Molecular phylogeny and morphological comparison of the deep-sea genus Alloptilella Li, Zhan & Xu, 2021 (Octocorallia, Pennatulacea)

A previously described species and a new one belonging to the recently described sea pen genus Alloptilella Li, Zhan & Xu, 2021, are here described and illustrated based on a morphological and molecular study of materials collected in the Tasman Sea (SW Pacific) and at Puerto Rico (Caribbean Sea), respectively. The species, Alloptilella moseleyi comb. nov. (Kölliker, 1880) and Alloptilella williamsi sp. nov., are in overall agreement with the generic diagnosis of Alloptilella., based on the type species, Alloptilella splendida Li, Zhan & Xu, 2021. A single relatively large colony (55 to 95 cm in total length) is known for each of the three Alloptilella species. The transferred and the new species differ from the type species in having an opposite, rather than alternate, placement of polyp leaves along the rachis, colouration of autozooids, and mesozooids (in the case of A. moseleyi comb. nov.), and spicular features (e.g. maximum sizes in different parts of the colony, presence/absence of tentacular sclerites). Alloptilella williamsi sp. nov. is the first species of the genus recorded so far from the Atlantic Ocean, all other described species being western Pacific. A molecular comparison based on a set of concatenated sequences of four markers (three mitochondrial genes (mtMutS, ND2, and COI) and a nuclear segment (28S)) relates the species here studied to the published sequences of Alloptilella splendida, within the named Clade II of previous general phylogenetic studies on the octocoral Order Pennatulacea. Alloptilella is a monophyletic grouping, sister group of the genus Scytalium Herklots, 1858. The available molecular information of the genus Ptilella Gray, 1870, is reinforced with sequences (mtMutS, ND2 and 28S) from specimens of Ptilella inflata (Kükenthal, 1910) collected off the Namibian coast (SE Atlantic).


Introduction
Recent molecular studies in pennatulacean octocorals pointed out the polyphyly of the genus Pennatula Linnaeus, 1758 (Dolan et al. 2013;Kushida and Reimer 2018;García-Cárdenas et al. 2020). Two additional contributions proposed nomenclatural acts to resolve that situation: (1) the sequencing of certain North Eastern Atlantic specimens identified as Pennatula grandis Ehrenberg, 1834, resulted in the restoration of the old synonymized genus, Ptilella Gray, 1870(García-Cárdenas et al. 2019; and (2) the existence of a divergent lineage from the genus Pennatula (sequences of the type species Pennatula phosphorea Linnaeus, 1758, and its close relatives) was corroborated with the description of a new genus based on North Western Pacific material, Alloptilella Li, Zhan & Xu, 2021(Li et al. 2021. In both cases, molecular comparisons helped segregate morphological features previously considered within the morphological variability of the genus Pennatula (Kükenthal 1915;Williams 1995).
The morphological description and molecular characterization of Alloptilella splendida Li, Zhan & Xu, 2021, force the morphological and molecular revision of controversial material that was preliminary identified as Pennatula sp. (GenBank accession numbers DQ302870 for mtMutS and DQ302943 for ND2) from the Tasman Sea (SW Pacific) deposited at the Northern Territory Museum (Darwin, NT, Australia). This latter material is closely allied (from a molecular point of view) to Alloptilella, and was the basis of different discussions concerning the polyphyly of the genus Pennatula.
An additional specimen with a similar morphology and colouration pattern of A. splendida was collected during a deep-sea expedition in the North Western Atlantic. This material was deposited at the United States National Museum (Smithsonian Institution, Washington, USA), and the preliminary morphological and molecular study also confirmed its relatedness to the genus Alloptilella.
The goal of the present paper is the description of a previously discovered and a new species of the genus Alloptilella based on western South Pacific and western North Atlantic specimens, respectively. A morphological and phylogenetic comparison of this material with A. splendida was carried out.

