Chinese Journal of Oceanology and Limnology   2017, Vol. 35 issue(4): 803-814     PDF       
http://dx.doi.org/10.1007/s00343-017-5320-5
Institute of Oceanology, Chinese Academy of Sciences
0

Article Information

LI Yang(李阳), ZHAN Zifeng(詹子锋), XU Kuidong(徐奎栋)
Morphology and molecular phylogeny of Paragorgia rubra sp. nov. (Cnidaria: Octocorallia), a new bubblegum coral species from a seamount in the tropical Western Pacific
Chinese Journal of Oceanology and Limnology, 35(4): 803-814
http://dx.doi.org/10.1007/s00343-017-5320-5

Article History

Received Nov. 6, 2015
accepted in principle May. 6, 2016
accepted for publication Dec. 15, 2016
Morphology and molecular phylogeny of Paragorgia rubra sp. nov. (Cnidaria: Octocorallia), a new bubblegum coral species from a seamount in the tropical Western Pacific
LI Yang(李阳)1, ZHAN Zifeng(詹子锋)1, XU Kuidong(徐奎栋)1,2,3        
1 Department of Marine Organism Taxonomy and Phylogeny, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China;
2 Laboratory for Marine Biology and Biotechnology, Qingdao National Laboratory for Marine Science and Technologys, Qingdao 266071, China;
3 University of Chinese Academy of Sciences, Beijing 100049, China
ABSTRACT: A new species of bubblegum coral, Paragorgia rubra sp. nov., discovered from a seamount at a water depth of 373 m near the Yap Trench is studied using morphological and molecular approaches. Paragorgia rubra sp. nov. is the fourth species of the genus found in the tropical Western Pacific. The new gorgonian is red-colored, uniplanar, and measures approximately 530 mm high and 440 mm wide, with autozooids distributed only on one side of the colony. Paragorgia rubra sp. nov. is most similar to P. kaupeka Sánchez, 2005, but differs distinctly in the polyp ovals with large and compound protuberances (vs. small and simple conical protuberances) and the medullar spindles possessing simple conical protuberances (vs. compound protuberances). Moreover, P. rubra sp. nov. differs from P. kaupeka in the smaller length/width ratio of surface radiates (1.53 vs. 1.75). The genetic distance of the mtMutS gene between P. rubra sp. nov. and P. kaupeka is 0.66%, while the intraspecific distances within Paragorgia Milne-Edwards & Haime, 1857 except the species P. regalis complex are no more than 0.5%, further supporting the establishment of the new species. Furthermore, the ITS2 secondary structure of P. rubra sp. nov. is also different from those of congeners. Phylogenetic analyses indicate Paragorgia rubra sp. nov. and P. kaupeka form a clade, which branched early within Paragorgia and diversified approximately 15 Mya.
Key words: gorgonian     taxonomy     new species     genetic distance     molecular phylogeny    
1 INTRODUCTION

Members in the family Paragorgiidae, commonly known as bubblegum corals, are one of the dominant megafaunal taxa in hard bottom environments, such as continental shelves and seamounts. They can form habitats for a variety of invertebrates and fishes, playing a fundamental ecological role in benthic environments (Koslow et al., 2001; Heifetz, 2002). For example, Buhl-Mortensen and Mortensen (2005) recorded a total of 1 264 individuals representing 47 species associated with 13 colonies of Paragorgia arborea (Linnaeus, 1758). Thus, exploring bubblegum corals and other habitat-forming organisms are helpful for understanding the biota of associated environments.

During the survey on the seamount benthos in the tropical Western Pacific, we discovered a remarkable red gorgonian attached to the rocky bottom at a water depth of 373 m. The species was also inhabited by a single echinoderm of the order Euryalida. The specimen was collected by the submersible Remotely Operated Vehicle (ROV) FaXian. Based on morphological and molecular studies, we described the specimen as Paragorgia rubra sp. nov. and investigated its phylogenetic position and evolution.

2 MATERIAL AND METHOD 2.1 Collection and morphological examination

The specimen was collected from a seamount (8°51.36′N, 137°46.76′E; tentatively named as Yap-3 seamount) near the Yap Trench by the submersible ROV FaXian (Discovery) during a cruise of the R/V KeXue (Science) in the tropical Western Pacific in December 2014. The specimen was photographed in situ before sampled, and stored in 75% ethanol after collection. The type specimen (Y30157) is deposited in the Marine Biological Museum of Chinese Academy of Sciences in Qingdao, China.

