Ecology and trophic role of Oncholaimus dyvae sp. nov. (Nematoda: Oncholaimidae) from the lucky strike hydrothermal vent field (Mid-Atlantic Ridge)

and Abstract Background: Nematodes are an important component of deep-sea hydrothermal vent communities, but only few nematode species are able to cope to the harsh conditions of the most active vent sites. The genus Oncholaimus is known to tolerate extreme geothermal conditions and high sulphide concentrations in shallow water hydrothermal vents, but it was only occasionally reported in deep-sea vents. In this study, we performed morphological, genetic and ecological investigations (including feeding strategies) on an abundant species of Oncholaimus recently discovered at Lucky strike vent field on the Mid-Atlantic Ridge at 1700 m water depth. Results: We described this species as Oncholaimus dyvae sp. nov.. This new species differs from all other members of the genus by the combination of the following characters: body length (up to 9 mm), the presence of a long spicule (79 μ m) with a distally pointed end, a complex pericloacal setal ornamentation with one precloacal papilla surrounded by short spines, and a body cuticule with very fine striation shortly posterior to the amphid opening. Overall, O. dyvae sp. nov. abundance increased with increasing temperature and vent emissions. Carbon isotopic ratios suggest that this species could consume both thiotroph and methanotrophic producers. Furthermore sulfur-oxidizing bacteria related to Epsilonproteobacteria and Gammaproteobacteria were detected in the cuticle, in the digestive cavity and in the intestine of O. dyvae sp. nov. suggesting a potential symbiotic association. Conclusions: This study improves our understanding of vent biology and ecology, revealing a new nematode species able to adapt and be very abundant in active vent areas due to their association with chemosynthetic micro-organisms. Faced by the rapid increase of anthropogenic pressure to access mineral resources in the deep sea, hydrothermal vents are particularly susceptible to be impacted by exploitation of seafloor massive sulfide deposits. It is necessary to document and understand vent species able to flourish in these peculiar ecosystems.


Background
Deep-sea hydrothermal vents are unique and severe environments. Hydrothermal fluids are formed by cold sea water which infiltrates oceanic crust. These fluids are heated and enriched with reduced chemicals sorting at very high temperatures (> 400°C, [1]). Vent ecosystems are formed by organisms able to cope with these extreme conditions (high concentrations of reduced compounds, heavy metals and radionuclides, low oxygen level, and elevated temperatures [2,3]). Nevertheless the severity of this environment, vents are particularly richer (in term of biomass and productivity) than the adjacent deep-seafloor [4]. This elevated macromegafaunal density results from the obligate exploitation of a localized food source that is produced primarily by microbial chemolithoautotrophy, which in turn is directly dependent on reducing substances from vent fluids [5,6]. Hydrothermal vent fauna interacts directly with microorganisms either through symbiosis, which provides nutrition for the most dominant large invertebrate species (e.g., siboglinid tube worms, bathymodiolin mussels, vesicomyid clams, shrimps [7]), through direct grazing on the microbial communities [6], or, in the case of species like the crab Xenograpsus testudinatus [8], indirectly, as they feed on microbial grazers.
Nematodes are an important component of hydrothermal vent communities worldwide in term of abundance and biomass [9][10][11]. Hydrothermal vent nematode communities are usually composed by a dozen of nematode taxa and dominated by two or three of them [4,9,10,[12][13][14][15][16]. Mussel bed fields of the Lucky Strike vent field (Mid-Atlantic Ridge) are usually dominated by two nematode genera: Cephalochaetosoma and Halomonhystera [9]. The genus Oncholaimus (only occasionally found in deep-sea vents [13]) was recently reported in deep-sea Atlantic hydrothermal vents [10,17]. Both studies showed that Oncholaimus can be very abundant, reaching remarkable biomass values for the total nematode community due to its big size [10].
Nevertheless Oncholaimus is an important component of vent communities, information about diversity, trophy and ecological role of this group is almost unknown. In this study, we performed morphological and genetic investigations on this abundant species of Oncholaimus reported at Lucky strike vent field on the Mid-Atlantic Ridge at 1700 m water depth (Fig. 1), describing it as a new species for science. We also analyzed its abundance, biomass and diet and we explored its ecological role in the Lucky Strike hydrothermal vent field.

