Introduction
Living organisms are classified into three domains , namely archaea, eukarya and bacteria, although phylogenetic proximity among these domains has been the subject of debate in the research community. A molecular clock analysis has estimated that archaeal evolution originated between 3.8 to 2.5 Ga ago, and archaea are thought to have evolved at a slower rate than both bacteria and eukarya (
Feng et al., 1997;
John et al., 2003;
Dvornyk et al., 2003;
Hedges and Kumar, 2004;
Schopf, 2006). The genetic and molecular characteristics of the archaea domain are shared with bacteria and eukarya to originate as a separate kingdom (
Doolittle et al., 1996;
Huynen and Bork, 1998;
Hedges 2002). The distinct physiology and metabolic pathways of the archaea domain also distinguish this domain from other kingdoms, however they share a common ancestry. Hyperthermophilic anaerobes are most likely ancestors of archaebacteria (
Graham et al., 2000;
Daubin et al., 2002). Based on 16S and 18S rRNA sequence data, this domain is classified into three main types: the crenarchaeota, which are characterized by their ability to tolerate extremes in temperature and acidity; the euryarchaeota, which include methane-producers and halophiles; and the korarchaeota, a catch-all group for archaea which includes uncultured microbes from terrestrial hot springs (
Woese et al., 1978;
1990). Waters et al. (
2003) have recently isolated a novel archaebacteria,
Nanoarchaeum equitans, which belong to a new type, the nanoarchaeota.
Methanogens are the largest group of archaebacteria, and are characterized by their unique energy metabolism. Methanogenesis occurs in oxygen-free as well as low redox potential environments since methanogens are very strict anaerobes and have several novel groups of coenzymes which are not found in any other group of eubacteria (
Thauer and Bonacker, 1994;
Thauer, 1998;
Karthigeyan et al., 2007;
Thauer et al., 2008;
Chellapandi et al., 2009;
Chellapandi and Sivaramakrishnan, 2011). Methanogens are therefore extensively exploited for the production of methane from various feedstocks by anaerobic digestion processes (
Chellapandi, 2004;
Patel and Chellapandi, 2008;
Chellapandi et al., 2008;
2010a). Five orders of methanogens have been identified: Methanopyrales, Methanococcales, Methanobacteriales, Methanomicrobiales and Methanosarcinales (
Thauer et al., 2008). Although methanogens have most of the basic features of prokaryotes, they have some very typical characteristics that define their evolution. This review describes evolutionary aspects of methanogens based on comparisons of their biochemical and metabolic pathways with those of other closely related genomes.
Genomic standpoint of archaebacteria
The modern age of archaea began in 1996 with the whole-genome shotgun sequencing of the first archaeal genome,
M. jannaschii, which was completed by Bult et al. (
1996). Genome sequencing projects during the period from 2006 to 2009 identified additional novel genomes that may define completely new lineages in phylogenetic trees. Archaeal genomes ranged in size from 0.84 to 2.3 × 10
9 daltons and the chromosome is a single, circular DNA molecule. Archaea also contain plasmids that are smaller than bacterial chromosomes. The G+ C content of archaea varies between 21% to 68% shows a marked genotypic diversity. Functional genes are often organized into operon-like structures and 23S & 16S rRNA and tRNA genes contain introns (
Edgell and Doolittle, 1997). A crucial component of the prokaryotic world is the mobilome, the enormous collection of viruses, plasmids and other selfish elements which are in constant exchange with more stable chromosomes and serve as horizontal gene transfer (HGT) vehicles (
Koonin et al., 1997;
Deppenmeier et al., 2002;
Koonin and Wolf, 2008). A gene-centric perspective, therefore, describes the evolution of mobilome elements during prokaryotic evolution in terms of the gene as a distinct evolutionary unit that is subject to selection on its own and that can compete with other genes. The evolutionary impact of phage-related genes on the genomic and metabolomic complexities of methanogens have been studied (unpublished) and will provide new ideas concerning the genomic evolution of methanogens and their relatedness to foreign genetic elements of distantly related organisms. Genomic features of methanogens in different public domain databases, including coding genes, non-coding genes, protein families, metabolic pathways etc, are presented in Tables 1 and 2. These resources will help to provide for a more complete understanding of methanogens for constructing metabolic pathways and deriving novel evolutionary concepts.
