An anaerobic digestate sample (2 L) was collected in May 2011 from an anaerobic digester in the Biogas Research Center, Hankyong National University, Republic of Korea, which was built in early 2008 and was kept running for about 4 years [8]. Pig slurry from a pig farm was the main substrate used by the anaerobic digester. The anaerobic digestate from the digester was characterized, showing a total volatile fatty acid concentration of 4,500 ppm, pH of 8.2, and chemical oxygen demand of 70,000 ppm. The anaerobic digestate was inoculated into medium to isolate methanogens.

Methanogens are highly sensitive to oxygen and need strict anoxic conditions; therefore, pre-reduced media are critical for their growth and isolation. The methods used for the preparation of media and substrate solutions and culture techniques were those of Hungate [9], as modified by Balch et al [10]. For enrichment culture and isolation, basal medium was prepared with the modification from Sowers and Schreier [11] and contained the following compounds (per liter distilled water): KCl, 0.335 g; MgCl₂·6H₂O, 2.75 g; MgSO₄·7H₂O, 3.450 g; NH₄Cl, 0.25 g; CaCl₂·2H₂O, 0.14 g; K₂HPO₄, 0.14 g; NaCl, 18.0 g; trace element solution, 10.0 mL; vitamin solution, 10.0 mL; Fe(NH₄)₂(SO₄)₂·7H₂O, 0.002 g; NaHCO₃, 5.0 g; methanol, 5.0 mL; sodium acetate, 2.5 g; yeast extract, 2.0 g; trypticase peptone, 2.0 g; resazurin, 0.001 g; cysteine hydrochloride, 0.5 g; and Na₂S·9H₂O, 0.5 g. The trace element solution was made up of the following (in one liter of distilled water): ZnSO₄·7H₂O, 0.1 g; MnCl₂·4H₂O, 0.03 g; H₃BO₃, 0.3 g; CoCl₂·6H₂O, 0.2 g; CuCl₂·2H₂O, 0.01 g; NiCl₂·6H₂O, 0.02 g; and Na₂MoO₄·2H₂O, 0.03 g. First, the minerals were dissolved with 500 mL distilled water and adjusted to pH 3.4 and then the solution was made up to a volume of 1,000 mL. The vitamin solution was made up of the following (in one liter of distilled water): biotin, 0.02 g; folic acid, 0.02 g; pyridoxine hydrochloride, 0.1 g; thiamine hydrochloride, 0.05 g; riboflavin, 0.05 g; nicotinic acid, 0.05 g; calcium pantothenate, 0.05 g; vitamin B₁₂, 0.01 g; p-aminobenzoic acid, 0.05 g; and thioctic acid, 0.05 g. Clarified sludge fluid was prepared following the method from Battumur et al [8]. The sludge fluid (5%, v/v) was used to supply unknown growth factors. The medium was prepared under a H₂/CO₂ (80:20, v/v) gas phase at 173 kPa (25 psi). The medium was adjusted to pH 6.8 to 7.0 and sterilized by autoclaving at 121°C for 30 min.

Enrichments were achieved in the basal medium, which was adjusted to pH 7.0 under H₂/CO₂ (80:20, v/v) gas phase. Vials containing medium (45 mL) were inoculated with a 5-mL inoculum. The growth of non-methanogenic organisms was inhibited by adding streptomycin sulfate (7×10⁴ IU/L) and benzyl penicillin (2×10¹⁰ IU/L) to the medium. The inoculated medium was cultured in the dark at 38°C for two weeks. After detection of high amounts of methane in the culture, 5 mL of the culture was anaerobically transferred into new, sterile basal medium. Roll tubes containing the basal medium with 1.8% agar were prepared followed by four successive transfers. Well-isolated colonies were withdrawn with Pasteur pipettes and transferred to culture tubes containing the basal medium under anaerobic condition. The culture tubes were closed with butyl-rubber stoppers and repressurized with sterile filtered H₂/CO₂ (80:20, v/v) at 173 kPa (25 psi). The organism was re-isolated with agar medium from the liquid cultures. It was inoculated on bacterial growth medium to check the purity of methanogen cultures. The bacterial growth medium contained peptone (2.5 g/L), yeast extract (2.5 g/L), D-glucose (0.5 g/L), D-cellobiose (0.5 g/L), and D-xylose (0.5 g/L) in the basal medium. For further purification of the isolated strains KOR-3, -4, -5, and -6, an antibiotic mixture was prepared that contained four antibiotics (benzyl penicillin, 0.5 mg/mL; streptomycin sulfate, 0.5 mg/mL; vancomycin-HCl, 0.2 mg/mL; and ampicillin, 0.2 mg/mL). After purification, the isolates were incubated in the basal media without antibiotics. We isolated ten methanogens from the isolation process. Strains KOR-3, -4, -5, and -6 were microscopically distinct, with a coccal shape compared to the rod shape of other isolated methanogens.

