Growth and Yields of Dechlorinators, Acetogens, and Methanogens

digester treating wastewater of a palm oil plant in North-Sumatra, Indonesia. ..... M. Ahsanul Islam , Elizabeth A. Edwards , Radhakrishnan Mahade...
0 downloads 0 Views 289KB Size
Environ. Sci. Technol. 2007, 41, 2303-2310

Growth and Yields of Dechlorinators, Acetogens, and Methanogens during Reductive Dechlorination of Chlorinated Ethenes and Dihaloelimination of 1,2-Dichloroethane MELANIE DUHAMEL† AND ELIZABETH A. EDWARDS* Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario, Canada

The population dynamics of a mixed microbial culture dechlorinating trichloroethene (TCE), cis-1,2-dichloroethene (cDCE), 1,2-dichloroethane (1,2-DCA), and vinyl chloride (VC) to ethene were studied. Quantitative PCR revealed that Dehalococcoides, Geobacter, Sporomusa, Spirochaetes, and Methanomicrobiales phylotypes grew in shortterm experiments. Both Geobacter and Dehalococcoides populations grew during TCE dechlorination to cDCE, but only Dehalococcoides populations grew during further dechlorination to ethene. The cell yields for Dehalococcoides determined in this study were similar on an electron equivalent basis regardless of the chlorinated compound transformed: (0.9 ( 0.3) × 108 16S rRNA gene copies/microelectron equivalent (µeeq) ethene produced during cDCE dechlorination, (1.5 ( 0.3) × 108 copies/µeeq ethene produced during VC dechlorination, and (1.6 ( 0.8) × 108 copies/ µeeq ethene produced during 1,2-DCA dihaloelimination. The yield for the Geobacter population on TCE was estimated to be (1 ( 0.5) × 108 copies/µeeq cDCE produced. Calculations showed that the Geobacter population was likely responsible for approximately 80% of the TCE dechlorinated to cDCE in this experiment. Acetogenesis by a Sporomusa population was the main competition to dechlorination for reducing equivalents. Sporomusa did not transform any chlorinated substrates tested, but was capable of converting methanol to acetate and hydrogen for dechlorination. Understanding the functions of various populations in mixed communities may explain why Dehalococcoides spp. are active at some sites and not others, and may also assist in optimizing the growth of bioaugmentation cultures, both in the laboratory and in the field.

Introduction Bioaugmentation with cultures containing Dehalococcoides is an effective means of remediating chlorinated ethenecontaminated sites where halorespiring organisms are not naturally present or abundant (1, 2). However, despite the * Corresponding author phone: (416) 946-3506; fax: (416) 9788605; e-mail: [email protected]. † Current affiliation: Geosyntec Consultants, Guelph, Ontario, Canada. 10.1021/es062010r CCC: $37.00 Published on Web 03/03/2007

 2007 American Chemical Society

many studies characterizing reductively dechlorinating communities (e.g., 3-14), little has been confirmed about the ecological relationships between the populations detected. While there have been several qualitative studies using molecular biological techniques to show variation in chlorinated ethene-dechlorinating communities in time and space (3-5), quantitative assessments resulting in growth yields for specific populations other than Dehalococcoides spp. have yet to be performed. The general consensus is that fermenters transform electron-donating substrates to hydrogen and acetate, which are then used by Dehalococcoides, often regarded as the primary halorespiring population. Sulfate-reducers, acetogens, and methanogens are competitors for available hydrogen. Yet perhaps these other populations assist in the dechlorination process, either directly or by providing essential micronutrients to halorespiring species. Understanding the functions of various populations in mixed communities may explain why Dehalococcoides spp. are active at some sites and not others, and may also assist in optimizing the growth of bioaugmentation cultures, both in the laboratory and in the field. In a previous study (6), we detected a Geobacter population in KB-1 subcultures maintained on tetrachloroethene (PCE) and TCE that had a close phylogenetic relationship to the PCE-to-cDCE-dechlorinating species Geobacter lovleyi strain SZ (15). The objectives of the present study were to determine if this population was indeed dechlorinating, as well as to clarify the roles of other putative non-dechlorinating populations in KB-1 and to estimate cell yields. The dynamics of a KB-1 subculture enriched on TCE were explored in batch systems throughout the course of reductive dechlorination. Dihaloelimination of 1,2-DCA was also tested because at least 177 of the 1,244 sites on the U.S. EPA’s Final National Priority List are contaminated with TCE and 1,2-DCA (http:// cfpub.epa.gov/supercpad/cursites/srchsites.cfm, accessed July 11th, 2006). In these timecourse experiments, a small inoculum was used so that the growth of populations key to each dechlorination step would become obvious above initial levels. Populations of interest were measured by quantitative PCR, allowing the yields of several populations to be calculated simultaneously in a mixed culture.

Materials and Methods Analytical Procedures. Chlorinated ethenes, ethene, and methane were analyzed by headspace gas chromatography (16). Acetate was analyzed by ion chromatography (16). Cultures. The KB-1/TCE culture was enriched and maintained under anaerobic conditions as previously described (6, 16). A pure culture of the Sporomusa population in KB-1 was obtained during attempts to isolate a VCdechlorinating Dehalococcoides population (16) by sequential liquid transfers and anaerobic plating. The purity of the Sporomusa culture was established by repeated dilutions and plating, uniform colony morphology, molecular analyses, and the inability of the culture to produce methane or to dechlorinate VC (Supporting Information). Sporomusa Transformation Experiment. This experiment was conducted to determine whether the Sporomusa population in KB-1 was capable of dechlorinating chlorinated ethenes or 1,2-DCA. Experimental procedures are described in the Supporting Information. Timecourse Experiment. Each treatment was in triplicate. Sterile anaerobic mineral medium (190 mL) was placed in each 250 mL bottle, followed by a 10 mL aliquot (5% v/v) of a KB-1/TCE-methanol culture (sterile controls were not inoculated). Bottles were sealed with Mininert caps (VICI VOL. 41, NO. 7, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2303

