Sediment samples were collected at depth of 5,800 m (142.274°E, 10.761°N) in the Mariana Trench during the cruise aboard R/V “ZHANG JIAN HAO” in December 2017. A box corer (with a base area of 400 cm² and a height of 25 cm) attached to the Hadal Lander II was used to collect sediment samples from the Trench as described in Luo et al. (2018). To collect sediment from the seafloor, 2 h after the lander reached the seafloor, the box corer was slowly driven into the seafloor until it reached ~25 cm below the sediment surface. A lid was then released to seal the box corer, and the lander was recovered. After recovering on board, the sediment core was immediately subsampled and stored in 50-ml sterile centrifuge tubes in a cold room at 4°C on board the ship.

Approximately 2.5 g of the collected sediment from the middle of the core was used to set up enrichment cultures for bacterial isolation at each pressure (10, 20, 50, and 100 MPa). The sediment was added into 100 ml MMC medium (24 g/L NaCl, 7.0 g/L MgSO₄·7H₂O, 1 g/L NH₄NO₃, 0.7 g/L KCl, 2.0 g/L KH₂PO₄, 3.0 g/L Na₂HPO₄, and pH 7.5) (Liu et al., 2019), supplemented with 0.1 ml of a vitamin mixture (including 40 mg/L 4-aminobenzoic acid, 10 mg/L biotin, 100 mg/L nicotinic acid, 50 mg/L pantothenic calcium, 150 mg/L pyridoxine dihydrochloride, 100 mg/L thiamine chloride dihydrochloride, and 50 mg/L cyanocobalamin), 0.1 ml of a trace element mixture [including 10 mg/L H₃BO₃, 5 mg/L MnCl₂·4H₂O, 4 g/L FeSO₄·7H₂O, 190 mg/L CoCl₂·6H₂O, 24 mg/L NiCl₂·6H₂O, 2 mg/L CuCl₂·2H₂O, 200 mg/L ZnSO₄·7H₂O, 36 mg/L Na₂MoO₄·2H₂O, and 13 ml/L HCl (7.7 M)], and 125 mg/L n-alkane mixture (hexadecane, heptadecane, and octadecane as 1:1:1, w/w) as the carbon sources, and filled into disposable plastic syringes, which were sterilized by ultraviolet light before usage. After heat sealing, these disposable plastic syringes were placed into stainless steel high-pressure vessels (Feiyu Science and Technology Exploitation Co. Ltd., Nantong, China) and incubated at 4 °C and multiple pressures (10, 20, 50, and 100 MPa).

After 6 months of incubation, the enrichment cultures were diluted (1:1,000) and then coated in MMC-alkane-agar plates at 4°C and atmospheric pressure for bacterial isolation. After another 6 months, bacterial colonies, presenting no more than 10 per plate, were picked, inoculated on R2A-agar plates (containing 0.5 g/L tryptone, 0.5 g/L yeast extract, 0.5 g/L soluble starch, 0.5 g/L glucose, 0.5 g/L casein amino acids, 0.3 g/L sodium pyruvate, and 15 g/L agar power, pH = 7.2), and then cultured at 16°C. Such adjustments in temperature, pressure, and culture medium could accelerate its growth for better identification of its colony morphology. After 1 week, strain C2-1 (= MCCC 1K03735 = KCTC 72289) was isolated from the R2A-agar plates.

For cellular fatty acid analysis, strain C2-1 and its closely related reference strains (Thalassolituus marinus IMCC1826^T, Thalassolituus oleivorans DSM 14913^T, Oleibacter marinus DSM 24913^T, and Oceanobacter kriegii IFO 15467^T) were cultured in MB for 3 days at 35°C. Fatty acids were saponified, methylated, and extracted using the standard protocol of MIDI (Sherlock Microbial Identification System, version 6.0). Fatty acid methyl esters were analyzed by gas chromatography (Agilent Technologies 6850) and identified by using the RTSBA6.0 database of the Microbial Identification System (Athalye et al., 1985). Polar lipids of strain C2-1^T were extracted and separated on silica gel 60 F₂₅₄ aluminum-backed thin-layer chromatography plates (10 × 10 cm; Merk 5554) and further analyzed according to Minnikin et al. (1984). The first dimension of the solvent system was chloroform/methanol/water (65:24:4, by vol.), and the second dimension was chloroform/glacial acetic acid/methanol/water (80:15:12:4, by vol.). Then, the plates were sprayed with 5% phosphomolybdic acid (w/v, dissolved in alcohol) and heated at 160°C for 10–15 min to reveal total lipids. The respiratory quinones were extracted using the method described by Minnikin et al. (1984) and analyzed by HPLC as described by Tindall (1990).