Growth temperature range, pH range, NaCl concentration range, substrate utilization, and antibiotic tolerance were investigated to determine the physiological characteristics of strains KOR-3, -4, -5, and -6. Growth was verified by observing optical density (OD) at 660 nm (OD₆₆₀) with a spectrophotometer (V-530, Jasco, Tokyo, Japan) and by measuring the concentration of methane in the gas phase using a gas chromatograph (GC-2010A, Shimadzu, Kyoto, Japan). All experiments for physiological studies were repeated twice.

Basal medium, with acetate and methanol omitted, was prepared under N₂ and used for substrate-utilization studies. Anaerobic stocks of filter-sterilized substrates (sodium formate, sodium acetate, trimethylamine, methanol, 2-propanol, and isobutanol) were prepared and added separately at final concentrations of 50 mM. Freshly grown cultures of each isolate were inoculated at 10% (v/v) and vials were incubated at 38°C for 20 days. The basal medium under H₂/CO₂ served as the control.

The optimal temperature for growth in the basal medium was determined at optimal pH. Vials inoculated with 10% (v/v) culture were incubated at temperatures ranging from 20°C to 50°C. The vials were pressurized every other day with H₂/CO₂ to ensure an adequate supply of substrate.

The optimal pH for growth in the basal medium was determined at the optimum temperature, with pH values ranging from 4.0 to 9.0. Values above pH 4.0 were produced by adding sterile Na₂CO₃ to media at pH 4.0 until the required value was reached. pH 4.0 medium was prepared by omitting NaHCO₃ from the basal medium and cooling it under a CO₂ headspace.

The sensitivity of strains KOR-3, -4, -5, and -6 to ampicillin, penicillin G, spectromycin, kanamycin, tetracycline, and chloramphenicol (all at a concentration of 100 μg/mL) was tested. Aliquots (5 mL) of the cultures were inoculated into fresh media containing one of the six antibiotics. Strains KOR-3, -4, -5, and -6 were incubated for 1 week at 38°C. The effect of antibiotics was determined by comparing the growth of cultures containing these antibiotics to that of the control.

The salinity range of the isolates was tested at NaCl concentrations ranging from 0.5% to 3.0% NaCl, with 0.5% intervals. Media with various concentrations of NaCl were prepared by adding a sterile anoxic stock solution of 58.44 g/L NaCl to the media.

An Olympus BX41 phase-contrast microscope (Olympus, Tokyo, Japan) was routinely used to observe cells. A standard Gram-stain kit (BBL Microbiology Systems, Becton Dickinson, Bergen County, NJ, USA) was used to determine the Gram-staining character. Motility was determined by the hanging-drop method, using a glass cavity slide.

Culture samples of strains KOR-3, -4, -5, and -6 grown in the basal media were used for DNA isolation (FastDNA SPIN kit for soil, MP Biomedicals, Irvine, CA, USA) following the manufacturer’s instructions. DNA integrity was checked on 1% agarose gels and DNA concentration was determined using a Nanodrop (ND 2000, Thermo Fisher Scientific, Waltham, MA, USA).

