first enrichment phase (Fig. 1a), active microcosms were transferred three times into mineral medium described by Zinder (1998) and amended with PCE as an electron acceptor and lactate as an electron donor and carbon source. The first transfer (0.1) was prepared by a 10% dilution of the original microcosm (0) in 27-mL tubes filled with 10 mL of medium. For the second transfer (0.2), 5% inoculum (0.1) was used. The third transfer (0.3) was prepared by a 1% dilution of the previous transfer (0.2) in 120-mL bottles filled with 50 mL of medium.
In the second enrichment phase (Fig. 1b), enrichment culture (0.3) was transferred on different electron acceptors. Cultures were prepared in 50-mL glass bottles, filled with 25 mL of medium (Zinder, 1998) and were amended with NaHCO3 (1 g L−1), Na2S (25 mg L−1) and vitamins (Zinder, 1998). Cultures received either PCE, TCE, cis-DCE, trans-DCE, 1,1-DCE or VC as an electron acceptor (10 μmol, as pure solvent or gas) and were inoculated (4% v/v). Two series of enrichment cultures were prepared. In the first series, acetate (3 mM) served as the carbon source and hydrogen (overpressure) served as the electron donor. In the second series, lactate (4 mM) was used as an electron donor and carbon source. Cultures were incubated at 20 °C without shaking.
Active cultures were further transferred on the same electron acceptor, electron donor and carbon source several times (Fig. 1b). Additionally, the second transfer of the VC/lactate culture (culture 12.2) was retransferred on VC, acetate and hydrogen, as the initial transfer (6.1) to this electron donor and acceptor was not active (Fig. 1c).
The third transfer (1.3) enriched on PCE, acetate and hydrogen received three additional doses of PCE (a total of 50 μmol) and was used as the inoculum in the ‘substrate specificity experiment’, where it was subsequently retransferred on to all chlorinated ethenes (PCE, TCE, cis-DCE, trans-DCE, 1,1-DCE or VC). The medium was prepared as described above (second enrichment phase) and hydrogen and acetate were added as an electron donor and a carbon source, respectively. Cultures were prepared by a 4% dilution of culture 1.3 (PCE/acetate+H2) (Fig. 1b). All treatments were set up in triplicate. Negative controls (not inoculated treatments) were prepared in triplicate for each chlorinated compound. The concentrations of chlorinated ethenes were analyzed by GC-FID in headspace samples taken at the start of the experiment and at subsequent dechlorination points.
DNA extraction and PCR amplification
Bacterial biomass was obtained from the liquid culture by centrifugation at 16 100 g for 30 min. DNA was extracted from the pellet using the DNeasy Tissue Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions for Gram-positive cells. DNA was eluted in 60 μL of RNAse-free distilled water. 16S rRNA genes were PCR amplified using HotStar Taq polymerase (Qiagen) and ‘universal’ bacterial primers 27f (Lane, 1991) and 1378r (Heuer et al., 1997). The thermocycler program was as follows: initial denaturation at 95 °C for 15 min, followed by 32 cycles of primer annealing at 51 °C for 30 s, chain extension at 72 °C for 50 s, denaturation at 95 °C for 30 s and a final extension at 72 °C for 30 min. The second semi-nested PCR was performed with bacterial primers GC968f (Nübel et al., 1996) and 1378r to obtain an amplicon size appropriate for DGGE analysis. The following thermocycling program was used: initial denaturation at 95 °C for 15 min, followed by 30 cycles of primer annealing at 55 °C for 1 min, chain extension for 1 min at 72 °C, denaturation for 30 s at 95 °C and a final extension at 72 °C for 30 min.
A DGGE analysis of the 16S rRNA gene PCR products was performed as described by Imfeld et al. (2008) using the DCode Universal Mutation Detection System (Bio-Rad, Munich, Germany). The PCR products were directly applied to an 8% (w/v) acryl-bisacrylamide gel (37.5 : 1, Merck) with a 30–60% linear urea/formamide denaturing gradient (7 M urea and 40% formamide (v/v) as 100% denaturants). The PCR products were separated by electrophoresis for 5.5 h at a constant temperature (60 °C) in 1 × TAE buffer at 200 V. The gels were stained for 20 min in 0.01% Sybr Green I (Molecular Probes, Leiden, the Netherlands) in 1 × TAE solution at room temperature. DGGE patterns were visualized by UV excitation and recorded using the Chemi Doc system and quantity one software (Bio-Rad). DNA retrieval from the distinct bands was performed as described earlier (Nikolausz et al., 2005). DNA was reamplified with the original PCR primers used for the DGGE and sequenced with the reverse primer as described below.
