Members of the methanotrophic genus Methylomarinum inhabit inland mud pots

Proteobacteria capable of converting the greenhouse gas methane to biomass, energy, and carbon dioxide represent a small but important sink in global methane inventories. Currently, 23 genera of methane oxidizing (methanotrophic) proteobacteria have been described, although many are represented by only a single validly described species. Here we describe a new methanotrophic isolate that shares phenotypic characteristics and phylogenetic relatedness with the marine methanotroph Methylomarinum vadi. However, the new isolate derives from a terrestrial saline mud pot at the northern terminus of the Eastern Pacific Rise (EPR). This new cultivar expands our knowledge of the ecology of Methylomarinum, ultimately towards a fuller understanding of the role of this genus in global methane cycling.


Sampling and enrichment
Two mud pots in the Davis-Schrimpf seep field (latitude 33.200, longitude -115.579) were sampled on 9 September 2012. Temperature and pH were determined for each pot (Table 1). Approximately 400 ml of mud (∼30% solids) from each pot was sampled into sterile 1l polycarbonate bottles. The bottles were capped and transported to the laboratory at ambient temperature.
On 12 September 2012, a 10 ml slurry from each bottle was transferred to 30 ml serum bottles, capped with black butyl stoppers, and sealed with aluminum crimps. Methane was added to a final headspace concentration of 50%, and the bottles were kept at 22 C.
On 30 January 2016, additional samples were collected from 6 mud pots and associated gryphons (mud volcanos less than 3 m in height) in the Davis-Schrimpf seep field (Sites 1-6, Fig. S1) as described above, except that 50 ml polypropylene tubes were used for collection, and the collected volume was approximately 30-50 ml per sample. Salinity, temperature, and pH were recorded for each sample (Table 1).

Media
A modified nitrate mineral salts (modified NMS) media was used for cultivation. This medium is similar to nitrate mineral salts (NMS, ATCC 1306), except that trace elements from DSMZ 141 replaced the trace elements from medium ATCC 1306. Vitamins were added (DSMZ 141, 1,000Â), and sodium chloride was added to 2%. The medium was sterilized by filtration. To solidify medium for plates, Bacto agar (Becton Dickinson and Company, Franklin Lakes, NJ, USA) was added to 1.5% final concentration and the medium was autoclaved.

Isolation of methanotrophic species
In June 2015, approximately 20 ml of slurry from each enrichment bottle was streaked onto modified NMS agar plates. Plates were incubated in a cultivation jar (Oxoid anaerobic jar, Thermo Fisher Scientific, Grand Island, NY, USA) with 50% methane and 50% air in the headspace. Incubations proceeded for fourteen days at 22 C, and individual colonies of distinct morphology were targeted for additional analysis.

Molecular identification of methane-oxidizing colonies
Colonies were probed for the presence of the pmoA gene (which provides a molecular indicator of methane oxidation potential) via polymerase chain reaction (PCR). Single colonies were picked with sterile pipet tips into 50 ml of sterile phosphate-buffered saline (PBS, 10 mM Na 2 HPO 4 , 1.8 mM KH 2 PO 4 , 137 mM NaCl, 2.7 mM KCL, pH 7.4). Two ml of this suspension was used as a template in PCR; final reaction volume was 20 ml. Reactions were performed in 1Â Taq reaction buffer, with 0.2 mM each dNTP, 10 mM of forward and reverse primers (pmoA189f and mb661r, Costello & Lidstrom, 1999), synthesized by Integrated DNA Technologies, Coralville, IA, USA) and 1 unit Taq polymerase (New England Biolabs, Ipswich, MA). Reactions were heated to 95 C for 10 min to lyse bacterial cells, and then cycled at 95 C (1 min), 54 C (1 min), and 72 C (1 min). After 30 cycles, a final extension step at 72 C (1 min) was included. Four ml of the product was electrophoresed through a 1% (w/v) agarose gel, stained with SYBR safe dye (Thermo Fisher Scientific, Grand Island, NY, USA), and visualized with ultraviolet transillumination. Bacterial colonies that gave a PCR amplicon of ∼500 base pairs were retained for further analysis.
For determination of 16S rRNA gene sequences, PCR was performed as above except that primers Bac8f and Uni1492r (Lane, 1991, synthesized by Integrated DNA Technologies, Coralville, IA, USA) replaced pmoA189 and mb661r, and a ∼1,400 bp product was expected. When using PCR to amplify target genes from DNA (instead of bacterial cells), the initial denaturation step was reduced from 10 to 3 min.

