Hydrogenase Chip Design Variations for Each Version

The Hydrogenase Chip: A Tiling Oligonucleotide DNA Microarray Technique for Characterizing Hydrogen Producing and Consuming Microbes in Microbial Communities

Supplemental Methods Section

Hydrogenase Chip Design Variations for Each Version

For Hydrogenase Chip version one, 199 genes involved in formate metabolism were retrieved from IMG/M by searching for annotations of genes encoding pyruvate formate lyase, formate hydrogen lyase, and formate dehydrogenase. These genes were added to the design after all retrieved hydrogenase genes were included and space for more probes remained.

For Hydrogenase Chip version two, 57 reductive dehalogenase (rdhA) genes were retrieved from the NCBI non-redundant nucleotide database to fill space after all hydrogenase genes were included.

For Hydrogenase Chip version 3, additional gene sequences were obtained from three sources: by retrieving all annotated hydrogenase and reductive dehalogenase gene sequences from all “Dehalococcoides” genomes in IMG version 2.9 (these were then not subjected to CD-HIT clustering), by retrieving hydrogenase sequences from published clone libraries (Boyd et al. 2009., Xin et al. 2008), and from the uptake [NiFe]-hydrogenase sequence from the genome of Microcoleus chthonoplastes PCC 7420. The reductive dehalogenase gene sequences were derived from all “Dehalococcoides” genomes in IMG version 2.9.

Operation of Reductive Dechlorinating Chemostat

The Point Mugu dehalogenating enriched microbial culture was maintained in a chemostat reactor following ten years of batch cultivation as described in Yu, Dolan, and Semprini (Yu et al., 2005). This batch culture was used to inoculate the first in a series of three identically-operated chemostats, the third of which (“PM-5L”) provided culture for this experiment after nearly 250 days of operation.

The PM-5L chemostat consists of a 5L (nominal) GL-45 Kimax reactor fitted with a three-hole Teflon cap (Kontes Glass Co., Vineland, NJ) that is compatible with PEEK tubing and fittings. It is an all-liquid reactor stirred using a 2” Teflon-coated magnetic stir bar. The influent feed solution was introduced to the chemostat via PEEK tubing, which was connected to a Hamilton 100-ml gas tight syringe driven by an Orion M361 syringe pump (Thermo Electron Corp., Beverly, MA). A dilution rate of 0.0186/day was maintained. The influent feed was a base of sterile basal anaerobic medium described by Yang and McCarty (1998), adjusted to double the buffering capacity (1 g/L K2HPO3 and 3 g/L Na2CO3) and increase the reductant, H2S, to 20 mg/L. The medium was amended with PCE (saturated, 1.15 mM), and sodium lactate (4.3 mM) as electron donor. Sulfate was not added to the influent feed. The chemostat was equipped to anaerobically transfer culture to batch reactors through PEEK tubing by pressurizing the chemostat with furnace-treated, anaerobic N2 gas. At the time of the batch experiment, a pseudo-steady state had been reached in the chemostat. Electrons from lactate were distributed as follows: 67% to acetate, 8% biomass, 2% VC, 16% ethene, and 7% unaccounted for, with a residual liquid H2 concentration of 3.25 nM.

Chemical Analytical Methods

Reactor headspace samples were used to monitor chlorinated aliphatic hydrocarbons (CAHs), ethene, and H2. CAHs and ethene were measured with an HP-6890 gas chromatograph (GC) equipped with a photoionization flame ionization detector and a 30m-0.53mmGS-Q column (J&W Scientific, Folsom, CA), with helium as the carrier gas (15 mL/min). The headspace samples were injected to the GC using a100μL gastight syringe (Hamilton, Leno, NV). The GC oven was initially set at 150°C for 2 min, heated at 45°C/min to 220°C, and held at 220°C for 1.44 min. Hydrogen concentrations in headspace gas samples (100 μL) were determined using an HP-5890 GC series II with a thermal conductivity detector (TCD), operated isothermally at 220°C. Gas samples were chromatographically separated with a Carboxen 1000 column (15 ft-1/8 in, Supelco, Bellefonte, PA) using argon as the carrier gas at 15 mL/min. Liquid samples (0.25 mL) were taken and diluted ten times with deionized water to measure sulfate and acetate concentrations. Sulfate in the aqueous phase was monitored with a Dionex DX-500 ion chromatograph (Sunnyvale, CA) equipped with an electrical conductivity detector and a Dionex AS14 column. Acetate was measured with a Dionex-500 HPLC chromatograph equipped with UV/VIS detector and an Alltech Prevail Organic acid column. Sulfide measurements were obtained with 0.3mL samples using the methylene blue colometric HACH method 8131.

