We report complete genome sequence of a mesophilic hydrogenotrophic methanogen Methanocella paludicola, the first cultured representative of the order Methanocellales once recognized as an uncultured key archaeal group for methane emission in rice fields. The genome sequence of M. paludicola consists of a single circular chromosome of 2,957,635 bp containing 3004 protein-coding sequences (CDS). Genes for most of the functions known in the methanogenic archaea were identified, e.g. a full complement of hydrogenases and methanogenesis enzymes. The mixotrophic growth of M. paludicola was clarified by the genomic characterization and re-examined by the subsequent growth experiments. Comparative genome analysis with the previously reported genome sequence of RC-IMRE50, which was metagenomically reconstructed, demonstrated that about 70% of M. paludicola CDSs were genetically related with RC-IMRE50 CDSs. These CDSs included the genes involved in hydrogenotrophic methane production, incomplete TCA cycle, assimilatory sulfate reduction and so on. However, the genetic components for the carbon and nitrogen fixation and antioxidant system were different between the two Methanocellales genomes. The difference is likely associated with the physiological variability between M. paludicola and RC-IMRE50, further suggesting the genomic and physiological diversity of the Methanocellales methanogens. Comparative genome analysis among the previously determined methanogen genomes points to the genome-wide relatedness of the Methanocellales methanogens to the orders Methanosarcinales and Methanomicrobiales methanogens in terms of the genetic repertoire. Meanwhile, the unique evolutionary history of the Methanocellales methanogens is also traced in an aspect by the comparative genome analysis among the methanogens.
Citation: Sakai S, Takaki Y, Shimamura S, Sekine M, Tajima T, et al. (2011) Genome Sequence of a Mesophilic Hydrogenotrophic Methanogen Methanocella paludicola, the First Cultivated Representative of the Order Methanocellales. PLoS ONE 6(7): e22898. doi:10.1371/journal.pone.0022898
Editor: Lennart Randau, Max-Planck-Institute for Terrestrial Microbiology, Germany
Received: January 25, 2011; Accepted: July 8, 2011; Published: July 29, 2011
Copyright: © 2011 Sakai et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was partly supported by a Grant-in-Aid for Young Scientists (A) (21687006) from the Japan Society for the Promotion of Science (JSPS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Methanocella paludicola is a mesophilic hydrogenotrophic methanogen within the order Methanocellales, isolated from rice field soil in Japan . The order Methanocellales is the recently proposed euryarchaeotal order, and is composed only of one genus Methanocella. Before the taxonomic description of M. paludicola and proposal of the order Methanocellales, this euryarchaeotal lineage had long been recognized as uncultured archaeal group Rice Cluster I (RC-I). It was indicated by many molecular ecological investigations that the members of RC-I would play a key role in the methane production in rice fields .
Based on the phylogenetic analysis using 16S rRNA gene sequences, the methanogens are classified into six orders, Methanopyrales, Methanococcales, Methanobacteriales, Methanomicrobiales, Methanosarcinales and Methanocellales , . It is widely accepted that the phylogenetic organization of methanogens generally responds to the taxonomic classification based on morphological, physiological and metabolic traits of a diversity of methanogens , . The order Methanocellales is closely related to the orders Methanosarcinales and Methanomicrobiales. The phylogenetic analyses of 16S rRNA gene and the methanogen-specific marker methyl-CoM reductase gene (mcrA) sequences suggest that the order Methanocellales is the most closely related to the order Methanosarcinales . Whereas, the methanogenic substrate utilization of the order Methanocellales, that is one of the most important physiological characteristics for methanogens, is similar with the order Methanomicrobiales (hydrogenotrophic) rather than the order Methanosarcinales (hydrogenotrophic, aceticlastic and methylotrophic). These results suggest that the order Methanocellales represents an evolutionary intermediate between the hydrogenotrophic (Methanomicrobiales) and versatile [hydrogenotrophic, aceticlastic and methylotrophic] (Methanosarcinales) methanogens.
To date, the genome sequences are determined for all the methanogens' orders other than the order Methanocellales , , , , . The comparative genome investigations have demonstrated new perspectives in evolution of methane-production metabolism and methanogens. For instance, based on the phylogenetic analysis using seven core enzymes of methanogenesis and a protein clustering method using a spectral clustering procedure, Anderson et al.  suggested that the methanogenic archaea should be classified into three distinct classes. The Class I consisted of the orders Methanopyrales, Methanococcales and Methanobacteriales, and the Class II and Class III were composed of the order Methanomicrobiales and the order Methanosarcinales, respectively. Although the genome sequence of RC-I (an environmental genotype of RC-IMRE50) is already available by metagenomic approach , the coordination between the genomic information and the morphological, physiological and metabolic features characterized in a pure culture is inevitable for the comparative genomic analysis. In this study, therefore, we report the complete genome sequence of M. paludicola representing the first pure culture of the order Methanocellales. The genome sequence and genomic components are characterized by through the comparison with the previously reported genome sequence of a member of the order Methanocellales RC-IMRE50 and other methanogens' genomes. The genome sequence of M. paludicola would provide new insights into understanding how the members of the order Methanocellales (i.e. M. paludicola and RC-IMRE50) are distinct from each other, and how the order Methanocellales is different from other methanogen orders.
Results and Discussion
General genome features of M. paludicola
The general features of the M. paludicola genome and the previously reported complete genome sequence of RC-IMRE50 are listed in Table 1. The genome of M. paludicola is a single circular chromosome consisting of 2,957,635 bp in length, which is slightly smaller than the genome of RC-IMRE50. Two copies of the 16S-23S-5S rRNA operons with 2 more distantly located 5S rRNA genes are present in the genome of M. paludicola. A total of 48 tRNA genes containing putative introns are scattered over the genome. The origin and terminus of replication in the genome of M. paludicola was predicted based on cdc6 (MCP_0001) and ORB (origin recognition box). A cluster of repeats was identified in the upstream region of the Cdc6 gene and was contained a mini-ORB like sequence that was found in other archaea . Interestingly, the extremely low G+C content regions are present especially in the lower half of the genome (Fig. 1). The analysis using SIGI-HMM suggested a potential acquisition by horizontal gene transfer. Whereas, the upper half region containing the origin of the genome has the higher G+C content value indicating the relatively well conservation of the upper half of the genome. Indeed, a genome plot analysis of M. paludicola and RC-IMRE50 revealed a high level of synteny especially near the origin of the genome (see Fig. S1 in the supporting information). The 3004 predicted protein-coding sequences (CDSs) of the M. paludicola genome were identified by a combination of coding potential prediction and similarity searches, with an average length of 856 bp, covering 87.4% of the genome. Through the similarity and domain searches of the predicted protein products, specific functions were assigned for 1467 genes (48.8% of the protein coding genes). The remaining 1537 genes (51.2%) were assigned to hypothetical proteins. The distribution of genes into COGs functional categories was not much different between M. paludicola and RC-IMRE50, however the genes involving in chemotaxis are markedly disparate (Table S1 in the supporting information). The genes for chemotaxis and flagella formation were identified in the genome of RC-IMRE50, while it was not found in M. paludicola. This finding is in agreement with the characteristic that cell motility was not observed in M. paludicola.
