Tetrachloromethane-Degrading Bacterial Enrichment Cultures and Isolates from a Contaminated Aquifer

Abstract: The prokaryotic community of a groundwater aquifer exposed to high concentrations of tetrachloromethane (CCl4) for more than three decades was followed by terminal restriction fragment length polymorphism (T-RFLP) during pump-and-treat remediation at the contamination source. Bacterial enrichments and isolates were obtained under selective anoxic conditions, and degraded 10 mg·L−1 CCl4, with less than 10% transient formation of chloroform. Dichloromethane and chloromethane were not detected. Several tetrachloromethane-degrading strains were isolated from these enrichments, including bacteria from the Klebsiella and Clostridium genera closely related to previously described CCl4 degrading bacteria, and strain TM1, assigned to the genus Pelosinus, for which this property was not yet described. Pelosinus sp. TM1, an oxygen-tolerant, Gram-positive bacterium with strictly anaerobic metabolism, excreted a thermostable metabolite into the culture medium that allowed extracellular CCl4 transformation. As estimated by T-RFLP, phylotypes of CCl4-degrading enrichment cultures represented less than 7%, and archaeal and Pelosinus strains less than 0.5% of the total prokaryotic groundwater community.


Introduction
Tetrachloromethane (carbon tetrachloride, CCl4) has been widely used for decades as industrial degreasing agent, pesticide, flame retardant and in dry cleaning [1]. Prolonged large-scale use of CCl4 has caused significant environmental soil and groundwater contamination. Because of its hazardous effects on human health [2] and its implication in depletion of atmospheric ozone, industrial production and use of CCl4 is regulated under the Montreal Protocol since 1987. Nevertheless, worldwide CCl4 emissions remain high (up to 39 Gg/year between 2000 and 2012 [3,4]). CCl4 is highly recalcitrant and shows the longest half-life for abiotic degradation among halogenated aliphatic compounds [5]. Microbial CCl4 degradation occurs exclusively under anoxic conditions [6], and can be mediated, often extracellularly, by many biomolecules including corrinoid [7,8], quinone [9,10] and humic acid [11] derivatives; cytochromes [12]; nickel-and zinc-containing porphinoids [13]; riboflavin [14]; and siderophores [15]. Biodegradation of CCl4 by taxonomically diverse, both strict and facultative anaerobic bacteria, as well as by methanogenic Archaea, was reported under a wide range of conditions including denitrification, sulfate reduction, iron reduction, fermentation, acetogenesis and methanogenesis [6,16]. CCl4-degrading microorganisms growing with CCl4 as the sole carbon source have not been reported, but several studies showed some incorporation of CCl4 carbon as biomass via CCl4 degradation intermediates or products [6,17,18]. Nevertheless, CCl4 transformation is highly toxic to degrading organisms, notably because of the formation of reactive free radicals upon dehalogenation, which lead to formation of stable deleterious adducts both in the cytoplasm and in the cell membrane [6]. Processes that would allow productive CCl4 metabolism without such toxic effects would be expected to confer a selective advantage to their host in CCl4-contaminated environments. Therefore, isolating new microorganisms able to degrade CCl4, and stimulating the development of such organisms, or adding selected microbial strains known for their sustained CCl4-degrading activity in contaminated environments, represent attractive goals for cost-effective in situ remediation of CCl4-contaminated sites [6,19].
The purpose of this study was to investigate the CCl4 biodegradation potential of the prokaryotic community from a sub-surface aquifer located in Alsace, France, which was heavily contaminated by an accidental spill of 4000 liters of CCl4 in 1970. Infiltration into the groundwater, which remained undetected for over 20 years, led to the gradual development of a 12-km long plume. Maximum CCl4 concentrations at the contamination source exceeded more than thousand-fold the WHO drinking water guideline value of 4 μg·L −1 . Physico-chemical "pump-and-treat" remediation at the contamination source was performed between August 2006 and June 2009, at which point operational costs became critical. Here, terminal restriction fragment length polymorphism (T-RFLP) [20] was used to identify changes in diversity in the microbial community over three sampling campaigns carried out at the pollution source over a 15-month period during pump-and-treat remediation, and to follow several CCl4-degrading enrichment cultures obtained from samples taken at the contamination source. This led to the isolation of the novel CCl4-degrading bacterial strain Pelosinus sp. TM1.

