Genomic insights into a versatile deep-sea methanotroph constituting the rare biosphere of a Brazilian carbonate mound complex

ABSTRACT

Recent discoveries of aerobic methanotrophs in non-seep carbonate-rich environments in the deep sea suggest that these organisms may persist as part of the rare biosphere. Recovering rare, active methanotrophs through targeted culturing is essential for understanding their persistence under the oligotrophic non-seep conditions and for uncovering their genomic adaptations related to the survival in energy-limited ecosystems. In our study, using metagenomic analysis of enrichment cultures from the Alpha Crucis Carbonate Ridge, we discovered Methylotuvimicrobium crucis sp. nov., a novel methanotroph representing the rare biosphere in native sediments, described in accordance with the SeqCode rules. Recent discoveries of aerobic methanotrophs in non-seep carbonate-rich environments in the deep sea suggest that these organisms may persist as part of the rare biosphere. Recovering rare, active methanotrophs through targeted culturing is essential for understanding their persistence under the oligotrophic non-seep conditions, and for uncovering their genomic adaptations related to the survival in energy-limited ecosystems. In our study, using metagenomic analysis of enrichment cultures from the Alpha Crucis Carbonate Ridge, we discovered Methylotuvimicrobium crucis sp. nov., a novel methanotroph representing the rare biosphere in native sediments, described in accordance with the SeqCode rules. Phylogenomic analysis revealed <95% of Average Nucleotide Identity (ANI) to described species, with genomic evidence of deep-sea specialization including: (i) stress adaptation through cold-shock proteins (CspA) and DNA repair systems (UvrD/LexA), (ii) metabolic versatility via complete methane oxidation (pmoABC), nitrogen fixation (nifHDK), and sulfur cycling (sox/sqr) pathways, and (iii) niche partitioning through biofilm formation (GGDEF/EAL) and heavy metal resistance (CopZ/CzcD). Comparative genomics identified a 1,234-gene deep-sea core shared with Methylotuvimicrobium sp. wino1, enriched in mobile elements (TnpA, prophages) suggesting horizontal gene transfer drives adaptation. While undetected in situ amplicon surveys, Methylotuvimicrobium crucis exhibited enrichment under methane availability, demonstrating its role as a latent methane filter. These findings contribute to the understanding of the ecological significance of aerobic methanotrophs in deep-sea systems, revealing how rare microbial taxa with genomic plasticity have the potential to influence biogeochemical cycling in deep carbonate-rich environments.

IMPORTANCE

Microbial communities in deep-sea sediments play crucial roles in global biogeochemical cycles, yet they remain poorly characterized due to the challenges of sampling and culturing under extreme conditions. This study provides a comprehensive overview of microbial diversity and functional potential in carbonate-rich deep-sea sediments, with an emphasis on methane-oxidizing bacteria. By combining high-throughput metagenomics and comparative genomics, we reconstructed high-quality genomes from previously uncharacterized microbial consortia, including novel members of the genus Methylotuvimicrobium. Our findings shed light on the ecological strategies of methanotrophs in oxygen-limited environments and expand the genomic representation of key players in carbon cycling.

INTRODUCTION 

Deep-sea geomorphic structures, including carbonate mounds, pockmarks, and mud volcanoes, serve as archives of past or intermittent fluid and gas seepage, shaping distinct ecosystems in regions such as the Southwest Atlantic (1). These environments are characterized by dynamic physicochemical conditions, marked by fluctuating hydrocarbon fluxes and heterogeneous substrates, which drive microbial niche differentiation (2, 3). Notably, carbonate mounds (authigenic formations often linked to microbial activity) act as long-term carbon sinks while harboring microbial consortia critical to biogeochemical cycling (4). Despite their ecological significance, the microbial diversity and metabolic interactions within these systems, particularly in understudied regions like the Brazilian margin, remain poorly resolved (5–7). Central to these ecosystems are methylotrophic microorganisms, which utilize reduced one-carbon compounds (e.g., methane, methanol) to fuel carbon, nitrogen, and sulfur transformations (8, 9). Methanotrophs, in particular, mitigate methane emissions and influence carbonate precipitation, yet their distribution and activity in non-seep deep-sea sediments are enigmatic (10). While amplicon sequencing studies in the Alpha Crucis Carbonate Ridge (ACCR) and adjacent Santos Basin have detected taxa such as Methylomirabilota, these approaches often fail to resolve low-abundance methanotrophs, leaving their functional roles speculative (5, 7). For instance, Methylomirabilota populations in ACCR sediments lack canonical methane oxidation genes, suggesting metabolic plasticity or methodological limitations in detecting rare taxa (6, 7). This knowledge gap underscores the need to explore the “rare biosphere,” that is, microbial taxa present at low abundances but with potential outsized ecological impacts. Recent genomic advances reveal that rare microbes can harbor conserved adaptive traits, enabling persistence in oligotrophic environments (11–13). Such taxa may act as keystone species in carbonate-rich systems, maintaining metabolic functions during periods of resource scarcity. For example, reclassified genera like Methylotuvimicrobium (formerly Methylomicrobium) exhibit genomic adaptations to energy limitation, hinting at strategies for survival in different habitats (14). However, whether these taxa contribute to biogeochemical cycles without active seepage remains untested. Here, we hypothesize that (i) aerobic methanotrophs constitute a cryptic component of the rare biosphere in deep-sea sediments under non-seep conditions, evading detection by standard amplicon sequencing; (ii) these organisms possess a conserved core genome optimized for oligotrophy, enabling persistence in low-energy environments; and (iii) the rare biosphere’s metabolic versatility, particularly in deep-sea carbonate mound ecosystems, enables functional resilience, with aerobic methanotrophs sustaining carbon, nitrogen, and sulfur cycling through alternative metabolic pathways during periods of methane scarcity. To test this, we integrate culture-dependent enrichments, metataxonomics, and metagenomics of sediments cultures from the Santos Basin. By coupling these approaches, we aim to resolve the genomic and functional landscape of aerobic methanotrophs, assess their contributions to biogeochemical processes, and evaluate the ecological resilience of microbial consortia in deep-sea carbonate ecosystems.

