Microbial Diversity 2011

Characterization of methanogenic communities and nickel requirements for methane production from Wood Hole marshes and isolation of a novel of the order from Eel Pond mud

Jennifer Glass California Institute of Technology [email protected]

July 28, 2011 Microbial Diversity Course Marine Biological Laboratory

1 Highlights of Mini-Project • Isolation of anaerobic Citrobacter sp. (F1_G2) from School Street Marsh that seems to be producing acetate (further testing required). Glycerol stock made. • Isolation of methanogen with 85% similarity to petrolearius from Eel Pond mud (S1_G3). Glycerol stock made. • Construction of mcrA clone libraries and community analysis for Cedar Swamp (165 clones) and School Street Marsh (62 clones). • 454 pyrosequencing and community analysis of sediment samples from Cedar Swamp (JG4: 6166 total OTUs and 224 methanogen OTUs), School Street Marsh (JG2: 5279 total OTUs and1 methanogen OTU), Great Sippewissett (JG3: 4131 total OTUs and 3 methanogen OTUs) and marine mud from the south coast of Martha’s Vineyard (JG1: 4192 total OTUs and 0 methanogen OTUs). • CARD-FISH analyses of complex populations of Methanosarcinales and other and bacteria. Acquisition of two new probes (MSMX860 and MS1414) for the course. • Attempt at nickel limitation experiment was not useful for testing hypothesis likely due to nickel carryover and contamination.

Introduction Methane is a very potent greenhouse gas and the only organisms known to produce it in significant quantities are methanogenic archaea. The marshes and salt ponds around Woods Hole offer excellent environments for sampling and enrichment of , as exemplified by the brilliant pyrotechnic displays every summer at Cedar Swamp during the Volta experiment. Although methanogens play a vital role in carbon cycling and require high concentrations of micronutrients such as iron, molybdenum, cobalt and nickel for growth (Schönheit et al., 1979; Sowers and Ferry, 1985), little is known about the environmental availability of such trace metals in the ecosystems when methanogens live or how methanogens acquire such metals. Furthermore, little is known about how differences in the primary substrate that methanogens use for growth (H2 CO2, acetate or methyl-compounds) affects their micronutrient/trace metal requirements. Nickel is a particularly important metal for methanogens compared to other microbes because methanogenesis involves four Ni enzymes: [NiFe] hydrogenase, Ni-dependent

2 carbon monoxide dehydrogenase, acetyl SCoA synthase and most importantly methyl-SCoM (Mcr) which catalyzes the last step of methanogenesis. During this project, I developed a hypothesis that methanogens might have higher nickel requirements when grown on H2 CO2 than when grown on acetate or methyl-compounds because additional Ni-Fe hydrogenases are required for hydrogenotrophic growth via aceticlastic or methylotrophic methanogenesis (Thauer et al., 2008). The goals of this mini-project were as follows: (1) isolate and characterize a methanogen using the group enrichments started at the beginning of the course; (2) test the nickel hypothesis by measuring methane production of pure cultures of methanogens grown on H2 CO2 vs. methylated compounds; (3) use CARD-FISH to analyze the spatial structure of methanogens in mixed culture and (4) perform community analysis of methanogenic diversity in a range of natural environments around the Woods Hole/Falmouth area.

Methods and Materials Sample sites Samples were collected from the following sites around Woods Hole, Massachusetts: Great and Little Sippewissett Marsh, Eel Pond, Cedar Swamp, School Street Marsh and marine sediment (kindly provided by Erik Zettler) from the south coast of Martha’s Vineyard (Fig. 1). The first round of sampling occurred on June 15, 2011 at School Street Marsh and Little Sippewissett Marsh as part of the class field trip. Each of the four student groups collected sediment from the two sampling localities. In addition, group 3 collected a sample on June 13, 2011 from a patch of black, smelly, fine mud from Eel Pond directly across from the Loeb building near the parking lot and boats. Anaerobic chemotrophic group enrichments from these sites later became classified as the “freshwater” (School Street Marsh) and “seawater” (Little Sippewissett for groups 1, 2 and 4; Eel Pond for group 3) cultures (see below). The second round of sampling occurred on July 3-4, 2011 at Cedar Swamp, School Street Marsh and Great Sippewissett Marsh. Samples of sediment collected at Cedar Swamp, School Street Marsh and Great Sippewissett were frozen at -80°C immediately upon return to the lab. Along with a sample of marine sediment (C235AA-002-SG) collected on June 20, 2011 by Erik Zettler from 41°14’94009N/70°53’93055W, these samples were used for constructing mcrA clone libraries and for 454 pyrosequencing. At Great Sippewissett, anaerobic enrichment cultures

3 were set up within several hours after sampling (see below) whereas at School Street Marsh and

Cedar Swamp, enrichment samples were taken with an N2-flushed syringe in the field and immediately inoculated into sterile anaerobic freshwater media in 160-mL serum bottles. The marine sediment enrichment culture was inoculated ~2 weeks after collection.