Construction of 16S rRNA gene clone libraries
PCR products amplified from genomic DNA with primers 27f and 1378r were purified using Qiaquick spin columns (Qiagen) according to the manufacturer's protocol. Cloning was performed using the pGEM-T Promega Cloning Kit (Promega, Madison, WI). Recombinant plasmids were extracted by boiling (5 min at 98 °C) an inoculation loop of bacterial cells in 50 μL water and pelleting the debris by centrifugation (2 min at 15 000 g). The supernatant was transferred into fresh tubes. The inserts were further amplified using M13(−20) and M13rev primers, screened and grouped by amplified ribosomal DNA restriction analysis (ARDRA) using the restriction enzymes HinP1I and BsuRI (Fermentas, Lithuania, Vilnius) as described earlier (Nikolausz et al., 2004). Clone libraries were prepared for cultures amended with acetate and H2 and PCE (1.1, 1.2, 1.3), TCE (2.3), cis-DCE (3.3) or 1,1-DCE (4.3) and lactate and VC (12.3) as indicated in Fig. 1 and Supporting Information, Table S2. A total of 20, 47, 49, 19, 19, 20 or 17 clones were picked for the clone libraries of cultures 1.1, 1.2, 1.3, 2.3, 3.3, 4.3 or 12.3, respectively.
Determination of nucleotide sequences
The partial sequences of the 16S rRNA genes from unique ARDRA types were obtained using the PCR primer 1378r. For each clone library, except for cultures 1.4 and 12.3, three clones representing the Dehalococcoides like ARDRA type were sequenced. For cultures 1.4 and 12.3, four and one clones were sequenced, respectively. The primers used for the almost complete 16S rRNA gene sequence determination were 27f, 1378r, 907r, 1114f, 533f, 803f (Lane, 1991) and 338R (Amann & Ludwig, 2000). The sequencing reactions were performed using the Big Dye Terminator Cycle Sequencing Kit V3.1 (Applied Biosystems, Foster City, CA), according to the manufacturer's protocol. Sequencing products were separated on a Model 3100 Genetic Analyzer (Applied Biosystems). The sequence alignment was performed using the multalign online software tool (Corpet, 1988). Analyses of the sequences and homology searches were performed using the blast algorithm with the BLAST server on the National Centre for Biotechnology Information (Altschul et al., 1998). Additional phylogenetic analyses were conducted using mega4 software (Tamura et al., 2007). The DNA sequences obtained in this study were submitted to GenBank and have accession numbers AM981291–AM981298.
Single-nucleotide primer extension (SNuPE)
A simple SNuPE assay was developed to investigate two variable positions of the Dehalococcoides-type sequences (Fig. S1). DhcPMT1 (5′-TTTACTGCCCCGCGAAACGG-3′) and DhcPMT2 (5′-TTTTTGACAGAACAATAGGTTGCAA-3′) primers were designed to anneal upstream from variable positions 1 and 2, respectively. (Bold letters indicate the non-complementary tails added as mobility modifiers. The role of these tails is described in Fig. S1.) The DhcPMT1 primer extension results in G incorporation into the dominant sequence type and an A extension in sequence variant 1, while DhcPMT2 results in C incorporation into the dominant sequence type and a T extension in sequence variant 2 (Fig. S1).
Cyclic primer extension reactions were performed as described before (Nikolausz et al., 2008), with a slight modification. Briefly, SNuPE reactions were carried out in a final volume of a 10-μL reaction mixture containing 1.8 μL 5 × sequencing buffer (Applied Biosystems), 1.2 μL SNaPshot multiplex kit reagent (Applied Biosystems), 4 μL of purified PCR products, 1 μL primer solution or primer mixture (10 μM of each primer) and 2 μL dH2O. The SNuPE reactions were carried out with 35 cycles of denaturation at 96 °C for 10 s, annealing at 55 °C for 5 s and extension at 60 °C for 30 s. In order to remove unincorporated ddNTPs, 1 U of shrimp alkaline phosphatase was added to each reaction and reactions were incubated at 37 °C for 1 h and then at 75 °C for 15 min. SNuPE reactions were run in duplicate to ensure reproducibility. Half microliter of post-treated extension products was mixed with 9 μL formamide and 0.5 μL GeneScan-120 LIZ internal size standard (Applied Biosystems). The mixture was denatured at 95 °C for 5 min and quickly cooled on ice. DNA fragment separation was performed on an ABI PRISM 3100 Genetic Analyzer using a 36-cm capillary filled with a denaturing POP6 polymer with filter set E5 (Applied Biosystems) (Nikolausz et al., 2008).