The MMC liquid media (30 ml), prepared as described earlier, was distributed in air-tight syringes and inoculated with freshly grown strain C2-1. The cultures were supplemented with a mixture of C_{16 − 18} n-alkanes as the sole source of carbon, and the culture medium was then filled into disposable plastic syringes, which were sterilized by ultraviolet light before usage. After heat sealing, these disposable plastic syringes were placed into stainless steel high-pressure vessels (Feiyu Science and Technology Exploitation Co. Ltd., Nantong, China). High-pressure incubations were conducted at 20°C and atmospheric pressure (0.1 MPa) or at in situ temperature and pressure (4°C and 58 MPa) for 10 days.

Once the incubation was terminated, the cultures were extracted for n-alkanes using dichloromethane (DCM) twice. The extracts were combined, dried under a gentle stream of nitrogen, and then redissolved in 500 μl DCM. The extracted n-alkanes were analyzed on a Hewlett-Packard 7890B gas chromatography with an HP 5977A Mass Selective Detector. Analytical separation of the n-alkanes was accomplished using a HP-5ms capillary column (50 m × 0.25 mm × 0.25 μm). The column temperature was programmed from 70 to 130°C at 20°C/min and then to 310°C at 4°C/min (and held for 15 min). Helium was used as a carrier gas, and the flow rate was set at 1 ml/min. The concentrations of individual n-alkanes were determined based on the chromatographic response of the alkanes relative to that of the internal standard (squalene, Sigma).

The genomic DNA of strain C2-1 was extracted as described by Fang et al. (2017). The genome of C2-1 was sequenced using a combination of PacBio RS II Single Molecule Real Time (SMRT) and Illumina HiSeq 2500 platforms by MajorBio (Shanghai Co., Ltd., China).

For Illumina sequencing, approximately 1μg genomic DNA was sheared into 400–500 bp fragments using a Covaris M220 Focused Acoustic Shearer following the manufacturer's protocol. Illumina sequencing libraries were prepared from the sheared fragments using the NEXTFLEX™ Rapid DNA-Seq Kit.

For Pacific Biosciences sequencing, an aliquot of 15 μg DNA was spun in a Covaris g-TUBE (Covaris, MA) at 6,000 RPM for 60 s using an Eppendorf 5424 centrifuge (Eppendorf, NY). DNA fragments were then purified, end-repaired, and ligated with SMRTbell sequencing adapters following the manufacturer's recommendations (Pacific Biosciences, CA). The resulting sequencing library was purified three times using 0.45 × volumes of Agencourt AMPure XP beads (Beckman Coulter Genomics, MA) following the manufacturer's recommendations. Next, a ~10 kb insert library was prepared and sequenced on one SMRT cell using standard methods.

The data generated from PacBio and Illumina platforms were used for bioinformatics analysis. The complete genome sequence was assembled using both the PacBio reads and the Illumina reads. The original image data are transferred into sequence data via base calling, which is defined as raw data or raw reads and saved as a FASTQ file. A statistic of quality information was applied for quality trimming, by which the low-quality data can be removed to form clean data. The reads were then assembled into a contig using the hierarchical genome assembly process (HGAP) and canu (Koren et al., 2017). The last circular step was checked and finished manually, generating a complete genome with a single seamless chromosome. Finally, error correction of the PacBio assembly results was performed using the Illumina reads using Pilon (Walker et al., 2014).

The NCBI prokaryotic genome annotation pipeline (Tatusova et al., 2016) was used in ORF prediction and gene annotation. The predicted protein sequences were also aligned with the Clusters of Orthologous Groups of proteins (COG) (Galperin et al., 2021) and TransporterDB 2.0 (Elbourne et al., 2017) databases using the BLASTp software with the following parameters: identity, 50%; query-cover, 80%; and e-value 1e-5. The annotation of the Kyoto Encyclopedia of Genes and Genomes (KEGG) was assigned with BlastKOALA (Kanehisa et al., 2016). The genomic islands were predicted using IslandViewer 4 (Bertelli et al., 2017).

