Article pubs.acs.org/est
Transformation of Chloroform in Model Treatment Wetlands: From Mass Balance to Microbial Analysis Yi Chen,†,‡,§ Yue Wen,*,† Junwei Zhou,† Qi Zhou,† Jan Vymazal,‡ and Peter Kuschk§ †
Key Laboratory of Yangtze Water Environment of Ministry of the State Education, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, P.R. China ‡ Department of Landscape Ecology, Faculty of Environmental Sciences, Czech University of Life Sciences, Prague 16521, Czech Republic § Department of Environmental Biotechnology, Helmholtz Centre for Environmental Research UFZ, Permoserstrasse 15, 04318 Leipzig, Germany S Supporting Information *
ABSTRACT: Chloroform is one of the common disinfection byproducts, which is not susceptible to degradation and poses great health concern. In this study, the chloroform removal efficiencies and contributions of sorption, microbial degradation, plant uptake, and volatilization were evaluated in six model constructed wetlands (CWs). The highest chloroform removal efficiency was achieved in litteradded CWs (99%), followed by planted (46−54%) and unplanted CWs (39%). Mass balance study revealed that sorption (73.5−81.2%) and microbial degradation (17.6−26.2%) were the main chloroform removal processes in litter-added CWs, and that sorption (53.6−66.1%) and plant uptake (25.3−36.2%) were the primary contributors to chloroform removal in planted CWs. Around 60% of chloroform got accumulated in the roots after plant uptake, and both transpiration and gas-phase transport were expected to be the drivers for the plant uptake. Sulfate-reducing bacteria and methanogens were found to be the key microorganisms for chloroform biodegradation through cometabolic dechlorination, and positive correlations were observed between functional genes (dsrA, mcrA) and biodegradation rates. Overall, this study suggests that wetland is an efficient ecosystem for sustainable chloroform removal, and that plant and litter can enhance the removal performance through root uptake and microbial degradation stimulation, respectively.
1. INTRODUCTION Chlorine disinfection is a worldwide technology for controlling microorganisms in water and wastewater. Chlorination of water and wastewater has been reported to produce disinfection byproducts (DBPs), some of which are carcinogenic and consequently raise health and regulatory concern.1,2 Chloroform is the most frequently detected DBPs in the effluent of drinking water treatment plants (DWTP) and wastewater treatment plants (WWTP), and there is an urgent need for regulating the level of chloroform by the government of many countries due to its potential health risks to humans (i.e., bladder cancer and adverse reproductive effects).3 In particular, chloroform is also the most commonly detected volatile organic compound (VOC) in groundwater.4 Thus, development of methods to efficiently remove chloroform from water and wastewater is potentially of much practical value. Adsorption and microbial degradation (aerobic/anaerobic) have been applied to eliminate chloroform from wastewater and groundwater. Activated carbon, charcoal, and lignite have been demonstrated to be efficient adsorbents for chloroform removal.5−7 Unlike adsorption, microbial degradation of chloroform is relatively slow owing to its highly chlorinated structure.8 Previous studies have demonstrated that chloroform © XXXX American Chemical Society
can hardly serve as a carbon and energy source for microbial growth, and a primary substrate is needed to drive its biodegradation through cometabolism.8 It has been suggested that chloroform can be biodegraded via either oxidative or reductive pathway. In aerobic environment, some bacteria are capable of transforming chloroform to CO2 using propane, butane, and ammonia as primary substrates,9−11 whereas in anaerobic environment, methanogens, sulfate reducers, and fermenting bacteria can convert chloroform to lower chlorinated methanes using volatile fatty acids (VFAs) as primary substrates.8,12,13 However, the high cost of both adsorbent and primary substrate has limited the use of adsorption and biotechnology in engineering applications. Constructed wetlands (CWs) have advantages such as low implementation costs, simple operation, and efficient removal of emerging contaminants.14−16 Previous studies have shown that the use of CWs with added plant litter as tertiary treatment systems resulted in a good chloroform removal efficiency, with Received: December 31, 2014 Revised: March 18, 2015 Accepted: April 22, 2015
A
DOI: 10.1021/es506357e Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
