Novel facultative Methylocella strains are active methane consumers at terrestrial natural gas seeps

Abstract

Background: Natural gas seeps contribute to global climate change by releasing substantial amounts of the potent greenhouse gas methane and other climate-active gases including ethane and propane to the atmosphere. However, methanotrophs, bacteria capable of utilising methane as the sole source of carbon and energy, play a significant role in reducing the emissions of methane from many environments. Methylocella-like facultative methanotrophs are a unique group of bacteria that grow on other components of natural gas (i.e. ethane and propane) in addition to methane but a little is known about the distribution and activity of Methylocella in the environment. The purposes of this study were to identify bacteria involved in cycling methane emitted from natural gas seeps and, most importantly, to investigate if Methylocella-like facultative methanotrophs were active utilisers of natural gas at seep sites.

Results: The community structure of active methane-consuming bacteria in samples from natural gas seeps from Andreiasu Everlasting Fire (Romania) and Pipe Creek (NY, USA) was investigated by DNA stable isotope probing (DNA-SIP) using ¹³C-labelled methane. The 16S rRNA gene sequences retrieved from DNA-SIP experiments revealed that of various active methanotrophs, Methylocella was the only active methanotrophic genus common to both natural gas seep environments. We also isolated novel facultative methanotrophs, Methylocella sp. PC1 and PC4 from Pipe Creek, able to utilise methane, ethane, propane and various non-gaseous multicarbon compounds. Functional and comparative genomics of these new isolates revealed genomic and physiological divergence from already known methanotrophs, in particular, the absence of mxa genes encoding calcium-containing methanol dehydrogenase. Methylocella sp. PC1 and PC4 had only the soluble methane monooxygenase (sMMO) and lanthanide-dependent methanol dehydrogenase (XoxF). These are the first Alphaproteobacteria methanotrophs discovered with this reduced functional redundancy for C-1 metabolism (i.e. sMMO only and XoxF only).

Conclusions: Here, we provide evidence, using culture-dependent and culture-independent methods, that Methylocella are abundant and active at terrestrial natural gas seeps, suggesting that they play a significant role in the biogeochemical cycling of these gaseous alkanes. This might also be significant for the design of biotechnological strategies for controlling natural gas emissions, which are increasing globally due to unconventional exploitation of oil and gas.

Keywords: Biological methane; DNA stable isotope probing; Facultative methanotrophs; Geological methane; Methylocella; Natural gas; Propanotrophs; Soluble methane monooxygenase; XoxF-methanol dehydrogenase.

Fig. 1

Relative abundance of bacterial genera in native environmental samples from Pipe Creek and Andreiasu Everlasting Fire, based on 16S rRNA gene amplicon sequencing. Genera with a relative abundance of higher than 5% are shown in bold and with an asterisk (*). 16S rRNA gene amplicon sequence data for Andreiasu Everlasting Fire were reported in Farhan Ul Haque et al. [59]

Fig. 2

Consumption of methane by environmental samples from natural gas seep sites of Pipe Creek (a) and Andreiasu Everlasting Fire (b). Microcosms containing environmental samples were incubated under ¹³C-methane (red circles) or ¹²C-methane (blue circles) as the only sources of C or energy without any supplementary nutrients. The total amount of methane was injected into the headspace in two spikes (approximately 100 μmol/gram of fresh sample per spike). Data points show the mean (with error bars showing the standard errors) of duplicate incubations for each substrate

Fig. 3

Community profile of the enriched heavy (C-13 H) and light (C-13 L) DNA fractions of ¹³C-methane incubations from DNA-SIP experiment with the Pipe Creek samples (a) and Andreiasu Everlasting Fire samples (b), analysed by 16S rRNA gene amplicon sequencing. Sequencing community profiles of heavy (C-12 H) and light (C-12 L) fractions of control incubations with ¹²C-methane are also presented. Taxa represented by black borders and in parenthesis are identified as “¹³C-labelled” in that experiment. Taxa present at a relative abundance lower than 1% in any replicate of C-13 H fraction are included in “Others”. Data presented here are the mean of biological duplicates

