Correlation of Key Physiological Properties of Methanosarcina Isolates with Environment of Origin

ABSTRACT

It is known that the physiology of Methanosarcina species can differ significantly, but the ecological impact of these differences is unclear. We recovered two strains of Methanosarcina from two different ecosystems with a similar enrichment and isolation method. Both strains had the same ability to metabolize organic substrates and participate in direct interspecies electron transfer but also had major physiological differences. Strain DH-1, which was isolated from an anaerobic digester, used H₂ as an electron donor. Genome analysis indicated that it lacks an Rnf complex and conserves energy from acetate metabolism via intracellular H₂ cycling. In contrast, strain DH-2, a subsurface isolate, lacks hydrogenases required for H₂ uptake and cycling and has an Rnf complex for energy conservation when growing on acetate. Further analysis of the genomes of previously described isolates, as well as phylogenetic and metagenomic data on uncultured Methanosarcina in anaerobic digesters and diverse soils and sediments, revealed a physiological dichotomy that corresponded with environment of origin. The physiology of type I Methanosarcina revolves around H₂ production and consumption. In contrast, type II Methanosarcina species eschew H₂ and have genes for an Rnf complex and the multiheme, membrane-bound c-type cytochrome MmcA, shown to be essential for extracellular electron transfer. The distribution of Methanosarcina species in diverse environments suggests that the type I H₂-based physiology is well suited for high-energy environments, like anaerobic digesters, whereas type II Rnf/cytochrome-based physiology is an adaptation to the slower, steady-state carbon and electron fluxes common in organic-poor anaerobic soils and sediments.

IMPORTANCE Biogenic methane is a significant greenhouse gas, and the conversion of organic wastes to methane is an important bioenergy process. Methanosarcina species play an important role in methane production in many methanogenic soils and sediments as well as anaerobic waste digesters. The studies reported here emphasize that the genus Methanosarcina is composed of two physiologically distinct groups. This is important to recognize when interpreting the role of Methanosarcina in methanogenic environments, especially regarding H₂ metabolism. Furthermore, the finding that type I Methanosarcina species predominate in environments with high rates of carbon and electron flux and that type II Methanosarcina species predominate in lower-energy environments suggests that evaluating the relative abundance of type I and type II Methanosarcina may provide further insights into rates of carbon and electron flux in methanogenic environments.

INTRODUCTION

The physiology and ecology of Methanosarcina species is of substantial interest because Methanosarcina plays a key role in anaerobic digestion, an important bioenergy strategy, and they contribute to methane production in soils and sediments that are an important source of atmospheric methane (1–3). Methanosarcina are unique among methanogens in their broad range of substrate utilization, which typically includes acetate and/or methylated compounds (methanol, methylamines) and, in some instances, H₂ (2). All Methanosarcina species that have been evaluated can also accept electrons for the reduction of carbon dioxide to methane via direct interspecies electron transfer (4). In addition to producing methane, Methanosarcina may impact the biogeochemistry of anaerobic environments through the reduction of extracellular electron acceptors, such as Fe(III) and humic substances (5–7).

The two Methanosarcina species that have been studied in greatest detail are Methanosarcina barkeri and Methanosarcina acetivorans (3, 8). Remarkably, these two members of the same genus have distinct mechanisms for energy conservation (3, 9). M. barkeri physiology centers around H₂ metabolism, even when acetate is provided as the substrate (10). M. barkeri requires three hydrogenases for growth on H₂, Ech (energy converting), Frh (F₄₂₀-reducing), and Vht (methanophenazine-linked) (3, 10, 11). Each hydrogenase interacts with different electron carriers that are important for the reduction of carbon dioxide to methane. During acetate metabolism, the Ech hydrogenase complex generates H₂ in the cytoplasm, and the Vht complex reoxidizes this H₂ after it diffuses across the cell membrane (3). This “intracellular H₂ cycling” generates a proton gradient while coupling the electron transfer between oxidative and reductive components of the methane production pathway (3).

In contrast, M. acetivorans is unable to use H₂ as an electron donor (12), and it does not employ H₂ cycling in acetate metabolism (3, 8). M. acetivorans does have several gene clusters coding for Frh/Fre and Vht/Vhx hydrogenases, but it lacks genes for the Ech hydrogenase complex (11). Furthermore, no detectable hydrogenase enzyme activity has been detected in M. acetivorans (13). Instead, the six-subunit Rnf (Rhodobacter nitrogen fixation) complex serves as a membrane-bound electron transport chain that generates a sodium ion gradient that drives ATP production during acetoclastic methanogenesis (8, 9).