Material and methods
The material described here was collected during two different cruises, one of them in the Tasman Sea (RV Tangaroa) and the second one in the Caribbean Sea (NOAA ship Okeanos Explorer). The specimen from the Caribbean Sea was collected by the ROV Deep Discoverer (NOAA).
For comparative purposes, specimens of the morphologically similar genera Pennatula and Ptilella collected in different survey programmes from the Artic to the Antarctica were also examined: Subantarctic and Antarctic (EASIZ, CLIMANT, BIOROSS), North Eastern Atlantic-Arctic The colonies were fixed on board in ethanol 95% for morphological examination and further molecular studies. Sclerites of different parts of the colonies were prepared for SEM study employing the standard methodology described by different authors (e.g. Bayer and Stefani 1988), and permanent mounts were made for light microscopy. Calyces, tentacles, and other selected fragments were prepared in semipermanent mounts using clove oil as mounting medium for corroboration of the presence of sclerites and examination of their arrangement in a MOTIC B3 light microscope (Li et al. 2020). Sclerites were mounted on stubs, coated with goldpalladium under a Leica ACE600, and observed with a Zeis EVO SEM at the General Research Services of Microscopy at the University of Seville. Colony and sclerite terminologies mainly follow Bayer et al. (1983) and Li et al. (2021).
Total genomic DNA was extracted from the ethanol (EtOH)-preserved specimen using the E.Z.N.A. DNA kit (OmegaBiotech) following the manufacturer's instructions. Three mitochondrial regions, mtMutS, a homologue of the bacterial DNA mismatch repair gene mutS (=msh1), the NADH-dehydrogenase subunits 2 (ND2) and the cytochrome c oxidase subunit I (COI), plus a nuclear region, the large subunit ribosomal RNA gene (28S) were sequenced. These four markers were concatenated, representing the largest multiloci segment used today in sea pen phylogenies, previous contributions used mtMutS+ND2 (Dolan et al. 2013;Kushida and Reimer 2018) or mtMutS+COI+28S (García-Cárdenas et al. 2020). The start of the mtMutS region was amplified using the primers ND42599F and MUT3458R (France and Hoover 2002;Sánchez et al. 2003). The start of the ND2 region was amplified using the primers 16S47F and ND2-1418R (McFadden et al. 2004). The COI region was amplified using the primers COII8068F and COIOCTR (France and Hoover 2002;McFadden et al. 2004). The 28S nuclear ribosomal gene (28S rDNA) was amplified using the primers 28S-Far and 28S-Rar (McFadden and van Ofwegen 2013). Each PCR used 1 U of MyTaq Red DNA Polymerase (Bioline), 10 μM of each primer, approximately 30 ng of genomic DNA, and was brought to a final volume of 25 μL with H 2 O for molecular biology (PanReac-AppliChem). MtMutS PCR was carried out using the following cycle profile: initial denaturation at 95°C for 1 min, 35 cycles of denaturation at 95°C for 15 s, annealing at 55°C for 15 s, and extension at 72°C for 10 s, and a final extension at 72°C for 5 min. The ND2, COI, and 28S PCRs used the same cycle profile, but 51°C, 58°C, and 50°C as annealing temperatures, respectively. PCR products were purified using ExoSAP-IT™ PCR Product Cleanup Reagent (ThermoFisher Scientific) following the manufacturer's instructions, before strong amplifications were sent to Macrogen Spain for sequencing in both directions. All chromatograms were visualized and sequence pairs matched and edited using Sequencher v4.0.
A preliminary Maximum Likelihood (ML) comparison based on mtMutS (ca. 400 pennatulacean sequences, not shown) placed the studied taxa here close to a controversial sequence initially attributed to Pennatula sp. (GenBank accession number DQ302870, Li et al. 2021, López-González 2021 and the recently proposed genus and species Alloptilella splendida (GB MZ198005), within an informally named Clade 2 (Dolan et al. 2013;Kushida and Reimer 2018). These initially identified main Clades were referenced using Roman numerals (as Clade II in this case) by García-Cárdenas et al. (2019, López-González (2021), and López-González and Drewery (2022). Since that Pennatula sp. specimen is studied and identified here as Pennatula moseleyi Kölliker, 1880, it is now proposed to be ascribed to the genus Alloptilella (see below). This molecular placement is in agreement with morphological features observed in this material, and diagnosed by Li et al. (2021) for the genus Alloptilella. Therefore, for the present study, we will only include sequences of Clade II in order to not repeat images and ideas already discussed earlier (Li et al. 2021;López-González 2021). The available molecular information concerning the genus Alloptilella covered only mtMutS and ND2 markers (see Dolan et al. 2013 andKushida andReimer 2018 as Pennatula sp.2), but Li et al. (2021) sequenced also COI and 28S when describing Alloptilella splendida.
The set of new sequences obtained in this study and those homologous from GenBank (see Table 1) were aligned using MUSCLE, implemented in MEGA 6 (Tamura et al. 2013). The dataset included 43 mtMutS, 35 ND2, 27 COI, and 16 28S pennatulacean sequences. According to previous cited molecular phylogenies, the basal resolution of the main sea pen clades is poor. Thus, sequences of two ellisellids from GenBank were selected as an out-group. In addition to the most variable mitochondrial marker (mtMutS), three multiloci datasets were analysed: mtMutS+ND2, mtMutS+ND2+COI, and mtMutS+ND2+COI+28S. All the aligned matrices had 45 sequences (43 pennatulaceans and two ellisellids), but varied in the total number of bases: 698 for mtMutS, 1239 for mtMutS+ND2, 2013 for mtMutS+ND2+COI, and 2846 for mtMutS+ND2+COI+28S.
The phylogeny reconstruction was obtained by applying Maximum Likelihood and Bayesian inference (BI) methods. After alignment, pairwise genetic distances based on the Kimura 2-parameter (K2P) model of nucleotide substitution (Kimura 1980) were obtained in order to compare them with previous analyses at genus and family levels, following the comparisons of Pante and France (2010), Pante et al. (2012), and López-González (2020). The best nucleotide substitution model was selected using Modeltest implemented in MEGA6, according to Akaike Information Criterion (AIC) and hierarchical likelihood ratio test (hLRT) values. The Maximum Likelihood method was carried out in MEGA6 using the NNI (Nearest Neighbour Interchange) heuristic method and 1000 bootstrap replications. The selected nucleotide substitution model was GTR+G for all multiloci and mtMutS matrices. The Bayesian Inference was carried out with MrBayes v3.1.2 (Huelsenbeck and Ronquist 2001;Ronquist and Huelsenbeck 2003), using the substitution model GTR+G (lset nst=6 rates=gamma) and 10 7 generations, and discarding 25% of the initial trees. According to Hillis and Bull (1993), for ML bootstraps, those values <70% should be considered as low, between 70 and 94% moderate, and ≥95% high. According to Alfaro et al. (2003), for Bayesian posterior probabilities, those values <0.95 should be considered as low and ≥0.95 as high. The stationarity of the chains and convergence of the two runs were monitored for each parameter by Tracer (v.1.7.1) (Rambaut et al. 2018), determining whether the effective sample size (ESS) of all parameters was larger than 200, as recommended.

