Abstract
The lack of cultured isolates and microbial genomes from the deep seabed means that very little is known about the ecology of this vast habitat. Here, we investigated energy and carbon acquisition strategies of microbial communities from three deep seabed petroleum seeps (3 km water depth) in the Eastern Gulf of Mexico. Shotgun metagenomic analysis revealed that each sediment harbored diverse communities of chemoheterotrophs and chemolithotrophs. We recovered 82 metagenome-assembled genomes affiliated with 21 different archaeal and bacterial phyla. Multiple genomes encoded enzymes for acetogenic fermentation of aliphatic and aromatic compounds, specifically those of candidate phyla Aerophobetes, Aminicenantes, TA06 and Bathyarchaeota. Microbial interactions in these communities are predicted to be driven by acetate and molecular hydrogen, as indicated by a high abundance of fermentation, acetogenesis, and hydrogen utilization pathways. These findings are supported by sediment geochemistry, metabolomics and thermodynamic modelling of hydrocarbon degradation. Overall, we infer that deep-sea sediments experiencing thermogenic hydrocarbon inputs harbor phylogenetically and functionally diverse communities potentially sustained through anaerobic hydrocarbon, acetate and hydrogen metabolism.
Deep-sea sediments, generally understood to be those occurring in water depths greater than ~500 meters, represent one of the largest habitats on Earth. In recent years, culture-independent 16S rRNA gene surveys and metagenomic studies have revealed these sediments host a vast abundance and diversity of bacteria and archaea 1-8. Cell numbers decrease with sediment depth and age, from between 106 and 1010 cm-3 in the upper cm at the sediment-water interface to below 104 cm-3 several kilometers below the ocean floor 9,10. However, due to a lack of cultured representatives and genomes recovered from deep-sea sediments, it remains largely unresolved how microorganisms survive and function in these nutrient-limited ecosystems. Energy and carbon sources are essential requirements that allow the buried microorganisms to persist. With sunlight penetration not reaching the deep seabed, photosynthetic processes do not directly support these communities 11. It has therefore been proposed that deep sea benthic and subsurface microbes are primarily sustained by complex detrital organic matter, including carbohydrates, proteinaceous compounds, and humic substances, derived from the overlying water column via sedimentation 11-13.
Another important potential carbon and energy source in deep-sea sediments are petroleum geofluids that migrate from subseafloor reservoirs up to the seafloor 14. Petroleum compounds include smaller gaseous molecules, such as methane, propane and butane, and larger aliphatic and aromatic liquids. Numerous studies have investigated the role of methane oxidation in seabed sediments, which is mediated by anaerobic methanotrophic archaea (ANME), generally in syntrophy with bacteria respiring sulfate or other electron acceptors 4,6,8,15,16. In contrast, little is known about the degradation of larger alkanes or aromatic compounds by deep seabed microorganisms. Vigneron et al. 2 performed a comparative gene-centric study of hydrocarbon and methane seeps of the Gulf of Mexico, and suggested that microorganisms in deep cold seeps (water depth ~1 km) can potentially utilize a range of non-methane hydrocarbons. However, due to the absence of metagenome binning in that study, relevant metabolic functions were not assigned to specific pathways or taxa.
In addition to organic carbon compounds, microbial life in deep-sea sediments is also supported by inorganic electron donors. Some microorganisms have been isolated from deep sediments that are able to sustain themselves by oxidizing elemental sulfur, hydrogen sulfide, carbon monoxide, ammonia and molecular hydrogen (H2) 6,8,11. Of these, H2 is a particularly important energy source given its production in large quantities by biological and geochemical processes. H2 can be generated as a metabolic byproduct of fermentation, together with volatile fatty acids such as acetate, during organic matter degradation 9,17. H2 can also be produced abiotically via serpentinization, radiolysis of water, or thermal alteration of sedimentary organic matter 18. For example, the radiolysis of water by naturally occurring radionuclides (e.g. 40K and 238U) is estimated to produce 1011 mol H2 per year 8,19. Depending on the availability of electron acceptors, H2 oxidation can be coupled to sulfate, nitrate, metal, and organohalide respiration, as well as acetogenesis and methanogenesis 8,11.
To develop understanding of the role of hydrocarbon substrates metabolic processes in supporting microbial life in deep-sea sediments, we performed metagenomic, geochemical and metabolomic analyses of three deep seabed sediments (water depth ~3km). The three sites exhibited different levels of migrated thermogenic hydrocarbons. Metagenomes generated from sediment samples of each site were assembled and binned to obtain metagenome-assembled genomes (MAGs) and to reconstruct metabolic pathways for dominant members of the microbial communities. Complementing this genome-resolved metagenomics, a gene-centric analysis was performed by directly examining unassembled metagenomic data. Through the combination of metagenomics with geochemistry and metabolomics, with supporting thermodynamic modeling, we provide evidence that (1) deep-sea sediments harbor phylogenetically diverse heterotrophic and lithotrophic microbial communities; (2) some members from the candidate phyla are engaged in degradation of aliphatic and aromatic thermogenic hydrocarbons; and (3) microbial community members are likely interconnected via acetate and hydrogen metabolism.