General morphology and anatomy were examined under a stereo dissecting microscope. Sclerites were isolated respectively from the polyps, the colony surface and the medulla by digestion of the tissues in sodium hypochlorite, and then were washed with distilled water and 70% ethanol. For each sample, 20 sclerites were randomly chosen and measured at the magnification of 100× or 200× using a light microscope. To investigate the ultrastructure of sclerites, sclerites were air-dried, mounted on carbon double adhesive tape, and coated for the Scanning Electron Microscope (SEM). SEM scans were obtained using Hitachi S-3400N SEM at 5 kV and the optimum magnification for each kind of sclerites. Terminology follows Bayer et al. (1983).

2.2 DNA extraction and sequencing

Total genomic DNA was extracted from the polyps of the only specimen using the TIANamp Marine Animal DNA Kit (Tiangen Bio. Co., Beijing, China) following instructions. Mitochondrial genomic regions examined included two different regions of the large subunit ribosomal RNA (16S) and the 5′-end of the NADH dehydrogenase subunit 2 (nad2), the 5′-end of the cytochrome coxidase subunit I (cox1), the 5′-end of the DNA mismatch repair protein—mutS— homolog (mtMutS), 3′-end of the NADH dehydrogenase subunit 6 (nad6), the nad6-nad3 intergenic spacer (int), the 5′-end of the NADH dehydrogenase subunit 3 (nad3). The nuclear genomic region examined was the complete internal transcribed spacer-2 (ITS2). The PCR products were purified using the TIANgel Midi Purification Kit (Tiangen Bio. Co., Beijing, China). The PCR amplification followed the procedure described by employing cox1 primers following Folmer et al. (1994) and McFadden (unpublished), and the rest mitochondrial and one nuclear pairs of octocoral primers following Herrera et al. (2010) (see Supplementary Table A1). Amplified and cloned DNA was sequenced in both directions with the ABI 3730 DNA Analyzer sequencing facility by Shanghai Sangon Biological Engineering and Technical Service Company, Shanghai, China. The resulting sequences GenBank numbers and lengths are as follows: 16S (KX505986, 303 bp), nad2 (KX505987, 912 bp), mtMutS (KX505989, 747 bp), nad6-int-nd3 (KX505990, 628 bp), ITS2 (KX505991, 286 bp) and cox1 (KX505988, 580 bp).

2.3 Phylogenetic analyses

Three sequence alignments were created for the subsequent genetic distance and phylogenetic analyses: (1) the mtMutS gene sequences of all available Paragorgia associated with published articles in GenBank (153 sequences from 14 of 17 species) and two Sibogagorgia (three sequences) as outgroups (Table 1, Alignment A1); (2) the concatenated sequences of mitochondrial 16S, nad2, cox1, mtMutS and nad6-int-nd3 regions mainly following Herrera et al. (2012) plus the new sequences in the present study (Table A2, Alignment A2); and (3) the ITS2 sequences of all available Paragorgia associated with published articles in GenBank (54 sequences from eight of 17 species) plus one Sibogagorgia sequence (Table A9, Alignment A3). Sequences of each region were aligned using MAFFT v.7 (Katoh and Standley, 2013) with the G-INS-i and Q-INS-i algorithms for the mitochondrial and ITS2 regions, resulting in three datasets that comprised 610 nucleotide positions for mtMutS, 2 740 nucleotide positions for the concatenated sequences and 309 nucleotide positions for ITS2, respectively. The haplotypes of the mtMutS gene and ITS2 were calculated using DNASP 5.0 (Librado and Rozas, 2009) with alignment gaps included as an informative state (Giribet and Wheeler, 1999), respectively. Genetic distances, calculated as uncorrected "p" distances within each species and among species, were estimated using the represent haplotype sequences of mtMutS and ITS2 by MEGA v.6 (Tamura et al., 2013).