Figure 1. Circular representation of the M. paludicola genome.
From the inside, the 1st and 2nd circles show the GC skew (values greater than zero are indicated in green and smaller in pink) and the G+C % content (values greater than the average percentage in the overall chromosome or plasmid are shown in blue and smaller in sky blue) in a 10-kb window with 100-b step, respectively. The 3rd and 4th circles show the presence of RNAs (rRNA, tRNA, and small RNA genes) and CDSs aligned in the clockwise and counterclockwise directions are indicated in the upper and lower sides of the circle, respectively. Different colors mean different functional categories: red for metabolism; green for genetic information and processing; blue for membrane transport; orange for cellular processes; pink for miscellaneous and mobile elements, gray for poorly characterized function and purple for RNA genes.doi:10.1371/journal.pone.0022898.g001
Table 1. General features of M. paludicola and RC-IMRE50 genome.doi:10.1371/journal.pone.0022898.t001
The M. paludicola genome contains sufficient genes to encode a full methanogenesis pathway using H2 and CO2 (Fig. 2). Similarly to other obligate hydrogenotrophic methanogens, formate dehydrogenase complexes (MCP_1569 to 1570 and MCP_2406 to 2407) and a formate transporter (MCP_2408), a key enzyme for the growth on formate as an alternative methanogenic substrate, were also found in the M. paludicola genome. No homologous genes for alcohol dehydrogenase, which is involved in methanogenesis from primary or secondary alcohols, were found. Although an incomplete pathway of potential methanogenesis from methanol was found in the genome of RC-IMRE50 , none of the corresponding genes for utilizing methanol and other C1 compound was found in the M. paludicola genome. The gene structure for the methanogenesis pathway in the M. paludicola genome is consistent with the substrate utilization of M. paludicola only using H2 and CO2 or formate as the sole energy source for methane production .
Figure 2. Predicted functional systems and general metabolic pathway of M. paludicola.
Overview of the presence of genes for homologues of key enzymes of the carbon, sulfur and nitrogen metabolism as well as of selected electron-accepting complexes and transporters. The KEGG database was used for the reconstruction of metabolic pathways. Arrows and dot-lined arrows indicate presence and absence of the genes, respectively. The lines indicate general pathways found in methanogens (black), the unique pathways identified in M. paludicola and RC-IMRE50 (red), the pathways found in RC-IMRE50 but not in M. paludicola (blue). Fd, ferredoxin; MF, methanofuran; H4MPT, tetrahydromethanopterin; Fdh, formate dehydrogenase; Fmd, formylmethanofuran dehydrogenase; Ftr, formylmethanofuran:H4MPT formyltransferase; Mch, methenyl-H4MPT cyclohydrolase; Mtd, F420-dependent methylene-H4MPT dehydrogenase; Mer, methylene-H4MPT reductase; Mtr, methyl-H4MPT: coenzyme M methyltransferase; Mcr, methyl-coenzyme M reductase; Hdr, heterodisulfide reductase; Ech, energy-converting hydrogenase; Frh, F420-reducing hydrogenase; Mvh, non F420-reducing hydrogenase; Pgm, phosphoglucomutase; PgmB, ß-phosphoglucomutase; Gpi, glucose-6-phosphate isomerase; SuhB, inositol-1-monophosphatase/fructose-1,6- bisphosphatase; Pfp, pyrophosphate–fructose 6-phosphate 1-phosphotransferase; Pfk, 6-Phosphofructokinase; Fba, fructose-bisphosphate aldolase; TriA, triosephosphate isomerase; Gap, glyceraldehyde-3-phosphate dehydrogenase; AcyP, acylphosphatase; Pgk, phosphoglycerate kinase; Gpm, phosphoglycerate mutase; Eno, enolase; PpsA, phosphoenolpyrivate synthase; Pyk, pyruvate kinase; Por; pyruvate ferredoxin oxidoreductase; Pdh, pyruvate dehydrogenase; Acd, acetyl-CoA synthetase (ADP-forming); Acs, acetyl-CoA synthetase; Pyc, pyruvate carboxylase; Icd, isocitrate dehydrogenase; Fum, fumarate hydratase; Phi, 3-hexulose-6-phosphate isomerase; Fae, formaldehyde-activating enzyme; Hps, 3-hexulose-6-phosphate synthase; Rpi, ribose-5-phosphate isomerase; Prs, ribose-phosphate pyrophosphokinase; Apt, adenine phosphoribosyltransferase; Deo, thymidine phosphorylase; eif2B, translation initiation factor; Sat, sulfate adenylyltransferase; CysC, adenylylsulfate kinase; CysH, phosphoadenosine phosphosulfate reductase.doi:10.1371/journal.pone.0022898.g002
Most methanogens contain three distinct types of [NiFe]-hydrogenases, called energy converting hydrogenase, F420-reducing hydrogenase and F420-nonreducing hydrogenase. Some hydrogenotrophic methanogens additionally contain a [Fe]-hydrogenase, which is also known as H2-forming methylene-tetrahydromethanopterin (CH2-H4MPT) dehydrogenase (Hmd) . All these hydrogenases are involved in methanogenesis pathway . Genes for the hydrogenases except the Hmd were detected in the genome of M. paludicola. Among all those hydrogenases, the F420-nonreducing hydrogenase co-operates with a heterodisulfide reductase, and reduces heterodisulfide of coenzyme M and coenzyme B (CoM-S-S-CoB) that is synthesized in the last step of methanogenesis pathway. The heterodisulfide reductase/hydrogenase systems are different in between the obligate hydrogenotrophic methanogens such as Methanothermobacter thermautotrophicus and the members of the order Methanosarcinales . The cytoplasmic cytochrome-lacking F420-nonreducing hydrogenase (Mvh) is tightly bound to HdrABC type heterodisulfide reductase. This system is installed in the obligate hydrogenotrophic methanogens . The membrane-bound cytochrome-containing F420-nonreducing hydrogenase (Vht) co-operates with the HdrDE type heterodisulfide reductase, which is operative in the member of the order Methanosarcinales . Most of the methanogens except for some Methanomicrobiales members, which include some members do not possess F420-nonreducing hydrogenase, have either of the heterodisulfide reductase/hydrogenase systems , , . However, both F420-nonreducing hydrogenase genes (MCP_2492 to 2493, 1575 and MCP_2773 to 2774, 2767 for Mvh; MCP_2934 to 2936 and 0374 for Vht) were found in the M. paludicola genome as well as in the RC-IMRE50. The result implies that existence of isozyme systems for heterodisulfide reductase/hydrogenase in the M. paludicola and the Methanocellales members are atypical of the previously known methanogens.
