Interactions between Perchlorate and Nitrate Reductions in the

ARTICLE pubs.acs.org/est

Interactions between Perchlorate and Nitrate Reductions in the Biofilm of a Hydrogen-Based Membrane Biofilm Reactor He-Ping Zhao,*,† Steve Van Ginkel,† Youneng Tang,† Dae-Wook Kang,† Bruce Rittmann,† and Rosa Krajmalnik-Brown† †

Swette Center for Environmental Biotechnology, Biodesign Institute at Arizona State University, P.O. Box 875701, Tempe, Arizona 85287-5701, United States

bS Supporting Information ABSTRACT: We studied the microbial functional and structural interactions between nitrate (NO3) and perchlorate (ClO4) reductions in the hydrogen (H2)-based membrane biofilm reactor (MBfR). When H2 was not limiting, ClO4 and NO3 reductions were complete, and the MBfR’s biofilm was composed mainly of bacteria from the ε- and β-proteobacteria classes, with autotrophic genera Sulfuricurvum, Hydrogenophaga, and Dechloromonas dominating the biofilm. Based on functional-gene and pyrosequencing assays, Dechloromonas played the most important role in ClO4 reduction, while Sulfuricurvum and Hydrogenophaga were responsible for NO3 reduction. When H2 delivery was insufficient to completely reduce both electron acceptors, NO3 reduction out-competed ClO4 reduction for electrons from H2, and mixotrophs become important in the MBfR biofilm. β-Proteobacteria became the dominant class, and Azonexus replaced Sulfuricurvum as a main genus. The changes suggest that facultative, NO3-reducing bacteria had advantages over strict autotrophs when H2 was limiting, because organic microbial products became important electron donors when H2 was severely limiting.

’ INTRODUCTION Perchlorate (ClO4) is a strong oxidizing agent that has been widely used in rocket propellants, explosives, and fireworks.1 ClO4 contamination in water is a health problem because it interferes with the production of thyroid hormones that are needed for pre- and postnatal growth and development as well as for normal metabolism in adults.2 Although the US EPA has not yet established a maximum contaminant level (MCL) for ClO4, some states have established cleanup levels ranging from 2 to 18 μg/L for ClO4 in drinking water.3 Strategies for removing ClO4 from water include activated carbon, membrane filtration, ion exchange (IX), chemical or electrochemical reduction, and biological reduction.1,4,5 Biological processes have gained interest in the past decade due to their ability to destroy ClO4 and their lower cost.3 Among the biological-reduction methods is the H2-based membrane biofilm reactor (MBfR), in which H2 gas is delivered to bacteria by diffusion through the wall of a gas-transfer fiber. H2 is an electron donor that drives the respiratory reduction of one or a mixture of oxidized contaminants, e.g., nitrate (NO3), ClO4, selenate, chromate, and chlorinated solvents.69 The biological reduction of ClO4 involves ClO4 reductase (pcrABCD), which reduces ClO4 to chlorite (ClO2),10 and chlorite dismutase (cld), which catalyzes the reduction of chlorite r 2011 American Chemical Society

(ClO2) to chloride (Cl) and oxygen (O2),11 which is subsequently reduced to H2O. Because the cld gene is not specific to perchlorate-reducing bacteria (PRB), targeting a pcr gene is most appropriate for detecting PRB.12 A frequent cocontaminant with ClO4 is ammonium nitrate (NH4NO3), which is a main component in rocket fuel and explosives. In these situations, the concentration of NO3 typically is 2 to 5 orders of magnitude higher than ClO4.13 NO3 and its immediate reduction product nitrite (NO2) also are serious water contaminants. The USA drinking-water standards for NO3 and NO2‑ are 10 and 1 mg N/L, respectively.14 Bacteria also readily reduce NO3 and NO2‑ to harmless N2 gas, a process called denitrification. NO3 reductases (Nar or Nap) reduce NO3 to NO2; NO2 is reduced to nitric oxide (NO) by either a copper (Cu-Nir) (nirK) or a cytochrome cd1 NO2‑ reductase (Cd-Nir) (nirS); NO reductase (Nor) reduces NO to nitrous oxide (N2O); and N2O reductase (Nos) reduces N2O to N2.15,16 Figure S1 in the Supporting Information summarizes Received: July 25, 2011 Accepted: October 21, 2011 Revised: October 19, 2011 Published: October 21, 2011 10155

dx.doi.org/10.1021/es202569b | Environ. Sci. Technol. 2011, 45, 10155–10162

Environmental Science & Technology

ARTICLE

Figure 1. (A) Perchlorate, nitrate, and nitrite concentrations in MBfR. (B) Nitrate and perchlorate percentage removals.

