Genomic Insights into Moderately Thermophilic Methanotrophs of the Genus Methylocaldum

Considering the increasing interest in understanding the biotic component of methane removal from our atmosphere, it becomes essential to study the physiological characteristics and genomic potential of methanotroph isolates, especially their traits allowing them to adapt to elevated growth temperatures. The genetic signatures of Methylocaldum species have been detected in many terrestrial and aquatic ecosystems. A small set of representatives of this genus has been isolated and maintained in culture. The genus is commonly described as moderately thermophilic, with the growth optimum reaching 50 °C for some strains. Here, we present a comparative analysis of genomes of three Methylocaldum strains—two terrestrial M. szegediense strains (O-12 and Norfolk) and one marine strain, Methylocaldum marinum (S8). The examination of the core genome inventory of this genus uncovers significant redundancy in primary metabolic pathways, including the machinery for methane oxidation (numerous copies of pmo genes) and methanol oxidation (duplications of mxaF, xoxF1-5 genes), three pathways for one-carbon (C1) assimilation, and two methods of carbon storage (glycogen and polyhydroxyalkanoates). We also investigate the genetics of melanin production pathways as a key feature of the genus.


Introduction
Microbial methane oxidation is a key process in the carbon cycle at local and global scales.Methane-utilizing bacteria (methanotrophs) inhabiting high-temperature ecosystems include members of the genera Methylococcus, Methylothermus, and Methylocaldum from the phylum Pseudomonadota (Proteobacteria) and the genus Methylacidiphilum in the phylum Verrucomicrobia [1][2][3].Thermophilic and thermotolerant species of the Methylocaldum genus (Gammaproteobacteria class in the family Methylococcales) are commonly identified as important members of the bacterial communities in soils from a range of geographical locations [4], including hot springs [5], landfill cover soils [6,7], tin-mining ponds [8], oil sands [9], flooded rice fields [10], and marine sediments [11].The genus Methylocaldum was introduced by Bodrossy et al. (1997) [5].The genus currently includes four species: M. szegediense, M. tepidum, M. gracile [5], and M. marinum [11].The isolate M. szegediense O-12 was obtained from a sample of cow manure from a farm near Pushchino in the Moscow region in Russia.The pure culture was obtained from an initial enrichment, followed by a serial dilution to extinction technique [12].M. szegediense O-12 grows at a temperature range of 37-59 • C, with optimal growth at 55 • C. M. szegediense Norfolk was isolated from biofilter soil enrichments from a landfill in Norfolk County Council, UK, and pmoC homologs.Genomes of Methylocaldum, Methylobacter, and Methylococcus available on IMG/MER were used to search for the range of copies of pmoC.The search was based on IMG annotations using the different available identifiers for pmoC: TIGR03078, KO10946, and gene product name.A table containing the Anvi'o v7 results for synteny and functional genomic comparison of the genomes can be found in Supplementary Table S1.
XoxF phylogeny.The alignment was generated using the multiple alignment program for amino acid or nucleotide sequences v7.511 [17,18], choosing the E-INS-i method [16][17][18].The tree topology was obtained after 100 bootstrap replications using the maximum likelihood method and general time reversible model [20] using MEGA X software v10.2.6 [19,21].The tree with the highest log likelihood (−175,992.24)[22] was used to generate the summarized version of it.
Fdh phylogeny.The alignment was generated using the multiple alignment program for amino acid or nucleotide sequences MAFFT v7.511 [17,18], choosing the E-INS-i method [16][17][18].The tree topology was obtained after 100 bootstrap replications using the maximum likelihood method and the Whelan and Goldman + Freq.model [30].The tree with the highest log likelihood (−50,331.00)(Supplementary Figure S1) was used to generate the cartoon version of it.The percentage of trees in which the associated taxa clustered together is shown next to the branches.This analysis of 193 amino acid sequences was conducted in MEGA X v10.2.6 [19,21].There were a total of 1638 positions in the final dataset.

Methylocaldum Phylogeny
The phylogeny of Methylocaldum strains based on available 16S rRNA sequences shows five clades (Figure 1A).Four of these clades correspond to currently characterized species within the Methylocaldum genera (Methylocaldum tepidum LK6 T , Methylocaldum szegediense OR2 T , Methylocaldum gracile VKM 14L T , and Methylocaldum marinum S8 T ), while the fifth clade formed by strains BFH1, dr65, and r6f does not have a type strain.A comparison based on the presence or absence of orthologous genes between the three strains analyzed in this study revealed that the M. szegediense O-12 and Norfolk strains share 90% and 93%, respectively, of their predicted protein sequences as orthologous, while only 47% of the predicted protein sequences in M. marinum S8 were orthologous to genes in the M. szegediense O-12 and Norfolk strains (Figure 1B).The average nucleotide identity times the alignment coverage (ANI x AC) percentage between the three strains confirmed that M. szegediense O-12 and Norfolk are members of the same species (≥90%), while M. marinum S8 shared only 38% with the M. szegediense strains (Figure 1B).