Cytochrome content is another major difference between M. barkeri and M. acetivorans. All Methanosarcina species contain b-type cytochromes that are essential components for methane formation (14). However, M. acetivorans also contains multiheme c-type cytochromes, whereas M. barkeri does not (15). Most notably, M. acetivorans, which is able to conserve energy to support growth from electron transfer to extracellular electron acceptors, requires a multiheme membrane-bound cytochrome A (MmcA) for extracellular electron transport (6, 7). In contrast, attempts to grow M. barkeri with extracellular electron acceptors were unsuccessful (5). M. mazei, which is similar to M. barkeri in lacking an Rnf complex and its ability to grow on H₂ (16, 17) does contain a 5-heme c-type cytochrome, but it is of unknown function. Deletion of the gene for this in M. mazei, or its homolog in M. acetivorans, had no impact on methane production or extracellular electron exchange (7, 18).

We hypothesized that the physiological differences in Methanosarcina species could have ecological consequences. We combined isolation and characterization of new Methanosarcina strains with genomic analysis of previously described Methanosarcina isolates and molecular analyses of the distribution of Methanosarcina phyla and metabolic genes in diverse environments. The results suggest that Methanosarcina can be separated into two types that not only have distinct physiologies, but also preferred habitats.

RESULTS AND DISCUSSION

Similar isolation methods but different environments yield Methanosarcina with distinct physiologies.

As detailed in Materials and Methods, strains DH-1 and DH-2 were enriched and isolated under similar conditions with the exception that temperatures for enrichment and isolation were environmentally appropriate. Strain DH-1 was recovered from anaerobic digester sludge and was enriched and isolated at 37°C, whereas strain DH-2 came from subsurface sediments and thus was enriched and isolated at 25°C.

Both strain DH-1 and strain DH-2 are closely related to previously described Methanosarcina species (see Fig. S1 in the supplemental material). The 16S rRNA and mcrA gene sequences from strain DH-1 are 99.59% and 95.29% identical, respectively, to M. vacuolata (Fig. S1). Genome analysis confirmed that DH-1 is most similar (97.73%) to M. vacuolata (Fig. 1), which was isolated from a mesophilic anaerobic digester (19). The 16S rRNA and mcrA gene sequences of strain DH-2 are 99.82% and 96% identical, respectively, to M. subterranea strain HC-2, which was isolated from groundwater in a deep subsurface diatomaceous shale formation (20). The genome of M. subterranea is currently not available; however, comparison of the strain DH-2 genome to other Methanosarcina genomes with the FastANI tool (average nucleotide identity [ANI], a measure of nucleotide-level genomic similarity between the coding regions of two genomes) (21) showed that DH-2 is most similar to M. lacustris (86.47%), which was isolated from anoxic lake sediments (22). Further phylogenetic comparisons of genomic data with Anvi’o software (23) indicated that strain DH-2 clustered with M. lacustris and Methanosarcina sp. strain WH1, which was isolated from sandy marsh sediments (24) (Fig. 1). Both DH-1 and DH-2 have a typical Methanosarcina coccoid morphology (Fig. 2).

FIG 1.

Phylogenetic tree constructed from concatenated proteins from genomes of the 31 representative Methanosarcina species. Organisms highlighted in red represent type I Methanosarcina, while those highlighted in blue represent type II Methanosarcina species.

FIG 2.

(A and B) Images of strain DH-1 (A) and strain DH-2 (B) obtained with scanning electron microscopy (lower panels), phase contrast (upper left), and fluorescence microscopy (upper right). Bar, 10 μm.

Both isolates were capable of acetotrophic growth and also utilized methanol, monomethylamine, dimethylamine, and trimethylamine as substrates for methanogenesis (Table 1). Neither species grew with formate, ethanol, or dimethyl sulfide (DMS) as substrates. Both DH-1 and DH-2 were capable of growing in coculture with Geobacter metallireducens converting ethanol to methane (Fig. 3). These findings are consistent with previous demonstrations that other Methanosarcina species can accept electrons for carbon dioxide reduction to methane from G. metallireducens via direct interspecies electron transfer because G. metallireducens is incapable of metabolizing ethanol with the production of H₂ or formate (18, 25, 26). Both species grew optimally at 33°C (Fig. S2A) with only slight differences in salinity and pH optima (Fig. S2B and C).

TABLE 1.

Comparison of Methanosarcina sp. DH-1 and DH-2 to other characterized Methanosarcina species

Characteristic	Data for strain:a
	1	2	3	4	5	6
Cell diameter (μm)	2.0–2.8	1.3–2.4	1.5–2.0	2.9–3.9	0.9–1.4	1.0–2.0
Temp range (°C)	15–42	25–37	20–45	20–40	10–40	18–42
Optimum temp (°C)	33	33	40–42	35	35	37–40
pH range	3–6.8	6.3–8.0	5–7.5	<5.0–7.5	5.9–7.4	6.0–8.0
Optimum pH	6.0	6.7–7.5	6–6.5	6.5	6.6–6.8	7.5
Tolerance of NaCl (M)	0–0.36	0 to >0.53	0–0.8	0 to >1	0–0.6	0.1–0.6
Optimum NaCl (M)	0.07	0.17	0.1	0.2–0.6	0.1–0.2	0.1–0.2
Utilization of:b
H2/CO2	+	−	+	−	−	+
Methanol	+	+	+	+	+	+
Acetate	+	+	+	+	−	+
Formate	−	−	ND	−	−	−
Dimethyl sulfide	−	−	ND	+	ND	−
Monomethylamine	+	+	+	ND	+	+
Dimethylamine	+	+	+	ND	+	+
Trimethylamine	+	+	+	+	+	+
Ethanol	−	−	ND	ND	ND	ND

^aStrain 1, Methanosarcina strain DH-1 (data from this study); strain 2, Methanosarcina strain DH-2 (data from this study); strain 3, Methanosarcina barkeri MS^T (90); strain 4, Methanosarcina siciliae C2J (91); strain 5, Methanosarcina subterranea HC-2^T (20); strain 6, Methanosarcina vacuolata Z-761^T (19).