the reactions and enzymes for all the reduction reactions mentioned here. Nitrate and perchlorate appear to have complex interactions as respiratory electron acceptors, and past studies have given contradictory patterns. Herman and Frankenberger17 and Choi and Silverstein18 reported that NO3 inhibited ClO4 reduction, particularly when the electron donor was limited. However, Nerenberg et al.19 showed that inhibition between NO3 and ClO4‑ reductions was trivial in a H2-based MBfR as long as sufficient H2 was available. Interactions between NO3 and ClO4 may depend on the members of the microbial community and what respiratory genes they possess. PRB isolated so far are phylogenetically diverse, with members in the α, β, γ, and ε subclasses of the Proteobacteria phylum.1,4 However, most of the known PRBs are in the β subclass and closely related to each other: 16S rRNA gene sequences diverging by less than 1%.4 Most of the known PRB are denitrifiers,1,9,1921 although a few exceptions have been noted: e.g., Pseudomonas chloritidismutans. sp. nov. and Pseudomonas sp. PDA.21,22 Among the bacterial strains that reduce NO3 and ClO4, several reduce ClO4 to chloride ion (Cl) using dissimilatory nitrate reductases, along with specialized perchlorate reductases.11,23 For example, perchlorate and nitrate reductases were found in the culture perclace.20 However, the presence of both types of reductases does not necessarily mean that both reduction processes occur simultaneously. For instance, Xu et al.21 observed a small rate of denitrification activity by Dechlorosoma sp. KJ when grown on chlorate (ClO3) or ClO4, but strain KJ enriched with NO3 as the sole electron acceptor did not reduce ClO4 immediately. They concluded

that separate substrate-induced enzyme systems were present for NO3 and ClO4 in strain KJ. The objective of this study was to understand interactions between simultaneous NO3 and ClO4 reductions in the biofilms of a H2-based MBfR. We studied reduction patterns of NO3 and ClO4 under sufficient or limiting availability of H2 and with different relative inputs of NO3 and ClO4. We used quantitative real-time PCR (qPCR) to monitor changes in the microbial community according to specific groups: total bacteria by 16S rRNA gene, PRBs by the functional gene pcrA, and denitrifiers by the functional genes nirK and nirS. We also analyzed changes in bacterial community structure and diversity in the MBfR through pyrosequencing. Based on our new understanding, we are able to resolve the apparently contradictory results in past studies.

’ MATERIALS AND METHODS Experimental Setup. We used an MBfR setup similar to that of Chung and Rittmann.7 The MBfR (Figure S2 in the Supporting Information) was a 55-mL continuous-flow system consisting of two glass tubes connected with Norprene tubing and plastic fittings. One glass tube contained 32 hollow-fiber membranes (nonporous polypropylene fiber, 200 μm OD, 100110 μm ID, wall thickness 5055 μm; produced by Teijin, Ltd., Japan), each with a 25-cm active length. The fibers were glued into a H2supply manifold at the bottom, while the top of each fiber was sealed. A second glass tube contained 10 25-cm-long fibers that served as sampling coupons for microbial ecology studies; they were similarly glued into a H2 supply manifold. The MBfR was completely mixed due to a high recirculation rate of 100 mL/min 10156

dx.doi.org/10.1021/es202569b |Environ. Sci. Technol. 2011, 45, 10155–10162


ARTICLE

achieved with a peristaltic pump (Master Flex, model 7520-40, Cole-Parmer Instrument Company, U.S. A); therefore, the concentration in the MBfR was equal to the effluent concentration. A peristaltic pump (Rainin Dynamax Peristaltic pump, model RP-1) and PVC tubing (Rainin Silicone pump tube, Yellow Blue) were used to provide an influent feed rate of 0.25 mL/min. The H2 gauge pressure was 17.5 psi (1.2 atm) for all experiments. Start up and Continuous Operation. We inoculated the MBfR with 1 mL of diluted sludge obtained from our laboratory and enriched the community by circulating 10 mg/L ClO4 in a mineral salt medium (described below) for 24 h. In stage 1, we fed the MBfR continuously with a target influent ClO4 concentration of 1 mg/L until steady state was reached. In stage 2, we added 2.3 mg/L NO3N and operated the MBfR to steady state in terms of effluent concentrations of NO3 and ClO4. Subsequently, we systematically changed the influent concentrations of NO3 and ClO4 in stages 3 through 6 until each reached steady state: stage 3: 1 mg/L ClO4 and 11.3 mg N/L, stage 4: 10 mg/L ClO4 and 0 mg/L N; stage 5: 10 mg/L ClO4 and 11.3 mg N/ L; and stage 6: 10 mg/L ClO4 and 22.6 mg N/L. Actual concentrations varied slightly from the targets and are presented in Figure 1 in Results and Discussion. The medium pH was adjusted to 7.5 ( 0.2 with sodium hydroxide and contained the following mineral salts (analytical grade or purer) per L of demineralized water: CaCl2•2H2O 2 mg, NaSO4 2 mg, NaCl 0.2 g, MgCl2 0.1 g, NaHCO3 0.2 g, FeSO4•7H2O 1 mg, NaH2PO4 0.95 g, Na2HPO4 0.7 g, and 1 mL trace element solution (50 mg of MnCl2•4H2O, 100 mg of ZnSO4•7H2O, 300 mg of H3BO3, 200 mg of CoCl2•6H2O, 20 mg of CuCl2•2H2O, 30 mg of NiCl2•6H2O, 30 mg of Na2MoO4•2H2O, 30 mg of NaSeO3•5H2O, 50 mg of Na2WO4•2H2O per liter). Perchlorate and Nitrate Analyses. We took liquid samples from the MBfR with 5-mL syringes and filtered them immediately through a 0.2-μm membrane filter (LC+PVDF membrane, Pall Life Sciences Acrodisc Syringe Filters, USA). We assayed for NO3, NO2, ClO3, and ClO2 using ion chromatography (Dionex ICS 3000) with an AS18 column and AG18 precolumn, an eluent concentration of 22 mM KOH, and a 1 mL/min flow rate. We assayed ClO4 using ion chromatography (Dionex ICS 2000) with an AS 16 column and AG16 precolumn, an eluent concentration of 35 mM KOH, and a 1.5 mL/min flow rate. We measured the pH of all influent and effluent samples with a pH meter (837 pH lab, Metrohm, Switzerland); pH maintained stable for all stages at 7.88.0. Flux Calculations. We calculated the NO3, ClO4, and O2 removal fluxes (g/m2-d) according to eq 1 J ¼ ðSo -SÞQ =A