Figure 1. Map of sampling localities in the Woods Hole and Falmouth area, Cape Cod, Massachusetts

Anaerobic chemotrophic media Anaerobic basal modular medium was used throughout this project. This media was made by mixing 1 mL 0.1% resazurin with 1L of 1x seawater (342.2 mM NaCl; 14.8 mM

MgCl2Ÿ2H2O; 1.0 mM CaCl2Ÿ2H2O; 6.71 mM KCl) or freshwater (17.1 mM NaCl; 1.97 mM

MgCl2Ÿ2H2O; 0.68 mM CaCl2Ÿ2H2O; 6.71 mM KCl) basal medium in 1-L Pyrex bottle. The following additions were made: 20 mL 1 M MOPS (pH 7.2) to maintain the pH, 1 mL 0.1M potassium phosphate (pH 7.2), 1 mL 1000x trace element solution (made without Ni for the nickel experiments) and 0.7 g of sodium acetate (for a final concentration of 5 mM acetate). This media was purged for 1 hour under a stream of anaerobic 80% N2: 20%CO2 gas and then brought

4 into the anaerobic chamber. The following anaerobic stock solutions were then added: 70 mL 1

M NaHCO3 to adjust the pH to 7.0 under 20% CO2, 10 mL of 0.5 M NH4Cl, 2.0 mL of 0.2 M cysteine-HCl and 2.0 mL of Na2S to reduce the medium and serve as a source of sulfur. Once the medium turned from pink to clear in color, indicating complete reduction, 30 mL of media was distributed into 160-mL serum bottles. Bottles intended for nickel experiments were acid-washed in 20% hydrochloric acid and then rinsed 3x in MilliQ water. An anaerobic stock solution of nickel (30 mM NiCl2Ÿ6H2O) was added to final concentrations of 0 (no added Ni), 10, 100, 500 and 1000 nM Ni. For experiments with trimethylamine and methanol, anaerobic stock solutions were added to serum bottles a final concentration of 10 and 62 mM, respectively. Bottles were stoppered with blue butyl stoppers (HCl-washed for nickel experiments) and crimped inside the anaerobic chamber. Bottles were then autoclaved and allowed to cool before inoculation. Plates were made using the same media recipe described above, but with Gelrite added to the seawater media and agar added to the freshwater media.

Inoculation for anaerobic chemotrophic enrichments For group enrichments, bottles containing 30-mL of anaerobic basal modular media containing either no added or 62 mM methanol were decrimped, purged with N2/CO2 gas (80%:20%) for 10 minutes and inoculated with 5-mL of a prepared soil slurry (2-3 g of soil or sediment into 15 mL of media). Bottles without added methanol were then overpressured with

H2/CO2 (80%:20%) gas to 7 psi (in later transfers, higher overpressures were used, up to 30 psi). The inventory of inoculated cultures included 4 freshwater (from School Street Marsh) cultures grown on H2 CO2, 4 freshwater cultures grown on methanol, 4 seawater (from Little

Sippewissett) cultures grown on H2 CO2 and 4 seawater cultures grown on methanol, totaling to 16 bottles. The annotation used for labeling cultures in this mini-project was “F” or “S” for freshwater or seawater, “1” for H2 CO2 growth, “4” for methanol growth and “G1, G2, G3 or G4” for group 1, 2, 3, or 4.

Enrichment and isolation of methanogens All sixteen group cultures were transferred after 6 days of growth by withdrawing 1-mL of culture with an N2-purged syringe and injecting it into 30 mL of fresh sterile anaerobic basal modular medium. Cultures were monitored for growth by visually checking for turbidity, by

5 observing F420 autofluorescence under the microscope and by measurement of methane production by gas chromatography. The culture was grown for another 2 days and plated onto agar plates for freshwater samples and Gelrite plates for seawater samples, otherwise of the same composition as the liquid media. Plates were grown inside the anaerobic chamber in H2S incubators with the same headspace previously used for incubation. Small colonies appeared after several days of growth in the H2S incubators. These colonies were checked for F420 autofluorescence and colony PCR with universal and archeal primers (see below). I restreaked single colonies from all plates onto new plates after approximately 1 week of growth. After another week, single colonies were picked and transferred into sterile anaerobic basal modular media.

Figure 2. Protocol for isolation of methanogens used during the group enrichments (first three steps) and the steps conducted in this mini-project (last three steps). Numbers indicate the number of different cultures at each step.

DNA extraction DNA was extracted from four sediment samples (Great Sippewissett, Cedar Swamp, School Street Marsh and marine sediment) and from pure liquid cultures using the Mo-Bio

6 PowerSoil® DNA Isolation Kit following manufacturer’s directions. DNA concentration and quality was checked with a NanoDrop. DNA from all four natural samples was submitted for 454 pyrosequencing (see below). For colony PCR, a single colony grown on a plate was picked with a sterile toothpick and dipped into the PCR reaction mixture (see below) and was then boiled at 95°C for 5 minutes immediately prior to PCR.

Polymerase chain reaction Polymerase chain reaction (PCR) was performed with Promega 2x PCR MasterMix using three primer sets. For universal 16s amplification, the primers 515F (5’- GTGYCAGCMGCCGCGGTAA-3') and 1391R (5'-GACGGGCGGTGTGTRCA-3') were used. For bacterial 16s amplification, the primers 8F (5'-AGAGTTTGATCCTGGCTCAG-3') and 1492R (5'-GGTTACCTTGTTACGACTT-3') were used. For archeal 16s amplification, the primers 2F (5'-TTCCGGTTGATCCYGCCGGA-3') and 1492R were used. For mcrA amplification, primers ordered in 2009 were used: LFS-mcrA-F (slightly modified at the 3’ end from the literature sequence: 5’-GGTGGTGTMGGATTCACACARTAY-3’) and LFS-mcrA-R (5’-TTCATTGCRTAGTTWGGRTAGTT-3’) were used (Luton et al., 2002). For 16s amplification, an annealing temperature of 46°C was used. For mcrA amplification, the annealing temperature was 50°C for School Street Marsh and Great Sippewissett samples and 55°C for the Cedar Swamp sample. The rest of the PCR program was based on the recommended protocol of Luton et al. (2002). PCR products from colony PCR and pure cultures were purified using the Qiagen MinElute DNA purification kit and were then diluted to 20 ng µL-1 and submitted for sequencing. Samples for clone library construction were stored at -20°C until further processing (see below).