Results
Sediment geochemistry
This study tested three petroleum-associated near-surface sediments (referred to as Sites E26, E29 and E44) sampled from the Eastern Gulf of Mexico 20. Petroleum content and other geochemical characteristics were analyzed for each of the three sites (Table 1). All sites had high concentrations of aromatic compounds and liquid alkanes; aromatic compounds were most abundant at Site E26, while liquid alkanes were at 2.5-fold higher concentration at Sites E26 and E29 than Site E44. Alkane gases were only abundant at Site E29 and were almost exclusively methane (CH4). CH4 sources can be inferred from stable isotopic compositions of CH4 and molar ratios of CH4 to higher hydrocarbons 15. Ratios of C1/(C2+C3) were greater than 1,000 and d13C values of methane were more negative than −60‰, indicating that the CH4 in these sediments is predominantly biogenic 15,21. GC-MS revealed an unresolved complex mixture (UCM) of saturated hydrocarbons in the C15+ range in all three sites. Such UCM signals correspond to degraded petroleum hydrocarbons and may indicate the occurrence of oil biodegradation at these sites 22. Signature metabolites for anaerobic biodegradation of alkanes and aromatic compounds 23 were also detected, including benzoate, toluate and methyl- or trimethylsilyl esters (Table S1).
High concentrations of sulfate (>20 mM) were detected at each of the three sites (Table 1), consistent with sulfate being present in high concentrations in seawater and diffusing into the sediments. H2 and acetate concentrations were both below limits of detection (0.015-0.1 nM and 2.5 µM, respectively); this is consistent with previous observations in deep-sea sediments showing that H2 and acetate is present at extremely low steady-state concentrations due to tight coupling between producers and consumers 3,15.
Deep-sea sediments harbor phylogenetically diverse bacterial and archaeal communities
Illumina NextSeq sequencing of genomic DNA from deep-sea sediment communities produced 85,825,930, 148,908,270, and 138,795,692 quality-filtered reads for Sites E26, E29, and E44, respectively (Table S2). The 16S rRNA gene amplicon sequencing results suggest the sediments harbor diverse bacterial and archaeal communities, with Chao1 richness estimates of 359, 1375 and 360 amplicon sequence variants (ASVs) using bacterial-specific primers, and 195, 180 and 247 ASVs using archaeal-specific primers, for Sites E26, E29 and E44, respectively (Table S3 and Figure S1). Taxonomic profiling of these metagenomes using small subunit ribosomal RNA (SSU rRNA) marker genes demonstrated that the most abundant phyla in the metagenomes were, in decreasing order, Chloroflexi (mostly classes Dehalococcoidia and Anaerolineae), Candidatus Atribacteria, Proteobacteria (mostly class Deltaproteobacteria), and Candidatus Bathyarchaeota (Figure 1a). While the three sites share a broadly similar community composition, notable differences were Ca. Bathyarchaeota and Proteobacteria being in higher relative abundance at the sites with more hydrocarbons (E29 and E26; Table 1), whereas the inverse is true for Actinobacteria, the Patescibacteria group, and Ca. Aerophobetes that are all present in higher relative abundance at Site E44 where hydrocarbon levels are lower. Additional sampling is required to determine whether these differences are due to the presence of hydrocarbons or other factors.
Assembly and binning for the three metagenomes resulted in a total of 82 MAGs with >50% completeness and <10% contamination based on CheckM analysis 24. Reconstructed MAGs comprise taxonomically diverse members from a total of six archaeal and 15 bacterial phyla (Figure 2 and Table S4). Within the domain Bacteria, members of the phylum Chloroflexi are highly represented in each sample, especially from the classes Dehalococcoidia and Anaerolineae. Within the domain Archaea, members of phylum Bathyarchaeota were recovered from all three sites. Most other MAGs belong to poorly understood candidate phyla that lack cultured representatives, including Aminicenantes (formerly OP8), Aerophobetes (formerly CD12), Cloacimonas (formerly WWE1), Stahlbacteria (formerly WOR-3), Atribacteria (formerly JS1 and OP9), TA06 and the Asgard superphylum including Lokiarchaeota, Thorarchaeota, and Heimdallarchaeota.
Among those phyla, candidate phylum TA06 is the only one not yet given provisional names. Also known as GN04 or AC1, it was originally discovered in a hypersaline microbial mat 25. First genomic representatives of this phylum were recovered from estuarine sediments 26 with a small number of other MAGs recently reported to belong to this lineage 27,28. Due to the paucity of available MAGs and misclassifications based on 16S rRNA gene sequences, members of TA06 are often ‘confused’ with members of the phylum WOR-3 (Stahlbacteria) 28. In addition to the phylogenetic inference here based on 43 concatenated protein marker genes (Figure 2), the placement of two bins within the original TA06 phylum is further supported by genome classification based on concatenation of 120 ubiquitous, single-copy marker genes 29 as well as classification of 16S rRNA genes using the SILVA database 30 (Tables S4 and S5).