Clone library construction
To identify the major contaminating species in an early enrichment culture, PCR primers Bac8f and Uni1492r (Lane, 1991) targeting the 16S rRNA gene were used to generate a  Fig. S1. b The pH values of gryphons were not determined due to the high viscosity of these samples. c Salinity was measured in the supernatant derived from pelleted sediment, using a hand held refractometer.
∼1,400 bp PCR product. This product was purified from 1% agarose gel electrophoresis using the Wizard DNA extraction kit (Promega, Madison, WI, USA) following manufacturer's recommendations. The gel-purified product was cloned into the pSMART GCHK vector using the pSMART system (Lucigen, Middleton, WI, USA) and following manufacturer's recommendations. Forty-eight colonies containing the 16S rRNA gene insert were sequenced commercially.

Sequencing
To prepare amplicons for sequencing, PCR products were diluted with sterile water to 100 ml, and separated from primers on 96-well filter plates under vacuum (EMD Millipore, Billerica, MA, USA). Cleaned amplicons were resuspended in 30 ml of 10 mM Tris pH 7.5, and DNA concentrations were measured by Nanodrop (Thermo Scientific, Wilmington, DE, USA). DNA amplicons were sequenced commercially (Laragen, Culver City, CA, USA).

Purification of Methylomarinum strain SSMP-1
To generate a pure culture of Methylomarinum strain SSMP-1 (Salton Sea Mud Pots), colonies that gave a positive pmoA PCR signal were inoculated into 3 ml of sterile liquid modified NMS and incubated under 50% methane headspaces in Balch style tubes. Inocula were grown with gentle rotation at 22 C until they reached slight but visible turbidity. These cultures were diluted via a tenfold dilution series to an estimated 1,000 cells per ml (4 dilution steps). At this point in the dilution series, 2 fold dilutions were made past extinction (an additional 14 tubes). This set of 18 tubes was incubated with rotation at 22 C, and the 'most dilute' tube that ultimately grew was subjected to this same dilution-to-extinction procedure, until molecular (amplification and sequencing of the 16S rRNA gene) and microscopic (fluorescent in situ hybridization) inspection failed to detect any contaminants. At this point Methylomarinum strain SSMP-1 was considered pure. The isolate was stored at -80 C as described in Hoefman et al. (2012).

Growth determinations
The growth rate of Methylomarinum strain SSMP-1 was determined by measuring the change in optical density (450 nm) of liquid cultures grown at 10, 22, 28, 37, and 45 C. Duplicate tubes were grown at each temperature and results were averaged. The effect of trace metals on the growth of Methylomarinum strain SSMP-1 was determined by measuring the change in optical density (450 nm) of liquid cultures amended with 1Â, 2Â, 4Â, 6Â, 8Â, 10Â and 20Â concentrations of DSMZ141 trace element solution. Duplicate tubes were grown for each amended media, and results were averaged.

DNA extraction
To extract DNA from cell cultures, 4-10 ml cultures at an optical density of 0.3 were pelleted in polypropylene tubes in a Beckman Coulter Allegra X-15R tabletop centrifuge at 4,000 RPM for 20 min. The supernatant was discarded and the cell pellets were subjected to lysozyme-proteinase K treatment followed by a phenol-chloroform extraction protocol and ethanol precipitation (Wilson, 2001). The resulting DNA was treated with an additional cesium chloride and ethidium bromide cleanup step (Saunders & Burke, 1990), to attain a level of purity suitable for molecular analysis.
To extract DNA from environmental samples, the Power Soil DNA extraction kit was used following manufacturer's specifications (MO BIO Laboratories, Carlsbad, CA, USA). Approximately 0.2 g of sediment was extracted per sample. A FastPrep bead beater (Bio101, Thermo Fisher Scientific, Waltham, MA, USA) was used to disrupt cells during this procedure.
To observe motility in Methylomarinum strain SSMP-1, log phase cells were removed from their stoppered vessel with an 18-gauge needle attached to a 3 ml syringe. A drop (∼25 ml) of this culture was deposited gently onto a glass slide and covered with a 20 Â 60 mm coverslip. This wet-mounted preparation was viewed using phase contrast on an Olympus BX51 microscope under oil immersion at 1,000Â magnification.