Table 1 Interspecific and intraspecific uncorrected pairwise distances at mtMutS among haplotypes of the species of Paragorgia and Sibogagorgia Stiasny, 1937 (%)

The HKY+G evolutionary model was the bestfitted model for mtMutS, selected by AIC as implemented in jModeltest2 (Darriba et al., 2012). Maximum likelihood (ML) analysis was carried out using PhyML-3.1 (Guindon et al., 2010). Node support came from a majority-rule consensus tree of 1 000 bootstrap replicates. For the ML bootstraps, we considered values < 70 as low, 70‒94 as moderate and ≥95 as high following Hillis and Bull (1993). Bayesian inference (BI) analysis was carried out using MrBayes v3.2.3 (Ronquist and Huelsenbeck, 2003) on CIPRES Science Gateway. Posterior probability was estimated using four chains running 10 000 000 generations sampling every 100 generations. The first 25% of sampled trees were considered burn-in trees according to Herrera et al. (2012), and Tracer v1.6 was used to confirm that the remaining trees yielded an effective sample size (ESS) of >200 for all parameters. All remaining trees were used to calculate posterior probabilities (PPs) using a majority rule consensus. For the Bayesian posterior probabilities, we considered values < 94% as low and ≥95% as high following Alfaro et al. (2003). The accession numbers of the mtMutS sequences were listed next to the species names in the phylogenetic tree (Fig. 3a).

Concatenated alignment sequences of mitochondrial 16S, nad2, cox1, mtMutS and nad6-int-nd3 regions were used for the molecular clock analysis (Table A2). The HKY+G evolutionary model was selected as the best-fitted model for the mitochondrial concatenated sequences. A BayesianMCMC joint estimation of divergence times was performed in BEAST 1.8.4 (Drummond et al., 2012) for the mitochondrial concatenated sequences, following the parameter sets of Herrera et al. (2012): 1) employing an uncorrelated relaxed lognormal molecular clock model and the Yule model of constant speciation rate (Yule, 1925; Gernhard, 2008); and 2) using the close family Coralliidae as the calibration point with an initial value of 85.3 million years ago (Mya) and a standard deviation of 0.7. Posterior probability was estimated using four Markov chains running 10 000 000 generations sampling every 100 generations. The first 25% of sampled trees were considered burn-in trees according to Herrera et al. (2012), and Tracer v1.6 was used to confirm that the remaining trees yielded an effective sample size for all parameters. FigTree v1.3.1 (Rambaut, 2006) was used for visualization.

2.4 Secondary structures

Consensus structure of Paragorgia ITS2 regions was predicted using the LocARNA Sever (available from http://rna.informatik.uni-freiburg.de/LocARNA/ Input.jsp, Will et al., 2012), which predicts structures from an alignment of related RNA sequences. With the guidance of the consensus structure, the secondary structures of ITS2 sequences were predicted using MFOLD (Zuker, 2003) with default parameters. From the output files, the skeleton that had the highest negative free energy value and presented essentially the same arrangement as the consensus structure was selected.

3 RESULT 3.1 Taxonomy

Class Anthozoa Ehrenberg, 1831

Subclass Octocorallia Haeckel, 1866

Order Alcyonacea Lamouroux, 1812

Family Paragorgiidae Kükenthal, 1916

Genus Paragorgia Milne-Edwards & Haime, 1857

Paragorgia rubra sp. nov. (Figs. 1, 2)

Figure 1 External morphology (a–d) and internal anatomy (e, f) of Paragorgia rubra sp. nov. in situ (c, d) and after ethanol preservation (a, b, e, f) a, c, d: side of the holotype with autozooids. The white animal inhabiting the gorgonian is an echinoderm species of the ophiuroid order Euryalae; b: side of the holotype without autozooids; e: transversal cross-section of a terminal branch, showing a polyp, outer cortex (OC), inner cortex (IC), medulla (M), medullar canals (MC), and boundary canals (BC); f. a polyp and aperture, showing the arrangement of red polyp sclerites. Scale bars=100 mm (a, b), 1 mm (e), and 0.5 mm (f).
Figure 2 Sclerites of Paragorgia rubra sp. nov. a. sclerites from the polyps; b. radiate sclerites from the colony surface; c. sclerites from the colony medulla. Scale bars=20 μm (a, b) and 50 μm (c).

Material examined Holotype: Y30157, collected on 24 December 2014 from the station FX-Dive 21 (8°51.36′N, 137°46.76′E) of Yap-3 seamount near the Yap Trench at a water depth of 373 m, rocky bottom.