The growth of the culture on all chlorinated ethenes (PCE, TCE, cis-DCE, trans-DCE, 1,1-DCE and VC) was tested in separate experiments. The cultures were prepared as described above using hydrogen and acetate as an electron donor and a carbon source, respectively. The fourth transfer (1.4) enriched on PCE, acetate and hydrogen served as an inoculum (4% v/v). Cultures amended with PCE and VC were prepared in triplicate, while all the others were prepared in duplicate. Negative controls (uninoculated treatments) were prepared in duplicate for each chlorinated compound separately. Two types of living controls were set up in duplicate: (1) with PCE as an electron acceptor and no donor and carbon source and (2) without an electron acceptor, but with hydrogen and acetate as an electron donor and carbon source.
Dechlorination products were measured by GC-FID analysis, as described above. Growth was quantified by quantitative real-time PCR (qPCR) and samples were taken at the beginning of the experiment and at subsequent dechlorination points. One milliliter of sample was centrifuged for 30 min at 16 100 g (4 °C) and the pellet was resuspended in 500 μL of molecular-grade distilled water. After 15 min of centrifugation at 16 100 g (4 °C), the supernatant was discarded and the pellet was stored at −80 °C until DNA extraction.
Dehalococcoides 16S rRNA gene copies were quantified by qPCR using a Rotor-Gene 3000 (Corbett-Research, Qiagen Benelux, Venlo, the Netherlands) and the ABsolute™ QPCR SYBR Green Mix (ABgene, Leusden, the Netherlands) with the Dehalococcoides-specific primers Dco728F (5′-AAGGCGGTTTTCTAGGTTGTCAC-3′) and Dco944R (5′-CTTCATGCATGTCAAAT-3′) (Smits et al., 2004). The following thermocycler program was applied: initial denaturation at 95 °C for 15 min, 40 cycles of 10 s at 95 °C, 20 s at 50 °C and 20 s at 72 °C, with a final extension step at 72 °C for 5 min. The qPCR was performed in 15-μL reaction volumes in duplicate. Calibration of the qPCR was performed as described by Dijk et al. (2008). Previously cloned 16S rRNA gene PCR amplicons of D. ethenogenes strain 195 served as a template for the preparation of the standard curve.
Enrichment of the BTF culture
The sampling campaign in Bitterfeld took place in February 2003. Microcosms were prepared from groundwater derived from well BVV3051 located at the fringe of a PCE and TCE plume, but with high DCE concentrations (Nijenhuis et al., 2007). In enrichment phase I, all transfers were made on PCE as an electron acceptor and lactate as an electron donor (see Fig. 1a). Duplicate microcosms became active after several months. A first (0.1) transfer was performed in September 2003, and in February 2004, a second (0.2) transfer was performed. In culture 0.2, dechlorination activity started after <1 month and PCE was dehalogenated to ethene without the accumulation of intermediate products. Taxon-specific 16S rRNA gene-based PCR amplification indicated the presence of Dehalococcoides, Dehalobacter and Desulfuromonas in culture 0.2. The amplification signal for Dehalobacter was very weak (Nijenhuis et al., 2007; Imfeld et al., 2008). Culture 0.2 was transferred to larger vials in March 2004 (transfer 0.3) to produce more biomass for further investigations.
The second enrichment phase was targeted for the enrichment and isolation of the Dehalococcoides sp. detected in the PCE/lactate enrichment culture. Two enrichment series were prepared using chlorinated ethenes (PCE, TCE, cis-, trans-, 1,1-DCE or VC) as electron acceptors and two different electron donors and carbon sources (Fig. 1b). The first series aimed to enrich the Dehalococcoides; therefore, hydrogen and acetate were used as an electron donor and carbon source. The second series was amended with lactate – the electron donor used in the first enrichment phase. The third transfer culture (0.3) from the first enrichment phase was used as the inoculum for all treatments.