MATERIALS AND METHODS

Study area and sample collection

Sediment sampling was conducted in the Santos Basin, located in the Southeast region of the Brazilian continental margin, occupying approximately 3.52 × 10⁵ km². This basin covers the continental margin of the Southwest Atlantic and is limited to the north by Alto de Cabo Frio and to the south by the Florianópolis Platform. Samples were collected in November 2019 on board the R/V Alpha Crucis of the Oceanographic Institute of the University of São Paulo (IO-USP) during the development of the project Biology and Geochemistry of Oil and Gas Seepages, SW Atlantic (BiOIL) (15). Sampling was developed in three distinct regions over the upper slope: (i) Area 1—the Tupana Carbonate Ridge (TCR), a 35 km-long lineament of carbonate mounds with occurrence of cold-water coral (16); (ii) Area 2—the ACCR, a ring-shaped carbonate ridge where there is evidence of seepage of hydrocarbons (4, 15); and (iii) Area 3—an extensive pockmark field (PF), originated from salt tectonics, with the occurrence of an exhumed salt diapir (1, 17) (Fig. 1). Sediment samples were collected in triplicates using a stainless steel box-corer (50 cm × 50 cm) (Table S1). The sediment cores were sliced into 0–15 cm layers with sterile spatulas and stored in evacuated Hungate tubes. The headspace was filled with argon, and samples were preserved at 5°C.

Fig 1.

Bathymetric map of the southwestern Santos Basin, offshore southeastern Brazil, highlighting key geomorphological features. (A) Regional bathymetric overview indicating the locations of the three study areas: Area 1 (TCR), Area 2 (ACCR), and Area 3 (PF). The inset map shows the location of the study area along the Brazilian margin. (B) Detailed bathymetric map from TCR in Area 1. (C) Bathymetric view of the ACCR in Area 2. Red dots represent sampling stations. (D) Bathymetric map of the PF in Area 3. Color scales represent water depth (A) and seafloor elevation (B–D), with depth contours and bathymetric gradients indicating topographic variability. Coordinates are in decimal degrees. The maps were generated using QGIS Geographic Information System software, version 3.32.3.

Cultivation of methanotrophic consortia

The 24 sediment samples were inoculated under aerobic and anaerobic conditions, yielding 48 enrichment cultures for methanotrophic cultivation. Aerobic and anaerobic cultures were performed in an nitrate mineral salts medium (NMS) culture medium with copper addition (18) in BOD TE-371 (Tecnal, Brazil), at 20°C. After autoclaving, the following were added: 10 mL, per liter of Phosphate Solution (26 g KH₂PO₄ and 62 g NaHPO₄.7H₂O in 1L of stock solution), 1.4 mg of CuCl₂ 2H₂O per liter of medium, and 10 mL of sediment extract prepared from the deep-sea sediment obtained. The headspace of the aerobic cultures (24 samples) was composed of methane:air (Whyte Martins, USA) in the proportion of 1:1, while the headspace in the anaerobic cultures (24 samples) was composed of methane:argon in the proportion of 1:1, totaling 48 cultures. During the suspension of laboratory activities in 2020 due to COVID-19, the cultures were kept in methane atmosphere, in an incubator BOD TE-371 (Tecnal, Brazil) at 5°C to reduce microbial metabolism and favor the preservation of the consortia. The incubation period totaled 9 months.

Methane oxidation rate estimation 

Methane present in the headspace was quantified using a gas chromatograph (GC6850, Agilent Technologies) equipped with a flame ionization detector and a 0.1 mL sample loop according to (19). Calibration curves were generated from serial dilutions of high-purity methane (99.5%, White Martins) in ambient air, ranging from 1:1 to 1:10. Both the conversion of chromatographic peak areas to nanomolar concentrations and the calculation of methane oxidation rates were performed according to Nakamura et al. (19). A fresh methane standard was run at the beginning of each measurement day to ensure accuracy. To minimize carryover and cross-contamination, the chromatographic column was flushed between analyses with one to three syringes of room air filtered through 0.2 µm membranes. When necessary, highly concentrated methane samples were diluted with 0.2 µm filtered argon gas. In both aerobic and anaerobic incubations, methane was introduced to the headspace of the serum bottles by replacing half of the gas volume (30 mL out of 60 mL total), resulting in an initial concentration of approximately 500,000 ppmv (50% vol/vol) CH₄. When converted to dissolved concentration under equilibrium conditions, this corresponds to about 4.44 × 10⁷ nmol·L⁻¹ (44.4 mM) of methane available at the start of the experiments. This value was applied equally to all consortia, with the difference being the balancing gas: filtered air in aerobic incubations and argon in anaerobic incubations. Methane consumption was monitored throughout the incubation, and measurements were continued until complete depletion of the initially added methane, which in our experiments occurred after approximately 1 week.

DNA extraction and sequencing of the 16S rRNA gene by Illumina MiSeq

Genomic DNA was extracted using the DNeasy PowerBiofilm Kit (MoBio, USA), following the manufacturer’s instructions. Approximately 15 mL of each culture was concentrated in a refrigerated centrifuge at 20°C (12,000 × g for 15 min), and the pellet was resuspended in lysis solution. The extracted DNA was quantified using the Qubit dsDNA HS Assay (Thermo Fisher Scientific, USA). The 16S ribosomal RNA (rRNA) gene was amplified using universal primers 515F-Y (5′-GTGYCAGCMGCCGCGGTAA-3′) and 926R (5′-CCGYCAATTYMTTTRAGTTT-3′) (20). PCR reactions (25 µL) contained 2 µL of template DNA, 0.2 µM of each primer, and 1× Ultra Mix 2× PCRBio (PCR Biosystems). Thermocycling conditions were initial denaturation at 95°C for 2 min, 30 cycles of 95°C for 45 s, 50°C for 45 s, and 68°C for 90 s, followed by a final extension at 68°C for 5 min. PCR products were purified using AMPure XP beads (Beckman Coulter) and subjected to a second PCR for Illumina adapter ligation, followed by another AMPure XP purification step. The final pooled libraries were quantified by quantitative PCR using the KAPA Library Quantification Kit for Illumina (Roche). Sequencing was performed on a NextSeq 2000 platform (Illumina) with 300 bp paired-end reads, at NGS Soluções Genômicas facility (Piracicaba, SP, Brazil). The sediment samples in supplementary data were extracted, sequenced, and analyzed in a previous study (5).