Measurement
See Table 3.

Etymology
This species is named in honour of the project DYVA (Deep-sea hYdrothermal Vent nematodes as potential source of new Antibiotics) that supported this study.

Males
Body length 9430 μm, maximum diameter 95 μm, with a smooth cuticle (Fig. 2a, Table 1). Habitus very long and slender, slightly anteriorly tapered, but more pronounced in tail region; body cuticle smooth in light microscopy but showing fine striation shortly posterior to the amphid opening (see Fig. 3c and Fig. 4f).

Females
Very similar to males in appearance, body length up to 8598 μm, maximum body diameter 139 μm (Fig. 2a). No caudal setae. Single anterior ovary, reflexed (Fig. 2b). Gravid female with 2 fertilized eggs (217 μm long). Vulva (Fig. 4e) at 64-69% of body length from anterior end. Demanian system with uvette (connection between main duct and posterior end of uterus) clearly visible (Fig. 5d) Fig. 1 Study area and sampling sites. a Location of the Lucky Strike vent field on the Mid-Atlantic Ridge at 37°17 N, 32°16 W. b Samples were collected using the ROV Victor6000 manipulator arm. c Locations of the sites where different substrata were deployed around the Eiffel Tower edifice (sites 1, 2, 3 and 4) and sites from which Bathymodiolus azoricus assemblages were sampled (Eiffel Tower, Cypress, Y3) are reported and connected to exterior through lateral openings or copulation pores (Fig. 4f).

Differential species diagnosis
Oncholaimus dyvae sp. nov. differs from all other members of the genus by the combination of the following characters: body length (up to 9 mm), the presence of a long spicule (79 μm) with a distally pointed end, a complex pericloacal setal ornamentation with one precloacal papilla surrounded by short spines, and a body cuticule with very fine striation shortly posterior to the amphid opening. According to the World Register of Marine Species, Oncholaimus accommodates 126 species names of which 85 are considered valid species. These species can be clustered in accordance with their dimensions (most Oncholaimus species are less than 5 mm long). Only [17], which previously identified this hydrothermal vent species as O. scanicus, we consider that the severe disparity of habitats, together with evident morphological differences (conspicuous pre-cloacal midventral papilla, two postcloacal lateral rows of six short conical elevations) are strong characters to separate these species.

DNA taxonomy
The maximum likelihood (ML) tree shows that Oncholaimus dyvae sp. nov. forms a clade and clusters with Oncholaimus sp. (accession numbers HM564402, HM564475, AY854196, LC093124, KR265044), Viscosia sp. (accession number FJ040494) and other Oncholaimidae (accession number KR265043, FJ040493, AY866479, HM564620, HM564605) (Fig. 6). The genera Oncholaimus seems to be polyphyletic, but this is maybe due to artefactual misidentification of specimens. The sister group of the new species is not clear because of a poor support and resolution. Moreover, it should be stressed that very few species of Oncholaimoidea are represented in GenBank. DNA taxonomy through Poisson Tree Process (PTP) provides a total of 29 units (from ML solution) and 30 units (from Bayesian solution) and corroborates the fact that all the individuals of O. dyvae sp. nov. belong to one single species, which is also different from any other species already present in GenBank (Fig. 6).

Oncholaimus dyvae sp. nov. abundance and biomass
No Oncholaimus dyvae sp. nov. was recorded from the organic and inorganic substrata located at site 4 (a sediment area located between the Eiffel Tower and Montsegur edifices), and very low abundance and biomass were reported at site 1 (4.  (Fig. 7a). Regarding the effect of temperature, a quadratic relationship

Carbon and nitrogen stable isotope ratios
In all three Eiffel Tower samples, Oncholaimus dyvae sp. nov.'s δ 15 N was lower than the one of photosyntheticderived organic matter (Table 3, Fig. 8), and markedly higher than the one of Bathymodiolus azoricus tissues (over 12‰; Table 3, Fig. 8). δ 15 N values of O. dyvae sp. nov. were 3.9 to 5.8 ‰ higher than those of Beggiatoa bacterial mats. Moreover, analysed O. dyvae sp. nov. specimens were clearly more 13 C-enriched than Beggiatoa mats or B. azoricus tissues (up to 6 ‰, Table 3). Interestingly, isotopic ratios of O. dyvae sp. nov. fluctuated from one sample to another, and those fluctuations closely matched those of their bivalve hosts (Table 3, Fig. 8), as shown by almost identical net differences between O. dyvae sp. nov. and B. azoricus muscle for both isotopic ratios and in all 3 samples (Table 3).