McrA clone library construction For the three sediment samples (Great Sippewissett, Cedar Swamp and School Street Marsh) that gave positive mcrA amplification (single band at 400-500 bp), DNA bands were cut out from the agarose gel and purified using the Millipore Montage DNA gel extraction kit. If DNA was stored at 4°C for several hours after purification, 10 µL of DNA was combined with 10 µL of 2x Promega 2x PCR MasterMix at 72°C for 15 minutes to readenylate the ends of the DNA before cloning. Cloning was performed by combining 1 µL of pCR4-TOPO vector with 1

7 µL of salt solution and 4 µL of amplified and adenylated DNA and incubating at room temperature for 30 minutes. Following incubation, 2 µL of the ligation mixture was added to ~70 µL of electrocompetent TOPO cloning cells and transformation was performed by electroporation (BioRad Gene Pulser Xcell Electroporation System). SOC media (200 µL) was added and the cells were incubated for 1 hour at 37°C with shaking. A volume of 20-100 µL of cells were then plated with sterile beads onto LB + ampicillin plates. Plates were incubated at 37°C overnight and kept at 4°C until colonies were picked into 96-deep well plates containing 1.25 mL of sterile LB + ampicillin media per well and submitted for sequencing on the capillary sequencer (Applied Biosystems model 3730) at the Josephine Bay Paul Center at the Marine Biological Laboratory in Woods Hole, MA.

454 Pyrosequencing (protocol written and performed by Chuck Pepe-Ranney) SSU rRNA genes were amplified with barcoded primers that also incorporated the Roche 454 Ti adapter sequences (see below). The barcode was on the forward primer. Each barcode was 9 nt long. The barcode was TTAATTCGG for the sample of marine sediment (JG1); TTATTACCG for the sample of School Street Marsh mud (JG2); TCGGAACCG for the sample of Great Sippewissett mud (JG3) and TCCTACCGG for the sample of Cedar Swamp sediment (JG4). The primer targets were 515F and 907R on the E. coli 16S gene. The forward primer (X denotes barcode, lowercase is the linker between barcode and 16S primer) was: 5'CGTATCGCCTCCCTCGCGCCATCAGXXXXXXXXgaGTGYCAGCMGCCGCGGTAA-3' The reverse primer (lowercase denotes linker between adapter and 16S primer) was: 5'-CTATGCGCCTTGCCAGCCCGCTCAGggCCGYCAATTCMTTTRAGTTT-3' Samples of marine sediment (JG1, plate 1), School Street Marsh (JG2, plate 1) and Great Sippewissett (JG3, plate 1) were amplified for 32 cycles. The Cedar Swamp (JG4, plate 4) was amplified for 36 cycles. The PCR program utilized a touch-down annealing temperature for the first 10 cycles from 58-68°C, followed by 12 cycles of three-step PCR (denaturation, annealing, elongation) and then by 10-14 cycles of two-step PCR (annealing and elongation at the same temperature). Phusion HF polymerase (2X MasterMix) was used to amplify the gene with 8% DMSO and 0.5 µM primers in the final reaction volume. The template was normalized to 15 ng/µL (for DNA above 15 ng/uL) and 2 µL of each template was used for PCR. Post amplification DNA was quantified using the PicoGreen assay and then pooled ~125 ng of each

8 PCR product. That pool was concentrated down to 100 µL using the vacufuge and gel purifed using the Montage Kit (Millipore). The gel-purified pool was then shipped to Penn State University for sequencing.

Nickel limitation experiments Nickel limitation experiments were performed on a pure culture grown up from a single colony of F1_G2 classified as Citrobacter sp. and on an enrichment culture from the bottom layer of a mat from Great Sippewissett transferred 3x into anaerobic seawater media kindly provided by Kim Gallagher. This culture was classified as Methylococcoides methylutens (99% similarity; see K. Gallagher course report from 2011). It was originally enriched on trimethylamine (TMA, 10 mM) and then subsequently transferred into methanol (62 mM). For both experiments, 1 mL was taken from the original starting culture with a nitrogen-flushed syringe and inoculated into sterile anaerobic media. The Citrobacter nickel experiment was performed in HCl-washed 160 mL serum bottles with H2 CO2 overpressure to 12 psi in duplicate at six conditions (calculated from amount of nickel added assuming no contamination, although that assumption was likely wrong as discussed below): no added Ni, 10 nM Ni, 100 nM Ni, 500 nM Ni, 1000 nM Ni and 1000 nM Ni without any H2 CO2 overpressure as a control. The Methylococcoides nickel experiment was also performed in HCl-washed 160 mL serum bottles in duplicate with either H2 CO2 overpressure to 12 psi, 10 mM TMA or 62 mM methanol at five conditions: no added Ni, 100 nM Ni, 500 nM

Ni, 1000 nM Ni and 1000 nM without any substrate added (no H2 CO2, TMA or methanol). Growth was monitored by GC for methane production or HPLC for acetate production (see below).

