Ophiotholia (Echinodermata: Ophiuroidea): A little-known deep-sea genus present in polymetallic nodule fields with the description of a new species

Introduction The Clarion Clipperton Zone (CCZ) in the Northeast Pacific Ocean holds the largest deposits of polymetallic nodules at abyssal depths. These nodules are rock formations containing valuable metals and minerals targeted for mining. They further provide diverse habitat for a range of deep-sea species. Little is known so far on the taxonomy, natural history and biogeography of these deep-sea animals which is vital for accurate assessment of the risk of species extinctions from large-scale mining. One of the most abundant megafaunal groups in the CCZ is the Ophiuroidea (brittle stars), of which Ophiotholia is one of the more abundant genera found in the area. The genus Ophiotholia has a world-wide distribution and currently holds six species. Methods Material collected from seven scientific cruises to the CCZ was examined, morphologically, together with comparative material from all the known species. The small size and the damage caused during sampling often impeded their identification. The specimens were also genetically analyzed using a fragment of the mitochondrial COI gene. Scanning Electron Microscope images of the key microstructural characters were made using selected specimens from CCZ as well as from the comparative material. Result and discussion One morphotype was identified as the known species Ophiotholia supplicans Lyman, 1880, while the second is new to science and is described in this paper. The umbrella spines and the arrangement of their articulations on the lateral arm plate, were selected as the most relevant morphological characters in the taxonomy of the genus Ophiotholia and a revised identification key of all characters from all known Ophiotholia species is provided as a table in the supplement material. The identification and description of such a little-known genus improves the evaluation of the biodiversity not only in the CCZ but also for the deep sea.

Among taxa collected in the CCZ, the class Ophiuroidea (Echinodermata) are widespread and abundant (Simon-Lledóet al., 2019b;Christodoulou et al., 2020). Ophiuroids inhabit all kind of habitats, from the tropics to the poles and from shallow waters to deep-sea ecosystems (O'Hara, 2007;Stöhr et al., 2012;Bribiesca-Contreras et al., 2019;Laming et al., 2021). Because of their global distribution, ophiuroids serve as excellent model organisms for the understanding of large-scale diversity and evolution patterns (Hunter and Halanych, 2008;Cho and Shank, 2010;Galaska et al., 2017;Galaska et al., 2019;Jossart et al., 2019;O'Hara et al., 2019;Stöhr et al., 2020;Eichsteller et al., 2022). The world ophiuroid database (accessed 13.09.2022) lists around 2100 described species of which 100 exclusively dwell on the abyssal plains (Stöhr et al., 2012). Christodoulou et al. (2020) identified 43 putative ophiuroid species in the CCZ using COI barcodes. The most abundant species were Ophiosphalma glabrum (Lütken & Mortensen, 1899) and Silax daleus (Lyman, 1879) followed by 3 unidentified species belonging to Ophiotholia spp. The genus Ophiotholia was erected by Lyman in 1880 to accommodate the type species of the genus, Ophiotholia supplicans, and was first placed in the family Ophiuridae. Although Lyman (1880) noticed the unique characters and body shape of this genus that set it apart from the other known genera of the family Ophiuridae, it was Perrier (1893) that created a new family, Ophiohelidae, to which Ophiotholia was assigned. Two more genera were then assigned to the family of Ophiohelidae: Ophiomyces Lyman, 1869 andOphiohelus, Lyman, 1880 with the latter being the type of the family. In 1915, Clark transferred all three genera to the family of Ophicanthidae Ljungman, 1867, ignoring Ophiohelidae and Ophiomycetidae, a new family created to include the same genera by Verrill, 1899. Fell (1960 and Paterson (1985), although agreeing with Verrill (1899) that the three genera are quite distinct, assigned the genera to the family of Ophiacanthidae with Paterson (1985) reestablishing the subfamily Ophiohelinae (Perrier, 1891 amended) to include them. Martynov (2010) reviewed the segregating characters and suggested that the three genera, Ophiohelus, Ophiomyces, and Ophiotholia, be transferred from the Ophiacanthidae to new family but it was O'Hara et al. (2018), who re-established the family of Ophiohelidae within the order Ophioscolecida after restructuring the higher taxonomy of the class Ophiuroidea based on genetic and morphological data. Ophioscolecida are unique in having reduced radial shields concealed by disc scales or being absent (O'Hara et al., 2018). Ophiohelidae can be easily separated from the only other family in the order, Ophioscolecidae Lütken, 1869, by their unique body form, which form is perfectly adapted to living buried in the soft sediment with their arms protruding outside for feeding. The most unique character within this family is the so called "parasol-spine" on the posterior part of the arms. The function of these spines is unclear, but Lyman (1880) described them as supplementary spines or pedicellaria. Because of the species-specific structure of these spines, they are often used as a species delimitation character in the two genera in which they occur (Ophiohelus and Ophiotholia).