Synteny of Methane Monooxygenase (MMO) Gene Clusters
The identification of key genes necessary for methane oxidation in bacteria revealed the presence of several copies of the pmoC gene, which is usually part of the pmoCAB operon for the particulate methane monooxygenase (pMMO).Five copies were found in M. marinum S8, six were found in M. szegediense O-12, and five were found in M. szegediense Norfolk (Figure 2).In the M. marinum S8 and M. szegediense Norfolk strains, two pmoC genes formed part of the complete gene cluster for the synthesis of the pMMO, whereas only one pmoC gene formed part of the complete gene cluster in M. szegediense O-12 (Figure 2).All pmoC genes forming part of pMMO operons had high sequence identity to each other and were denominated as type 1 (purple ribbon, Figure 2).Furthermore, the three Methylocaldum genomes contained paralogs of pmoC as independent genetic components in different loci of their genomes, which have been previously named as stand-alone copies of pmoC [31].Except for a single stand-alone pmoC in M. szegediense O-12 assigned to type 1, the rest of the stand-alone copies of pmoC have different nucleotide sequences than the one forming the pMMO cluster (type 1) and form two separate groups of paralogs, denominated as type 2 and 5 (pink and magenta ribbons, respectively, Figure 2).Furthermore, both M. szegediense strains exhibit two stand-alone pmoC type 2 relatively next to each other.Additionally, two stand-alone pmoC without high homology to any other pmoC sequences within the three genomes were found in M. marinum S8 (type 4) and M. szegediense O-12 (type 3), shown in black-edged triangles in Figure 2.Only M. marinum S8 exhibits a complete gene cluster for the soluble methane monooxygenase (sMMO).When compared with other methanotrophs with genomes annotated on IMG/MER (Methylobacter [16] and Methylococcus [10]), Methylocaldum strains have between five and six coding sequences (CDS) assigned to pmoC, while Methylobacter strains have between one and two and Methylococcus have between two and three pmoC in the available genomes.

Synteny of Methane Monooxygenase (MMO) Gene Clusters
The identification of key genes necessary for methane oxidation in bacteria revealed the presence of several copies of the pmoC gene, which is usually part of the pmoCAB operon for the particulate methane monooxygenase (pMMO).Five copies were found in M. marinum S8, six were found in M. szegediense O-12, and five were found in M. szegediense Norfolk (Figure 2).In the M. marinum S8 and M. szegediense Norfolk strains, two pmoC genes formed part of the complete gene cluster for the synthesis of the pMMO, whereas only one pmoC gene formed part of the complete gene cluster in M. szegediense O-12 (Figure 2).All pmoC genes forming part of pMMO operons had high sequence identity to each other and were denominated as type 1 (purple ribbon, Figure 2).Furthermore, the three Methylocaldum genomes contained paralogs of pmoC as independent genetic components in different loci of their genomes, which have been previously named as stand-alone copies of pmoC [31].Except for a single stand-alone pmoC in M. szegediense O-12 assigned to type 1, the rest of the stand-alone copies of pmoC have different nucleotide sequences than the one forming the pMMO cluster (type 1) and form two separate groups of paralogs, denominated as type 2 and 5 (pink and magenta ribbons, respectively, Figure 2).Furthermore, both M. szegediense strains exhibit two stand-alone pmoC type 2 relatively next to each other.Additionally, two stand-alone pmoC without high homology to any other pmoC sequences within the three genomes were found in M. marinum S8 (type 4) and M. szegediense O-12 (type 3), shown in black-edged triangles in Figure 2.Only M. marinum S8 exhibits a complete gene cluster for the soluble methane monooxygenase (sMMO).When compared with other methanotrophs with genomes annotated on IMG/MER (Methylobacter [16] and Methylococcus [10]), Methylocaldum strains have between five and six coding sequences (CDS) assigned to pmoC, while Methylobacter strains have between one and two and Methylococcus have between two and three pmoC in the available genomes.Black-edge triangles indicate stand-alone copies of pmoC without high identity with any other pmoC within the three genomes.The gene cluster for soluble methane monooxygenase, sMMO, which is only found in M. marinum S8, is also shown.