^b+, positive result; −, negative result; ND, not determined.

FIG 3.

Ethanol consumption and methane and acetate production in defined cocultures established with G. metallireducens and either strain DH-1 (A) or DH-2 (B) on the fourth transfer of the cocultures. The mol CH₄/mol ethanol yields of 1.2 and 1.0 mol CH₄/mol for the cocultures with strains DH-1 and DH-2, respectively, are within the range of methane recoveries previously reported for G. metallireducens/Methanosarcina cocultures (18, 25, 26). Error bars represent triplicate samples.

The important physiological distinction between strains DH-1 and DH-2 revolved around H₂. Strain DH-1 could grow with H₂ as the electron donor and carbon dioxide as the electron acceptor, but strain DH-2 could not (Table 1). These results are consistent with genomic differences (Table S2). Strain DH-1 has all of the genes coding for the three hydrogenase complexes that M. barkeri requires for growth on H₂. These include the Ech (energy converting) hydrogenase operon (echABCDEF), duplicate operons of the Frh (F₄₂₀-reducing) hydrogenase (frhBGDA, freAEGB), and the Vht (methanophenazine-linked) hydrogenase (vhtDCAG, vhxGAC) (3, 10, 11, 27, 28). In contrast, strain DH-2 does not have Ech hydrogenase genes and only has a gene for the beta subunit of the Frh complex, eliminating the potential for growth on H₂ or acetate metabolism via H₂ cycling. Strain DH-2 does have all six of the genes coding for subunits of the Rnf complex (rnfABCDEG) necessary for energy conservation during acetate conversion to methane in M. acetivorans (3, 8, 9). Strain DH-1 lacks the Rnf genes, suggesting that it relies on H₂ cycling during acetate metabolism, similar to M. barkeri (3, 27, 28).

The results demonstrated that strains DH-1 and DH-2 have substantially different physiologies, despite being enriched and isolated on the same medium. The primary difference between them was the environment from which they were recovered.

Environment of origin is predictive of physiological type for other Methanosarcina isolates.

In order to further evaluate whether the physiology of Methanosarcina isolates can be related to their environment of origin, we analyzed the genomes of previously described Methanosarcina isolates. This analysis further demonstrated that Methanosarcina species segregate into two physiological groups, which we designate type I and type II (Table 2). The primary physiological distinction between the type I Methanosarcina, of which M. barkeri is the most studied example, and type II Methanosarcina, exemplified by M. acetivorans, is the role of H₂ in metabolism. The type I Methanosarcina species have the ability to consume H₂ as an electron donor and possess the Ech hydrogenase necessary for metabolism of acetate via H₂ cycling (10, 28) but lack an Rnf complex that is essential for energy conservation during acetate conversion to methane in the absence of H₂ cycling (8, 29). In contrast, the type II Methanosarcina species possess an Rnf complex and typically do not grow on H₂ (Table 2). The exception to this generalization is M. lacustris, which is reported to grow on H₂ (22) but contains an Rnf complex as well as multiheme c-type cytochromes (NCBI GenBank accession number CP009515.1) that are characteristic of type II Methanosarcina, as discussed in detail below.

TABLE 2.

Genome and physiological characteristics of Methanosarcina strains available in pure culture^a