Bioinformatic analysis of the 16S rRNA gene

The 16S rRNA gene sequences were analyzed for their quality using the FastQC software. Bioinformatic analysis was performed using the QIIME2 software, v. 2020.2. Raw sequences were imported into QIIME 2 (v. 2020.2, https://docs.qiime2.org/) (5, 21) using the qiime tools import script via a manifest file. The sequences were summarized using the qiime demux summarize command, and the DADA2 software was used to obtain the observed Amplicon Sequence Variants (ASVs) (22). Through the quality analysis generated by FastQC, both the forward and reverse sequences were truncated at position 240, and the barcodes were removed, using the qiime dada2 denoise-paired script. The taxonomy was signed using the qiime feature-classifier script using the SILVA v.138.2 database (22, 23). Phylogenetic distances were calculated using the script qiime phylogeny align-to-tree-mafft-fasttree, using the MAFFT aligner (24).

The alpha and beta diversity metrics were calculated using the qiime diversity core-metrics-phylogenetic script that calculates core diversity metrics. In the phyloseq package (v. 3.6.3) (24, 25) of R v. 3.6.3 (R Development Team, 2018), the Simpson and Shannon indices were calculated and visualized via box plot. The non-parametric Kruskal-Wallis test was performed to determine whether there were significant differences in richness and evenness. Beta diversity was measured by the Bray-Curtis distance and visualized via non-metric multidimensional scaling (NMDS). Permutational multivariate analysis of variance (PERMANOVA) tests were performed to determine if there were differences in the dissimilarity of treatment applied to the cultures. To identify ASVs that are differentially abundant between treatments, the DESeq2 software was used, which performs differential expression analysis for sequence count data (26). The 16S rRNA gene sequencing data are available in the National Center for Biotechnology Information Sequence Read Archives under BioProject ID PRJNA1228598.

Metagenomic sequencing of microbial consortia

Based on the results of the 16S rRNA gene-based taxonomic profiling, microbial consortia in which methanotrophic microorganisms were detected were prioritized for metagenomic sequencing. Following this criterion, five consortia were selected, two from anaerobic cultures (_AN) and three from aerobic cultures (_A): 685B1A, 686B3A, 684B1AN, 688B3AN, and 683B3A. This strategy aimed to maximize the recovery of metagenome-assembled genomes (MAGs) from methanotrophic taxa of interest. The metagenomic libraries were prepared using the Illumina DNAPrep Kit (Illumina, San Diego, CA, USA) and sequencing was performed on the Illumina HiSeq platform (NextSeq 2 × 100 p) at NGS Soluções Genômicas facility (Piracicaba, SP, Brazil).

Recovering MAGs

The samples were kept separate for the assembly and recovery of the MAGs, to avoid cross the combination of contigs from different consortia in the same MAG. Analyses were performed on the Kbase online platform (https://www.kbase.us/) (27). After the cutting step using the Trimmomatic v. 0.36 software (28), the metaSPAdes v. 3.15.3 software (29) was used for metagenomic assembly with a minimum contig length of 2,000 bp. The binning step was performed using three software: MaxBin2 v2.2.4 (30), CONCOCT v1.1 (31), and MetaBAT2 v. 2.3.0 (32). The consensus binning was performed using DAS-Tool v.1.0.7 (33), integrating results from MaxBin2, MetaBAT2, and CONCOCT. The quality of the resulting bins was assessed using CheckM v. 1.1.3 (34). Bins were categorized according to the quality standards for MAGs proposed by Bowers et al. (35) as high-quality (completeness >90%, contamination <5%), medium-quality (completeness >50%, contamination <10%), and low-quality (completeness <50%, contamination <10%). All MAGs were taxonomically annotated using the GTDBTk v2.0.0 software (36). The annotation of contigs for the prediction of coding regions (coding DNA sequence [CDSs]) was performed using the RAST v1.073 (37) PATRIC (38) and PGAP (39), databases. The functional annotation of the MAGs was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database using data from the DRAM (Distilled and Refined Annotation of Metabolism) tool v. 1.2.4 (40). Genes related to the methane, nitrogen, and sulfur cycles were selected to be explored during the analyses.

Comparative genomics between MAGs of the genus Methylotuvimicrobium

Two of the recovered MAGs were taxonomically classified within methanotrophic bacterial lineages. These genomes were selected for comparative genomic analysis with four publicly available methanotrophic genomes retrieved from the GenBank database (Table S2).

The tools average nucleotide identity (ANI), average amino acid identity (AAI), and DNA-DNA in silico hybridization (DDH) were used to compare the MAGs and the reference genomes. The calculation of the ANI and AAI values was performed on the “Kostas Lab” website (http://enve-omics.ce.gatech.edu/g-matrix/), and the calculation of the DDH distance was performed on the genome-to-genome website (GGDC) (http://ggdc.dsmz.de/home.php). The pangenome analysis was performed with the anvi’o v.8 pipeline to compare reference genomes and MAGs, identifying the core region and singletons. The phylogenomic analysis was performed using the software Insert Genome Into SpeciesTree v2.2.0 (available in Kbase), which allows the construction of a species tree using a set of 49 universal core genes defined by families of Clusters of Orthologous Groups genes (41). Pangenome analysis was performed using the anvi’o v. 8 pipeline in order to conduct genome comparison and gene cluster identification. To determine the content of unique and shared genes among the genomes of methanotrophic microorganisms, clusters of orthologous genes were analyzed using the OrthoVenn3 program (42). The functional annotation of MAGs was performed using the KEGG database and the DRAM software (40) to search for complete metabolic pathways, comparison of genes of interest, and search for methanotroph genes.

