Mitochondrial DNA Sequence Evidence Supporting the Recognition of a New North Atlantic Pseudostichopus Species (Echinodermata: Holothuroidea)

J. Mar. Biol. Ass. U.K. (2004), 84,1077^1084 Printed in the United Kingdom

Mitochondrial DNA sequence evidence supporting the recognition of a new North Atlantic Pseudostichopus species (Echinodermata: Holothuroidea)

O O P Francisco A. Sol|¤s-Mar|¤n*$, David S.M. Billett , Joanne Preston and Alex D. Rogers *Coleccion Nacional de Equinodermos, Laboratorio de Sistematica y Ecologia de Equinodermos, Instituto Ciencias del Mar y O Limnologia, Universidad Nacional Autonoma de Mexico, Apdo. Post. 70-305, CP. 04510, Mexico City. Southampton Oceanography P Centre (SOC), School of Ocean and Earth Sciences (SOES), Empress Dock, Southampton, SO14 3ZH, UK. British Antarctic Survey, High Cross, Madingley Road, Cambridge, CB3 0ET, UK. $Corresponding author, e-mail: [email protected]

A new species of the synallactid sea cucumber genus Pseudostichopus is described, P. aemulatus sp. nov., based on genetic (DNA sequences of the mitochondrial gene Cytochrome Oxidase I [COI] gene) and morphological characters. A comparative molecular study with two other species of the same genus (P. villosus and P. mollis) and from a di¡erent family (Isostichopus fuscus) was carried out in order to clarify its taxonomic identity. The nucleotide distance between P. aemulatus sp.nov.andP. villosus and P. mollis is su⁄cient to support distinct species status. The estimated di¡erence in the number of amino acids, coded for by a partially sequenced COI gene, within the species of the family Synallactidae ranged from 4 to 18. The phylogenetic analysis clearly supports separate species status of these sympatric morphotypes, as indicated by the morphological analysis.

INTRODUCTION ossicle morphology of Pseudostichopus species are ambig- Holothurians are amongst the most conspicuous uous and frequently contradictory. Di¡erences in general elements of the deep-sea megafauna. Many of their body shape and internal anatomy between Atlantic Pseudostichopus specimens are often small and again may morphological characters show a high degree of variation simply represent intraspeci¢c variation. some of which may arise as a result of artefacts of methods In this study we aimed to clarify the morphological used for collection and preservation of specimens. The identity of some of the north-east Atlantic Pseudostichopus deep-sea species of the family Synallactidae are typical of species, and to augment the morphological data with this situation. This family contains 19 controversial DNA sequences of the mitochondrial gene Cytochrome described genera including: Amphigymnas, Bathyplotes, Oxidase I (COI) gene. The data indicate that the Benthothuria, Hansenothuria, Mesothuria, Paroriza, Pelopatides, smaller, common morphotype of Pseudostichopus from the Pseudostichopus, Synallactes, Zygothuria, Allopatides, Bathyzona, Porcupine Abyssal Plain is a distinct species and that in Capheira, Dendrothuria, Filithuria, Galatheathuria, Perizona, Pseudothuria, Scotothuria. The genus Pseudostichopus The¤el, this case what has been interpreted as intraspeci¢c varia- 1886 is one of the oldest established synallactid taxa and tion within species is interspeci¢c variation between known and cryptic species of holothurians. The systematic its species also demonstrate a high level of putative intra- implications of this for deep-sea and shallow-water speci¢c variation. holothurians are discussed. Many synallactid holothurians were collected by the RRS ‘Discovery’ and RRS ‘Challenger’ from the deep-sea £oor of the Porcupine Abyssal Plain, north-east Atlantic MATERIALS AND METHODS during a number of European Union co-sponsored research programmes between 1991 and 1999, including Sampling the BENGAL project (High-resolution temporal and Pseudostichopus specimens were examined from 14 spatial study of the BENthic biology and Geochemistry of samples taken on the Porcupine Abyssal Plain to the south- a north-eastern Atlantic abyssal Locality) (Billett & Rice, west of Ireland, north-east Atlantic, ranging from 4764 to 2001). The samples included numerous specimens of 4849 m in depth (Table 1). Rice (1992, 1996, 1997), Billett P. v i l l o s u s The¤el, 1886, and many individuals of a smaller (2000) and Billett & Rice (2001) provide descriptions of Pseudostichopus morphotype (Billett et al., 2001). the study area, and the sampling strategy and gear used. Whether this smaller morphotype represented onto- In addition to these abyssal north-east Atlantic speci- genetic variation amongst one of the other Atlantic mens, Isostichopus fuscus was collected using SCUBA diving Pseudostichopus (P. depressus He¤rouard, 1902, P. lapidus from a shallow-water site (15m depth) in the East Paci¢c, He¤rouard, 1923, P. marenzelleri He¤rouard, 1923, P. occultatus o¡ the coast of Jalisco, Mexico (218N) in the year 2000. Marenzeller, 1893 and P. v i l l o s u s The¤el) or whether it Specimens used in the morphological analysis were represented a previously unrecognized taxon was uncer- ¢xed in 4^8% bu¡ered formalin for at least 24 hours and tain. Previous descriptions of the external features and then transferred to 70% ethanol. Specimens collected for