Since M. paludicola is an obligate hydrogenotrophic methanogen and three HdrABC genes for heterodisulfide reductases were found in the genome (MCP_1576 to 1578, MCP_1989 to 1991 and MCP_2768 to 2770), a complete set of the Mvh and HdrABC genes in the M. paludicola genome would be associated with the potential in vivo function of the heterodisulfide reductase/hydrogenase system. Meanwhile, the membrane-bound cytochrome-containing F420-nonreducing hydrogenase (Vht) co-operates with the HdrDE type heterodisulfide reductase. The homolog gene for HdrD, which contained the catalytic site for the reduction of the disulfide substrate , was found in the M. paludicola genome (MCP_0282). This gene conserved two CCG domains and two [4Fe-4S] clusters as observed in the HdrD gene of Methanosarcina barkeri , . Thus, it seems likely that the possible M. paludicola HdrD would have the same function as the Methanosarcinales HdrD. However, none of the homolog genes for the potential HdrE that contained a b-type cytochrome and interacted with electron donor was present in the M. paludicola genome. The absence of HdrE is common between the Methanocellales genomes of M. paludicola and RC-IMRE50. As the active site for heterodisulfide reductase activity is conserved in the HdrD, the heterodisulfide reductase/hydrogenase system (Vht/HdrD) may be operative in M. paludicola without the potential HdrE function. In Methanosarcina members, the heterodisulfide reductase/hydrogenase system (Vht/HdrDE) builds up an electrochemical gradient, then proton translocating ATP synthetase synthesizes ATP from the generated proton motive force . Phylogenetic analysis of the ATP synthetases in M. paludicola (MCP_0339) and RC-IMRE50 (RCIX2030) indicated that ATP synthetases of Methanocellales members more closely related to the ATP synthase of Methanosarcina mazei (MM_0780) (see Fig. S2 in the supporting information), which was experimentally confirmed as proton-translocating ATP synthase . Moreover, the alignment of ATP synthase showed that the gene of M. paludicola does not contain sodium-binding site that was characterized in Ilyobacter tartaricus  (see Table S2 in the supporting information). This sodium-binding site was not also found in the ATP synthase of M. mazei, while it was found in the ATP synthase of M. thermautotrophicus, which possess the experimentally characterized sodium-translating ATP synthase . Therefore it seems more likely that M. paludicola possess proton-translating ATP synthase. Moreover, Ech hydrogenase, which provides reducing equivalent of the first step of methanogenesis from H2 and CO2 in M. barkeri , is present in M. paludicola (MCP_0477 to 0482) as well as RC-IMRE50. Ech hydrogenase is also assumed to be proton-translocating . Therefore M. paludicola and Methanocellales members seem likely to use proton-translocating system. The Methanosarcina-like heterodisulfide reductase/hydrogenase system (Vht/HdrD) is more likely to operate in M. paludicola and Methanocellales members. Nevertheless further biochemical studies are necessary to clarify the in vivo function of these genes.
The lack of genes encoding carbon monoxide dehydrogenase/acetyl-CoA synthase (CODH/ACS) is a distinctive feature of the M. paludicola genome (Fig. 2). Those genes are present in RC-IMRE50 genome (LRC456–463). Thus the capability of acetate assimilation may be different between the two members of the Methanocellales methanogens. Since M. paludicola does not possess CODH/ACS, it metabolically loses a bypass between methyl-tetrahydromethanopterin (CH3-H4MPT) and acetyl-CoA and is not able both to assimilate CO2 as the sole carbon source and to utilize acetate as the methanogenic source. It was pointed out as a notable growth characteristic that M. paludicola required acetate for the growth . The genome harbors the genes for acetyl-CoA metabolisms, e.g. AMP forming acetyl-CoA synthases (MCP_0419 and MCP_0935) and ADP forming acetyl-CoA synthetases (MCP_0448 to 0449, and MCP_2918). As predicted in Methanosphaera stadtmanae , acetate is likely assimilated into acetyl-CoA from which many cell compounds could be synthesized. Indeed, tracer experiments using the 13C-labeled acetate or bicarbonate showed that both acetate and bicarbonate were incorporated into the growing cells of M. paludicola, while acetate was more abundantly assimilated than bicarbonate (see Table S3 for the supporting information). These results indicate that M. paludicola require acetate as the primary carbon source but also mixotrophically grows with acetate and CO2. The carbon dioxide serves not only as the absolute electron acceptor for methanogenesis (energy source) but also as the additional carbon source. Carbon dioxide might accessorily be incorporated by the following reactions, such as pyruvate ferredoxin oxidoreductase (Por) and ribulose-1,5-bisphosphate carboxylase-oxygenase that involve in pyruvate metabolism and AMP metabolism, respectively (Fig. 2).