Description Growth form and size: Uniplanar colony, dichotomous, approximately 530 mm long and 440 mm wide in preservation. Holdfast nearly rectangle, side length 33–54 mm. The main stem approximately 100 mm long before the first large side branch arising. Two small side branches arising prior to the large branch, one measuring 33 mm long arising immediately above the holdfast, the other one 100 mm long (Fig. 1c, but not shown in Fig. 1a, b) about 60 mm apart from the holdfast. The main stem somewhat compressed, 42 mm wide and 37 mm deep before the first small branch arising, 33 mm wide and 38 mm deep between the two small branches. Terminal branches 5–90 mm in length, 3–5 mm in diameter exclusive autozooid calices. Medulla in the terminal branches usually perforated by 3–5 main canals (Fig. 1e). Subsurface exhibiting about 20 boundary canals around the medulla (Fig. 1e).

Polyps: Autozooids expanded and exhibited white color in live state; retracted into coenenchyme when preserved, but more or less exsert. Autozooids approximately 1.2–2.0 mm wide and 1.5 mm long in preservation, and scattered only on the side of the colony that faces towards the current to capture food particles (Fig. 1a, c, d). Autozooid polyp tentacles densely armed with ornate ovals, anthostele with ovals relatively sparsely arranged in eight lines (Fig. 1f).

Sclerites: Ovals usually with large compound tuberculate ornaments and an indistinct waist (Fig. 2a). Ovals up to 150 μm in length (85±9.1 μm, n=20), and 1.5–2.8 times (averaging 2.14) longer than width (40±6.2 μm, n=20). Coenenchyme with radiates with 6-, 7-and 8-rays, small, less than 80 μm in length (57±9.0 μm, n=20), and 1.4–1.8 times (averaging 1.53) longer than width (38±5.4 μm, n=20) (Fig. 2b). Radiates asymmetrical as enlargement or differentiation of some rays (Fig. 2b). Medulla with long and slim spindles having low conical protuberances (Fig. 2c). Medulla with spindles up to 390 μm long (283±68 μm, n=20), and 3.5–7.8 times (averaging 5.38) longer than width (53±7.5 μm, n=20).

Color: Medulla translucent to white, interspersed with red (Fig. 1e). Coenenchyme red both in vivo and in ethanol preservation (Fig. 1). Polyps white (Fig. 1cf). Ovals from polyps light red, radiate sclerites from the colony surface light red, spindles from the colony medulla colorless or light red.

Etymology The Latin adjective ruber (red) referring to the red color of the species.

Distribution and ecology Found only on a rocky bottom on the Yap-3 seamount, which is located near the Yap Trench in the tropical Western Pacific. The water depth was 373 m, water temperature was 8.96℃, and salinity was 36.8. The specimen was inhabited by an echinoderm individual of the order Euryalae.

Species comparisons Paragorgia rubra sp. nov. is most similar to P. kaupeka Sánchez, 2005, a species discovered in the Southwestern Pacific Ocean, in its red color, the autozooid clusters arranged towards one side, and the surface sclerites which are composed of asymmetrical radiates (Sánchez, 2005). These characters unite the two species in a group differing from the other species of Paragorgia. Nevertheless, P. rubra sp. nov. differs distinctly from P. kaupeka in the ornaments of polyp ovals (with large and compound protuberances vs. with small and simple conical protuberances) and medullar spindles (with simple conical protuberances vs. compound protuberances) (Sánchez, 2005). Moreover, P. rubra sp. nov. differs from P. kaupeka also by the smaller length/width ratio of surface sclerites (1.53 vs. 1.75) (Sánchez, 2005).

In comparison with the three species (P. coralloides Bayer, 1993; P. sibogae Bayer, 1993; P. splendens Thomson and Henderson, 1906) found in the Central Indo-Pacific, P. rubra sp. nov. differs from P. coralloides mainly in the surface sclerite rays (>5 lobes vs. globular) and medullar sclerite size (maximum length 0.39 mm vs. 0.15 mm); from P. sibogae in the diameter of terminal branches (3–5 mm vs. 1.3 mm) and polyp sclerite size (maximum length 150 μm vs. 85 μm); and from P. splendens mainly in the surface radiates (asymmetrical vs. symmetrical), length/width ratio of surface radiates (1.53 vs. 1.8) and medullar sclerite size (maximum length 0.39 mm vs. 0.7 mm) (Bayer, 1993; Sánchez, 2005).