Within the genus Ophiotholia there are six extant species, Ophiotholia spathifer (Lyman, 1879); Ophiotholia supplicans Lyman, 1880; Ophiotholia mitrephora H.L. Clark, 1910;Ophiotholia montana Litvinova 1981;Ophiotholia gibbosa Litvinova, 1992 and Ophiotholia odissea Litvinova, 1992. There is also one recently described extinct species Ophiotholia aurora Thuy and Meyer, 2013 that was found in the Hauptrogenstein Formation and coeval formations in Switzerland, dated to the middle Jurassic. Because all Ophiotholia species seem to be very rare and not well documented in the literature, the known distribution of the genus is limited between 60°N and 60°S (Lyman, 1882;Litvinova, 1992;Martynov, 2010). The depth in which the species have been collected previously varied between 216 m to 4100 m. Due to their fragile structure, they often get damaged during sampling (Litvinova, 1992), the parasol spines can be lost, and thus many specimens are only identified to higher taxonomic levels Christodoulou et al., 2019;Christodoulou et al., 2020).
The present study aims to describe a new species within the genus Ophiotholia as well as reviewing the existing species of the genus using revised morphological characters. Besides morphology, mitochondrial COI sequences were used to explore the genetic diversity of Ophiotholia in the CCZ area.
2 Material and methods 2.1 Specimen collection and processing material ( Figure 1 and Table S1). Ophiuroid specimens were collected during various cruises from the UKSRL, NORI-D and BGR licence areas, as well as the APEI3, using a Brenke-type epibenthic sledge (EBS) or Box Corer following standard deployment procedures (Brenke, 2005) and as described in detail by Christodoulou et al. (2020) for the EBS samples and by Glover et al. (2015) and Bonifaćio et al., 2020 for the Box core samples. The voucher specimens of the Ophiotholia spp. collected from the different CCZ areas are kept in the German Centre of Marine Biodiversity Research (Wilhelmshaven), the Senckenberg Research Institute and Natural History Museum Frankfurt (SMF) and the Natural History Museum London (NHM), while the type material from the other extant species of Ophiotholia are deposited at either the Natural History Museum in London, UK, or at the Zoological Museum of Moscow University, Russia (ZMMU).

Morphological analysis
Specimens were morphologically examined and photographed using a Leica M125 stereo microscope. Selected specimens were used for scanning electron microscope (SEM) analysis and essential micromorphological characters were recorded either with a VEGA3 TESCAN, CamScan II, and JSM-6380 SEMs after the samples were gold-palladium-coated. Images of the whole body were taken after the selected specimens were cleaned with distilled water in order to remove any detritus. Ossicles were isolated by macerating arms or portions of arms in a mixture of (1:1) distilled water and household bleach (NaClO). In general, disintegration of the arm skeleton was completed within one hour following immersion. The disarticulated arm plates were rinsed in distilled water to remove any traces of NaClO. After rinsing, the plates were air-dried, and the selected Map with the recorded distributions from all recent species of the genus Ophiotholia. The colour of the upper map indicates: orange, distributions recorded for species with just morphological data; yellow and turquoise, distributions recorded from species with morphological and molecular data. A unique combination of colour and symbol is assigned to each species. Detailed map of the Clarion Clipperton Zone showing the distribution of the cooccurring species Ophiotholia saskia sp. nov. and Ophiotholia supplicans. The distribution map of the Ophiotholia species was created using QGIS (version 3.16, http://www.qgis.org), considering the data of the International Seabed Authority (2020/2021). ossicles, as well as the bodies, were mounted on aluminum stubs for SEM examination using a glue foil. The orientation of the plates on the stubs was chosen to best display the relevant characters on a single picture per view (ad/abradial, distal/proximal, dorsal/ventral). An image of each character per individual were taken if possible. The scanned habitus and ossicles mount on the stubs as well as the remaining ossicles of the disintegrated arms were deposited in the German Centre for Marine Biodiversity Research, Senckenberg, Wilhelmshaven, or the Zoological Museum of Moscow University. The images were edited and compiled with Adobe Photoshop Version 23.2.1. The most important characters were listed in Supplementary  Table 2 and give an overview for each character for each species.