It was noteworthy that most of the genetic components for tricarboxylic acid (TCA) cycle were absent in the M. paludicola genome (Fig. 2). The oxaloacetate can be generated from pyruvate by pyruvate carboxylase (MCP_0183 and 0184). However, as similar to the RC-IMRE50 genome, the only two enzymes, isocitrate dehydrogenase (MCP_0437) and fumarase (MCP_0508 and 0509) seemed to be encoded. Therefore M. paludicola is not able to synthesize 2-oxoglutarate, the precursor of glutamate. Generally, in many methanogens, the incomplete TCA cycle starts from oxaloacetate to proceeds in either reductive or oxidative direction leading to 2-oxoglutarate. It is suggested that the members of the order Methanosarcinales adopt the oxidative direction of the TCA cycle mediated by the enzymes of citrate synthase, aconitase, and isocitrate dehydrogenase . In contrast, the reductive pathway of TCA cycle would operate in the obligately hydrogenotrophic methanogens for the production of 2-oxoglutarate via malate, fumarate, and succinate . However, as described above, neither pathway was reconstructed from the M. paludicola genome. The reconstructed TCA cycle from the genome sequence let us to predict that M. paludicola may require some intermediates of TCA cycle or their derivatives such as 2-oxoglutarate and L-glutamate. Although the previous study described that yeast extract was not absolutely required for the growth of M. paludicola , it has recently become evident that the complete lack of supplementation of yeast extract finally results in the discontinuity of M. paludicola culture. The further growth experiments indicated that the supplementation of yeast extract was compensated with L-glutamate. Therefore, M. paludicola requires L-glutamate for their growth. The CDSs for amino acid transporter (MCP_2742 to 2744) were found in the genome. Due to their quite incomplete genetic sets of TCA cycle, M. paludicola and its relatives probably require at least L-glutamate for the growth as a donor of amino groups.
The absence of the TCA cycle in M. paludicola also indicates the inability to produce NADH in addition to amino acids. However, M. paludicola encodes pyruvate dehydrogenase (MCP_1717 to 1720), which catalyzes the NAD-linked oxidative decarboxylation of pyruvate concomitantly with the formation of acetyl-CoA (Fig. 2). Although the pyruvate dehydrogenase has usually been found in aerobe and facultative anaerobes, M. paludicola as well as RC-IMRE50 genomes encoded the enzyme. Since two genomes lack the genes for TCA cycle, pyruvate dehydrogenase might function as the alternative system for NADH production.
Nitrogen and Sulfur metabolism
Nitrogen fixation is catalyzed by a nitrogenase consisting of two components: dinitrogenase reductase called Fe protein encoded by nifH and dinitrogenase called MoFe protein encoded by nifDK . Although a full component of the genes for nitrogenase were found in RC-IMRE50, only homologues of nifH were found in M. paludicola (MCP_0364 and MCP_0905). The nifH genes are widely found in many methanogens regardless of their nitrogen fixation capability. It has been suggested that the nifH gene product of non-N2-fixing methanogen is not associated with the N2-fixing function but may be involved in the biosynthesis of factor 430 . In fact, the phylogenetic analysis showed that the nifH genes of M. paludicola clustered with the nifH genes from non-N2-fixing methanogens (see Fig. S3 in the supporting information). Since M. paludicola possess an ammonium transport system (MCP_0585–0588), it would take in ammonium from outside the cell. Meanwhile, one of the nifH gene of RC-IMRE50 was clustered with the nifH genes from nitrogen-fixing bacteria Clostridium acetobutylicum  (see Fig. S3 in the supporting information), and the nifH gene was part of an operon with other nif genes e.g. nifE, nifK. Therefore the nitrogen fixation capability might be an inter-species physiological difference among the member of the order Methanocellales.
Although many methanogens are recognized to be sensitive to sulfite , some methanogens not only tolerate the existence of sulfite but also utilize sulfite as a sole sulfur source , . To date, the reduction of sulfate to sulfide has been reported only for Methanothermococcus thermolithotrophicus . Nevertheless, the genetic and enzymatic components remain uncertain. The potential assimilatory sulfate reduction pathway from sulfate to sulfide was reconstructed in the M. paludicola genome (Fig. 2). The CDSs for adenylyltransferase (MCP_1659), adenylylsulfate kinase (MCP_1347), phosphoadenosine phosphosulfate reductase (MCP_0638 and MCP_1358) and sulfite reductase (MCP_1732) were identified. Meanwhile, we did not find the genes for a coenzyme F420-dependent sulfide reductase that was characterized in M. jannaschii . These genetic components for potential assimilatory sulfate reduction pathway shared both in the M. paludicola and RC-IMRE50 genomes. All other previously reported genomes of methanogens lack the genes encoding adenylyltransferase. Moreover, the adenylylsulfate kinase was encoded only in the genomes of M. paludicola and RC-IMRE50 and Methanosaeta thermophila. Although the sulfur detoxification and assimilation of the Methanocellales members should be experimentally characterized by the growth experiments, the genome-predicting sulfur metabolisms represent one of the distinctive features of the Methanocellales members.
Methanogens are usually characterized by the most severe anaerophilia of all the life in this planet. Thus, it has been believed that the ecological niches of methanogens are quite limited in the O2-abundant modern earth. However, a few species are known to be relatively resistant to the O2 exposure , . In such methanogens, possible antioxidant systems such as superoxide dismutase , catalase  and F420H2 oxidase  have been characterized. In addition, different kinds of antioxidant enzymes have been identified in bacteria and other archaea, e.g. superoxide reductase including neelaredoxin, desulfoferrodoxin and desulforedoxin, and peroxiredoxins including alkyl hydroperoxide and thiol-specific peroxidase . The candidate genes coding these antioxidant enzymes in the M. paludicola and other methanogens' genomes are listed (see Table S4 in the supporting information). The M. paludicola genome encoded superoxide reductase (MCP_0733) and rubredoxin (MCP_2757), which could catalyze the reduction of superoxide to hydrogen peroxide, and rubrerythrin (MCP_0328 and MCP_2368) and peroxiredoxins (MCP_0070, MCP_0461, MCP_0633 and MCP_2051) for detoxification of hydrogen peroxide to water. In addition, F420H2 oxidase (FprA) (MCP_2921) would be involved in O2 detoxification by catalyzing a four-electron reduction of O2 to H2O . Although, RC-IMRE50 genome additionally encodes the genes for superoxide dismutase, catalase and desulfoferrodoxin, none of the homologous genes were found in M. paludicola. The greater genetic complement for antioxidant enzymes of the RC-IMRE50 genome may point that the antioxidant system might be different among the Methanocellales members. Since superoxide dismutase and catalase are generally found in aerobes and facultative anaerobes, the absence of those genes in the M. paludicola genome indicates that M. paludicola would more prefer to inhabit strictly anaerobic condition, and needs the apparatus for monitoring oxygen concentration in their habitats. Actually, as similar to other methanogen genomes , M. paludicola possesses histidine kinase containing PAS domains that can be involved in sensing the redox condition changes . The domain architectures of histidine kinases of M. paludicola and RC-IMRE50 differ in pattern (see Table S5 in the supporting information). It also suggests that antioxidant system might be different between the two methanogens.