After 40 days, dechlorination was observed in the following cultures: cis-DCE/H2+acetate (3.1), cis-DCE/lactate (9.1) and VC/lactate (12.1). Within the next 15 days, dechlorination products were detected in other cultures: TCE/H2+actetate (2.1), 1,1-DCE/H2+actetate (4.1), TCE/lactate (8.1) and 1,1-DCE/lactate (10.1). Treatments with PCE as an electron acceptor (1.1 and 7.1) only became active after 80 days. Cultures amended with trans-DCE (5.1 and 11.1) as well as cultures with VC/H2+acetate (6.1) did not become active over a 2-year period. As the VC/H2+acetate treatment was not active, we used the VC/lactate treatment (12.1) as inoculums for a VC/H2+acetate (13.1) culture. The latter culture became active and was further transferred on VC/H2+acetate (Fig. 1c). All active cultures from the first enrichment series – PCE/H2+acetate (1.1), TCE/H2+acetate (2.1), cis-DCE/H2+acetate (3.1) and 1,1-DCE/H2+acetate (4.1) – were successfully transferred on the same combination of substrates at least three times (Fig. 1b).
The third transfer of the culture with PCE/H2+acetate (1.3) was transferred on all chlorinated ethenes (PCE, TCE, cis-, trans-, 1,1-DCE or VC) to verify that the culture maintained the ability to carry out all dechlorination reactions. In this experiment, treatments amended with PCE, TCE, cis-DCE and VC exhibited dechlorination activity. These four chlorinated ethenes were degraded with similar efficiencies; 300–600 μmol L−1 of each were dechlorinated to ethene in approximately 100 days (Fig. 2). Systematic losses of PCE and TCE (Fig. 2a and b) were observed during the experiment. Similar losses were also detected in control treatments, suggesting that they might be caused by sampling or adsorption. Only about 30% of the PCE and 40% of the TCE added to the flasks remained after approximately 100 days in sterile controls (Fig. 2a and b). For cis-DCE and VC, the losses were much lower and approximately 60% of added cis-DCE and 75% of VC was still present after approximately 100 days in the negative controls (Fig. 2c and d). In all negative controls, the observed losses did not coincide with product formation. No dechlorination products were detected in the cultures in which 1,1-DCE and trans-DCE were used as electron acceptors.
Several cultures, mainly from the first and second transfers from enrichment phase II (Fig. 1b), were subjected to DGGE analysis of 16S rRNA genes to assess the changes in the bacterial community throughout the experiment. DGGE analysis was used to identify the dominant microorganisms involved in reductive dechlorination as well as to compare the differences in the microbial community structure between the different treatments. The brightest bands in the denaturing gradient gel (labeled in Fig. 3) were excised, and the reamplified DNA fragments were partially sequenced and compared with known sequences using blastn (Table S2).
We observed a significant change in the community structure of the enrichment cultures compared with the original groundwater. Imfeld et al. (2008) carried out a DGGE analysis of the groundwater from well BVV3051 and detected 17 bands. All enrichment cultures investigated in this study were of very low complexity because, in most cases, only two bands per lane were detected. The most important differences were observed between the treatments with lactate and the treatments supplied with acetate and hydrogen. The brightest band in cultures growing on lactate corresponded to a Clostridium-like bacterium. In the cultures growing on acetate and hydrogen, the sequences obtained from the dominant bands were closely related to Dehalococcoides strain BAV1. The Clostridium band was not visible in the H2+acetate cultures, while the Dehalococcoides band could be seen in some lactate treatments, for example VC/lactate (12.3); however, the signal was very weak.
The DGGE separation patterns of cultures growing on various chlorinated ethenes were very similar, but the signals for Dehalococcoides were much stronger in the first transfer cultures (1.1, 2.1, 3.1) compared with the second transfer (1.2, 2.2, 3.2, 4.2). The higher content of DNA in the first transfer corresponds with the consumption of a larger dose of the potential electron acceptor. This suggests that Dehalococcoides strain BTF08 is capable of growth with PCE, TCE and cis-DCE. It was also observed that cultures from the second transfer lost some community members, which were still present in the first transfer, although with a very weak signal on the DGGE gel (see lanes 2.1 and 3.1). The cultures from the second transfer [PCE/H2+acetate (1.2), TCE/H2+acetate (2.2), cis-DCE/H2+acetate (3.2) and 1,1-DCE/H2+acetate (4.2)] contained only visible bands corresponding to Dehalococcoides.
