Neanthes goodayi sp. nov. (Annelida, Nereididae), a remarkable new annelid species living inside deep-sea polymetallic nodules

A new species of abyssal Neanthes Kinberg, 1865, N. goodayi sp. nov., is described from the Clarion-Clipperton Zone in the central Pacific Ocean, a region targeted for seabed mineral exploration for polymetallic nodules. It is a relatively large animal found living inside polymetallic nodules and in xenophyophores (giant Foraminifera) growing on nodules, highlighting the importance of the mineral resource itself as a distinct microhabitat. Neanthes goodayi sp. nov. can be distinguished from its congeners primarily by its distinctive, enlarged anterior pair of eyes in addition to characters of the head, pharynx and parapodia. Widespread, abundant, and easily recognisable, N. goodayi sp. nov. is also considered to be a suitable candidate as a potential indicator taxon for future monitoring of the impacts of seabed mining.


Introduction
Exploration of our deep oceans for potential new industrial activities has increased rapidly in recent decades with the so-called 'blue growth' economy (European Commission 2020). Critical to a sustainable blue economy is baseline knowledge on the environmental characteristics of these exploration areas, in particular knowledge of the species that live there (Glover et al. 2018). This is especially the case in the central Pacific abyss Clarion-Clipperton Zone (CCZ), a region targeted for seabed mineral exploration for polymetallic nodules, where basic faunistic and taxonomic data are notably lacking and many animals likely undescribed or undocumented (Glover et al. 2018). Here, we describe a new nereidid annelid from the abyss that is not only important for understanding the general baseline biology of the region, but also presents a remarkable natural history -living inside the polymetallic nodules themselves. As the species is relatively large and easy to recognise, it should be added to a list of nodule-dwelling fauna that could be used as indicators in future environmental assessments (Lim et al. 2017). Information on the existence, abundance and distribution of these species could be essential to environmental monitoring and conservation measures in the region.
The CCZ lies in international waters and lacks strictly defined boundaries; however, it is generally accepted to encompass the region between the Clarion and Clipperton Fracture Zones, with multiple polymetallic nodule exploration contracts for seabed minerals issued by the International Seabed Authority (ISA 2018), extending from 115° W (the easternmost extent of the UK-1 polymetallic nodule exploration area) to approximately 158° W (the westernmost extent of the COMRA polymetallic nodule exploration area). As such, we hereafter use a working definition of the CCZ as comprising the box: 13° N, 158° W; 18° N, 118° W; 10° N, 112° W; 2° N, 155° W -an area spanning almost 6 million km 2 , approximately 1.4% of the ocean's surface.
Polymetallic nodules are small mineral accretions (usually 5-10 cm in diameter, but occasionally exceeding 20 cm) rich in cobalt, manganese, copper and nickel, among numerous other metals of economic interest (Hein et al. 2013). These nodules sit on the sea floor, often half submerged in sediment, providing the only hard substrate in an otherwise soft sediment environment, contributing to a high habitat heterogeneity compared with regions of the deep sea without nodules or hard substrate. Nodules provide microhabitats for meio-and macrofaunal groups such as annelids and crustaceans (Thiel et al. 1993;Gooday et al. 2017;Kersken et al. 2019), in addition to sites of attachment for sessile megafauna (e.g., Relicanthus sp. anemones) (Amon et al. 2016).
Nereididae de Blainville, 1818 is among the most diverse families within Annelida, with over 40 valid genera and up to 750 valid species (Read & Fauchald 2020a). Members of the family are broadly omnivorous, and most species appear to be facultatively motile, rarely leaving mucus-built tubes and burrows unless disturbed or when conditions become unfavourable (Fauchald & Jumars 1979;Jumars et al. 2015). Sexually mature individuals may develop into pelagic morphs (epitokes), which are thought to have much greater motility. However, not all nereidids form epitokes during reproduction, and not all epitokes are pelagic, with the degree of modification varying between species and sexes (Bakken & Wilson 2005).
The genus Neanthes Kinberg, 1865 is one of the most diverse genera within the family, with over 80 currently accepted species (Read & Fauchald 2020b), and can be distinguished from similar genera such as Hediste Malmgren, 1867 and Nereis Linnaeus, 1758 by morphological characters primarily relating to the presence or absence of certain chaetal types, for example in lacking compound falcigers in notopodial fascicles (as in Nereis), but possessing homogomph spinigers in ventral neuropodial fascicles (absent in Hediste and Nereis) (Bakken & Wilson 2005).
However, Neanthes is considered to be polyphyletic (Bakken & Wilson 2005) and a generic revision based on phylogenetic analyses is needed to resolve its taxonomy (Bakken & Wilson 2005;Bakken 2006; Glasby et al. 2011;Shimabukuro et al. 2017;Villalobos-Guerrero 2019). The majority of species of Neanthes have been described from shallow or intertidal waters, with only 13 species reported from depths greater than 200 m (Khlebovich 1996;Shimabukuro et al. 2017;Hsueh 2019). Notably, Thiel et al. (1993), when examining nodules collected from the South Pacific (outside of the CCZ) as part of the DISCOL project, reported two unidentified species of Neanthes when first describing polymetallic nodule crevices as a discrete microhabitat; these were among six annelid taxa that were only found within interstitial mud from nodule crevices, and not from the surrounding soft sediment.
In this study, we describe a new species of abyssal Neanthes observed to reside either directly within nodule crevices, within mud balls on nodule surfaces or burrowing within xenophyophores (giant foraminiferans) growing on nodules. This species is notable in that it highlights the potential importance of nodule microhabitats for macrofaunal-sized animals, and is also one of the most abundant and widespread annelid species collected as part of the ABYSSLINE ('ABYSSal baseLINE') UK-1 environmental survey project. Easily recognisable, it is a critical 'target taxon' for further assessments of biogeography and population connectivity patterns, the subject of a separate study (Dahlgren et al., unpublished data).