Precision Sampling). Methanol was added as the electrondonating substrate for all treatments including sterile controls. Cultures were re-amended with methanol when methane production and/or dechlorination stalled. An inoculated, methanol-only treatment was monitored to determine which populations would grow independently of dechlorination. Potential electron acceptors (TCE, cDCE, VC, or 1,2-DCA) were added targeting equal electron equivalents (eeq) to 230 µM (aqueous) TCE (310 µeeq). The following conversions were used for calculating potential electron acceptor concentrations: 6 eeq/mol TCE, 4 eeq/mol cDCE, 2 eeq/mol VC, and 2 eeq/mol 1,2-DCA (Table SI-5). When about 70-80% of the initial amount of each electron acceptor in the 1,2-DCA, VC, and cDCE treatments had been transformed to ethene, cultures were sampled for DNA (50 mL) and ion chromatography (1 mL). Methanol and sterile control treatments were sampled concurrently. Sampling for DNA was performed only once for these treatments, as this consumes significant liquid volume, thus limiting the number of times they can be sampled. Bottle sizes and sampling frequencies were a compromise between having adequate volume, glovebox space, and time to sample. Only the TCE treatments were sampled more than once for DNA to monitor growth during each step of dechlorination. Thus, TCE treatments were set up with a larger liquid volume (1-2 L) than the others to minimize the effects of more frequent sampling. DNA and ion chromatography samples (51 mL total volume) were removed from each culture (using a disposable 60 mL syringe fitted with a 21 gauge needle [Becton Dickinson]) as closely as possible to each of the following points: end of TCE dechlorination, end of cDCE dechlorination, end of VC dechlorination, and one week later, in order to observe which other organisms consumed any excess electron donors. The first TCE treatment was set up in 2 L (liquid volume) media bottles with plastic screwcaps with a hole drilled to accommodate a black butyl rubber stopper. TCE sorbed heavily to these stoppers, so the TCE treatment was repeated using 1 L bottles modified to fit Mininert caps. The second TCE treatment was amended with twice the amount of methanol as the first in order to avoid donor limitations observed partway through the first treatment. Quantitative PCR methods developed for the populations detected in the KB-1 cultures were described previously (6). A single time 0 DNA extraction was performed on 50 mL of the inoculum culture. Since a 5% (v/v) inoculum was added to each bottle, it was assumed that the initial 16S rRNA gene copies/µL for each population in the bottles was 5% of the copies/µL obtained for the time 0 sample. This assumption was verified in separate experiments by analyzing dilutions of the inoculum by quantitative PCR. Yield Calculations. Observed cell yields were calculated from the change in 16S rRNA gene copy numbers measured by quantitative PCR divided by the change in concentration of electron acceptors over the same time period. It was assumed that decay was negligible during the short duration of the experiment (16 days), although decreases in copy numbers were observed for some populations as shown in Figures 1-3, SI-1 and SI-2. Throughout the text, observed yields are reported in 16S rRNA gene copy numbers per µeeq of electron acceptor reduced, as these were the parameters measured. For comparison between different organisms and to theoretical and literature values, yields in various units were converted to grams dry organic mass of cells/eeq electron acceptor reduced assuming: (1) a wet cell volume; (2) that cells have an average density equal to water; and (3) that cells are 80% water and 90% organic by dry weight (Table 2, ref 17). Wet cell volumes were estimated from the size and shape of related organisms for which microscopic images have been published (Table 2). The number of 16S gene copies 2304

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 7, 2007

per genome was used to convert copy numbers to cells, although this value was not available for all organisms. Different combinations of cell volumes and 16S gene copy numbers per genome were used to calculate yields for the Sporomusa and Methanomicrobiales populations to show the sensitivity of the estimated yield to these parameters (Table 2). Cell yields can also be predicted from thermodynamic considerations as presented in ref 17. Such predicted yields are useful to assess the efficiency of energy recovery from a particular donor/acceptor couple in a microbial system. Balanced stoichiometric equations including cell growth based on theoretically computed or measured growth yields can be easily derived, and sample calculations are presented in the Supporting Information. Thermodynamically predicted cell yields for processes relevant to this paper are shown in Table SI-6 (assuming standard conditions). Standard Gibbs’ free energies of formation at pH 7 were modified to reflect the actual concentrations of reactants and products (17) and resulting yields under these conditions are derived in Table SI-7 and reported in Table 2.

Results Sporomusa Experiment. No growth was observed in either uninoculated or purged controls. The Sporomusa culture did not dechlorinate any of the chlorinated compounds tested (data not shown). Acetate production occurred with methanol and/or H2 as electron donors. Sporomusa growth yields calculated from this data set were similar to those calculated from corresponding treatments in the timecourse experiment (Supporting Information Table SI-2). VC- and particularly cDCE-amended cultures had lower yields than other treatments, suggesting that these compounds inhibited the Sporomusa population. Timecourse Experiment. The inoculum culture for the timecourse experiment was deliberately chosen to be diverse compared to most other KB-1 subcultures (6), having seven populations each representing g1% of the measured 16S rRNA gene copies in its DNA sample. The relative abundance of each population in the inoculum was as follows: 30% Geobacter, 26% Sporomusa, 17% Dehalococcoides, 12% Spirochaetes (identified to taxonomic class), 9% Methanomethylovorans, 5% Methanomicrobiales (identified to taxonomic class), 1% Acetobacterium, and