16S rRNA gene amplicons (~1,350 bp) were obtained using the following primer pair: Ar109f (5′-ACKGCTCAGTAACACGT-3′) [12] and Ar1383r (5′-CGGTGTGTGCAAGGAGCA-3′) [13]. Reaction mixtures contained the following components in a final volume of 20 μL within a 200-μL PCR tube: 2 μL PCR buffer (Takara, Tokyo, Japan), 2 μL dNTP mix (35 mM), 0.5 μL each primer (10 pmol/μL), 0.1 μL Taq DNA polymerase (Takara, Japan), 1.0 μL template DNA sample (100 ng), and 13.9 μL molecular grade water (Severn Biotech Ltd, Kidderminster, Worcester, UK). The PCR was started by immediately placing the reaction tubes into the preheated (94°C) thermal cycler (PCR Thermal Cycler Dice, Takara, Japan). The thermal program was as follows: an initial denaturation step (94°C, 4 min) followed by 30 cycles of denaturation (94°C, 30 s), annealing (55°C, 30 s), and extension (72°C, 90 s). After a final extension step (72°C, 6 min), samples were kept at 4°C until further analysis. Each PCR run included a positive control, DNA extracted from pure cultures, and a negative PCR control where molecular grade water was substituted for the DNA template. The DNA product was analyzed using gel electrophoresis (1.2% w/v agarose gel stained and run at 100 V for 30 min in 1×Tris-acetate-ethylene-diamine-tetraacetic acid [EDTA] buffer with 5 μL aliquots of each DNA product). The tris-acetate-EDTA (TAE) buffer was composed of 40 mM Tris base, 20 mM acetic acid, and 0.5 M EDTA (pH 8.0). The gel was run with 2.5 μL of Ladder I DNA quantification marker (TNT research, Seoul, Korea). The gel was imaged using the Bio Imaging System (Daihan, Seoul, Korea) and photographs were taken using WiseCapture II software (Daihan, Korea).

The mcrA genes fragments were amplified using the primer combinations MLf (5′-GGTGGTGTMGGATTC ACACARTAYGC WACAGC-3′) and MLr (5′-TTCATTGCRTAGTTWGGRTAG TT-3′), yielding ~490 bp amplicons [14]; ME1 (5′-GCMATGCA RATHGGWATGTC-3′) and ME2 (5′-TCATKGCRTA GTTDGG RTAGT-3′), yielding ~740 bp amplicons [15]; and MR1 (5′-GACC TCCACTWCGT VAACAACGC-3′) and ME2, yielding amplicons of ~1,100 bp [16]. Denaturation, annealing, and extension were carried out at 96°C (15 s), 55°C (30 s), and 72°C (90 s), respectively, with MLf/MLr primers, and 94°C (40 s), 50°C (45 s), and 72°C (90 s), respectively, with ME1/ME2 and MR1/ME2 primers.

PCR products were purified with the AccuPrep PCR purification kit (Bioneer, Daejeon, Korea). Sequencing of PCR products was performed using the BigDye terminator cycle sequencing kit on ABI 3730XL capillary DNA Sequencer (Applied Biosystems, Thermo Fisher Scientific Inc., Carlsbad, CA, USA). 16S rRNA and mcrA gene sequences from the isolated strains were compared to that of similar sequences obtained from GenBank using the BLAST program. Phylogenetic analysis was conducted using MEGA 4.0 [17]. We examined nine additional 16S rRNA sequences (M. mazei [NR041956], M. acetivorans [NR044724], Methanoculleus receptaculi [NR043961], Methanomicrobium mobile [NR 044726], Methanosphaera stadtmanae [M59139], Methanobacterium alcaliphilum [NR028228], Methanobacterium subterraneum [NR028247], Methanobacterium formicicum [JQ973735], and Methanobacterium formicicum [AF169245]). Phylogeny was further confirmed by mcrA gene and mcrA protein sequences (M. mazei [AB703645], M. barkeri [Y00158], M. lacustris [AY260439], Methanopyrus kandleri [U57340], Methanobacterium kanagiense [AB551869], Methanobacterium formicicum [EF465103], and Methanobacterium formicicum [JX141395]).

The 16S rRNA and mcrA gene sequences of strains KOR-3, -4, -5, and -6 determined in this study were deposited in the GeneBank database under No JX560177-JX560180, KC292221-KC292223, and KC150017.