The 120 conserved bacterial marker genes of the GTDB taxonomy were used to study the phylogeny of strain C2-1 and its related strains. The sequences of 120 concentrated proteins in the genomes were predicted using GTDB-Tk (database version: Release 07-RS207) (Chaumeil et al., 2019) and separately aligned using Clustal Omega (Sievers and Higgins, 2021). The aligned sequences were manually degapped. If one concentrated protein was not identified in some genome, we would then add the corresponding number of “–,” depending on the length of the sequence after degap. Then, the degapped alignments of each concentrated protein were tandemly connected (Tang et al., 2021). The phylogenetic tree was constructed using FastTree2 with the neighbor-joining method (Price et al., 2010), and a bootstrap analysis with 1,000 replicates was performed to assess the robustness of the tree. Finally, the phylogenetic tree was plotted using iTOL (Letunic and Bork, 2021).

In addition, to identify the genus-specific insertions and deletions of genus “UBA2009,” we separately aligned the 120 conserved proteins predicted by GTDB-tk (Chaumeil et al., 2019) using Clustal Omega (Sievers and Higgins, 2021) and then identified the insertions and deletions based on the alignments using lab-in Perl scripts.

To study the geographic distribution of the genus Venatorbacter, we first identified three signature proteins of this genus (FOT50_RS00155, FOT50_RS08930, and FOT50_RS12870 in C2-1), which were annotated as hypothetical proteins, and found in almost all the Venatorbacter genomes, but absent in any other microbes (BLASTp identity, 50%; query-cover, 80%). Then, these signature proteins were used to query our in-house database which includes the protein sequences from 4,830 publicly available environmental metagenomic datasets (human-associated samples were excluded). If one sample contained at least two signature proteins, the reads would be trimmed by Sickle v1.33 (https://github.com/najoshi/sickle) and assembled using MEGAHIT (Li et al., 2016) with the following parameters: min kmer 31; max kmer, 149; and kmer step 6. The reads were aligned to the contigs using Bowtie 2 (Langmead and Salzberg, 2012). A BamM (v.17.3, https://github.com/ecogenomics/BamM) filter was used to screen the reads mapped to the contigs with coverage ≥90% and identity ≥95%. The average per-base-pair coverage of the contigs was calculated by BamM “parse” with the parameter “tpmean” to remove the top and bottom 10% coverage regions. The contigs longer than 2.5 kb were used to calculate contig coverages and perform binning.

The protein sequences were predicted from the contigs using Prodigal (Hyatt et al., 2010) and then aligned with the signature proteins. The t-SNE algorithm-based dimension reduction from the tetranucleotide frequency matrix and visualization were performed using the R package mmgenome2 (https://github.com/KasperSkytte/mmgenome2). The signature protein-containing contigs were highlighted when visualization and manual binning were adopted to carefully divide the boundaries between metagenome-assembled genomes (MAGs). Rebinning was performed by plotting all the contigs from each bin based on their abundances plus GC content, and contigs with inconsistent coverage were manually deleted. The phylogeny of the MAGs was checked by GTDB-tk (Chaumeil et al., 2019), and the MAGs assigned as “UBA2009” were retained. Completeness and contamination of the MAGs were assessed using CheckM (Parks et al., 2015), and the MAGs with both >30% completeness and <10% contamination were finally retained.

The clean datasets of SAMN03609707 and SAMN03609696 were downloaded from the NCBI SRA database and trimmed from raw reads with Trimmomatic (Bolger et al., 2014) for further analysis by removing adapter sequences, low-quality reads (Q value <20), ambiguous “N” nucleotides, and fragments of <35 bp. Then, the ribosomal reads were identified and removed using SortMeRNA (Kopylova et al., 2012). The non-ribosomal reads were mapped to the coding regions of their corresponding MAGs (SRR2046235_bin1 and SRR2046236_bin1) using Bowtie 2 (Langmead and Salzberg, 2012). The coverages of the genes were calculated by the same methods in metagenomic analysis.