ð1Þ

in which So and S are the influent and effluent NO3, ClO4, or O2 concentration (g/L), Q is the influent flow rate to the MBfR system (L/d), and A is the membrane surface area (m2). The H2 flux was calculated from the removal fluxes and reaction stoichiometry shown in eqs 2-414 NO3 - þ 3:0H2 þ 0:23CO2 þ Hþ ¼ 0:48N2 þ 0:046C5 H7 O2 N þ 3:4H2 O

ð2Þ

ClO4 - þ 0:11Hþ þ 0:11NO3 - þ 0:53CO2 þ 5:48H2 ¼ Cl- þ 5:15H2 O þ 0:11C5 H7 O2 N ð3Þ

O2 þ 2:4H2 þ 0:028NO3 - þ 0:14CO2 þ 0:028Hþ ¼ 0:028C5 H7 O2 N þ 2:3H2 O

ð4Þ

We also compared the actual H2 flux against the maximum H2 flux that can be delivered through the polypropylene fiber at the applied H2 pressure of 17.5 psig (i.e., 0.21 e eq/m2-d24) to indicate if H2 delivery was limiting. Biofilm Sampling and DNA Extraction. We collected biofilm samples for all stages when the reactor reached a steady state in terms of ClO4 and NO3 concentrations. We cut off one ∼5cm-long section from the coupon fiber and then sealed the remaining fiber by tying the end into a knot. We then placed the section of fiber into a 15-mL Falcon centrifuge tube with 5.0 mL DNA-free water, vortexed the Falcon tube for 15 min at the highest speed (Eppendorf 5810R, Eppendorf, Germany) at 4 °C to detach the biofilm from the fiber, and then centrifuged the tube for 20 min at 8000 rpm (5900 g) at 4 °C (Eppendorf 5415R, Eppendorf, Germany). We carefully decanted the supernatant and removed the fiber to another 15-mL Falcon tube with 5 mL of DNA-free water. We again vortexed and centrifuged the Falcon tube to precipitate all biofilm. We then combined the pellets formed in both Falcon tubes into a 1.5-mL Eppendorf centrifuge tube for downstream DNA extraction using the DNeasy Blood & Tissue Kit (QIAGEN, USA) and followed the manufacturer’s recommendations. We quantified the DNA with a spectrophotometer (NanoDrop ND-1000, NanoDrop Technologies, USA) and documented its yield and purity based on the 260/280 nm absorbance ratio.25 PCR Conditions and Construction of Control Plasmids. We constructed plasmids containing target fragments and used them to generate standard curves based on serial dilutions containing between 108 and 102 target gene copies, calculated directly from the concentration of extracted plasmids. For plasmid construction, we used degenerate primers that target the perchlorate reductase gene (pcrA), nitrite reductase genes (nirS and nirK), and the 16S rRNA gene.12,2628 We present the primer names, primer sequences, and PCR conditions for each target gene in Table S1. Quantification of pcrA, nirK, nirS, and 16Sr RNA genes. The primers and qPCR conditions for pcrA, nirK, nirS, and the16S rRNA gene were previously described.12,29,30 We used the SYBR Premix Ex Taq Kit (Takara Bio Inc., Japan) and performed qPCR amplification in 20-μL reaction volumes following the protocols in Table S1. Each reaction tube contained 10 μL of SYBR Premix Ex Taq Mix, 8.6 μL of H2O, 0.2 μL of each forward and reverse primer (1 pmol/μL), and 1 μL of DNA template. The copy numbers of the biofilm samples were calculated by comparison to standard curves. Negative controls included water instead of template DNA in the PCR reaction mix. We performed triplicate PCR reactions for all samples and negative controls. The slopes of the plasmid standard curves and efficiency values for quantification by qPCR were the following: 3.404 and 0.97 for pcrA, 3.437 and 0.95 for nirK, 3.411 and 0.96 for nirS, and 3.352 and 0.99 for 16S rDNA, respectively. We calculated the Pearson Correlation between ClO4- and NO3 flux and gene copy numbers (Table S2 in the Supporting Information).31 We estimated cell ratios from the gene copy numbers using the following assumptions: one pcrA copy and two nirS copies per cell based on the genome of the perchlorate- and nitrate-reducing Dechloromonas aromatica RCB,32 respectively; one nirK copy per copper denitrifier cell;33 and an average of seven copies of 16S rRNA gene per bacterial cell.34 10157



ARTICLE

Table 1. Average Acceptor and Donor Fluxes for Each Stage in the MBfRb ClO4 flux stages (g/m2-d)

a

NO3-N

O2

electron donor (H2)

electron donor consumed

flux

electron donor consumed

flux

(e eq/m2-d)

(gN/m2-d)

(e eq/m2-d)

(g/m2-d)

(e eq/m2-d)

(e eq/m2-d)

(e eq/m2-d)a

electron donor consumed maximum H2 flux

actual H2 flux

1

0.05

0.004

NA

NA

0.51

0.038

0.21

0.042

2

0.05

0.004

0.10

0.021

0.51

0.038

0.21

0.063

3

0.04

0.003

0.60

0.146

0.51

0.038

0.21

0.187

4

0.53

0.030

NA

NA

0.51

0.038

0.21

0.068

5

0.31

0.016

0.72

0.156

0.51

0.038

0.21

0.210

6

0.12

0.007

0.84

0.189

0.51

0.038

0.21

0.234

: calculated from eqs 2-4. b ND = not detected. NA = not applicable.