Methane analysis by gas chromatography (GS) Methane concentrations were analyzed on a Shimadzu GC 2014 gas chromatograph by injection of 0.1 mL of headspace (from a total of 130 mL headspace) with a gas-tight syringe and detection by flame ionization. Stoppers of pure cultures were sterilized with ethanol and flaming before withdrawal of headspace. GC sampling needles were autoclaved prior to use for pure culture samples. The methane peak eluted at ~0.35 minutes. The methane peak area (in millions) was converted to concentration in mmol/L using a “fudge factor” provided by S. Zinder of 0.24.

9

High-performance liquid chromatography (HPLC) Acetate concentrations in liquid media were analyzed on a Shimadzu LC-2010C HT HPLC. First, 1 mL of liquid culture was centrifuged for 10 minutes at maximum speed. Then 0.9 mL of supernatant was pipetted into an HPLC sample vial and combined with 0.1 mL of 5 N

H2SO4 and analyzed by HPLC.

Catalyzed-Reporter Deposition-Fluorescent In Situ Hybridization (CARD-FISH) Samples of primary enrichments, first transfers and pure cultures of samples F1_G2 and F4_G2 were targeted for CARD-FISH analysis, along with first transfers of S1_G4 and S4_G4. Additionally, primary enrichments of samples from marine sediment, School Street Marsh, Great Sippewissett and Cedar Swamp started in July 3, 2011 during the second round of sampling were taken for CARD-FISH analysis. Liquid enrichment cultures were preserved by combining 1 mL of sample with 100 µL of 20% paraformaldehyde and incubation for 1 hour at room temperature. Samples were centrifuged for 10 minutes at 10,000 x g, resuspended in 1x PBS and vortexed. This step was repeated 3x. Samples were stored in a 1:1 mix of PBS/ethanol at -20°C. Initial enrichments samples with abundant sand particles were sonicated 7 times for 30 sec on ice. Muddy enrichment culture were sonicated 1 time for 30 sec on ice. After sonication, sediment- rich samples were diluted 1:100 and 500 µL of sample was filtered onto 0.2µm 25-mm diameter Millipore Isopore polycarbonate membrane filters. For samples without sediment, 100 µL of undiluted culture was filtered. Filters were then washed 2x with sterile-filtered 1x PBS and allowed to air dry. One-eighth filter slices were checked with DAPI staining to ensure that sufficient cells were present for CARD-FISH analysis. This was the case for 6 out of 12 filters. Filters were then stored at -20°C until embedding. Embedding was performed by lightly spraying low boiling point 0.1% agarose (35-40°C) onto the filters. Filters were then air dried on Whatman paper. Permeabilization was performed with lysozyme for filters destined for bacterial probe CARD-FISH and with either HCl or proteinase-K with filters destined for archeal probe CARD- FISH. For the lysozyme permeabilization, the filter was incubated in fresh lysozyme solution (10 mg mL-1 in 0.05 M EDTA, pH 8.0; 0.1 M Tris-HCl, pH 8.0) for 60 min at 37°C. For HCl permeabilization, the filter was incubated in 0.1 M HCl for 30 sec at room temperature and then

10 washed well in 1x PBS. For proteinase-K permeabilization, the filter was incubated for 5 minutes in fresh proteinase-K solution (5.6 µL of proteinase-K stock; 0.05 M EDTA, pH 8.0; 0.1 M Tris- HCl, pH 8.0; 0.5 M NaCl) followed by 5 mins of inactivation in 0.01 M HCl at room temperature and a wash in MilliQ water. Endogenous peroxidases were inactivated by incubating all filters in

0.15% H2O2 in methanol for 30 min at room temperature. Filters were then washed in 96% ethanol and dried on Whatman paper. Five CARD-FISH probes were used for hybridization: EUBI-III for bacteria (35% formamide (FA)), ARC915 for archaea (35% FA), Eury806 for marine euryarcheota (0% FA), MSMX860 for Methanosarcinales (50% FA) newly purchased for the course and MS1414 for (45% FA) kindly provided by the Max Planck Institute for Marine Microbiology. FISH hybridization was performed at 46°C for 2 hours on a parafilm-wrapped slide in a 50-mL Falcon tube containing a Kim-Wipe soaked in a formamide- water mixture of the proper formamide concentration for the hydridization buffer. The filters were then washed in prewarmed washing buffer (10 min; 48°C) and 1x PBS for 15 min at room temperature. For CARD amplification, 1 mL of amplification buffer was combined with 10 µL of

0.15% H2O2 and 1 µL of Alexa 488 or 594 dye and incubated at 46°C for 30 min. The filter was then washed with 1x PBS for 5-10 mins and then washed thoroughly with MilliQ and 96% ethanol and allowed to dry thoroughly. When performing double hybridizations, another inactivation step was included at this point (0.15% H2O2 in methanol for 30 min at room temperature) and then a second round of hybridization-washing-amplification was performed with another probe and dye. Lastly, filters were stained with DAPI on a slide and stored at -20°C for at least 30 mins before microscopy. A Zeiss Axio Imager M2 was used for slide imaging.