Fieldwork
Specimens were collected across two cruises, the first UK Seabed Resources ABYSSLINE cruise (AB01) sampling the UK-1 exploration contract area aboard the RV Melville, October 2013, and the second cruise (AB02) sampling the UK-1 and OMS (Ocean Mineral Singapore) exploration contract areas as well as an area to the north designated as Area of Particular Environmental Interest 6 (APEI-6) onboard RV Thomas G. Thompson, February-March 2015 (Fig. 1). A comprehensive description of the DNA taxonomy methodological pipeline used here is provided in Glover et al. (2016). In summary, a range of oceanographic sampling gear, including box corer, epibenthic sledge (EBS), ROV and multiple corer, were used to collect deep-sea benthic specimens from the UK-1, OMS and APEI-6 areas. Geographic data from sampling activities were recorded on a central GIS database. A 'cold-chain' pipeline was used in the live-sorting of specimen samples aboard both vessels, where material was constantly maintained in chilled, filtered seawater held at 2-4 °C. Specimens underwent preliminary identification at sea and were live-imaged using digital cameras attached to stereo microscopes . All specimens were then stored in individual microtube vials containing an aqueous solution of 80% nondenatured pre-chilled ethanol, which were numbered, barcoded into a database and stored chilled until return to the Natural History Museum, London, UK.