and the duration of the experiment was 100 d, which included 20 batches (each batch lasted for 5 days). Before each batch, the residual chlorine in the influent was quenched with 10% sodium sulfite to prevent chloroform production in wetlands. 2.3. Adsorption Equilibrium Experiments. Before the start of the experiment, fresh litter and gravel (with biofilm) were autoclaved at 120 °C for 20 min to prevent any microbial activity. Subsequently, 57.0 g of gravel (ϕ 8−13 mm) and 0.35 g of litter were, respectively, added to 50 mL centrifuge glass tubes containing different chloroform concentrations (50, 100, 150, 200, and 250 μg L−1). Each sample was shaken with contact time varied from 0 to 72 h (150 rpm, 298 ± 1 K), and 10 mL of the supernatant were collected and the residual chloroform concentration was quantified at time intervals. Once the equilibrium time (48 h) had been determined, isotherms were constructed and the data were fitted using the Langmuir and Freundlich models. A control test was performed without the gravel and litter, and all sorption experiments were conducted in triplicate. 2.4. Inhibition Degradation Experiment. Thirty 250 mL serum bottles (5 sets × 6 bottles), each containing 300 g of gravel (collected from W2), were simultaneously spiked with 65 μg L−1 chloroform without headspace. The bottles were previously flushed with pure N2 to achieve anaerobic environment and were divided into five sets: A (autoclaved at 120 °C for 20 min to prevent any microbial activity), B (unautoclaved), C (addition of 20 mM molybdate as sulfatereducing bacteria (SRB) inhibitor), D (addition of 10.5 g L−1 2bromoethanesulfonic acid (BES) as methanogens (MG) inhibitor), and E (addition of 20 mM molybdate +10.5 g L−1 BES). The experiment was conducted at a pH of 7.0 and the serum bottles were stirred at 150 rpm in dark at 25 °C. Subsequently, the bottles were successively withdraw and 10 mL supernatant were collected from each bottle in an anaerobic glovebox at the following time intervals: 0, 6, 12, 24, 48, and 72 h. Six serum bottles without gravel were prepared as control to monitor the hydrolysis and volatilization of chloroform. The control experiment showed that the abiotic loss from hydrolysis and volatilization was negligible (Figure 2), thus indicating that the chloroform removal in A−E batches were mostly owing to sorption and/or biodegradation. 2.5. Sampling Procedure and Analysis of Aqueous, Solid and Gaseous Samples. Water samples were collected in the midpoint of the microcosms at predetermined time intervals (0−120 h) during each batch. Then, water samples were membrane-filtered (0.22 μm) and stored in 15 mL amber glass vials. The vials were filled headspace-free and capped with polytetrafluoroethylene (PTFE) lined silicone septa until extraction. The chloroform emitted in each wetland was trapped by passive samplers (RADIELLO, RAD130, Supelco), and the trapped chloroform was recovered using carbon disulfide before analysis.17 For solid samples (gravel, litter, and plant), the adsorbed chloroform was extracted using hexane and acetone mixture (50:50) for 5 min, and 1.5 mL of the supernatant was used for analysis. The extraction recoveries were 70−85% for gravel and 45−65% for plant and litter. The chloroform and biotransformation products (DCM and MCM) were measured in duplicate using a gas chromatograph (GC; Agilent 7890A; 30 × 0.32 mm HP-5 phenyl methyl silicone column) with an electron capture detector (ECD), as described previously.17 The limit of detection (LOD) for all the analytes was 0.1 μg·L−1 and the recovery was between 80% and 120%. The calculations of chloroform removal rates for plant uptake,
the half-lives of chloroform in CWs (1.0−1.7 days) in orders of magnitude shorter than those observed in other systems (i.e., aquifers, maximum: 550 days).17 However, the main mechanism of chloroform removal in CWs is still not clear because the knowledge on the contributions of microbial degradation, sorption, and other elimination processes is limited. Quantitative evaluation of organic contaminant distribution among wetland compartments (i.e., aqueous phase, gaseous phase, vegetation, and bed substrate) is analytically difficult, especially for trace organic pollutants (ng−μg L−1).16,18,19 To date, the ultimate fate of DBPs in CWs is still unclear because of the lack of complete mass balance. Furthermore, the influence of plant biomass on the fate of DBPs and the transformation of DBPs in wetland plants are not well-known. Microbial degradation is considered to be an elimination pathway for sustainable removal of DBPs in CWs; however, the contribution of specific microorganisms as well as the abundance of key genes responsible for DBPs biodegradation is still unknown. Thus, the objective of the present study was to evaluate the fate of a model DBP (chloroform) in different types of lab-scale CWs, including planted and unplanted CWs with or without litters. Based on the calculation of mass balance, the contaminant removal processes could be subdivided into (1) nondestructive processes such as sorption, volatilization, and plant uptake and (2) destructive removal processes such as microbial degradation and its relationship with functional genes (dsrA and mcrA). In particular, the roles of microbial degradation as sustainable contaminant removal mechanisms in CWs were examined.