Bioinformatics with BLAST and QIIME A command-line BLAST search was performed on mcrA clone libraries against the curated FunGene mcrA database (fungene.cme.msu.edu). First, vector sequences were removed from each clone using the function Cross Match. Then, the FunGene mcrA database was downloaded and all gene hits from uncultured organisms were removed. The following commands were then used to BLAST all of the clones against the database to determine the closest cultured match to each clone:

11 $ formatdb -i fungene_mcrA_no_uncultured_db.fasta -o T -n formatted_fungene_mcrA_db -p F $ blastall –i mcrA_clone_library.fasta –d formatted_fungene_mcrA_db -p blastn -v 1 -b 1 -m 8 -o mcrA_top_hits.fa.m8 454 amplicon data were analyzed using the computer program QIIME. First each library was split using the command split_libraries.py. Then the sequences were clustered using the command pick_otus.py. Clustering was performed at 97% identity using uclust. The script pick_rep_set.py was used to pick a representative sequence from each of the 97% identity clusters. was assigned on the selected reads using the RDP classifier through the command assign_taxonomy.py. The threshold percentage was changed to 50% to optimize for archaeal 16s hits at the genera level. Finally, an OTU table was built using the script make_otu_table.py and summarized at the genera level using the command summarize_taxa.py.

Results and Discussion Isolation of a methanogen from Little Sippewissett Salt Marsh Of the 16 samples designed to enrich for methanogens from the group enrichments, only one (S1_G3) yielded a pure methanogenic culture. The majority of the group enrichments grew colonies after 1-2 weeks of growth on plates, but colony PCR revealed that these colonies were primarily acetogenic bacteria. The first round of colony PCR from spread plates using universal 16s primers yielded a sequence for F1_G4 with 100% assignment classification in the Ribosomal Project Database (RDP; http://rdp.cme.msu.edu) to Acetobacterium. Similarly, DNA from pure liquid cultures (F1_G2 and F1_G3) grown up from single colonies from streak plates did not amplify with mcrA or 16s archaeal primers, but did amplify with bacterial 16s primers. Sequencing revealed that they had 100% assignment classification to Citrobacterium (F1_G2) and Acetobacterium (F1_G3; see Appendix A for 16s sequences). A DAPI-stained image of Citrobacteria is provided in Fig. 3. It seems that citrobacteria and acetogenic bacteria were selected for by the freshwater anaerobic media; it was very difficult to remove them in our enrichment and isolation protocol. Nothing grew up in ~2 weeks from the other colonies inoculated into liquid media (F4_G2, F4_G4, S1_G1, S4_G1, S4_G2 and S4_G3).

12

Figure 3. DAPI-stained image at 100x of Citrobacter pure culture. Colony PCR with 16s archeal primers from streak plates did not amplify DNA from the 7 colonies tested; all amplification was contamination by Vibrio. Colony PCR with mcrA primers yielded a positive signal for 1 out of 13 of the colonies tested from the plates streaked for isolation (Fig. 4A,B). Unfortunately, the colonies from this plate (SW1_G3) were not tested with archeal PCR primers so no 16s sequence is currently available. The product of mcrA primer amplification had 85% similarity to Methanoplanus petrolearius (isolated from an oil field in the Gulf of Guinea (Brambilla et al., 2010; Ollivier et al., 1997)) and 83% and 82% similarity to organophilum and Methanogenium boonei, respectively (Fig. 4C). One liquid S1_G3 culture produced ~4 mM methane day-1 by the end of the course, whereas the second liquid culture (picked from another colony on the same plate) produced very little methane (<0.2 mM methane day-1) (Fig. 5). Microscopy of the SW1-G3 isolate revealed rapidly fading F420 autofluorescence and small peanut-shaped rods (~1-2 µm; Fig. 4D). The sequence of the mcrA gene from this isolate is given in Appendix B. A 30% glycerol stock of this culture in anaerobic seawater media was made and frozen at -80°C in Loeb for experimentation in future Microbial Diversity courses.

13 Figure 4. Final steps of isolation of a new strain of Methanosarcinales from Eel Pond. (A) Streak plate with S1_G3 colonies after ~2 weeks of growth. (B) mcrA DNA amplification with colony PCR of one of the colonies. (C) Phylogenetic tree showing the relationship of the isolate S1_G3 with other cultured methoagens. (D) 100x phase DIC image of the methanogen isolate.

Figure 5. Methane production by pure cultures of S1_G3 grown on H2 CO2.

14 Nickel limitation experiments Nickel limitation experiments generally did not show any nickel dependence on methane or acetate production. No methane was produced by the Citrobacter sp. culture (which at the time was thought to be a methanogen) (Fig. 6). Acetate was produced by the Citrobacter culture across all Ni concentrations tested (~35 mM was present after 8 days of growth), but not when

H2 CO2 overpressure was omitted (Fig. 7). No significant methane production was measured for the Methanococcoides culture grown on H2 CO2 (Fig. 8). This was predicted by literature reports that Methanococcoides does not grow on H2 CO2 (Sowers and Ferry, 1983). Methane production was highest for Methanococcoides grown on TMA (Fig 8). The methane production rate flattened on day 4-5 and an addition of TMA and methanol was made to bring the culture to the same final concentrations as previously added (10 mM TMA and 62 mM methanol). This produced a large increase in methane production in the TMA culture but not in the methanol culture (Fig. 8). There was likely significant nickel carry-over and contamination in these cultures that prevented the onset of nickel limitation. Although the serum bottles and blue butyl stoppers were washed in 20% hydrochloric acid for 24 hours, nickel contamination was likely brought in via needles used for GC measurements and from the 1 mL of original inoculation at 1000 nM Ni concentration.