RESULTS

Comparing aerobic and anaerobic cultures of methanotrophs

Aerobic cultures presented 133 exclusive ASVs, while anaerobic cultures presented 112 exclusive ASVs, and 133 were shared among all consortia (Table S1). The microbial composition was similar between treatments in which Proteobacteria (92.6%) and Bacteroidota (7.29%) were the most abundant phyla (Fig. S2). Other phyla such as Firmicutes, Actinobacterota, and Bdelovibriota accounted for less than 1% of the total abundance. The most abundant family was Alcavoracaceae, followed by Pseudoalteromonadaceae, Sphingomonadaceae, Pseudomonadaceae, and Marinobacteraceae. The Methylomonadaceae family was also identified among the 10 most abundant in enrichments (Fig. 2A). Through the analysis of Beta diversity using the Bray-Curtis metric (Fig. 2B), it was possible to identify that the samples are dissimilar regarding the treatment applied to the cultures. Dissimilarity was tested for statistical significance using a PERMANOVA (Betadisper P = 0.2788). The microbial composition of the aerobic cultures showed a significant difference (P = 0.001) compared to the anaerobic cultures, indicating that the applied treatment was responsible for generating a variation in the composition of the microbial communities.

Fig 2.

Methane-related genera composition, methane oxidation, and beta-diversity in enrichment cultures from deep-sea carbonate systems. (A) Relative abundance of methane-related genera (e.g., Methylobacterium, Methylophaga, Methylovulum, Methylomicrobium, and Methylomonas) in aerobic and anaerobic cultures from ACCR, TCR, and PF. (B) Methane concentration in the headspace (nmol·L⁻¹) over time in representative aerobic enrichment cultures (e.g., 685B1A and 686B3A), with incubation periods ranging from 24 to 168 h. (C) NMDS based on methane-related genera composition, showing the separation of aerobic and anaerobic samples (analysis of similarities P = 0.001; Betadisper P = 0.2788).

The enrichment cultures from the ACCR, TCR, and PF revealed three canonical aerobic methanotroph genera (Methylomicrobium, Methylomonas, and Methylovulum) alongside facultative methylotrophs (Methylobacterium and Methylophaga) (Fig. 2A). These taxa were undetected in original in situ sediment profiles (5), Fig. S3), supporting their classification as rare biosphere members that proliferate under selective cultivation. Taxonomic composition diverged sharply with oxygen availability: Methylobacterium dominated aerobic cultures, whereas anaerobic incubations favored Methylovulum and Methylophaga (Fig. 2B). NMDS ordination confirmed significant community dissimilarity (PERMANOVA, P = 0.001), with no dispersion bias (Betadisper, P = 0.2788), underscoring oxygen’s role in structuring methane-driven consortia.  Successful samples in the enrichment and recovery of MAGs from microorganism groups associated with methanotrophy refer to the collection sites 685 and 686, both located in the Alpha Crucis Carbonatic Ridge (ACCR) region. The Alpha Crucis carbonate mound was the region that presented the greatest diversity of genera of strict or facultative methanotrophic microorganisms (Fig. 1B) when compared to Tupana Carbonate Mound and the PF. It was only possible to recover MAGs from methanotrophic microorganisms (Methylotuvimicrobium) from the ACCR probably due to the formation of this area of carbonate mounds.

Metagenomic sequencing of consortia enriched with methanotrophic microorganisms and MAGs

After sequencing the five metagenomes, 355,540,109 reads were generated . The value of reads generated per sample can be seen in Table S3. The result obtained after assembly revealed that the number of contigs varied between 2,053 and 6,159, the percentage of GC content varied from 51.63% to 56.65%, and the values of L50 and N50 varied from 102 to 408 and 151,495 to 46,552, respectively. All parameters analyzed in the assembly evaluation are recorded in Table S4.

We obtained a total of 90 MAGs, 78 high-quality MAGs, eight medium-quality MAGs, and four low-quality MAGs. The taxonomic annotation revealed the presence of 18 families of the domain bacteria, distributed in three different phyla. Seventy-four MAGs were taxonomically assigned as belonging to the phylum Proteobacteria, followed by 14 from the Bacteroidota and one from the Actinobacteriota, distributed in the classes Gammaproteobacteria (46 MAGs), Alphaproteobacteria (27 MAGs), Bacteroidia (13 MAGs), and Actinobacteria (1 MAG) (Fig. S4). Additionally, two MAGs were taxonomically assigned to the genus Methylotuvimicrobium; both recovered from microbial consortia within the ACCR called LECOM 001 and LECOM 002.

Genomic analysis of MAGs from Methylotuvimicrobium sp. LECOM 001 and LECOM 002

Both MAGs showed a high completeness value, with 98.91% corresponding to LECOM 001 and 99.25% to LECOM 002 and low contamination values 2.53% and 2.5%, respectively. The recovered genomes are circular chromosomes of 5,020 Mb (LECOM 001) and 5027 Mb (LECOM 002), with a content of 48.31% and 48.2% of GC content, respectively (Table 1). These values were similar when compared with the GC content of other genomes of the same genus, such as 48.6% of Methylotuvimicrobium sp. wino1, and the reference strains, Methylotuvimicrobium buryatense strain 5 GB1C (48.7%) and Methylotuvimicrobium alcaliphilum str. 20Z (48.7%). Through the annotation, a total of 5,084 CDSs (or DNA coding regions) corresponded to Methylotuvimicrobium sp. LECOM 001 and 5,109 CDSs to LECOM 002. The annotated hypothetical proteins totaled 2,183 and 2,206, respectively (Table 1).

TABLE 1.