Pyrosequencing. We used primers 939F and 1492R to target

the conserved V6 and V7 regions of 16S rRNA gene. Amplicon pyrosequencing was performed with standard 454/GS-FLX protocols by Research and Testing Laboratories LLC (Texas, USA).35 After a sequencing run and base calling, we trimmed the sequences using Mothur Software36 for downstream analysis. To assign 454 pyrosequencing tags to their closest relative in a reference database with 44,011 nonidentical V6V7 sequences extracted from 119,480 bacterial rRNA genes, we carried out three main data-processing steps. We first aligned all qualified sequences using the Silva alignment37 and excluded sequences shorter than 200 bp and chimeric sequences detected by modified ChimeraSlayer.38 We then clustered the sequencing readouts at 95% similarity with farthest algorithm to obtain the number of unique phylotypes and calculated rarefaction curves with the Mothur software. Finally, we classified sequences using the RDP Classifier software at the 50%-confidence threshold.39 For each biofilm sample, we generated rarefaction curves and calculated the Chao1 estimator to estimate the true total number of OTUs (Operational taxonomic unit) with infinite sampling.40

’ RESULTS AND DISCUSSION Interference Effects between Nitrate and Perchlorate. Figure 1-A shows the influent and effluent concentrations of NO3 and ClO4 for the entire experiment, while Figure 1-B shows the removal percentages of NO3 and ClO4. Table 1 summarizes the steady-state ClO4 and NO3 fluxes (calculated by eq 1), along with the H2 fluxes (from eqs 2-4). Based on the H2 fluxes, stages 1, 2, and 4 clearly had sufficient H2 delivery, but stages 5 and 6 were H2 limited, because the actual H2 flux was close to the theoretical maximum H2 flux of 0.21 e eq/m2-d. The removal percentage of ClO4 was 100% when the NO3 input was zero (stage 1) or low (stage 2), but it decreased to 69% when the NO3 loading increased (stage 3) and the total H2 flux was 89% of the maximum. The removal of NO3 also was reduced from 100% in stage 2 to 93% in stage 3. Thus, when the H2 delivery rate was not high enough for full reduction of both acceptors, the percentage removals of both declined, but ClO4 reduction had the larger proportional decline. Returning the NO3 loading to zero in stage 4 allowed perchlorate removal to return to 100%, even though the perchlorate loading was over 10 times greater than in stage 1; the H2 flux was only 32% of the maximum for stage 4. Stages 5 and 6 had much higher NO3 loadings (Table 1), which caused the H2 flux to be essentially equal to the maximum.

The impact of limited H2 delivery was much greater on ClO4 than on NO3, although both had substantially lowered percent removals in stage 6 (Figure 1). This trend can be seen by noting that the increases in NO3 fluxes for stages 5 and 6 were accompanied by roughly corresponding decreases in the ClO4 fluxes (Table 1). The trends shown in Table 1 suggest that competition for electron equivalents from H2 favored NO3 reduction over ClO4 reduction. This is consistent with the MBfR findings of Ziv-El and Rittmann,41 who reported that the electron acceptors had a clear H2-utilization priority when H2 was limiting: O2 > NO3 > NO2 > ClO4. Nerenberg et al.19 also reported that full NO3 removal was essential before ClO4 removal started in an MBfR. For heterotrophic systems, Choi and Silverstein18 suggested that competition for electrons by NO3 was a factor in the inhibition of ClO4 reduction, while Giblin et al.42 found that NO3 and ClO4 were simultaneously reduced to below detectable levels when the electron donor (acetate) was not limiting. Functional Gene Abundance. Figure 2 shows the pcrA, nirS, nirK, and 16S rRNA gene copy numbers in the different stages. The 16S rRNA gene increased continuously through all stages, indicating an increase of microbial biomass, which is consistent with increased H2 flux through all stages (except stage 4, when NO3 was removed when ClO4 was increased 10-fold). Although the loading of ClO4 was held nearly steady from stage 1 to stage 3, the copy number of pcrA gene (representing PRBs) gradually increased; presumably, the increase represents accumulation of more biomass over time. The increase in pcrA gene numbers from stage 3 to stage 4 was about a 13-fold, reaching 1.2 107 copies/cm2; this increase mirrored the ∼10fold increase of the ClO4 flux (Table 1). In last two stages, perchlorate reduction was inhibited by higher concentration of nitrate, but the perchlorate fluxes were still higher than in first 3 stages. Correspondingly, pcrA gene abundance decreased by around 30% but still was higher than for the first 3 stages. Overall, the pcrA gene tracked the rate of perchlorate removal, and the Pearson Correlation was 99.7%. The copy numbers of the nirK gene (copper-based NO2 reductase) did not change significantly, probably because nirK was not a good marker for denitrifiers in the MBfR biofilm. However, nirS (cytochrome-based NO2 reductase) increased steadily, approximately in proportion to the 16S rRNA gene. The abundance of nirS also correlated with the ClO4 flux, suggesting that the denitrifiers contained the nirS and pcrA genes, which enabled them to simultaneously reduce NO3 and ClO4. This 10158