A small colony C2-1 was picked from the MMC-alkane-agar plates and then inoculated on R2A-agar plates, which formed clear, round colonies with regular, slightly raised edges. The transmission electron micrograph (TEM) showed that the cell of strain C2-1 was 1.15–1.52 μm in length, 0.19–0.37 μm in width, and 1.43–2.78 μm in flagellar length (Supplementary Figure S1). Cultured in the R2A medium, the optimum growth pressure for strain C2-1 was 0.1 MPa, with tolerance up to 50 MPa, indicating that C2-1 was a piezotolerant strain (Supplementary Figure S2).

The major cellular fatty acids of strain C2-1 (>5.0%) included Summed Feature 3 (C_16:1ω7c and/or C_11:1ω6c; 50.9%), C_16:0 (23.8%), and Summed Feature 8 (C_18:1ω7c and/or C_18:1ω6c; 8.0%; Supplementary Table S1). The polar lipids of strain C2-1 were composed of phosphatidylglycerol (PG) and phosphatidylethanolamine (PE) (Supplementary Figure S3, Supplementary Table S1). The respiratory quinone of strain C2-1 was identified as ubiquinone Q-9.

The alignments of 16S rRNA gene sequences showed that strain C2-1 is closely related to Thalassolituus alkanivorans TMPB967, Thalassolituus marinus IMCC1826, and Venatorbacter cucullus ASxL5^T, with identities of 100, 96.98, and 96.29, respectively (Supplementary Table S2). However, the 16S rRNA similarity between TMPB967 and the Thalassolituus type strain T. oleivorans MIL-1^T was only 95.74%, suggesting that the genus Thalassolituus might be non-monophyletic.

Considering that Venatorbacter cucullus ASxL5^T was reported with heterogeneous 16S rRNA sequences, we obtained the complete genome sequences of strain C2-1 (Table 1, Supplementary Figure S4) and constructed a phylogenetic tree based on 120 conserved protein sequences (known as GTDB taxonomy). It showed that strain C2-1, TMPB967, IMCC1826, TTBP476, and 41 metagenome-assembled genomes (see below) were classified as “UBA2009” in GTDB taxonomy and formed a branch together with Venatorbacter cucullus ASxL5^T, a recently reported Oceanospirillales predator isolated from a bovine slurry tank. This branch was divided from the Thalassolituus strains, T. oleivorans MIL-1^T, R6-15, 4BN06-13, and K188 (Figure 1).

Moreover, the GC contents of these “UBA2009” genomes (including MAGs) were 52.43%−56.12%, which are higher than those of Thalassolituus strains (45.52%−46.62%; Supplementary Table S3). In addition, some components of cellular fatty acids in strain C2-1 and IMCC1826 were also distinct from those of neighboring genera. For instance, C_12:0 was one of the major fatty acids in T. oleivorans DSM 14913^T, O. marinus DSM 24913^T, and O. kriegii IFO 15467 (5.7%−12.1%) but was only 1.3% and 1.9% in strain C2-1 and IMCC1826, respectively, whereas C_12:1 3-OH was not detectable in other three strains but was up to 3.9 and 7.8% in strain C2-1 and IMCC1826, respectively (Supplementary Table S1). These pieces of evidence suggest that strain C2-1, TMPB967, ST750PaO-4, IMCC1826, and TTBP476 should be classified as members of Venatorbacter, and seven clades (Clade I-VII) of Venatorbacter were further proposed based on their phylogeny (Figure 1). The ANI and DDH values among the seven clades of Venatorbacter and Thalassolituus species are shown in Supplementary Tables S4, S5, respectively.

To further study the phylogenetic markers of the genus Venatorbacter, the 120 conserved protein sequences from Venatorbacter and the neighboring genera were aligned. One deletion with two amino acid residues in leucyl-tRNA synthetase (TIGR00396) was only identified in the genus Venatorbacter (Supplementary Figure S5). Moreover, three signature proteins (SPs) of Venatorbacter (FOT50_RS00155, FOT50_RS08930, and FOT50_RS12870 in C2-1) were identified and were found in all the Venatorbacter genomes but absent in any other microbes in the NCBI nr database. Such SPs as molecular markers could be used to search for Venatorbacter species from large-scale metagenomic datasets.