General features of the Methylotuvimicrobium LECOM 001, LECOM 002, wino1, M. buryatense, and M. alcaliphilum genomes^a

	Chromosome	Chromosome	Chromosome	Chromosome	Chromosome
Features
Taxonomy	M. crucis LECOM 001	M. crucis LECOM 002	Methylotuvimicrobium sp. wino1	M. buryatense strain 5GB1C	M. alcaliphilum str. 20Z
Number of contigs	100	112	1	1	2
Completeness	99.25%	98.91%	100%	100%	100%
Contamination	2.5%	2.53%	1.4%	0.6%	1%
GC content	48.2%	48.31%	48.60%	48.7%	48.7%
Estimated chromosome size (bp)	5,027,165	5,020,519	5,056,496	4,998,879	4,796,711
Protein coding genes (CDS)	5,109	5,084	5,003	5,015	4,794
Number of hypothetical proteins	2,206	2,183	2,091	2,187	1,954
				44	43
RNAs t	39	38	44	154	114
CRISPR repeaters	278	278	236	150	111
CRISPR spacers	276	276	232	1	2

^a This table presents a comparative overview of key genomic features of M. crucis LECOM 001 and LECOM 002, alongside three reference genomes from GenBank: Methylotuvimicrobium sp. wino1, M. buryatense 5GB1C, and M. alcaliphilum 20Z. Features include assembly statistics (number of contigs, estimated chromosome size), genome quality metrics (completeness, contamination), GC content, number of predicted protein-coding sequences (CDSs), number of hypothetical proteins, tRNA counts, and CRISPR system components (number of repeats and spacers).

The comparative analysis of overall genome-relatedness indices, including ANI, AAI, and DDH, showed that Methylotuvimicrobium sp. LECOM 001 and LECOM 002 correspond to the same species (Table 2). Following the current parameters for species delimitation (43), organisms considered to belong to the same microbial species share >95% mean ANI, >95% mean AAI, and >70% similarity (in silico) genome-to-genome hybridization (DDH).

TABLE 2.

Genomic similarity metrics (DDH, ANI, and AAI) between M. crucis strains LECOM 001 and LECOM 002 and reference genomes from GenBank^a

		M. crucis LECOM 001			M. crucis LECOM 002
Access Number	Reference genome	DDH (%)	ANI (%)	AAI (%)	DDH (%)	ANI (%)	AAI (%)
GCA_000968535.1	M. alcaliphilum 20Z	56.40	93.00	92.14	56.40	93.01	92.22
GCA_005931095.1	M. buryatense 5GB1C	52.30	91.40	91.34	52.30	91.43	91.32
NZ_CP024202.1	Methylotuvimicrobium sp. wino1	66.40	92.63	92.94	66.50	92.65	93.00
GCA_050611805.1	M. crucis LECOM 002	100.00	100.00	99.97	–b	–	–
GCA_050611825.1	M. crucis LECOM 001	–	–	–	100.00	100.00	99.97

^a Values of DNA–DNA in silico hybridization (DDH), average nucleotide identity (ANI), and average amino acid identity (AAI) are shown for pairwise comparisons between M. crucis strains LECOM 001 and LECOM 002 and selected methylotuvimicrobium reference genomes retrieved from GenBank.

^b “–” indicates comparisons that were not performed because they refer to the same strain.

The DDH, ANI, and AAI values indicate that the MAGs recovered in this study do not belong to any species of Methylotuvimicrobium available in GenBank. These results suggest that the MAGs found may represent new species of the genus, with genomic similarities to MAGs from marine environments but with unique characteristics.

Given these genomic distances, we propose that the MAGs LECOM 001 and LECOM 002 represent a novel species within the genus Methylotuvimicrobium, for which the name Methylotuvimicrobium crucis sp. nov. is proposed, according to SeqCode rules and recommendations (44). The specific epithet crucis refers to the Alpha Crucis carbonate mound, the site from which the genomes were obtained, and also to the Southern Cross constellation (Crux). Phylogenomic analyses support the monophyly of this taxon, which forms a well-supported clade with Methylotuvimicrobium sp. wino1, its closest known relative, while remaining genomically distinct based on all measured criteria (ANI, AII, and DDH). Phylogenomic analysis revealed that the MAGs of Methylotuvimicrobium sp. LECOM 001 and LECOM 002 formed a clade with 100% bootstrap support and three other MAGs from the same genus (Fig. 3C). A distinct deep-sea clade was also observed, consisting of MAGs recovered from carbonate mounds in the Santos Basin and a MAG retrieved from deep-sea sediments. This phylogenetic grouping reinforces the similarities between the MAGs from deep-sea environments, suggesting a potential common adaptation to these ecosystems.

Fig 3.

Comparative genomic and functional insights into deep-sea methylotuvimicrobium lineages. (A) Comparative genomics: pangenome analysis of Methylotuvimicrobium strains, including two newly recovered genomes from deep carbonate mounds (LECOM 001 and LECOM 002), compared with reference genomes. The core genome, shared by all strains, comprises 2,746 protein clusters, while a subset of 358 clusters defines the deep-sea core, exclusive to the deep-sea lineage. Additionally, 983 protein clusters are unique to a strain recovered from the ACCR. The inner ring includes metadata for each genome, including completeness, GC content, and total genome length. Singleton genes (unique to a single genome) are also indicated. (B) Metabolic profile: presence/absence matrix of key metabolic pathways across methylotuvimicrobium genomes, inferred from functional gene annotations. Pathways are grouped by metabolic categories, including methane oxidation, nitrogen and sulfur transformations, carbon fixation, amino acid biosynthesis, central carbon metabolism (e.g., TCA cycle), and stress response mechanisms. Colored dots represent the presence of genes or pathways, highlighting functional differentiation among deep-sea and non-deep-sea strains. (C) Deep-sea core gene functions: left panel shows a maximum-likelihood phylogenetic tree, highlighting the deep-sea clade (blue). The right panel presents the functional classification of the 358 protein clusters that compose the deep-sea core genome. Functional categories include transcriptional regulation, DNA modification, stress response, cell adhesion, metal ion binding, and horizontal gene transfer-related functions, underscoring potential adaptations to deep-sea environmental pressures.