Comparative Abundance CDS Assigned to COG Categories
According to the pangenome analysis of the three Methylocaldum strains, from a total of 14,284 CDS predicted for the three genomes, 2214 CDS formed orthologous groups (OG) of amino acid sequences based on a ≥70% threshold of combined homogeneity index (derived from geometrical and functional homology; Figure 3).Each genome had at least one gene copy as part of each OG assigned as shared among the three strains.Based on their relative abundances, the main shared COG categories (comprising >5% of the total gene calls that could be categorized) were genes related to categories C (8.58%) involved in energy production and conversion; E (7.36%) amino acid transport and metabolism; H (5.74%) coenzyme transport and metabolism; J (8.04%) translation, ribosomal structure and biogenesis; M (7.09%) cell wall/membrane biogenesis; O (6.23%) post-translational modification; P (6.32%) inorganic ion transport and metabolism; and R (5.42%) classified only as general function.
On the contrary, the CDS found to be unique to each genome, based on a low combined homogeneity index of their amino acidic sequences when compared to their counterparts in the other genomes, were mainly assigned to COG categories M, cell wall/membrane biogenesis (11.57%);X, mobilome including prophages and transposons (10.9%);T, signal transduction mechanisms (8.15%); P, inorganic ion transport and metabolism Black-edge triangles indicate stand-alone copies of pmoC without high identity with any other pmoC within the three genomes.The gene cluster for soluble methane monooxygenase, sMMO, which is only found in M. marinum S8, is also shown.

Comparative Abundance CDS Assigned to COG Categories
According to the pangenome analysis of the three Methylocaldum strains, from a total of 14,284 CDS predicted for the three genomes, 2214 CDS formed orthologous groups (OG) of amino acid sequences based on a ≥70% threshold of combined homogeneity index (derived from geometrical and functional homology; Figure 3).Each genome had at least one gene copy as part of each OG assigned as shared among the three strains.Based on their relative abundances, the main shared COG categories (comprising >5% of the total gene calls that could be categorized) were genes related to categories C (8.58%) involved in energy production and conversion; E (7.36%) amino acid transport and metabolism; H (5.74%) coenzyme transport and metabolism; J (8.04%) translation, ribosomal structure and biogenesis; M (7.09%) cell wall/membrane biogenesis; O (6.23%) post-translational modification; P (6.32%) inorganic ion transport and metabolism; and R (5.42%) classified only as general function.

Conjugation
Several predicted functions were only found in the two moderate thermophilic M. szegediense strains, including K12056, corresponding to the conjugal transfer mating pair stabilization protein TraG, which is one of the pilus assembly proteins in bacterial type IV secretion systems (T4SS) [32].While M. szegediense Norfolk has only two conjugal transfer proteins, TraG and TraN, the M. szegediense O-12 genome contains two TraG, as well as TraW, TraF, TraE, and TraL, which participate in pilus assembly; TraA (pilin); TraN and TraU, which are responsible for mating pair stabilization; TraV, TraK, and TraB, which form the core complex; TraD and TraC, which are ATPase proteins reported to be complexed in the cytosol and inner membrane; and also TrbL, which is a subset of proteins described as F-like T4SS [32,33].

O-Antigen Biosynthesis
Interestingly, two homologs of O-antigen biosynthesis protein RfbC (K20444; COG0438|COG1216) were found only in the M. szegediense strains.RfbC is predicted to catalyze the incorporation of sugars and their products to form the O-antigen polysaccharides in the lipopolysaccharides (LPS) [34].The enzyme's specificity toward rare sugars enables the production of unusual LPS, thus providing the remarkable diversity of cell envelop recognition patterns observed among bacteria [35][36][37][38][39]. RfbC is a member of the glycosyltransferase family 2 (GT2).A search for other glycosyltransferases belonging to that family revealed the presence of six other OGs shared between the M. szegediense strains and absent in M. marinum S8.Additionally, the three Methylocaldum strains share

Conjugation
Several predicted functions were only found in the two moderate thermophilic M. szegediense strains, including K12056, corresponding to the conjugal transfer mating pair stabilization protein TraG, which is one of the pilus assembly proteins in bacterial type IV secretion systems (T4SS) [32].While M. szegediense Norfolk has only two conjugal transfer proteins, TraG and TraN, the M. szegediense O-12 genome contains two TraG, as well as TraW, TraF, TraE, and TraL, which participate in pilus assembly; TraA (pilin); TraN and TraU, which are responsible for mating pair stabilization; TraV, TraK, and TraB, which form the core complex; TraD and TraC, which are ATPase proteins reported to be complexed in the cytosol and inner membrane; and also TrbL, which is a subset of proteins described as F-like T4SS [32,33].