Quantitative PCR
To enumerate Methylomarinum from cultures and environmental samples, qPCR primers specific to Methylomarinum were designed (Methylomarinum_pmoA267f: ACTRTTTCTGTATTRGCGCTG, and Methylomarinum pmoA375r: AAYCARGTTTGAT GGGAATACG). Primers were synthesized by Integrated DNA Technologies, Coralville, IA, USA. At least 1 mismatch in each primer is predicted for all other validly published methanotrophic genera, thus, as a paired set, these primers are not expected to easily amplify from methanotrophs outside of Methylomarinum. The primers were empirically validated in this regard against genomic DNA corresponding to Methylomarinum strain SSMP-1, Methylosinus trichosporium strain OB3b (Genbank accession number AF186586), Methylocystis parvus strain OBBP (AF533665), Methylomicrobium album strain BG8 (FJ713039), Methylosarcina lacus strain LW14 (AY007286), Methylosarcina fibrata (AF177325), Methylomonas sp. strain LW13 (AF150793), Methylomonas methanica strain S1 (U31653), Methylomicrobium alcaliphilum strain 2OZ (FO082060), Methyloprofundus sedimenti strain WF1 (KF484908), and from the nitrosifiers Nitrosococcus oceani (U96611) and Nitrosomonas europaea (AF037107). Optimal annealing temperature for this primer pair was 53 C. The 'no template control' used PCR grade water in place of DNA template, and generated a Cq of 39 or greater in quantitative PCR. DNA templates from environmental samples were assayed at 1-fold, 4-fold, and 16-fold to verify that the assay fell within the linear range (i.e. no inhibitory substances in the DNA extract). The standard curve was generated against 10 fold dilutions of genomic DNA from Methylomarinum strain SSMP-1. Assuming a pmoA copy number of 1 and genome size of 4.3 Â 10 6 bp, which is the size of the Methylomarinum vadi strain IT4 genome, the standard curve ranged from 20 to 2 Â 10 7 copies per microliter. The dynamic range of the assay encompassed these values.
Quantitative PCR was performed using the Power SYBR Universal mastermix (Thermo Fisher Scientific, Grand Island, NY, USA) and a Biorad CFX96 Real time C1000 Touch Thermal Cycler System. Prior to cycling, DNA was denatured for 3 min at 95 C. Then, forty cycles of qPCR (denaturing at 95 C, for 15 s and annealing and extending at 57 C for 1 min) were performed. A dissociative melt curve (0.5 per minute) was included to evaluate amplicon homogeneity. Measurements were made in triplicate and averaged; standard deviations were calculated. The products of qPCR were sequenced commercially (Laragen, Culver City, CA, USA).
To reconstruct a phylogenetic tree for the A subunit of particulate methane monooxygenase, pMMO-A sequences from cultured methanotrophs within Methylococcaceae, Methylothermaceae, and select environmental sequences were aligned using MUSCLE (http://www.phylogeny.fr; Dereeper et al., 2008). The resulting alignment was analyzed via maximum-likelihood, with the AMO-A sequences of Nitrosomonas cryotolerans ATCC 49181 (AF314753) and Nitrosospira briensis strain C-128 (U76553) as the outgroup (not shown). Sequences generated in this study were deposited to Genbank under accession numbers KU740209-KU740211 and KX060584.

Isolation of Methylomarinum strain SSMP-1
We routinely place environmental samples under methane, as one means of building resources for investigating microbial methane oxidation. Two such incubations, from samples collected at the Davis-Schrimpf seep field (Fig. 1) near the southeast shore of the Salton Sea, were established in September 2012.
In June 2015, a small amount of sediment slurry from each of these incubations was inoculated onto solid modified NMS medium and placed under 50% methane headspace. After 14 days, approximately 22 distinct colony types were identified that ranged in size, coloration, and shape. These colonies were probed for the particulate methane monooxygenase gene (pmoA), whose product, pMMO-A, catalyzes aerobic methane oxidation. Of the 22 colonies probed, two generated strong pmoA amplicons. Both colonies derived from the "Site 2" mud pot (Fig. S1). The sequences of these two amplicons were identical, and were 89% identical (at the DNA level) to the pmoA sequence of the marine methanotroph Methylomarinum vadi IT-4.
Both colonies were streaked three times successively on agar in an effort to purify them from co-cultivating bacteria; however, this approach did not yield a pure culture. One of the colonies was inoculated into liquid media and purified via the dilution-to-extinction method twice, until no contaminating species were evident (Fig. 2). At this point, both the pmoA and the 16S ribosomal genes were amplified from the culture. The 16S rRNA gene demonstrated 96.7% identity to Methylomarinum vadi IT-4, and the pMMO-A protein scores were 97% (identity) and 99% (similarity) when compared to Methylomarinum vadi IT-4 (Fig. 3). These results place strain SSMP-1 within Methylococcaceae, and within Methylomarinum, in Gammaproteobacteria.

