Title: Western Region Hazardous Substance Research Center Project 2-OSU-07. Continuous Flow Column Studies of Reductive Dehalogenation with Two Different Enriched Cultures: Kinetics, Inhibition, and Monitoring of Microbial Activity

Investigators: Lewis Semprini, Mark E. Dolan, and Alfred Spormann

Institution: Oregon State University and Stanford University

Research Category: Groundwater, TCE, PCE, Vinyl Chloride, DNAPL, Bioremediation

Project Period: January 2003 – August 2007


Objectives: This joint project between Oregon State University and Stanford University evaluated the transformation of chlorinated ethenes in continuous-flow column studies with the Evanite (EV), Point Mugu (PM) and Victoria Strain (VS) cultures that have been developed and kinetically characterized in previous WRHSRC studies. The overall objectives of the project were to 1) evaluate if the predicted performance of the enrichment cultures was achieved and to test methods that may distinguish the VS from the EV culture; 2) apply molecular methods, such as FISH and Real-Time PCR, to determine the spatial distribution of the cultures and quantify the dehalogenating biomass within the column; 3) apply RNA-based methods to determine energetically based TCE and VC-dehalogenating activity temporally and spatially within the columns; 4) apply molecular based activity tests, such as transformation of fluorinated analogs, to determine dehalogenating activity that develops within the column; and 5) compare the results from modeling, molecular, and activity based results.

Summary of Findings:

Evanite Strain (EV) Culture

Continuous-flow column experiments were conducted to evaluate the reductive dechlorination of tetrachloroethene (PCE) in Hanford aquifer material after bioaugmentation with the Evanite (EV) culture. Three sets of column studies were performed with the Evanite culture. In Study-1 the culture was added to Hanford aquifer material where the iron content of the solids had not been pre-reduced. In Study-2 the effects of pre-reduction of the aquifer material with Na2S solution was evaluated and a series of transient tests were performed to determined how the columns responded to gradual increases in PCE and TCE concentrations. In Study-3 the aquifer material was changed to silica sand that had low iron content.

Study 1. In Study-1, prior to culture addition, PCE (0.07 mM) and bromide were added to a column amended with synthetic Hanford groundwater that contained 0.2 mM sulfate and 0.34 mM lactate. Little microbial activity was observed during this period with indigenous microbes from the Hanford aquifer material based on the absence of sulfate reduction and lactate removal. After six weeks, the column was inoculated with the EV enrichment culture. The injected lactate concentrations were increased from 0.34 mM to 0.67 mM, and to 1.34 mM, after six, eight, and sixteen weeks, respectively. PCE dechlorination to TCE and cis-DCE was observed when lactate concentrations increased from 0.35 mM to 0.67 mM on week eight. Rapid increases in cis-DCE concentration in the effluent to levels higher than the influent PCE concentration indicated enhanced PCE desorption and subsequent reduction to cis-DCE. When the lactate concentration was increased to 1.34 mM, propionate production was observed from lactate fermentation, and cis-DCE reduction to vinyl chloride (VC) and ethene occurred. The results indicated that competing electron acceptors, such as ferric iron, may have been responsible for the stall in cis-DCE transformation, but with prolonged treatment transformation to ethene occurred. At the end of 170 days, the column was destructively tested in an anaerobic glove box. Microcosms were constructed with spatial samples from the column and rates of PCE and VC transformation were determined. Spatial samples of the aquifer material were shipped to Stanford for molecular-based analysis.

Microcosm results indicated that most of the PCE and TCE transformation occurred near the column inlet. Microcosms supplied with PCE had the highest TCE and c-DCE formation rates in the sample taken closest to the column inlet and exhibited an essentially exponential decrease of an order of magnitude along the column length. VC transformation rates obtained from microcosms fed VC showed essentially uniform ETH production rates along the column length. The uniform spatial distribution of VC transformation activity compared to PCE transformation activity may have been due in part to the time chosen for column solids analysis. At 170 days, VC transformation to ETH was just being completely developed in the column and VC was present throughout the column, whereas rapid PCE transformation most likely depleted PCE concentrations within the first few centimeters from the inlet. Results of the molecular-based analysis of parallel column samples are provided below.