ARTICLE

Figure 2. (A) pcrA, nirK, nirS, and 16S rRNA gene copy numbers per cm2 fiber. (B) Estimated cells densities associated with the flux of H2.

interpretation is further supported by the strong correlation of the nirS gene copy numbers with the perchlorate flux (97.6%) and with the pcrA gene copy numbers (98.7%). The coexistence of ClO4 and NO3 reduction genes has been reported for denitrifiers.20,21 However, the response of PRBs and denitrifiers to ClO4 and NO3 is species-specific. Sun et al.43 reported that Dechloromonas aromatica did not distinguish between perchlorate and nitrate when grown with perchlorate, but it reduced only nitrate when grown with nitrate. Chaudhuri et al.44 found that nitrate-grown cells of Dechlorosoma suillum exhibited an extended lag when transferred to medium containing equimolar amounts of perchlorate and nitrate, while perchlorate-grown D. suillum preferentially reduced nitrate and did not reduce perchlorate until nitrate was removed completely. In contrast, Dechloromonas agitata strain CKB reduced perchlorate equally well in the presence or absence of nitrate. While total bacteria continued to grow from stages 1 to 6, the ratios of PRBs and denitrifers to total bacteria decreased from stage 1 to stage 6. The reason probably is that the MBfR became mixotrophic as heterotrophic bacteria continually grew using soluble microbial products from the autotrophs,45 especially in stages 5 and 6, when heterotrophs gained a selective advantage against autotrophs as H2 became limiting. The Bacterial Community Structure of the MBfR Biofilm. Pyrosequencing yielded a total of 22,572 high-quality sequences for the 16S rRNA gene for all stages, and Figure S3 in the Supporting Information shows the rarefaction curves based on an OTU definition of 95% sequence similarity. Figure 3 shows the taxonomic breakdown at the class level for all stages. Overall,

the MBfR biofilms contained sequences from 23 bacterial classes. The majority of the sequences belonged to β-, γ-, and ε-proteobacteria, with the remainder spread among Holophagae, Sphingobacteria, and other minor groups. From stage 1 to stage 4, β- and ε-proteobacteria were the dominant classes, but, when H2 became limiting in stages 5 and 6, the percentage of ε-proteobacteria class dramatically decreased (from ∼40% in stage 4 to 3% in stage 5 and 0.4% in stage 6), while β-proteobacteria became the dominant class (from ∼45% in stage 4 to 86% in stage 5 and 91% in stage 6). Not surprisingly, most of the bacteria known to reduce perchlorate and nitrate belong to the β-proteobacteria class.1,4,5,9 Comparing with MBfR community structures at the genus level allows us to infer functions, based on the assumption that members that are closely related phylogenetically share similar metabolic capabilities. Figure 4 shows the relative proportions of the most abundant bacterial genera in the different stages. The genus Azonexus (β-proteobacteria) was strongly enriched in the last two stages, when the H2 delivery reached its maximum flux, ClO4 flux decreased, but the highest NO3 fluxes in the study were achieved. Previously described strains from the genus Azonexus use NO3 as an electron acceptor and contain examples of autotrophy (Azonexus caeni)46 and heterotrophy (Azonexus sp. LT-1T).47 Sulfuricurvum (ε-proteobacteria) represented around 40% of all sequences in the first four stages but became a minority member in the last two stages. Sulfuricurvum were reported in autotrophic denitrification conditions48,49 and could oxidize sulfide, elemental sulfur, thiosulfate, or hydrogen. The genus Hydrogenophaga (β-proteobacteria) was another dominant genus during all stages, and it increased slightly from 10159



ARTICLE

Figure 3. Class-level abundance of pyrosequences from the MBfR biofilms. The relative abundance of a class within a biofilm community is defined as the number of sequences affiliated with that class divided by the total number of sequences in the reactor’s biofilm.

Figure 4. Predominant bacterial genera by relative abundance (%) in the MBfR biofilms.

stage 4 to stage 6. Park et al.50 reported that an unculturable H2-oxidizing Hydrogenophaga was dominant in a denitrifying microbial community in an autotrophic biofilm reactor. Zhang et al.51 suggested that Hydrogenophaga may play an important role in autohydrogenotrophic denitrification in the MBfR. Hydrogenophaga also appeared in a heterotrophic fluidized bed reactor using ethanol as electron donor.52 With insufficient electron donor in stage 5 and stage 6, the dramatic decrease in Sulfuricurvum sequences stands in contrast to the large increase in Azonexus and Hydrogenophaga sequences. The abundance of the nirS gene did not decrease, probably because the decline in Sulfuricurvum was balanced by an increase in Azonexus or Hydrogenophoga. The MBfR appeared to become

more mixotrophic in stages 5 and 6, when heterotrophs gained a selective advantage as H2 became limiting. The dominant genus shift to Azonexus may have occurred because Azonexus prefers to reduce nitrate under heterotrophic conditions; Hydrogenophaga is known to be a facultative denitrifier, while Sulfuricurvum is only autotrophic.53 Dechloromonas (β-proteobacteria) is the main genus known to reduce NO3 and ClO4.9,32 Nerenberg et al.9 found that the abundance of Dechloromonas increased with increasing ClO4 loading in an MBfR also fed NO3- or O2-fed MBfR. We found that, although Dechloromonas remained stable in the first 4 stages, it decreased from stage 5 to stage 6, when the H2 supply was insufficient and the ClO4 flux declined; meanwhile Hydrogenophaga 10160