Molecular phylogenetics of methanogens
The abundance of genomic information has defined a starting point for new insights into the multilevel organization of organisms and their metabolic evolution (
Forst and Schulten, 1999;
2001;
Heymans and Singh, 2003;
Chellapandi et al., 2010b;
Chellapandi and Kalaimathy, 2010;
Chellapandi and Dhivya, 2010;
Razia et al., 2010). Phylogenetic trees based on single genes represent evolutionary time, but the evolution of organisms is non-uniform due to HGT event. Figure 1 shows a whole genome-based phylogenetic tree of methanogens showing that all the groups are clustered and formed separate clades.
Pyrococcus furiosus DSM 3638 is used as an outgroup organism for the phylogenetic classification of methanogens.
Whole-genome-based evolutionary trees are less sensitive to the evolutionary history of genes than multi-gene analyses and are more representative of true relatedness than single-gene-based trees yet (
Bansal, 1999;
Brown et al., 2001;
Daubin et al., 2002). Phylogenic classification based on orthologous genes (
Fitz-Gibbon and House, 1999;
Snel et al., 1999;
Ma and Zeng, 2004) and protein content (
Tekaia et al., 1999;
Vedhagiri et al., 2009;
Chellapandi and Dhivya, 2010) is more reliable, but is limited to global similarity in gene and protein content. Estimating various aspects of gene content evolution, such as the size of ancestral genomes and the amount of gene duplication are, of course, not independent in organisms. Snel et al. (
2002) have described the evolution of gene content based on the presence and absence of genes in the archaea and proteobacteria and in other complete genomes, a process which has been shaped by gene loss, gene duplication, HGT, gene fusion/fission and gene genesis. Although the phylogenetic tree represents the evolution histories of organisms, it is based on the history of single genes and not necessarily that of the whole organism. Methanogens not only have common 16S rRNA sequences, but also share elements of metabolic pathways. As per earlier reports, 16S rRNA based phylogeny has provided some biological insight into the evolutionary process (
Vedhagiri et al., 2009;
Razia et al., 2011).
Molecular evolution of metabolic pathway contents
According to the Woese hypothesis, metabolic genes are anticipated to travel laterally, even during modern evolutionary times. Molecular embracing of metabolic genes as well as HGT of complete pathways between organisms can be inferred by phylogenetic analysis. Independent gene duplication has been suggested to be a plausible evolutionary process for initiating a metabolic pathway. Hence, a new tree based on metabolic pathway content reflects similarities in metabolic network repertoires related to the biology of the organisms (
Tatusov et al., 1997;
Hong et al., 2004). A clear evolutionary distinction between the informational and the metabolic facets of the cell has been indicated by genomic studies (
Olsen and Woese, 1997). Genes have been related to metabolic pathways, enabling determination of whether a pathway is present in a given organism, and making possible the estimation of the minimal set of genes necessary for life (
Koonin et al., 1997;
Tatusov et al., 1997;
Macario et al., 1999;
Downs, 2006). Phylogenetic proximity between methanogens and proteobacteria has recently revealed the sharing of metabolic functions by protein family divergence (
Karthigeyan et al., 2007;
Chellapandi and Sivaramakrishnan, 2011). A major fluctuation in protein secondary structures may also suggest an evolutionary basis for regeneration and degeneration of some metabolic modules in methanogens (
Chellapandi et al., 2009). Chellapandi and Sivaramakrishnan (
2011) suggested that the mosaic nature of some functional domains in protein superfamilies indicates the evolutionary relatedness of mesophilic methanogens with thermophilic archaea/halophilic archaea, which may result from selective pressure. Metabolic trees therefore represent an integrative approach for the comparison of the evolution of the metabolism with the evolution of the genome, helping to define new relationships in the tree of life (
Aguilar et al., 2004;
Nielsen and Oliver, 2005). Genomes can be compared in terms of functional capabilities, which indicate the relative abundance of all protein families or functional families across selected genomes, with the option of displaying this finformation either as a heat map or in a matrix format. Table 3 presents the current metabolic information of completely sequenced genomes of methanogens, which are available in KEGG, a metabolic pathway database(www.genome.jp/kegg/).