Molecular analysis
Cytochrome c oxidase I (COI) sequences analysed herein derive from Christodoulou et al. (2019), Christodoulou et al. (2020), Glover et al. (2016) and new COI sequences produced during this study. More specifically, the Ophiotholia dataset includes 42 sequences of O. saskia sp. nov. (Ophiotholia sp.1) and three sequences of O. supplicans (Ophiotholia sp.28) prepared by Christodoulou et al. (2019), Christodoulou et al. (2020). In addition, 18 new sequences of O. saskia sp. nov. and five sequences of O. supplicans were produced herein using methods described in Christodoulou et al. (2020); The arm tissue was extracted using E.Z.N.A DNA tissue Kit (Omega), the COI gene was amplified using the AccuStart PCR SuperMix polymerase (ThermoFisher Scientific) and the echinoderm specific LCOech1aF1 (Layton et al., 2016) forward primer paired with the universal jgHCO2198 reverse primer (Geller et al., 2013). Sequencing was executed at Macrogen (Amsterdam, Netherlands). Another 6 sequences were added from the NHM dataset , four specimens labeled as "NHM_076", which can be assigned to O. supplicans, and two specimens referred as "NHM_303" which can be assigned to O. saskia sp. nov. Due to the age, the type of fixation or the insufficient amount of tissue, our efforts were unsuccessful in obtaining sequences from other species of the genus Ophiotholia. The COI sequences were assembled and edited using Geneious v.9.1.8 (Kearse et al., 2012), whereas all sequences were aligned using MAFFT v7.017 (Katoh et al., 2002) with the default settings. The sequences, trace files, collection data and photos for each specimen are listed in the dataset "Clarion Clipperton Zone -Specimens of the genus Ophiotholia (Echinodermata : Ophiuroidea)" (CCZOTHOL; doi: dx.doi.org/10.5883/DS-CCZOTHOL). The new sequences produced are also available in GenBank (Sayers et al., 2019), accession numbers: OP575327-OP575373, OQ152220, OQ152221 (Supplement Table 1).

Putative species delimitation and genetic distances
Different species delimitation methods were used to cluster Ophiotholia specimens into possible genetic species and classify the individuals based on the dataset of Christodoulou et al. (2020). (1) The method "General Mixed Yule Coalescent" (GMYC, Pons et al., 2006) was applied using the R package SPLITS (Fujisawa and Barraclough, 2013), with the single-threshold model. The required ultrametric tree based on bayesian analysis was produced using the software BEAST v.2.6.6 with parameters as set out in Christodoulou et al. (2020). Additionally, the only other available sequence of the genus Ophiotholia, (Ophiotholia spathifer, NIWA14067) was included. As an outgroup, three species of the close related genus Ophiomyces were used (AB1_EB4_11_22 "Ophiomyces sp.", UF16941 "Ophiomyces frutectosus", BP34 "Ophiomyces delata". (2) Barcode index numbers (BINs) were assigned to the registered DNA dataset using the BOLD v.4 workbench (http://www.boldsystems.org; Ratnasingham and Hebert, 2013). The genetic distances between and within Ophiotholia species were calculated under the simple pdistance model in MEGA 11 (Tamura et al., 2021). The genetically delimited species were checked for morphological diagnostic characteristics to confirm the species definition by morphological evidence in parallel to genetic analysis. A minimum spanning haplotype network based on the COI sequences was produced in PopART (Version 1.7; available on: http://popart.otago.ac.nz/ index.shtml).