Characteristics of Methylomarinum strain SSMP-1
The doubling time of Methylomarinum strain SSMP-1 in liquid modified NMS was measured at 10, 22, 28, 37, and 45 C. The strain did not grow at 10, 37, or 45 C. At 22 C, the strain doubled every 15 h, while at 30 C the strain doubled every 18 h. These results indicate a lower optimal growth temperature for Methylomarinum strain SSMP-1 than that reported for Methylomarinum vadi IT-4, which has an optimal growth temperature of 37-43 C (Hirayama et al., 2013). As Methylomarinum strain SSMP-1 approached saturation, it formed large flocs, in sheets approximately 3-4 mm across. Because hydrothermal features often exhibit elevated metal concentrations, the tolerance of Methylomarinum strain SSMP-1 to elevated metals was preliminarily assessed, by amending culture media with up to 20-fold higher trace element solution. Even at the highest end of this amendment series, Methylomarinum strain SSMP-1 grew with near normal kinetics (not shown). In contrast, the reference marine organism Methyloprofundus sedimenti (Tavormina et al., 2015) was inhibited for growth at a 6-fold elevation in trace element concentration.
When grown on solid modified NMS media, Methylomarinum strain SSMP-1 formed colonies within two weeks that were approximately ½ mm across. Colonies were smooth and shiny at two weeks but rough and irregular at four weeks. Non-uniform pink pigmentation developed as incubations proceeded past 2 weeks (Fig. S2).
Individual cells of Methylomarinum strain SSMP-1 were short, thick rods approximately 1.3-1.5 mm long (Fig. 2), similar to the size reported for Methylomarinum vadi IT-4 (Hirayama et al., 2013). At a magnification of 1,000Â, cells appeared granular (Fig. 2). Cells were clearly motile (Movie S1), in keeping with reported features of the type species of the genus (Hirayama et al., 2013). During the course of purifying Methylomarinum strain SSMP-1, cells were routinely imaged with FISH. These images indicated that Methylomarinum strain SSMP-1 initially existed in close association with smaller rod-shaped bacteria (Fig. 2A). These rods were typically surrounded by Methylomarinum strain SSMP-1, but occasionally occurred independently.
The contaminating rods were not detected in the pure culture (Fig. 2B). A 16S rRNA clone library derived from the enrichment culture revealed that approximately 83% of 16S rRNA genes affiliated with Methylomarinum, whereas 16% affiliated with genus Methylophaga (96.9% identity to accession KF790925; Mishamandani, Gutierrez & Aitken, 2014). This finding suggests that organisms capable of utilizing methanol as a carbon source may associate with Methylomarinum in the Davis-Schrimpf seep field environment.

Natural abundance of Methylomarinum in the Davis-Schrimpf seep field
Because Methylomarinum strain SSMP-1 was isolated following a three-year enrichment at room temperature, its significance to the in situ microbial community in the seep field is not clear. To assess the natural abundance of Methylomarinum in the seep field, DNA from six mud pots and associated gryphons was extracted and probed directly for the pmoA gene using quantitative PCR with primers specific to Methylomarinum. These primers did not amplify DNA from ten reference organisms (D. Fradet, 2016, unpublished data). Environmental samples for this analysis were collected in January 2016, after four years of drought in California, and the water level of the mud pots was significantly reduced relative to September 2012, when the original samples were collected (Fig. S3). Additionally, the samples were collected in January when ambient temperatures are near their lowest values of the year. Despite these distinctions between the two collection dates in this study, Methylomarinum pmoA genes were readily detected in all mud pots, and were most abundant at Site 2, the original source of Methylomarinum strain SSMP-1 (Table 1). At this site they were detected at over 10 4 copies per nanogram of DNA. Sequencing these amplicons confirmed that they are closely related to Methylomarinum, although several base changes exist between in situ amplified pmoA sequences and the Methylomarinum strain SSMP-1 pmoA sequence. A near-complete pmoA gene sequence was recovered from Site 2 DNA, to verify affiliation with Methylomarinum (Fig. 3).
A general trend existed between increasing salinity of the mud pots and decreased abundance of Methylomarinum, as assessed by abundance of the pmoA gene (Table 1). Also, a clear distinction in abundance existed between gryphons and mud pots. In gryphons, abundance of Methylomarinum did not exceed ∼50 copies per nanogram of DNA; this equates to three orders of magnitude less than the abundance of Methylomarinum in the Site 2 pot. The salinity of gryphons was between 0 and 1.2 in all cases, lower than the salinity in mud pots where values were between 5.0 and 17.2 ( Table 1). The reported tolerated salinity range of Methylomarinum vadi is 1.0-8.0, with an optimum of 2.5-3.0 (Hirayama et al., 2013), which fits reasonably well with the numbers reported here, from individual pots and gryphons within the seep field.