Batch Cultures

Batch experiments were carried out in 125mL Borosilicate glass bottles fitted with phenolic screw-on caps with gray chlorobutyl rubber septa (Wheaton Industries, Millville, NJ). 50mL of culture was anaerobically transferred from the PM-5L chemostat into six reactors via PEEK tubing. Each bottle was purged with a furnace treated 75:25 Ar/CO2 gas mixture (Airco, Inc. (Albany, OR)) for 15 minutes to remove any residual chloroethenes, ethene, or H2 following the transfer.

To enrich different populations in the culture (dehalogenating, sulfate-reducing, and combination of sulfate-reducing and dehalogenating microbes) three sets of duplicates were prepared: “P” bottles were amended with 15 μmol neat PCE (99.9%, spectrophotometric grade from Acros Organics (Pittsburgh, PA)), “S” bottles with 16 μmol sulfate (from a 228 mM Na2SO4 solution in media), and dual- electron acceptor “SP” bottles with 16 and 15 μmol sulfate and PCE, respectively. Sulfate and PCE were added to achieve equivalent electron acceptor level assuming sulfate was reduced to sulfide and PCE was reduced to ethene. Hydrogen (82 μmols, 99%, Airco, Inc. (Albany, OR)) was injected to the headspace, creating an initial H2 liquid concentration around 15,000 nM.

Reactors were incubated at 20°C with continuous shaking at 200 rpm. PCE and transformation products, H2, sulfate, and acetate concentrations were monitored over time using gas chromatography (GC) with a flame ionization detector, GC with a thermo conductivity detector, ion chromatography, and high-performance liquid chromatography, respectively. Hydrogen was added whenever aqueous concentrations approached 1000nM, and bottles were amended to their initial PCE and sulfate concentration after PCE and sulfate had been completely reduced to VC, ethene, and sulfide, respectively. After 44 days (and eight to ten amendments), all bottles were amended with electron acceptors to the same electron accepting capacity as in the SP bottles, i.e. 15μmol of both PCE and sulfate were added to all bottles. Bottles were sacrificed for molecular analysis 1.2 days into this dual substrate experiment. Microcosm cultures were transferred to 50 mL polypropylene centrifuge tubes, centrifuged for 30 minutes at 9000 rpm and 4°C. Supernatant was decanted, and the pellets were stored at -80°C until analysis. An initial sample for RNA an DNA extraction (denoted “C”) was also collected from the chemostat in this same manner at the time of harvesting culture for the batch experiments.

Reductive Dechlorinating Liquid Culture DNA/RNA Co-Isolation

Liquid batch and chemostat cultures were transferred to 50mL Falcon tubes, then centrifuged for 30 minutes at 9000 rpm and 4°C. Supernatant was decanted, and the pellets were stored at -80°C until analysis. Frozen cell pellets were resuspended in 1mL of the lysis solution described below for microbial mat nucleic acid extraction, and RNA/DNA extraction also carried out according to the same protocol.

Primer Design and PCR

All primers were designed using Geneious (Biomatters, Auckland, New Zealand) based on manual identification of conserved regions of the sequence alignment, and design of primers producing the desired sequence length and annealing temperature.