Genomes can also be compared using raw gene counts, or can be normalized in order to take into account the genome size. For Z-score normalization, a Z-score is computed for each protein/functional family x in a given genome: Zx = (x - meanx)/standard deviationx. Cluster of orthologous genes (COG) abundance of methanogens also highlights functional relationships and their evolutionary correspondence to closely related genomes, which are represented as heat maps in Fig. 2. Clicking on the cell in the heat map will retrieve the list of genes assigned to this particular family in this genome, while clicking on the identifier of the family displayed on right side of the column will add the corresponding family to the “Function Cart”, as available in IMG/M (integrated microbial genomes and metagenomes) (http://img.jgi.doe.gov/m). An Abundance Profile Search in IMG/M allows defining a profile for functions (COGs, Pfams) in a query genome in terms of their abundance compared to other related genomes. Abundance Profile tools can be used for comparing the functional capabilities of genomes. An overview of the relative abundance of protein families (COGs and Pfams) and functional families (Enzymes) across selected genomes is provided in the Abundance Profile Viewer. Abundance Profile Overview results can be resorted according to the abundance in other genomes by clicking on the corresponding column header.
Correlation matrices constructed according to existing COG entries and available metabolic gene information in COG can be useful to derive evolutionary insight into methanogens, and whether they are functionally related to closely-related species or distantly related organisms (Tables 4 and 5). Using a current literature survey and ongoing research activity, the following genetic information and metabolic facets have been comprehensively reviewed to given an overview of current knowledge on the evolutionary histories of methanogens.
Genetic information transfer system
The majority of genes involved in information transfer processes such DNA replication (
Edgell and Doolittle, 1997;
Ponomarev et al., 2003), and transcription & protein synthesis (
Koonin et al., 1997;
2000;
Apic et al., 2001;
Brochier et al., 2004;
2005;
Briones et al., 2005) show methanogens to be different from bacteria. The DNA replication machinery of methanogens appears to be most different since they lack the DNA polymerases and helicases typical of bacteria (
Edgell and Doolittle, 1997). The methanogen RNA polymerase is more complex, consisting of up to 14 subunits in which the core subunits (α, β and β') are similar to bacterial RNA polymerase, but several smaller subunits are not found in bacteria (
Brochier et al., 2005). Promoters have an A-T rich sequence -32 to -25 bp upstream of the transcriptional start site and a consensus sequence which resembles the eukaryotic TATA box. The translation machinery is generally quite similar between bacteria and archaea (
Olsen and Woese, 1997). Methanogens share amino acid homology with the histone proteins of eukarya (
Gaasterland and Ragan, 1998;
Brochier et al., 2005). Similar to bacteria, methanogens have a significant proportion of putative coding regions with no similarity to any sequence in any other organism and many corresponding proteins families have no known function (
Gaasterland and Ragan, 1998). The majority of encoded proteins are multi-domain in nature and have evolved through gene duplication, recombination and fusion (
Eisen, 1998;
Vothknecht and Tumbula, 1999;
Pagel, 1999;
Apic et al., 2001;
Koonin, 2005).
High quality phylogenetic classification of organisms has been based on properties computed from whole gene content (
Eisen, 1998) and or from relationships between genes of known and unknown functions (
Makarova and Koonin, 2003;
2007). Phylogenies based on the components of the replication (
Corbett and Berger, 2003;
Gadelle et al., 2003), translation (
Briones et al., 2005) and transcription machineries (
Brochier et al., 2005) strongly support the existence of a core of genes that have evolved mainly through vertical inheritance in methanogens. The common ancestor of archaea and bacteria possessed a RNA degradation complex; however, several protein subunits of methanogens have limited sequence similarity to the bacterial RNase holoenzyme (
Wilson et al., 2006).
Gene transfer (invading) system
Phage-related proteins are decisive for genomic and metabolomic complexities under selective pressures. Among methanogens, the M. acetivorans genome contains phage-related genes that may be responsible for the adsorption, integration, recombination, replication and transcription processes of invading phage genomes. Phylogenetic relatedness of these genes is found among members of the genera of Methanosarcina and Methanococcus, and enteric bacterial viruses (SP6, HK620, RB69, Felix 01, IF1 and HP2). Transposase, HNH endonuclease and terminase have phylogenetic proximity to distantly related bacteria, particularly pathogens and symbiotic bacteria. Phage-related proteins are functionally equivalent in the genomes of double stranded DNA viruses. It has been shown that phage-related genes in this genome can be acquired from distantly-related organisms by HGT and subsequently established as homologs in the present lineages before evolving into a new gene family (unpublished).