In summary, while there are considerable community-level differences between the three sample locations, the recovered MAGs share common taxonomic affiliations at the phylum and class levels. Guided by sediment geochemistry (Table 1), we subsequently analyzed the metabolic potential of these MAGs to understand how bacterial and archaeal community members generate energy and biomass in these natural petroleum-associated deep-sea environments. Hidden Markov models (HMMs) and homology-based models were used to search for the presence of different metabolic genes in both the recovered MAGs and unbinned metagenomes. Where appropriate, findings were further validated through metabolomic analyses, phylogenetic visualization, and analysis of gene context.
Capacity for detrital biomass and hydrocarbon degradation in sediment microbial communities
In deep-sea marine sediments organic carbon is supplied either as detrital matter from the overlying water column or as aliphatic and aromatic petroleum compounds that migrate upwards from underlying petroleum-bearing sediments 11. With respect to detrital matter, genes involved in carbon acquisition and breakdown were prevalent across both archaeal and bacterial MAGs.
These include genes encoding intracellular and extracellular carbohydrate-active enzymes and peptidases, as well as relevant transporters and glycolysis enzymes (Figure 3 and Table S6). The importance of these carbon acquisition mechanisms is supported by the detection of corresponding intermediate metabolites, such as glucose and amino acids, in all three sediments (Table S1). The ability to break down fatty acids and other organic acids via the beta-oxidation pathway was identified in 13 MAGs, including members of Chloroflexi, Deltaproteobacteria, Aerophobetes and Lokiarchaeota (Figure 3 and Table S6). These results align with many other studies suggesting that the majority of seabed microorganisms are involved in recycling of residual organic matter, including complex carbohydrates, proteins and lipids 13,31,32.
Unlike in other studies, the presence of petroleum hydrocarbons is a defining feature of the sediments investigated here and thus a key goal of this study was to identify the potential for microbial degradation of hydrocarbons as a source of energy and carbon. To this end, we focused on functional marker genes encoding enzymes that catalyze the activation of mechanistically sophisticated C-H bonds, to initiate hydrocarbon biodegradation 33. For anaerobic hydrocarbon degradation, four oxygen-independent C-H activation reactions have been characterized: (1) addition of ⁗ fumarate by glycyl-radical enzymes, e.g. for activation of alkylbenzenes and straight chain alkanes 34; (2) hydroxylation with water by molybdenum cofactor-containing enzymes, e.g. for activation of ethylbenzene 33; (3) carboxylation catalyzed by UbiD-like carboxylases, e.g. for activation of benzene and naphthalene 35; and (4) reverse methanogenesis involving variants of methyl-coenzyme ⁗ M reductase, e.g. for activation of methane and butane 36. Most of the evidence for mechanisms (1) – (3) has come from studies of hydrocarbon contaminated aquifers, whereas mechanism (4) has been studied extensively in marine sediments23, 37.
Evidence for glycyl-radical enzymes that catalyze fumarate addition was found in 15 out of the 82 MAGs based on identifying genes encoding alkylsuccinate synthase (AssA) (Figures 3 and 4a). The assA sequences identified, while phylogenetically distant from canonical fumarate-adding enzymes and pyruvate formate lyases (Pfl), form a common clade with Pfl-like AssA from Archaeoglobus fulgidus VC-16 and Abyssivirga alkaniphila L81 (Figure 4a). Both of these organisms have been shown experimentally to be capable of anaerobic alkane degradation 38,39. The putative assA genes identified here are present in all three samples regardless of hydrocarbon concentrations. They belong to MAGs affiliated with the bacterial phyla Aerophobetes, Aminicenantes and Chloroflexi as well as the archaeal phyla Bathyarchaeota, Lokiarchaeota and Thorarchaeota. The highest relative abundance of putative assA sequences was found in Site E29 as indicated by quality-filtered reads, which is consistent with this sediment containing the highest concentration of aliphatic compounds (Tables 1 and S7). Additional searching for other genes encoding fumarate-adding enzymes in the quality-filtered reads (e.g. bssA, nmsA, and canonical assA) did not return significant counts (Figure 4 and Table S7). Among the other three anaerobic hydrocarbon biodegradation mechanisms mentioned above, a MAG classified as Dehalococcoidia (Chloroflexi E29_bin2) contained genes encoding putative catalytic subunits of p-cymene dehydrogenase (Cmd) and alkane C2-methylene hydroxylase (Ahy) (Figures 3 and S2), known to support p-cymene and alkane utilization 37. Genes encoding enzymes catalyzing hydrocarbon carboxylation, reverse methanogenesis and aerobic hydrocarbon degradation (e.g. alkB, nahC and nahG) were not detected (Table S6). The latter result is expected due to the low concentrations of oxygen in the top 20 cm of organic rich seabed sediments 11.