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh 1078 F.A. Sol|¤s-Mar|¤n et al. DNA sequence of a new Pseudostichopus species

Table 1. List of stations used in the study. RRS ‘Discovery’ cruises 222(2), 226, 237 and RRS ‘Challenger’ cruises 79, 142 (Rice, 1992, 1996, 1997; Billett, 2000).

Position start Position end No. of Station number Date Latitude (N) Longitude (W) Latitude (N) Longitude (W) Depth (m) specimens

52701#42 24-05-91 48852.700 16838.500 48851.200 16828.500 4849^4843 85 12930#46 09-09-96 48847.210 16843.310 48849.490 16832.620 4837^4841 53 12930#78 16-09-96 48853.040 16830.490 48850.030 16841.920 4836^4840 82 13078#29 04-04-97 48856.200 16822.770 48847.350 16833.230 4844^4847 183 13627#10 30-09-98 48853.060 16842.060 49802.000 16853.030 4835^4837 68 54901#5 28-04-99 48844.870 16840.530 48848.160 16836.240 4835^4838 97 54901#7 29-04-99 48847.450 16848.880 48850.820 16846.040 4836^4838 2 54901#9 30-04-99 48846.890 16841.590 48850.580 16836.360 4837^4841 111 54903#1 03-05-99 49832.090 15856.020 49828.110 15856.520 4810^4817 545 54905#1 04-05-99 50832.650 16857.770 50828.660 16859.430 4764^4786 72 Total: 1298

molecular analyses were placed immediately in chilled (approximately 0.012% units/ml of Boeghringer water on-board ship and transferred to a temperature Mannheim, Cat. no. 1373-196). These preparations were controlled room (48C). Individuals were dissected to incubated for 2 h at 558C in an incubator and then a obtain a sample of longitudinal muscle from the body standard phenol/chloroform-isoamyl alcohol extraction wall, which was immediately placed in 99% ethanol. was carried out with precipitation of DNA by 2:1 ice-cold 100% ethanol plus 1:10 3M sodium acetate. Polymerase chain reactions (PCRs) were carried out in 20 ml total Identi¢cation volume using sterile water, and contained 160 mM each Taxonomic identi¢cation using external and internal dNTP, 10 mM Tris^HCL, pH 8.3, 40 mM KCL, 2 mM anatomy was carried out prior to DNA analysis, based on MgCl2,1mM each primer, 1 unit Taq-polymerase, and 10 original descriptions and keys. Total length (TL), as to 30 ng template DNA. The primers used to amplify the indicated in the results, was measured from the tip of the 30 end of the COI gene and their position in the mitochon- anterior part of the body to the posterior end. Width (W) drial map of the sea urchin Strongylocentrotus purpuratus was measured from the widest part of the body at the mid- (Stimpson, 1857) are shown in Table 2 (Jacobs et al., ventral region. For each set of measurements, the 1988; Arndt et al., 1996). maximum, minimum and median values were recorded. Polymerase chain reaction was performed using either a All measurement values are in millimetres. The specimens Perkin-Elmer 480 or a Hybaid PCR-Express thermocycler were deposited in the collections of the Natural History and comprised a 948C/4 min initial denaturing step Museum, London (NHM); the National Museum of followed by 30 cycles of 948C/1min, 508C/1min, and Natural History, Smithsonian Institution,Washington, DC, 728C/1min. A ¢nal elongation step of 728C/10min was USA (USNM); the Zoological Museum, Copenhagen, used. The ‘oil overlay’ and ‘hot lid’ methods worked Denmark (ZMC); Echinoderm National Collection, equally well. Products were visualized on a 1.5% agarose Universidad Nacional Autonoma de Mexico, Mexico City, gel stained with ethidium bromide. Polymerase chain (ICML-UNAM) and the Discovery Collections, South- reaction products were puri¢ed with Qiagen Qiaquick ampton Oceanography Centre, UK (SOC-DC). PCR puri¢cation columns (Cat. no. 28106), according to manufacturer’s guidelines. Cycle-sequencing reactions of 10 ml were prepared using Perkin-Elmer BigDye Termi- Polymerase chain reaction and sequencing nator Ready Reaction mixes (Cat. no. 4303152), using the Three species were screened in the present study, two manufacturer’s guidelines. The products of the cycle- Pseudostichopus species and Isostichopus fuscus (Ludwig, sequencing reactions were puri¢ed using Qiagen DyeEx 1875).The latter was used as outgroup for the phylogenetic Spin kits (Cat. no. 63104) and sequences visualized using analysis. DNA was extracted through digestion of a small a Perkin-Elmer ABI 377 automated DNA sequencing piece of muscle (approximately 200 mg) in 100 mM Tris^ machine. All samples were double-checked by reverse HCL, pH 8.0, 1.25% SDS, and 390 ng/ml proteinase K sequencing.