The whole-genome-level features of M. paludicola were characterized by comparative analyses with the closely related genotype (RC-IMRE50) and with the previously determined methanogens' genomes. The M. paludicola genome shared 70% of the CDSs with those of RC-IMRE50 genome. When a similar comparison was conducted in the genomes of different species within the genus Methanosarcina, the percentage was in the range of 62–73% (Data not shown). Thus, it seems likely that the genetic composition is well correlated with each other within the Methanocellales members as well as Methanosarcina species. Meanwhile, in the comparison with the genomes from other orders of methanogens, relatively high percentage was obtained from the comparisons with the members of the orders Methanosarcinales (37.3–45.6%) and Methanomicrobiales (35–42.7%), while the value with other orders was less than 33% (see Table S6 in the supporting information). These results strongly suggested the structural and compositional relatedness of the Methanocellales genomes with the Methanosarcinales and Methanomicrobiales genomes. Although the phylogenetic relatedness of the Methanocellales with the Methanosarcinales and Methanomicrobiales methanogens has been often pointed to by the phylogenetic analyses of rRNA and methyl-CoM reductase genes, the genomic compositional relatedness provides further insight into understanding the evolution of these modern lineages of methanogens.
In addition, the ordination of methanogens' genomes was generated using non-metric multidimensional scaling (NMDS) method (Fig. 3), which is a major branch of multivariate analysis. The NMDS ordinations attempt to place all samples in an arbitrary three-dimensional space, in which the relative distances between the samples indicate the corresponding pairwise similarity. Hence, the closely related organisms in the NMDS ordination would have the similar gene repertoires. In the NMDS map, the Methanocellales genomes made an independent group from other methanogen genomes, and were most closely related with the Methanomicrobiales genomes while the relatively closer relationship between the Methanocellales and the Methanosarcinales was also represented (Fig. 3). Since the highly conserved genetic elements are potentially relevant to the important physiological functions, the closer relationship in the NMDS map may represent the greater similarity in the conservative cellular functions and metabolisms, and the ecological niches. Thus, the close relationship of the Methanocellales genomes with the Methanomicrobiales and Methanosarcinales genomes in the NMDS map strongly suggests the general physiological relatedness of the Methanocellales to both Methanomicrobiales and Methanosarcinales methanogens. Interestingly, however, the Methanocellales genomes represented the unique position with height direction. The distinctiveness of the position is comparable to that of the Methanopyrus kandleri genome , of which the physiological uniqueness is characterized by the most hyperthermophily  and potentially the most ancient lineage of life . This result suggests that the Methanocellales methanogens would also have some of the exclusive characteristics from any other methanogenic groups. Insights from these genome-wide relationships and the molecular phylogenetic relationships among the Methanomicrobiales, Methanocellales and Methanosarcinales members will shed light on the evolution of these modern lineages of methanogens.
Figure 3. The ordination plot of genomes of methanogenic archaea using non-metric multidimensional scaling.
Distances were calculated from gene profiles of 18 representative archaeal genomes, based on protein families by the method described in Materials and Methods. The organisms are color-coded according to their affiliated orders: orange, Methanocellales; red, Methanosarcinales; green, Methanomicrobiales; blue, Methanobacteriales; brown, Methanococcales; purple, Methanopyrales. The organisms are represented with the numbers: 1, M. paludicola SANAE; 2, uncultured methanogenic archaeon RC-I, RC-IMR50; 3, M. a acetivorans C2A; 4, M. barkeri Fusaro; 5, M. mazei Go1; 6, M. burtonii ACE-M; 7, M. thermophila PT; 8, M. marisnigri; 9, M. palustris E1-9c; 10, M. boonei 6A8; 11, M. hungatei JF-1; 12, M. labreanum Z; 13, M. thermautotrophicus ΔH; 14, M. smithii PS; 15, M. stadtmanae MCB-3; 16, M. jannaschii JAL-1; 17, M. maripaludis S2; 18, M. kandleri AV19.doi:10.1371/journal.pone.0022898.g003
Materials and Methods
Methanocella paludicola strain SANAE
The detailed physiological description of M. paludicola strain SANAE (NBRC 101707, JCM 13418 and DSMZ 17711) was reported previously .
Genome sequencing, assembly, and gap closure
The genome of M. paludicola was sequenced using a conventional whole-genome shotgun strategy following the method of Takarada et al. . All plasmid clones were end-sequenced by using dye-terminator chemistry on an ABI Prism 3730 sequencer. Raw sequence data corresponding to approximately 10-fold coverage was assembled by using PHRED/PHARAP/CONSED software (http://www.pharap.org) . Gaps between the assembled sequences were primarily closed by primer walking on gap-spanning library clones or with PCR products from genomic DNA. Assessment of final assembly quality was completed as described previously .
Gene identification and annotation
Putative non-translated genes were identified by using Rfam , tRNAscan-SE , ARAGON  and SPRITSX . While, rRNA was identified by the BLASTN program . The potential protein sequences were predicted using a combination of GLIMMER  and GeneMarkS , and start sites were manually inspected and altered. These predicted CDSs were translated and were searched against the UniProt database  and the protein signature database, InterPro . The KEGG database  was used for pathway reconstruction. Signal peptides in proteins were predicted by using SIGNALP 3.0 , and transmembrane helices were predicted by using TMHMM . Horizontal gene transfer was predicted by using SIGI-HMM .
Incorporation of 13C-labeled acetate and bicarbonate into cell material
The cultivation for M. paludicola followed the procedure previously described . Basal medium contained 1 mM acetate and 30 mM NaHCO3. The 13C labeled acetate or bicarbonate was added to the culture at a final concentration of 5% (w/w) of non-labeled substrate. All cultivations were performed in the 120-ml serum vials containing 40 ml medium (pH 7.0 at 25°C) under an atmosphere of H2/CO2 (80/20 [v/v]). The temperature for cultivation was maintained at 37°C. Cells were corrected in the logarithmic growth phase with grass micro filter. The filtered cells were washed with 20 ml of 1.5% (w/v) NaCl solution containing 1 M HCl and were then washed with 20 ml 1.5% (w/v) NaCl solution. The filters were frozen and lyophilized. The incorporation of 13C-labeled acetate and bicarbonate in the lyophilized cells were examined by elemental-analysis-isotope-ratio-mass-spectroscopy(EA-IRMS). The stable isotope composition was determined by SI Science (Saitama, Japan) using isotope ratio mass spectrometer DELTA plus Advantage (Thermo Fisher Scientific).