O-Antigen Biosynthesis
Interestingly, two homologs of O-antigen biosynthesis protein RfbC (K20444; COG0438| COG1216) were found only in the M. szegediense strains.RfbC is predicted to catalyze the incorporation of sugars and their products to form the O-antigen polysaccharides in the lipopolysaccharides (LPS) [34].The enzyme's specificity toward rare sugars enables the production of unusual LPS, thus providing the remarkable diversity of cell envelop recognition patterns observed among bacteria [35][36][37][38][39]. RfbC is a member of the glycosyltransferase family 2 (GT2).A search for other glycosyltransferases belonging to that family revealed the presence of six other OGs shared between the M. szegediense strains and absent in M. marinum S8.Additionally, the three Methylocaldum strains share three OGs of GT2, and M. marinum S8 has four different GT2, which are not shared with the M. szegediense strains.Differences between the genetic toolkit for the biosynthesis of the O-antigen among Methylocaldum offer a starting point for future studies to elucidate the mechanisms of specialization in their different niches and how they can tolerate a wide range of temperatures.It has been shown that O-antigen structural differences can confer resistance to different types of stress, such as oxidative [40], mechanical [39], and osmotic stresses [41].Additionally, antigen molecular diversity is a current target of study to understand interdomain symbiosis [42].

Catalase
Among the distinct functions that were found to be unique for M. marinum S8 is catalase KatE (COG 0753, KEGG K03781), which can convert toxic hydrogen peroxide (H 2 O 2 ) into water and oxygen.H 2 O 2 is one of the three products of methanethiol (CH 3 SH) oxidation (HCOH, H 2 S, H 2 O 2 ), for which all three genomes have a homolog to the gene mtoX, a copper-dependent methanethiol oxidase (MTO) gene characterized in Hyphomicrobium sp.VS [43] and annotated in our analysis as K17285.CH 3 SH is a volatile organic sulfur compound that can co-occur with methane.It can be consumed by the verrucomicrobial methanotroph Methylacidiphilum fumariolicum SolV, which produces H 2 S that is subsequently oxidized for energy [44].It seems likely that this gene would be beneficial for methanotrophs since it could allow them to flourish in niches that release methane and CH 3 SH, an inhibitor of methane oxidation [45].

C1-Oxidation Pathways Are Highly Redundant
As described above, in addition to having two pmoCAB operons in the M. szegediense Norfolk and M. marinum S8 genomes and the sMMO operon in M. marinum S8, the three genomes also contain multiple stand-alone pmoCs with unknown functions.They also have multiple methanol dehydrogenases.The mxaFJGIRACKLD gene clusters in the three Methylocaldum strains encode the two subunits of methanol dehydrogenase mxaF (K14028) and mxaI (K14029) and the natural electron acceptor, cytochrome c L (mxaG).Moreover, the three strains share two OGs of additional methanol dehydrogenases, lanthanide-dependent MDHs, corresponding to xoxF types 5 and 3 according to the phylogenetic reconstructions (Figure 4) following previously xoxF type assignation [52][53][54].M. marinum has a third xoxF with no homology to genes in the M. szegediense strains that also belongs to xoxF type 3.The three Methylocaldum have the set of genes for the tetrahydromethanopterin (HMPT)-dependent pathway required for the oxidation of formaldehyde to formate.Gene redundancy was also found in this pathway.Based on the OG analysis, three different OGs for fae (K10713), which encodes the 5,6,7,8-tetrahydromethanopterin hydrolyase, were found among the Methylocaldum.One fae OG included three CDS from M. marinum S8, and two were included for M. szegediense O-12 and Norfolk.The other two OGs had only one CDS in each strain.Each strain had two methylene-tetrahydromethanopterin dehydrogenases mtdB (K10714) in separate OGs.Only one of the following genes was found in each of the strains: methylenetetrahydrofolate dehydrogenase mtdA (K00300), methenyltetrahydromethanopterin cyclohydrolase mch (K01499), formylmethanofuran-tetrahydromethanopterin N-formyltransferase ftr (K00672), formylmethanofuran dehydrogenase fwdABC (K00200, K00201, K00202), glycine hydroxymethyltransferase glyA (K00600), and coenzyme F420 hydrogenase subunit beta frhB (K00441), for which only M. marinum S8 had two CDS.
Interestingly, several formate dehydrogenases encoding genes (fdh) were found in the three strains.These enzymes catalyze the oxidation of formate to CO 2 , donating electrons to NAD + or cytochromes.The phylogenetic reconstruction based on the amino acid sequence of the major subunits (K00123) of the different formate dehydrogenases found in Methylocaldum revealed that they have four types of fdh (Figure 5 and Supplementary Figure S1).According to the previously described types of fdh in Methylobacterium extorquens AM1 [55,56], we detected the presence of three different fdh in the three Methylocaldum strains, corresponding to fdh types 2 and 3, including all their subunits.Additionally, FDH (K00122) was found in the three genomes as well.Moreover, M. marinum S8 has the major subunit for fdh type 1, although it was previously described to require the subunit B for functioning in the M. extorquens AM1 model [56,57].
The large rbcL/cbbL (K01601) and small rbcS/cbbS (K01602) subunits of the ribulose 1,5-bisphosphate carboxylase/oxygenase (RuBisCO) were found in the three strains, having a single gene difference between M. szegediense and M. marinum.This includes the presence of norQ and norD next to RuBisCO genes of the three Methylocaldum.M. marinum S8 also has pimeloyl-ACP methyl ester carboxylesterase between the RuBisCO and nor genes.The presence of rcb/cbb genes had been reported in other methanotrophs; however, its activity had not been demonstrated yet [58][59][60].
Similarly to other gammaproteobacterial methanotrophs, the regeneration of pentose phosphates from hexosephosphates requires fructose-bisphosphate aldolase, for which class I, ALDO (K01623), is present in all three Methylocaldum genomes.
As in other gammaproteobacterial methanotrophs, the Methylocaldum genomes contain some key genes for the serine pathway of formaldehyde assimilation.These include serine-glyoxylate transaminase, AGXT (K00830), which converts glyoxylate to hydroxypyruvate and glycine hydroxymethyltransferase, glyA (K00600), which uses 5,10methylenetetrahydrofolate and glycine as substrates to produce tetrahydrofolate and Lserine, which can continue to the carbon fixation pathway or sulfur or serine metabolism, respectively.Also, as part of the serine pathway, the three genomes encode malyl-CoA/(S)citramalyl-CoA lyase mcl (K08691), which produces acetyl-CoA and glyoxylate from Lmalyl-CoA as a substrate.However, all strains lack phosphoenolpyruvate carboxylase, ppc (K01595), which produces oxaloacetate from phosphoenolpyruvate, necessary for the serine pathway.Anaplerotic functions of the serine pathway have been previously proposed in Methylococcus capsulatus Bath.Like the strain Bath, all three genomes have the necessary genes to convert malyl-CoA (mcl, K08691) to glyoxylate, then to glycine (AGXT, K00830), and subsequently to serine (glyA, K00600) as part of the serine pathway.In the same pathway, they also have the genes to convert from D-glycerate to glycerate-2P (gck, K11529) to phosphoenolpyruvate (eno, K01689).