Pangenome analysis comparing recovered MAGs with reference MAGs identified a core region among the five genomes belonging to the same bacterial genus, a region that includes 15,342 genes (Fig. 3A). The pangenome analysis of the genus Methylotuvimicrobium revealed a core set of genes essential for the survival and adaptation of these species. Key genes in the core genome include those associated with aerobic methane oxidation, carbon fixation, cellular structure, intracellular transport and signaling, and DNA processing genes such as endonucleases. Several general functions were identified, including intracellular trafficking, secretion, vesicle biogenesis, translation and ribosomal structure, cell cycle control, cell wall and membrane biogenesis, and the transport and metabolism of nucleotides, carbohydrates, lipids, amino acids, and cofactors, highlighting the genus’s metabolic complexity. Stress-related genes include cold-shock proteins from the cspA family and CheY-like domains, which are involved in cold responses, chemotaxis, and sporulation regulation. Bacterial motility is supported by genes involved in the biosynthesis and functionality of flagella and Type IV pili, including components such as fliH, motA, motB, pilZ, pilU, and flgC, which are essential for movement and environmental interaction. The pangenome also features a robust DNA repair system, with genes involved in SOS response (lexA), UV damage repair (lhrB), and proteins such as uvrD and mutS that participate in mismatch repair and recombination. Genetic mobility is evidenced by transposases, prophage elements, and plasmid maintenance systems such as vapI, emphasizing the genomic plasticity of the genus. Additionally, genes related to chemotaxis and signal transduction, such as cheW, cheA, and cheB, were identified, enabling these bacteria to sense and respond to environmental chemical gradients.

The marine core

In addition to the core region, we identified a deep-sea core (1,234 genes) containing regions shared only among deep-sea Methylotuvimicrobium, including the two recovered MAGs and the Methylotuvimicrobium sp. wino1 genome (Fig. 3A). Our findings suggest that Methylotuvimicrobium from deep-sea environments exhibit strong adaptive mechanisms through horizontal gene transfer (HGT) and biofilm formation. For instance, the genomes harbor genes associated with the mobilome, including prophages and transposons, contributing to genomic adaptability. Key transposases identified include the TnpA family, YbfD/YdcC-associated transposase, and inactivated tnpA derivatives. The plasmid stabilization protein parE was annotated, suggesting mechanisms for plasmid stability in challenging environments. Phage-related proteins, such as gpD (phage protein D), FI (phage tail sheath protein), and gpI (P2-related tail formation protein), indicate the presence of prophages or remnants, potentially influencing host fitness. Furthermore, proteins involved in pilus formation, such as PilF/PilW (Type IV pilus assembly proteins) and TadD (Flp pilus assembly protein), emphasize the role of pili in bacterial adhesion and biofilm formation. These findings suggest that Methylotuvimicrobium from deep-sea environments exhibit strong adaptive mechanisms through HGT and biofilm formation. We also observed proteins involved in DNA repair and maintenance, including RpnC (recombination-promoting DNA endonuclease) and Nfi (deoxyinosine 3′-endonuclease), essential for repairing oxidative DNA damage and maintaining genomic stability under the high-pressure, low-temperature conditions of the deep sea. Additional proteins, such as Ssb (single-stranded DNA-binding protein) and Mod (adenine-specific DNA methylase), suggest mechanisms for DNA stabilization and protection against environmental stressors.  Transcriptional regulation is crucial for managing oxidative stress and metabolic flexibility. Regulators such as lysR, csgD, and merR (soxR) control gene expression in response to environmental signals, while sigma factors like rpoE (σ24) and helix-turn-helix domain regulators (yiaG and copY) fine-tune metabolic pathways. Essential cofactors, such as molybdenum (via moaA, moaD, and moaE) and cobalamin (via btuB), are synthesized, supporting survival in nutrient-limited environments. Enzymes like hisF, hisH, and gadA contribute to nitrogen metabolism. For cellular trafficking and secretion, Methylotuvimicrobium utilizes the Type II secretory pathway, with components like gspA, exeA, and gspD/PulD. Flp pilus assembly proteins (TadD and TadG) and the Vgrg protein are involved in surface attachment and biofilm formation.

Functional annotation of MAGs

Functional annotation of genes allowed observing which genes are related to important biogeochemical cycles in the environments in which these microorganisms are present. Comparing the five MAGs of the genus Methylotuvimicrobium, we were able to observe that there are metabolic differences when comparing the deep-sea group (LECOM 001, LECOM 002, and wino1) with the group from other environments. The pangenome and the analysis of orthologous proteins made it possible to identify a core genome that is shared only among M. wino 1, M. LECOM 001, and M. LECOM 002 (Fig. 3A), revealing a highly versatile metabolic capacity in relation to C1 compound utilization, central metabolism, and nitrogen/sulfur cycling.

 For aerobic methanotrophy and ammonia oxidation, genes for particulate methane monooxygenase activity pmoA-amoA, pmoB-amoB, and pmoC-amoC are consistently present across all strains (M. crucis LECOM 001, M. crucis LECOM 002, Methylotuvimicrobium sp. wino1, M. buryatense strain 5GB1C, and M. alcaliphilum str. 20Z), indicating the widespread capacity of these strains to oxidize methane aerobically.

For methylotrophy, nearly the full mxa gene cluster (mxaFJG__ACKLD, except mxaI) is present across all strains, which encodes for calcium-dependent methanol dehydrogenase (MDH), showing a robust ability to metabolize methanol into formaldehyde. Notably, mxaC is missing in M. crucis LECOM 002, potentially indicating a functional divergence in MDH assembly or activity. Other related genes, such as mdh1 and xoxF, are also present in all strains, exhibiting different forms of MDH. The gene mxaI is only present in M. alcaliphilum str. 20Z, suggesting slight variations in methanol oxidation pathways.