Glutamate auxotrophy test
To check glutamate auxotrophy of M. paludicola, L-glutamate (0.5 mM), instead of yeast extract, was added to the medium. Growth was determined by monitoring the concentration of methane by using a GC-3200G gas chromatograph (GL Science) with a thermal conductivity detector. The measurements were performed in duplicate.
Comparative genomic analysis
For comparative genome analysis, following methanogenic archaeal genomes were used: Methanopyrus kandleri AV19 (AE009439), Methanocaldococcus jannaschii JAL-1 (L77117), Methanococcus maripaludis S2 (BX950229), Methanothermobacter thermautotrophicus ΔH (AE000666), Methanobrevibacter smithii PS (CP000678), Methanosphaera stadtmanae MCB-3 (CP000102), Methanocorpusculum labreanum Z (CP000559), Methanospirillum hungatei JF-1 (CP000254), Methanoculleus marisnigri JR1 (CP000562), Methanosphaerula palustris E1-9c (CP001338), Methanoregula boonei 6A8 (CP000780), Methanosarcina acetivorans C2A (AE010299), Methanosarcina barkeri Fusaro (CP000099), Methanosarcina mazei Go1 (AE008384), Methanococcoides burtonii ACE-M (CP000300), Methanosaeta thermophila PT (CP000477) and uncultured methanogenic archaeon RC-I, RC-IMRE50 (AM114193). Proteome for all analyzed methanogens was clustered by the FORCE program . In order to examine genome-wide relationship of M. paludicola in term of gene repertories, the ordination of archaeal genomes was generated using non-metric multidimensional scaling (NMDS) method, which is a one of the method of multivariate analysis. NMDS ordinations attempt to place all samples in an arbitrary three-dimensional space, in which the relative distances between the samples indicate the corresponding pairwise similarity. Therefore, the closely related organisms in the NMDS ordination would have the similar gene repertories. Gene context of each genome were constructed as described previously .
Nucleotide sequence accession number
The sequence and annotation of the complete M. paludicola genome is available at GeneBank/EMBL/DDBJ under accession numbers AP011532.
Genome plot of the orthologous gene pairs between the genomes of M. paludicola and RC-IMRE50. Pairwise ortholog families were identified with the InParanoid program (Remm et al., 2001, J. Mol. Biol., 314: 1041–1052). Orange circle indicates the area peripheral to the origin of the DNA replication.
Phylogenetic tree of Na+-translocating and proton-translocating A1A0-ATPases in methanogenic archaea. The neighbor-joining phylogenetic tree was constructed on the basis of a sequence alignment of A subunits of ATPases. The names of microbes with experimentally characterized ATPases are shown in colored red for Na+-dependent enzymes and blue for H+-dependent enzymes. The accession numbers are shown in parentheses after each sequence name. The scale bar indicates the estimated number of base changes per amino acid position. The numbers at internal branches indicate the bootstrap probabilities with 1,000 resampled data sets.
Phylogenetic tree of nifH-deduced amino acid sequences showing the phylogenetic position of M. paludicola (indicated by bold type). The tree was constructed by the neighbor-joining method using the ARB software package (Ludwig et al., 2004, Nucleic Acids Res., 32: 1363–1371). The accession numbers are shown in parentheses after each sequence name. The scale bar indicates the estimated number of base changes per amino acid position. The symbols at branch nodes indicate bootstrap values. Bootstrap analysis was performed with 1,000 resampled data sets.
Numbers of genes associated with the general COG functional categories.
Sequence alignment of c/K subunits of V type ATPases. This table was made based on the data of Mulkidjanian et al. . Active site residues are indicated in colored as follows: conserved ion-binding acidic (Glu/Asp) residue in red; other Na+ ligands are in right blue. The hydrophobic residue corresponding to Val63 of Ilyobacter tartaricus c subunit is colored orange. The conserved small (Pro, Gly, Ala, Ser) residue, corresponding to Pro28 of I. tartaricus c subunit is colored pink. Predicted cation specificity of the c/K subunit. Ions whose binding has been experimentally studied are shown in bold.
Carbon isotope fractionation of M. paludicola. Control indicates the value for the non- labeled cells.
Antioxidant enzymes among methanogens. The number indicates multiple gene copies. -, not present; n.d., not determined. Catalases indicated are E (katE, type I monofunctional clade II large subunit hemed catalase), A (katA, type I monofunctional clade III small subunit hemeb catalase), and G (katG, type II bifunctional hemeb catalase/peroxidase). Type of superoxide dismutases indicated are C (sodC, Cu-Zn-containing periplasmic enzyme) and B (sodB, Fe-containing cytoplasmic enzyme).
The number of genes of histidine kinase and response regulator of M. paludicola and RC-IMRE50.
Distribution of shared genes of M. paludicola and RC-IMRE50 with other methanogenic archaeal genomes. Asterisks show the value indicates the number and percentage of the genes of M. paludicola and RC-IMRE50 shared with each methanogen genome.
Conceived and designed the experiments: S. Sakai YT HI KT. Performed the experiments: S. Sakai MS HK YK RN. Analyzed the data: YT S. Sakai TT NI ET ATH AS KM. Contributed reagents/materials/analysis tools: YT S. Shimamura. Wrote the paper: S. Sakai YT HI KT. Coordination of the project: HI NF.
- 1. Sakai S, Imachi H, Sekiguchi Y, Ohashi A, Harada H, et al. (2007) Isolation of key methanogens for global methane emission from rice paddy fields: a novel isolate affiliated with the clone cluster Rice Cluster I. Appl Environ Microbiol 73: 4326–4331.
- 2. Conrad R, Erkel C, Liesack W (2006) Rice Cluster I methanogens, an important group of Archaea producing greenhouse gas in soil. Curr Opin Biotechnol 17: 262–267.
- 3. Garrity GM, Holt JG (2001) Phylum AII. Euryarchaeota phy. nov. In: Boone DR, Castenholz RW, Garrity GM, editors. Bergey's manual of systematic bacteriology. 2nd ed. New York: Springer. pp. 211–355.
