Environ. Sci. Technol. 2008, 42, 4784–4789
Adaptation of Aerobic, Ethene-Assimilating Mycobacterium Strains to Vinyl Chloride as a Growth Substrate YANG OH JIN AND TIMOTHY E. MATTES* Department of Civil and Environmental Engineering, 4105 Seamans Center, University of Iowa, Iowa City, Iowa 52242
Received January 07, 2008. Revised manuscript received April 15, 2008. Accepted April 23, 2008.
Contamination of drinking water source zones by vinyl chloride (VC), a known human carcinogen and common groundwater contaminant, poses a public health risk. Bioremediation applications involving aerobic, VC-assimilating bacteria could be useful in alleviating environmental VC cancer risk, but their evolution and activity in the environment are poorly understood. In this study, adaptation of etheneassimilating Mycobacterium strains JS622, JS623, JS624, and JS625 to VC as a growth substrate was investigated to test the hypothesis that VC-assimilating bacteria arise from naturally occurring ethene-assimilating bacteria. VC consumption in the absence of microbial growth was initially observed in cultures grown in both ethene and 1/10-strength trypticase soy agar + 1% (w/v) glucose. After extended incubations (55-476 days), all strains commenced growth-coupled VC consumption patterns. VC-adapted cultures grown on 20 mM acetate subsequently retained their ability to assimilate VC. Three independent purity check methods (streak plates, 16S rRNA gene sequencing, and repetitive extragenic palindromic polymerase chain reaction) verified culture purity prior to and following VC adaptation. Overall, our results suggest that ethene-assimilating mycobacteria have a widespread ability to adapt to VC as a growth substrate.
Introduction Vinyl chloride (VC) is a known human carcinogen (1) that is often found in ambient groundwater (2). Contamination of drinking water with VC poses a cancer risk to exposed populations, either by ingestion of contaminated drinking water or inhalation during showering (3, 4). The toxicity and carcinogenicity of VC has prompted the US EPA to set the maximum contaminant level for VC in drinking water at 2 ppb (5). VC-contaminated groundwater can result from polyvinyl chloride (PVC) production facility releases (1, 3, 6) and leaching of VC monomer from PVC pipe made prior to 1977 (7). However, field evidence indicates that aqueous environmental VC contamination may also be derived from incomplete anaerobic reductive dechlorination of the more highly chlorinated ethenes such as tetrachloroethene (PCE) and trichloroethene (TCE) (8, 9). The potential for generation of a mobile VC plume that escapes anaerobic biotransformation during in situ reductive dechlorination of PCE and TCE and migrates toward down* Corresponding author fax: (319) 335-5660; e-mail:
[email protected]. 4784
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 13, 2008
gradient drinking water zones is an additional cause for public health concern. This concern has stimulated interest in microorganisms that degrade or detoxify VC in situ. Recent research has revealed several Dehalococcoides spp. that use VC as an electron acceptor under anaerobic conditions, reducing it to ethene and capturing the energy released for growth (10–12). Microbially mediated anaerobic oxidation of VC has also been reported (13–15); however, no specific bacteria have yet been implicated in this process. In contrast, aerobic bacteria that couple VC oxidation to growth (i.e., VC-assimilating bacteria) are easily enriched from environmental samples (16–20), which suggests they play a role in attenuating VC in the environment. Therefore, our goal is to develop a better understanding of VC-assimilating bacteria so that we may ultimately be able to better predict the biodegradation of VC in groundwater. To date, all organisms reported to grow on VC (e.g., Mycobacterium, Nocardioides, and Pseudomonas strains) also assimilate ethene and appear to use the same enzymes to metabolize both VC and ethene (16, 17, 21, 22). The enzymes catalyzing the initial steps of the VC and ethene pathways are well-known. Alkene monooxygenase (AkMO) attacks and transforms VC and ethene into VC epoxide (chlorooxirane) (16, 17) and epoxyethane (ethylene oxide) (19, 22–24), respectively. Epoxyalkane:coenzyme M transferase (EaCoMT) further assimilates VC epoxide and epoxyethane into central