Methanosarcina strain	Rnf complex present?	Ech hydrogenase present?	No. of multiheme c-type cytochromes	MmcA present?	Utilization of H2-CO2	Habitat	Reference(s) or source
Methanosarcina barkeri MS	No	Yes	0	No	Yes	Anaerobic sewage sludge digester	90, 92
Methanosarcina barkeri 227	No	Yes	0	No	Yes	Anaerobic sewage sludge digester	93
Methanosarcina barkeri CM1	No	Yes	0	No	Yes	Bovine rumen	94
Methanosarcina barkeri JCM 10043	No	Yes	0	No	Yes	Anaerobic sewage sludge digester	95
Methanosarcina barkeri_A strain 3	No	Yes	0	No	Yes	Unknown	96
Methanosarcina barkeri_B Fusaro	No	Yes	0	No	Yes	Mud from Lago del Fusaro Lake	96, 97
Methanosarcina barkeri_B Wiesmoor	No	Yes	0	No	Yes	Peat bog	98
Methanosarcina flavescens E03.2	No	Yes	0	No	Yes	Full-scale commercial biogas plant fed with maize silage, cattle manure, and dry poultry feces	99
Methanosarcina mazei Go1	No	Yes	1	No	Yes	Anaerobic sewage digester	100
Methanosarcina mazei 1.H.A.0.1	No	Yes	1	No	Yes	Sediment from the Columbia River Estuary	31
Methanosarcina mazei C16	No	Yes	1	No	Yes	Shoal mud of the southern North Sea	101
Methanosarcina mazei JCM 9314	No	Yes	1	No	Yes	Paddy field soil	102
Methanosarcina mazei JL01	No	Yes	1	No	Yes	Arctic permafrost	103
Methanosarcina mazei LYC	No	Yes	1	No	Yes	Swamp mud	104
Methanosarcina mazei S-6	No	Yes	1	No	Yes	Wastewater treatment plant sludge	90
Methanosarcina mazei SarPi	No	Yes	1	No	Yes	Rice paddy soil	104
Methanosarcina mazei SMA-21	No	Yes	1	No	Yes	Siberian permafrost-affected soil	105
Methanosarcina mazei Tuc01	No	Yes	1	No	Yes	Sediment from hydropower station reservoir	106
Methanosarcina spelaei MC-15	No	Yes	0	No	Yes	Floating biofilm on a sulfurous subsurface lake	107
Methanosarcina thermophila Ms 97	No	Yes	0	No	Yes	Sheep rumen	DSM 11855
Methanosarcina thermophila TM-1	No	Yes	0	No	Yes	Thermophilic digester sludge	108
Methanosarcina vacuolata Z-761	No	Yes	0	No	Yes	Mesophilic anaerobic digester	90
Strain DH-1	No	Yes	0	No	Yes	Sludge from anaerobic digester	This study
Methanosarcina vacuolata Kolksee	No	Yes	0	No	Yes	Lake Kolksee mud	109
Methanosarcina acetivorans C2A	Yes	No	4	MA0658	No	Methane-evolving sediments of a marine canyon	12
Methanosarcina horonobensis HB-1	Yes	No	3	Ga0072443_113899	No	Groundwater sampled from a subsurface Miocene formation	110
Methanosarcina horonobensis JCM 15518	Yes	No	3	Ga0128354_1005162	No	Groundwater sampled from a subsurface Miocene formation	110
Methanosarcina lacustris Z-7289	Yes	Yes	2	Ga0072454_11511	Yes	Anoxic lake sediments	22
Methanosarcina siciliae C2J	Yes	No	4	Ga0072451_11663	No	Submarine canyon sediments	91
Methanosarcina siciliae H1350	Yes	No	4	Ga0072474_11601	No	Production water from off shore oil well	111
Methanosarcina siciliae T4/M	Yes	No	3	Ga0072440_11646	No	Pristine lake sediment	111
Methanosarcina sp. WH1	Yes	No	3	Ga0072445_113060	No	Anoxic sandy sediments	24
Strain DH-2	Yes	No	4	Ga0399897_1739	No	Subsurface aquifer sediments	This study
Methanosarcina sp. MTP4	Yes	No	2	Ga0072449_113047	No	Salt marsh sandy sediments	112

^aFurther details regarding substrate utilization are provided in Table S1. Methanosarcina with white background represent type I Methanosarcina, whereas those shaded in gray are type II Methanosarcina.

Previously, the physiological differences between M. barkeri and M. acetivorans were attributed to the difference between the marine environment of M. acetivorans and the less saline “freshwater” environments of M. barkeri and close relatives (30). However, the distinction between type I and type II Methanosarcina does not appear to be related to salinity. One notable difference is that many of the type I Methanosarcina species have been recovered from methanogenic digesters (Table 2). In contrast, none of the type II Methanosarcina species are digester isolates (Table 2). However, it must be recognized that isolate recovery may not always be a reliable indication of which types of Methanosarcina are most abundant in specific environments. For example, M. mazei (a type I species) was isolated from sediments collected from the Columbia River Estuary, but molecular analysis revealed that M. mazei was relatively rare in this environment and that type II Methanosarcina such as M. lacustris, M. acetivorans, and M. horonobensis were much more abundant (31).

Metagenomic and phylotype data also support specific physiology-environment associations.

In order to evaluate the environmental distribution of Methanosarcina species without culture bias, previously published data on mcrA and 16S rRNA gene sequences from 20 different methanogenic environments (10 anaerobic digesters and 10 anoxic soils/sediments) were analyzed (Fig. 4). Type I species accounted for 78 ± 16.2% of the Methanosarcina isolates in the digester environments (Fig. 4A), and type II species accounted for 70 ± 15.9% of the Methanosarcina isolates in the sedimentary environments (Fig. 4B). A Chi-square test of independence and a Fisher’s exact test both confirmed that the sediment and digester Methanosarcina communities were significantly different (P < 0.0001, alpha = 0.05).

FIG 4.