Chemicals
Unless otherwise stated, all chemicals and reagents were of analytical grade and obtained from Sigma-Aldrich Co. (Saint-Louis, MO, USA) or Thermo Fisher Scientific (Waltham, MA, USA).

Sampling of Groundwater and Concentration of Microorganisms
Groundwater from the studied CCl4-polluted site was collected in October 2007 (T1), April 2008 (T2) and December 2008 (T3), i.e., 14, 20 and 28 months after the start of "pump-and-treat". Sampling was performed at the location of the initial spill by diverting 100 (T1 and T2) or 1000 (T3) liters of pumped groundwater before physico-chemical remediation. pH and redox potentials were measured using a pH 330i pH-meter (WTW, Weilheim, Germany) equipped with a SenTix Mic or a SenTix ORP electrode for pH or Eh determination, respectively. Mineral composition was determined by groundwater analyses carried out by the Centre d'Analyses et de Recherches in Strasbourg. Samples were concentrated by tangential filtration using a Prep/Scale TFF cartridge with a molecular weight cut-off of 30 kDa (samples T1 and T2; Millipore, Billerica, MA, USA), or with a Sartocon slice ultrafiltration set equipped with 5 polyethersulfone membranes of 0.1 m 2 (sample T3; cut-off 30 kDa; Sartorius, Göttingen, Germany). Sample volumes were further reduced from 300-400 mL to 10-20 mL by Centricon Plus-70 centrifugal filter devices (T1 and T2; Millipore), or by centrifugation at 10,000 g for 30 min in an Optima L90K ultracentrifuge equipped with a fixed angle 45Ti rotor (T3; Beckman, Brea, CA, USA). Cell densities were estimated using a DM 4000B epifluorescence microscope (Leica, Wetzlar, Germany) and fluorescent (SYBR Green I) intracellular DNA staining as described [21].

Cultivation Methods
Three culture media with different carbon and energy sources and electron acceptors were used. MBM medium (containing (per L) 2 g yeast extract, 2 g casitone and 10 mL methanol) corresponds to Methanosarcina barkeri medium n°120a from the German culture collection DSMZ. C medium contains 10 g yeast extract and 0.5 g tryptone per L and was prepared as described [22]. Pseudomonas stutzeri KC [23] and Methanosarcina barkeri DSM 1538 [24] were used as positive controls for microbial degradation of CCl4. Simulated groundwater medium (SGW), modified from [25] to reduce the concentration of chloride, contained (per L of deionized water) 0.35 g Na2SiO3·5H2O, 0.16 g Na2CO3, 0.006 g Na2SO4, 0.02 g KOH, 0.15 g MgSO4·7H2O, 0.0125 g Ca(NO3)2·4H2O, 1.6 g Na2HPO4, 0.8 g KH2PO4 and 1 mL of trace elements solution. pH was adjusted to 7.5-7.8 with 10 M NaOH prior to autoclaving at 121 °C for 10 min. Individual salt solutions and trace elements were added separately as sterile 100× concentrated stock solutions after autoclaving. When used as solid medium, 15 g·L −1 bacteriological agar (Euromedex, Souffelweyersheim, France) was added. After sterilization, medium for anoxic growth was transferred into 500 mL glass infusion bottles (Glasgerätebau Ochs, Bovenden, Germany) sealed with Teflon-coated rubber stoppers (Apodan, København, Denmark), then flushed under a continuous flow of nitrogen (Linde Gas, Saint-Priest, France) for 1 h. After transfer to an anoxic glovebox (Jacomex, Saint-Priest, France), media were reduced by addition of 0.3 g·L −1 Na2S·9H2O and 0.3 g·L −1 L-cysteine·HCl. The redox indicator resazurin (in aqueous 1 mM solution; Riedel-de Haën, Germany) was added to liquid media at a final concentration of 1 μM.
Liquid enrichment cultures (5 mL) were carried out at 25 °C in 17 mL Hungate tubes sealed with Viton rubber stoppers (Glasgerätebau Ochs, Bovenden, Germany) in MBM, C and SGW media. Several CCl4 concentrations (0, 10 and 200 mg·L −1 ), carbon sources (methanol, acetate, lactate, succinate, glucose, yeast extract, casitone or tryptone), and electron acceptors (oxygen, nitrate, sulfate, bicarbonate or organic compounds) were tested (Table S1). Oxic incubations were performed in a Microtron rotary shaker at 100 rpm (Infors, Bottmingen, Switzerland). Cultures under anoxic conditions were prepared in an anoxic glove box in Hungate tubes that were degassed using a vacuum pump connected to a nitrogen flow circuit (H. Lüdi + Co., Regensdorf, Switzerland) before addition of O2-depleted medium. Carbon and energy sources and electron acceptors were added as 100× concentrated sterile stock solutions in water as summarized in Table S1. Saturated aqueous CCl4 solutions (800 mg·L −1 ) were prepared using ultrapure CCl4 (purity > 99.9%; Fluka) and added last to culture media. Inoculation of enrichment cultures was done with 100 μL aliquots of concentrated groundwater samples from campaigns T1 and T2, representing between 0.5 and 1.0 L of the initial groundwater samples. Growth was measured at 600 nm directly in Hungate culture tubes with a Libra S6 spectrophotometer equipped with a 16 mm culture tube adapter (Biochrom, UK). Enrichment cultures for selected samples were obtained under the same conditions by successive reinoculation with 100 μL of stationary phase cultures (OD 0.2-1.0).