metabolism (19, 21, 22, 25, 26). The remaining enzymes and metabolic intermediates that participate in the VC- and ethene-biodegradation pathways have yet to be conclusively identified and characterized, but recent gene and protein expression studies have revealed candidates for detailed study (22, 27). Both pathways also appear to end with the formation of acetyl-CoA, which enters the TCA cycle (23). Under certain conditions, ethene concentrations can reportedly exceed 30 mg/L in soil (28), suggesting that etheneassimilating bacteria would be widespread in those environments. Ethene, generated in anaerobic groundwater via the activity of Dehalococcoides spp (29), can subsequently move into aerobic groundwater systems and provide an ecological niche for aerobic, ethene-degrading bacteria. It is proposed that, upon exposure to VC in the environment, aerobic, ethene-assimilating bacteria will gradually adapt to VC as a growth substrate. In support of this hypothesis, Pseudomonas aeruginosa strain DL1, isolated with ethene as the sole carbon and energy source, gradually transitioned from a cometabolic VC degradation pattern to a growth-coupled VC degradation pattern in the laboratory (30). However, a similar phenomenon has not been reported for other ethene-assimilating bacteria. Recently, four ethene-assimilating Mycobacterium strains (JS622, JS623, JS624, and JS625) were isolated from several locations in the United States. All four strains possessed highly similar EaCoMT genes, but none of the strains grew on VC (31) after incubation periods of 78 days (JS622), 52 days (JS623), 78 days (JS624), and 15 days (JS625) (32). Here, we report that strains JS622, JS623, JS624, and JS625 adapted to VC as growth substrate after extended incubations. We also show that following growth on 20 mM acetate, strains JS622, JS623, JS624, and JS625 retained VC-assimilating ability. The implications of these results are discussed.
Materials and Methods Chemicals, Bacterial Strains, Media, and Culture Conditions. Ethene (99%) was from Airgas and vinyl chloride (99.5%) was from Fluka. All other chemicals were reagent grade or molecular biology grade. VC-assimilating Myco10.1021/es8000536 CCC: $40.75
2008 American Chemical Society
Published on Web 06/03/2008
bacterium sp. strains JS60 (ATCC BAA-494), JS616 (ATCC BAA496), JS617(ATCC BAA-497), and JS621 (ATCC BAA-498) and ethene-assimilating Mycobacterium sp. strains JS622, JS623, JS624, and JS625 (31) were used in this study. Initially, cultures were grown in minimal salts medium (MSM), prepared as described previously (33) with filter-sterilized ethene (0.25 mM) as the sole source of carbon and energy. Cultures were incubated in a 2 L modified Erlenmeyer flask (25) at room temperature (21-23 °C), aerobically with shaking at 200-300 rpm, harvested at midexponential phase (OD600 ∼ 0.3), and stored at -80 °C for future experiments. Culture purity was checked by plating onto 1/10-strength trypticase soy agar + 1% (w/v) glucose (TSAG) and incubating at 30 °C until colonies appeared (approximately 14 days). Analytical Methods. Headspace samples of VC and ethene (100 µL) were analyzed via gas chromatography with flame ionization detection as described previously (22), and the peak area was compared to an external standard. Aqueous phase VC and ethene concentrations are reported. The Henry’s law constant used was 0.973 for VC (34) and 7.058 for ethene (30). Bacterial growth was measured by reading the optical density at 600 nm with a spectrophotometer (OD600). Bottles were shaken vigorously, liquid sample (0.7 mL) was removed to a cuvette, and OD600 was measured immediately. DNA Extraction and PCR Analysis. PCR analyses were employed to check the purity of VC-adapted Mycobacterium strains. A bead-beating DNA extraction method (31) was used with the following modifications: Cells were grown on TSAG, harvested, washed with 600 µL of STE buffer (100 mM NaCl, 10 mM Tris-Cl, 50 mM EDTA, pH 8.0), and added to a sterile 2 mL screwcap tube containing 1.5 mL of 0.1 mm zirconia/ silica beads wetted with STE buffer. Cell lysis was achieved by bead-beating at high speed for 2 min with a Bio-Spec MiniBeadbeater-8. The lysate was purified with sequential 25:24:1 (v/v/v) phenol/chloroform/isoamyl alcohol (PCI) and 24:1 (v/v) chloroform/isoamyl (or isobutyl) alcohol extractions and precipitated with 10 