in bold)
Colonies elongated and slender, pinnate. Axis circular in cross section, present throughout length of colony. Rachis bilaterally symmetrical, with a central dorsal track naked of zooids. Rachis-peduncle limit with a distinct thickening or swelling that forms an edged ring at the thickest point. Polyp leaves large and conspicuous, narrow to deltoid, opposite or alternately disposed and inserted obliquely, extending ventrally upward. Autozooids in a row along the ventral edge of polyp leaves. Anthocodiae retractile into permanent spiculiferous calyces alternatively placed (giving the false impression of being two series). Calyces tubular, with eight terminal teeth (sometimes of unequal development). Mesozooids in a ridge in continuation of the dorsal base of polyp leaves, and coordinated with siphonozooid patches, basally. Siphonozooids in welldefined patches (V-shaped to check mark-shaped) part on latero-dorsal side of rachis and part between polyp leaves, along the distal insertion of each polyp leaf, on rachis, not on polyp leaves surfaces. Sclerites as three-flanged needles on calyces, polyp leaves and rachis; inconspicuous three-flanged rods on peduncle and pharynx. Y-shaped, crossed and 5 or 6 radiates can be present in peduncle, minute bodies 0.02 mm can also be present. Three-flanged needles on autozooid tentacles can be present or absent.

Distribution and depth
Our knowledge on the distribution of this genus is limited, due to its recent description, the scarce number of collected specimens, the few species currently known and its distant type localities. West Pacific (tropical (~9°N, Y3 seamount), western and central Tasman Sea) and West Atlantic (Puerto Rico, Caribbean Sea); 559 to~1733 m depth (Kölliker 1880;Li et al. 2021;present account). Note that all GenBank sequences are considered here with the names as they appear in GenBank and their original publications (including numbers or letters). Annotations between parentheses (1, 2, … sp.1, sp.2) after the species name are included by the present author just to identify different specimens of a given species or different species under the same denomination in GenBank  underdeveloped autozooid calyces (udau) and single line of siphonozooids between two consecutive polyp leaves; c rachis, lateral view, showing the "check mark" shape of siphonozooids fields (msz), on the right to the dorsal naked track, and on the left those rows placed between consecutive polyp leaves; d detail of an autozooid calyx with teeth (tth) and autozooid tentacles (te) with a narrow band of red sclerites (tsc); e terminal part of an autozooid cut longitudinally to observe the white tentacles (te) with tentacular sclerites (tsc), calycular teeth (dp), and pharynx with red sclerites (phs); f detail of sclerites from peduncle (light microscopy) showing minute bodies (mb) and rods (rd) (see also Figure 5) López-González 2021; Li et al. 2021). The placement of the genus Alloptilella in the family Pennatulidae is here considered for practical reasons, to be in agreement with previous literature, and gross morphologies attributed to Pennatulalike forms, there being a need for a more global revision of the family delimitations of the pennatulacean genera reunited in the informally named Clade II.

Morphological description
Colony pinnate, elongated and erect, approximately 952 mm in length in preserved state, fragmented in four parts (Fig. 1a). Rachis bilaterally symmetrical, 854 mm in length (89.7% of overall length) and 9 mm in width (measured at mid-length of rachis, without polyp leaves), with a moderate thickening at rachis-peduncle limit (10 mm in width), more or less distinct in preserved state (Fig. 1a). Peduncle 98 mm in length (10.3% of total length). Rachis with a distinctive naked central dorsal track and a hidden ventral track occupied by insertions of polyp leaves. Rachis with approximately 126 pairs of nearly opposite polyp leaves (the fragmented and contracted states of the type make a more accurate count difficult), with the lowest pairs gradually decreasing in size ( Fig.  1b), averaging 18 pairs per 100 mm of length (at midrachis length). Polyp leaves inserted obliquely, extending ventrally upward. Fully developed polyp leaves narrow in lateral view, maximum length 40 mm, and dorsal edge slightly shorter than ventral edge, maximum width 6 mm (Fig. 1b, c). Axis present throughout colony, circular in cross section, 2.7 mm in diameter measured 12 cm above rachis-peduncle limit (I preferred not to further fragment this specimen looking for the value at the rachis-peduncle limit zone).
Autozooids with calyces, up to 28 in number in the largest polyp leaves, arranged in one row along the ventral edge of each developed polyp leaf (indicated by arrangement of gastrovascular cavities in transversal sections of the polyp leaves), but oriented alternately, appearing falsely to be placed in two rows (Fig. 1c). Anthocodiae retractile into permanent spiculiferous calyces. Calyces tubular, spiculate, armed with 8 strongly projecting teeth, up to 4 mm in length (including teeth) and 1.5 mm in maximum width at basis of teeth level (Figs. 1c and 2d, e), teeth up to 1.6 mm in length (usually around 1.1 mm). Four to six under-developed autozooids on the proximal ventralmost edge of each polyp leaf (Figs. 1d and Fig. 4 Alloptilella moseleyi comb. nov. (NTM-C014415). SEM photographs of sclerites; a Polyp leaves; b mesozooids; c siphonozooids; d rachis, dorsal track at mid length; e rachis, just above the rachis-peduncle limit 2a, b). Mesozooids distributed on the lateral side of rachis, in an elevated ridge of one line (two lines proximally) (Fig. 1b, c), along the rachis and proximal dorsal edge of polyp leaves, up to 14 in number, decreasing in size from polyp leaf to dorsal track, the last ones of similar size to siphonozooids, the largest one conical, 1.4 mm in height and 0.8 mm in width, spiculiferous. Siphonozooids minute, 0.17-0.26 mm in diameter (average 0.22 mm, N=20), numerous, arranged in~10 rows along the axillae of polyp leaves to the latero-dorsal sides of the rachis between polyp leaves, resembling a "check mark" where the long stroke is placed between consecutive polyp leaves (Figs. 1c, d, and 2c), appearing as a single line on the ventral side (Fig. 2b).
Sclerites distributed in various parts of the colony. Densely placed in peduncle, autozooid calices, around mesozooids and siphonozooids openings, and dorsal naked tracks (Figs. 1b, c, and 2c), but widely spaced in lateral (actually they are distal and proximal) sides of polyp leaves (Fig. 2a). Along the aboral side of tentacular axis the sclerites are disposed in a narrow line of 2-3 sclerites (Fig. 2d). Overall, in the proximal part of the rachis, the spicules are more densely packed than in the distal part, and the mesozooid ridges are better defined (Fig. 1b, d).
Sclerites from calyx body as three-flanged needles, up to 0.59 mm in length (Fig. 3a). Sclerites from calycular teeth as three-flanged needles, up to 0.97 mm in length (Fig. 3b). Sclerites from aboral sides of autozooid tentacle axis as three-flanged needles, up to 0.21 mm in length (Fig. 3c, e). Sclerites from autozooid's pharynx as three-flanged needles, up to 0.06 mm in length (Fig. 3d, f). Sclerites from polyp . SEM photographs of sclerites; a Peduncle, just below the rachispeduncle limit; b peduncle, midlength, notice minute bodies (*); c peduncle, lower end, notice minute bodies (*); d and e detail from two sclerites from c, a threeflanged rod with a minute body (*) and a 6-radiated twinning; f magnified minute bodies present in the lower part of peduncle, note also their three-flanged structure leaves as three-flanged needles, up to 1.1 mm in length (Fig. 3e). Sclerites from mesozooids as three-flanged needles, up to 1.0 mm in length (Fig. 4a). Sclerites from siphonozooids as three-flanged needles, up to 0.9 mm in length (Fig. 4b). Sclerites from mid-rachis dorsal track as three-flanged needles, up to 0.38 mm (Fig. 4c). Sclerites from the lowest part of the rachis, just over the rachispeduncle limit as three-flanged needles, up to 0.37 mm (Fig. 4d). Sclerites from peduncle as three-flanged rods up to 0.2 mm in length and some crosses up to 0.15 mm in length (Fig. 5a). Sclerites from mid-length of peduncle as three-flanged rods and some crosses, both types up to 0.16 mm in length (Fig. 5b), and a few minute bodies0 .02 mm (see Fig. 5b asterisk). Sclerites from lower basal end of peduncles as three-flanged rods, up to 0.13 mm in length, with numerous crosses, even 3-, 5-, and 6twinning (Fig. 5c, d, e), and abundant minute bodies0 .02 mm (Figs. 2f, and 5d arrow, f).