3.2 Genetic distances

In octocorals, the mtMutS gene is considered as the best marker for initial identification (McFadden et al., 2010; 2011). The mtMutS gene sequences of all Paragorgia species except P. sibogae Bayer, 1993, P. splendens Thomson and Henderson, 1906 and P. tapachtli Sánchez, 2005, were available for the present analyses (Table A3). Based on the mtMutS aligned region, most intraspecific genetic distances within the Paragorgia species except P. regalis Nutting, 1912 ranged between 0–0.44%, while those within P. regalis were 0–2% (Fig. 3b and Table A3–A8). No intraspecific variability was observed for P. kaupeka populations, while the genetic distance between the new species P. rubra sp. nov. and P. kaupeka was 0.66% (Table 1, A4). Based on the present data, the interspecific genetic distances between Paragorgia species ranges between 0–5.63%.

Figure 3 Consensus tree (a) inferred from the mtMutS sequences and uncorrected genetic distance ranges (b) of mtMutS within and between Paragorgia and Sibogagorgia cauliflora Herrera et al., 2010 a. the trees inferred with maximum likelihood (ML) and Bayesian (BI) methods; b. (Ⅰ) between P. alisonae, P. aotearoa, P. jamesi, P. johnsoni, P. maunga, P. regalis (except for KX008439), P. stephencairnsi, P. wahine and P. whero; (Ⅱ) between P. rubra sp. nov. and congeners except P. kaupeka; (Ⅲ) between P. arborea and P. pacifica; (Ⅳ) between P. arborea and congeners except P. pacifica; (Ⅴ) between P. rubra sp. nov. and P. kaupeka; (Ⅵ) between Paragorgia species; and (Ⅶ) between Paragorgia and Sibogagorgia cauliflora. Bars show the range of variation.

In the ITS2 distance analyses, the intraspecific genetic distances ranged between 0–0.37% (data only calculated from P. arborea), and the interspecific ones among the eight Paragorgia species ranged between 0–5.49%. The genetic distance between P. rubra sp. nov. and P. kaupeka was 1.1% (Table A9).

3.3 Phylogenetic inferences and divergence times

The ML and BI phylogenetic trees of the mtMutS gene were identical in topology and were thus combined into a consensus tree with both support values (Fig. 3). Paragorgia rubra sp. nov. and P. kaupeka Sánchez, 2005 formed a sister clade, which branched early with full node support (ML 100%, BI 1.00). Paragorgia regalis Nutting, 1912 populations were separated into five divergent clades in the tree, among which P. regalis KX008349 clustered with P. coralloides Bayer, 1993. The other populations of P. regalis formed parallel branches with P. alisonae Sánchez, 2005, P. aotearoa Sánchez, 2005, P. jamesi Herrera and Shank, 2016, P. johnsoni Gray, 1862, P. maunga Sánchez, 2005, P. stephencairnsi Sánchez, 2005, P. wahine Sánchez, 2005 and P. whero Sánchez, 2005 with high node support (Fig. 3a).

In the mitochondrial multi loci tree, the topologies within Paragorgia were largely congruent with those in the mtMutS gene tree (Fig. 4). The estimated divergence time between Paragorgia rubra sp. nov. and P. kaupeka was approximately 15.0 Mya, similar to the divergence time between P. alisonae and P. johnsoni (14.2 Mya) and that between P. regalis and P. yutilinux (15.7 Mya).

Figure 4 The mitochondrial multi loci tree generated with the Yule-model tree prior from BEAST Ultrametric tree shows the estimated times of divergence under this model. The close family Coralliidae was used as the calibration point. Each node in the tree is labeled with its Bayesian posterior probabilities. Node bars represent the 95% highest posterior density intervals, which is only available for nodes with posterior probability >0.5. The new species is indicated in bold.
3.4 Secondary structures

The consensus structure was predicted based on the ITS2 sequences of Paragorgia. As showed in Fig. 5, the consensus structure was a closed central loop with four main helices (Ⅰ, Ⅱ, Ⅲ and Ⅳ), in consistent with previous study (Herrera et al., 2012). Compared with the helices Ⅰ, Ⅱ and Ⅳ, helix Ⅲ was more variable in structure based on the low basepairing probability (marked with yellow and green colors in Fig. 5). Among the structures in Paragorgia, both P. rubra sp. nov. and P. kaupeka had a large terminal bulge in helix Ⅳ, which was distinguished from other congeners. Paragorgia regalis had two medium bulges in the helix Ⅲ, which is unique in this genus (Fig. 5). The structures in P. arborea, P. aotearoa, P. wahine and P. yutlinux were similar, and those in P. aotearoa and P. wahine were the same and neither compensatory base changes (CBCs) nor hemi-CBCs could be observed (Fig. 5).