M NH4-acetate and 2-propanol. The DNA pellet was washed with 70% ethanol and dissolved in 250 µL of 10 mM Tris-Cl, pH 8.0. DNA was diluted to 100 ng/µL in water for use as a template for PCR. The Taq PCR Master Mix (Qiagen) kit was used for PCR with 16S rRNA gene primers while the Taq PCR Core Kit (Qiagen) was used for repetitive extragenic palindromic polymerase chain reaction (REP-PCR). REP-PCR was carried out as described previously (35) with the following modifications: PCR mixtures (25 µL) contained 1 × buffer, 1.5 mM MgCl2, 200 µM dNTPs, 0.625 U Taq polymerase, 2 µM of each primer (REP1R-I and REP2-I), and 50 ng DNA template. REPPCR products were separated on a 1% agarose gel, stained with ethidium bromide, and photographed. Amplification of the 16S rRNA gene was carried out as described previously (36) with 0.5 µM of each primer (27f and 1492r) and 100 ng of DNA template per 100 µL reaction. DNA sequencing was performed by The University of Iowa DNA Facility. Sequencing data were compared to 16S rRNA gene sequences deposited in Genbank. BioEdit and ClustalX (37) were used for alignment and analysis of DNA sequences. VC Adaptation Experiments. Ethene-Grown Cultures. Mycobacterium sp. strain JS622, JS623, JS624, and JS625 cultures were grown on MSM-ethene (0.25 mM) in a 2 L modified Erlenmeyer flask (25), harvested at midexponential phase (OD600 ∼ 0.3), rinsed twice in MSM, resuspended to OD600 ∼ 0.1 in MSM, and transferred into 160 mL serum bottles for 72 mL of total culture volume. Filter-sterilized VC was injected into the bottles to provide initial VC concentrations of 0.8-1.0 mM. Cultures were incubated at room temperature (21-23 °C) with shaking of 200-300 rpm. The mass of VC and cell density in the bottle were monitored as described in the analytical methods. A bottle containing MSM
and VC (0.9 mM) was used to measure abiotic VC mass loss. All experiments were prepared in triplicate. TSAG-Grown Cultures. Ethene-grown Mycobacterium sp. strain JS622, JS623, JS624, and JS625 cultures were washed twice in MSM, transferred onto TSAG plates, and incubated at 30 °C. After colonies appeared (approximately 7-14 days), they were scraped from the plates, washed twice in MSM, resuspended to OD600 ∼ 0.1 in MSM in 160 mL serum bottles, and fed VC (0.6-1.8 mM) exactly as described above for ethene-grown cultures. One TSAG plate provided sufficient biomass for these experiments. Acetate-Grown Cultures. Both unadapted and VC-adapted strain JS622, JS623, JS624, and JS625 cultures were grown on acetate (20 mM), harvested at midexponential phase, rinsed twice in MSM, and resuspended to OD600 ∼ 0.1 in MSM and fed VC and ethene (0.8-1.0 mM) exactly as described above for ethene-grown cultures.
Results VC Biodegradation in Ethene- and TSAG-Grown Mycobacterium Cultures. In a previous study, Mycobacterium strains JS622, JS623, JS624, and JS625 were not reported to use VC as a growth substrate (31). Here, we sought to observe VC consumption patterns in ethene-grown JS622, JS623, JS624, and JS625 cultures, which was expected to occur in the absence of microbial growth (hereafter referred to as cometabolic VC biodegradation). Additionally, because TSAGgrown Mycobacterium strain K1 immediately utilized VC as a carbon and energy source (19) we investigated the response of TSAG-grown JS622, JS623, JS624, and JS625 to VC. Washed, ethene-grown JS622, JS623, JS624, and JS625 cultures fed 0.8-1.0 mM VC rapidly consumed VC for up to 14 days before VC degradation virtually ceased (Figure 1 and Figure S1 of the Supporting Information). During this approximately 14 day VC consumption period, we observed decreases in OD600 that stabilized after VC consumption ended (Figure 1 and Figure S1). However, OD600 decreases in VC-fed cultures closely followed OD600 decreases in control cultures that were not fed VC. The patterns of VC consumption and the lack of observable OD600 increases during the first 14 days of the experiment are both indicative of cometabolic VC biodegradation (30). In contrast to ethenegrown cultures, washed, TSAG-grown JS622, JS623, JS624, and JS625 cultures fed 0.6-1.8 mM VC consumed VC at a steady but initially much slower rate (Figure 1 and Figures S1-S3). Unadapted, TSAG-grown cultures also retained the ability to grow on ethene (data not shown). These data indicate that ethene- and TSAG-grown cultures did not