Separation of the DGGE PCR products from the first transfer cultures, PCE/H2+acetate (1.1), TCE/H2+acetate (2.1) and cis-DCE/H2+acetate (3.1), yielded two different bands with a sequence very similar to that of Dehalococcoides strain BAV1 (Fig. 3, Table S1 bands A3–A4 and A5–A7). The bands in the lower part of the gel (Fig. 3, bands A5–A7) probably represent single-stranded DNA, a side-product produced due to the asymmetric nature of the PCR used for DGGE (Nikolausz et al., 2005; Zhang et al., 2005). The sequences of the bands labeled with letter A (A1–A7) were identical (100% identity for 360-bp fragments). Similarly, the separation of the DNA from treatments growing on lactate resulted in two Clostridium bands (see lane 12.3, bands B1 and B2). The sequences of all bands labeled with letter B (B1–B4) were identical (100% identity for 360-bp fragments).
Purity of the cultures assessed by 16S rRNA gene clone libraries
Clone libraries of 16S rRNA genes were constructed from several cultures to investigate culture purity and the structure of the community (Fig. 1). The clones were screened and grouped by ARDRA and representative members were identified by sequence analysis (Table S2).
All cultures amended with hydrogen as an electron donor were dominated by Dehalococcoides. The amount of clones representing minority populations varied from approximately 2% in the 1.2 (PCE/H2+acetate) and 1.3 (PCE/H2+acetate) cultures to 10% in the 2.3 (TCE/H2+acetate) and 4.3 (1,1-DCE/H2+acetate) cultures. Most organisms constituting the minor population and detected in enrichment cultures were closely related to organisms associated with other dechlorinating consortia. Sulfurospirillum multivorans (X82931) is a known dehalorespiring bacterium capable of dechlorination of PCE to cis-DCE (Scholz-Muramatsu et al., 1995). The uncultured bacterium clone IA-23 (AJ488074) was detected in a bacterial consortium dechlorinating chlorobenzenes; the uncultured spirochete clone KB-1 (AY780558) is associated with the chlorinated ethene-degrading culture KB-1 described by Duhamel & Edwards (2006). The uncultured bacterium clone TANB5 (AY667250) was found in TCE-contaminated aquifer by Macbeth et al. (2004).
Unlike cultures growing on chlorinated ethenes and hydrogen, culture 12.3, which was amended with VC and lactate, was dominated by a bacterium closely related to the uncultured bacterium clone TANB5 (99% identity) and Clostridium lactatifermentans (AY033434) (95% identity). Dehalococcoides clones comprised <6% of this culture (Fig. S1; Table S1).
Variability of sequences
Almost complete 16S rRNA gene sequences (approximately 1260 bp) were obtained from three to four randomly selected clones corresponding to the Dehalococcoides ARDRA pattern from each clone library. Random base changes were detected in 58% of the clones (11 out of 19) (data not shown). To examine whether the sequence variability reflected the natural diversity of Dehalococcoides sequences in the cultures or whether it was due to a method-introduced error, one culture, PCE/H2+acetate (1.3), was investigated further. The 219-bp fragments of the 16S rRNA gene where variations occurred were obtained for 10 additional clones. Sequences were produced with forward and reverse primers to exclude sequencing errors. Additionally, the original PCR product used for clone library formation was sequenced directly. In 23% of the screened clones (three out of 13), base pair changes at random positions, as compared with the sequence of the original PCR product and the dominant sequence type, were detected. A SNuPE assay (Nikolausz et al., 2009a) was developed to investigate the two variable positions in the Dehalococcoides-type sequences (Fig. S1). In this assay, the fluorescently labeled ddNTP incorporation provides information about the presence of the different sequence variants in a PCR product. The first microbial ecology application of primer extension with four differently labeled ddNTPs was carried out by Wu & Liu (2007) for the multiplex detection of different Bacteroides spp. Recently, Nikolausz et al. (2008) developed and applied an SNuPE assay for the detection and typing of Dehalococcoides spp. sequences obtained from chloroethene-contaminated groundwater samples. Our recent study demonstrated the excellent specificity of the incorporated labeled nucleotide analogues in SNuPE (Nikolausz et al., 2009b), which is associated with a good dynamic range and the detection of a minority template (Nikolausz et al., 2008).