Methanogenesis
There are three pathways of biological methane production: the hydrogenotrophic pathway, the aceticlastic pathway and the methylotrophic pathway. The hydrogenotrophic pathway is found in all methanogens, while the other two pathways are limited to Methanosarcinales. The universal distribution of the hydrogenotrophic pathway suggests that hydrogenotrophic methanogenesis may be the ancestral form of biological methane production and that it may appear only once during evolution (
Bapteste et al., 2005). The ancestral state of the methanogenic pathway in Methanosarcina before the transfer event is likely attributable to one of the following two possibilities: (1) absence of an aceticlastic pathway and utilization of the numerous other methanogenic substrates available to Methanosarcinales (
Galagan et al., 2002); or (2) the presence of an acetoclastic pathway utilizing acetyl-CoA synthetase, which is still found in Methanosaetaceae.
Methanosarcina and Methanosaetaceae may have coevolved after the transfer event, optimizing growth according to different acetate concentrations (
Min and Zinder, 1989). Chellapandi and Sivaramakrishnan (
2011) have shown that energetic metabolisms of methanogens are shared ancestrally with primitive proteobacteria.
Reverse methanogenesis
Genes encoding proteins involved in the methyl reduction pathway of methanogenesis are homologous to those encoding oxidative tetrahydromethanopterin-methanofuran-dependent proteins in a methylotrophic bacterial species,
M. extorquens AM1 (
Vorholt et al., 2000;
Schmidt et al., 2010). Archaeal acetyl-CoA and carbon monoxide dehydrogenase are similar to subunits of the same enzymes found in the
Clostridium genus that can operate in the same synthetic direction under autotrophic methanogenesis (
Chistoserdova et al., 1998). Methyl-coenzyme M reductase and methyltransferase subunits show similarities to
Archeoglobus fulgidus and
Pyrococcus sp. sharing exclusively only one-third the number of signature clusters. The high proportion of unique genes in the methanogenic archaeal lineages is inconsistent with a random assortment mechanism, in which any gene can be transferred, lost or replaced. It suggests, rather, that the prediction of signatures supports the fetch phylogenetic conclusion that all methanogens are anciently diverged major lineages containing a substantial proportion of unique genes (
Graham et al., 2000).
The origins and evolutionary history of tetrahydromethanopterin-linked C
1 transfer reactions are still not well understood (
Hallam et al., 2003,
2004). The known phylogenetic diversity of these reactions has been expanded by the identification of genes highly divergent from those associated with cultivated Proteobacteria, Planctomycetes or Archaea (
Chistoserdova et al., 1998;
Kalyuzhnaya et al., 2005). Methyl coenzyme M reductase in methylotrophic methanogens implies the possibility of the existence of reverse methanogenesis (
Friedrich, 2005). Cellular levels of coenzyme F
420-dependent sulfite reductase are comparable to that of methyl-coenzyme M reductase, an enzyme essential for methanogenesis and a possible target for sulfite. This enzyme is found only in the hydrogenotrophic thermophilic methanogens,
M. kandleri and
M. thermautotrophicus, and is the likely ancestor of H
2F
420 dehydrogenases that serve as electron input units for membrane-based energy transduction systems of certain late evolving methanogens, including the dissimilatory sulfite reductases of bacteria and archaea (
Hartzell et al., 1985;
Thauer and Bonacker, 1994;
Nolling et al., 1996;
Thauer, 1998;
Johnson and Mukhopadhyay, 2005;
2007;
Thauer et al., 2008) and sulfate-reducing archaea (
Klein et al., 2001). Consequently, energy metabolism involving C
1 transfer reactions crosses the bacterial/archaeal boundaries for methylotrophy and methanogenesis. In addition, a close phylogenetic boundary between methanogens and methylotrophs is suggested by the reverse methanogenesis hypothesis. Structural flexibility at the substrate binding sites of each enzyme is thought to assign substrate-binding specificity in either the forward or reverse directions by inducing a specific conformational change (unpublished).
Sugar catabolic pathways
The conservation of glycolytic enzymes in methanogens is the result of convergent evolution from bacteria (
Verhees et al., 2003). Sugar catabolic pathways are absent in non-heterotrophic archaea, with the exception of glycogen-degrading methanogens. The independent emergence of aerobic respiration in euryarchaeotes is possibly due to the recruitment of bacterial genes by convergent evolution (
Brochier et al., 2005;
Kato et al., 2006). Unlike in the bacteria and eukarya, dihydrolipoamide dehydrogenase in methanogens is not part of a multimeric complex (
Brown and Doolittle, 1997). Fournier and Gogarten (
2008) defined the source of the transferred
ackA and
pta genes (cellulose metabolism) in the ancestor of
Methanosarcina, which may have been acquired from a cellulolytic clostridium belonging to a sister group of the