initially possess the ability to use VC as a growth substrate. VC Adaptation in Ethene- and TSAG-Grown Mycobacterium Cultures. Despite evidence that both ethene- and TSAG-grown JS622, JS623, JS624, and JS625 cultures could not initially grow at the expense of VC, we continued to monitor the cultures for extended periods. Between days 110 and 125, ethene-grown strain JS622 consumed the remaining VC (Figure 2). The culture was immediately fed VC (1.3 mM) and monitored. During a 25-35 day period, strain JS622 degraded this VC spike and a subsequent spike relatively rapidly with a concomitant increase in OD600 (Figure 2). After the culture consumed about 550 µmol of VC, 4 mL of the culture was transferred into a 160 mL serum bottle containing 68 mL of fresh MSM to avoid problems with low oxygen tension and/or low pH that result from VC biodegradation and HCl release in a closed system. JS622 cultures transferred to fresh MSM displayed VC consumption rates and OD600 increases (Figure 2) that were similar to VCassimilating bacteria maintained in our laboratory (e.g., JS60, data not shown). Overall, these observations provide strong evidence that VC metabolism in Mycobacterium sp. strain JS622 gradually shifted from a cometabolic VC degradation VOL. 42, NO. 13, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4785
FIGURE 1. Cometabolic VC biodegradation patterns and OD600 values in ethene- and TSAG-grown Mycobacterium strains JS622 and JS623: (9) abiotic VC control for ethene-grown cultures, (0) abiotic VC control for TSAG-grown cultures, ([) unfed ethene-grown cultures (OD600 control), (]) unfed TSAG-grown cultures (OD600 control), (b) ethene-grown cultures fed VC, and (O) TSAG-grown cultures fed VC. The data for ethene-grown cultures is the average of three replicates and error bars show the standard deviation. The data for TSAG-grown cultures is representative of two independent experiments.
FIGURE 2. VC consumption patterns and changes in OD600 during adaptation of ethene- and TSAG-grown Mycobacterium strain JS622 cultures to VC. Points 1 and 2 in the graphs indicates when 6-10% of the culture was transferred to fresh MSM and fed VC. Higher amounts of VC were added at point 1 in the TSAG-grown JS622 culture because the culture was transferred to a culture bottle containing 500 mL of MSM. The data for ethene-grown cultures is the average of three replicates, and error bars show the standard deviation. The data for TSAG-grown cultures is representative of two independent experiments. pattern to a growth-coupled VC degradation pattern over an approximately 150 day period. Similar VC adaptation patterns were observed in ethene-grown Mycobacterium strains JS623, JS624, and JS625, but the incubation periods required for VC adaptation varied (Table 1 and Figures S4-S6). The VC consumption rate in a TSAG-grown JS622 culture fed 0.8 mM VC also accelerated around day 100 (Figure 2). As observed in the ethene-grown JS622 culture fed with VC, the VC consumption rate in the TSAG-grown JS622 culture increased with subsequent VC injections with concurrent increases in OD600 values, indicating that growth on VC was occurring. At 135 days, the JS622 culture was transferred into fresh MSM-VC, where the culture continued to display a growth-coupled VC consumption pattern (Figure 2). 4786
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 13, 2008
TSAG-grown Mycobacterium strains JS623, JS624, and JS625 fed 0.7, 0.6, and 0.7 mM VC, respectively, displayed similar VC adaptation patterns in comparison to TSAG-grown strain JS622, but, with the exception of JS623, the VC adaptation time in each culture was longer (Table 1, Figures S4-S6). When initial VC concentrations of 1.5-1.8 mM were used in VC adaptation experiments, all strains, with the exception of TSAG-grown strain JS625, adapted to VC after longer incubation periods than observed when initial concentrations of 0.6-0.8 mM VC were used (Table 1). Retention of VC- and Ethene-Assimilation in VCAdapted Mycobacterium Strains. Previous reports indicated that growing certain VC-assimilating bacteria on nonselective media resulted in loss of VC and ethene degradation ability
TABLE 1. Observed VC Adaptation Times in Mycobacterium Strainsa TSAG-grown Mycobacterium strain
ethene-grown: 0.8-1.0 mM VC addedb
0.6-0.8 mM VC added
1.5-1.8 mM VC added
JS622 JS623 JS624 JS625