Figure 5 Predicted ITS2 RNA secondary structures for six Paragorgia species Ⅰ–Ⅳ mark the position of helices Ⅰ–Ⅳ. Arrows indicate the structural differences between the consensus and the congeners. Color bars (0–1) indicate basepairing probability. Newly sequenced species are in bold.
4 DISCUSSION

Paragorgiids are unusual octocorals as their axial medulla is formed by accumulation of unfused sclerites, while most branching gorgonians contain a proteinaceous or calcareous axial skeleton. To date, the family Paragorgiidae comprises the genera Paragorgia Milne-Edwards and Haime, 1857 and Sibogagorgia Stiasny, 1937. Among the 17 known species of Paragorgia, the six species P. alisonae Sánchez, 2005, P. aotearoa Sánchez, 2005, P. kaupeka Sánchez, 2005, P. maunga Sánchez, 2005, P. wahine Sánchez, 2005 and P. whero Sánchez, 2005 have been reported only in New Zealand; P. coralloides Bayer, 1993 and P. sibogae Bayer, 1993 have been found only in the tropical Western Pacific; P. splendens Thomson & Henderson, 1906 has been found in both the tropical Western Pacific and Indian Ocean; P. pacifica Verrill, 1922 has been found in the temperate Northern Pacific; P. jamesi Herrera and Shank, 2016, P. stephencairnsi Sánchez, 2005, P. tapachtli Sánchez, 2005 and P. yutlinux Sánchez, 2005 have been found in the Northeastern Pacific; P. regalis Nutting, 1912 is trans-Pacific, P. johnsoni Gray, 1862 is trans-Atlantic, and P. arborea (Linnaeus, 1758) has a bipolar distribution (Sánchez, 2005; van Ofwegen, 2015; Herrera and Shank, 2016). Paragorgia rubra sp. nov. is the fourth species found in the tropical Western Pacific in the North Hemisphere.

Consistent with morphology, molecular data indicate that P. rubra sp. nov. is closely related to P. kaupeka, forming a sister clade which branched early within Paragorgia. The phylogenetic relationships within the genus Paragorgia inferred from the mtMutS gene sequences indicate the populations of P. regalis are separated into four divergent clades, where P. regalis KX008349 showed a close relationship with P. coralloides. However, based on the broader sequence database of the present markers, the phylogenetic placements of P. johnsoni, P. maunga, P. yutilinux, P. aotearoa, P. wahine and P. regalis are still not resolved, as indicated by Herrera et al. (2012). Therefore, more molecular analyses, e.g., restriction-site associated DNA sequencing (RADseq) method, are needed to obtain further phylogenetic resolutions (Herrera and Shank, 2016).

Maximum uncorrected genetic distances of the mtMutS gene among conspecifics and minimum ones among congeners have been used to assess the intraand interspecific variation, with the distances greater than 1% confidently used to indicate cryptic species of Octocorallia (McFadden et al., 2011; Herrera et al., 2012). Based on this threshold, the genetic distances among populations of P. regalis are relatively high, in the range of 0–2.06%, indicating the presence of cryptic species (Fig. 3b and Table A3). Excluding P. regalis, the intraspecific genetic distances at mtMutS within the Paragorgia species ranged between 0–0.44%, which is congruent with the study of Herrera et al. (2012). Furthermore, the mtMutS gene within P. kaupeka was relatively conservative and no intraspecific variability among populations could be observed. Therefore, based on the intra-and interspecific genetic data from 14 of 17 Paragorgia species, the minimum distance of 0.66% between P. rubra sp. nov. and the most similar species P. kaupeka could be confidently regarded at interspecific level.

A similar pattern was observed for the ITS2 distances calculated from eight Paragorgia species (Table A4). The genetic distance between the two similar species P. rubra sp. nov. and P. kaupeka is much greater than the maximum intraspecific one within the Paragorgia species (1.1% vs. 0.37%). The predicted ITS2 secondary structures are also consistent with the variation pattern of genetic distances. Most species showed bulge differences in helix Ⅲ and/or helix Ⅳ structures, except that the structures in P. aotearoa and P. wahine were the same and no CBCs or hemi-CBCs were observed (Fig. 5). However, similar to the mitochondrial barcoding threshold, high similarities in ITS2 primary and secondary structures do not imply the absence of species boundaries, the criterion of which is also unidirectional (Coleman, 2009; Herrera et al., 2012).