Proportions of various Methanosarcina species based on mcrA, 16S rRNA, or both mcrA and 16S rRNA gene sequences from metagenomic libraries constructed from 20 different environments. Species were designated as type I or type II based on the physiology of pure culture isolates of the same species. Type I and type II species are designated with red and black symbols, respectively. All of the bioreactor environments had total organic carbon (TOC) concentrations of >5% (Table S3), while sediment TOC concentrations varied significantly. Bioreactor 1: low-salinity bioreactor (JGI GOLD IDs Ga0334882 to Ga0334890 mcrA); bioreactor 2: anaerobic solid waste digester (73, 74) (SRR8165483; mcrA); bioreactor 3: anaerobic digester in wastewater treatment plant (Gp0313021; mcrA); bioreactor 4: GAC-amended bioreactor treating municipal solid waste (MSW) (75) (SRR7687449 to SRR7687452; 16S rRNA and mcrA); bioreactor 5: bioreactor seeded with sewage sludge (76) (SRR5486931; mcrA); bioreactor 6: sewage sludge and household waste codigester (77) (ERR2586913 to ERR2586931; mcrA); bioreactor 7: cattle manure digester (78) (SRR3166092; mcrA); bioreactor 8: switchgrass digester (79) (JGI GOLD IDs Ga0134090 to Ga0134105; 16S rRNA and mcrA); bioreactor 9: bioreactor treating the dry organic fraction of MSW (80) (SRR5229592; 16S rRNA); bioreactor 10: anaerobic digester at WWTP (SRR3485656; 16S rRNA); sediment 1: mesotrophic meromictic freshwater lake, DOC ∼40 to 80 μM) (81) (IMG GOLD IDs Ga0247831 to Ga0247844); 16S rRNA and mcrA); sediment 2: paddy soil TOC 2.85% (82) (SRR11653212 to SRR11653222; 16S rRNA); sediment 3: estuary sediments, TOC >3.5% (31–33) (SRR1210425 to SRR1210426; mcrA); sediment 4: peat bog sediments, TOC 6 to 50% (36) (IMG GOLD ID Gp0348925; mcrA); sediment 5: mangrove sediment, TOC 0.7 to 11.4% (83, 84) (SRR3095812; mcrA); sediment 6: groundwater and sediments from uranium-contaminated aquifer, TOC <0.2% (38), mcrA (48); sediment 7: Amazon soil, TOC <2% (85) (SRR12110053 to SRR12110059; 16S rRNA); sediment 8: freshwater lake sediments, DOC ∼4 to 60 mg/liter) (86, 87) (IMG GOLD IDs Ga0031653 to Ga0031658; 16S rRNA); sediment 9: South Georgia marine sediments, TOC ∼0.65% (88) (SRR12815614 to SRR12815618; mcrA); sediment 10: Aarhus Bay marine sediments, TOC ∼2% (89) (SRR7119900 to SRR7119905; mcrA). Further details regarding organic carbon concentrations from various environments are available in Table S3 in the supplemental material.

The distribution of Methanosarcina in sediments was examined in more detail. Although all well-characterized type II isolates have been recovered from sediments, the molecular data demonstrated that the relative abundance of type II Methanosarcina was lower in sediments with higher organic content, such as estuaries, mangrove sediments, and peat bogs (sediments 1 to 5, Fig. 4B). Organic inputs into estuaries from agricultural and wastewater runoff promote algal blooms and enrich H₂-producing fermentative Bacteroidetes species (32–34), which would provide an electron source for hydrogenotrophic methanogenesis by type I Methanosarcina. Peat, rice paddy, and mangrove sediments, which have high concentrations of partially degraded plant debris, are also organic content rich (35–37). Type II was much more abundant in sediments expected to have smaller amounts of organic matter, such as subsurface aquifer sediments and soils/sediments with high sand content (sediments 6 to 10, Fig. 4B), which tend to have low organic carbon content (38, 39).

In order to further evaluate the possible correspondence between physiology and environment in Methanosarcina species, 144 assembled genomes from diverse environments were grouped into 32 different Methanosarcina species based on a 96% ANI cutoff value for species demarcation (Table S1). Type I or type II physiology was inferred from the presence or absence of Rnf and Ech and the ability to utilize H₂ as the electron donor for CO₂ reduction. Environments were characterized by their total organic content (TOC); those with TOC concentrations greater than 2% were considered high, and those below 2% were considered low (33, 36, 38–43) (Table S3). A mixed principal-component analysis comparing Methanosarcina strains with genomes that were at least 90% complete further confirmed that type I and type II Methanosarcina sort into two different groups based on genome characteristics, physiology, and habitat organic content (Fig. 5).

FIG 5.

Results from mixed principal-component analysis of 144 different Methanosarcina strains using the following observations: quantitative (number of multiheme c-type cytochromes) and qualitative (type I or type II, presence/absence of Rnf complex, presence/absence of Ech hydrogenase complex, presence/absence of MmcA homolog, organic content of environment from which organism was isolated/detected [high (>2% TOC) or low (<2% TOC)]). M1-M144 specifies the Methanosarcina strain described in Table S1; organisms with CheckM genome completeness scores of <90% were not included in the analysis.