Taxonomy
Order Ophioscolecida O'Hara, Hugall, Thuy, Stöhr & Martynov, 2016 Family Ophiohelidae Perrier, 1893 Genus Ophiotholia Lyman, 1880 Diagnosis Disc has a sac-like shape and is fully scaled. Radial shields are absent. Two to three large and flat teeth present. Jaws (oral plates) covered by numerous flat papillae. Lateral arm plate has a fragile ledge with a distalwards pointing ventral extension that separates all arm spine articulations from the distal edge and encloses the round to triangle shaped dorsal arm plate. Arm spine articulation with nearly vertical dorsal and ventral lobes, connected at their proximal ends, opening distal-wards into a horseshoe-shape, bordering a single muscle/nerve opening, dorsal and ventral lobe same shape (except dorsally bent in O. spathifer). Parasol-spine articulation with an elevated circular lobe and a single (possibly muscle or nerve) opening ( Figure 2). Parasol-spines present on the distal part of the arm, with a thin shaft and one to two rings of teeth (one or two sprocket wheels), replacing arm spines partly or totally in some species. Distribution: This species was reported from the central Indian ocean in depths of 420 m (Litvinova, 1992).
Remarks: The species can be distinguished by the following characters. The disc is covered with large plates each provided with a long triangular spine (larger than in O. montana). There are 3-4 serrated arm spines that are shorter than an arm joint ( Figure 3A) and are arranged slightly distal to the parasol spines. There can be up to six parasol spines distally, arranged in two rows ( Figures 3D, E). The blunt parasol spine teeth are arranged in one ring. Two, pointed tentacle scales. The oral papillae were described as "very large". The dental plate carries three teeth.
Ophiotholia mitrephora H.L. Clark, 1910 (Figure 4 Remarks: The species has 4-5 smooth arm spines at the arm base, that are as long as an arm joint ( Figure 4H). The arm spines become confluent with a row of parasol spines distally ( Figures 4D, E). The parasol spines start at the fourth to fifth arm joint and are around six times smaller than the arm spines. There are three to four parasol spines occurring on each lateral arm plate, each with one ring of broad but pointed teeth ( Figures 4A, B). The dorsal arm plate is triangular shaped ( Figure 4F). There are three tentacle scales, one on the adradial and two at the abradial side of basal tentacle pores. Each dental plate carries four teeth. Distribution: This species was reported from the northwest in depths of 1230-3070 m (Litvinova, 1992).
Remarks: The disc has a conical shape, lower than in O. saskia sp. nov. and O. supplicans ( Figure 5D), and is covered with long thin spines (shorter than in O. gibbosa). The jaw has one long, thin apical papilla and next to it on each side is a similar-shaped mouth papilla, followed by a gap and then by typically three long and thick, almost rectangular, mouth papillae arranged in a single row ( Figure 5B). The dental plate carries two pointed teeth. There are five arm spines on the first four arm joints; followed by four arm spines in the subsequent arm joints, arm spines about 1.5 times the segment length ( Figure 5C). These are in a row confluent with three parasol spines. The parasol spines are large and have two rings of about six teeth. Distribution: This species was reported from the northwest Pacific in depths of 700-750 m (Litvinova, 1992).
Remarks: The disc is cone shaped and it is covered with thin, fishlike scales that carry triangular, serrated spines ( Figure 6E). It has four rounded arm spines ( Figure 6D), 1.5 times the length of an arm joint. The ventral most spine is thicker than the others. The parasol spines start at the seventh to eight arm joint and form two alternating rows offset to each other. There are up to six per lateral arm plate. The arm spines are arranged slightly proximal to the parasol spines. Each parasol spine has two rings of short teeth ( Figures 6A, B).  Scanning electron microscope images of close ups from the habitus of Ophiotholia saskia sp. nov. SO240_46_5. (A) Possible muscular connection between the parasol spine and its articulation. Scale = 50 µm (B) Muscular connection between arm spine and its articulation. Scale = 100µm. Description of the holotype: The description is based on the measurements of the holotype (SMF6909, Figure 7) with variation in the species represented by two additional specimens in Figure 8 (AB2_EB1_13_21) and Figure 9 (SO240_46_5). The disk diameter is 1.5 mm and its height 2.5 mm (when the arms are stretched out). The arms (without spines) are 0.35 mm in width and 4.7 mm in length. The mouth-angles are high and narrow, so that the mouth-slits between them are wide ( Figures 7G, H): They have one pointed apical papilla with one pointed mouth