2. MATERIALS AND METHODS 2.1. Chemicals. Analytical standards for chloroform were purchased from Sigma−Aldrich. Dichloromethane (DCM) and chloromethane (MCM) standards were obtained from Dr. Ehrenstorfer GmbH (Germany). 2.2. Design and Operation of the CW. Six sequencing batch CW microcosms, each with a bulk volume of 0.045 m3 (length: 0.3 m, width: 0.3 m, height: 0.5 m) and a pore volume of 12 L, were established in this study. The six types of CWs were as follows: unplanted control (W0), litter-added microcosms (W1, 100 g; W2:200 g), planted microcosms (W3, 22 plants m−2; W4, 40 plants m−2), and planted plus litter-added microcosms (W5, 22 plants m−2, 100 g of litter). All the microcosms were filled with gravel (ϕ 8−13 mm, porosity = 0.4) and three of them (W3, W4, and W5) were planted with cattail (Typha latifolia). The wetland microcosms were maintained in a temperature-controlled (25 ± 1 °C) greenhouse since 2005. The details of the microcosm design had been illustrated in our previous study.17 Cattail litter was cut into pieces (1.0−2.0 cm), mixed uniformly with gravel, and used as a potential bioadsorbent and electron donor for reductive dechlorination of chloroform. The source and characteristics of litter and gravel are provided in Supporting Information (SI) Tables S1−S2. The feedwater for the CWs was the effluent from a neighboring WWTP with chlorine disinfection, and the concentration of chloroform was 35−70 μg L−1. The other inflow characteristics had been described in our previous study.17 The wetland microcosms were operated as a batch system with pulse loading, which were filled with wastewater at the start of each batch and gravity drained within 1 h prior to the next batch. All the treatments (W0−W5) were triplicated B
DOI: 10.1021/es506357e Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
R biodegradation = Δm − R volatilization − R sorption − R plant uptake
sorption, volatilization, and microbial degradation are listed in SI Table S3. 2.6. Functional Genes and Microbial Community Characterization. For the characterization of functional genes and microbial community, samples (200 g) were collected from three depths (5, 20, and 40 cm) of wetland microcosms on 10, 50, and 90 days, respectively. The samples collected from different depths were combined for DNA extraction using E.Z.N.A. Soil DNA Kit (OMEGA biotek). The key functional genes for methanogenesis (mcrA) and sulfate reduction (dsrA) were quantified by real-time PCR performed on ABI STEPONE PLUS detection system. Met1f/Met1r and dsr-1F 5′/dsr-500r 5′ were used as the primers for mcrA and dsrA, respectively, and the amplification procedure and thermal cycling conditions had been described in SI. To characterize the microbial community in the CWs, high-throughput 454 GSFLX pyrosequencing of the 16S rRNA gene was conducted according to standard protocols.21 The raw sequencing data have been deposited to the NCBI Sequence Read Archive (SRA) under Accession no. SRP034645. The details of pyrosequencing and analysis have been presented in SI and in a previous study.20 2.7. Chloroform Mass Balance. The removal efficiency for the water phase Rw (%) in the CWs was based on the influent and effluent loading rates of chloroform as follows: Rw =
L in − Lout × 100 L in
(5)
2.8. Statistical Analysis. All tests were performed at least in duplicate if not otherwise noted. ANOVA was used to test the significance of results using SPSS version 19.0 software, and p < 0.05 was considered to be statistically significant. The direct and indirect effects on chloroform biodegradation and their intermediate relationships were identified by Path Analysis using AMOS version 21 (SPSS, IBM).