A 400bp fragment of dsrA was PCR-amplified using forward primer Dsr-1F-GC (5’-ACSCACTGGAAGCACG-3’) and reverse primer Dsr-DGGE-Rev (5’-CGGTGMAGYTCRTCCTG-3’) as described by Leloup et al. (2009). PCR was performed in 50µL reactions containing 25µL 2X DreamTaq Green PCR Master Mix (Fermentas), 200nM of each primer, and 1µL of DNA in solution extracted from sample S.

Primers DMR-15600_F-717 (5’-MAARAACCCSCAYMCCCAG-3’) and DMR-15600_R-1720 (5’-GACRTGYACRSMRCAG-3’) targeting the hynA-1 gene in Desulfovibrio sp. were designed based on hynA-1 sequences from Desulfovibrio sp. related to Desulfovibrio magneticus (IMG identifiers 637123154, 637783027, 639819476, 643139751, 643538766, 643581839, 644801811, 644840066, and 645564504). These primers were used at a concentration of 500nM each to PCR-amplify hynA-1 from DNA sample S in 50µL reactions with 25µL 2X DreamTaq Green Master Mix, cycled with an initial 95ºC denaturation step for 3min, followed by 45 cycles with 30s at 95ºC, 30s at 48ºC, 30s at 72ºC, then a final 72ºC extension step of 10min. Gel extraction of the fragment with the expected amplicon size of approximately 1000bp was performed using the Wizard SV Gel and PCR Clean-up System according to manufacturers instructions (Promega).

PrimersHupL_F (5’-ATGCAGAAGATAGTAATTGAYC-3’) and HupL_R (5’-GCCAATCTTRAGTTCCATMR-3’) for amplification of a 1099bp fragment of the hupL gene from Dehalococcoides sp. used in the plasmid standard and primers HupL_Fq_56 (5’-AAGCCACCGTAGACGGCG-3’) and HupL_Rq_190 (5’-AGTGCCGTGRGAGGTGGG-3’) for a 134bp fragment for qPCR were designed based on sequence alignments of hupL from Dehalococcoides sp. genomes of strains 195, BAV-1, CBDB1, and VS (IMG gene object identifiers 637119679, 646445988, 637702682, and 640529159). HupL_F and HupL_R were used at a concentration of 500nM each to PCR-amplify hupL from DNA sample S in 50µL reactions with 25µL 2X DreamTaq Green Master Mix, cycled with an initial 95ºC denaturation step for 3min, followed by 45 cycles with 30s at 95ºC, 30s at 55ºC, 30s at 72ºC, then a final 72ºC extension step of 10min.

Cloning

The hynA-1 and dsrA PCR products were cloned using the TOPO TA cloning kit (Invitrogen), then Sanger-sequenced using the M13F primer (Elim Biopharm, Hayward, CA, USA). Using Geneious software, vector and primer sequence was trimmed from the sequences. Similar existing sequences were retrieved using the functional gene pipeline version 6.1 ( and NCBI BLAST (Altschul et al., 1990), then a muscle alignment (Edgar et al., 2004) and PHYML tree bootstrapped 100X (Guindon et al., 2003) were generated with the sequences and their closest relatives (Biomatters, Auckland, New Zealand). Of ten clones generated for dsrA, three representative dsrA sequences were submitted to GenBank (accession numbers HQ399561- HQ399563). For hynA-1, two clones were sequenced and submitted to GenBank with accession numbers HQ399559 and HQ399560.

Reverse Transcription - Quantitative PCR of Dehalococcoides sp. hupL

A 1/10 dilution series was used for quantification, with eight dilutions from a 1/10 dilution of the purified plasmid containing Dehalococcoides sp. hupL to a 1/108 dilution.

cDNA was synthesized using the Superscript III First-Strand Reverse Transcription Kit (Invitrogen) with random hexamers according to manufacturer’s instructions, albeit with a 3-hour 50ºC incubation. For each sample, in order to confirm that DNase treatment of the RNA was complete, a negative control cDNA synthesis reaction with no reverse transcriptase was performed.