When PCR products from the selected clones were used as templates, the expected incorporation of nucleotides was observed, in agreement with the sequencing data. The SNuPE analysis of the dominant sequence type resulted in the expected G and C incorporation with the assigned colors, blue and black, respectively (Fig. 4a). A (green)/C (black) and G (blue)/T (red) incorporation was observed when sequence variants 1 and 2 were used as templates for SNuPE, respectively (Fig. 4b and c). The SNuPE analysis of the PCR product used for the establishment of the clone library revealed the presence of only the dominant sequence type (Fig. 4d). We did not observe any traces of the two sequence variants represented by A and T incorporation. These sequence variants were most probably introduced by subsequent steps of cloning and PCR. This finding is in agreement with the observation of microvariation artifacts of cloning and PCR reamplification by Speksnijder et al. (2001).
Phylogenetic analysis of the dominant 16S rRNA gene sequence
In light of the results presented above, the variability of sequences was most probably introduced by the method. Therefore, the most abundant sequence was assumed to be the sequence representing the valid ribotype (AM981291) of Bitterfeld. The phylogenetic affiliation of the Dehalococcoides ribotype present in our culture is shown in Fig. 5. The almost complete 16S rRNA gene sequence of ribotype BTF08 has a 16S rRNA gene sequence identical to that of Dehalococcoides strain BAV1 (CP000688) and a highly similar sequence to strain CBDB1 (AJ965256). The sequence, consisting of 1293 bp, differs by 1 bp from the sequence of Dehalococcoides strain CBDB1. The sequence shares 1275 identical base pairs with the sequence of D. ethenogenes strain 195.
Growth-linked reductive dechlorination of chlorinated ethenes
16S rRNA gene-targeted qPCR confirmed the growth of strain BTF08 with PCE, cis-DCE, 1,1-DCE and VC as electron acceptors. Dechlorination of all four chlorinated ethenes was accompanied by an increase in 16S rRNA gene copies (Table 1). Dechlorination did not occur in the TCE- and trans-DCE-amended treatments, as well as in one of the three replicates amended with PCE. No increase in 16S rRNA gene copies was observed in inoculated controls lacking an electron acceptor, suggesting a dependence on chlorinated ethenes for growth.
Table 1. Growth of Dehalococcoides sp. in culture BTF08 on chlorinated ethenes as electron acceptors determined by qPCR
Cl− released (μmol L−1)
16S rRNA gene copies produced (× 107 mL−1)
16S rRNA gene copies produced μmol−1 Cl− released (× 107 mL−1 μmol−1 Cl− released)
PCE (cult. A) (1.2)
3.199 ± 0.393
0.316 ± 0.039
PCE (cult. B) (1.3)
4.158 ± 0.921
0.495 ± 0.110
cis-DCE (cult. A) (3.1)
3.914 ± 4.620
0.463 ± 0.547
cis-DCE (cult. B) (3.2)
5.179 ± 0.434
0.613 ± 0.051
1,1-DCE (cult. A) (5.1)
0.329 ± 0.045
0.328 ± 0.045
VC (cult. A) (6.1)
16.125 ± 1.537
2.853 ± 0.272
VC (cult. B) (6.2)
27.787 ± 2.649
3.351 ± 0.320
VC (cult. C) (6.3)
4.001 ± 0.919
0.585 ± 0.134
Figure 6 demonstrates the growth of strain BTF08 on cis-DCE, shown by the increase of the 16S rRNA gene copy numbers during dechlorination to ethene. Interestingly, dechlorination in this culture slowed down between days 106 and 151 and there was no evidence for further growth in this period. Most likely, the electron donor became depleted as both dechlorination and growth accelerated again after day 164, when hydrogen was added. This example suggests that the growth of strain BTF08 is also hydrogen dependent and that dechlorination cannot occur if this substrate is depleted.