Colour
Preserved colony is red, depending on the density of dark red sclerites (Fig. 1a). Red sclerites are present in dorsal track, polyp leaves, autozooid calyces, pharynx, mesozooids, siphonozooids, and aboral side of tentacles. Sclerites from peduncle are reddish at the upper parts and light red to translucent at the mid and proximal ends.

Morphological description of the holotype
Colony pinnate, elongated and erect (Fig. 6a) 620 mm in length in the preserved state (Fig. 6b, c). Rachis bilaterally symmetrical, 53 mm in length (~85% of overall length) and 5 mm in width (measured at mid-length of rachis, without polyp leaves), with a moderate thickening at the rachispeduncle limit (8 mm in width), more or less distinct in the preserved state (Fig. 6a). Peduncle 90 mm in length (~15% of total length). Rachis with a distinctive naked central dorsal track and a hidden ventral track occupied by insertions of polyp leaves, with a few siphonozooids in the space between consecutive polyp leaves (Fig. 6c). Rachis with approximately 50 pairs of opposite polyp leaves, with the lowest pairs gradually decreasing in size (Figs. 6a, and 7a, c), averaging 8 pairs per 100 mm of length (at mid-rachis length). Polyp leaves inserted obliquely, extending ventrally upward (Fig. 6c, d).
Fully developed polyp leaves close to right/obtuse triangular in lateral view, maximum length 50 mm, and dorsal edge slightly shorter than ventral edge, maximum width 20 mm (Fig. 6d, e). Axis present throughout colony, circular in cross section, 4 mm in diameter measured 10 mm above the rachispeduncle limit.
Autozooids with spiculiferous calyces, up to 29 in number in the largest polyp leaves, arranged in one row along the ventral edge of each developed polyp leaf (indicated by arrangement of gastrovascular cavities in transversal sections of the polyp leaves), but oriented alternately, appearing falsely to be placed in two rows (Figs. 6e,f,and 7c,d,e,f). Anthocodiae retractile into permanent spiculiferous calyces. Calyces tubular but wider distally, spiculate, armed with 8 strongly projecting teeth (sometimes of different development), up to 4.1 mm in length (including teeth) and 1.7 mm in maximum width at basis of teeth level (Figs. 6f, and  7d, e, f), teeth up to 0.95 mm in length (usually around 0.8 mm). Autozooids of lowermost polyp leaves increasing in size from proximal to distal in the polyp leaf, quickly reaching a high number of calyces (Fig.  7b). Mesozooids distributed on the dorsal side of the rachis, in a low ridge of a single line (rarely two in the proximities of the polyp leaf base), along the rachis and proximal dorsal edge of polyp leaves,~14 in number (coordinated with the lowermost polyp leaves, see Fig. 11e), decreasing in number from the proximal to the distal part of the rachis, decreasing in size from Whole colony, specimen damaged at rachis-peduncle limit; b detail of naked dorsal track (ndt) and latero-dorsal fields of siphonozooids (sph), and reduced number of mesozooids (msz), one or two of them on the proximal dorsal edge of polyp leaves, and lateral side of polyp leaves (pll) showing low density of sclerites, as well as the opposite disposition of polyp leaves; c reduced ventral naked track (vnt), showing two opposite polyp leaves with reduced in size proximal autozooids (aut) and line of single siphonozooids between consecutive polyp leaves (sph), note also the low density of red sclerites on the lateral sides of polyp leaves; d rachis mid-length in lateral view showing some of the largest polyp leaves; e detail of a couple of consecutive polyp leaves (ventral vision of rachis) showing distribution and density of red and yellow sclerites; f detail of some autozooid calyces with teeth, notice also the distribution of red sclerites basally. Photos a and d by T. Coffer USNM polyp leaf to dorsal track, the last ones of similar size to siphonozooids, difficult to distinguish (Fig. 7h). One of the two largest ones is conical, on the proximal part of the dorsal edge of polyp leaves, approximately up to 1.5 mm in height and 0.5 mm in width, spiculiferous (Figs. 6b, 7h, and 11d arrows). Siphonozooids minute, 0.2-0.3 mm in diameter (average 0.24 mm, N=20), numerous, arranged in approximately up to 6-8 rows dorsally, along the axillae of polyp leaves to the laterodorsal sides of the rachis between polyp leaves, resembling a "check mark" where one of the strokes is placed between consecutive polyp leaves, and the other runs dorsally, proximally limited by the ridge of mesozooids (Figs. 6b, 7h, and 11d, e, f). Siphonozooids usually do not form a continuous band dorsally, each dorsal "patch" decreases in the number of rows until just a single one, the band running between consecutive polyp leaves also decreases in number of rows, ending in a single one, which reaches the ventral track (Fig. 6c).