Considering the degradation of aromatic hydrocarbons, genes responsible for reduction of benzoyl-CoA were detected in 12 MAGs (Figures 3 and 4b). Benzoyl-CoA is a universal biomarker for anaerobic degradation of monoaromatic compounds as it is a common intermediate to biochemical pathways catalyzing this process 40. Benzoyl-CoA reduction to cyclohex-1,5-diene-1-carboxyl-CoA is performed by Class I ATP-dependent benzoyl-CoA reductase (BCR; BcrABCD) in facultative anaerobes (e.g. Thauera aromatica) or Class II ATP-independent reductase (Bam; BamBCDEFGHI) in strict anaerobes like sulfate reducers 41. The bcr genes detected are all Class I, and were found in bacterial MAGs (i.e., Dehalococcoidia, Anaerolineae, Deltaproteobacteria, Aminicenantes and TA06) and archaeal MAGs (i.e., Thermoplasmata and Bathyarchaeota) (Figures 3 and 4b). Genes for further transformation of dienoyl-CoA to 3-hydroxypimelyl-CoA were also identified (Figures 3 and 4b), i.e., those encoding 6-oxo-cyclohex-1-ene-carbonyl-CoA hydrolase (Oah), cyclohex-1,5-diencarbonyl-CoA hydratase (Dch) and 6-hydroxycyclohex-1-ene-1-carbonyl-CoA dehydrogenases (Had) 42. Together with the detection of 23 – 162 nM benzoate in these sediments (Table 1) these results strongly suggest that the organisms represented by these MAGs mediate the typical downstream degradation of aromatic compounds through the central benzoyl-CoA Bcr-Dch-Had-Oah pathway. However, the upstream pathways resulting in benzoate production from degradation of complex aromatic compounds were not resolved based on current data.
Widespread capacity for fermentative production and respiratory consumption of acetate and hydrogen
Analysis of MAGs from these deep-sea hydrocarbon-associated sediments suggests that fermentation, rather than respiration, is the primary mode of organic carbon turnover in these environments. Most recovered MAGs with capacity for heterotrophic carbon degradation lacked respiratory primary dehydrogenases and terminal reductases, with exceptions being several Proteobacteria and one Chloroflexi (Table S6). In contrast, 6 and 14 MAGs contained genes indicating the capability for fermentative production of ethanol and lactate, whereas some 69 MAGs contained genes for fermentative acetate production (Figure 3 and Table S6). These findings are consistent with other studies emphasizing the importance of fermentation, including acetate production, in deep-sea sediments 12,43.
Acetate can also be produced by acetogenic CO2 reduction through the Wood-Ljungdahl pathway using a range of substrates, including heterotrophic compounds 15. Partial or complete sets of genes for the Wood-Ljungdahl pathway were found in 50 MAGs (Figures 3 and S3), including those affiliated with phyla previously inferred to mediate acetogenesis in deep-sea sediments through either the tetrahydrofolate-dependent bacterial pathway (e.g. Chloroflexi and Aerophobetes) 7,44 or the tetrahydromethanopterin-dependent archaeal variant (e.g. Bathyarchaeota and Asgard group) 45,46. In addition, the signature diagnostic gene for the Wood-Ljungdahl pathway (acsB; acetyl-CoA synthase) is in high relative abundance in the quality-filtered metagenome reads at all three sites (Table S7). These observations are in agreement with mounting evidence that homoacetogens play a quantitatively important role in organic carbon cycling in the marine deep biosphere 45,47,48.
Evidence for H2 metabolism was also found in MAGs from all three sites. We screened putative hydrogenase genes from various subgroups in MAGs as well as unbinned metagenomic sequences (Figures 1, 3 and Tables S6, S7). Surprisingly few H2 evolving-only hydrogenases were detected, with only five Group A [FeFe]-hydrogenases and five Group 4 [NiFe]-hydrogenases detected across the bacterial and archaeal MAGs. Instead, the most abundant hydrogenases within the MAGs and quality-filtered unassembled reads were the Group 3b, 3c, and 3d [NiFe]-hydrogenases. Group 3b and 3d hydrogenases are physiologically reversible, but generally support fermentation in anoxic environments by coupling NAD(P)H reoxidation to fermentative H2 evolution 49-51. Group 3c hydrogenases mediate a central step in hydrogenotrophic methanogenesis, bifurcating electrons from H2 to heterodisulfides and ferredoxin 52; their functional role in Bacteria and non-methanogenic Archaea remains unresolved 51 yet their corresponding genes co-occur with heterodisulfide reductases across multiple archaeal and bacterial MAGs (Figure 3). Various Group 1 [NiFe]-hydrogenases were also detected, which are known to support hydrogenotrophic respiration in conjunction with a wide range of terminal reductases. This is consistent with previous studies in the Gulf of Mexico that experimentally measured the potential for hydrogen oxidation catalyzed by hydrogenase enzymes 53.