5 CONCLUSION

A bubblegum coral specimen collected from a seamount near the Yap Trench in the tropical Western Pacific was identified and described as a new species, Paragorgia rubra sp. nov. The new species differs from its congeners by the combination of terminal branch size, coenenchyme color, autozooid tentacle sclerites, surface sclerites and medulla sclerites. The genetic distance of the mtMutS gene between P. rubra sp. nov. and the closely related species P. kaupeka is higher than the intraspecific genetic distances between most congeners except the P. regalis species complex (0.66% vs. < 0.50%), supporting the establishment of the new species. Furthermore, the ITS2 secondary structure of P. rubra sp. nov. is different from those of other congeners. Paragorgia rubra sp. nov. and P. kaupeka branch early within Paragorgia, with a recent diversification time of approximately 15.0 Mya.

6 ACKNOWLEDGEMENT

We thank the assistance of the crew of R/V KeXue and ROV FaXian for sample collection. We thank our colleague Dr. LI Yuhang for assistance of SEM preparation. We also appreciate the anonymous reviewers for providing constructive comments and criticisms on earlier versions of the manuscript.

References
Alfaro M E, Zoller S, Lutzoni F, 2003. Bayes or Bootstrap? A simulation study comparing the performance of Bayesian Markov chain Monte Carlo sampling and bootstrapping in assessing phylogenetic confidence. Molecular Biology and Evolution, 20(2): 255–266. Doi: 10.1093/molbev/msg028
Bayer F M, Grasshoff M, Verseveldt J. 1983. Illustrated Trilingual Glossary of Morphological and Anatomical Terms Applied to Octocorallia. E. J. Brill/Dr. W. Backhuys, Leiden. 75p.
Bayer F M, 1993. Two new species of the gorgonacean genus Paragorgia (Coelenterata:Octocorallia). Precious Corals & Octocoral Research, 2: 1–10.
Buhl-Mortensen L, Mortensen P B. 2005. Distribution and diversity of species associated with deep-sea gorgonian corals off Atlantic Canada. In: Freiwald A, Roberts J M eds. Cold-Water Corals and Ecosystems. Springer-Verlag, Berlin Heidelberg. p. 849-879.
Coleman A W, 2009. Is there a molecular key to the level of "biological species" in eukaryotes? A DNA guide. Molecular Phylogenetics and Evolution, 50(1): 197–203. Doi: 10.1016/j.ympev.2008.10.008
Darriba D, Taboada G L, Doallo R, Posada D, 2012. jModelTest 2:more models, new heuristics and parallel computing. Nature Methods, 9: 772.
Drummond A J, Suchard M A, Xie D, Rambaut A, 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29(8): 1969–1973. Doi: 10.1093/molbev/mss075
Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R, 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit Ⅰ from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3(5): 294–299.
Gernhard T, 2008. The conditioned reconstructed process. Journal of Theoretical Biology, 253(4): 769–778. Doi: 10.1016/j.jtbi.2008.04.005
Giribet G, Wheeler W C, 1999. On gaps. Molecular Phylogenetics and Evolution, 13(1): 132–143. Doi: 10.1006/mpev.1999.0643
Gray J E, 1862. Notice of a second species of Paragorgia discovered in Madeira by Mr. James Yate Johnson. Annals and Magazine of Natural History, 10: 125–126.
Guindon S, Dufayard J F, Lefort V, Anisimova M, Hordijk W, Gascuel O, 2010. New algorithms and methods to estimate maximum-likelihood phylogenies:assessing the performance of PhyML 3.0. Systematic Biology, 59(3): 307–321. Doi: 10.1093/sysbio/syq010
Heifetz J, 2002. Coral in Alaska:distribution, abundance, and species associations. Hydrobiologia, 471(1-3): 19–28.
Herrera S, Baco A, Sánchez J A, 2010. Molecular systematics of the bubblegum coral genera (Paragorgiidae, Octocorallia) and description of a new deep-sea species. Molecular Phylogenetics and Evolution, 55(1): 123–135. Doi: 10.1016/j.ympev.2009.12.007
Herrera S, Shank T M, Sánchez J A, 2012. Spatial and temporal patterns of genetic variation in the widespread antitropical deep-sea coral Paragorgia arborea. Molecular Ecology, 21(24): 6053–6067. Doi: 10.1111/mec.12074