The growth yields of the cultures that were active in the experiment are summarized in Table 1. They ranged from (0.32±0.04) to (3.35±0.32) × 107 16S rRNA gene copies μmol−1 Cl− released, for PCE and VC, respectively. The growth yields μmol−1 Cl− released of the cultures grown on different chlorinated ethenes were in the same order of magnitude, with the exception of the two VC-amended batches (Table 1, VC culture A and B). The average 16S rRNA gene copy number μmol−1 Cl− released, for all treatments excluding the latter two, was 0.47 × 107 gene copies μmol−1 Cl− released. Cultures A and B grown on VC as an electron acceptor yielded 2.85 × 107 and 3.35 × 107 16S rRNA gene copies μmol Cl− released, respectively, which is approximately five to six times more than the yield in the third VC-amended replicate with 0.59 × 107 16S rRNA gene copies μmol−1 Cl− released (Table 1, VC culture C). It is difficult to explain the difference in growth in three replicates of the same culture. In the majority of cultures, growth and dechlorination may have been uncoupled in a stationary phase as seen previously for D. ethenogenes strain 195 (Maymó-Gatell et al., 1997), or method-based errors (e.g. DNA extraction, PCR amplification) may have led to the observed differences.
We enriched a novel Dehalococcoides-containing culture that dechlorinates PCE to ethene from the contaminated anaerobic aquifer in Bitterfeld (Eastern Germany). Because of its origin, the culture was named BTF08. To our knowledge, this work is the first attempt to enrich a Dehalococcoides strain capable of the conversion of lower chlorinated ethenes from a European field site. Most currently known Dehalococcoides isolates were derived from various locations in North America (Maymó-Gatell et al., 1997; Sung et al., 2006). Culture BTF08 was enriched using a standard liquid enrichment procedure. Microcosms were prepared from groundwater derived from Bitterfeld and amended with lactate and PCE as an electron donor and acceptor, respectively. After the third transfer, the enrichment culture was amended with acetate and hydrogen instead of lactate to eliminate lactic acid fermenters and to promote the growth of Dehalococcoides. Three additional transfers on this combination of substrates yielded a culture that contained over 98%Dehalococcoides. The BTF08 culture grows in purely synthetic medium containing salts, bicarbonate as a pH buffer, trace elements, vitamins, (including B12), hydrogen as an electron donor, acetate as a carbon source, chlorinated ethenes as electron acceptors and sodium sulfide as a reducing agent.
Substrate specificity and growth were investigated in two independent experiments: the first experiment, referred to as the ‘substrate specificity experiment’, (for details, see Microcosm and enrichment culture preparation), and the second experiment, referred to as the ‘growth assay’ (for details, see Growth assay). In both experiments, the Dehalococcoides culture enriched on PCE as an electron acceptor was retransferred on different chlorinated ethenes, including PCE, TCE, cis-, trans-, 1,1-DCE and VC. PCE, cis-DCE and VC were dechlorinated in both experiments and growth on these compounds was confirmed by qPCR. Dechlorination of 1,1-DCE failed in the ‘substrate specificity experiment’, but occurred in the ‘growth assay’, where it was coupled to an increase in 16S rRNA gene copies. As the treatment amended with TCE as an electron acceptor did not show activity in the ‘growth assay’, direct evidence for growth-dependent reductive dechlorination of this compound could not be provided. However, several transfers cultivated with this substrate as the sole electron acceptor were active (Fig. 1b, transfers 2.1–2.4), as were the treatments inoculated with the PCE pregrown culture in the ‘substrate specificity experiment’ (Fig. 2b). It is rather unlikely that some of the transfers pregrown on PCE lost the ability to degrade 1,1-DCE or TCE as dechlorination of both compounds failed only in one experiment. Perhaps some nutrients necessary for growth were lacking in these treatments or the solvents were added in concentrations inhibitory for growth (Maymó-Gatell et al., 1997; Nijenhuis, 2002). Trans-DCE was not dechlorinated by any of the transfer cultures set up in this study. These findings support that PCE, TCE, cis-DCE 1,1-DCE and VC can be used as metabolic electron acceptors by culture BTF08.