papillae next to it on each side. The following 6-8 mouth papillae are broader and larger, almost rectangular in shape, and arranged in two rows on each jaw (Figures 7H, 9E). Three broad and serrated teeth are visible ( Figure 9F). The oral plate is absent. The oval adoral shields are visible at the base of the jaws ( Figure 9E). The dorsal arm plates are broad, almost as long as wide, and rounded triangle in shape ( Figure 8I). They are well separated from each other by the contiguous lateral arm plates ( Figure 9C). The lateral arm plates are longer than wide, and meet broadly below, and form an obvious spinecrest at their outer edge ( Figures 7B, C, 8C, D, G). This spine-crest provides the articulation for the parasol-spines and is proximal to the arm spines ( Figures 7C, 8D). The distal side of the lateral arm plate forms a fragile ledge with a distalward pointing ventral extension, which separates all arm spine articulations from the distal edge, and further encloses a dorsal arm plate. The ventral arm plates are elongated and are rounded to rectangular, wider than long. The disk has a high conical shape, set with rows of minute spines between the arms, each spine on one disc plate. The disc plates are irregular, round, and large ( Figure 9A). No radial shields visible, which may account for the fact that the arms are raised vertically, encircling the high disk like a basket. Three to four pointed, slightly flattened, serrated arm-spines nearly as long as an arm segment, standing near the outer edge of lateral armplate and on a low spine-ridge, reduced to one spine in a row with the parasol-spines on distal arm segments ( Figures 7C, D, 9A, C). The articulations of the parasol-spines are arranged in three to four visible rows ( Figures 7C, 9C). From the third joint, there is a cluster up to 15, minute, parasol-spines on the inner side of spine-ridge, varying in length from 0.1 to 0.5 mm ( Figures 7A, 9B-D). The number increases distally. They are shaped like long-handled parasols, with an elongate shaft, surmounted by a disk with two rings of alternating pointed teeth, with up to 15 in each ring, and with a slight bulb at the base, which is inserted into articulations on the spine-ridge ( Figure 7A). Each tentacle pore on the first joints has one long, spine-like scale on its inner edge, about half the length of the arm spines ( Figure 9A). The arm vertebrae are longer than wide, with small, fragile and oval muscle attachments but long appendages that serve to connect with other ossicles. The length to width ratio increases distally ( Figure 7F).
Remarks: The species was described by Lyman in 1879 under the name Ophyomyces spathifer and was later transferred to the genus Ophiotholia by Litvinova, 1992 and synonymized with O. multispina. Fujita et al. (2010) have abolished the synominisation and assigned Ophiotholia spatifer as a species. Ophiotholia spathifer has six arm spines ( Figure 10E), which are gradually replaced distally from the 15th arm joint by the parasol spines ( Figures 10A, B). The parasol spines have one ring of thin pointed teeth ( Figure 10C). The dorsal arm plate has a triangular shape ( Figure 10F), whereas the ventral arm plate is rounded triangular, with convex distal edge and a thinner proximal lobe ( Figure 10D). There are two tentacle scales, one on each side of the tentacle pore.
Remarks: The species was used as the type species for Ophiotholia by Lyman in 1880 because of the uniqueness of its parasol-spines. Unfortunately, the holotype is so damaged that not all characters could be checked and consequently the type figures were also used to identify diagnostic characters. Ophiotholia supplicans can be distinguished from the other species by the following: The disc has a high cone shape and is sparsely covered by minute spines, which seem to stand on a small, fine disc scale. It has three arm spines, equal to the length of one arm segment. They are on the same flange of the lateral arm plate as the parasol spines ( Figure 11H). The parasol-spine articulations visible are not in defined rows (different to all other species). The parasol-spines have a short shaft and possesses a single ring of widely separated, blunt teeth ( Figures 11A, B). The ventral arm plate is elongated with a long and round structure on the distal side. The dorsal arm plate has an oval to triangle shape ( Figure 11D). There is one thin and pointed apical papillae in a line with three to four lateral oral papillae of the same shape. The outermost three to four mouth papillae are large, flat, paddle-shaped, and arranged in two rows on each jaw. The adoral shield appear in an oval shape distal to the mouth papillae. The coexisting new species O. saskia sp. nov. differs in having two rings of parasol teeth and two rows of distal lateral oral papillae.