Laboratory work
A total of 43 specimens were identified as conspecific using genetic data (see below) and considered in morphological analyses, with a portion of representative specimens selected as type material for more detailed analyses.
Specimen measurements taken included total length (TL), length to chaetiger 15 (L15), width of chaetiger 15 excluding parapodia (W15), and the total number of chaetigers for complete specimens. Paragnaths for each pharangeal area were counted, with paired areas that differed in numbers distinguished using a and b for the left and right side of the specimen respectively. The number of teeth on the jaws were also counted. For specimens where the pharynx was not everted, a longitudinal dissection was made in the mid-ventral region. For examination of parapodial features and modifications along the body, several parapodia were removed (from chaetigers 1, 3, 6, 10, every tenth chaetiger thereafter, and a posteriormost chaetiger, where possible) and mounted on glass slides. Parapodia were dissected from either the left or right side of the specimen based on intactness of features such as cirri.
Specimens were examined using compound and light microscopes, and photographed using attached digital cameras on both microscopes. Figures were assembled using Adobe Photoshop CS6 software. A fine white or black line was used to outline and highlight particular morphological features where they were unclear from the images alone. Standardised terminology of nereidid parapodial features follows Villalobos-Guerrero & Bakken (2018); the shape of pharangeal areas and ridge patterns follows Villalobos-Guerrero (2019).
In total 47 terminal taxa were used in the phylogenetic analyses, with 44 from Nereididae, and three taxa from Hesionidae Grube, 1850, another family within Nereidiformia, as the outgroup. While some earlier studies suggest that Chrysopetalidae Ehlers, 1864 is sister taxon to Nereididae (Dahlgren et al. 2000), later analyses have indicated that the Nereidiformia relationships are unresolved (Weigert & Bleidorn 2016), which justify the use of Hesionidae as the outgroup here. The program jModelTest (Posada 2008) was used to assess the best model for each partition (18S, 16S and COI) with BIC, which suggested GTR + I + G as the best model for all genes. The data was partitioned into the three parts (18S, 16S and COI) and this evolutionary model was applied to each partition. The parameters used for the partitions were unlinked. Bayesian phylogenetic analyses (BAs) were conducted with MrBayes ver. 3.2.6 (Ronquist et al. 2012). Analyses were run three times for 10 000 000 generations. Of these, the first 2 500 000 generations were discarded as burn-in. Tree files were interpreted with FigTree ver. 1.4.2 (available from http://tree.bio.ed.ac.uk/software/figtree/).

Data management
The management and transfer of specimen data between a central museum database, a molecular collections database, and external data repositories and aggregators (e.g., GenBank, World Register of Marine Species (WoRMS), Ocean Biodiversity Information System (OBIS), Global Biodiversity Information Facility (GBIF), Global Genome Biodiversity Network (GGBN), and ZooBank) was carried out through the usage of DarwinCore data standards (Wieczorek et al. 2012) including the GGBN DarwinCore extensions (Droege et al. 2016). See Glover et al. (2016) for further elaboration of this data pipeline. All specimens and DNA vouchers are archived in the Natural History Museum London collections. All specimen occurrence (and associated preparation) data are provided in a DarwinCore Archive (DWcA) in the supplementary material (Supp. file 1). Neanthes -Fauchald 1977: 89. -Wilson 19841988: 5. -Wu et al. 1985
Pharynx not everted. Jaws dark red-brown with 6 lateral teeth; All paragnaths brown, conical, arranged as follows (Fig. 2E): area I: 2, one large cone, one smaller cone distally; area II: 12 in cluster; area III: approx. 6 (area damaged), four cones in row with two smaller cones laterally; area IV: 13 in teardropshaped cluster, with curved line of cones extending from jaws posteriorly, ending in cluster of 7 cones; area V: no paragnaths; area VIa: 1; area VIb: 4, one large and three smaller cones in trapezoid arrangement; areas VII-VIII: 19, eight large cones in a single well-spaced row with 11 smaller cones scattered laterally. Areas VI-V-VI with λ-shaped ridge pattern.
Notochaetae all homogomph spinigers with long blades, of similar width towards toothed edge but drastically slendering to an aristate distal end (Fig. 3I); 4 present in anterior chaetigers, 5 in medial chaetigers, 3 in posterior chaetigers and absent from chaetiger 46. Supracicular neurochaetae with homogomph spinigers and heterogomph falcigers, both types present in all falcigers except final two chaetigers, where supracicular falcigers are absent. Homogomph spinigers similar in appearance to those of notopodia (Fig. 3J), though with blades reducing in length moving ventrally (shortest blades two-thirds as long as longest blade), numbering 4 on first two chaetigers, 3-5 on anterior and medial chaetigers and 2 on posterior chaetigers where fascicles remain. Heterogomph falcigers with knob-like tips (Fig. 3K) and blades roughly half the length of shortest spinigers, numbering 1 on anterior chaetigers, 2 on medial chaetigers and 1 on posterior chaetigers where fascicles remain.
Natatory chaetigers with distinctly enlarged, elongate modified parapodia (Fig. 6D). Noto-and neuropodia elongated basally, with ligules and lobes not significantly larger than on non-modified parapodia. Neuracicular ligule with lamellar structure distally. Both dorsal and ventral cirri notably elongate, with a pair of conical lobes emerging from the upper and lower base of each cirrus, not present on anterior chaetigers; dorsal cirri slightly papillated ( Fig. 6D-E). Both notopodial and neuropodial fascicles dense, up to 40 chaetae per fascicle, and with only a single chaetal type: long, simple sesquigomph spinigers  with ensiform (knife-shaped) blades (Fig. 6F). No gametes observed, though the presence of slightly papillated dorsal cirri on natatory chaetigers suggests that this specimen is a male (Read 2007).