15

Figure 6. Methane production by Citrobacter under a range of Ni concentrations. No methane was produced.

Figure 7. Acetate production by Citrobacter under a range of Ni concentrations.

16

Figure 8. Methane production by a mixed culture (predominantly Methanococcoides methylutens) grow on trimethylamine (TMA), methanol and H2 CO2 under a range of Ni concentrations. The downward arrows indicate the time points when additional substrate (TMA or methanol) were added to the cultures. CARD-FISH A total of 12 anaerobic chemotrophic samples were processed for CARD-FISH analyses (see methods). Of those 12, only 6 had sufficient cells on the filter for CARD-FISH: the first transfer of F1_G2, the first transfer of S4_G4, the pure F1-G2 culture (Citrobacter), the primary enrichment from Cedar Swamp, the primary enrichment from School Street Marsh and the primary enrichment from Great Sippewissett. In the first round of CARD-FISH, EUBI-III and ARC915 probes were used. This experiment gave the third indication (besides no methane production and no archeal 16s nor mcrA DNA amplification) that the microbe in the turbid F1_G2 pure culture was not a methanogen because the EUBI-III probe bound to all the cells in the culture, and the ARC915 probe did not bind to any cells. Primary enrichment samples from

17 Cedar Swamp turned out to have problems with autofluorescence seemingly coming from pieces of sediment in the sample. Problems were encountered with the S4_G4 CARD-FISH sample because the probes tested bound to all of the cells in the culture, suggesting that the permeabilization was too effective. Both HCl and proteinase-K permeabilizations were tested, but the same problem occurred in both. In the future, it would be useful to purchase recombinant pseudomurein endopeptidase (PeiW) for permeabilization since this enzyme has been found to be the most effective for CARD-FISH with methanogens (Kubota et al., 2008). The sample that worked the best for CARD-FISH with a variety of probes was the first transfer of F1_G2. It had high cell density (2.5±0.9 x 107 cells mL-1) and displayed specific binding for the bacterial probe (EUBI-III; 24±3 % of the total cells), the archaeal probe (ARC915; 9±5 %), the marine euryarcheota probe (Eury806; 19±3%), the Methanosarcinales probe (MSMX860; 6±4%) and the Methanosarcinaceae probe (MS1414; 2±1%). Either the bacterial or archaeal probe or both probes must have not bound to all the cells in the sample, unless there were high abundances of anaerobic eukaryotes in this sample, which seems unlikely. Furthermore, it is clear that the archaeal probe did not bind to all archaea since the marine euryarcheota probe bound to an additional 10% of the cells. Double hybridizations were performed with pairs of MSMX860-ARC915 (Fig. 9), MS1414-MSMX860 (Fig. 10) and MSMX860-Eury806 (Fig. 11) CARD-FISH probes. The ARC915 probe did not bind to all the Methanosarcinales (MSMX860 probe) cells as indicated by the red fluorescent cells without any tints of yellow (green + red) in Fig. 9. Archeal and Methanosarcinales cells tended to be coccoidal cells (although one imaged cell was an long filament, see upper left panel in Fig. 8) whereas unstained cells were dominantly rods and about 25% of them fluoresced with the EUBI-III probe (Fig. 10). The double hybridization with MS1414-MSMX860 probes showed abundant orange cells, indicating that both the probes were binding, with greater fluorescence deriving from the red probe (MS1414) (Fig 11). The cell shape of the Methanosarcinales was generally coccoidal. Several green or light green cells were detected, indicating that there were some Methanosarcinales present that were not members of the Methanosarcinaceae, suggesting that they were probably Methanosaeta as indicated by previous studies of the probe’s coverage (Crocetti et al., 2006). Many cells were clumps into Methanosarcinales-rich aggregates. The double hybridization with Methanosarcinales and

18 marine euryarcheotal probes showed distinct populations of red and green cells, with some overlap primarily in pink cells (Fig 12). Marine euryarcheotal cells were generally rod-like and clumped with cells stained with only DAPI (Fig 12).

Figure 9. Double hybridization images at 100x showing blue DAPI stained cells, red-dyed (Alexa 594) Methanosarcinales (MSMX860) cells and green-dyed (Alexa 488) Archeal (ARC915) cells.

19

Figure 10. Single hybridization CARD-FISH image at 100x showing blue DAPI stained cells and green-dyed (Alexa 488) bacterial (EUBI-III) cells.

Figure 11. Double hybridization CARD-FISH images at 100x and 63x showing blue DAPI stained cells, red-dyed Methanosarcinaceae (MS1414) cells and green-dyed (Alexa 488) Methanosarcinales (MSMX860) cells.

20

Figure 12. Double hybridization CARD-FISH images at 100x showing blue DAPI cells, red- dyed Methanosarcinales (MSMX860) cells and green-dyed (Alexa 488) marine euryarcheota (Eury806) cells. Methane production and community analysis of four sample sites around Woods Hole and Falmouth, MA Initial enrichments with the four samples collected during the second round of sampling from Cedar Swamp, Great Sippewissett, School Street Marsh and marine sediments were grown on H2 CO2 and measured for methane production rates. The Great Sippewissett sample produced more methane than the other samples and still appeared to be on the linear portion of the growth curve on the 9th day of growth at ~11 mM methane (Fig. 13). The Cedar Swamp sample produced the second highest amount of methane, but maxed out at ~3 mM methane on day 5. The samples from School Street Marsh and marine sediment produced minimal methane (< 2 mM) after 9 days of growth. These results are consistent with the original group 4 enrichment from School Street Marsh that produced only 1.7 mM methane after 15 days of growth. The

Great Sippewissett sample (from an anoxic pond muddy sediment) grown on H2 CO2 produced slightly more methane than the Little Sippewissett group enrichment sample (11 vs 9 mM).