Triplicate reactions were performed for each sample in 25µL reactions containing 12.5µL iQ SYBR Green Supermix (Biorad), 500nM of each primer, and 5µL of a 1/10 dilution of cDNA or standard plasmid dilution. Thermal cycling and fluorometry was performed using an iCycler iQ Real-Time PCR Detection System (Biorad), with a 3-minute intial denaturation step at 95ºC, followed by 40 cycles of 10 seconds of 95ºC denaturation and 45 seconds of annealing/extension at 61.5ºC.

Microbial Mat DNA/RNA Co-Isolation

The top 2mm layer of the mat cores collected during that study were placed in a 1mL a lysis solution consisting of 10mM EDTA, 50mM Tris-HCl, 4M guanidine thiocyanate, 2% sodium dodecyl sulfate, and 130mM -mercaptoethanol in a 2mL tube containing 0.1g 150-212µm acid-washed glass beads (Sigma-Aldrich, St. Louis, MO, USA). Tubes were vortexed at 4ºC at maximum speed for five minutes, then 1mL of pH 4.5 acid-phenol:chloroform:isoamyl alcohol in the ratio 125:24:1 (Ambion, Austin, TX, USA) was added. The solution was briefly vortexed, incubated at room temperature for five minutes, then centrifuged at 16,000 g (Eppendorf, Hamburg, Germany) for five minutes. The aqueous phase was removed and mixed with 40µL of RNase-free 3M sodium acetate (Ambion) and 1.7mL of -20ºC 100% ethanol. The mixture was incubated at -20ºC for one hour, centrifuged at 16,000g, 4ºC for 30 minutes. The liquid phase was decanted leaving a nucleic acid pellet that was air-dried for ten minutes then resuspended in 200µL of nuclease-free water. 100µL of this solution was stored at -20ºC for DNA analysis. 10µL TURBO DNase buffer and 2µL TURBO DNase (Ambion) were added to the remaining 100µL, then incubated at 37ºC for 30 minutes. An additional 2µL of DNase was added then incubated for a further 30 minutes. 120µL of the acid-phenol:chloroform:IAA solution were added, then the mixture was vortexed briefly, incubated at room temperature for one minutes, then centrifuged at 16,000g for two minutes. The aqueous phase was removed and mixed with 10µL 3M sodium acetate solution and 300µL ethanol, incubated at -20ºC for 30 minutes, then centrifuged at 4ºC, 16,000g for 30 minutes. The liquid was decanted, RNA solution resuspended in 50µL nuclease-free water. RNA and DNA were quantified in the solution using the Qubit fluorometer and broad-range double-stranded DNA and broad-range RNA Quant-it quantification kits (Invitrogen).

DNA Labeling and Hybridization

DNA was mixed with random hexamers (Invitrogen) at a concentration of 250ng/µL in 39µL of water, then incubated at 95ºC for 10 minutes. The mixture was placed on ice for 30 seconds, then 5µL of NEBuffer 2 (New England Biolabs (NEB), Ipswich, MA, USA), 2µL of a dNTP labeling mix (5mM dATP, cCTP, dGTP, dTTP (Invitrogen), 1.67mM amino-allyl labeled dUTP (Fermentas, Vilnius, Lithuania), and 4µL of Klenow Fragment (3´→5´ exo–) at 5000 units/mL (NEB). This Klenow reaction mixture was incubated for 16 hours at 37ºC, then stopped by the addition of 5µL 0.5M EDTA. The amino-allyl labeled DNA (aa-DNA) was purified using the QIAquick PCR Purification Kit (Qiagen) with a custom-made phosphate wash buffer in place of the Qiagen-supplied buffer PE (50mM KPO4, 80% ethanol, pH 8.5) then dried down at 45ºC in a SpeedVac (Thermo Scientific, Waltham, MA, USA). A Cy3 mono-reactive dye pack (GE Healthcare Biosciences, Piscataway, NJ, USA) was resuspended in 13µL Dimethyl-sulfoxide (DMSO). aa-DNA was resuspended in 10µL nuclease-free water, 0.5µL 1M sodium bicarbonate solution and 3µL of dissolved Cy3 dye, then incubated in the dark at room temperature for 60 minutes. Cy3-labeled DNA was again purified using the QIAquick PCR Purification Kit. Labeled DNA was quantified and Cy3 incorporation determined using a Nanodrop (Thermo Scientific). Labeled DNA was then hybridized to the test design DNA microarray at 65ºC for 17 hours according to the manufacturer’s protocol for one-color gene expression analysis (Agilent Technologies, Santa Clara, CA, USA). The hybridized microarray was scanned using a Genepix 4000B Microarray Scanner (Molecular Devices, Sunnyvale, CA, USA) and median probe intensity value used for further analysis.