Study 2. In column Study-2, the effects of the pre-reduction of iron in the aquifer material was evaluated along with a series of transient tests where PCE (0.09-0.27 mM) and TCE (0.38-1.52 mM) concentrations were gradually increased. The aquifer material was chemically pre-reduced with a 5 mM Na2S solution to eliminate easily excessable Fe(III) in the aquifer material. With pre-reduced aquifer material, PCE was rapidly dechlorinated to cis-DCE, VC and ETH, with essentially no stall at the cis-DCE stage of transformation. Redox capacity measurements showed highly reducing conditions were more rapidly achieved in the column and sulfate reduction was immediately observed after lactate addition was initiated. Also, complete reduction to ETH was observed at lower lactate injection concentrations. Immediately upon switching to TCE addition (0.38 mM), 94% transformation to ETH was observed. TCE was essentially completely transformed to ETH at a concentration of 1.52 mM at a lactate concentration of 1.42 mM. Electron mass balances showed that in Study-1, where aquifer solids were not pre-reduced only 4.4 % of the electron flow was associated with dechlorination reactions, while in the pre-reduced column about 22% of the electron flow was associated with dechlorination reactions. The results indicate that when aquifer solids were not pre-reduced, the dehalogenating microorganisms were likely being outcompeted by iron reducing microorganisms for the available hydrogen required as an electron donor to drive c-DCE transformation to VC and ethene.


Study 3. In column study 3, the EV culture was added to quartz sand (99.5% SiO2) with low iron content (40 mg/kg). PCE (0.09 mM) and bromide were added to the columns amended with synthetic Hanford groundwater that contained 0.2 mM sulfate. No retardation of PCE transport through the column was observed. After 13 days, the columns were inoculated with the EV enrichment culture and 0.67 mM lactate. Immediately after EV culture addition, PCE dechlorination to TCE, and cis-DCE was observed. Effluent c-DCE concentrations did not exceed the influent PCE concentrations indicating no PCE was sorbed to the sand. Sulfate reduction occurred after increasing the lactate concentration to 1.0 mM and sulfide concentration increased to about 160 uM in the column. Complete PCE transformation to 89% ETH and 11% VC was achieved after 100 days of column operation. High sulfide concentration in the aqueous phase may have inhibited VC transformation to ETH. After 267 days, lactate was replaced with 2.14 mM formate as an electron donor. No major improvement was observed in the extent of PCE dechlorination to VC and ETH, but a higher percentage of total electrons were used for dehalogenation reactions than with lactate as the electron donor.

Mass and electron balance calculations were performed for the CAHs, ETH, and organic acids exiting the different columns. Approximately 60% to 70% of the electron donor addition could be accounted with 4.4% being used in dehalogenation reactions in the column study with high iron-content aquifer material. Approximately 95% of the electron donor addition could be accounted for with 7.8% associated with dehalogenation reactions in the lactate-fed low iron-content aquifer material. With formate addition in the same column, 65% of the electron donor could be accounted for but 15.6% of the donor was associated with dehalogenation reactions. The results indicate where the aquifer solids with high iron were used, hydrogen produced from lactate fermentation was partly consumed by ferric iron reducing bacteria. Formate was a more efficient electron donor for dechlorination than lactate, possibly due to supporting less iron reduction or less growth of iron reducing populations.

Victoria Strain (VS) Culture

Study 4. In column Study-4 the reductive dechlorination of TCE (0.17-0.37 mM ) with Hanford aquifer material bioaugmented with the Victoria (VS) culture, was evaluated. The VS culture can dehalogenate TCE to ethene and obtains energy from all steps of the transformation process. The aquifer solids were chemically pre-reduced with a 5 mM Na2S solution. Shortly after VS culture addition, cis-DCE concentrations in the column effluent exceeded the influent TCE concentration indicating enhanced TCE desorption and transformation. The lactate concentration was increased from 0.67 to 1.24 mM to compensate for hydrogen consumption and sulfate reduction and increases in the influent TCE concentration. About 98% of TCE was transformed to ETH at concentrations of 0.17 and 0.37 mM within a hydraulic residence time of 3.6 days. After 152 days of operation the column was destructively sampled and spatial samples were sent to Stanford for molecular analysis.