21

Figure 13. Methane production from enrichments of sediment from Cedar Swamp, School Street Marsh, Great Sippewissett and marine sediment. A community analysis of methanogen population structure in the four sites sampled for this mini-project was conducted by comparing the results of mcrA clone libraries and 454 pyrosequencing. mcrA clone libraries were successfully obtained for Cedar Swamp (165 clones) and School Street Marsh (62 total clones). 454 pyrosequencing data was successfully obtained for marine sediment (JG1, 4192 total OTUs; 0 methanogen OTUs), School Street Marsh (JG2, 5279 total OTUs; 1 methanogen OTU), Great Sippewissett (JG3, 4131 total OTUs; 3 methanogen OTUs) and Cedar Swamp (JG4, 6166 total OTUs; 224 methanogen OTUs). The methanogenic communities were compared at the genera level from the 454 amplicon data and the clone library for Cedar Swamp (Fig. 14). The dominant methanogen order represented in both libraries was the Methanomicrobiales, followed by Methanosarcinales and Methanobacteriales. There was a large discrepancy in the dominant order between the two libraries: the 454 pyrosequencing found Methanosphaerula as the most abundant genera and the mcrA clone library gave as the dominant genera. This discrepancy could be the result of using different primers for amplifying data for the two libraries. Similarly, the School Street Marsh mcrA clone library was also dominated by Methanomicrobiales, but also contained significant Methanosarcinales clones and a few

22 Methanobacteriales clones as well (Fig 15). The Great Sippewissett sample did not clone properly and only 3 methanogen OTUs were found in the 454 library: 2 hits for Methanohalophilus and one hit for Methanosphaera.

Figure 14. Cedar Swamp methanogen community analysis at the genera level for the 454 pyrosequencing data (given as a percentrage out of a total of 224 methanogen OTUs) and the mcrA clone libraries (given as a percentrage out of a total of 165 total clones). “M” stands for the Methanomicrobiales order, “S” stands for Methanosarcinales order and “B” stands for “Methanobacteriales” order.

23

Figure 15. School Street Marsh methanogen community analysis at the genera level for the mcrA clone library (given as a percentage out of a total of 62 total clones). 454 pyrosequencing yieded only one methanogen OTU in the Methanococcus . “M” stands for the Methanomicrobiales order, “S” stands for Methanosarcinales order, “B” stands for “Methanobacteriales” order and “C” stands for Methanococcales order. Attempts at amplification of mcrA genes from the marine sediment were unsuccessful and 454 pyrosequencing yielded zero methanogen OTUs, suggesting that marine sediment had very low if any methogen populations. It is possible that the methanogens died off in the 2 weeks that the sample was stored at 4°C until DNA extraction or it could be that methanogens were very rare at this sediment depth if another community is dominant (likely sulfate reducers).

Future Improvements Below is a list of suggested changes to the methanogen isolation and study protocol for future Microbial Diversity courses at MBL. 1. Include antibiotics (i.e. rifampcin) in the anaerobic growth media to kill off acetogens from the very start of the enrichments.

24 2. If attempting to conduct metal limitation experiments, perform multiple transfers of methanogens into media lacking the metal of interest to dilute the carryover of metals from previous growth media. 3. Eel Pond and Cedar Swamp show evidence of having very robust and culturable methanogen populations (in the case of Eel Pond; attempts were not made to isolate a methogen from Cedar Swamp). It might be worthwhile to focus methanogen isolation efforts on those two localities instead or in addition to School Street Marsh and Little Sippewissett. 4. It is highly suggested that future courses study the putative Methanoplanus strain isolated from Eel Pond described in this mini-project. It may grow well on formate in

addition to H2 CO2 based on the formate use of its relatives Methanoplanus and Methanogenium. It is hoped that the (meta)genome sequencing attempt with this strain is successful and that the data can be used by future classes. 5. This mini-project has only scratched the very surface of the tremendous dataset contained in the 454 pyrosequencing data sets, particularly the one from Cedar Swamp has great interest for analyzing the abundant and diverse methanogenic community contained there. Hopefully future classes will benefit from access to this data (JG1-JG4 libraries).

Acknowledgements Many thanks to my fellow students, TAs, course faculty and course directors Steve Zinder and Dan Buckley for their time and helpful discussions. In particular, thanks to Suzanna Brauer and Annie Rowe for assistance with anaerobic techniques and culturing, Chuck Pepe- Ranney and Libusha Kelly for help with sequencing and bioinformatics, and Sara Kleindienst with CARD-FISH analyses. Kimberly Gallagher kindly provided her culture of Methanococcoides methylutens. Thanks to Carly Buchwald and the rest of Group 3 for thinking outside the box and inoculating their anaerobic chemotrophic culture with another type of mud! Thanks to Bill Metcalf and Victoria Orphan for helpful discussions. Erik Zettler kindly provided a sample of marine sediment. Funding for the course tuition was provided by a PEO Scholar Award and MBL’s Horace W. Stunkard Scholarship Fund.