RNA Labeling and Hybridization

The RNA amplification and labeling protocol was based on the Whole-Community RNA Amplification protocol (Gao et al., 2007). cDNA synthesis was performed with the Superscript III First-Strand synthesis kit (Invitrogen) following the manufacturer’s instructions, with the following modifications: 7µL of dissolved RNA (containing at least 500ng RNA) was used as starting material, and the primer added was 2µL 0.5µg/µL T7N6S primer (5’- AATTGTAATACGACTCACTATAGGGNNNNNN-3’), and the reverse transcription was performed overnight (18 hours) at 50ºC. The second strand cDNA was synthesized by adding 1µL of Klenow Fragment (New England Biolabs), 1µL 50ng/µL random hexamer solution (Invitrogen) to the cDNA first strand synthesis solution then incubated for two hours at 37ºC. The resulting cDNA was purified using the QIAquick PCR Purification Kit, quantified using the Qubit fluorometer and Quant-it broad-range double-stranded DNA kit, then dried down in a SpeedVac. This cDNA was used as a template for amino-allyl labeled RNA using the MEGAscript T7 Kit (Ambion) according to the manufacturer’s instructions, with the following modification: the 2µL UTP solution was replaced with 3µL of a 50mM 3:1 amino-allyl UTP : UTP solution mixture (Ambion). The resulting amino-allyl labeled cRNA was purified using the RNeasy Mini Kit (Qiagen) and eluted in 30µL nuclease-free water. A Cy3 mono-reactive dye pack (GE Healthcare) was resuspended in 13µL DMSO. 1.5µL 1M sodium bicarbonate and 3µL reactive Cy3 in DMSO were mixed with the amino-allyl-labeled cRNA then incubated in the dark at room temperature for 60 minutes. The resulting Cy3-labeled cRNA was purified using the RNeasy Mini Kit. Cy3 dye incorporation and cRNA quantity were determined using a Nanodrop. Labeled cRNA fragmentation, hybridization at 65ºC for 17 hours, and washing were performed according to the DNA microarray manufacturer’s instructions (Agilent Technologies, Santa Clara, CA). The hybridized microarray was scanned using a Genepix 4000B Microarray Scanner (Molecular Devices, Sunnyvale, CA, USA).

DNA Microarray Data Analysis Continued

For the reductive dechlorinating soil columns, both lactate and propionate amended time points were normalized to the formate sample in order to facilitate a meaningful three-way comparison. During analysis of the microbial mat samples, both the DNA and RNA probe intensities from the 20:00 time point were normalized to their respective 12:00 probe intensities. All reductively dechlorinating batch cultures were normalized to the chemostat sample.

Quantitative shifts in gene or transcript abundance between samples were determined by calculating ln(A/B) for each probe targeting a given gene, where A is the probe intensity from one sample and B from another, then finding the median of these numbers. This median value is referred to as the log intensity ratio. A log intensity ratio > 0 signifies a greater abundance in sample A, and a result below 0 signifies greater abundance in sample B. Since the set of ln(A/B) for a given gene generally fails tests of normality, statistics to determine significant changes in gene or transcript abundance based on an assumption that a distribution is normal could not be applied. Instead, the median absolute deviation was used as an estimate of variability in the measurement, and the binomial test (R function binom.test()) was applied to test the null hypothesis that the genes are in equal abundance (50% of probes with ln(A/B) > 0). Genes were considered of significantly different abundances in the two samples if the binomial test resulted in a null hypothesis probability of less than 0.01.