Gas Chromatography
CCl4 was quantified using a CP 3800 gas chromatograph connected to a flame ionization detector (GC-FID; Varian, Palo Alto, CA, USA). Headspace aliquots (500 μL) were collected from sealed Hungate culture tubes with a gastight 1750 syringe (Hamilton, Franklin, MA, USA) and injected onto the GC column (CP-Sil 5 CB, length 15 m; Varian). Separation of volatile compounds was achieved by isothermal elution at 30 °C for 1 min, followed by a linear temperature gradient (20 °C min −1 ) up to 220 °C. Injector and detector were maintained at 220 °C (splitless mode) and 300 °C, respectively, with nitrogen (N2) as the make-up gas (Linde Gas). Peak areas were analyzed with Galaxie Workstation software (Varian) and expressed as percentage of CCl4 compared to an abiotic reference tube prepared under the same conditions.

Determination of CCl4 Degradation Rates and Chloride Production
Specific CCl4 degradation rates in growing cultures of dechlorinating enrichments and of strain TM1 were calculated by determining the amount of CCl4 degraded per day and per mg of total cellular protein upon culturing under the respective conditions described in Table S1. Average degradation rates were determined per day of culture, and normalized per unit of cellular protein (average of at least three measurements during growth by the method described below). Similarly, rates of CCl4 degradation and chloride production were determined in cell suspensions of strain TM1 grown in SGW medium. Cell cultures (50 mL OD600 0.15-0.20) were centrifuged at 10,000 g for 15 min, resuspended (equivalent to 0.9 mg total protein) in 5 mL 10× diluted SGW medium in 30 mL sealed anoxic flasks under a nitrogen atmosphere, and exposed to 65 μM (10 mg·L −1 ) CCl4 at 25 °C. Headspace samples were taken for GC measurements as described above, and 550 μL aliquots of the cell suspension were removed for determination of protein content and chloride concentration. Briefly, aliquots were spun down and an appropriate dilution of the supernatant analyzed by ion chromatography on an AS18-AG18 column (Dionex, Sunnyvale, CA, USA) eluted with 34 mM NaOH at 1 mL per minute. Protein measurements were performed on cell pellets resuspended in 0.1 M NaOH, lysed at 95 °C for 10 min and centrifuged at 12,000 rpm for 5 min. Aliquots of the resulting solutions were analyzed with the bicinchoninic acid kit (Sigma) following the manufacturer's protocol, using bovine serum albumin (BSA) as a standard. For analysis of CCl4 degradation in spent medium, culture supernatants were separated from cells of a 500 mL culture grown in SGW medium by centrifugation for 15 min at 7650 g in a Sorvall Evolution RC centrifuge equipped with a SS-34 rotor (Thermo Fisher Scientific). Supernatants were filter-sterilized using 0.22 μm cellulose acetate syringe filters (VWR, Fontenay-sous-Bois, France). Corresponding cell pellets were washed and resuspended in an equivalent volume of 50 mM phosphate buffer, pH 7.5. Heat treatment was performed by autoclaving (121 °C, 10 min), and oxygen-exposed samples were obtained by contact with air for 90 min under sterile conditions. Part of the oxygen-exposed supernatant fraction was cysteine/Na2S-treated by addition of 0.3 g·L −1 Na2S·9H2O and 0.3 g·L −1 L-cysteine·HCl. Degradation rates in supernatants were expressed as μg CCl4 day −1 per mL of bacterial culture, normalized for a culture at OD600 = 1.0.