25 References Brambilla, E., Djao, O.D.N., Daligault, H., Lapidus, A., Lucas, S., Hammon, N., Nolan, M.e.a., 2010. Complete genome sequence of Methanoplanus petrolearius type strain (SEBR 4847T). Standards in Genomic Sciences 3, 203-211.

Crocetti, G., Murto, M., Bjorsson, L., 2006. An update and optimisation of oligonucleotide probes targeting methanogenic Archaea for use in fluorescence in situ hybridisation (FISH). Journal of Microbiological Methods 65, 194-201.

Kubota, K., Imachi, H., Kawakami, S., Nakamura, K., Harada, H., Ohashi, A., 2008. Evaluation of enzymatic cell treatments for application of CARD-FISH to methanogens. Journal of Microbiological Methods 72, 54-59.

Luton, P.E., Wayne, J.M., Sharp, R.J., Riley, P.W., 2002. The mcrA gene as an alternative to 16S rRNA in the phylogenetic analysis of methanogen populations in landfill. Microbiology 148, 3521–3530.

Ollivier, B., Cayol, J.L., Patel, B.K.C., Magot, M., Fardeau, M.L., Garcia, J.L., 1997. Methanoplanus petrolearius sp. nov., a novel methanogenic bacterium from an oil-producing well. FEMS Microbiology Letters 147, 51-56.

Schönheit, P., Moll, J., Thauer, R.K., 1979. Nickel, cobalt, and molybdenum requirement for growth of Methanobacterium thermoautotrophicum. Archives of Microbiology 123, 105-107. Sowers, K.R., Ferry, J.G., 1983. Isolation and characterization of a methylotrophic marine methanogen, Methanococcoides methylutens gen. nov., sp. nov. Applied and Environmental Microbiology 45, 684-690.

Sowers, K.R., Ferry, J.G., 1985. Trace metal and vitamin requirements of Methanococcoides methylutens grown with trimethylamine. Archives of Microbiology 142, 148-151. Thauer, R.K., Kaster, A.-K., Seedorf, H., Buckel, W., Hedderich, R., 2008. Methanogenic archaea: ecologically relevant differences in energy conservation. Nature Reviews in Microbiology 6, 579-591.

26 Appendix A

Partial sequence of 16s bacterial gene from F1_G2 isolate from School St Marsh grown on H2 CO2 (100% RDP classified as Citrobacter)

TGCAGTCGAACGGTAGCACAGAGGAGCTTGCTCCTTGGGTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACT GCCCGATGGAGGGGGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGAGGGGGACCTTC GGGCCTCTTGCCATCGGATGTGCCCAGATGGGATTAGCTAGTAGGTGGGGTAACGGCTCACCTAGGCGACGATCCCT AGCTGGTCTGAGAGGATGACCAGCCACACTGGAACTGAGACACGGTCCAGACTCCTACAGGAGGCAGCAGTGGGGAA TATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGAAGAAGGCCTTCGGGTTGTAAAGTACTTTCA GCGAGGAGGAAGGTGTTGTGGTTAATAACCGCAGCAATTGACGTTACTCGCAGAAGAAGCACCGGCTAACTCCGTGC CAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGTCTGTCA AGTCGGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATCCGAAACTGGCAGGCTAGAGTCTTGTAGAGGGGGGT AGAATTCCACGTGTAGCGGTGAAATGCGTACAGATCTGGAGGAAT

Partial sequence of 16s bacterial gene from F1_G3 isolate from School St Marsh grown on H2 CO2 (100% RDP classified as Acetobacterium)

TGCAAGTCGAACGAGAAGATTATGATTAAGCCTTCGGGCGAGAGAATAATTGGAAAGTGGCGAACGGGTGAGTAACG CGTGGGTAACCTGCCCTATGGAAAGGAATAGCCTCGGGAAACTGGGAGTAATGCCTTATGAAATATTGAAGTCGCAT GGCTTTAATATCAAACGCTCCGGTGCCATAGGATGGACCCGCGTCCCATTAGCTGGTTGGTGAGGTAACGGCTCACC AAGGCGACGATGGGTAACCGGTCTGAGAGGGCGAACGGTCACACTGGAACTGAGACACGGTCCAGACTCCTACGGGA GGCAGCAGTGGGGAATATTGCGCAATGGGGGAAACCCTGACGCAGCAATACCGCGTGAATGAAGAAGGTCTTCGGAT CGTAAAGTTCTGTTATTGGGGAAGAAGAAAAGACGGTACCCAAGGAGAAAGTCCCGGCTAACTACGTGCCAGCAGCC GCGGTAATACGTAGGGGACAAGCGTTGTCCGGATTTACTGGGCGTAAAGGGCACGCAGGCGGTTTTTTAAGTCAGAT GTGAAAGGTCCCGGCTCAACCGGGGAAATGCATTTGAAACTGGAGAACTTGAGTATTGGAGAGGCAAGTGGAATTCC TAGTGTAGCGGTGAAATGCGTAGAGATTAGGAGGAACACCAGTGGCGAAGGCGGCTTGCTGGACAAATACTGACGCT GAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCA

27 Appendix B

Alignment of three partial sequences of mcrA gene from SW1_G3 methanogen isolate from Eel Pond grown on H2 CO2

28