3. RESULTS AND DISCUSSION 3.1. Treatment Efficiency. In the present study, the chloroform removal in wetland microcosms followed two patterns: (1) Initial stage (days 1−60, batches 1−12): the chloroform removal was rapid but unstable and (2) Terminal stage (days >90, batches >18): the chloroform removal was stable but slow. Given that the chloroform removal kinetics followed a similar pattern after day 90, batch 20 (days 96−100) was chosen as the stable period for analysis. As shown in SI Figure S1, the chloroform removal efficiencies varied among treatments from 38.6% (W0) to 99.1% (W2) during the stable period. When compared with the litter-added units (W1, W2, W5), the planted units (W3, 45.6%; W4, 53.8%) exhibited significantly lower chloroform removal efficiencies, and the densely planted unit presented more efficient chloroform removal than the sparsely planted one. The removal of chloroform in the CWs followed firstorder kinetics (R2 > 0.6), and the attenuation rate constants ranged from 0.067 d−1 for W0 to 0.456 d−1 for W2 (Figure 1).
(1)
The chloroform mass balance in the CWs was calculated using eqs 2 and 3 as follows: Δm = L in − Lout
(2) n
Δm = A−1 × nT −1 ×
∑ (Ci(in) × Vi(in) − Ci(out) × Vi(out)) i=1
(3)
where Lin/out (μg·m−2·d−1) is the areal chloroform load for the inflow or outflow, respectively; Δm is the chloroform removal rate (μg·m−2·d−1); i is the number of batch in sequence (i = 1, 2, 3...20); n is the total number of batches; Ci(in) and Ci(out) are the chloroform concentrations in the influent and effluent of batch i, respectively (μg·L−1); Vin and Vout are the volume of pore water in the influent and effluent, respectively (L); A is the area of gravel (m2; A = 0.09 m2); and T is the retention time for every batch (d; T = 5 days). The chloroform removal rate (Δm) (eq 2) can also be given as the sum of the rates of microbial degradation (Rbiodegradation), volatilization (Rvolatilization), sorption on bed substrate and litter (Rsorption), plant uptake (Rplant uptake), photolysis (Rphotolysis), and hydrolysis (Rhydrolysis) (see eq 4):
Figure 1. First-order attenuation rates and half-lives of chloroform in W0−W5 microcosms during the stable period (batches 18, 19, and 20). Error bars represent ± standard deviation of the mean for measurements made during the stable period (n = 3). W0: unplanted control; W1: 100 g cattail litter; W2: 200 g cattail litter; W3: planted with Typha latifolia, 22 plants/m2; W4: planted with Typha latifolia, 40 plants/m2; W5: 100 g cattail litter +22 plants/m2.