Species delimitation and genetic divergence
A total of 78 sequences were used for the genetic analyses of which 20 were novel sequences (BOLD dataset: DS-CCZOTHOL, 68 sequences; GenBank: 10 sequences) and ranging from 597 bp to 658 bp in length (87% have a length of 658 bp). These barcodes clustered into two clades in the baysian tree corresponding to the two   (Figure 12). The clustering is supported by a posterior probability value above 99%. Mean interspecific p-distance between the two species was 23.7% (s=0.016) ranging from 21.7% to 24.9%, while mean intraspecific variability was 5% for O. supplicans and 1% for O. saskia sp. nov. So, a barcoding gap is recognizable. The BINs and GMYC both resulted in seven groups (Figure 12). Both methods also show the separation between the morphological assigned clusters O. saskia sp. nov. and O. supplicans. Within the cluster of O. saskia sp. nov. all specimens were considered to be one group. The cluster of O. supplicans is divided into six groups, one large group represented by seven specimens, three groups with two specimens and two singletons. A total of 59 and 11 haplotypes were recovered out of the 62 and 14 individuals of O. saskia sp. nov. and O. supplicans respectively (Figure 12). The haplotype diversity (Hd) in the medium joining network was p = 0.0898503. The haplotypes of O. saskia sp. nov. were found in three different CCZ mining license areas (UKSRL, BGR and NORI-D), while haplotypes of O. supplicans were found in four different license areas (UKSRL, BGR, IOM and NORI-D) and one APEI (APEI3). The number of mutations separating the two species was 120. It should be noted that the haplotypes of O. supplicans are separated by a higher number of mutations (2-33) than the haplotypes of O. saskia sp.nov. (max. six mutations between haplotypes).

Discussion
In the current study, a new species of Ophiotholia from the CCZ is described using an consolidated taxonomic approach (morphological and molecular analyses). The remaining six species of the genus are revised based on morphological data and genetic data (for O. supplicans). Only freshly collected material belonging to O. saskia sp. nov. and O. supplicans was suitable for molecular work. Specimens of the five other known Ophiotholia species were too old or not properly preserved for sequencing. The holotype of O. supplicans was too damaged and fragile for tissue sampling or detailed morphological analyses. The presence of "one-ringed parasol spines" and the structure of the oral part of the species "Ophiotholia sp.28" (as reported by Christodoulou et al., 2020) is consistent with the drawings of Lyman (1880) in its original description. Consequently, we can confirm that "Ophiotholia sp. 28" is Ophiotholia supplicans. Additionally, Christodoulou et al. (2020) suggest the occurrence of a third Ophiotholia species ("Ophiotholia sp.41") in the CCZ. However, because of the degraded condition and young age (postlarvae) of the specimens investigated, it was not possible to morphologically verify whether it is a known or new species, and its potential description as a new species will have to await the collection of more complete and adult specimens. The morphological study could show for the first time that there is a connection of each parasol-spines with the interior system of the ophiuroids (Figure 2). We suggest that it is a muscular connection, which allows the specimens to move the spines. The purpose of these spines is still unclear. In freshly collected specimens, of O. spathifer a curtain of mucus or a thin skin enclosing each parasolspine was observed (personal observations Eichsteller and O'Hara). The parasol-spines could therefore have a respiratory or sensatory purpose, as they occur on the tips of the arms which are outside the sediment while the rest of the animal lives buried inside the sediment.