Given the genomic evidence for hydrogen and acetate production in these sediments, we investigated whether any of the MAGs encoded terminal reductases to respire using these compounds as electron donors. In agreement with the high sulfate concentrations (Table 1), the key genes for dissimilatory sulfate reduction (dsrAB) were widespread across the metagenome reads, particularly at Site E29 (Table S7). These genes were recovered from MAGs affiliated with Deltaproteobacteria and Dehalococcoidia (Table S6). We also identified 31 novel reductive dehalogenase (rdhA) genes across 22 MAGs, mainly from Aminicenantes and Bathyarchaeota (Figure 3 and Table S6), suggesting that organohalides – that can be produced through abiotic and biotic processes in marine ecosystems 54 – may be electron acceptors in these deep-sea sediments. All MAGs corresponding to putative sulfate reducers and dehalorespirers encoded the capacity to completely oxidize acetate and other organic acids to CO2 using either the reverse Wood-Ljungdahl pathway or TCA cycle (Figure 3 and Table S6). Several of these MAGs also harbored the capacity for hydrogenotrophic dehalorespiration via Group 1a and 1b [NiFe]-hydrogenases (Figure 3). In addition to these dominant uptake pathways, one MAG belonging to the epsilonproteobacterial genus Sulfurovum (E29_bin29) included genes for the enzymes needed to oxidize either H2 (group 1b [NiFe]-hydrogenase), elemental sulfur (SoxABXYZ), and sulfide (Sqr), using nitrate as an electron acceptor (NapAGH); this MAG also has a complete set of genes for autotrophic CO2 fixation via the reductive TCA cycle (Figure 3 and Table S6).). In contrast, the capacity for methanogenesis appears to be relatively low and none of the MAGs contained mcrA genes. The genes for methanogenesis were detected in quality-filtered unassembled reads in all three sediments (Figures 1d and S4) and were mainly affiliated with acetoclastic methanogens at Site E29, and hydrogenotrophic methanogens at the other two sites (Figures 1d and S4). Overall, the collectively weak mcrA signal in the metagenomes suggests that the high levels of biogenic methane detected by geochemical analysis (Table 1) is primarily due to methanogenesis in sediment layers deeper than the top 20 cm.
Thermodynamic modelling of hydrocarbon degradation
Both the geochemistry data and biomarker gene survey suggest that hydrocarbon degradation occurs in the three deep-sea sediments sampled (Tables 1 and S1). Recreating the environmental conditions for cultivating the organisms represented by the retrieved MAGs is a challenging process, preventing further validation of the hydrocarbon degradation capabilities (and other metabolisms) among the majority of the lineages represented by the MAGs retrieved here 48. Instead, we provide theoretical evidence that hydrocarbon degradation is feasible in this environment by modelling whether these processes are thermodynamically favorable in the conditions typical of deep sea sediments, namely high pressure and low temperature.
As concluded from the genome analysis and supported by metabolomics (Table 1), it is likely that most hydrocarbon oxidation occurs through fermentation rather than respiration. Taking hydrogen production and the Wood-Ljungdahl pathway into consideration (Figures 3 and 4), we compared the thermodynamic constraints on hydrocarbon biodegradation for two plausible scenarios: (1) fermentation with production of hydrogen and acetate, and (2) fermentation with production of acetate alone. Hexadecane and benzoate are used as representative aliphatic and aromatic compounds, respectively, based on the geochemistry results (e.g. C2+ alkane detection) and genomic analysis (e.g. bcr genes) 47,55. The calculated results show that the threshold concentrations of acetate that result in favorable energetics (ΔG′ < 0 kJ mol-1) for fermentative co-generation of acetate and hydrogen require acetate to be extremely low in a hexadecane degradation scenario (< 10-12 mM acetate) and acetate to be at moderate levels in a benzoate degradation scenario (< 3.8 mM acetate) (Figure 5). By contrast, for fermentation leading to production of only acetate, its concentration can be as high as 470 mM in a benzoate degradation scenario and as high as 300 mM in a hexadecane degradation scenario (Figure 5). Fermentative degradation of hexadecane to hydrogen and acetate in the deep seabed could therefore be less favorable than acetate production alone via the Wood-Ljungdahl pathway Thus, if microbial communities consume hexadecane or more complex hydrocarbons as carbon and energy sources, it is likely that they employ the Wood-Ljungdahl pathway to produce acetate. However, other reactions such as fermentation to H2 still cannot be excluded, e.g., for less complex hydrocarbons such as benzoate and related compounds.
Discussion
In this study, metagenomics revealed that most of the Bacteria and Archaea in the deep-sea sediment microbial communities sampled belong to candidate phyla that lack cultured representatives and sequenced genomes (Figures 1 and 2). As a consequence, it is challenging to link phylogenetic patterns with the microbial functional traits underpinning the biogeochemistry of deep seabed habitats. Here, we were able to address this by combining de novo assembly and binning of metagenomic data with geochemical and metabolomic analyses, and complementing our observations with thermodynamic modeling. Pathway reconstruction from 82 MAGs recovered from the three deep-sea near surface sediments revealed that many community members were capable of anaerobic hydrocarbon degradation as well as acquiring and hydrolyzing residual organic matter (Figure 3), whether supplied as detritus from the overlying water column or as autochthonously produced necromass (Figure 6). Heterotrophic fermenters and acetogens were in considerably higher relative abundance than heterotrophic respirers, despite the abundance of sulfate in the sediments (Table 1). For example, while genomic coverage of putative sulfate reducers is relatively low (< 1% of the communities), the most abundant MAG at each site were all putative acetogenic heterotrophs, i.e. Dehalococcoidia E26_bin16, Actinobacteria E44_bin5, and Aminicenantes E29_bin47 for Sites E26, E44 and E29 respectively (~3.3-4.5% relative abundance, Table S4). Therefore, in contrast with coastal sediments 56, microbial communities in the deep seabed are likely influenced by the capacity to utilize available electron donors more so than by the availability of oxidants.