Differences in mode of extracellular electron exchange.

The different environmental niches of type I and type II Methanosarcina may also be reflected in their different strategies for extracellular electron transfer. These differences in electron transfer strategies are apparent when one focuses on the presence or absence of multiheme c-type cytochromes that are known to facilitate extracellular electron transfer in many other species. Analysis of all of the available Methanosarcina genomes revealed that only type II Methanosarcina species have genes for the seven-heme, membrane-associated, c-type cytochrome MmcA (Table 2, Table S1), shown to be essential for extracellular electron transfer in M. acetivorans (7). Notably, M. barkeri, which lacks MmcA, reduced extracellular electron acceptors, such as Fe(III) and the humics analog anthraquinone-2,6-disulfonate, but unlike M. acetivorans, could not gain energy from the growth of these extracellular electron acceptors (5).

Other multiheme c-type cytochromes were detected in type II Methanosarcina but not in type I Methanosarcina, with the exception of one c-type cytochrome found in all type I M. mazei strains (Table S4). The function of this cytochrome, which is also present in type II Methanosarcina, is unknown. Gene deletion studies have indicated that this cytochrome is not involved in methane production or extracellular electron exchange (7, 18). The presence of this c-type cytochrome in M. mazei, but not other type I Methanosarcina species, and the fact that M. mazei appears to be one of three exceptions to type I and type II Methanosarcina aligning in distinct phylogenetic groups (Fig. 1), indicates that the evolution of type I and type II Methanosarcina physiologies warrants further study.

Implications.

In summary, the results of multiple lines of investigation, including both cultivation- and non-cultivation-based approaches, indicate that distinct differences in Methanosarcina physiology have ecological consequences. Type I Methanosarcina species are best suited for growth in environments with high rates of organic matter degradation, such as anaerobic digesters and organic-rich sediments. A physiology that revolves around H₂ is probably sustainable in such environments because H₂ is expected to be well above the nanomolar steady-state levels found in soils and sediments with lower energy input. However, in methanogenic soils and sediments, where steady-state H₂ concentrations are only ca. 10 nM (44), methanogens that specialize in the use of H₂ as an electron donor and have higher affinities for H₂ uptake than Methanosarcina species (14) can be expected to outcompete Methanosarcina. It is also likely that the H₂-utilizing specialists can consume H₂ that type I Methanosarcina species produce during acetate metabolism. Type II Methanosarcina may save energy by not expressing hydrogenases that would be of little use for H₂ uptake in organic-poor soils and sediments, but then require the Rnf-based alternative to intracellular H₂ cycling to conserve energy from acetate metabolism. It is also more likely that alternative electron acceptors such as Fe(III) and oxidized humic substances will become intermittently available in soils and sediments with low organic content due to intermittent drying, oxygen inputs from animal burrowing and plants, or changes in oxygen availability in the overlying water. The expression of the multiheme c-type cytochrome MmcA, which enables growth via extracellular electron transfer, may confer an additional advantage to type II Methanosarcina in such environments.

Thus, the relative importance of type I and type II Methanosarcina may provide further insights into rates of carbon and electron flux in methanogenic environments. The important physiological differences in Methanosarcina should be recognized when interpreting their role in anaerobic environments.

MATERIALS AND METHODS

Culture media and growth conditions.

Methanosarcina strains DH-1 and DH-2 were cultured under strictly anaerobic conditions in modified DSMZ 120 medium (https://www.dsmz.de/microorganisms/medium/pdf/DSMZ_Medium120.pdf), in which concentrations of Na₂S · 9H₂O and l-cysteine · HCl were adjusted to 0.5 mM and 1 mM, respectively. Yeast extract, Casitone, and resazurin were not added to the medium (25), and all cultures were incubated in an oxygen-free 80:20 N₂:CO₂ atmosphere at 37°C. Sodium bicarbonate (NaHCO₃; 2 g/liter) was added to all cultures except when pH tolerance was being tested. Either acetate (40 mM) or methanol (100 mM) was used as the electron donors and carbon sources unless otherwise noted.

Geobacter metallireducens (ATCC 53774) was routinely cultured under strict anaerobic conditions with ethanol (20 mM) provided as the electron donor and Fe(III) citrate (55 mM) as the electron acceptor at 30°C under anaerobic conditions (N₂:CO₂, 80:20) as previously described (45).

For coculture experiments, G. metallireducens and either Methanosarcina strain DH-1 or DH-2 were anaerobically grown with ethanol (20 mM) provided as the sole electron donor and CO₂ as the acceptor at 30°C as previously described (25, 26, 46).

Isolation of Methanosarcina strains.