Species delimitation methods based on the COI gene have been widely used for ophiuroid species (e.g. Ward et al., 2008;Khodami et al., FIGURE 9 Scanning electron microscope images of close ups from the habitus of Ophiotholia saskia sp. nov. 2014; Boissin et al., 2017;O'Hara et al., 2019;Christodoulou et al., 2020;Eichsteller et al., 2022). The methods used (GMYC, BIN) corroborated the morphological distinction of O. saskia sp. nov. and O. supplicans in the present study. The separation of these two clades was well supported in the phylogenetic tree (posterior probability >99%). We also highlighted the presence of a barcode-gap (higher interspecific genetic distances compared to the intraspecific ones) as well as a high number of mutations between the two species in the haplotype network. The extended dataset "CCZOTHOL" did not support the third Ophiotholia species "Ophiotholia sp.41" found in Christodoulou et al. (2020) but instead included it in the cluster of O. supplicans. BOLD and GMYC have recovered six subdivisions within O. supplicans implying the presence of more species. However, it has been shown in previous studies (Boissin et al., 2017;Christodoulou et al., 2020) that GMYC and BIN can overestimate the number of species. The low intra-cluster divergence (2.2%) at the initial cluster step of refined single linkage (RESL) in BIN methodology (Ratnasingham and Hebert, 2013;Song et al., 2018) could be the reason why in some cases the BIN method overestimated species number. Furthermore, it can be that GMYC has not performed optimally as the ultrametric tree was estimated with a single locus. Finally, no morphological differences were found that could justify the presence of more species. While the mean intraspecific genetic variability within O. supplicans is 5% and, although arguable significantly higher than O. saskia which is 1%, it's still within the limits of intraspecific variability in Ophiuroidea (Khodami et al., 2014;Boissin et al., 2017;Christodoulou et al., 2020), while species with higher intraspecific values are known from previous studies (eg. Layton et al., 2016;Boissin et al., 2017). The few individuals within these subdivisions come from different study areas. It has been shown in other studies that specimens of the same species can have a high variation in the haplotype analyses over a short distance (Peŕez-Portela et al., 2013;Galaska et al., 2017;Weber et al., 2019). Additionally, the haplotype network of the specimens belonging to O. saskia sp. nov. shows a high haplotypes diversity, thus it can be expected that O. supplicans has a similar high haplotype diversity as well. To avoid unnecessarily splitting, we follow a conservative approach and classify the Ophiotholia specimens found in CCZ and, which are not O. saskia, as one species under O. supplicans until more data support otherwise. It is possible that our dataset of O. supplicans is too small to exhibit a stable number of species within this subdivision. Even if there is an accumulation of individuals aligning with species of Ophiotholia sp. 41 (Christodoulou et al., 2020), this subdivision is still closely related to O.
supplicans. In many cases the animals were immature and damaged (sometimes only a single arm) which limited our ability to use morphology. A larger dataset is required to better resolve all the species boundaries within Ophiotholia.
Understanding the diversity of seafloor habitats and the associated fauna is critical to making informed decisions about deep-sea conservation (Brix et al., 2019;McQuaid et al., 2020;Jones et al., 2021;Washburn et al., 2021). In the CCZ, the possible extraction of the abundant and valuable nodules could not only result in loss of substratum but will also result in seabed upheaval and possible biodiversity loss, and perhaps an ongoing effect on the (Ratnasingham and Hebert, 2013) local food web (Levin et al., 2016;Niner et al., 2018;Brix et al., 2019;Simon-Lledóet al., 2019a;FIGURE 11 Scanning electron microscope images of the ossicles from Ophiotholia supplicans.  Stratmann et al., 2021). Although ophiuroids are reported to take the role of the "generalist" in recolonization after mining disturbance, this refers mostly to the large, mobile, deposit feeders recognized in image based analyses (Jones et al., 2017;Simon-Lledóet al., 2019a;Simon-Lledóet al., 2019b). Knowledge at species level is important to determining disturbance impacts. Species of the class Ophiuroidea represent all kinds of feeding types and show different lifestyles (e.g. Pearson and Gage, 1984;O'Hara et al., 2008;Mosher and Watling, 2009;Stöhr et al., 2012;Laming et al., 2021;Manjoń-Cabeza et al., 2021). So, it is incorrect to generalize the behavior of the whole class. The smaller and fragile specimens of the genus Ophiotholia are living buried in the upper layer of the seafloor, with their arms erected into the water column. This behavior suggests a less mobile and possibly filter feeding lifestyle, as likely in other sessile fauna (Lyman, 1880;Litvinova, 1992). The impact on the sediment redeposition on the surrounding benthic fauna is not well understood, (Jones et al., 2017;Simon-Lledóet al., 2019a;Stratmann et al., 2021), so it is unclear what will happen to the benthic communities, including the Ophiotholia species. But it is most likely that not only the targeted area but also surrounding areas will be influenced by the sediment plume, for example benthic communities could be smothered or nutrients could be introduced to an otherwise nutrient poor area and hence alter the local community (Simon-Lledóet al., 2019a;Washburn et al., 2021). There is also a risk from mobilized toxic metals and machinery contaminates in-and outside the mining area (Peukert et al., 2018). In 1989 an area off Peru was disturbed in 4150 m during an intensive disturbance study DISCOL (disturbance and recolonization experiment, Thiel and Schriever, 1990). After revisiting the site more than 20 years later, studies on the biodiversity show little recovery in the disturbed area and only the presence of some mobile species, including large ophiuroids (Miljutin et al., 2011;Simon-Lledóet al., 2019a). Although not all circumstances of habitat condition and data collection are the same in both areas (CCZ and DISCOL), similar recovery can be expected. Thus, mobile, deposit feeding ophiuroids may recover, but ophiuroids with a different lifestyle, like species of Ophiotholia might not resettle.