In this context, multiple lines of evidence indicate degradation of aliphatic or aromatic petroleum compounds as carbon and energy sources for anaerobic populations in these deep-sea hydrocarbon seep environments (Table 1, Figures 3-5). Whereas capacity for detrital organic matter degradation is a common feature in the genomes retrieved in this study, and from many other environments 26, anaerobic hydrocarbon degradation is a more exclusive feature that was detected in 23 out of 82 MAGs. Evidence of anaerobic alkane oxidation via fumarate addition and hydroxylation pathways, as well as anaerobic aromatic compound degradation by the Class I benzoyl-CoA reductase pathway, was found in all three sediments. The ability to utilize hydrocarbons may explain the ecological dominance (high relative abundance) of certain lineages of Bacteria and Archaea in these microbial communities (Figure 1a), as many of those phyla have previously been found to be associated with hydrocarbons in various settings. For example, Aerophobetes have been detected in other cold seep environments 7, Aminicenantes are often found associated with fossil fuels 27, and Chloroflexi harboring genes for anaerobic hydrocarbon degradation have been found in hydrothermal vent sediments 4. While Archaea have been reported to mediate oxidation of methane and other short-chain alkanes in sediments 5,36, few have been reported to anaerobically degrade larger hydrocarbons 37. The finding of Bathyarchaeota and other archaeal phyla potentially capable of anaerobic hydrocarbon degradation extends the potential hydrocarbon substrate spectrum for Archaea. More broadly, these findings extend the breadth of bacterial and archaeal lineages that putatively degrade hydrocarbons. Current knowledge of anaerobic hydrocarbon degradation remains limited, with the majority of studies focused on environments subject to anthropogenic hydrocarbon contamination, most notably groundwater aquifers 37. It is possible that microorganisms inhabiting deep-sea sediments harbor novel mechanisms for anaerobic hydrocarbon degradation that may be relevant for biotechnology and bioremediation in a variety of other settings, e.g., other cold habitats. Future studies of genome-enabled hydrocarbon degradation using samples such as the sediments studied here may elucidate this further.
Genomic analyses of 12 MAGs harboring genes for central benzoyl-CoA pathway reveal that they are likely a mixture of obligate fermenters and sulfate reducers. The finding that these organisms use the ATP-consuming class I, not the reversible class II, benzoyl-CoA reductase is surprising. It is generally thought that strict anaerobes must use class II BCRs because the amount of energy available from benzoate oxidation during sulfate reduction or fermentation is not sufficient to support the substantial energetic requirement of the ATP-dependent class I BCR reaction 42. However, acetogenic fermentation of hydrocarbons may explain how the Class I reaction could be thermodynamically favorable, as shown in Figure 5. In agreement with this, there are reported exceptions to the general Class I vs Class II observations, such as the hyperthermophilic archaeon Ferroglobus placidus that couples benzoate degradation via the Class I system with iron reduction 42, and fermentative deep-sea Chloroflexi strains DscP3 and Dsc4 that contain genes for class I benzoyl-CoA reductases 44. Indeed, acetogens can utilize many different substrates and have relatively high ATP yields, as well as thermodynamic efficiencies toward heterotrophic substrates, which is consistent with the proposed importance of acetogens in energy-limited seafloor ecosystems 45,47.
Based on the evidence presented here, we propose that acetate and hydrogen are the central intermediates underpinning community interactions and biogeochemical cycling in these deep-sea sediments (Figure 6). Maintaining low acetate and hydrogen concentrations in the environment is important for promoting continuous fermentation of organic substrates, consistent with thermodynamic constraints (Figure 5). Acetate and hydrogen in sediment porewater were below detection limits, consistent with the high turnover rates of both compounds. This may correspond with the genomic potential within these microbial communities for the coupling of acetate consumption to sulfate reduction, organohalide respiration and acetoclastic methanogenesis, as suggested in other studies 55,57. Some community members also appear to be capable of H2 consumption, including via putative heterodisulfide reductase-coupled hydrogenases. In turn, hydrogen oxidation can support autotrophic carbon fixation and therefore may provide a feedback loop for regeneration of organic carbon. Acetate- and hydrogen-oxidizing community members are likely to promote upstream fermentative degradation of necromass and hydrocarbons (Figure 6).
Overall, this metagenome dataset extended the knowledge of metabolic potential of microbial communities inhabited in deep-sea sediments that receive an input of thermogenic hydrocarbon. They are mostly likely sustained through fermentation, acetogenesis and hydrogen metabolisms. More importantly, as supported by geochemical data, metabolomic analysis, and thermodynamic modelling, our findings expand the diversity of microbial lineages with the potential for anaerobic hydrocarbon degradation through e.g. the activity of glycyl-radical enzymes. Together with the recent discovery of anaerobic butane degradation in gas-rich hydrothermally-heated sediments 36, it can be inferred that anaerobic degradation of hydrocarbons heavier than methane might be more widespread than previously expected and may significantly contribute to energy and carbon budgets in dark deep-sea sediments 58.