Strain DH-1 (type I) was isolated from sludge collected from an anaerobic digester operating at a wastewater treatment plant located in Pittsfield, MA. Initial enrichment cultures were established by inoculating 0.1 g sludge into 156-ml serum bottles containing the modified DSMZ 120 medium described above, but with acetate (40 mM) provided as the electron donor. Enrichments were incubated under an 80:20 N₂:CO₂ atmosphere for 60 days at 37°C in the presence of antibiotics (kanamycin [200 μg/ml], erythromycin [200 μg/ml], and penicillin-G [50 μg/ml]).

Strain DH-2 (type II) was isolated from sediments and groundwater collected from a 24-acre experimental site located on the premises of an old uranium ore processing facility in Rifle, Colorado. This site has uranium concentrations in the water table that are 2 to 8 times higher than the water contamination limit established by UMTRA (Uranium Mill Tailings Remedial Action). Many in situ uranium bioremediation experiments have been conducted at this site (47). Sediments and groundwater for DH-2 methanogenic enrichments were collected in September 2011 from well CD-01, which was downgradient from the point of injection of ∼15 mM acetate into the subsurface to stimulate U(VI) reduction (48). Then, 5 g wet sediment and 5 ml aquifer groundwater were added to 40 ml modified DSMZ 120 medium with acetate (40 mM) in 156-ml serum bottles in an anaerobic chamber under an 80:20 N₂:CO₂ atmosphere. To reduce growth of bacteria, antibiotics (kanamycin [200 μg/ml], erythromycin [200 μg/ml], and penicillin-G [50 μg/ml]) were added to the enrichment cultures. All sediment enrichments were incubated for 60 days at 25°C.

After initial DH-1 and DH-2 methanogenic enrichments were established, serial dilutions to extinction were carried out at 37°C or 25°C, respectively, in 9 ml modified DSMZ 120 medium with acetate (40 mM) as the substrate for growth. Ampicillin (1 mg/ml), gentamicin (20 μg/ml), and tetracycline (10 μg/ml) were added to the media to suppress the growth of bacteria. The highest dilution that grew after the fourth serial transfer was then transferred to solidified modified DSMZ 120 medium (2% agar, wt/vol) in Hungate anaerobic roll tubes amended with 0.02% (wt/vol) yeast extract. Isolated single colonies were selected from each tube and resuspended in 2 ml of liquid medium and grown at 37°C or 25°C.

The purity of both strains was confirmed with microscopy and 16S rRNA gene sequencing with primers Arc344F (5′-ACGGGGYGCAGCAGGCGCGA-3′) and Arc915R (5′-GTGCTCCCCGCCAATTCCT-3′) (49). The PCR was also conducted with primers targeting bacterial 16S rRNA genes (49) to ensure that cultures were not contaminated with bacteria.

Metabolic and growth profile.

For determination of the pattern of substrate utilization, a sterile anoxic stock solution of each substrate was added to a final concentration of 10 to 50 mM in modified DMSZ 120 medium. Hydrogen as an electron donor was provided as an H₂-CO₂ mixture (80:20; at 10⁵ pascals pressure). The ability to utilize various substrates was confirmed by growth at 37°C after at least four transfers with the substrate being investigated.

The effects of pH values, temperatures, and salt concentrations on the growth of both strains were tested with modified DMSZ 120 medium with methanol (100 mM) as the substrate. Specifically, optimal salt concentrations for both strains were evaluated at 37°C with the addition of appropriate volumes from a sterile anoxic stock solution (300 g/liter NaCl). Three different buffering systems were needed to adjust DSMZ 120 medium without NaHCO₃ supplementation for pH optimum experiments. pH was adjusted with the addition of HCl or NaOH to the following buffering systems: for pH ≤ 5.0, no buffer additions were needed; for pH 5.0 to 7.0, 100 mM 2-(N-morpholino) ethanesulfonic acid (MES, pK_a = 5.98) was added; for pH 6.5 to 8.0, 100 mM 3-(N-morpholino) propanesulfonic acid (MOPS, pK_a = 6.98) was added; for pH 8.0 to 8.5, 100 mM Tricine (pK_a = 7.80) was added. After incubation at 37°C, all pHs remained within 0.1 unit of the original value.

Analytical techniques.

Methanosarcina cultures were monitored by changes in optical density measured in a split-beam, dual-detector spectrophotometer (Spectronic Genosys2; Thermo Electron Corp.) at an absorbance of 600 nm (7). Growth of G. metallireducens cultures on Fe(III) was monitored by the formation of Fe(II) over time with a ferrozine assay in the same spectrophotometer at an absorbance of 562 nm as previously described (50, 51).

Ethanol and methanol concentrations were monitored with a gas chromatograph equipped with a headspace sampler and a flame ionization detector (Clarus 600; PerkinElmer, Inc., California). Methane in the headspace was measured by gas chromatography with a flame ionization detector (GC-8A; Shimadzu) as previously described (52). Acetate concentrations were measured with a Shimadzu high-performance liquid chromatograph (HPLC) with an Aminex HPX-87H ion exclusion column (300 mm by 7.8 mm) and an eluent of 8.0 mM sulfuric acid.

Microscopy.