The higher attenuation rate constant observed in litter-added CWs indicated that cattail litter could significantly enhance chloroform removal in CWs. These attenuation rates are comparable with those measured previously in nitrifying biofilters,22 but lower than those determined in pure culture systems.23−25 Figure 1 shows that the half-lives of chloroform in the CWs ranged from 1.5 days (W2) to 10.3 days (W0), which is lower than those previously measured in aquifer sites (1−550 days).26 The short half-lives of chloroform in the planted (6.6− 7.2 days) and litter-added CWs (1.5−3.1 days) suggested that
Δm = R biodegradation + R volatilization + R sorption + R plant uptake + R photolysis + R hydrolysis
(4)
Given that the hydrolysis constant of chloroform is very low (kHYD= 2.32 × 10−8 h−1) and that penetration of sunlight into the substrate of CWs is difficult, both hydrolysis and photolysis are considered to be insignificant in the chloroform mass balance. Thus, the rate of microbial degradation of chloroform can be obtained from the following equation: C
DOI: 10.1021/es506357e Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
chloroform by plants. In addition to transpiration, dissolved volatile compounds in water can also be taken up by plants via gas-phase transport in aerenchyma.31 In the present study, the transpiration-driven uptake of chloroform was 159 (W3) and 210 μg m−2 d−1 (W4) according to the Dettenmaier model,32 which was lower than the detected uptake rates (W3:184.0 μg m−2 d−1, W4:307.3 μg m−2 d−1) (SI Table S4), indicating the presence of other pathways (i.e., gas-phase transport) that account for the uptake of chloroform in planted CWs. Similarly, Bendix et al. also found that T. latifolia could accelerate gasphase transport along the soil−plant−atmosphere system with pressurized internal mass flows33 and may enhance the uptake rates of chloroform in planted CWs. Plant uptake with subsequent phytodegradation is a sustainable removal mechanisms of organic chemicals in CWs. In this study, very limited accumulations of dechlorination metabolites (SI Figure S4b) and inorganic chlorine (1.34 × 10−4 mol g−1) in plant tissues indicated that phytodegradation might play a minor role on chloroform removal in planted CWs. Further studies based on 14 C-chloroform are deserved to explore its fate and metabolites in wetland plants. Considering the chloroform plant uptake rates (184 μg m−2 d−1) and uptake capacity (123.7 μg g−1) obtained in the seasonal study, annual harvest is needed to avoid the risk of chloroform release back to water. 3.2.3. Volatilization. The chloroform emissions in CWs were measured using passive samplers. As shown in SI Table S4, the calculated volatilization rates were higher in planted CWs (13.1−21.5 μg m−2 d−1), when compared with those in unplanted ones (3.4−3.8 μg m−2 d−1). Phytovolatilization might be the main contributor to the increase in chloroform emissions in planted CWs, considering that surface volatilization was similar between different systems. Volatilization could be regarded as the primary mechanism for chloroform removal in the surface flow constructed wetlands (SF CWs) owing to the relatively high Henry constant (3.67 × 10−3 m3 mol−1) for chloroform and quick diffusion rates at the water−air interface.34 However, in the present study, chloroform emissions only accounted for 2.0−2.5% of the total removal of chloroform in planted systems, suggesting that volatilization had a minor effect on chloroform removal in planted subsurface flow constructed wetlands (SSF CWs). When compared with SF CWs, the low emissions observed in the SSF CWs in this study might probably be owing to the water level below the bed surface, resulting in the limitation of water−air transfer via diffusion in the SSF CWs. This result is in accordance with the previous findings obtained using planted gravel filters and soil filters.29,35 Alternatively, the retention of chloroform in plant tissues (i.e., roots, stems, and leaves) via hydrophobic sorption could also inhibit volatilization from leaves. 3.2.4. Microbial Degradation. The microbial degradation rates in the CWs were obtained from the mass balance equation (eq 5). As shown in SI Table S4, the microbial degradation rates of chloroform were 275.0, 288.6, and 200.2 μg m−2 d−1 in W1, W2, and W5, respectively, which were significantly higher than those in CWs without litter (27.7−64.6 μg m−2 d−1), indicating that the addition of cattail litter could significantly enhance the microbial degradation of chloroform. Molecular H2 is a preferred electron donor for reductive dechlorination of chloroform, and it is expected to be produced by fermentation of organic carbon in litter-added CWs as the accumulation of VFAs (i.e., acetic acid)36 and the detection of H2 producers (i.e., Clostridium sp.). SRB and MG were widely distributed in litter-added CWs (SI Figure S5−S6), and it could jeopardize