To study the environmental distribution of bacteria in this novel genus, we employed the aforementioned signature proteins of Venatorbacter to search against the NCBI nr database and our in-house database, including the protein sequences predicted from ~5,000 metagenomic assemblies. As many as 41 MAGs of Venatorbacter were retrieved, which accounted for 0.1%−15.82% of the total reads (Supplementary Table S6). We also collected 16 16S rDNA amplicons which showed >97% identities with isolated Venatorbacter strains (Supplementary Table S6). In addition, the sampling information of strain MT-18 was also included, which we recently isolated from sediment at 11,000 m of the Mariana Trench (unpublished data).

The environmental distribution of Venatorbacter is shown in Figure 2. It is clear that the genus Venatorbacter is widely distributed in various environments, including seawater (38 of sample locations), marine sediments (3), hydrothermal vent plumes (2), Antarctic ice (1), groundwater (13), and marine sponges (1) (Supplementary Table S6). Their distribution in the marine environments ranged from surface seawater to the deepest place on earth (11,000 m). Notably, the Venatorbacter MAGs were the most abundant microbes in samples from the Von Damm hydrothermal vent plume at a depth of 2,041 m (SAMN03609667) and 2,238 m (SAMN03609666) but were absent in the reference seawater sample from 2,337 m (SAMN03609668), indicating their important ecological function in the hydrothermal vent ecosystems, which will be further discussed below.

In addition, different Venatorbacter clades showed distinct distributions. Clades I and II were only observed in marine water less than 800 m in depth; Clades III (including strain IMCC1826^T) and IV were only found in coastal waters and marine sediment, respectively, and Clade V was found in groundwater and coastal water. Interestingly, Clade VI (including C2-1, TMPB967, TTBP476, ST750PaO-4, MT-18, and six MAGs) was only observed in the marine water, sediments (>750 m), and groundwater from Aspo HRL located in the southeast of Sweden (Lat N 57 26 4 Lon E 16 39 36) at depth of 415.19 m (GCA_018826465.1), suggesting that strains in Clade VI could adapt well to and play an important role in the deep biosphere.

Given its wide distribution in the environment, we infer that Venatorbacter has diverse metabolic capabilities and possibly a wide range of environmental adaptation strategies. The reconstructed metabolic pathways of strains C2-1 and IMCC1826 showed the potential utilization of various carbon sources, including carbohydrates (fructose, sucrose, and glycerate), amino acids, oligopeptides, and phospholipids (Figure 3). In addition, the utilization of additional carbohydrates and organic acids was tested by Biolog GEN III and API 20NE, including glucose, mannose, xylitol, L-arabinose, D-arabitol, γ-amino butyric acid, and urocanic acid (Supplementary Table S7). In addition, the genes of type II secretion system (T2SS), preprotein translocation system, and many signal peptide-fused enzymes were identified in strains C2-1 and IMCC1826, suggesting that Venatorbacter were able to degrade extracellular components as carbon sources (see discussion).

Moreover, we found that C2-1 and most Venatorbacter genomes (including MAGs) retained two genes of non-heme diiron integral membrane n-alkane monooxygenases (alkB), alkB-dependent rubredoxin (alkG), and rubredoxin reductase (alkT) (Williams and Austin, 2022), suggesting that they could utilize n-alkane as carbon sources. Our results showed that ~50% of the n-alkanes were degraded at surface conditions (20°C and 0.1 MPa) after 10 days, and ~25% were degraded at in situ conditions (4°C and 58 MPa) when strain C2-1 was cultured with C₁₆-C₁₈ n-alkanes as the sole carbon source (Figure 4), indicating that Venatorbacter species could be n-alkane degraders under both surface water and deep-sea conditions.