New methanogens were isolated from an anaerobic digester that uses pig slurry. Pure cultures were obtained after repeated transfer in the presence of antibiotic mixtures for at least half a year and up to two years. On solid media, colonies were irregularly shaped, white (KOR-3 and -4) or yellow (KOR-5 and -6), and about 0.8 mm in diameter after incubation for two weeks. The cultures were comprised of small and irregular cocci (diameter, 0.5 μm to 1.5 μm) occurring singly and in pairs, which is similar to previous reports for M. mazei (Table 1) [4]. The cell wall was Gram-negative. Although a few cells of strains KOR-3, -4, -5, and -6 demonstrated sarcina, the shape of most cells were that of irregular cocci, which is similar to reports by Cairó et al [5] and Lai et al [18]. M. mazei could be distinguished from other Methanosarcina species by their ability to grow as single cells under appropriate condition, and Xun et al [19] found medium that could induce the disaggregation of M. mazei S-6. The variable disaggregation of all isolated strains in the study may have been because the culture condition was not appropriate for the disaggregation of the strains. It was reported the enzyme disaggregatase could be released into the medium and play a role in disaggregation [4]. Therefore, culture conditions may influence the release or activity of this enzyme.

KOR-3 and KOR-4 strains were able to use acetate but were unable to use formate, methanol, 2-propanol, isobutanol, or H₂/CO₂ as a substrate for growth and methane production (Table 1). KOR-5 and KOR-6 strains consumed acetate, methanol, and trimethylamine to produce methane but did not use H₂/CO₂, formate, 2-propanol, or isobutanol as a growth substrate (Table 1). Acetate is known to be a precursor for more than 70% of CH₄ formation in most anaerobiosis [20]. Methanosarcina sp. and Methanosaeta sp. were identified as two types of methanogens capable of metabolizing acetate [21]. It was reported that M. barkeri, M. mazei, and M. vacuolata could be distinguished from other methanogens by their ability to produce methane from trimethylamine, methanol, and acetate [22]. The most acetotrophic Methanosarcina species described, including M. acetivorans [23] and Methanosarcina thermophile [24] are unable to oxidize H₂ or reduce CO₂ to methane. From both our data and that of others, strains of M. mazei, such as strains O1M9704, GFJ07, TMA, S-6T, JC3, JC2, LYC, N2M9705, and KOR-3, -4, -5, and -6, could not use formate as a substrate. Based on morphology and substrate utilization of the four isolated strains in this experiment, this may be a basic characteristic of M. mazei.

The growths of strains isolated in this study were observed at a temperature range of 20°C to 40°C, with the fastest growth occurring at 39°C (Table 1). The pH range used for growth was 4.0 to 9.0, and the optimum pH for growth was 6.8 to 7.2 for the strains. The temperature and pH values of strains KOR-3, -4, -5, and -6 were mostly consistent with that of other M. mazei strains. Oren [22] reviewed that the optimum temperature and optimum pH of M. mazei strains ranged from 30°C to 40°C and 6.8 to 7.2, respectively. However, some strains related to M. mazei, M. lacustris, or Methanomethylovorans hollandica can be found at lower temperature (5°C to 6°C) and have been isolated from tundra wetland soil, a Russian pond polluted with paper-mill waste, anoxic sediments from a Swiss lake, and from an anaerobic digester for cattle manure operated at 6°C [16]. Although the best known strains of M. mazei grow fast at neutral pH, some can optimally grow at pH 8.5 and tolerate alkaline conditions up to pH 9.5 in places such as cow dung [25]. The strains isolated in this study can also tolerate pH value up to 9.0.

All of the isolated strains used in our experiments could grow well in salinity up to 3.0% (w/v). The optimum NaCl concentration for KOR-3, -5, and -6 and KOR-4 was 1.0% and 0.5% (w/v), respectively (Table 1). Certain species of Methanosarcina are halophilic or halotolerant in nature, such as Methanosarcina thermophila, which can grow in up to 7.0% (w/v) NaCl, with optimum growth occurring at 3.5% (w/v) NaCl [21] and M. barkeri A-12 and A-13, which can grow in up to 3% NaCl, with optimum growth occurring at 0.06% NaCl [26]. Thakker and Ranade [27] also described isolate LN 1, which grew in 3% NaCl. It is known that the halophilic or halotolerant characteristics of Methanosarcina are due to osmotic adaptation using glycine betaine, α-glutamate, N^ɛ-acetyl-β-lysine, and K⁺ as compatible solutes [28]. Compared to these species, the optimal growth for strain KOR-4 was 0.5% (w/v) salt, which is lower than that of the other isolated strains used in this study—even though KOR-4 can also grow with up to 3% NaCl.