Nitrogen Metabolism
All three strains of Methylocaldum have the structural genes for nitrogenase (nif H (K02588), nif D (K02586), and nif K (K02591)) in an operon with nif E, nif N, nif X, and nif Q genes and also a nif -specific ferredoxin III (TIGR02936) (Figure 6B).Two additional homologs of nifH and nifK genes were also found in the three strains, although their genetic context was different from each other and may not be related to nitrogen fixation.The genetic elements required for nitrate assimilation were found in the three strains, having as a difference the additional presence of the nrtABC gene cluster (K15576, K15577, K15578) for the nitrate/nitrite transport system in M. marinum S8 only (Figure 6A).The three strains have the norCB gene cluster (K02305, K04561) for the reduction of nitric oxide (NO) to nitrous oxide (N 2 O); however, only M. szegediense O-12 and Norfolk had the nirK (K00368) for dissimilatory nitrite reductase for denitrification, and none of them had the alternative nirS (K15864) nitrite reductase.Similarly to other methanotrophic species, homologs of hydroxylamine dehydrogenase (K10535) responsible for NH 2 OH oxidation to NO 2 , as well as hydroxylamine reductase reducing NH 2 OH to form NH 4 + (K05601), are present in the three genomes.Additionally, the three strains have the genetic potential to assimilate NH 4 + through the glutamate cycle, with their genes for glutamine synthetase (GS) glnA (K01915) and glutamate synthase (GOGAT) gltBD (K00265, K00266) [58].The three strains have the genetic potential for the reversible conversion of glutamate to 2oxoglutarate/α-ketoglutarate (which is an intermediate of the Krebs cycle) and ammonia through their glutamate dehydrogenase GDH2 (K15371).Additionally, the three strains also have the gene for alanine dehydrogenase ald (K00259), which had been described to participate in the reductive amination of pyruvate in other methanotrophs, under high NH 4 + environments [58].The two M. szegediense strains also have the gene for another glutamate dehydrogenase gdhA (K00262), which has been demonstrated to be required for Streptococcus pneumoniae for adaptation to high temperatures (40 assimilate NH4 + through the glutamate cycle, with their genes for glutamine synthetase (GS) glnA (K01915) and glutamate synthase (GOGAT) gltBD (K00265, K00266) [58].The three strains have the genetic potential for the reversible conversion of glutamate to 2oxoglutarate/α-ketoglutarate (which is an intermediate of the Krebs cycle) and ammonia through their glutamate dehydrogenase GDH2 (K15371).Additionally, the three strains also have the gene for alanine dehydrogenase ald (K00259), which had been described to participate in the reductive amination of pyruvate in other methanotrophs, under high NH4 + environments [58].The two M. szegediense strains also have the gene for another glutamate dehydrogenase gdhA (K00262), which has been demonstrated to be required for Streptococcus pneumoniae for adaptation to high temperatures (40 °C) [61].