Point Mugu (PM) Culture

Study-5. Column Study-5 was conducted to evaluate the anaerobic transformation of trichloroethene (TCE) and trichlorofluoroethene (TCFE) in a continuous flow column packed with aquifer material and bioaugmented with the Point Mugu (PM) mixed culture that transforms TCE to ethene. The effect of adding different concentrations of lactate as an electron donor and changing electron donor addition to formate was also evaluated. TCE (0.17 mM) was initially fed along with lactate as an electron donor. Upon addition of the PM culture, sulfate reduction and lactate fermentation occurred and, after 130 days of flow through the column, TCE was converted to VC (50%) and ethene (50%) within a hydraulic residence time of 1.5 days. Upon adding TCFE (0.038 mM), ethene and VC concentrations decreased and 1,2-cis-dichloroethene (c-DCE) was produced. This coincided with TCFE being transformed to dichlorofluoroethene (DCFE), chlorofluoroethene (CFE) and fluoroethene (FE). With prolonged addition, TCE and TCFE were transformed to VC (80%) and ethene (20%) and CFE (80%) and FE (20%), respectively. Significant inhibition of TCE transformation resulted from TCFE addition, likely due to competitive inhibition kinetics among the chlorinated and chloro-fluorinated compounds. However, consistent with previous studies, DCFE, CFE and FE production correlated with c-DCE, VC and ethene production, respectively, showing that fluorinated analogs successfully tracked the transient transformation conditions. In this column experiment, the effect of electron donors as lactate and formate concentrations were tested during the course of the experiment to address limitations on TCE dechlorination. About 96% of the TCE was transformed to ETH at a lactate concentration of 1.0 mM, but at a reduced lactate concentration of 0.67 mM, partial transformation of TCE to cis-DCE and VC was observed. Addition of formate lessened iron reduction and resulted in more effective electron flow into dehalogenation reactions.

Molecular Studies: Spatial Distribution of the genus Dehalococcoides and species subpopulations

Study-1. The microbial community composition in column study 1 with the EV culture was analyzed with special focus on the genus Dehalococcoides. Aquifer solids from the column were sampled for molecular analysis after 170 days of column operation. The column was split in six 5 cm sections.. DNA extracted from each section served as template in real-time PCR assays to quantify Dehalococcoides organisms along the column profile. The relative abundance of Dehalococcoides species as a percentage of total Eubacteria increased from 0.5% in the first 5 cm to about 4% towards the column outflow.

The population composition of the genus Dehalococcoides was analyzed with the help of functional gene primers specific to certain Dehalococcoides strains, e.g. strain VS, and strain BVA-1. Targeted were key genes of reductive dehalogenation that have been genetically and biochemically characterized to catalyze the complete dechlorination of trichloroethene (TCE) and/or vinyl chloride (VC) to ethene. Primer sets were designed to target the vinyl chloride reductase of Dehalococcoides sp. strain VS (vcrA_VS), the vinyl chloride reductase of Dehalococcoides sp. strain BVA-1 (vcrA_BVA-1), and the trichloroethene reductase of Dehalococcoides ethenogenes strain 195, Dehalococcoides sp. strain FL2, and Bacterium PM-VC1, RC-VC2, and YK-TCE1 (tceA_195+).

The three Dehalococcoides subpopulations differ dramatically in abundance along the column profile. Whereas strain BVA-1 makes up to two third of all Dehalococcoides cells in the last 10 cm closest to the column outflow, tceA containing relatives of Dehalococcoides ethenogenes strain 195 decrease in abundance from 6% to 0.2% and 1% towards the column end. The vcrA gene of Dehalococcoides strain VS showed a more equal distribution over the column profile decreasing from 20% to about 10% with a low of 4% around 25 cm from the column inflow.

Monitoring of gene expression associated with reductive dehalogenation

Study-1. The primers for the TCE and VC reductive dehalogenases were further used in real-time reverse transcription (RT-) PCR experiments to study gene expression along the vertical horizon of the column. RNA was extracted from each 5 cm section. The RNA was reverse transcribed into cDNA which served as template in real-time PCR assays to estimate the relative expression of the described reductive dehalogenases. The expression data for each gene was normalized to the gene abundance determined in the DNA quantification experiments as described in the previous paragraph. Despite their low gene abundance the trichloroethene reductase (tceA) showed the highest expression of all three dehalogenases under investigation. The relative expression of tceA from Dehalococcoides ethenogenes strain 195 and others peaked 10 to 15 cm from the column inflow. The highest PCE reduction rates were measured in sediments obtained within 5 cm of the column inflow, so it is consistent to observe the highest tceA expression further into the column where most of the PCE has already been reduced. Vinyl chloride reduction as measured by vcrA expression of strain VS and BVA-1 showed a more uniform activity pattern over the column profile with slightly elevated expression of the vcrA of strain BVA-1 towards the reactor outflow.