Cells were routinely examined by phase-contrast and fluorescence microscopy (BV-2A filter set) with a Nikon E600 microscope. For scanning electron microscopy, cells were harvested during the late exponential phase of growth and fixed with 2.5% (vol/vol) glutaraldehyde in 0.1 M phosphate buffer (pH 6.8) for 12 h at 4°C. Then, cell pellets were washed three times with 0.1 M phosphate buffer and dehydrated with solutions of increasing ethanol concentrations (35%, 50%, 70%, 80%, 90%, 95%, and 100% [vol/vol]) and a hexamethyldisilazane/ethanol solution (1:1) as previously described (53). Cells were then immersed in pure hexamethyldisilazane and dried with a stream of high-purity nitrogen. Scanning electron microscopy was conducted with an ultrahigh-resolution field emission scanning electron microscope (FEI Magellan 400; Nanolab Technologies, California, USA).

Genome extraction and analysis.

For extraction of genomic DNA, cultures (50 ml in 156-ml serum bottles) were divided into 50- ml conical tubes, and cells were pelleted by centrifugation at 4,500 rpm (Sorvall Heraeus 75006445 rotor) for 15 min. After centrifugation, cell pellets were resuspended in 10 ml TE sucrose buffer (10 mM Tris, pH 8.0, 1 mM EDTA, and 6.7% [wt/vol] sucrose), and DNA was extracted from the cell pellets as previously described (54). DNA concentrations were evaluated with the Qubit double-stranded DNA (dsDNA) high-sensitivity (HS) assay kit (Life Technologies) and sent for whole-genome sequencing by Molecular Research LP (MR DNA).

Initial raw nonfiltered DH-1 and DH-2 libraries contained 14,766,691 and 15,568,552 reads, respectively, that were ∼100 bp long. Reads were quality checked by visualization of base quality scores and nucleotide distributions with FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Sequences from all of the libraries were trimmed and filtered with Trimmomatic (55), resulting in an average of 14,585,756 and 15,342,415 quality reads for the DH-1 and DH-2 libraries, respectively. All paired-end reads were then merged with FLASH (56), resulting in 6,630,882 and 7,008,483 reads with an average read length of 152 bp. Contigs were then assembled from these quality control (QC)-filtered and merged reads with SeqMan NGen (DNAStar) and MEGAHIT software (57) with an overlapping base length of 50 bp and a minimum contig length of 500 bp.

These contigs were then submitted to Integrated Microbial Genomes and Microbiomes (IMG/MER) for preliminary annotation (img.jgi.doe.gov). The current Methanosarcina sp. DH-1 draft genome sequence contains 51 contigs with 4,061 total genes and 3,957 protein-coding genes. The draft genome sequence of strain DH-2 is composed of 37 contigs with 4,130 total genes and 4,043 protein-coding genes. The quality of the assembled genomes was determined using SeqMan NGen software (DNAStar) and the genome quality assessment program CheckM (58) on the KBase website (www.kbase.us) (Table S4).

Phylogenetic and statistical analyses.

Initial gene sequence analyses were done with tools available on the Integrated Microbial Genomes (IMG) website (img.jgi.doe.gov) and through comparisons to GenBank nucleotide and protein databases with BLASTn and BLASTx algorithms (59, 60). Some protein domains were identified with NCBI conserved domain search (61) and Pfam search (62) functions. Transmembrane helices were predicted with TMpred (63) TMHMM (64) and HMMTOP (65), and protein localization and signal peptide predictions were made with PSORTb v. 3.0.2 (66), SOSUI (67) and SignalP v. 4.1 (68).

16S rRNA gene alignments were generated with MAFFT (69), and phylogenetic trees were constructed with MEGA7 software using the maximum likelihood method with 100 bootstrap replicates (70). Average nucleotide identities (ANI) for all Methanosarcina genomes were obtained with FastANI (21), and whole-genome-based phylogenetic trees were generated in Anvi’o (23). The Anvi’o output was then imported into iTOL (https://itol.embl.de/) and FigTree v.1.4.4 (http://tree.bio.ed.ac.uk/software/figtree).

All metagenomes and genomes analyzed for environmental comparisons were downloaded from the NCBI SRA database (https://www.ncbi.nlm.nih.gov/sra) or from the JGI IMG Integrated Microbial Genomes and Microbiomes database (https://img.jgi.doe.gov). The software program Prodigal (71) was used to identify open reading frames in unassembled genomes. Databases with mcrA and 16S rRNA nucleotide sequences were built from the various Methanosarcina species with the makeblastdb function using NCBI BLAST-2.2.31+ standalone software (72). All statistical analyses were done with XLSTAT, Statistical Software for Excel.

Data availability.

Genome sequences for strains DH-1 and DH-2 have been submitted to the NCBI genome database under BioProject numbers PRJNA561374 and PRJNA561362 and BioSample numbers SAMN12616996 and SAMN12616828, respectively. The genomes have also been submitted to the IMG under IMG genome IDs 2835727744 and 2835723613.