chloroform could be efficiently removed in a hydraulic retention time (HRT) normally used in CWs. 3.2. Removal Pathways. 3.2.1. Sorption. As shown in SI Table S4, chloroform was widely detected in the gravel, litter, and plant tissues after operation for 100 days. The highest sorption rates were observed in the litter-added CWs, and the sorption rates ranged from 811 to 955 μg m−2 d−1. The efficient sorption in the litter-added CWs might be owing to the hydrophobic interactions between chloroform and organic carbon in the litter (SI Table S2) and biofilm of gravel. Previous studies had also reported that organic-rich biomass exhibited good performance with regard to chloroform sorption with removal efficiencies of 74−99%.27,28 In the present study, the sorption isotherm showed that the sorption of chloroform by gravel and litter was well described by the Freundlich model, with the sorption coefficient (kF) of 94.0 and 0.35 L kg−1 for litter and gravel, respectively (SI Figure S3). Furthermore, the Freundlich affinity constant (n ≈ 1) calculated in this study suggested a linear sorption of chloroform to gravel and litter, implying that sorption affinity is independent of concentration and can be determined by the solid−water distribution coefficient (kd). Moreover, the measured aqueous chloroform concentrations (35−70 μg L −1 ) in the influent were significantly lower than the equilibrium aqueous concentration (428 μg L−1) predicted by adsorbed concentration on the gravel (0.15 μg g−1) and kd values measured with autoclaved gravel (0.35 L kg−1). Thus, sorption may serve as a long-term sink for chloroform removal in CWs. Considering that biotransformation might be coupled to sorption, the sink functions of CWs may be more pronounced than predicted. 3.2.2. Plant Uptake. As shown in SI Table S4, the chloroform uptake rates ranged from 179.2 to 307.3 μg m−2 d−1 in the planted CWs, suggesting that plant uptake is an important mechanism that can contribute to the removal of chloroform. Previous studies had suggested that the direct uptake of organics by plants via transpiration is generally indicated by low to intermediate log Kow values ranging from 0.5 to 3.0.29 Such compounds are lipophilic enough to move through the lipid bilayer of the membranes, as well as adequately water-soluble to dissolve into the cell fluids. Chloroform is a compound with moderate hydrophobicity and a log Kow value of 1.97, and could enter and easily translocate within the plant via passive transport mechanism. In the present study, to understand the distribution and transport of chloroform in plant tissues, the content and concentrations of chloroform were detected in plant tissues at the end of the experiment. As shown in SI Figure S4a, chloroform was detected in both belowground and aboveground parts of the plant, and the chloroform content in the roots (1079−1750 μg) was significantly higher than that in the stems (292−530 μg) and leaves (284−486 μg). This finding suggested that at least 40% of chloroform was transported from the roots to stems and leaves, which is higher than the ratio of other organic pollutants reported in a previous study.30 Considering its moderate hydrophobicity (log Kow = 1.97), chloroform might have been loosely bound to the root surface, which could have facilitated its translocation within the plant via the transpiration stream.31 The water balance in W3 and W4 revealed transpiration rates of 8.5 and 9.4 L m−2 d−1, respectively, making up 31−35% of the inflow water. The high rate of plant transpiration (31−35% of inflow) was positively correlated to the efficient chloroform uptake (25−36% of chloroform removal) in planted CWs, suggesting that transpiration is the main driver for the uptake of D
DOI: 10.1021/es506357e Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