- 4. Sakai S, Imachi H, Hanada S, Ohashi A, Harada H, et al. (2008) Methanocella paludicola gen. nov., sp. nov., a methane-producing archaeon, the first isolate of the lineage ‘Rice Cluster I’, and proposal of the new archaeal order Methanocellales ord. nov. Int J Syst Evol Microbiol 58: 929–936.
- 5. Boone DR, Whitman WB, Rouvière P (1993) Diversity and taxonomy of methanogens. In: Ferry JG, editor. Methanogenesis. New York: Chapman & Hall. pp. 35–80.
- 6. Liu Y, Whitman WB (2008) Metabolic, phylogenetic, and ecological diversity of the methanogenic archaea. Ann N Y Acad Sci 1125: 171–189.
- 7. Anderson I, Ulrich EL, Lupa B, Susanti D, Porat I, et al. (2009) Genomic characterization of Methanomicrobiales reveals three classes of methanogens. Plos One 4: e5797.
- 8. Bult CJ, White O, Olsen GJ, Zhou L, Fleischmann RD, et al. (1996) Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science 273: 1058–1073.
- 9. Maeder DL, Anderson I, Brettin TS, Bruce DC, Gilna P, et al. (2006) The Methanosarcina barkeri genome: comparative analysis with Methanosarcina acetivorans and Methanosarcina mazei reveals extensive rearrangement within methanosarcinal genomes. J Bacteriol 188: 7922–7931.
- 10. Slesarev AI, Mezhevaya KV, Makarova KS, Polushin NN, Shcherbinina OV, et al. (2002) The complete genome of hyperthermophile Methanopyrus kandleri AV19 and monophyly of archaeal methanogens. Proc Natl Acad Sci USA 99: 4644–4649.
- 11. Smith DR, Doucette-Stamm LA, Deloughery C, Lee H, Dubois J, et al. (1997) Complete genome sequence of Methanobacterium thermoautotrophicum ΔH: functional analysis and comparative genomics. J Bacteriol 179: 7135–7155.
- 12. Erkel C, Kube M, Reinhardt R, Liesack W (2006) Genome of Rice Cluster I archaea - the key methane producers in the rice rhizosphere. Science 313: 370–372.
- 13. Capaldi SA, Berger JM (2004) Biochemical characterization of Cdc6/Orc1 binding to the replication origin of the euryarchaeon Methanothermobacter thermoautotrophicus. Nucleic Acids Res 32: 4821–4832.
- 14. Shima S, Pilak O, Vogt S, Schick M, Stagni MS, et al. (2008) The crystal structure of [Fe]-hydrogenase reveals the geometry of the active site. Science 321: 572–575.
- 15. Thauer RK (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144: 2377–2406.
- 16. Ferry JG, Kastead KA (2007) Methanogenesis. In: Cavicchioli R, editor. Archaea: molecular and cellular biology. Washington DC: ASM press. pp. 288–314.
- 17. Setzke E, Hedderich R, Heiden S, Thauer RK (1994) H2: heterodisulfide oxidoreductase complex from Methanobacterium thermoautotrophicum. Composition and properties. Eur J Biochem 220: 139–148.
- 18. Deppenmeier U (2004) The membrane-bound electron transport system of Methanosarcina species. J Bioenerg Biomembr 36: 55–64.
- 19. Stojanowic A, Mander GJ, Duin EC, Hedderich R (2003) Physiological role of the F420-non-reducing hydrogenase (Mvh) from Methanothermobacter marburgensis. Arch Microbiol 180: 194–203.
- 20. Künkel A, Vaupel M, Heim S, Thauer RK, Hedderich R (1997) Heterodisulfide reductase from methanol-grown cells of Methanosarcina barkeri is not a flavoenzyme. Eur J Biochem 244: 226–234.
- 21. Hamann N, Mander GJ, Shokes JE, Scott RA, Bennati M, et al. (2007) A cysteine-rich CCG domain contains a novel [4Fe-4S] cluster binding motif as deduced from studies with subunit B of heterodisulfide reductase from Methanothermobacter marburgensis. Biochemistry 46: 12875–12885.
- 22. Thauer RK, Kaster A-K, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6: 579–591.
- 23. Pisa KY, Weidner C, Maischak H, Kavermann H, Müller V (2007) The coupling ion in the methanoarchaeal ATP synthases: H+ vs. Na+ in the A1A0 ATP synthase from the archaeon Methanosarcina mazei Gö1. FEMS Microbiol Lett 277: 56–63.
- 24. Mulkidjanian AY, Galperin MY, Makarova KS, Wolf YI, Koonin EV (2008) Evolutionary primacy of sodium bioenergetics. Biol Direct 3: 13.
- 25. Šmigáň P, Majerník A, Polák P, Hapala I, Greksák M (1995) The presence of H+ and Na+-translocating ATPases in Methanobacterium thermoautotrophicum and their possible function under alkaline conditions. FEBS Lett 371: 119–122.
- 26. Meuer J, Kuettner HC, Zhang JK, Hedderich R, Metcalf WW (2002) Genetic analysis of the archaeon Methanosarcina barkeri Fusaro reveals a central role for Ech hydrogenase and ferredoxin in methanogenesis and carbon fixation. Proc Natl Acad Sci USA 99: 5632–5637.
- 27. Fricke WF, Seedorf H, Henne A, Krüer M, Liesegang H, et al. (2006) The genome sequence of Methanosphaera stadtmanae reveals why this human intestinal archaeon is restricted to methanol and H2 for methane formation and ATP synthesis. J Bacteriol 188: 642–658.
- 28. Weimer PJ, Zeikus JG (1979) Acetate assimilation pathway of Methanosarcina barkeri. J Bacteriol 137: 332–339.
- 29. Shieh JS, Whitman WB (1987) Pathway of acetate assimilation in autotrophic and heterotrophic methanococci. J Bacteriol 169: 5327–5329.
- 30. Schindelin H, Kisker C, Schlessman JL, Howard JB, Rees DC (1997) Structure of ADP •AIF4−-stabilized nitrogenase complex and its implications for signal transduction. Nature 387: 370–376.
- 31. Staples CR, Lahiri S, Raymond J, Von Herbulis L, Mukhophadhyay B, et al. (2007) Expression and association of group IV nitrogenase NifD and NifH homologs in the non-nitrogen-fixing archaeon Methanocaldococcus jannaschii. J Bacteriol 189: 7392–7398.