Juveniles
Several small, possibly juvenile specimens were observed; paratypes NHM_127, NHM_171, NHM_1254, TL = 1.0-2.5 mm, L15 = max. 2.2 mm, W15 = max. 0.2 mm, 10-18 chaetigers (Fig. 5A-D). Posterio-dorsal tentacular cirri extending to chaetiger 6. Eyes poorly developed in these specimens, with anterior eye pair observed only as faintly pigmented crescents (Fig. 5B-D), lenses not obvious; posterior eye pair not visible in smallest specimens (Fig. 5C-D). The identity of these specimens was confirmed with genetic data. Due to their size and the delicate nature of specimens, pharyngeal and parapodial dissections were not conducted to preserve specimen integrity. Fig. 7. Phylogenetic analysis of Nereididae Blainville, 1818, 50% majority rule tree from the Bayesian analyses using 18S, 16S and COI, with posterior probability values on nodes. Forty-five taxa from GenBank were included, using three taxa from another family within Nereidiformia, Hesionidae Grube, 1850, as outgroup.

Genetic data
All 43 individuals were sequenced for 16S and COI. The gene 16S was successfully sequenced in all but six specimens. COI sequencing was less successful; however, each specimen had coverage of at least one of the two genes. All specimens formed a single clade with low intraspecific divergence. Several specimens were also sequenced for 18S in order to assess deeper taxonomic relationships. This species was genetically distinct from all other species included in our phylogenetic analyses, and forms the basal branch of a clade including Neanthes fucata (Savigny, 1822) and five species of Perinereis Kinberg, 1865 (Fig. 7).