SRB inhibitor had a more negative influence on chloroform removal than MG inhibitor. This finding suggested that both SRB and MG contributed to chloroform biodegradation, and that the contribution of SRB was higher than that of MG, which is in accordance with the previous results obtained using anaerobic sludge.13,37 In Test E, the chloroform removal rate significantly decreased and was close to that obtained in Test A (autoclaved) when both the inhibitors were added, suggesting that SRB and MG were mainly responsible for chloroform biodegradation in CWs. Despite the facts mentioned above, the dechlorination of chloroform by Dehalobacter cannot be completely ruled out in this study. Microbial Community and Functional Genes. The taxonmic classifications of bacterial and archaeal genera from the CWs were summarized in SI Figures S5−S6. Totally, 325 bacterial genera and 16 archaeal genera were observed by highthroughput sequencing. Among them, 19 bacterial genera and 11 archaeal genera were detected as the dominant genera (>1% of total composition) in CWs. The genus level identification illustrated that Methanosaeta (an aceticlastic MG) and Desulfovibrio (a hydrogenotrophic SRB) were respectively the main archaeal (39.9%) and bacterial genera (0.64%) in the gravel of litter-added CW (W1) (SI Figures S5−S6). Previous literature reported that both Methanosaeta and Desulfovibrio could involve in the cometabolic transformation of chloroform,37,38 supporting the superior chloroform degradation rates observed in litter-added CWs. Simultaneous dihaloelimination and hydrogenolysis was reported to be catalyzed by methylCoM reductase (mcr), a key enzyme of methanogenesis.39 In order to further prove the roles of MG and SRB on chloroform biodegradation, the functional genes related to methyl-CoM reductase A (mcrA) and dissimilatory sulfite reductase A (dsrA) were quantified by real-time PCR. Results showed that the numbers of copies of both dsrA and mcrA were significantly higher in the litter-added CWs than those in the CWs without litter (SI Figure S8). Notably, the chloroform biodegradation rates increased with the copy numbers of both the genes, and both these parameters were significantly linearly correlated to each other (Figure 3). It should be noted that the strong correlation does not represent their causal relationship. In this study, structural equation model and path analysis were applied to test different causal models explaining the variation in chloroform degradation rates using predictive variables. Model
the dehalorespiratory of chloroform via competition with halorespirers (i.e., Dehalobacter) for H2.8 The possible pathways of H2 competition in CWs were proposed in SI Figure S7. In contrast, SRB and MG were also reported to benefit the chloroform degradation through cometabolism.13,37 The role of MG and SRB in chloroform biodegradation in CWs was determined using microbial inhibitors. Inhibition Degradation Experiment. As shown in Figure 2, in Test A (autoclaved), about 70% of the chloroform was
Figure 2. Chloroform degradation under various inhibition conditions. A: autoclaved; B: unautoclaved; C: SRB inhibition; D: MG inhibition; E: coinhibition of SRB and MG; Control: without added gravel. BES: 2-bromoethanesulfonic acid.
removed within 80 h, suggesting that abiotic effect was dominant in the chloroform removal process in CWs. As volatilization and hydrolysis are considered to be marginal in airtight serum bottles (Figure 2), the rapid removal of chloroform in the autoclaved test could possibly be owing to the adsorption on the gravel biofilm. In Test B (unautoclaved), complete removal of chloroform was achieved within 80 h. From the comparison between Tests A and B, it could be concluded that biodegradation attributed to 30% of the total removal of chloroform. Furthermore, in both the tests, the chloroform removal rates decreased to some extent when SRB inhibitor (molybdate) or MG inhibitor (BES) was added, and
Figure 3. Relationships between the chloroform biodegradation rates and copy numbers of sulfate-reduction genes (dsrA) (a) and methanogenesis genes (mcrA) (b). E
DOI: 10.1021/es506357e Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology suggested that the copy numbers of dsrA and mcrA genes had significant direct effects, while other variations (i.e., redox potential) had indirect or no effects on the chloroform biodegradation (SI Figure S9). These findings suggested that the abundant dsrA and mcrA genes might contribute to effective chloroform degradation in litter-added CWs. Reductive dehalogenases (RDase; i.e. CfrA) genes identified from Dehalobacter strains were demonstrated to be responsible for chloroform dechlorination.40,41 In the present study, Dehalobacter sequences were always below the detection limit of Roche 454 high-throughput pyrosequencing (one per 20 thousand bacterial 16S rRNA gene sequences). Considering that the bacteria concentrations in CWs are about 9.7 × 105 1.1 × 106 cell mL−1,42 only the bacteria concentrations over 50 cell mL−1 can be detected by the pyrosequencing approach. According to this, the presence of chloroform-respiring Dehalobacter (