Sclerites distributed in various parts of the colony. Sclerites are densely placed on peduncle, around mesozooid and siphonozooid openings (Fig. 6b), but widely spaced in dorsal naked tracks (Fig. 6b), lateral (actually they are distal and proximal) sides of polyp leaves (Fig. 6c). Sclerites are completely absent along the aboral side of tentacular axis and pinnulae (Fig. 7f, g).
Sclerites from calyces body as three-flanged needles, up to 0.6 mm in length (Fig. 8a). Sclerites from calycular teeth as three-flanged needles, up to 1.2 mm in length (Fig. 8b). Sclerites from autozooid's pharynx as three-flanged needles, up to 0.06 mm in length (Fig. 8c, d). Sclerites from polyp leaves as three-flanged needles, up to 0.75 mm in length (Fig. 9a). Sclerites from mesozooids as three-flanged needles, up to 0.95 mm in length (Fig. 9b). Sclerites from siphonozooids as three-flanged needles, up to 0.57 mm in length (Fig. 9c). Sclerites from the mid-rachis dorsal track as three-flanged needles, up to 0.13 mm (Fig. 9d). Sclerites from the lowest part of rachis, just above the rachis-peduncle limit as three-flanged needles, up to 0.14 mm (Fig. 9e). Sclerites from peduncle as three-flanged rods (sometimes pointed, flanges mainly noticeable at the ends, often simply as slits) and some Y-shaped and crosses, up to 0.055 mm in length (Fig. 10a, b). Sclerites from mid-length of the peduncle are similar in shape diversity and size to that of the upper part, 5 or 6 radiated can be present (Fig. 10c, d). Sclerites from the lower distal end of the peduncle mainly as minute bodies (~0.02 mm) and scarce rods of similar shape and size to those in the upper parts of the peduncle.

Colour
The preserved colony is reddish, depending on the density of dark red sclerites (Figs. 6a and 11). Red sclerites are present in dorsal track, polyp leaves, basal part of autozooid's calyces, pharynx, and siphonozooids. Sclerites from distal part of calyces, including calycular teeth, are yellow. Sclerites from mesozooids are mainly yellow, but some reddish and some bicoloured (ends yellow and centre red) are also present. Sclerites from the peduncle are light red to translucent in colour.

Living habitus and environment
The holotype of Alloptilella williamsi sp. nov. was collected by the ROV Deep Discoverer (Fig. 11a) (see also all photographic information associated with the registration code USNM 1550625 on the Web site "Search the Department of Invertebrate Zoology Collections" of the of the Smithsonian Institution, National Museum of Natural History (https:// collections.nmnh.si.edu/search/iz/)). The fully expanded specimen (Fig. 11b) combining the deep and light red colours (depending on the density of red sclerites) with the distinct series of yellow autozooids on the ventral edge of polyp leaves, as well as the yellow ridges of mesozooids on laterodorsal surfaces (Fig. 11d-e), the rachis peduncle limits (Fig.  11e) is moderately expanded, as well as the polyp-free dorsal track and mid to lower polyp leaves (Fig. 6e). The autozooids, placed in a single row on the ventral edge of polyp leaf, are alternatively oriented (see Fig. 6c). The rachis of the type material presents damages at mid-length, probably due to extractive human activities on these bottoms, showing also clear signs of recovery (Fig. 11c). According to the expedition report (Wagner et al. 2019: p. 72), the habitat observed during the EX1811-Dive02 included moderate slopes with thin sediment (Fig. 11b), associated benthic fauna included, apart from some sponges (hexactinelids and demosponges) and the abundant echinoderms (crinoids, cidarids, brittle stars, and sea stars), hydrocorals, octocorals, antipatharians, and solitary scleractinians (not in high density). The octocorals Detail of the rachis-peduncle limit, latero-ventral view, showing the progressive decreasing size of polyp leaves; b detail of one of the nearly fulldeveloped polyp leaves at the proximal part of rachis, showing the progressive reduction in size of autozooid calyces; c detail of the rachispeduncle limit, latero-dorsal view, showing the progressive decreasing size of polyp leaves, the larger size of autozooid at dorsal placements, and the pronounced mesozooid ridges (msr); d detail of autozooid calyces, notice teeth and red sclerites basally; e part of an autozooid calyx cut longitudinally in clove oil to observe sclerite arrangement and unequal development of calycular teeth (tth); f distal part of autozooid cut longitudinally, showing unequal development of teeth (tth), tentacles without sclerites (te), and pharynx with red sclerites (phs); g half-distal part of a tentacle in clove oil showing internal hollow canals (ihc) along main tentacular axis and pinnulae, as well as absence of sclerites; h dorsal view of mesozooid ridges (msr) coordinated with siphonozooid (sph) fields of three consecutive polyp leaves, notice yellow colour of large mesozooids (msz) on the dorsal proximal edge of polyp leaves, and the scarce density of red sclerites on its lateral sides (pll)