Methods
Sampling and geochemical measurements
The three marine sediment samples used in this study were collected from the near-surface (top 20 cm) of the seafloor in the Eastern Gulf of Mexico as part of a piston coring survey, as described previously 20. Samples for hydrocarbon characterization were sectioned on board the research vessel immediately following piston core retrieval, flushed with N2 and sealed in hydrocarbon-free gas tight metal canisters then frozen until analysis. Interstitial gas analysis was later performed on the headspace in the canisters using GC with Flame Ionization Detector (GC-FID). Sediment samples for gas/liquid chromatography and stable isotope analysis were frozen, freeze-dried and homogenized then extracted using accelerated solvent extraction (ACE 200). Extracts were subsequently analyzed using GC/FID, a Perkin-Elmer Model LS 50B fluorometer, GC/MS and Finnigan MAT 252 isotope mass spectrometry as detailed elsewhere 59.
Sulfate and chloride concentrations were measured in a Dionex ICS-5000 reagent-free ion chromatography system (Thermo Scientific, CA, USA) equipped with an anion-exchange column (Dionex IonPac AS22; 4 × 250 mm; Thermo Scientific), an EGC-500 K2CO3 eluent generator cartridge and a conductivity detector. Organic acids were analysed in the 0.2 µm filtered sediment porewater using a Thermo RS3000 HPLC fitted with an Ultimate 3000 UV detector. Separation was achieved over an Aminex HPX-87H organic acid column (Biorad, USA) under isocratic conditions (0.05 mM H2SO4) at 60°C with a run time of 20 minutes. Organic acids were compared to the retention time of known standards and the limit of detection for acetate was determined to be 2.5 µM.
For the analysis of metabolites, sediment was spun down, the supernatant collected, diluted 1:1 in pure methanol, and filtered through 0.2 µm Teflon syringe filters. Extracts were separated using Ultra High-Performance Liquid Chromatography (UHPLC) equipped with a hydrophilic interaction liquid chromatography column (Syncronis HILIC, Thermo Fisher). A Thermo Fisher Scientific Q-Exactive HF mass spectrometer in negative-mode electrospray ionization was used to collect high-resolution full-scan MS data from 50-750 m/z at 240,000 resolution with an automatic gain control (AGC) target of 3e6 and a maximum injection time of 200 ms. In addition, benzoate ion (m/z 121.02943) was subjected to fragmentation using collision induced dissociation (CID) with a collision energy of 10eV at 120,000 resolution (m/z 121.02943 > 77.03948). For CID experiments, an AGC target of 1e6 was used with a maximum injection time of 100 ms. Metabolites were further identified using accurate mass and retention times of standards using Thermo Xcalibur software and MAVEN freeware. Larger compound lists were assigned identification using a combination of MAVEN and the KEGG database 60.
DNA extraction and sequencing
For the three sediment samples, DNA was extracted from 10 g of sediment using the PowerMax Soil DNA Isolation Kit (12988-10, QIAGEN) according to the manufacturer’s protocol with minor modifications for the step of homogenization and cell lysis i.e., cells were lysed in PowerMax Bead Solution tubes for 45 s at 5.5 m s-1 using a Bead Ruptor 24 (OMNI International). DNA concentrations were assessed using a Qubit 2.0 fluorometer (Thermo Fisher Scientific, Canada). Metagenomic library preparation and DNA sequencing was conducted at the Center for Health Genomics and Informatics in the Cumming School of Medicine, University of Calgary. DNA fragment libraries were prepared by shearing genomic DNA using a Covaris sonicator and the NEBNext Ultra II DNA library preparation kit (New England BioLabs). DNA was sequenced on a ~40 Gb (i.e. 130 M reads) mid-output NextSeq 500 System (Illumina Inc.) 300 cycle (2 × 150 bp) sequencing run.
To provide a high-resolution microbial community profile, the three samples were also subjected to 16S rRNA gene amplicon sequencing on a MiSeq benchtop sequencer (Illumina Inc.). DNA was extracted from separate aliquots of the same sediment samples using the DNeasy PowerLyzer PowerSoil kit (MO BIO Laboratories, a Qiagen Company, Carlsbad, CA, USA) and used as the template for different PCR reactions. The v3-4 region of the bacterial 16S rRNA gene and the v4-8 region of the archaeal 16S rRNA gene were amplified using the primer pairs SD-Bact-0341-bS17/SD-Bact-0785-aA21 and SD-Arch-0519-aS15/SD-Arch-0911-aA20, respectively 61 as described previously 20 on a ~15 Gb 600-cyce (2 × 300 bp) sequencing run (for results see Figure S1).
Metagenomic assembly and binning
Raw reads were quality-controlled by (1) clipping off primers and adapters and (2) filtering out artifacts and low-quality reads as described previously 62. Filtered reads were assembled using metaSPAdes version 3.11.0 63 and short contigs (<500 bp) were removed. Sequence coverage was determined by mapping filtered reads onto assembled contigs using BBmap version 36 (https://sourceforge.net/projects/bbmap/). Binning of metagenome contigs was performed using MetaBAT version 2.12.1 (--minContig 1500) 64. Contaminated contigs in the produced bins were further removed based on genomic properties (GC, tetranucleotide signatures, and coverage) and taxonomic assignments using RefineM version 0.0.22 65. Resulting bins were further examined for contamination and completeness using CheckM version 1.0.8 with the lineage-specific workflow 24.