Discussion
Ecology of Oncholaimus dyvae sp. nov. in deep-sea hydrothermal vents Species of Oncholaimus have been reported around the world in association with shallow-water hydrothermal vents (3 m water depth; [19]). Some Oncholaimus species tolerate extreme geothermal and hypersaline conditions as well as high sulfur concentrations. One example is the species Oncholaimus campylocercoides found in hydrothermal sources of the Aegean, Baltic and North Seas [20]. This species can produce secretions containing sulfur when exposed to hydrogen sulfide, thereby potentially reducing its toxicity. It was hypothesized that the accumulation of elementary sulfur also provides an energy "reserve" for subsequent oxidation into thiosulfate, sulfite, or sulfate in normoxic conditions [20] although no evidence has been provided so far. To date, large nematodes belonging to Oncholaimus have rarely been reported from deep-sea hydrothermal vents [13,17]. The genus Oncholaimus was for the first time reported in very high abundance at the Lucky Strike vent field [10]. Also Tchesunov [17] reported Oncholaimus from two other deep-sea hydrothermal vents of the Atlantic Ocean (Menez Gwen and Lost City, Mid Atlantic Ridge). In the present study, we reported a high density of Oncholaimus and we describe the new species Oncholaimus dyvae sp. nov. from the organic colonization substrata deployed at the most active sites around the Eiffel Tower [10]. We also reported Oncholaimus dyvae sp. nov. in very high abundance associated with Bathymodiolus assemblages at Eiffel Tower while their abundances were much lower at Cypress and Y3. Overall, O. dyvae sp. nov. abundance increased with higher temperature and vent emission (Fig. 7) and no individual was found at the inactive colonization site ( Table 2). Such data seems to indicate that the distribution of O. dyvae sp. nov. is linked with the presence of hydrothermal activity although not at all active sites. Indeed, in a recent paper on Eiffel Tower faunal assemblages, no Oncholaimus was reported from the 12 sampling units [9]. However, the presence of another genus belonging to the family Oncholaimidae (Viscosia) was reported in one of their samples [9]. In terms of biomass, our values are several orders greater than deepsea nematode fauna, but comparable to the Condor seamount nematofauna, where high nematode biomass values (due to the presence of large Comesomatidae nematodes) were recorded at the seamount bases [21].
The δ 15 N of Oncholaimus dyvae sp. nov. was lower than the one of photosynthetic-derived organic matter. While this food source could partly contribute to nematode diet, it is therefore unlikely to be a major food item. Moreover, despite the fact that O. dyvae sp. nov. was found in extraordinarily high densities associated to Bathymodiolus azoricus' byssus, the very substantial δ 15 N difference (over 12‰) between the nematode and its bivalve host rules out the existence of a direct trophic link between them. We can hypothesize that the byssus could act as shelter offering camouflage and protection from predation or simply as a physically suitable threedimentional substratum in an environment where most substrates are bare (i.e. no sediment), particularly near the vents and sources of emissions. In addition, the byssus may act as a trap for organic matter. Nitrogen stable isotope ratios suggest that Beggiatoa bacterial mats could constitute a feasible food source for O. dyvae sp. nov. Moreover, δ 13 C and δ 15 N of O. dyvae sp.nov. were comparable with those of other detritivore animals sampled at the Lucky Strike vent, such as the polychaete Amphisamytha lutzi, or the gastropods Protolira valvatoides and Pseudorimula midatlantica [22]. The fact that fluctuations in stable isotope ratios of C and N in O. dyvae sp. nov. closely match those observed in the tissues of  endosymbiotic mussels B. azoricus could nevertheless indicate that some kind of nutritional relationship exists between the nematode and its bivalve host. The nature of this relationship remains an open question, but could involve the nematode feeding on bivalve-associated bacteria. These results, together with the fact that no potential preys were identified in the surrounding food web [22], suggest that O. dyvae sp. nov. is a detritivore/bacterivore, which partly relies on free-living chemoautotroph microbes. Moreover, O. dyvae sp. nov.'s δ 13 C was clearly higher than Beggiatoa mats or B. azoricus tissues. This clearly indicates that CBB thiotrophs are not the nematode's sole food source. It could instead rely on a mix of thiotrophs and methanotrophs micro-organisms, explaining its intermediary δ 13 C values. Furthermore, sulfuroxidizing bacteria related to Epsilonproteobacteria and Gammaproteobacteria were detected in the cuticle, in the digestive cavity and in the intestine of O. dyvae sp. nov. suggesting a potential symbiotic association [23]. More research is needed to evaluate the relative importance of both of these groups in the nematode diet. The polynoid annelid Branchinotogluma mesatlantica has been identified as a potential predator of O. dyvae sp. nov. (unpublished data), but more evidence is needed to validate this hypothesis. As recently shown for Oncholaimus moanae, nematodes can be a high quality food source to predators thanks to their high amount of highly unsaturated fatty acids (HUFAs) [24]. In the deep sea and hydrothermal vents, such source may play an even more important role for food webs as basal sources have usually low amount of HUFAs [25,26]. Unlike most metazoans, nematodes have been shown to be able to biosynthesize HUFAs from acetate [27,28]. Often neglected for their size, nematodes can thus represent an important trophic component of vent communities.