Environmental Science & Technology and Azonexus showed large increases. The decrease of pcrA gene in the last two stages and dominant genus shift away from Dechloromonas are consistent with each other. They also infer that Hydrogenophaga and Azonexus were not responsible for reducing ClO4 but reduced only NO3. Thus, Dechloromonas played the most important role in ClO4 reduction. In summary, NO3 inhibited ClO4 reduction when H2 became limiting. Then, the two reductions competed for electrons from H2, and ClO4 reduction declined more severely than did NO3 reduction. Although the nirK gene was not a good marker for denitrifiers in the MBfR biofilm, the nirS gene reflected well the increase in denitirification over time. PRB were well reflected by the pcrA gene. With sufficient H2 flux, the MBfR was dominated by ε- and β-proteobacteria. The main genera responsible for NO3 reduction were Sulfuricurvum and Hydrogenopohaga, and Dechloromonas was mainly responsible for ClO4 reduction. Azonexus displaced Sulfuricurvum as the dominant genus when H2 was limiting, making β-proteobacteria the dominant class. Azonexus and Hydrogenophaga appeared to gain their competitive advantage during severe H2 limitation by carrying out mixotrophic denitrification. Thus, facultative bacteria have clear advantages over strict autotrophs when the inorganic donor (H2) is in short supply.

’ ASSOCIATED CONTENT

bS

Supporting Information. Tables S1 and S2 and Figures S1S3. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 480727-0673. Fax: 480-727-0889. E-mail: hzhao21@ asu.edu.

’ ACKNOWLEDGMENT This research was supported by grant number ER-200541 from the Environmental Security Technology Certification Program (ESTCP), United States Department of Defense, via a subcontract with CDM. The authors thank Erica Hartmann and Jie Sheng for reviewing the manuscript. ’ REFERENCES (1) Logan, B. E. Assessing the outlook for perchlorate-remediation. Environ. Sci. Technol. 2001, 35, 482A–487A. (2) United States Environmental Protection Agency, Federal Facilities Restoration and Reuse, Perchlorate. 2008. Available from http:// www.epa.gov/swerffrr/documents/perchlorate/htm. (3) Gu, B.; Coates, J. D. Perchlorate Environmental Occurrence: Interactions and Treatment; Springer: Boston, Maryland, U.S.A., 2006. (4) Coates, J. D.; Achenbach, L. A. Microbial perchlorate reduction: rocket-fuelled metabolism. Nat. Rev. 2004, 2, 569–580. (5) Logan, B. E. A review of chlorate- and perchlorate-respiring microorganisms. Biorem. J. 1998, 2, 69–79. (6) Rittmann, B. E.; Nerenberg, R.; Lee, K. C.; Najm, I.; Gillogy, T. E.; Lehman, G. E.; Adham, S. S. Hydrogen-based hollow-fiber membrane biofilm reactor (MBfR) for removing oxidized contaminants. Water. Sci. Technol. 2004, 4, 127–133. (7) Chung, J.; Rittmann, B. E. Bio-reductive dechlorination of 1,1, 1-trichloroethane and chloroform using a hydrogen-based membrane biofilm reactor. Biotechnol. Bioeng. 2007, 97, 52–60.