Bacterial Phenotyping
Substrate utilization profiling of strain TM1 was performed under anaerobic conditions with an AN MicroPlate (Biolog, Hayward, CA, USA) according to the manufacturer's recommendations, except that incubation was set to 25 °C for 72 h in an anoxic glove box. Staining of spores was carried out as described [26], except that a fuchsine solution (basic fuchsine 10 g·L −1 , phenol 50 g·L −1 , ethanol 100 mL·L −1 ) was used for staining instead of 1% aqueous safranin.

DNA Purification
Total DNA from environmental samples or enrichment cultures was purified and isolated from 10 7 to 10 9 cells trapped in agarose plugs [20]. DNA concentrations were estimated with a NanoDrop ND-1000 (Thermo Fisher Scientific). DNA from single colonies on solid media was extracted from approximately 10 8 cells suspended in 40 μL sterile ultrapure water or NaOH 0.05 M, and heated at 95 °C for 10 min. The resulting solution was cooled on ice for 5 min, diluted (1/15) in sterile ultrapure water and used as template for PCR amplification.

T-RFLP Analysis
PCR reaction mixtures, cycles and primers for bacterial T-RFLP were previously described [20]. Primers 6-FAM-21f and 958r [27] were used for T-RFLP analysis of the archaeal 16S rRNA gene. Preparation of PCR-amplified 0.9 kb 16S rRNA gene fragments and T-RFLP analysis were previously described [20]. The T-RF database was constructed using the "Microbial community analysis III" interface (MiCA III; [28]), with the following parameters: primers 27f and 926r, restriction enzymes AluI and HhaI, RDP release 10 update 27 (700,829 good quality bacterial SSU 16S rRNA fragments), and 10 mismatches allowed within 15 bases from 5′ end of primers. Data obtained with different fluorochromes for internal marker and sample fragments permitted determination of the total number of T-RFs and their relative proportion. The peak integration threshold was set to 500 ppm of the total integrated fluorescence. Bacterial diversity was evaluated using the Shannon diversity index (H′) [29], as follows: where pi is the relative area of T-RFLP peak i and S is the total number of integrated peaks (richness) in the sample. The distribution of T-RFLP peak areas was also evaluated using a measure of evenness (E): The maximum value E = 1.0 would describe a sample with equal frequencies of all peaks in its T-RFLP profile.

Denaturing High Performance Liquid Chromatography (D-HPLC) Analysis
D-HPLC was performed as described [20], using a gradient of 58% to 67% buffer B in 18 min, at a temperature of 59.5 °C and a flow rate of 0.45 mL·min −1 , for separation of 16S rRNA gene fragments.