Our results show that real-time (RT-) PCR can be used to quantify abundance and activity of genes involved in reductive dehalogenation of tetrachloroethene under bioremediation conditions. The protocol and oligonucleotide primers developed in this study assemble a powerful tool for the in situ monitoring and evaluation of laboratory scale bioreactors and field sites undergoing bioremediation.

Microbial diversity of PCE/TCE dechlorinating continuous flow columns

Study-2 and Study-4. The microbial community composition of two different dechlorinating column studies was analyzed by constructing 16S rRNA gene clone libraries. As described above, the columns were bioaugmented with either the TCE reducing Dehalococcoides sp. strain VS or the PCE reducing Evanite enrichment culture that contains at least two other Dehalococcoides strains in addition to strain VS. Both flow systems were operated with lactate as electron donor. Details of the operation of the columns are described in the previous section, column study 2 and column study 4. The clone libraries were constructed from 3 different horizontal sections of each of the two flow columns. Samples were taken from a location close to the column inflow port (0-2 cm), a middle section (8-10 cm) and a section near the column outflow (25-30 cm). Column solids from each section were used for DNA extraction. DNA from each column section served as template in PCR reactions with general bacterial 16S rRNA gene primers. The resulting PCR products were cloned and sequenced. No PCR products were produced using general archaeal primers indicating that Archaea were not detectable by these methods.

For the PCE-column (EV culture) a total of 200 full length 16S rRNA gene sequences comprising 29 operational taxonomic units (OTUs) were obtained, where an OTU represents a collection of sequences not more than 3% different from each other. Based on the rarefaction curve for this clone library, the probability of observing a novel OTU through additional sequencing was about 14.5%. From the column operated with TCE (VS culture) a total of 281 full length 16S rRNA gene sequences were obtained. The 16S rRNA gene sequences from the TCE-column comprise 32 OTUs with an 11.4% probability of an additional clone sequence falling into a not yet targeted OTU. Based on a chao1-estimate the total microbial community richness is 41-58 OTUs for the PCE column and 42-50 OTUs for the TCE column. For the PCE reducing column 106 clone sequences were obtained for the 2 cm section and 47 clone sequences were obtained from each of the two other column locations. In the TCE dechlorinating flow column 92 clones were retrieved from the inflow section, 94 from the middle and 95 from the outflow segment.

Both column libraries revealed a similar overall abundance of Acetobacterium and Clostridium novyi relatives, although their distribution among the different sections of each column varied. Both OTUs have their highest sequence representation in the inflow sections of the columns. OTUs found in the PCE and TCE dechlorinating column libraries with different abundance were Sedimentibacter, uncultured Thermomicrobia, Desulfitobacterium, Dehalococcoides and Clostridium sphenoides relatives. Small subunit rRNA sequences of Azospira-Dechlorosoma relatives were only found in the PCE dechlorinating column library, whereas relatives of Geobacter gribiciae, uncultured Bacteroidetes, and Desulfovibrio alcoholovorans could only be found in the TCE dechlorinating soil column. The middle and outflow section of the PCE dechlorinating column were dominated by uncultured Thermomicrobia and Dehalococcoides relatives. Sequences of both OTUs account for more than 50% of all clones obtained from these two sections. The middle and outflow section of the TCE dechlorinating column is dominated by Sedimentibacter and Geobacter gribiciae relatives, making up two thirds of all clone sequences retrieved from these two column segments.

Based on the column experiments and the microbial community analysis, we developed a working hypothesis of the flow column ecosystem as a whole and the interspecies interactions we think are essential to microbial reductive dehalogenation. In our model, the primary electron donor lactate is fermented to propionate, acetate, CO2, and hydrogen by a diverse population of fermenting bacteria, including Clostridia. Homoacetogenic bacteria, like Acetobacterium relatives, use CO2 and hydrogen to produce acetate. Acetate and hydrogen can serve as electron donors for the groups of dehalogenating microorganism, which compete with the homoacetogenic bacteria for hydrogen. While reductively dechlorinating, microorganisms like Desulfuromonas species require acetate as electron donor whereas Dehalococcoides spp. are solely dependent on the use of hydrogen as electron donor for halorespiration. In addition to hydrogen supply by the community, survival and sustained dechlorination activity of Dehalococcoides spp. is also dependent on the supply of several limiting cofactors.