Conclusions
This study improves our understanding of vent biology and ecology, revealing a new nematode species able to adapt and be very abundant in active vent areas due to their association with chemosynthetic micro-organisms. Faced by the Fig. 7 Relationship between Oncholaimus dyvae sp. nov. abundance and environmental conditions. a Relationship between Oncholaimus dyvae sp. nov. abundance (expressed in logarithmic scale) and substratum. b Relationship between Oncholaimus dyvae sp. nov. abundance (expressed in logarithmic scale) and habitat temperature for each of the seven analysed sites, including model fit with 95% confidence interval rapid increase of anthropogenic pressure to access mineral resources in the deep sea, hydrothermal vents are particularly susceptible to be impacted by exploitation of seafloor massive sulfide deposits [29]. It is necessary to document and understand vent species able to flourish in these peculiar ecosystems.

Methods
The study area and sampling collection The Lucky Strike vent field is situated on the Mid-Atlantic Ridge, south of the Azores (Table 4 and Fig. 1). This vent field is composed by three volcanic cones enclosing a lava lake (with a diameter of 200 m [30]). It hosts around 20 active edifices including "named as" Eiffel Tower, Cypress and Y3. Eiffel Tower and Y3 can be found on the eastern side of the field, while Cypress is located on its western side. Eiffel Tower is an active edifice characterized by an eight of 11 m. In the Lucky Strike vent field, Eiffel Tower is the most studied edifice. Samples were collected during three oceanographic cruises: Momarsat 2011 (29/06-23/07/2011), Biobaz 2013 (2-20/08/2013) and Momarsat 2014 (13-31/07/2014) on the research vessel "Pourquoi Pas? ". Sample collection was carried out using the remotely operated vehicle (ROV) Vic-tor6000. During the Momarsat 2011 cruise, samples for nematode morphological identification, abundance and biomass analyses were collected from inorganic (slates reported as A), and organic (woods reported as B and bones reported as C) colonisation substrata [10]. At the Eiffel Tower we performed a long-time experiment involving the deployment of these substrates at increasing distances from inactive to active hydrothermal vent are (Sites 1, 2, 3, 4, see details in [10]). The sites less active were site 1 (3 m from the edifice) and 3 (4-5 m from the edifice) located at the base south of the Eiffel Tower. Site 1 was caracterized by very few organisms while site 3 was on a crack with diffuse venting, where Bathymodiolus azoricus mussels and microbial mats were recorded. The most active site was site 2 located on the north-west flank of Eiffel Tower and near to fluid exits surrounded by dense Bathymodiolus azoricus assemblages. Site 4 was an external sedimentary site located between the Eiffel Tower and Montségur edifices. The other nematode samples came from a broad-scale study on the structure of Bathymodiolus azoricus assemblages at Lucky Strike initiated by Sarrazin et al. in 2012 (unpublished data) ( Table 4 and Fig. 1). These assemblages were collected from the Eiffel Tower, Y3 and Cypress edifices   (Table 4 and Fig. 1) during Biobaz 2013 and Momarsat 2014 cruises. The fauna was sampled using Victor's suction sampler and arm grab following the protocol described in Cuvelier et al. [31]. Once brought on board, faunal samples from each location were washed over stacked sieves (1 mm, 250 μm and 63 μm mesh size) and stored with filtered seawater at 4°C temperature. Bathymodiolus specimens were individually carefully washed over sieves and, together with the sieves, byssus was checked at stereomicroscope.