Carbon Storage
The carbon storage inventory includes glycogen biosynthesis glgABC as well as the genes for polyhydroxybutyrate biosynthesis (Supplementary Figure S2).The key genes for polyhydroxyalkanoate synthase phaC (K03821), as well as acetyl-CoA C-acetyltransferase phaA (K00626) and 3-oxoacyl-[acyl-carrier-protein] reductase phaB (K00023), were present in the three Methylocaldum strains.Furthermore, additional phaC homologs were found in the three strains, indicating possible PHB co-polymer biosynthesis.The ability to produce polyhydroxyalkanoates has been recently observed for axenic cultures of Methylocaldum, supporting the genetic observations [62].In addition to genes for polymer biosynthesis, all three strains contain genes encoding pathways of sucrose biosynthesis.

Pyomelanin/HGA-Melanin Proposed Production
A signature characteristic of Methylocaldum strains is the development of light to dark brown-colored colonies/culture [63].Previous analyses showed that M. szegediense O-12 synthesizes a tyrosine-derived melanin-like pigment upon a decrease in growth temperature (at a suboptimal temperature of 42 • C) [64].Because of the lack of the canonical melC1 for tyrosinase (K00505) for the production of eumelanin in Methylocaldum, here it is proposed that this color corresponds to a type of melanin, known as pyomelanin or HGA-melanin [65].The three Methylocaldum strains have genes encoding the tyrosine degradation pathway via homogentisic acid (HGA), hppD (K00457), and hmgA (K00451).In this pathway, the HGA intermediate can continue to be reincorporated into central metabolism as fumarate and acetoacetate [66] or accumulated and, through an spontaneous autoxidation process, converted to benzoquinone, which polymerizes to form pyomelanin [67,68] (Figure 7).The gene encoding histidinol-phosphateaminotransferase hisC (K00817), which produces the HGA precursor 4-hydroxyphenylpyruvate, has four CDS assigned to it in M. marinum S8, three CDS in M. szegediense O-12 and two CDS in Norfolk.
ferase phaA (K00626) and 3-oxoacyl-[acyl-carrier-protein] reductase phaB (K00023), were present in the three Methylocaldum strains.Furthermore, additional phaC homologs were found in the three strains, indicating possible PHB co-polymer biosynthesis.The ability to produce polyhydroxyalkanoates has been recently observed for axenic cultures of Methylocaldum, supporting the genetic observations [62].In addition to genes for polymer biosynthesis, all three strains contain genes encoding pathways of sucrose biosynthesis.

Pyomelanin/HGA-Melanin Proposed Production
A signature characteristic of Methylocaldum strains is the development of light to dark brown-colored colonies/culture [63].Previous analyses showed that M. szegediense O-12 synthesizes a tyrosine-derived melanin-like pigment upon a decrease in growth temperature (at a suboptimal temperature of 42 °C) [64].Because of the lack of the canonical melC1 for tyrosinase (K00505) for the production of eumelanin in Methylocaldum, here it is proposed that this color corresponds to a type of melanin, known as pyomelanin or HGAmelanin [65].The three Methylocaldum strains have genes encoding the tyrosine degradation pathway via homogentisic acid (HGA), hppD (K00457), and hmgA (K00451).In this pathway, the HGA intermediate can continue to be reincorporated into central metabolism as fumarate and acetoacetate [66] or accumulated and, through an spontaneous autoxidation process, converted to benzoquinone, which polymerizes to form pyomelanin [67,68] (Figure 7).The gene encoding histidinol-phosphateaminotransferase hisC (K00817), which produces the HGA precursor 4-hydroxyphenylpyruvate, has four CDS assigned to it in M. marinum S8, three CDS in M. szegediense O-12 and two CDS in Norfolk.