16S rRNA Gene Cloning, Sequencing and Analysis
PCR reactions from enrichment cultures were performed with primers 27f and rd1 (5′-AAGCTTAAGGAGGTGATCCAG-3′), adapted from [31]. Zero Blunt TOPO PCR cloning kit (Invitrogen) was used for cloning obtained 16S rRNA gene PCR amplicons. Selection of individual clones for sequencing was based on AluI or HhaI restriction profile polymorphism of PCR amplicons obtained from cloned fragments with T3 and T7 universal primers, and visualized on NuSieve 3:1 Agarose gels (2% equivalent agarose; Tebu-bio, Le Perray-en-Yvelines, France). Sequences of PCR products, purified with ExoSAP-IT (USB Corporation, USA) following the manufacturer's protocol, were obtained with primers 27f, 534r or rd1 on an ABI Prism 3130 XL sequencer (Applied Biosystems,

Accession Numbers
The 16S ribosomal RNA gene sequences obtained for selected bacterial strains enriched from the studied site were submitted to the EMBL database under accession numbers FN689722 to FN689728 and KP219716.

CCl4-Degrading Enrichments and Isolation of Pelosinus sp. TM1
To identify CCl4-degraders among the prokaryotic community of concentrated groundwater samples, enrichment cultures under 33 different selective conditions were set up for each of the collected samples in campaigns T1 and T2 (11 different conditions, in the presence of 0, 10 and 200 mg·L −1 CCl4; Table S1). Removal of CCl4 in enrichment cultures was only observed anaerobically in nutrient-rich C and MBM media with initial CCl4 concentrations of 10 mg·L −1 , and growth was observed 24 to 72 h after inoculation. Delayed and slower growth without degradation activity was observed at 200 mg·L −1 . In five enrichment cultures (designated C1 to C5), degradation of CCl4 was complete between six and 11 days after inoculation. Sustained CCl4 degradation in successive cultures was assessed by inoculating aliquots of stationary phase CCl4-degrading enrichments into fresh medium. Two types of degradation curves were observed after successive transfers ( Figure 1). Initially, cultures C4 and C5 degraded CCl4 fastest in the first enrichment cultures, but eventually lost their capacity for CCl4 degradation upon successive inoculations under the same culture conditions (Table S2; Figure 1). Evaluation of the diversity of Bacteria and Archaea in the five CCl4-degrading enrichment cultures by 16S rRNA gene T-RFLP analysis evidenced: (i) different fingerprints for each enrichment with exclusively bacterial content (no Archaea); (ii) shifts in community structure after successive transfers; and (iii) predominant phylotypes identified by database comparisons ( [28], Table 1).  [32]. When no sequence data were available, T-RFLP database analysis was performed using a size confidence interval of ±1 nt [28]; b n.a., not applicable. Strains TM1, TM2 and TM5 were isolated as single colonies from corresponding CCl4-degrading enrichments plated out on solid C medium (Table S2). Strain TM2 was affiliated to Klebsiella sp. basing on its 16S rRNA gene sequence (Table 1). Several enterobacteria catalyzing CCl4 degradation were described previously, including Klebsiella pneumoniae strain L17 (98% sequence identity with strain TM2), a dissimilatory iron-reducing bacterium (DIRB) capable of CCl4 degradation under anoxic conditions in presence of ferric oxides and humic substances [33]. Strain TM5 showed a 16S RNA gene sequence similar to representatives of Peptostreptococcaceae (Table 1), which comprise Acetobacterium woodii, Moorella thermoacetica and Clostridium sp. strain TCAIIB, previously described to dechlorinate CCl4 by a transformation process involving corrinoids [18,22,34]. Initially, strain TM5 displayed the fastest degradation rate observed in this study (Table S2), but its ability to degrade CCl4 was progressively lost upon transfer of the enrichment culture. Strain TM1, the most stable CCl4-degrading strain isolated in this work, was affiliated to the genus Pelosinus basing on 16S rRNA gene sequence analysis (Table 1).
Strains TM1, TM2 and TM5 and representatives of the other taxa identified in dechlorinating consortia represented less than 7% of the prokaryotic groundwater community analyzed by T-RFLP, with proportions ranging from 0.1% to 2.0% of the groundwater community (Table S2; Figure S1A). Diversity indices of the prokaryotic population of the CCl4-contaminated aquifer from water samples of the contamination source obtained at months 14, 21 and 29 after beginning of pump-and-treat showed little difference, confirming that major shifts in endogenous bacterial populations did not occur during "pump-and-treat" ( Figure S1; Table S3).