B. Unusual codon usage in vinyl chloride reductase genes of Dehalococcoides species

The enzymes responsible for catabolic reduction of vinyl chloride, vinyl chloride reductases (VC-RDase), are the key enzymes for complete microbial reductive dehalogenation of chloroethenes, including the groundwater pollutants tetrachloroethene and trichloroethene. Analysis of codon usage of VC-RDase genes showed that these genes are highly unusual, characterized by a low fraction of G+C at the third position (GC3). The third position of codons in VC-RDase genes is biased toward the nucleotide T, even though available Dehalococcoides genome sequences lack tRNAs that match to codons with T at the third position. The codon usage of VC-RDase genes is clearly distinct from genes shown to be highly expressed in recent proteomic analysis for Dehalococcoides strain 195. The comparatively high level of abnormality in codon usage of VC-RDase genes suggests a recent evolutionary history that is different from most Dehalococcoides genes, including those encoding other reductive dehalogenases. One explanation is that VC-RDase genes may have been recently acquired from a heretofore unknown microorganism.



Journal Articles

Azizian, M., S. Behrens, A. Sabalowsky, M. Dolan, A. Spormann, L. Semprini. Continuous-Flow Column Study of Reductive Dehalogenation of PCE upon Bioaugmenation with the Evanite Enrichment Culture. (in review 2007).

McMurdie, P. J., S. F. Behrens, S. Holmes, and A. M. Spormann (2007). Unusual Codon Bias in Vinyl Chloride Reductase Genes of Dehalococcoides Species. Appl. Environ. Microbiol. 73:2744-2747.


Conference Abstracts and Presentations

Azizian, Mohammad F., Mark E. Dolan, Peter Ruiz-Haas, James D. Ingle, and Lewis Semprini. (2007). Effect of pre-reduction of aquifer material on PCE reductive dechlorination in a continuous-flow column study. American Chemical Society, Division of Environmental Chemistry, Vol. 47 No.1, 560-565.

Azizian, M., S. Behrens, A. Sabalowsky, M. Dolan, A. Spormann, L. Semprini (2006). Continuous-Flow Column Studies of Reductive Dehalogenation of CAHs with Evanite Enriched Culture: Kinetics, Inhibition, and Monitoring of Microbial Activity (abstract and poster), Subsurface Biosphere Initiative (SBI) workshop/IGERT Retreat, Newport, Oregon (June 18-21).

Behrens, S., J. McMurdie, G. Meshulam, A. Spormann (2005). Evaluation of a CARD-FISH Protocol for the Quantification of Dehalococcoides sp. in Soil. 2005 ASM General Meeting (June 5-9).

Behrens, S., Azizian, M., McMurdie, J., Sabalowsky, A. Dolan, M., Semprini, L., Spormann, A. M. (2006). Monitoring Gene Abundance and Expression of Reductive Dehalogenases Involved in Complete Dechlorination of PCE Under Continuous Flow Conditions. Poster presentation at the 11th International Symposium on Microbial Ecology, Vienna, Austria (August).

Sabalowsky, A.R. and L. Semprini (2005). Alkynes as Reversible Inhibitors for Probing Mechanisms of Reductive Dehalogenation of Chloroethenes. Joint International Symposia for Subsurface Microbiology (ISSM 2005) and Environmental Biogeochemistry (ISEB XVII), Jackson Hole, Wyoming (August 14-19).

Semprini, L., M. Azizian, A. Sabalowsky, M. Dolan, P. Ruiz-Hass, J. Ingle, S. Behrens, A. Spormann (2005). A Continuous Flow Column Study of Anaerobic PCE Transformation with the Evanite Culture and Hanford Aquifer Solids. Joint International Symposia for Subsurface Microbiology (ISSM 2005) and Environmental Biogeochemistry (ISEB XVII), Wyoming (August 14-19).

Semprini, L. Mark E. Dolan and M. F. Azizian (2007). Anaerobic transformation of trichloroethene and trichlorofluoroethene in a continuous flow column study. American Chemical Society, Division of Environmental Chemistry, ACS national meeting, Boston, MA (August 19-23).

Semprini, L. and M. Azizian. Continuous-Flow Column Studies to Evaluate the Effects of Different PCE and TCE Concentrations and Lactate Addition on Reductive Dehalogenation (2007). (abstract and poster), Partners in Environmental Technology Technical Symposium and Workshop, Washington D.C. (December 4-6 ).


Supplemental Keywords: biotransformation; groundwater; NAPL; VOCs; bioremediation; chlorinated solvent; remediation technologies; in-situ; characterization