Etymology
The species epithet is chosen in honour of Dr. Gary C. Williams, curator of invertebrates at the California Academy of Sciences (California, USA), in recognition of his important Fig. 9 Alloptilella williamsi sp. nov. Holotype (USNM 1550625). SEM photographs of sclerites; a Polyp leaves; b mesozooids; c siphonozooids; d rachis, dorsal track at mid-length; e rachis, just above the rachis-peduncle limit contribution to our knowledge of pennatulacean diversity and worldwide distribution (Williams 2011).

Geographical and depth distribution
At present, Alloptilella williamsi sp. nov. is only known from the Caribbean Sea (NW Atlantic) from 559.10 m depth.

Phylogenetic analyses
In the mtMutS+ND2+COI+28S hypothesis (Fig. 12), the BI and ML topologies are identical. The multiloci sequences obtained for Allopitlella moseleyi comb. nov and A. williamsi sp. nov. were reunited in a distinct and well-supported clade (Bootstrap (Bst) 99%, Posterior Probability (PP) 1) with other multiloci sequences recently published as Alloptilella splendida, the type material of the type species of the genus Alloptilella. Alloptilella is then easily identifiable as a monophyletic genus, being the sister group of all sequences of another distinctly morphological and molecular genus, Scytalium Herklots, 1858. Alloptilella and Scytalium are in a well-supported clade (Bst 97%, PP 1) which is the sister group of a poorly supported Ptilella-Distichoptilum clade (Bst 29%, PP 0.51). The position of the genus Pennatula in this phylogenetic hypothesis is distinctly separated from Ptilella and Alloptilella, Ptilosarcus being the sister group of Pennatula, with a low to moderate support from the ML inference (Bst 68%), but a high support (PP 0.99) from the Bayesian inference. Different genera currently included in the families Renillidae (Renilla), Virgulariidae (Acanthoptilum, Actinoptilum), and Stachyptilidae (Gilibelemnon) are placed in between the Pennatula-Ptilosarcus and the Ptilella-Distichoptilum and Alloptilella-Scytalium subclades (see Fig. 12). Fig. 10 Alloptilella williamsi sp. nov. Holotype (USNM 1550625). SEM photographs of sclerites; a Peduncle, just below the rachispeduncle limit; b cross from a, magnified; c peduncle, midlength; d rods and cross from c, magnified The topologies obtained using the BI and ML methods for the mtMutS, mtMutS+ND2, and mtMutS+ND2+COI hypotheses are almost identical (SMFigs. 1, 2), and somewhat more resolved (but moderately supported) in the multiloci datasets. The three genera discussed here (Pennatula, Ptilella, and Alloptilella) are well supported in all BI trees (PP 1), while Bst values (ML trees) varied depending on the dataset used (98,91,and 94% for Pennatula;100,99,and 100% for Ptilella;and 92,99,and 96% for Alloptilella, for each of the above listed hypotheses, respectively).
In the mtMutS tree (SMFig. 1, left), species distinction is not always observed for all Ptilella and Alloptilella species. The three Pennatula species used for this tree are clearly differentiated, while P. grandis and P. inflata share identical sequences, and only P. grayi is differentiated. Concerning Alloptilella species, A. moseleyi comb. nov. and A. williamsi sp. nov. share the same sequence, and then, only A. splendida is differentiated. As in the four markers hypothesis, Scytalium is the sister group of Alloptilella (PP 1, Bst. 96%).
In the mtMutS+ND2 hypothesis (SMFig. 1, right), the topologies of the BI and ML trees are also very similar and are in general agreement with the mtMutS hypothesis. However, in this case, all the Ptilella and Alloptilella species are differentiated (more clearly visible in Ptilella, where at least three colonies per species are sequenced). Ptilella grandis and P. inflata are sister groups (PP 1, Bst 96%), and they are the sister group of P. grayi (PP 1, Bst 99%). A specimen identified as P. cf. inflata from eastern North Atlantic (GenBank MK919666, complete mitochondrial sequence) is here clearly aligned with P. grandis. Ptilella inflata is recognized as a southern West Indian species, also present in South Africa and Namibia (Kükenthal 1910;Williams 1995;López-González et al. 2001). The present sequenced colonies of P. inflata were collected off Namibian coasts, 418-439 m in depth (López-González et al. 2001). In this hypothesis, Alloptilella species are minimally differentiated from a molecular point of view (see "Discussion") forming a polytomy. As in the mtMutS hyphothesis, Pennatula species are well defined.
A phylogenetic hypothesis considering only mitochondrial markers (mtMutS+ND2+COI) have the same main topology, but is reinforced at species level by the additional differences in COI marker (SMFig. 2). Ptilella and Pennatula species remain well separated, while Alloptilella species also maintain the same polytomy observed in mtMutS and mtMutS+ND2 hypotheses.
In order to provide additional comparison, the most common multiloci dataset previously analysed in pennatulacean phylogenetic papers (mtMutS+ND2 and mtMutS+COI+28S) has also been explored using the K2 nucleotide substitution model (see SMTables 1 and 2). In the mtMutS+ND2 data matrix, the mean intraspecific genetic distances in Alloptilella, Ptilella, and Pennatula are 0.08%, 0.33%, and 0.27% respectively, while Alloptilella is 2.22% distant from Ptilella and 3.61% distant from Pennatula, and Pennatula is 2.96% distant from Ptilella (see SMTable 1 for ranges). In the analysis of the mtMutS+COI+28S data matrix, the mean intraspecific genetic distances in Alloptilella, Ptilella, and Pennatula are also low at 0.02%, 0.39%, and 1.06%, respectively, while Alloptilella is 27.62% distant from Ptilella and 29.11% distant from Pennatula, and Pennatula is 9.54% distant from Ptilella (see SMTable 2 for ranges). A final analysis including the multiloci mtMutS+COI (SMTable 3) corroborates the idea that genetic distances at genus level are improved by the inclusion of the 28S segment, as overall genetic distances are similar and of the same order of magnitude as those obtained using the mtMutS+ND2 dataset.
Artificial key to Alloptilella species (see also