Characteristics of CCl4-Degrading Pelosinus sp. TM1
Strain TM1 was shown to contain at least three different ribosomal RNA operons (Table 1). Although a single colony type was observed on solid medium, T-RFLP profiles with two AluI T-RF bands of 126 and 156 nt and two HhaI T-RFs of 110 and 580 nt were repeatedly observed (Table S2).
Complementary D-HPLC [20] and cloning steps allowed the identification of two 16S rRNA sequences of 1556 bp differing by only five bases, as well as a longer sequence of 1656 bp. The three 16S rRNA genes shared 99% overall sequence identity except at the 5′-end of the gene, in which the longer sequence featured a 120 bp sequence instead of the 20 bp sequence shared by the two shorter sequences. Additional HhaI and AluI restriction sites in the longer sequence insert (Figure 2A) provided the explanation for the observed T-RFLP fingerprints. Taxonomically closest strains from the genus Pelosinus also feature sequence inserts at the 5′-end of their 16S rRNA genes ( Figure 2B) [35,36].  (Table S1, Figure 3). No growth was observed after heat treatment (100 °C for 8 min), and spores were not detected. In C medium at 25 °C, strain TM1 reached an OD600 nm of about 0.6 in stationary phase (Figure 3), and retained its viability under anoxic conditions after exposure to oxygen for one week (incubation under sterile conditions in an air-exposed rotary shaker at 100 rpm at 25 °C). Strain TM1 metabolized 79 carbon sources out of the 95 of the Biolog "anaerobe identification test panel". Carbon sources that were not utilized included N-acetyl-D-glucosamine, N-acetyl-mannosamine, adonitol, D-arabitol, i-erythritol, D-glycerol, D-trehalose, glyoxylic acid, itaconic acid, urocanic acid, L-alanyl-L-threonine, L-valine, L-aspartic acid, inosine, thymidine and uridine. The range of growth-supporting carbon sources of strain TM1 was identical to that of the corresponding type strain P. propionicus TmPM3 T [37], except that strain TM1 was additionally able to metabolize succinate, propionate, acetate and formate, but unable to use D-glycerol. CCl4 degradation by Pelosinus sp. TM1 paralleled growth in C medium, with a degradation rate of 2.2 μg CCl4 day −1 mg protein −1 (Figure 3; Table 2). A similar specific degradation rate of 2.3 ± 0.3 μg CCl4 day −1 mg protein −1 was found after growth under more oligotrophic groundwater-simulating conditions (SGW medium supplemented for growth yield purposes with pyruvate (10 mM) as the potential electron donor, 10 times diluted yeast extract compared to C medium, and nitrate (10 mM) as nitrogen source). CCl4 degrading ability of strain TM1 was further assessed by estimating the amounts of chloride production from CCl4 degradation in cell suspensions of cultures grown in SGW medium with 10 mM pyruvate. Rates of CCl4 removal (0.8 nmol ± 0.2 CCl4 h −1 mg −1 ) and chloride production (2.8 nmol ± 0.4 Cl − h −1 mg −1 ) were compatible with a stoichiometry of four chloride ions produced per CCl4 consumed ( Figure S2). Notably, Pelosinus sp. TM1 was unable to degrade other chlorinated methanes ( Figure S3). Optical density (open symbols) and CCl4 degradation (filled symbols); cultivation with 10 mg·L −1 CCl4 in anoxic C medium (circles), C medium exposed to oxygen for one week (triangles), and anoxic SGW medium (squares). Error bars represent the standard error of 3 technical replicates. Filter-sterilized, oxygen-exposed <0.1 n.d.
Living cells of Pelosinus sp. TM1 were not required for CCl4 degradation, since filter-sterilized supernatants of stationary phase cultures displayed a similar CCl4 degradation rate as cell suspensions of the strain ( Table 2). The ability of cultures and cell suspensions of Pelosinus sp. TM1, as well as of cell-free supernatants, to degrade CCl4 following heat treatment, oxygen exposure or addition of the reducing agents L-cysteine·HCl or Na2S was tested. Exposure to oxygen suppressed CCl4 degradation (cell-free supernatant and resuspended cells in phosphate buffer; Table 2). Treatment of oxygen-exposed supernatants with reducing agents restored CCl4 degradation to levels similar to those of non-treated anoxic supernatants. Heat treatment of supernatants had no effect on CCl4 degradation activity. Taken together, these data indicate that Pelosinus sp. TM1 synthesized a thermostable molecule that can be released from cells in the supernatant, is redox-active (i.e., oxidized by oxygen and reduced by sulfide) and was at least in part responsible for CCl4 degradation.