Clostridium thermocellum/
C. cellulolyticum clade by HGT.
Photophosphorylation system
Photophosphorylation is generally thought to have originated twice by independent events once within eubacteria and once within euryarchaeotes. A primary photosynthetic apparatus (carotenoids) is a primitive property of both groups (
Lake et al., 1985). The origin and early evolution of the photosynthetic electron transfer chain and its phosphorylation mechanism were energetic mechanisms of archaea (
Masinovsky et al., 1992). Genomic and phylogenetic relationships between archaeal and proteobacterial proteorhodopsins support its probable lateral transfer between planktonic bacteria and methanogenic archaea (
Beja et al., 2000;
Frigaard et al., 2006).
Heavy metal assimilation pathways
A well established assimilation pathway for cadmium, copper and nickel ions is represented in the genome of M. barkeri str. Fusaro. Ionic cobalt and tungsten transport and detoxification systems also exist in the genomes of M. maripaludies and M. mazei Go1, respectively. As a consequence of the existence of such heavy metal assimilation systems, metalloenzymes of methanogens are required to generate products that serve as substrates for methanogenesis and other energetic processes. Unlike bacteria, methanogenic archaea have unique metabolic process for metal transport and detoxification reactions, and membrane transporters functionally resemble to bacterial ABC transporters. Although few metalloenzymes are unique to closely related methanogens, several metabolic modules are shared among methanogens, thermophilic archaea, sulfur utilizing archaea and some members of the proteobacteria. It has been shown that heavy metals are coupled to energy-driven metabolism in methanogens (unpublished).
Endosymbiosis system
Methanogens are metabolically distant from free-living bacteria and are sometimes placed close to the metabolically highly specialized group of obligate bacterial pathogens (
Aguilar et al., 2004). van Hoek et al. (
2000) found an association between anaerobic heterotrichous ciliates and endosymbiotic methanogenic archaea. This association is dependent on the presence of hydrogenosomes that provide intracellular hydrogen as a substrate for methane formation. The common ancestors of all anaerobic heterotrichs coexist in methanogenic endosymbionts. The relative contribution of the genomic dynamics of vertically inherited genes increases with increasing amounts of HGT as the size of the ancestors decreases less drastically than gene loss with increased HGT (
Snel et al., 2002).
Transmembrane system
Archaeoglobus fuldigus and
M. jannaschii have substantial differences in transmembrane domains for their regulatory, transport, and sensory functions (
Kyrpides et al., 1996;
Klenk et al., 1997). The membrane compositions of memebers of the alcohol phosphatidyltransferase family, including phosphatidylserine synthase, phosphatidylglycerol synthase and phosphatidylinositol synthase of bacteria and eukarya phylogenetically resemble the archaetidylglycerol synthase and archaetidylinositol synthase (
Daiyasu et al., 2005). Archaea synthesize unusual solutes such as beta-amino acids, nepsilon-acetyl-beta-lysine, mannosylglycerate and di-myo-inositol phosphate and uptake glycine betaine. Studies of the molecular basis of osmoadaptation and its regulation in methanogens is still in its infancy, but genomics and functional genome analyses combined with classical biochemistry are shedding light on the processes that confer osmoadaptation in methanogens (
Müller et al., 2005;
Morii et al., 2009).
Summary
Since much of the genetic patrimony of the methanogens appears to be unique, many genomic studies are currently in progress to identify the functions of methanogen-specific genes and thereby gain a better understanding of cellular processes in methanogens. Additional genomic data-sets would help clarify whether some adaptations are dependent on specific amino acid sequences or structural contexts, and whether methanogens have unique amino acid sequences and protein folding mechanisms for adaptation to extreme environmental conditions. Moreover, comparative studies of protein superfamilies and metabolic pathways in methanogens are needed before the evolutionary histories of this group can be defined. Understanding metabolic behaviors of methanogenic archaea, with a special emphasis on evolution, would assist in their exploitation in industrial applications, particularly methane production and in bioremediation process. Hypothetical gene transfer and defense systems against invading foreign genetic elements of methanogens would be useful for genetic engineering research and the development of spoligotype arrays (
Chellapandi and Ranjani, 2011). Obviously, evolutionarily-directed enzyme design would support the development of synthetic enzymes with applications in biotransformation processes and green chemistry (
Chellapandi, 2011). Therefore, the molecular evolutionary hypothesis provides a new horizon for the growth of research advancing metabolomics and metabolic engineering in methanogens.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(1073KB)