Discussion
The sequences in GenBank identified as Pennatula sp. (DQ302870 for mtMutS and DQ302943 for ND2) belong to the same material here identified as Pennatula moseleyi Kölliker, 1880. According to the current knowledge, the assignation of this material to Kölliker's species is reasonable, and is also the most conservative proposal. This species is transferred here to the genus Alloptilella as A. moseleyi comb. nov. Due to its controversial position in the published phylogenetic trees (e.g. Dolan et al. 2013;Kushida and Reimer 2018;García-Cárdenas et al. 2020;López-González 2021;Li et al. 2021), total DNA of that specimen was extracted and sequenced again for this paper; only the previously known segments (mtMutS and ND2) were successfully amplified and Fig. 11 In situ photographs of Alloptilella williamsi sp. nov. Holotype (USNM 1550625); a ROV Deep Discoverer on the bottom during the captures of the type material of A. williamsi sp. nov.; b whole extended colony (lasers are 10 cm apart), notice damage at the mid-rachis length; c detail of damage on the rachis; d detail of the upper part of rachis, polyp leaves extended, notice the shape of siphonozooid fields, alternate orientation of autozooids, low density of sclerites on the polyp leaves' lateral surfaces, and poor development of mesozooid ridges, notice one or two mesozooids on the proximal dorsal edge of each polyp leaf (arrows); e detail of the expanded rachis-peduncle limit and lowest polyp leaves, notice the progressively decreasing size of polyp leaves, and the welldeveloped and yellow-coloured mesozooid ridges; f general view of the distal part of rachis, showing the opposite disposition of polyp leaves; g precise moment of collection by the arm of the ROV Deep Discoverer.  Table 2 for complete list of species and GenBank accession numbers. *This specimen is probably conspecific with P. grandis (see "Discussion", SMFigs.1 and 2) Table 2 Comparative table of Alloptilella species discussed in this paper. Alloptilella species are listed according to their date of description. Notes: n.d., no data; (1) Kölliker (1880: p. 6); (2) Li et al.  A. splendida (2) A. williamsi sp. nov. Sclerites of lowermost part of rachis, above rachis-peduncle limit Three-flanged needles, up to 0.37 mm at the lowest part n.d.
Three-flanged needles, up to 0.14 mm at the lowest part, more robust (wider) than upper part (see Fig. 9 D and E comparatively) Sclerites of peduncle Short three-flanged rods up to 0.2 mm in length (slightly shorter at mid-length), some crosses up to 0.15 mm in length, more abundant at the mid and basal parts (where they can be 5 or 6 radiated), and minute bodies~0.02 mm in length, also more abundant at basal end Short three-flanged rods, up to 0.08 mm in length Short three-flanged rods (sometimes pointed, flanges mainly noticeable at the ends, often simply as slits) up to 0.06 mm in length (of similar size at mid length), Y-shaped sclerites, crosses up to 0.055 mm in length, are present, and become more abundant and even, and 5 or 6-radiated at the middle parts. sequenced. Both obtained sequences were in agreement with the previously published ones.
Alloptilella is identifiable as a monophyletic genus, being the sister group of all sequences of another distinct morphological and molecular genus recently revisited, the genus Scytalium (see Li et al. 2021;López-González 2021). The resurrection of the genus Ptilella (see García-Cárdenas et al. 2019) and the description of the genus Alloptilella resolve the polyphyletic situation of Pennatula observed in recent phylogenetic studies focused on the order Pennatulacea (e.g. Dolan et al. 2013;Kushida and Reimer 2018;García-Cárdenas et al. 2020). The realignment of Pennatula species in three different genera (based on molecular and morphological features) does not resolve the fact that the numerous named species of Pennatula are still in need of revision. Attending to the relatively high numbers of proposed species and the potential number of species recognizable as valid nowadays (Williams 1995(Williams , 2011, this is a task to be carried out in the future with the help of all available taxonomic tools. Molecular information seems to be essential in that revision, or at least the re-description of the already supposedly wellknown species, such as Pennatula phosphorea Linnaeus, 1758, P. aculeata Danielssen, 1860, and P. rubra (Ellis, 1764), among others (see García-Cárdenas et al. 2019).
The description of the genus Alloptilella was based on Alloptilella splendida, a species inhabiting seamounts in the Tropical Western Pacific at up to 879 m depth. The erection of a new Pennatula-like genus from a molecular point of view was completely necessary. This species is morphologically and chromatically close to Pennatula naresi Kölliker, 1880(see Kölliker 1880. García-Cárdenas et al. (2019) suggested that the latter species could belong to the genus Ptilella, but more morphological and molecular studies were necessary. The original description of P. naresi included enough details on the distribution and colour of the sclerites in the autozooids, mesozooids (named ventral zooids), and siphonozooids (named lateral zooids). The gastrovascular cavities of autozooids are arranged in the polyp leaves in a single line, and the yellow calyces are oriented in alternate directions giving the impression of being arranged in two series. The mesozooids, armed with yellow sclerites, form a row of polyps decreasing in size "which begin at the ventral margin of the pinnule at a little distance from its attachment, run obliquely upon the sides of the rachis, and end with a longitudinal streak". The siphonozooids occupy the space between mesozooid rows also running between consecutive polyp leaves. Kölliker (1880) stated that only autozooids and mesozooids (as ventral zooids) are armed with yellow sclerites, as A. splendida. Li et al. (2021) mainly differentiate A. splendida and P. naresi in the placement of mesozooids, restricted to the dorsal area in A. splendida and also occupying the proximalmost part of the polyp leaf edge dorsally. The taxonomic placement of P. naresi is controversial from a morphological point of view. The arrangements of autozooids, mesozooids, and siphonozooids agree with Alloptilella; hence, a complete morphological redescription of its type material is desirable.