Remarks
This species is most consistent with the genus Neanthes Kinberg, 1865, most recently defined by Ibrahim et al. (2019). Previous analyses based on morphological parsimony suggested that neither of the three most species-rich nereidid genera, Neanthes, Nereis and Perinereis, can be considered monophyletic, with many generic characters displaying high homoplasy (Bakken & Wilson 2005). Molecular phylogenetic analyses carried out in this study supported the polyphyly of Neanthes, as sequences of species currently regarded as Neanthes, both from the ABYSSLINE material and from GenBank, rarely grouped together and were evenly distributed throughout a tree that included 11 other nereidid genera.
Neanthes goodayi sp. nov. can be differentiated from the majority of its congeners by the notably large anterior pair of eyes. Only N. heteroculata (Hartmann-Schröder, 1981), described from abyssal (4700 m) waters off the Bay of Biscay in the northeastern Atlantic, appears to possess comparably large anterior and minute posterior pairs of eyes. Neanthes heteroculata and N. goodayi sp. nov. also display similarities with regard to several other characters, such as the appearance of the prostomium, antennae and tentacular cirri, in addition to the types of chaetae present and their appearance and arrangement. Based on an examination of the type material of N. heteroculata, N. goodayi sp. nov. differs in having distinctly rounded, spherical to ovoid palpophores (e.g., Fig. 4B), with palpophores in N. heteroculata found to be narrower, bluntly conical in shape. Furthermore, the dorsal cirri are relatively short in N. heteroculata, not exceeding the length of the notopodial ligules, whereas they exceed the length of the notopodial ligules in at least anterior and posterior chaetigers in N. goodayi sp. nov.
Notably, N. heteroculata is one of a handful of species of Neanthes reported from the deep sea. Of the 84 currently valid species of Neanthes (Read & Fauchald 2020b) only 13 have been reported from depths greater than 200 m (Khlebovich 1996;Shimabukuro et al. 2017;Hsueh 2019). Of these, N. goodayi sp. nov. also resembles N. papillosa (Day, 1963), described from deep (2745 m) waters off Cape Town, South Africa. Neanthes papillosa similarly possesses an enlarged anterior pair of eyes relative to the posterior pair, in addition to long tentacular cirri, relatively elongate, conical parapodial ligules, and dorsal cirri that exceed the length of the notopodial ligules, becoming longer on posterior chaetigers. The holotype of N. papillosa is noted to have pale, poorly chitinised paragnaths, thus making them difficult to observe (Day 1963). However, despite having fewer paragnaths in number across all areas, they appear to be organised in similar arrangements as in N. goodayi sp. nov., such as a single row of paragnaths on areas VII-VIII (single row of large cones in N. goodayi sp. nov. with varying numbers of smaller cones laterally). However, N. papillosa can primarily be differentiated from N. goodayi sp. nov. in that the anterior pair of eyes does not appear to be as strikingly large as in N. goodayi sp. nov. or N. heteroculata; thus, there is less disparity between the anterior and posterior eye pairs in size. Additionally, N. papillosa can be further distinguished in that it does not bear homogomph falcigers and that parapodial lobes of midbody and posterior chaetigers bear numerous club-shaped papillae; however, it is worth considering that some characters of N. papillosa may be reproductive modifications, as the holotype is described from a single epikotous female specimen.
Neanthes goodayi sp. nov. also bears similarities to N. vitiazi Khlebovich, 1996 from abyssal waters (3342-4160 m) of southern Japan, primarily in terms of broadly similar paragnath distributions, bearing homogomph falcigers and in having a large anterior pair of eyes, which are illustrated as rings without strong pigment. Neanthes vitiazi differs in that it has long, digitate median ligules positioned at right angles to the notoacicula on midbody and posterior chaetigers. Neanthes vitiazi is also described as having brown pigmentation on parapodial appendages and dense spot-like pigmentation on the apodous anterior segment; N. goodayi sp. nov. similarly bears two pigmented spots on the dorso-lateral anterior margin of this segment; however, these are relatively small, whereas the spots in Neanthes vitiazi span much of the length of the segment and are placed dorsally, behind the eyes.
The geographically most proximal deep-water species, N. mexicana Fauchald, 1972, described from abyssal waters off Baja California, and N. sandiegensis Fauchald, 1977 from the San Diego Trough (728-855 m), can also be differentiated from N. goodayi sp. nov. Neanthes mexicana was originally described from a single damaged specimen, re-examined and revised by de León-González & Solís-Weiss (2000) with the addition of several nereidids collected from abyssal waters off California USA agreeing with the type specimen. Neanthes mexicana is described as bearing a single pair of very large red eyes, with diffuse pigment spots posterior to the eyes noted to perhaps represent the posterior eye pair (Fauchald 1972). In ABYSSLINE specimens, the appearance of the posterior eye pair was variable, ranging from discrete dark spots to more faint, irregular shapes, occasionally with one or both eyes absent all together, particularly in smaller specimens. The eye morphology of N. mexicana therefore falls within the variation observed in the ABYSSLINE samples. Neanthes mexicana and N. goodayi sp. nov. also share similarities in terms of parapodial morphology, with all parapodial ligules broadly conical to somewhat triangular in shape (see de León-González & Solís-Weiss 2000: fig. 3). However, N. mexicana differs from N. goodayi sp. nov. in terms of palp morphology (long, digitate palpostyles), the arrangement and number of paragnaths (4 cones in areas II and IV versus 12 cones in both areas in N. goodayi sp. nov.,) and in lacking homogomph falcigers.
While none of the morphologically most similar or geographically proximal congeners had genetic data available for comparison, morphological differences existed in each case. Neanthes goodayi sp. nov. can be differentiated from other deep-water Neanthes spp. primarily in terms of eye morphology: N. articulata Knox, 1960, N. donggungensis Hsueh, 2019, N. kerguelensis (McIntosh, 1885 and N. suluensis Kirkegaard, 1995 bear two relatively small, subequal eye pairs, whereas N. bioculata (Hartmann-Schröder, 1975) bears a single pair of small eyes; N. abyssorum Hartman 1967, N. kermadeca (Kirkegaard, 1995, N. shinkai Shimabukuro et al., 2017 andN. typhla (Monro, 1930) are recorded as lacking eyes altogether and can be further differentiated from N. goodayi sp. nov. in terms of paragnath distribution, among other characters (see Shimabukuro et al. 2017 for comparative morphological table of most deep water Neanthes spp.).