125-128 55-60 110-122 176-224
112 63 158 476
132 118 222 293
a Estimated as the days required to degrade the first spike of VC. b Range represents the results of two independent experiments (both of which contained bottles in triplicate)
(20, 26). We subsequently tested the stability of the VCassimilation phenotype in VC-adapted JS622, JS623, JS624, and JS625 cultures after growth to midexponential phase on a nonalkene substrate. Both unadapted and VC-adapted variants of strains JS622, JS623, JS624 and JS625 were grown to midexponential phase (OD600 ∼ 0.3) on 20 mM acetate, a physiological condition that represses VC biodegradation gene expression and results in low VC biodegradation enzyme activity in other VC-assimilating bacteria (19, 21, 22, 25). Acetate-grown cultures were subsequently washed, resuspended in MSM, and fed VC (0.8-1.0 mM). Acetate-grown, VC-adapted strain JS622 grew readily on VC within 6 days without a lag period, while acetate-grown, unadapted strain JS622 did not grow on VC during the 24 day duration of the experiment (Figure 3). However, both unadapted and VCadapted strain JS622 grew equally well on ethene after growth on acetate (Figure S7). VC-adapted strain JS623 retained the ability to grow on VC within 7 days after growth on acetate, while acetate-grown, unadapted JS623 did not grow on VC during the 16 day duration of the experiment (Figure S3). Acetate-grown, VC-adapted, strains JS624 and JS625 also retained the ability to grow on VC (Figures S8, S9). Acetategrown JS623 cultures behaved differently than JS622 in that a 2-10 day lag period was observed before growth on VC (Figure 3) and a 3 day lag was observed before growth on ethene (Figure S7). Purity Confirmation of VC-Adapted Mycobacterium Strains. Because of long incubation times in VC adaptation experiments and the presence of VC-assimilating Mycobacterium strains in our laboratory, we used several independent approaches to confirm the purity of VC-adapted Mycobacterium strains. First, VC-adapted cultures were plated onto TSAG and incubated at 30 °C until colonies appeared, and the resulting colony morphology was compared to that of ethene-grown cultures prior to VC adaptation. In all but one case, the colony morphology and color of VC-adapted cultures was identical to those of unadapted cultures, indicating that our sterile technique was sound. One ethenegrown, VC-adapted JS623 culture displayed a different colony morphology and color (yellowish, opaque, and irregular colonies) from that of the unadapted culture (opaque, creamy, and irregular colonies), but this was not a reproducible phenomenon. Using 16S rRNA sequencing and REP-PCR methods (described below), both the yellowish, opaque, and irregular colonies and the opaque, creamy, and irregular colonies were identified as Mycobacterium strain JS623. The reason for the differences in colony morphology and color in a pure culture of strain JS623 is unknown at this time. Partial 16S rRNA sequences retrieved from the unadapted and VC-adapted variants of each Mycobacterium strain were 100% identical, which further suggested that the VC-adapted strains were pure. As a final purity check, REP-PCR profiles (19) of Mycobacterium strains JS60, JS616, JS617, JS621, JS622,
JS623, JS624, and JS625 were visualized as an alternative means of tracking potential VC-assimilating contaminants in our VC adaptation experiments. Each Mycobacterium strain displayed a unique REP-PCR pattern (Figure S10). However, comparison of the REP-PCR profiles of unadapted and VCadapted Mycobacterium strains revealed that they were the same (Figure 4).
Discussion Overall, the data presented here is the first to demonstrate that pure, ethene-assimilating Mycobacterium cultures, initially incapable of growth on VC, can adapt to VC as a sole growth substrate. Therefore, the results support the hypothesis that VC-assimilating bacteria arise from a naturally occurring population of ethene-assimilating bacteria and further suggests that VC adaptability is widespread among ethene-degrading mycobacteria. Reports of aerobic VC degradation in groundwater (38) and at VC-contaminated sites (8, 39) could thus be explained by adaptation of indigenous ethene-assimilating bacterial populations to VC. The VC concentrations considered in this study (37.5-112.5 mg/L) exceed VC concentrations typically observed in the field (