Identification by DNA sequencing of the three morphologically similar genera is well established, although molecular differentiations among the three Alloptilella species are low. The mtMutS sequences of the Caribbean (A. williamsi sp. nov.) and Tasman (A. moseleyi comb. nov.) species are identical, while A. splendida (Tropical Western Pacific) differs in a single base. The ND2 sequences of A. williamsi sp. nov. and A. moseleyi comb. nov. (not available for A. splendida) differ also in a single base. The COI sequences of A. splendida and A. williamsi sp. nov. differ in two bases (only 539 bases are known for A. splendida). Remarkably, the nuclear 28S differs in a single base between A. splendida (Tropical Western Pacific) and A. williamsi sp. nov. (Caribbean Sea) (only 677 bases are known for A. splendida), while no information is currently available for A. moseleyi comb. nov.
A similar situation occurs among the three species of Ptilella (P. grandis, P. grayi, and P. inflata) (see Table 2, Fig. 12, SMFigs. 1, 2 in this paper and García-Cárdenas et al. 2019). For this paper, two mitochondrial (mtMutS and ND2) segments and one nuclear (28S) segment from SW Atlantic specimens of Ptilella inflata have been sequenced. Ptilella grandis and P. inflata share an identical sequence of mtMutS, while P. grayi shows differences in two bases. ND2 sequences of P. grandis differ in four bases with P. grayi, but only in two with P. inflata, while P. grayi and P. inflata differ also in two bases. Two bases differentiate P. grandis and P. grayi COI sequences (not available for P. inflata). Surprisingly, differences in the nuclear 28S are also low, P. grayi differs in three bases with P. grandis, and in four bases with P. inflata, while P. grandis and P. inflata only differ in a single base.
From a chromatic point of view, Alloptilella splendida and A. williamsi sp. nov. share the same distribution pattern of coloured sclerites. Autozooids and mesozooids have yellow sclerites, while the rest of the sclerites in the colony are red, being light red or translucent in the lower parts of the peduncle. In the case of A. moseleyi, sclerites in all parts of the rachis are red, having a similar colour pattern in the peduncular sclerites to its congeners. However, A. splendida and A. moseleyi have red three-flanged sclerites along the aboral side of the tentacles, while in A. williamsi sp. nov. tentacular sclerites are completely absent.
From a morphological point of view (apart from the abovementioned chromatic differences), both western Pacific species (Allopilella moseleyi and A. splendida) have a similar appearance (see Table 2 for a complete comparative overview). The polyp leaves seem to be placed in a more opposite than alternate arrangement in A. moseleyi. Polyp leaves seem to be narrowest and proportionally longer in Schiaparelli, Annenina Lortz, Julian Gutt, Enrique Isla, Victor Díaz-del-Río, and José Luis Rueda. The study of the Antarctic specimens and the final conception of this paper were carried out under the project CTM2017-83920-P (DIVERSICORAL) funded by the Spanish Ministry of Economy, Industry and Competitiveness. The author also acknowledges José Martín Garrido and María Encarnación Rubio Pérez for their technical support in completing DNA extractions and amplifications of Ptilella specimens. Thanks are also due to Yang Li and Kuidong Xu (Institute of Oceanology, Chinese Academy of Sciences) for their kind comments and additional detailed graphic information on the type material of Alloptilella splendida. Mr. Tony Krupa is thanked for reviewing the English version. Finally, the author would like to thank the two anonymous reviewers and the Editorial Office of Marine Biodiversity for all their constructive comments and suggestions, which helped improve the quality of an early version of the manuscript.
Funding Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. The collection of some of the specimens here studied was carried out thanks to the Spanish Projects POL2006-06399/CGL (Polarstern ANT XXIII/8 -CLIMANT), and LIFE07/NAT/ E/000732 LIFE+INDEMARES. This research is supported by the project CTM2017-83920-P (DIVERSICORAL), Spanish Ministry of Economy, Industry and Competitiveness.

Declarations
Conflict of interest The author declares no competing interests.
Ethical approval All applicable international, national, and/or institutional guidelines for animal testing, animal care, and use of animals were followed by the author.
Sampling and field studies All necessary permits for sampling and observational field studies have been obtained by the author (or responsible researchers of the different research programmes) from the competent authorities and are mentioned in the acknowledgements.
Data availability The data generated and analysed during this study are deposited in public repositories.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.