Ecology
Neanthes goodayi sp. nov. was found at depths ranging from 4000 to 4400 m living in crevices of polymetallic nodules (Fig. 8A-B), burrowing in xenophyophore foraminifera growing on nodules ( Fig. 8C-E) or in mud balls on nodule surfaces (Fig. 8F-H). As in other nereidids, the strong eversible jaws, together with large eyes, indicate an active and predatory behaviour. While we were able to observe live, moving specimens kept at cold temperatures even after recovery from 4000 m water depth, behaviours such as predation were not observed. Polymetallic nodules are thought to contain a diverse meiofaunal community of nematodes, copepods and other small crustaceans; thus, it is possible that N. goodayi sp. nov. is a 'sit and wait' predator that is able to remain inside the nodules and detect prey passing overhead through extremely small variations in light (from local bioluminescence, detected by the large eyes) or other physio-chemical cues.

Distribution
Eastern Clarion Clipperton Fracture Zone, Central Eastern Pacific.

Discussion
It is perhaps remarkable that one of the more obvious and charismatic animals living on and inside the most investigated mineral resource on the deep seafloor has not been described until now. However, the CCZ region, despite a large number of expeditions and considerable sampling effort, has clearly never received appropriate taxonomic attention (Glover et al. 2018). Only in recent years has any effort been made to describe polychaete species, with 29 new species described in two recent papers (Bonafácio & Menot 2018;Wiklund et al. 2019). Such descriptions are essential to future investigations of population connectivity and resilience, extinction risk modelling, ecosystem function, natural history, ecology and life history (Glover et al. 2018).
The more obvious macrofauna that live on polymetallic nodules are likely to be useful in the future for monitoring the impacts of seabed mining, if it were to start. In this regard, Neanthes goodayi can be added to this list of potential 'indicator taxa' alongside the recently described nodule-dwelling sponge, Plenaster craigi Lim & Wiklund, 2017. Like P. craigi, N. goodayi sp. nov. is relatively easy to recognise during routine examination of nodules, and is sufficiently abundant to be counted in replicated samples. The smaller macrofauna dwelling in the sediments around the nodules is still extremely difficult to identify without using genetic methods and as such can only really be identified by specialists. The presence or absence of nodule-dwelling taxa such as P. craigi or N. goodayi sp. nov. may prove to be a useful measure of ecosystem health.