For nitrogen metabolism, various transporters and metabolic pathways of nitrogen substrates were predicted in strain C2-1 and IMCC1826^T, including ammonia, putrescine, urea, ethanolamine, methylamine, nitrate, nitrite, and cyanate (Supplementary Table S7). Interestingly, a complete pathway of denitrification was identified in almost all Venatorbacter genomes (including MAGs), comprising nitrate reductase (narGHI), nitrite reductase (nirS), nitric oxidase reductase (norBC), and nitrous oxide reductase (nosZ). Denitrification by strain C2-1 was further tested by our experiment (Supplementary Table S7), and similar results were obtained on IMCC1826^T in a previous study (Choi and Cho, 2013). Conversely, all the genomes from the neighboring genera (Oleibacter, Thalassolituus, Oceanobacter, Bacterioplanes, Oleispira, and Bermanella) were predicted to be deficient in denitrification. Considering that the expression of narGHI is typically induced at very low oxygen concentrations (Hoffmann et al., 2021), our results suggested a genus-specific ecological function of Venatorbacter in suboxic or anoxic environments, e.g., the oxygen-minimal zone (OMZ), where ST750PaO-4 was isolated from (BioSample: SAMN15097647), and the deeply buried marine sediments, where C2-1 and MT-18 were isolated from.

To reveal how Venatorbacter species in Clade VI adapt to the deep ocean environments, we studied the genes in C2-1 that were potentially involved in adaptation to stresses like high hydrostatic pressure (HHP) and low temperature (Supplementary Table S8). Previous studies have shown that betaine is one of the compatible solutes commonly found in microorganisms for osmoprotection, salt stress, and cold stress protection and has been found to be accumulated in deep-sea bacteria under HHP (Martin et al., 2002; Methe et al., 2005; Rath et al., 2020). In this study, we found that the genome of C2-1 contained genes encoding betaine aldehyde hydrogenase, choline dehydrogenase, choline transporting subunits, and a Bet I family regulator which regulates the biosynthesis of betaine from choline-containing compounds. A gene of the BCCT family transporter was also identified from its genome, which controls the input of betaine and other compatible solutes from the environment. The gene cluster (ectABC) encoding the enzymes for biosynthesis of ectoine (1,4,5,6-tetrahydro-2-methyl-4-pyrimidinecarboxylic acid) was detected in the genome of strain C2-1. Ectoine serves as a protectant in many bacterial cells against stresses such as high salinity, low temperature, and high temperature (Bursy et al., 2008; Graf et al., 2008; Ma et al., 2017). In addition, it plays an important role in stabilizing enzymes, DNA, and cytoplasmic membranes, which are prone to be affected at high pressure.

It has been shown that β-hydroxybutyrate (PHB) could protect deep-sea bacteria from high-pressure-induced protein structure changes (Martin et al., 2002). Although no biosynthetic gene of PHB was identified in C2-1, the genes encoding PHB depolymerase family esterase and D-(-)-3-hydroxybutyrate oligomer hydrolase were found, suggesting that strain C2-1 could potentially utilize PHB as an osmotic regulator in adaptation to variable osmotic environments by adjusting intercellular PHB molecular weight. In addition, strain C2-1 contained many chaperone genes including hslR, dnaK, groEL, and cspD. These chaperones were found to be induced at high pressure and/or low temperature, to help maintain protein folding (Simonato et al., 2006; Oger and Jebbar, 2010).

To date, four types of enzymes in the terminal oxidation of aliphatic hydrocarbons have been found in Alcanivorax, including alkB in C_{5 − 18} alkane oxidation (Liu et al., 2011; Wang and Shao, 2013), the soluble cytochromes P450 (CYP153) in C_{8 − 16} oxidation (Liu et al., 2011), flavin-binding monooxygenase almA in C_{22 − 36} oxidation (Liu et al., 2011), and flavin mononucleotide-binding monooxygenase ladA in C_{15 − 36} oxidation (Li et al., 2008). Many hydrocarbonoclastic species employ multiple n-alkane oxidases to utilize different alkanes, e.g., Alcanivorax borkumensis SK2 possesses two alkB, three cyp153 (P450), and one almA genes (Schneiker et al., 2006).

In this study, we only identified alkB genes in the Venatorbacter genomes, and the long alkane transporters reported in T. oleivorans MIL-1^T (Gregson et al., 2018) were not identified in Venatorbacter. These lines of evidence indicate that Venatorbacter has a narrowed n-alkane utilization capacity and could be a non-obligate n-alkane degrader. Such non-obligate characteristics were also reported in the Ketobacter bin 12 (“Alcanivorax bin 12” in the original article), which was the most abundant microbe in near-bottom water of the Mariana Trench (10,500 m), and potentially could utilize n-alkanes, carbohydrates, amino acids, and steroids (Liu et al., 2019).