ARTICLE

(8) Nerenberg, R.; Kawagosh,i, Y.; Rittmann, B. E. Kinetics of a hydrogen-oxidizing, perchlorate-reducing bacterium. Water. Res. 2006, 40, 3290–3296. (9) Nerenberg, R.; Kawagoshi, Y.; Rittmann, B. E. Microbial ecology of a perchlorate-reducing, hydrogen-based membrane biofilm reactor. Water. Res. 2008, 42, 1151–1159. (10) Kengen, S. W. M.; Rikken, G. B.; Hagen, W. R.; Van Ginkel, C. G.; Stams, A. J. M. Purification and characterization of perchlorate reductase from the chlorate-respiring strain GR-1. J. Bacteriol. 1999, 181, 6706–6711. (11) Van Ginkel., C. G.; Rikken, G. B.; Kroon, A. G. M.; Kengen, S. W. M. Purification and characterization of chlorite dismutase: a novel oxygen-generating enzyme. Arch. Microbiol. 1996, 166, 321–326. (12) Nozawa-Inoue, M.; Jien, M.; Hamilton, N. S.; Stewart, V.; Scow, K. M.; Hristova, K. R. Quantitative detection of perchloratereducing bacteria by real-time PCR targeting the perchlorate reductase gene. Appl. Environ. Microbiol. 2008, 74, 1941–1944. (13) Gu, B.; Ku, Y. K.; Brown, G. M. Treatment of perchlorate contaminated groundwater using highly selective, regenerable ionexchange technology: a pilot-scale demonstration. Fed. Facil. Environ. J. 2003, 14, 75–94. (14) Rittmann, B. E.; McCarty, P. L. Environmental Biotechnology: Principles and Applications; McGraw-Hill Book Co.: New York, 2001. (15) Richardson, D. J.; Berks, B. C.; Russell, D. A.; Spiro, S.; Taylor, C. J. Functional, biochemical and genetic diversity of prokaryotic nitrate reductase. Cell. Mol. Life Sci. 2001, 58, 165–178. (16) Richardson, D. J.; Watmough, N. J. Inorganic nitrogen metabolism in bacteria. Curr. Opin. Chem. Bio. 1999, 3, 207–219. (17) Herman, D. C.; Frankenberger, W. T. Bacterial reduction of perchlorate and nitrate in water. J. Environ. Qual. 1999, 28, 1018– 1024. (18) Choi, H.; Silverstein, J. Inhibition of perchlorate reduction by nitrate in a fixed biofilm reactor. J. Hazard. Mater. 2008, 159, 440–445. (19) Nerenberg, R.; Rittmann, B. E.; Najm, I. Perchlorate reduction in a hydrogen-based membrane-biofilm reactor. J. Am. Water Works Assoc. 2002, 94, 103–114. (20) Giblin, T.; Frankenberger, W. T. Perchlorate and nitrate reductase activity in the perchlorate-respiring bacterium perclace. Microbiol. Res. 2001, 156, 311–315. (21) Xu, J.; Trimble, J. J.; Steinber, L.; Logan, B. E. Chlorate and nitrate reduction pathways are separately induced in the perchloraterespiring bacterium Dechlorosoma sp. KJ and the chlorate-respiring bacterium Pseudomonas sp. PDA. Water Res. 2004, 38, 673–680. (22) Wolterink, A. F. W. M.; Jonker, A. B.; Kengen, S. W. M.; Stams, A. J. M. Pseudomonas chloritidismutans sp. Nov., a non-denitrifying, chlorate-reducing bacterium. Int. J. Syst. Evol. Microbiol. 2002, 52, 2183–2190. (23) Rikken, G. B.; Kroon, A. G. M.; Van Ginkel, C. G. Transformation of perchlorate into chloride by a newly isolated bacterium: reduction and dismutation. Appl. Microbiol. Biotechnol. 1996, 45, 420–426. (24) Tang, Y. N.; Zhou, C.; Van Ginkel, S.; Rittmann, B. E. Determining the hydrogen-diffusion coefficients of the fibers used in H2-based membrane biofilm reactors. Water. Res. 2011. (25) Tataurov, A. V.; You, Y.; Owczarzy, R. O. Predicting ultraviolet spectrum of single stranded and double strand deoxyribonucleic acids. Biophys. Chem. 2007, 133, 66–70. (26) Throbaeck, I. N.; Enwall, K.; Jarvis, A.; Hallin, S. Reassessing PCR primers targeting nirS, nirK and nosZ genes for community surveys of denitrifying bacteria with DGGE. FEMS Microbiol. Ecol. 2004, 49, 401–417. (27) Braker, G.; Fesefeldt, A.; Witzel, K. P. Development of PCR primer systems for amplification of nitrite reductase genes (nirK and nirS) to detect denitrifying bacteria in environmental samples. Appl. Environ. Microbiol. 1998, 64, 3769–3775. (28) Ritalahti, K. M.; Amos, B. K.; Sung, Y.; Wu, Q. Z.; Koenigsberg, S. S.; L€ offler, F. E. Quantitative PCR targeting 16S rRNA and reductive dehalogenase genes simultaneously monitors multiple Dehalococcoides strains. Appl. Environ. Microbiol. 2006, 72, 2765–2774. 10161