All of the strains tolerated ampicillin, penicillin G, kanamycin, spectromycin, and tetracycline. In contrast, chloramphenicol inhibited cell growth. Clarens et al [29] reported that M. mazei, such as strains JC3, S-6, and MC3, was completely inhibited by chloramphenicol. Hilpert et al [30] found that archaea were insensitive to many antibiotics that act against eubacteria and eukaryotes, such as those inhibiting the synthesis or cross-linkage of the peptide subunit of murein or those that inhibit the synthesis of RNA. Thus, the strains KOR-3, -4, -5, and -6 are also insensitive to antibiotics used in our experiment, with the exception of chloramphenicol. As Hilpert et al [30] mentioned in their report, the protein inhibitor chloramphenicol, together with lasalocid and monensin, interferes with the cell membrane function of methanogens. However, it remains to be shown whether this insensitivity to chloramphenicol is due to the impermeability of the cytoplasmic membrane or to inactivation of the antibiotic by the cell, rather than the absence of a particular target for the antibiotic [30].

PCR of 16S rRNA yielded an amplicon size of 1,350 bp. mcrA Gene-based amplification was also used for identification purposes and it resulted in a product size of 1,100 bp. A large fragment of the 16S rRNA gene (~1,350 bp) was obtained from isolates and sequenced. Comparative 16S rRNA gene sequence analysis showed that the strains were affiliated with the order Methanosarcinales. The closest relatives of strains KOR-3, -4, and -6 were M. mazei (99%, sequence similarity) and M. acetivorans (94%) (Figure 1). The 16S rRNA sequence similarity of KOR-5 was 98% compared with M. mazei showing a small difference between KOR-5 and KOR-6 which they have no difference in physiological characteristics in Table 1. MCR, the terminal enzyme complex in the methane generation pathway that catalyzes the reduction of a methyl group bound to coenzyme M, is associated with the release of methane and is thought to be unique to, and ubiquitous in, methanogens [31]; therefore, it makes an ideal tool for their specific detection. Thus, mcrA gene, which encodes one peptide of the MCR complex, is a suitable candidate for the development of PCR-based detection of methanogens [14] and use of this functional gene serves to complement analysis of the 16S rRNA gene to provide an improved assessment of methanogen diversity [32]. Application of the mcrA gene for analysis of Methanosarcinales has been previously performed [33]. For PCR amplification of the mcrA gene, we used primers MLr/MLf, ME1/ME2 [15], and MR1/ME2 [16] to obtain nearly the full-length mcrA gene. The mcrA gene sequence also indicated that the strains were a member of the order Methanosarcinales. The closest relatives based on mcrA gene sequences were M. mazei (99%), M. acetivorans (94%), and M. barkeri (92%) (Figure 2). Additionally, we rendered the sequence part of the mcrA gene (1,100 bp) to amino acids (AA) and then assembled a molecular phylogenetic tree based on the MCR II alpha subunit sequences of the mcrA gene. The AA sequences of the mcrA gene-based tree further demonstrated that the strains were members of the order Methanosarcinales (Figure 3). No major differences were found between the DNA and AA sequences or between the different algorithms that were used (Figures 2, 3). The phylogenetic tree based on the 16S rRNA gene (Figure 1) sequence showed two major clusters. Cluster 1 and 2 in the tree were completely different from each other. Cluster 1 comprised two sub-clusters and KOR-3, -4, -5, and -6 in sub-cluster 1 showed 100% similarity with the mcrA gene of M. mazei (Figure 2). The protein sequences (Figure 3) showed two major clusters: cluster 1 comprised two sub-clusters with KOR-3, -4, -5, and -6 in sub-cluster 1, showing 100% similarity to that of M. mazei. All three approaches showed similar phylogenetic results and, thus, are in accordance with [14] who stated that the mcrA gene sequence can be used an alternative to 16S rRNA-based sequences methods, demonstrating far greater diversity than that of 16S rRNA gene sequences in the methanogen population.

On the basis of morphology, physiological characteristics, and phylogenetic analyses described above, the strains were identified as new strains of M. mazei, namely, KOR-3, -4, -5, and -6.