The growth yields of the Dehalococcoides ribotype BTF08 ranged from (0.32±0.04) to (3.35±0.32) × 107 16S rRNA gene copies μmol−1 Cl− released, corresponding to the values reported in the literature for Dehalococcoides-containing cultures. For example, He et al. (2005) reported values of (7.3±0.2) to (8.4±0.8) × 107 16S rRNA gene copies μmol−1 Cl− released for a pure culture of strain FL2 growing with different chlorinated ethenes as electron acceptors. Another strain, BAV1, yielded (6.2±0.3) × 107 16S rRNA gene copies μmol−1 Cl− released, with VC as an electron acceptor (He et al., 2003b). The Dehalococcoides strains KB-1/VC-H2 and VS, present in enrichment cultures using VC as the electron acceptor, gave much better yields: (5.6±1.4) × 108 and (5.2±1.5) × 108 gene copies μmol−1 Cl−, respectively (Cupples et al., 2003; Duhamel et al., 2004). The yields measured in this study are similar to the values reported for strain FL2 and one to two orders of magnitude lower than the yields of Dehalococcoides strains present in the KB-1/VC-H2 and VS cultures. The growth of strain BTF08 on VC corresponds well with the yield obtained by He et al. (2003b) for BAV1.
The Dehalococcoides ribotype present in the enrichment culture is affiliated with the Pinellas group of Dehalococcoides, but it differs from the other isolates with regard to electron acceptor utilization. Although our ribotype has a 16S rRNA gene sequence identical to strain BAV1, it can, in contrast to strain BAV1, use the higher chlorinated ethene PCE as an electron acceptor. The 16S rRNA gene sequence of ribotype BTF08 is also highly similar to the sequence of strain CBDB1 (only 1 bp different), which grows on chlorinated aromatic compounds. However, the Pinellas group members and their dechlorination capabilities cannot be distinguished solely by 16S rRNA gene sequence analysis (Duhamel et al., 2002; He et al., 2003b, 2005; Ritalahti & Löffler, 2004).
While investigating the 16S rRNA gene sequences of the Dehalococcoides present in this culture, we observed that some Dehalococcoides-type sequences differed by 1 or 2 bp from the dominant Dehalococcoides sequence. To investigate this sequence variability, a SNuPE assay was developed and revealed that the variations in the Dehalococcoides sequences in the Bitterfeld enrichment culture were most likely due to an error introduced by the method. This finding is in agreement with the observation of microvariation artifacts of cloning and PCR reamplification (Speksnijder et al., 2001). However, it does not exclude the presence of other low-abundance Dehalococcoides-like bacteria in Bitterfeld and it warrants the careful interpretation of microheterogeneity results obtained using the cloning and sequencing approach.
The capability of culture BTF08 to degrade PCE to environmentally benign ethene and inorganic chloride is highly interesting, particularly for the bioremediation of chloroethene-contaminated sites. At many field sites, the biodegradation of PCE and TCE is incomplete and results in the production of cis-DCE and VC. Accumulation of these intermediate dechlorination products is a major problem in bioremediation as lower chlorinated ethenes are more toxic than PCE and TCE. In particular, VC is a known human carcinogen (ASTDR, 2007). Incomplete dechlorination is usually observed if the organisms that reductively dechlorinate the less chlorinated ethenes are not present or not active at the contaminated site. When requisite microorganisms are absent, ‘bioaugmentation’ can be implemented. In this cleanup approach, a reductively dechlorinating enrichment culture is added to the site (Harkness et al., 1999; Ellis et al., 2000; Major et al., 2002; Chartrand et al., 2005; Morrill et al., 2005). Like currently applied commercial consortia, such as BioDechlor INOCULUM, which contains Dehalococcoides strains FL2, BAV1 and GT (Ritalahti et al., 2005), or KB-1™ (Duhamel et al., 2004), our Dehalococcoides-containing enrichment culture, BTF08, also exhibits high potential for such an application.
We would like to thank our students Nathalie Perez, Marta Matusiak and Aaron Lee for assistance with the experiments and our technical assistants Ines Mäusezahl and Kerstin Ethner for help with laboratory work. We would like to thank Hans Richnow for critical comments and suggestions and are grateful to Gabriele Diekert for a discussion and helpful comments. We also would like to express our gratitude to Dirk Springael for enabling us to carry out the qPCR analyses in his laboratories and to Kelly Fletcher for her critical reading of the manuscript. The project was financially supported by a European Union Marie Curie Early Stage Training Fellowship (contract number MEST-CT-2004-8332) and by the Helmholtz Centre for Environmental Research – UFZ.