DISCUSSION
The fumaroles, mud pots, and gryphons in the Salton Sea basin are emission sources of one-carbon compounds, including methane and carbon dioxide. The Davis-Schrimpf seep field provides a terrestrial analog of submarine hydrothermal features in the Guaymas Basin, within the Gulf of California (Svensen et al., 2009). Yet, methane (and carbon dioxide) cycling in the Davis-Schrimpf field is essentially uncharacterized. The present work demonstrates, firstly, that methane-oxidizing bacteria reside in abundance in the Davis-Schrimpf seep field; secondly, that the ecological range of genus Methylomarinum includes inland mud pots; and thirdly, that the Davis-Schrimpf seep field shares not only morphological and geochemical elements with deep-sea hydrothermal features, but also microbial ecological elements.
Hydrothermal features along seismic zones source CO 2 and CH 4 to the environment, and the biological constraints that limit the escape of these gases to the atmosphere are not fully characterized. In deep marine systems, methane emissions are naturally remediated by sediment-hosted Archaea and Bacteria. Methane that escapes sediments can be oxidized within the water column, which hosts a distinct methane-oxidizing assemblage below 200 m water depth (Tavormina, Ussler & Orphan, 2008;Tavormina et al., 2013). Methane-oxidizing microbial assemblages in shallow ocean waters have yet to be clearly or definitively identified. Within the Davis-Schrimpf seep field, where methane is sourced from shallow saline water, methane that is not remediated in situ is expected to directly enter the atmosphere. Thus, methane-oxidizing assemblages in the mud pots within this field are of particular interest, for their role in limiting the escape of methane to the atmosphere.
Both Methylomarinum vadi strain IT-4 and Methylomarinum strain SSMP-1 were isolated from hydrothermal features, along seismic zones. Methylomarinum vadi strain T2-1 was isolated from mud at several meters water depth, near Hiroshima, Japan. Molecular signatures (pmoA) of related uncultured organisms (Fig. 3 this work; Abdallah et al., 2014) have been reported from the seawater:brine interface in the Atlantis II Deep brine pool in the Red Sea, at a water depth of approximately 2,000 m. Red Sea brine pools are characterized by hypersalinity, low oxygen, acidity, and elevated concentrations of metals (Bougouffa et al., 2013). Molecular signatures (16S rRNA) with slightly lower relatedness to the Methylomarinum 16S rRNA gene have been detected from hydrothermal vent sites, sediment incubations amended with manganese, and from hypersaline sediments from the Gulf of Mexico ( Fig. 3; Beal, House & Orphan, 2009;Hirayama et al., 2007;Li et al., 2013;Lloyd, Lapham & Teske, 2006;Schauer et al., 2011). Taken together, the available evidence may suggest that genus Methylomarinum preferentially establishes in environments characterized by elevated salinity or temperature, reduced pH, and/or high metal content. Initial queries here into the roles of temperature, salinity, and metals suggest that genus Methylomarinum may be particularly tolerant of elevated metal concentrations. Indeed, analysis of genomic potential among available methanotrophic genomes reveals that the number of genes conferring resistance to cadmium, zinc, and cobalt ranges from a low of 17 in Methylohalobius to a high of 34 in Methylococcus. Importantly, Methylomarinum also encodes 34 genes relevant to cadmium, zinc, and cobalt tolerance (D. Fradet, 2016, unpublished data). This result places Methylomarinum at the uppermost end of the range. Direct quantitation of Methylomarinum from habitats ranging in salinity, pH, and methane and metal concentrations, will likely provide additional insight into the physicochemical or geological factors that promote the establishment of this genus in the environment. Such studies, coupled with controlled laboratory investigations with cultured members of Methylomarinum, will ultimately improve our understanding of the role of this genus in global methane cycling.