Discussion
Here, three genomes of thermophilic/thermotolerant methanotrophs of the genus Methylocaldum were analyzed.Most methanotrophs display metabolic plasticity in response to the availability of key nutrients (nitrogen, phosphates) or metals (copper, tungsten, lanthanides).However, all three members of the Methylocaldum genus are superior to other comparable methanotrophs in the number of paralogs for key enzymes as well as the number of metabolic pathways for C 1 -carbon conversions and storage.The comparison revealed several key findings regarding their genetic diversity and metabolic potential.
All three strains possess multiple gene clusters for particulate methane monooxygenase (pMMO), suggesting redundancy in methane oxidation capabilities.This redundancy may provide metabolic flexibility, allowing Methylocaldum strains to thrive in diverse environmental conditions.The analysis of the genes involved in nitrogen metabolism revealed the presence of nitrogen fixation genes in all three strains, indicating their potential ability to fix atmospheric nitrogen.Additionally, genes involved in nitrate assimilation and nitric oxide reduction are present, suggesting versatility in nitrogen metabolism.
The genome comparison of the three Methylocaldum strains revealed insights into their genetic composition and metabolic potential.M. szegediense O-12 and Norfolk exhibit similar genomic characteristics, including genome size, GC content, and predicted protein-coding genes.However, M. marinum S8 diverges significantly, with a larger genome size and a lower percentage of orthologous genes shared with the M. szegediense strains.Phylogenetic analysis based on 16S rRNA sequences and ANI x AC confirmed the close relationship between M. szegediense O-12 and Norfolk, while M. marinum S8 formed a distinct clade.Genome synteny analysis highlighted variations in methane monooxygenase gene clusters among the strains, suggesting potential differences in methane oxidation pathways.Comparative abundance analysis revealed shared and unique functional categories, with genes related to energy production, metabolism, and transport being predominant.In terms of the genetic toolkit for methane oxidation, the Methylocaldum strains display redundancy in key enzymes involved in methane metabolism, including multiple copies of pmoC genes and methanol dehydrogenases.Additionally, differences in the presence of formate dehydrogenase genes suggest potential variations in formate utilization strategies among the strains.The genomes also encode pathways for single-carbon assimilation, including the ribulose monophosphate (RuMP), serine, and Calvin-Benson-Bassham (CBB) cycles.Nitrogen metabolism genes indicate the capacity for nitrogen fixation and assimilation, with differences in nitrate transport systems among the strains.Carbon storage pathways involve genes for glycogen and polyhydroxybutyrate biosynthesis, suggesting adaptations for carbon storage under varying environmental conditions.Unique functions identified in M. marinum S8 include catalase, involved in the detoxification of hydrogen peroxide, and a nearly complete de novo cobalamin biosynthetic pathway, enabling independent synthesis of the essential cofactor.These genomic features may confer advantages for survival and adaptation in specific ecological niches.Furthermore, M. marinum S8 harbors additional hydrogenase subunits and genes related to molybdopterin biosynthesis, indicating potential metabolic capabilities for hydrogen metabolism and cofactor biosynthesis.The presence of putative stress response genes and heavy metal-binding proteins suggests adaptation to diverse environmental conditions.O-antigen biosynthesis genes unique to the M. szegediense strains offer insights into cell envelope diversity and potential adaptation mechanisms.These differences may contribute to niche specialization and stress resistance in different environments.Lastly, the proposed production of pyomelanin/HGA-melanin in Methylocaldum strains suggests a potential mechanism for pigment production and adaptation to suboptimal growth conditions.Genes involved in tyrosine degradation pathways indicate the capacity for melanin synthesis via homogentisic acid accumulation.The significance of preservation of orthologs in organisms inhabiting very different ecological niches remains to be elucidated.However, it is tempting to propose that the evolution of C1 genes was driven by an adaptation to tolerate dramatic changes in temperature.
In summary, genomic analysis of Methylocaldum strains reveals extensive genetic diversity and metabolic versatility, with implications for their ecological roles and environmental adaptation strategies.Further research is needed to elucidate the functional significance of these genomic features and their contributions to the ecological success of Methylocaldum in diverse habitats.

Figure 1 .
Figure 1.(A) Maximum likelihood tree representing the phylogenetic relationship of Methylocaldum based on 16S rRNA gene sequences.Leaf names with the strains analyzed in this article are in bold.Identifiers for each IMG gene ID are at the tip of each leaf.(B) Dendrogram showing the association between cluster of groups of orthologous genes, represented by thin vertical black lines, which, when grouped, form the rectangular blocks depicted on each genome.The clustering is based on the presence or absence of the orthologous genes on each genome.Average nucleotide identity (ANI) times the alignment coverage (AC) percentage of each genome when compared with one of the other Methylocaldum strains are shown inside of blue or light blue squares.

640695723 2 2890036157 7 2889962382 12 AF215633Figure 1 .
Figure 1.(A) Maximum likelihood tree representing the phylogenetic relationship of Methylocaldum based on 16S rRNA gene sequences.Leaf names with the strains analyzed in this article are in bold.Identifiers for each IMG gene ID are at the tip of each leaf.(B) Dendrogram showing the association between cluster of groups of orthologous genes, represented by thin vertical black lines, which, when grouped, form the rectangular blocks depicted on each genome.The clustering is based on the presence or absence of the orthologous genes on each genome.Average nucleotide identity (ANI) times the alignment coverage (AC) percentage of each genome when compared with one of the other Methylocaldum strains are shown inside of blue or light blue squares.

Figure 2 .
Figure 2. Synteny of methane monooxygenase (MMO) gene clusters in Methylocaldum genomes.Schematics for genomes of M. marinum S8, M. szegediense O-12, and M. szegediense Norfolk depicting the locus of MMO gene clusters.Ribbons connecting genomes show pmoC genes with high identity as part of pMMO gene cluster (purple) and as stand-alone genetic component (pink and magenta).Black-edge triangles indicate stand-alone copies of pmoC without high identity with any other pmoC within the three genomes.The gene cluster for soluble methane monooxygenase, sMMO, which is only found in M. marinum S8, is also shown.