Herrera S, Shank T M, 2016. RAD sequencing enables unprecedented phylogenetic resolution and objective species delimitation in recalcitrant divergent taxa. Molecular Phylogenetics and Evolution, 100: 70–79. Doi: 10.1016/j.ympev.2016.03.010
Hillis D M, Bull J J, 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Systematic Biology, 42(2): 182–192. Doi: 10.1093/sysbio/42.2.182
Katoh K, Standley D M, 2013. MAFFT multiple sequence alignment software version 7:improvements in performance and usability. Molecular Biology and Evolution, 30(4): 772–780. Doi: 10.1093/molbev/mst010
Koslow J A, Gowlett-Holmes K, Lowry J K, O'Hara T, Poore G C B, Williams A, 2001. Seamount benthic macrofauna off southern Tasmania:community structure and impacts of trawling. Marine Ecology Progress Series, 213: 111–125. Doi: 10.3354/meps213111
Librado P, Rozas J, 2009. DnaSP v5:a software for comprehensive analysis of DNA polymorphism data. Bioinformatics, 25(11): 1451–1452. Doi: 10.1093/bioinformatics/btp187
Linnaeus C. 1758. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tomus Ⅰ. Editio decima, reformata, Holmiae, Laurentii Salvii. 824p.
McFadden C S, Benayahu Y, Pante E, Thoma J N, Nevarez P A, France S C, 2011. Limitations of mitochondrial gene barcoding in Octocorallia. Molecular Ecology Resources, 11(1): 19–31. Doi: 10.1111/men.2010.11.issue-1
McFadden C S, Sánchez J A, France S C, 2010. Molecular phylogenetic insights into the evolution of Octocorallia:a review. Integrative and Comparative Biology, 50(3): 389–410. Doi: 10.1093/icb/icq056
Milne-Edwards H, Haime J. 1857. Histoire Naturelle des Coralliaires ou Polypes Proprement Dits, volumen 1. Librairie Encyclopédique de Roret, Paris. 326p.
Nutting C C, 1912. Descriptions of the Alcyonaria collected by the U. S. Fisheries steamer "Albatross, " mainly in Japanese waters, during 1906 Proceedings of the United States National Museum, 43: 1–104.
Rambaut A. 2006. Figtree. Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, http://tree.bio.ed.ac.uk/software/figtree.
Ronquist F R, Huelsenbeck J P, 2003. Mrbayes 3:Bayesian phylogenetic inference under mixed models. Bioinformatics, 19(12): 1572–1574. Doi: 10.1093/bioinformatics/btg180
Sánchez J A, 2005. Systematics of the bubblegum corals (Cnidaria:Octocorallia:Paragorgiidae) with description of new species from New Zealand and the Eastern Pacific. Zootaxa, 1014(1): 1–72. Doi: 10.11646/zootaxa.1014.1
Stiasny G. 1937. Die Gorgonacea der Siboga-Expedition. Supplement Ⅱ, Revision der Scleraxonia mit ausschluss der Melitodidae und Coralliidae. Siboga-Expedition Monographs, 13b: 1-138.
Tamura K, Stecher G, Peterson D, Filipski A, Kumar S, 2013. MEGA6:molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution, 30(12): 2725–2729. Doi: 10.1093/molbev/mst197
Thomson J A, Henderson W D. 1906. An account of the alcyonarians collected by the Royal Indian Marine Survey Ship Investigator in the Indian Ocean. Ⅰ. The Alcyonarians of the deep sea. The Trustees of the Indian Ocean Museum, Calcutta. 132p.
van Ofwegen L. 2015. Paragorgiidae Kükenthal, 1916. Accessed through: World Register of Marine Species, http://www.marinespecies.org/aphia.php?p=taxdetails&id=125281. Accessed on 2015-07-22.
Verrill A E, 1922. Alcyonaria and Actiniaria. Report of the Canadian Arctic Expedition 1913-18, 8(G): 1–164.
Will S, Joshi T, Hofacker I L, Stadler P F, Backofen R, 2012. LocARNA-P:accurate boundary prediction and improved detection of structural RNAs. RNA, 18(5): 900–914. Doi: 10.1261/rna.029041.111
Yule G U, 1925. A mathematical theory of evolution, based on the conclusions of Dr. J. C. Willis, F.R.S. Philosophical Transactions of the Royal Society B:Biological Sciences, 213(402-410): 21–87. Doi: 10.1098/rstb.1925.0002
Zuker M, 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Research, 31(13): 3406–3415. Doi: 10.1093/nar/gkg595