It was reported that increased HHP selectively inhibited obligate hydrocarbon-degraders and downregulated the expression of β-oxidation-related proteins (i.e., the main hydrocarbon-degradation pathway), resulting in low cell growth and CO₂ production (Scoma et al., 2019). Metabolically, HHP induces the accumulation of citrate and dihydroxyacetone, suggesting reduced rates of aerobic oxidation of fatty acids in the TCA cycle at HHP (Scoma et al., 2019). Therefore, the utilization of other carbon sources not only alleviates the accumulation of β-oxidation products at HHP but also increases the availability of nutrients, which might especially benefit the survival of Clade VI species of Venatorbacter.

To understand the most predominant ecological function of Venatorbacter, we studied the metatranscriptomic datasets from the aforementioned Von Damm hydrothermal vent plumes where Venatorbacter species were highly enriched (SAMN03609707 from 2,041 m and SAMN03609696 from 2,238 m, respectively). There were 26.55 and 4.5% of non-ribosomal RNA metatranscriptomic reads aligned to the corresponding MAGs (SRR2046235_bin1 and SRR2046236_bin1), respectively. This result suggests that Venatorbacter species were not only abundant but also greatly active in the deep-sea hydrothermal vents.

For the sample SAMN03609707, 21.86% of the 4,054 protein-coding genes of the Venatorbacter MAG showed more than the average transcriptional depth (73.5×). As for metabolisms, although the alkB genes were identified in the MAG SRR2046235_bin1, their expression was not detected in situ (<2×), suggesting that Venatorbacter species might not be a main n-alkane degrader in the Von Damm hydrothermal vents. However, the transcripts of carbohydrate-utilizing enzymes were identified, including D-glycerate kinase (EC: 2.7.1.31), fructose kinase (EC: 2.7.1.4), and beta-N-acetylhexosaminidase (EC: 3.2.1.52). In addition, the transcripts of ABC transporters of phospholipid (mlaABCDE), nucleoside (bmpA and nupABC), betaine/proline (proXWV), lysine/arginine/ornithine/histidine/octopine (pa5152–5155), and oligopeptide (oppABCDEF) were also identified. It suggested that Venatorbacter species have the capability of utilizing a wide variety of non-alkane organic substrates in deep-sea hydrothermal vents. More importantly, the type II secretion system (T2SS) is a multi-protein complex used by many bacteria to move substrates across their cell membrane, which spans the bacterial cell envelope and extrudes substrates through an outer membrane secretin channel using a pseudopilus (Naskar et al., 2021). The mshA and mshI involved in T2SS showed expression of >1,000×, and secA and secY involved in preprotein translocation exhibited expression of >700×, suggesting that protein export of Venatorbacter species was highly active in deep-sea hydrothermal vents (Supplementary Table S9). Accordingly, signal peptide-fused M6 family metalloprotease, peptidoglycan DD-metalloendopeptidase, phospholipase, and endonuclease were robustly expressed (>200×) in this MAG (SRR2046235_bin1). In particular, such extracellular M6 metalloprotease showed an expression of 679×, which is higher than most ribosomal proteins (Supplementary Table S9). Our experiments further confirmed that both strain C2-1 and IMCC1826 could grow with peptone as the sole carbon source (Supplementary Figure S6). These lines of evidence illustrated that Venatorbacter species may have a significant contribution to secondary productivity by degrading various kinds of extracellular macromolecules.

In addition, most of those genes involved in the protein secretion system and extracellular enzymes are conserved in the streamlined genome of V. cucullus ASxL5^T (with merely ~2/3 genome size of strain C2-1), suggesting that strain ASxL5^T specialized to be a bacterial predator via a massive gene loss. Considering clade VI and clade VII Venatorbacter species are closely related in phylogeny, strain C2-1 with the first complete genome of putatively non-predative Venatorbacter can be used in a comparative study with predative strains ASx5O and ASxL5 to uncover the genome reduction from a free-living ancestor to a predator in future.