Downstream processing of formaldehyde is supported by the presence of genes faE, hxlB, hxlA, AGXT, and gck, which mediate formaldehyde detoxification and assimilation via the ribulose monophosphate and serine pathways (serABC and glyA). The same consistency is observed in genes involved in formate to CO₂ conversion (fdoG, fdhF, fdwA, fdsD, and fdwB) and CO₂ reduction (fwdA, fmdA, fwdB, fmdB, fwdC, fmdC, and ftr) via the Wood-Ljungdahl pathway, highlighting the organism’s capacity to thrive in fluctuating redox conditions. Genes associated with methanofuran biosynthesis (mfnB, mfnE, mfnF, and mfnD) are present in all strains, which may reflect HGT or ancestral inheritance. While methanofuran is classically essential for methanogenesis, in aerobic methanotrophs, these genes could be repurposed to support C1 metabolism, facilitating the transfer of formyl groups and enabling the utilization of low-concentration methane or other C1 compounds in carbonate-rich, non-seep environments.

Core carbon metabolism is supported by complete tricarboxylic acid cycle (TCA) cycle components such as porA, porB, porC, and porG, along with genes involved in gluconeogenesis and the Embden-Meyerhof-Parnas pathway, including ackA, fbaA, eno, gpmI, mch, acs, fbp, pps, and ppsA. This suggests robust energy generation and carbon flux capacity across diverse substrates. Notably, porA, porB, porC, and porG are absent in M. buryatense strain 5GB1C and M. alcaliphilum str. 20Z, indicating potential differences in TCA cycle functionality or alternative metabolic pathways.

The genes nifHDK for nitrogen fixation are present in M. crucis LECOM 001, M. crucis LECOM 002, Methylotuvimicrobium sp. wino1, and M. alcaliphilum 20Z, indicating their capacity for nitrogen fixation, a key process for converting atmospheric nitrogen into a biologically usable form. However, these genes are absent in M. buryatense 5GB1C, suggesting that this strain may not be capable of nitrogen fixation under natural conditions and might rely on other nitrogen sources.

Genes for nitrite and nitrate assimilation and dissimilation were identified, including nasC, nasA, nrtA, nasF, cynA, NRT2, narK, and nrtP, which are present in all genomes, suggesting they can utilize nitrate and cyanate as nitrogen sources in limited ammonium availability. In addition, the presence of hao, nirB, and nirD across all strains supports the potential for nitrite reduction and hydroxylamine oxidation associated with dissimilatory nitrogen metabolism linked to energy conservation under low-oxygen conditions. Additionally, denitrification capability was detected by the presence of norB and norC. Finally, genes cynT and cynS involved in reaction of cyanate with bicarbonate to produce ammonia and CO₂ are present in all strains. These features confirm a conserved mechanism across all strains to support a flexible nitrogen metabolism suited for fluctuating redox and nitrogen-limited habitats for both nitrogen uptake and reduction.

Sulfur metabolism is similarly well represented across all strains. For assimilatory sulfate reduction, several key genes, including cysU, cysW, cysD, cysH, cysIJ, and cysE, are present in all genomes. These genes play crucial roles in the reduction of sulfate to sulfide. Regarding oxidation of reduced sulfur compounds, genes such as sqr and fccA are also present in all genomes. Sulfide oxidation is a critical process for converting hydrogen sulfide (H₂S) into elemental sulfur or sulfate, contributing significantly to sulfur cycling. Sulfur oxidation involves several genes, such as sox system genes (soxX, soxY, soxZ, soxA, soxB, soxC, and soxD) (45). Among these, soxX, soxA, and soxB are conserved across all strains. Additionally, the gene SELENBP1, associated with organic sulfur compound degradation, is present in all strains, suggesting that these bacteria can degrade organic sulfur compounds and an additional layer of redox adaptability.

DISCUSSION

Rare but resilient: the ecological role of aerobic methanotrophs in deep-sea carbonate systems

Our study provides evidence that aerobic methanotrophic bacteria, although underrepresented in standard amplicon sequencing, are important components of the rare biosphere in carbonate mounds and pockmark areas.  Although aerobic methanotrophs typically rely on oxygen as the terminal electron acceptor, their persistence and detection under anoxic incubation conditions can be explained by two non-exclusive mechanisms. Some facultative methylotrophs are metabolically versatile and capable of utilizing alternative electron acceptors, such as nitrate, when oxygen is absent. Additionally, small amounts of residual oxygen may have been present during the initial stages of incubation (“oxygen carryover”), allowing limited microaerophilic growth or maintenance of these taxa. These mechanisms could account for the detection of facultative methylotrophs in anaerobic enrichment cultures, even though oxygen was not deliberately supplied (46, 47). Only after using enrichment techniques was it possible to detect the presence of some methanotrophic groups, including Methylomicrobium, Methylomonas, and Methylovulum, which were not detected in the initial analyses. Although the 16S rRNA amplicon data annotated sequences as Methylomicrobium, the corresponding MAGs were classified as Methylotuvimicrobium. This is consistent with the limited taxonomic resolution of short amplicon regions compared to genome-based analyses, especially for recently reclassified or closely related genera (48, 49). This difference highlights the limitations presented by culture-independent methods in detecting low-abundance taxonomic groups and the importance of combining culture-dependent and culture-independent techniques to reveal specific microbial groups.

Furthermore, the results suggest that aerobic methanotrophic groups remain components of the rare biosphere, potentially contributing to nutrient cycling in the sediment even in the absence of methane seeps, but can respond rapidly to episodes of methane escape. Our study reinforces the relevance of studying regions such as carbonate mounds and pockmark areas to better understand the aerobic methanotrophic groups that make up the rare biosphere and their ecological role in these regions.