Environmental Science & Technology (29) Kandeler, E.; Deiglmayr, K.; Tscherko, D.; Bru, D.; Philippot, L. Abundance of narG, nirS, nirK, and nosZ genes of denitrifying bacteria during primary successions of a glacier foreland. Appl. Environ. Microbiol. 2006, 72, 5957–5962. (30) Dionisi, H. M.; Harms, G.; Layton, A. G.; Gregory, I. R.; Parker, J.; Hawkins, S. A.; Robinson, K. G.; Sayler, G. S. Power analysis for realtime PCR quantification of genes in activated sludge and analysis of the variability introduced by DNA extraction. Appl. Environ. Microbiol. 2003, 69, 6597–6604. (31) Tuomivirta, T. T.; Yrjaelae, K.; Fritze, H. Quantitative PCR of pmoA using a novel reverse primer correlates with potential methane oxidation in Finnish fen. Res. Microbiol. 2009, 160, 751–756. (32) Coates, J. D.; Chakraborty, R.; Lack, J. G.; O’Connor, S. M.; Cole, K. A.; Bender, K. S.; Achenbach, L. A. Anaerobic benzene oxidation coupled to nitrate reduction in pure culture by two strains of Dechloromonas. Nature 2001, 411, 1039–1043. (33) Philippot, L. Use of functional genes to quantify denitrifiers in the environment. Biochem. Soc. Trans. 2006, 34, 101–103. (34) Fogel, G. B.; Collins, C. R.; Li, J.; Brunk, C. F. Prokaryotic genome size and ssu rDNA copy number: estimation of microbial relative abundance from a mixed population. Microb. Ecol. 1999, 38, 93–113. (35) Sun., Y.; Wolcott, R. D.; Dowd, S. E. Tag-encoded FLX amplicon pyrosequencing for the elucidation of microbial and functional gene diversity in any environment. Methods Mol. Biol. 2011, 733, 129–141. (36) Schloss, P. D.; Westcott, S. L.; Ryabin, T.; Hall, J. R.; Hartmann, M.; Hollister, E. B.; Lesniewski, R. A.; Oakley, B. B.; Parks, D. H.; Robinson, C. J.; Sahl, J. W.; Stres, B.; Thallinger, G. G.; Van Horn, D. J.; Weber, C. F. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microb. 2009, 75 (23), 7537–7541. (37) Pruesse, E.; Quast, C.; Knittel, K.; Fuchs, B. M.; Ludwig, W.; Peplies, J.; Gl€ockner, F. O. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res. 2007, 35 (21), 7188–7196. (38) Haas, B. J.; Gevers, D.; Earl, A. M.; Feldgarden, M.; Ward, D. V.; Giannoukos, G.; Ciulla, D.; Tabbaa, D.; Highlander, S. K.; Sodergren, E.; Methe, B.; DeSantis, T. Z.; Human. Microbiome Consortium; Petrosino, J. F.; Knight, R.; Birren, B. W. Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons. Genome Res. 2011, 21 (3), 494–504. (39) Cole, J. R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R. J.; Kulam-Syed-Mohideen, A. S.; McGarrell, D. M.; Marsh, T.; Garrity, G. M.; Tiedje, J. M. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009, 37, 141–145. (40) Chao, A.; Lee, S. M. Estimating the number of classes via sample coverage. J. Am. Stat. Assoc. 1992, 87, 210–217. (41) Ziv-El, M. C.; Rittmann, B. E. Systematic evaluation of nitrate and perchlorate bioreduction kinetics in groundwater using a hydrogenbased membrane biofilm reactor. Water Res. 2009, 43, 173–181. (42) Giblin, T.; Herman, D.; Deshusses, M. A.; Frankenberger, W. T. Removal of perchlorate in ground water with a flow-through bioreactor. J. Environ. Qual. 2000a, 29, 578–583. (43) Sun, Y.; Gustavson, R. L.; Ali, N.; Weber, K. A.; Westphal, L. L.; Coates, J. D. Behavioral response of dissimilatory perchlorate-reducing bacteria to different electron acceptors. Appl. Microbiol. Biotechnol. 2009, 84, 955–963. (44) Chaudhuri, S. K.; O’Connor, S. M.; Gustavson, R. L.; Achenbach, L. A.; Coates, J. D. Environmental factors that control microbial perchlorate reduction. Appl. Environ. Microbiol. 2002, 68, 4425–4430. (45) Ergas, S. J.; Reuss, A. Hydrogenotrophic denitrification of drinking water using a hollow fiber membrane bioreactor. J. Water Supply: Res. Technol.--AQUA 2001, 50, 161–171. (46) Quan, Z. X.; Im, W. T.; Lee, S. T. Azonexus caeni sp. Nov., a denitrifying bacterium isolated from sludge of a wasterwater treatment plant. Int. J. Syst. Evol. Microbiol. 2006, 56, 1043–1046. (47) Thrash, J. C.; Pollock, J.; Torok, T.; Coates, J. D. Description of the novel perchlorate-reducing bacteria Dehalorobacter hydrogenphilus

ARTICLE

gen. nov., sp. Nov. and Propionivibrio militaris, sp. Nov. Appl. Microbiol. Biotechnol. 2010, 86, 335–343. (48) Sahu, A. K.; Conneely, T.; Nuesslein, K. R.; Ergas, S. J. Biological perchloreate reduction in packed bed reactors using elemental sulfur. Environ. Sci. Technol. 2009, 43, 4466–4471. (49) Kodama, Y.; Watanabe, K. Sulfuricurvum kujiense gen. nov., sp. Nov., a facultatively anaerobic, chemolithoautotrophic, sulfur-oxidizing bacterium isolated from an underground crude-oil storage cavity. Int. J. Syst. Evol. Microbiol. 2004, 54, 2297–2300. (50) Park, H. I.; Choi, Y. J.; Pak, D. Autohydrogenotrophic denitrifying microbial community in a glass beads biofilm reactor. Biotechnol. Lett. 2005, 27, 949–953. (51) Zhang, Y. H.; Zhong, F. H.; Xia, S. Q.; Wang, X. J.; Li, J. X. Autohydrogenotrophic denitrification of drinking water using a polyvinyl chloride hollow fiber membrane biofilm reactor. J. Hazard. Mater. 2009, 170, 203–209. (52) Hwang, C.; Wu, W. M.; Gentry, T. J.; Carley, J.; Carroll, S. L.; Schadt, C.; Watson, D.; Jardine, P. M.; Zhou, J.; Hickey, R. F.; Criddle, C. S.; Fields, M. W. Changes in bacteria community structure correlate with initial operating conditions of a field-scale denitrifying fluidized bed reactor. Appl. Microbiol. Biotechnol. 2006, 71, 748–760. (53) Campbell, B. J.; Engel, A. S.; Porter, M. L.; Takai, K. The versatile e-proteobacteria: key players in sulphidic habitats. Nat. Rev. 2006, 4, 458–468.

10162


Interactions between Perchlorate and Nitrate Reductions in the

Recommend Documents