Nematode sorting and fixation
Nematodes were sorted directly on board of the research vessel under a stereomicroscope from the colonisation substrata or from the mussel assemblages. We selected one of the most abundant species (previously identified as Oncholaimus sp.1 [10]). Due to its size (several millimeters), the species could easily be separated from the other species directly at the stereomicroscope. A set of specimens was fixed in 4% formaldehyde for morphological description. Another set of individuals was immediately frozen at − 80°C for molecular and stable isotope analyses. Other individuals were prepared for Scanning Electron Microscopy (SEM) studies: nematodes were fixed in 2.5% glutaraldehyde for 16 h at 4°C, then transferred in a sodium azide solution (0.065 g in 150 ml filtered sea water) and stored at 4°C until use.

Nematode morphological analysis
In order to confirm that all nematodes were conspecific, we performed a detailed morphological examination for a subset of the population. Several nematodes were mounted on slides for detailed morphological observations using the formalin-ethanol/glycerol method [32,33]. Drawings and photos were made with a Leica DM IRB inverted light microscope equipped with live-camera (Image-Pro software) and on Zeiss AxioZoom microscope equipped with live-camera (Zen software Nematode DNA extraction, PCR, and sequencing As an additional control to check that all nematodes in the population with the same morphology indeed belonged to the same species, we used a DNA taxonomy approach based on DNA sequence data from 42 individuals. DNA extraction was performed with Chelex®. Each nematode was first incubated in 35 μl in a 5% Chelex® solution supplemented with 1 μl of proteinase K for 1 h at 56°C followed by 20 min at 95°C. The sample was then vortexed for 15 s: the Chelex® solution contains styrene-divinylbenzene beads that help grind the tissues and release DNA. The samples were centrifuged and the supernatants were recovered and stored at − 20°C. The polymerase chain reaction (PCR) amplifications for 18S rDNA (the small subunit of ribosomal DNA) and for 28S rDNA (the large subunit of ribosomal DNA) were carried out in a final volume of 25 μl using following mix: 2 μl of extracted DNA was added to 5 μl of 5x PCR buffer and 0.1 μl Taq polymerase (5 U/μl -Promega). For the 18S rRNA, 10 mM of each dNTP, 62.5 mM of MgCl 2 and 10 μM of each of the two primers were added. For the 28S rDNA, 5 mM of each dNTP, 50 mM of MgCl 2 and 20 μM of each of the two primers were added. The following primers were used: 18S1.2a (5′ -CGATCAGATACCGCCCTAG -3′), 18Sr2b (5′ -TACAAAGGGCAGGGACGTAAT-3′)), D2Ab (5′-ACAAGTACCGTGAGGGAAAGTTG-3′) and D3B (5′-TCGGAAGGAACCAGCTACTA-3′). The PCR cycles were 2 min at 94°C then 30 cycles of 1 min denaturation at 94°C, 1 min annealing at 55°C and 2 min extension at 72°C, followed by 10 min at 72°C. All amplification products were run on a 0.8% agarose gel to verify the size of the amplicons. Then, 20 μl of each PCR product were sent to GATC Biotech for sequencing. Each nematode was sequenced in both forward and reverse direction. We could obtain 18S sequences (592 bp) for 40 of the 42 animals and 2 sequences of 28S (607 pb). Chromatograms were checked using the FinchTV software package (©Geospiza Inc.), and all sequences were deposited in GenBank with accession number from KY451633 to KY451672.