Fig. 4

Phylogenetic analysis based on 16S rRNA sequences from new Methylocella isolates (bold red) along with other known Methylocella (bold black) strains. Accession number for the nucleotide sequences are given in brackets. Sequences were aligned using Mega 7.0 and the optimal tree (drawn to scale, with branch lengths measured in the number of substitutions per site) with the sum of branch length = 0.65 is shown where the evolutionary history was inferred using the neighbour-joining method (1067 positions in the final dataset)

Fig. 5

a The soluble methane monooxygenase genes in Methylocella sp. PC1 and PC4 isolates compared with their homologues in Methylocella silvestris BL2. Names of the genes are given above, and amino acid identities to their homologous proteins in Methylocella silvestris BL2 are shown in brackets. b Phylogenetic analysis of MmoX (alpha subunit of soluble methane monooxygenase) from new Methylocella isolates (bold red) along with MmoX from other known Methylocella (bold black) strains based on the derived amino acid sequences. Protein names and accession numbers for the sequences are given in brackets. Amino acid sequences were aligned using Mega 7.0 and the optimal tree (drawn to scale, with branch lengths measured in the number of substitutions per site) with the sum of branch length = 0.53 is shown where the evolutionary history was inferred using the neighbour-joining method (392 positions in the final dataset). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) ranged from 59 to 100

Fig. 6

a Gene clusters encoding for the two types of lanthanide-dependent methanol dehydrogenases XoxF3 and XoxF5 in Methylocella sp. PC1 and PC4. Names of the genes are given above. Another copy of xoxF5 is also present in the genomes of Methylocella PC1 and PC4, but only as a singleton gene (see Additional file 1: Figure S4). b PCR amplification showing the absence of mxaF in Methylocella sp. PC1 (lane 1) and Methylocella sp. PC4 (lane 2) along with positive control Methylocella silvestris BL2 (lane 3) and no template control (lane 4). Lane M represents 1 kb DNA ladder. c Growth of Methylocella sp. PC4 with (red) and without (blue) lanthanum. Cultures were grown in DNMS medium with methanol (20 mM) as the only source of carbon and energy. Each point shows the average of duplicate cultures with error bars (invisible if smaller than symbol size) showing the standard errors. A control without any carbon substrate showing no growth of cells (green) was also performed

Fig. 7

a Gene operon encoding the propane monooxygenase in Methylocella sp. PC1 and PC4 compared with their homologues in Methylocella silvestris BL2. Names of the genes are given above and amino acid identities to their homologous proteins in Methylocella silvestris BL2 are shown in brackets. b Gene operon putatively encoding butane monooxygenase in Methylocella sp. PC1 and PC4 compared with their homologues in Thauera butanivorans. c Phylogenetic analysis of the alpha subunit of propane monooxygenase (PrmA) and putative butane monooxygenase (BmoX) from Methylocella PC1 and PC4 isolates (red and blue, respectively). PrmA and BmoX from known closely related strains (bold black) and other soluble diiron monooxygenases (ThmA, alpha subunit of tetrahydrofuran monooxygenase; Blr3677 and Rsp2792, putative monooxygenases; PmoC, alpha subunit of propene monooxygenase; MmoX, alpha subunit of methane monooxygenase; DmpN, alpha subunit of phenol hydroxylase; TomA3, alpha subunit of toluene ortho-monooxygenase; IsoA, alpha subunit of isoprene monooxygenase; and TouA, toluene o-xylene monooxygenase component) from different bacteria are also presented. Compressed MmoX sequences are same as the known methane monooxygenases presented in Fig. 5b. Protein names and accession number for the sequences are given in brackets. Amino acid sequences were aligned using Mega 7.0 and the optimal tree (drawn to scale, with branch lengths measured in the number of substitutions per site) with the sum of branch length = 5.7 is shown where the evolutionary history was inferred using the neighbour-joining method (316 positions in the final dataset). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) ranged from 60 to 100