Discussion
To our knowledge, Pelosinus sp. TM1 is the first isolate of the genus shown to be capable of dehalogenation. Nevertheless, P. fermentans R7 was shown to provide an essential corrinoid cofactor for Dehalococcoides reductive dehalogenases in a trichloroethene-dechlorinating syntrophic consortium [38], and several strains of the genus Pelosinus were isolated from sites contaminated with halogenated compounds including chlorinated alkanes [35,[38][39][40]. Pelosinus sp. TM1 afforded stable and complete dehalogenation of CCl4 (Table 2, Table S2), at rates comparable to those previously observed for both Escherichia coli K12 and for a consortium of various anaerobic bacteria under anoxic conditions (0.2 and 0.3 μg day −1 mg protein −1 , respectively) [41,42].

CCl4 Degradation by Pelosinus sp. TM1 is a Co-Metabolic Process
A CCl4-degrading microbial process in which CCl4 is used as a carbon or energy source for growth has not been identified [6,41,42]. Several lines of evidence suggest that Pelosinus sp. TM1 also involves a co-metabolic process. In terms of carbon, the quantities and rate of degraded CCl4 under the tested conditions will at most only make a minor contribution to the biomass produced (e.g., 3 mg protein in the presence of 3.9 μg CCl4 carbon). In terms of energy, and although growth-promoting metabolism of CCl4 may be thermodynamically favorable [6], the obtained data suggest that it is unlikely that CCl4 is used by Pelosinus sp. TM1 as an electron acceptor in energy-generating metabolism for growth. First, comparing the molar ratio of electron donor pyruvate (10 mM) to the amounts of CCl4 added to cultures in SGW medium (65 μM), CCl4 would have been growth limiting when used as an electron acceptor. Second, considerable amounts (>60%) of the initial CCl4 remained once Pelosinus sp. TM1 reached stationary phase under the chosen culture conditions. Finally, CCl4 degradation also occurred in spent medium, where it resisted heat treatment (Table 2), further suggesting that, at least in part, dehalogenation of CCl4 in cultures of Pelosinus sp. TM1 may be an abiotic process. Several examples of extracellular degradation of CCl4 mediated by microbially produced low molecular weight compounds are known [6]. For example, under denitrifying conditions, P. stutzeri KC excretes the transition metal-chelator pyridine-2,6-bis(thiocarboxylate) (PDTC) which, when complexed to Cu(II), mediates extracellular dechlorination of CCl4 [15]. In addition, vitamin B12 and related corrinoid cofactors, which are produced and released to the extracellular compartment by a variety of microorganisms, also catalyze degradation of CCl4 [7,34,43,44]. The heat-stable molecule produced by Pelosinus sp. TM1 mediating CCl4 degradation might indeed be related to corrinoid analogues, and such cofactors were shown to be produced by Pelosinus strains [38], but further investigations will be required to determine its identity.

Conclusions
Diversity analysis of the prokaryotic community of an aquifer contaminated by CCl4 for decades and enrichment and isolation of CCl4-degrading bacteria in the presence of CCl4 showed that only a minor fraction of the bacteria from this environment were capable of CCl4 degradation in the laboratory. Among those, strains from genera Klebsiella and Clostridium had already been associated previously with CCl4 degradation. In contrast, Pelosinus sp. TM1 isolated here is the first representative of the Pelosinus genus shown to sustain CCl4 degradation.