- 32. Chen J-S, Toth J, Kasap M (2001) Nitrogen-fixation genes and nitrogenase activity in Clostridium acetobutylicum and Clostridium beijerinckii. J Ind Microbiol Biotechnol 27: 281–286.
- 33. Balderston WL, Payne WJ (1976) Inhibition of methanogenesis in salt marsh sediments and whole-cell suspensions of methanogenic bacteria by nitrogen oxides. Appl Environ Microbiol 32: 264–269.
- 34. Daniels L, Belay N, Rajagopal BS (1986) Assimilatory reduction of sulfate and sulfite by methanogenic bacteria. Appl Environ Microbiol 51: 703–709.
- 35. Rothe O, Thomm M (2000) A simplified method for the cultivation of extreme anaerobic Archaea based on the use of sodium sulfite as reducing agent. Extremophiles 4: 247–252.
- 36. Johnson EF, Mukhopadhyay B (2005) A new type of sulfite reductase, a novel coenzyme F420-dependent enzyme, from the methanarchaeon Methanocaldococcus jannaschii. J Biol Chem 280: 38776–38786.
- 37. Kiener A, Leisinger T (1983) Oxygen sensitivity of methanogenic bacteria. Syst Appl Microbiol 4: 305–312.
- 38. Zinder SH (1993) Physiological ecology of methanogens. In: Ferry JG, editor. Methanogenesis. New York: Chapman & Hall. pp. 128–206.
- 39. Takao M, Yasui A, Oikawa A (1991) Unique characteristics of superoxide dismutase of a strictly anaerobic archaebacterium Methanobacterium thermoautotrophicum. J Biol Chem 266: 14151–14154.
- 40. Shima S, Sordel-Klippert M, Brioukhanov A, Netrusov A, Linder D, et al. (2001) Characterization of a heme-dependent catalase from Methanobrevibacter arboriphilus. Appl Environ Microbiol 67: 3041–3045.
- 41. Seedorf H, Dreisbach A, Hedderich R, Shima S, Thauer RK (2004) F420H2 oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme F420-dependent enzyme involved in O2 detoxification. Arch Microbiol 182: 126–137.
- 42. Limauro D, Pedone E, Pirone L, Bartolucci S (2006) Identification and characterization of 1-Cys peroxiredoxin from Sulfolobus solfataricus and its involvement in the response to oxidative stress. FEBS J 273: 721–731.
- 43. Galperin MY (2005) A census of membrane-bound and intracellular signal transduction proteins in bacteria: bacterial IQ, extroverts and introverts. BMC Microbiol 5: 35.
- 44. Zhulin IB, Taylor BL, Dixon R (1997) PAS domain S-boxes in Archaea, Bacteria and sensors for oxygen and redox. Trends Biochem Sci 22: 331–333.
- 45. Takai K, Nakamura K, Toki T, Tsunogai U, Miyazaki M, et al. (2008) Cell proliferation at 122°C and isotopically heavy CH4 production by a hyperthermophilic methanogen under high-pressure cultivation. Proc Natl Acad Sci USA 105: 10949–10954.
- 46. Yu Z, Takai K, Slesarev A, Xue H, Wong J-T (2009) Search for primitive Methanopyrus based on genetic distance between Val- and Ile-tRNA synthetases. J Mol Evol 69: 386–394.
- 47. Takarada H, Sekine M, Kosugi H, Matsuo Y, Fujisawa T, et al. (2008) Complete genome sequence of the soil actinomycete Kocuria rhizophila. J Bacteriol 190: 4139–4146.
- 48. Ewing B, Hillier L, Wendl MC, Green P (1998) Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res 8: 175–185.
- 49. Griffiths-Jones S, Bateman A, Marshall M, Khanna A, Eddy SR (2003) Rfam: an RNA family database. Nucleic Acids Res 31: 439–441.
- 50. Schattner P, Brooks AN, Lowe TM (2005) The tRNAscan-SE, snoscan and snoGPS web servers for the detection of tRNAs and snoRNAs. Nucleic Acids Res 33: W686–689.
- 51. Laslett D, Canback B (2004) ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences. Nucleic Acids Res 32: 11–16.
- 52. Sugahara J, Yachie N, Arakawa K, Tomita M (2007) In silico screening of archaeal tRNA-encoding genes having multiple introns with bulge-helix-bulge splicing motifs. RNA 13: 671–681.
- 53. Altschul FS, Madden LT, Schäffer AA, Zhang J, Zhang Z, et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402.
- 54. Delcher AL, Harmon D, Kasif S, White O, Salzberg SL (1999) Improved microbial gene identification with GLIMMER. Nucleic Acids Res 27: 4636–4641.
- 55. Besemer J, Lomsadze A, Borodovsky M (2001) GeneMarkS: a self-training method for prediction of gene starts in microbial genomes. Implications for finding sequence motifs in regulatory regions. Nucleic Acids Res 29: 2607–2618.
- 56. Bairoch A, Apweiler R, Wu CH, Barker WC, Boeckmann B, et al. (2005) The universal protein resource (UniProt). Nucleic Acids Res 33: D154–159.
- 57. Mulder NJ, Apweiler R, Attwood TK, Bairoch A, Bateman A, et al. (2005) InterPro, progress and status in 2005. Nucleic Acids Res 33: D201–205.
- 58. Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori M (2004) The KEGG resource for deciphering the genome. Nucleic Acids Res 32: D277–280.
- 59. Bendtsen JD, Nielsen H, von Heijne G, Brunak S (2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340: 783–795.
- 60. Krogh A, Larsson B, von Heijne G, Sonnhammer EL (2001) Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305: 567–580.
- 61. Waack S, Keller O, Asper R, Brodag T, Damm C, et al. (2006) Score-based prediction of genomic islands in prokaryotic genomes using hidden Markov models. BMC Bioinformatics 7: 142.
- 62. Wittkop T, Baumbach J, Lobo FP, Rahmann S (2007) Large scale clustering of protein sequences with FORCE -A layout based heuristic for weighted cluster editing. BMC Bioinformatics 8: 396.
- 63. Takaki Y, Shimamura S, Nakagawa S, Fukuhara Y, Horikawa H, et al. (2010) Bacterial lifestyle in a deep-sea hydrothermal vent chimney revealed by the genome sequence of the thermophilic bacterium Deferribacter desulfuricans SSM1. DNA Res 17: 123–137.