To accurately assess the risk of species extinctions from largescale mining, the knowledge of the taxonomy, genetic structure, biogeography and basic natural history of deep-sea species is crucial. In a recent review (Engel et al., 2021) the "Taxonomic impediment" is discussed and shows the need for additional taxonomic work in the time of biodiversity loss and study of lesser Ultrametric tree resulting from the Bayesian analysis of the CCZOTHOL dataset. The posterior support (≥ 99%) is indicated at the branch. The results of both species delimitation analyses (BINs, GMYC), are shown through the bar. The tree is combined with the haplotype network of Ophiotholia saskia sp. nov. and Ophiotholia supplicans; produced by PopART (Leigh and Bryant, 2015). The haplotype network is based on COI data of 75 samples. The size of the circle indicates the number of haplotypes. The colours are assigned to the different areas in the CCZ (red = UKSRL; green = BGR; yellow = NORI-D, blue = IOM, violet = APEI). The dashed boxes around each subnetwork indicate the morphological clusters in the Neighbour-Joining tree (Turquoise = Ophiotholia supplicans; Yellow= Ophiotholia saskia). known habitats. Our study provides a better understanding of the specimens of the genus Ophiotholia as well as contributing to the list of species that occur in the CCZ and thus to the assessment of biodiversity in the area. A well-established taxonomic background supports the identification of common benthic communities with possible indicator species and can improve the models on e.g., food web analyses or ecosystem changes (Stratmann et al., 2021). Species descriptions and barcoding are also essential to informing future eDNA and imaging studies, as well as contributing to large scale analyses as shown in Woolley et al. (2021) and Uhlenkott et al. (2022).

Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: GenBank, OP575327-OP575373, OQ152220, OQ152221, and in the Barcode of Life Database: doi: dx.doi.org/10.5883/DS-CCZOTHOL.

Author contributions
The present study was conceptualized by AE, TO'H, MC and PMA. PMA provided the samples and acquired the funding for this project. GB-C also provided samples from the NHM London. AE, MC and GB-C planned and conducted all laboratory experiments. AE obtained the SEM images and configurated the plates of all species. AM and TK executed the data curation and SEM data obtaining of the specimens from the ZMMU. AE did the analyzing of the molecular data. TO'H and AM contributed with their taxonomic expertise in Ophiuroidea. AE wrote the initial draft of the manuscript; all authors contributed to the final version of the manuscript. All authors contributed to the article and approved the submitted version.

Funding
Research for this article was funded by Deutsche Forschungsgemeinschaft DFG, grant number MA2557/12-1 within the priority program "Taxon-Omics: New Approaches for Discovering and Naming Biodiversity" (SPP 1991). As well as funded by the EU JPIO-Oceans project MinigImpact-2 (German BMBF contract 03F0707E and 03F0812E). This is publication number 93 from the Senckenberg am Meer Molecular and Metabarcoding Laboratory. Also, funding was received from UK Seabed Resources and The Metals Company to the Natural History Museum, NORCE and University of Gothenburg to support the collection of samples and DNA sequencing made available to this study. GB-C has received support from TMC the metals company Inc. (The Metals Company) through its subsidiary Nauru Ocean Resources Inc.
(NORI). NORI holds exploration rights to the NORI-D contract area in the CCZ regulated by the International Seabed Authority and sponsored by the government of Nauru. This is contribution TMC/NORI/D/005. The work of AM was performed under a research project of the MSU Zoological Museum (18-1-21 No. 121032300105-0). The work of TK was conducted under the Koltzov Institute of Developmental Biology RAS (IDB) basic research program in (No. 0088-2021.