DNA taxonomy
We aimed at testing whether the new morphologicaly identified species could be supported as one unique molecular evolutionary entity and second to test their novelty with available sequences (NCBI database). In order to achieve this, we used a DNA taxonomy approach [34], namely the Poisson Tree Process (PTP [35]). All 18S sequences belonging to the superfamily Oncholaimoidea were retrieved, belonging to 13 named genera. Only five sequences are identified to species level, corresponding to three species: Calyptronema maxweberi, Pontonema vulgare and Viscosia viscosa. All sequences present in NCBI overlapping with the amplified fragment were used in the analysis. In order to eliminate redundancy, if some sequences were identical to others, we selected only one for each group of identical sequences. Also, for the putative new species, we reduced the dataset to unique sequences only. All unique non-redundant sequences were aligned using MAFFT [36] with Q-Ins-i settings, suitable for ribosomal markers. To perform PTP [35], we used as input a maximum likelihood (ML) reconstruction in RAxML 8.2.2 (Randomized Axelerated Maximum Likelihood [37]). The alignment for RAxML included also seven outgroup sequences from closely related nematode groups: Dintheria tenuissima (Bastianiidae), Deontostoma magnificum (Leptosomatidae), Ironus dentifurcatus (Ironidae), Litinium sp. (Oxystominidae), Rhabdodemania sp. (Rhabdodemaniidae) and Metenoploides sp. and Thoracostomopsis sp., (Thoracostomopsidae). We used GRT + G + I as evolutionary model, with 500 bootstrap resampling. The outgroups were then removed before performing PTP.

Nematode biomass
The nematode biomass was calculated from the biovolume of all the individuals collected per replicate using the Andrassy formula (V = L × W 2 × 0.063 × 10 − 5 , where V is expressed in nL (10 − 9 L) with body length, L, and width, W, expressed in μm [38]). The carbon contents were identified as representing 40% of the dry weight [39]. From each sample, 200 collected nematodes were isolated using an AxiozoomV16 stereomicroscope (400× magnification) using the software Zen 2012 (blue edition). We did not perform any test on biomass values, similar to what we did for abundance values, given that biomass was strictly correlated to abundance (Pearson's correlation r = 0.99).

Statistical analyses
We tested whether the abundance of the investigated nematode species was influenced by substratum and by temperature. For the effect of substratum, a categorical variable with four levels, slate, wood, bone, or Bathymodiolus, sampled in different sites, was used as an explanatory variable in a nested Analysis of Variance (ANOVA): given the inherent pseudo replication in the sampling design, we included the confounding effect of site in the random structure. For the effect of temperature, we removed the confounding effect of site by averaging abundance values for each site before performing the test. We then tested, using ANOVA test between models, whether a linear or a quadratic relationship between abundance and average temperature for each site could be a better fit to the data. All tests were run in R 3.1.2 [40], and abundance data were used in the models after logarithmic transformation.

Carbon and nitrogen stable isotopes ratios
We studied feeding ecology of the selected nematode species by analyzing carbon and nitrogen stable isotope ratios. On board, Bathymodiolus azoricus mussels (gill, muscle and byssus), filamentous bacterial mats as well as nematode individuals from the different species associated with Bathymodiolus were frozen (− 80°C) for isotopic analyses. In the laboratory, samples were rinsed, after fixation, in distilled Milli-Q water. For nematodes, the whole body was used and several animals were pooled to reach the minimum weight required for stable isotope analyses (10 to 20 nematodes per sample). Samples were freeze-dried and ground into a homogeneous powder using a ball mill. Tissue was precisely weighed (0.4 ± 0.1 mg) in tin capsules. Samples were analysed on a Flash EA 1112 elemental analyser coupled to a Thermo Scientific Delta V Advantage stable isotope ratio mass spectrometer (EA-IRMS). Analytical precision based on the standard deviation of replicates of internal standards was ≤0.1‰ for both δ 13 C and δ 15 N. Values are expressed in δ (‰) notation with respect to VPDB (δ 13 C) and atmospheric air (δ 15 N): δX (‰) = [(R sample / R standard ) -1] × 10 3 , where X is either 13 C or 15 N, R sample is the 13 C/ 12 C or 15 N/ 14 N isotope ratio in the sample and R standard is the 13 C/ 12 C or 15 N/ 14 N isotope ratio for the VPDB standard (δ 13 C) or atmospheric air (δ 15 N).