Figure 2 .
Figure 2. Synteny of methane monooxygenase (MMO) gene clusters in Methylocaldum genomes.Schematics for genomes of M. marinum S8, M. szegediense O-12, and M. szegediense Norfolk depicting the locus of MMO gene clusters.Ribbons connecting genomes show pmoC genes with high identity as part of pMMO gene cluster (purple) and as stand-alone genetic component (pink and magenta).Black-edge triangles indicate stand-alone copies of pmoC without high identity with any other pmoC within the three genomes.The gene cluster for soluble methane monooxygenase, sMMO, which is only found in M. marinum S8, is also shown.

Figure 3 .
Figure 3. Comparative abundance of COG categories for Methylocaldum genomes.The relative percentage of each COG category is depicted by colored horizontal bars and bubbles according to the color-coded legend.Horizontal bars represent the percentage of coding sequences (CDS) assigned to each COG category among the three strains (shared) and the uniqueness of each strain (S8, O-12, and Norfolk).The total number of CDS considered for this comparison is at the bottom of each column.* indicates shared CDS; N/A indicates not applicable for the shared genes column.Bubble sizes and colors depict COG categories that represented >5% of the total when unassigned CDS were not considered.

AFigure 3 .
Figure 3. Comparative abundance of COG categories for Methylocaldum genomes.The relative percentage of each COG category is depicted by colored horizontal bars and bubbles according to the color-coded legend.Horizontal bars represent the percentage of coding sequences (CDS) assigned to each COG category among the three strains (shared) and the uniqueness of each strain (S8, O-12, and Norfolk).The total number of CDS considered for this comparison is at the bottom of each column.* indicates shared CDS; N/A indicates not applicable for the shared genes column.Bubble sizes and colors depict COG categories that represented >5% of the total when unassigned CDS were not considered.

Figure 4 .
Figure 4. Phylogenetic reconstruction based on all xoxF and mxaF genes found in the three Methylocaldum genomes analyzed (highlighted in bold).The corresponding IMG gene IDs are at the tip of each leaf.Collapsed clades did not contain any genes from the three analyzed genomes.

Figure 4 .
Figure 4. Phylogenetic reconstruction based on all xoxF and mxaF genes found in the three Methylocaldum genomes analyzed (highlighted in bold).The corresponding IMG gene IDs are at the tip of each leaf.Collapsed clades did not contain any genes from the three analyzed genomes.

Figure 5 . 12 2832925093Figure 5 .
Figure 5. (A) Maximum likelihood tree reconstructing the phylogenetic relationship between the amino acidic sequences of four different types of formate dehydrogenase major subunits (K00123).

Figure 6 .
Figure 6.Nitrogen metabolism genes in Methylocaldum strains.(A) Diagram representing the presence and absence of genes involved in nitrogen cycling.(B) Synteny of nitrogenase gene cluster present in the three Methylocaldum strains.Pairwise identity percentage is indicated inside ribbons connecting homologous genes.

Figure 6 .
Figure 6.Nitrogen metabolism genes in Methylocaldum strains.(A) Diagram representing the presence and absence of genes involved in nitrogen cycling.(B) Synteny of nitrogenase gene cluster present in the three Methylocaldum strains.Pairwise identity percentage is indicated inside ribbons connecting homologous genes.

Figure 7 .
Figure 7. Proposed pyomelanin/HGA-melanin synthesis pathway via accumulation of homogentisic acid (HGA) in Methylocaldum.(A) Schematic representation of tyrosine degradation pathway having HGA as intermediate synthetized by 4-hydroxyphenylpyruvate dioxygenase (hppD, HPD, K00457), which can be degraded via homogentisate 1,2-dioxygenase (hmgA, HMG, K00451) to be reincorporated into central metabolism as fumarate and acetoacetate or accumulated and converted to pyomelanin/HGA-melanin via polymerization of the monomer benzoquinone acetate.(B) Synteny of gene cluster where the hppD gene (red-edge arrows) was present in the three Methylocaldum strains.Pairwise identity percentage is indicated inside ribbons connecting homologous genes.

szegediense Norfolk 2508832326 Methylocaldum szegediense O-12 2832925093 Methylocaldum marinum S8 2832924041 Methylocaldum marinum S8
Figure 4. Phylogenetic reconstruction based on all xoxF and mxaF genes found in the thre Methylocaldum genomes analyzed (highlighted in bold).The corresponding IMG gene IDs are at th tip of each leaf.Collapsed clades did not contain any genes from the three analyzed genomes.8069803325Methylocaldum