Genomic adaptations and metabolic versatility of deep-sea methanotrophs 

The conserved genomic content shared between the deep-sea Methylotuvimicrobium strains (LECOM 001, LECOM 002, and wino1) distinguishes these lineages from their shallow-water relatives, reflecting adaptations to high pressure, low temperature, and oligotrophic conditions. This genomic core underscores the metabolic versatility that enables these organisms to thrive in extreme environments. This metabolic versatility is supported by various genomic features that reflect specific adaptations to deep-sea conditions (50). The environment imposes severe physiological constraints, including high hydrostatic pressure, low nutrient availability, and oxidative stress (51–53). To endure these challenges, the Methylotuvimicrobium MAGs encode multiple stress-resistance mechanisms, including DNA repair systems (RpnC, Nfi, and UvrD) to counteract pressure-induced damage, cold-shock proteins (CspA) to stabilize protein function at low temperatures, and heavy metal resistance genes (CopZ and CzcD) to manage trace metal scarcity. The presence of sulfur assimilation pathways (e.g., cysJI and sqr) suggests an ability to respond to fluctuating redox conditions in sediment environments. Together, these features mirror those found in other piezophilic microbes and point to a conserved deep-sea survival toolkit that enables persistence in this extreme yet ecologically significant habitat. Another striking feature of the deep-sea Methylotuvimicrobium MAGs is the enrichment of mobile genetic elements (e.g., TnpA and ParE) and prophage remnants (gpD and FI), suggesting that HGT contributes significantly to their adaptation ability. The presence of Type IV pilus assembly proteins (PilF and TadD) and cyclic di-GMP signaling domains (GGDEF and EAL) further points to biofilm formation as a core survival strategy, likely facilitating attachment to sediment particles and enhancing access to limited nutrients (54, 55). These traits are consistent with those observed in other deep-sea microorganisms, where genetic exchange and surface adhesion are crucial for persistence in energy-limited environments (3). The genomic features of these Methylotuvimicrobium strains, spanning stress tolerance, metabolic flexibility, and ecological interactions, may reflect a finely tuned adaptation to the deep-sea niche. Their capacity for methane oxidation, nitrogen fixation, and sulfur metabolism positions them as potential keystone taxa in deep-sea biogeochemical cycles. Rather than transient or passive microbial community members, these organisms appear to be specialized residents shaped by genomic streamlining and functional redundancy. While their ecological significance is becoming clearer, important questions remain about the in situ regulation of these pathways and the extent to which they influence deep-sea productivity. Targeted approaches combining high-pressure cultivation with metatranscriptomics will be key to unlocking these dynamics and refining our understanding of their roles in global nutrient cycling. The genomic multifunctionality of M. crucis, spanning sulfur assimilation (sox and sqr), denitrification (norBC), and nitrogen fixation (nifHDK), challenges the paradigm of methanotrophs as methane-dependent specialists. Instead, their metabolic plasticity positions them as keystone taxa capable of sustaining elemental cycling in energy-limited carbonate mounds, even without active seepage. The ACCR’s geological setting, marked by inferred methane migration (4), suggests these methanotrophs act as latent methane filters, potentially mitigating emissions during episodic fluxes. Their nitrogen-fixing capacity mirrors diazotrophic methanotrophs in cold seeps (56), hinting at a role in priming nutrient-poor sediments for microbial colonization. The observed genomic adaptations in the M. crucis lineages such as biofilm formation (GGDEF/EAL domains), mobile elements, and stress-response genes potentially support resilience to oligotrophy and fluctuating methane availability. Collectively, the genomic features highlight the deep-sea core potential for methane oxidation, formaldehyde assimilation, full respiratory and fermentative carbon metabolism, flexible nitrogen and sulfur cycling, and biofilm formation as ecological adaptations for persistence and competitiveness in redox-variable systems among substrate pulses, which is critical for survival in dynamic deep-sea habitats.

Future studies should prioritize in situ activity assays, such as stable isotope probing of methane assimilation or nitrogen fixation rates, to resolve their contributions to carbon sequestration and microbial food webs. Additionally, quantifying their influence on carbonate precipitation could clarify linkages between methanotrophy and long-term carbon burial in non-seep systems.

Conclusions

This study uncovers the hidden ecological role of active deep-sea methanotrophs in the rare biosphere of SW Atlantic carbonate-rich sediments. By integrating cultivation with genomics, we propose M. crucis sp. nov., a novel obligate methanotrophic species represented by MAGs from the ACCR, the first methanotroph genomes reported for marine sediments in this underexplored region. Strikingly, these methyl-dependent taxa thrive in an area without active methane seeps, challenging paradigms about their survival in energy-limited habitats. Genomic evidence further reveals their potential for nitrogen fixation, suggesting a dual role in methane and methanol oxidations and nutrient cycling that could sustain microbial consortia in deep-sea carbonate ecosystems.

This discovery highlights a critical blind spot in marine microbial surveys: functionally specialized taxa like M. crucis evade detection in standard metabarcoding studies due to their low abundance yet may drive biogeochemical processes in cryptic niches. The presence of a putative new Methylotuvimicrobium species in SW Atlantic carbonates raises fundamental questions: Are transient methane pulses or alternative energy sources subsidizing these communities? Does their metabolic plasticity enable persistence during methane scarcity?

Future studies should target in situ activity assays to validate the proposed nitrogen fixation capability and methane metabolism under dynamic conditions. Additionally, systematic exploration of SW Atlantic cold seeps and carbonate-hosted methane reservoirs is urgently needed to resolve the mismatch between methanotroph occurrence and detectable methane fluxes. By bridging cultivation with omics, this work expands the genomic diversity of Methylotuvimicrobium. It also advocates for a unified framework to decode the ecological strategies of microbial “dark matter” in deep-sea sediments.

As the first genomic record of methanotrophs in SW Atlantic carbonates, M. crucis exemplifies the untapped functional diversity lurking in geologically complex marine regions, a frontier for understanding microbial resilience and its role in mediating Earth’s methane balance.

DATA AVAILABILITY

Raw amplicon-sequencing reads and MAGs recovered from metagenomic data are available on National Center for Biotechnology Information’s Read Archive under accession PRJNA1234549. The register of M. crucis sp. nov. can be found at the following link https://seqco.de/i:49843.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/msystems.01311-25.

Supplemental Material. msystems.01311-25-s0001.docx.

Tables S1 to S5; Fig. S1 to S4.

Table S6. msystems.01311-25-s0002.xlsx.

Methane consumption data.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.