Annotation
For MAGs, genes were called by Prodigal (-p meta) 66. Metabolic pathways were predicted against the KEGG GENES database using the GhostKOALA tool 67 and against the Pfam, TIGRfam and custom HMM databases (https://github.com/banfieldlab/metabolic-hmms) using MetaErg (https://sourceforge.net/projects/metaerg/). The dbCAN web server was used for carbohydrate-active gene identification (cutoffs: coverage fraction: 0.40; e-value: 1e-18) 68. Genes encoding proteases and peptidases were identified using BLASTp against the MEROPS database release 12.0 (cutoffs: e-value, 1e-20; sequence identity, 30%) 69. Genes involved in anaerobic hydrocarbon degradation were identified using BLASTp against a custom database (Table S8) (cutoffs: e-value, 1e-20; sequence identity, 30%). Hydrogenases were identified and classified using a web-based search using the hydrogenase classifier HydDB 70.
Full-length 16S rRNA genes were reconstructed from metagenomic reads using phyloFlash version 3.1 (https://hrgv.github.io/phyloFlash/) together with the SILVA SSU 132 rRNA database 30. Diversity calculations were based on separate 16S rRNA gene amplicon library results 20. Functional and taxonomic McrA gpkgs were used to assess the diversity of methanogens against the metagenomic reads using GraftM with default parameters 71. Genes encoding the catalytic subunits of hydrogenases, dsrA, acsB, assA, nmsA and bssA were retrieved from metagenomic reads through diamond BLASTx 72 queries against comprehensive custom databases 46,70,73 (cutoffs: e-value, 1e-10; sequence identity, 70%).
Phylogenetic analyses
For taxonomic classification of each MAG, two methods were used to produce genome trees that were then used to validate each other. In the first method the tree was constructed using concatenated proteins of up to 16 syntenic ribosomal protein genes following procedures reported elsewhere 74; the second tree was constructed using concatenated amino acid sequences of up to 43 conserved single-copy genes following procedures described previously 75. Both trees were calculated using FastTree version 2.1.9 (-lg -gamma) 76 and resulting phylogenies were congruent. Reference genomes for relatives were accessed from NCBI GenBank, including genomes selected from several recent studies representing the majority of candidate bacterial and archaeal phylogenetic groups 4,65,77-80. The tree in Figure 2 was inferred based on concatenation of 43 conserved single-copy genes (Database S1). Specifically, it was built using RaxML version 8 81 implemented by the CIPRES Science Gateway 82 and it was called as follows: raxmlHPC-HYBRID -f a -n result -s input -c 25 -N 100 -p 12345 -m PROTCATLG -x 12345. The phylogeny resulting from RAxML is consistent with the taxonomic classification of MAGs that resulted from FastTree. Interactive tree of life (iTOL) version 3 83 was used for tree visualization and modification.
For phylogenetic placements of functional genes, sequences were aligned using the MUSCLE algorithm 84 included in MEGA7 85. All positions with less than 95% site coverage were eliminated. Maximum likelihood phylogenetic trees were constructed in MEGA7 using a general time reversible substitution model and uniform rates among sites. These trees were bootstrapped with 100 replicates.
Taxonomic classification of MAGs inferred to belong to candidate phylum TA06 after phylogenetic analyses were additionally confirmed by performing classify workflow using GTDB-Tk version 0.0.6+ (https://github.com/Ecogenomics/GtdbTk).
Thermodynamic calculations
The values of Gibbs free energy of formation for substances were taken from Madigan et al. 86 and Dolfing et al. 55. The pH used in all calculations was 8.0 as reported in a previous thermodynamic study of deep marine sediments 47, partial pressure was 300 atm based on water depths at the three sites (http://docs.bluerobotics.com/calc/pressure-depth/), and temperature was set as 4°C to represent deep sea conditions 87. Calculations followed accepted protocols for determining reaction kinetics and thermodynamics 88.
Data availability
DNA sequences (amplicon sequences, genomes and raw sequence reads) have been deposited in the NCBI BioProject database with accession number PRJNA415828 and PRJNASUB3936075 (https://www.ncbi.nlm.nih.gov/bioproject/). The authors declare that all other data supporting the findings of this study are available within the article and its supplementary information files, or from the corresponding authors upon request.
Acknowledgements
The work was supported by a Genome Canada Genomics Applied Partnership Program (GAPP) grant to CRJH. I.A.L is supported by an Alberta Innovates Translational Health Chair.
Metabolomics data were acquired by R.A.G. at the Calgary Metabolomics Research Facility (CMRF), which is supported by the International Microbiome Centre and the Canada Foundation for Innovation (CFI-JELF 34986). We thank Xiaoli Dong and Marc Strous for establishing bioinformatics workflow and pipelines, the Centre for Health Genomics and Informatics at University of Calgary for NextSeq sequencing, Alexander Probst for providing the database of 16 syntenic ribosomal proteins, and Nina Dombrowski and Brett Baker for providing a custom blast database for hydrocarbon degradation genes.
Footnotes
The authors declare no conflict of interest.
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.
- 79.
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