Table 2. Oligonucleotid primers sequences used in the present study.

Mitochondrial map position Primer Sequence 50 30 (Strongylocentrotus purpuratus) Direction

COIef ATAATGATAGGAGG[A/G]TTTGG 6000^6019 Forward COIer GCTCGTGT[A/G]TCTAC[A/G]TCCAT 6692^6673 Reverse

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh DNA sequence of a new Pseudostichopus species F.A. Sol|¤s-Mar|¤n et al. 1079

Table 3. Sequences used in molecular analysis.

Species No. sequences GenBank accession no. Reference

Pseudostichopus mollis 1 U32221 Arndt et al. (1996) Pseudostichopus villosus 3 AF486434, AF486435, This paper AF486436 Pseudostichopus aemulatus sp. nov. 4 AF486430, AF486431, This paper AF486432, AF486433 Isostichopus fuscus 6 AF486424, AF486425, This paper AF486426, AF486427, AF486428, AF486429

Table 4. Nucleotide substitution rate matrix. cies were set to A¼0.2804, C¼0.2079, G¼0.1735 and, T¼0.3382. ACGT For all searches, starting branch lengths were obtained using the Rogers^Swo¡ord approximation method and a A 7.0575 19.5206 47.9867 molecular clock was not enforced. 3.9184 91.2120 1.0 RESULTS

SYSTEMATICS Molecular data analysis Order ASPIDOCHIROTA Grube, 1840 Fa m i l y SYNALLACTIDAE Ludwig, 1894 The potential interspeci¢c variation of the putative Genus Pseudostichopus The¤el, 1886 species of Pseudostichopus was analysed using a 597 base Pseudostichopus aemulatus sp. nov. Sol|¤s-Mar|¤n & Billett pair partial COI sequence with all bases included (bp 1^ (Figures 1A^I) 597, GenBank accession nos. AF486424^AF486436). The analysis sample set comprised COI sequences for P. mollis, Pseudostichopus sp. Billett, 1988: 196^197; Billett et al., 2001: P.villosus,thePseudostichopus morphotype and the outgroup 325^348. Isostichopus fuscus (Table 3). These represent consensus sequences determined from the original sequence data set Typematerial of 14 individuals in which all identical sequences were Holotype: adult male, (TL¼84 mm, W¼25 mm), RRS collapsed. The COI amino acid sequence was used for the ‘Discovery’ Station 54901#5, 28 April 1999, semi-balloon best possible alignment. All sequences were aligned using otter trawl 14, from 48844.870N16840.530Wto48848.160N Clustal X (Thompson et al., 1997) and edited by eye using 16836.240W, from 4835 to 4838 m depth [NHM MacClade (Maddison & Maddison, 1989, 1992). 2001.7066]. Phylogenetic analysis was performed on the nucleotide Paratypes: NHM 2001.7067^7069, all measurements sequence data. Non-parametric bootstrap analysis was are in mm (3 specimens, TLmax¼103, TLmin¼60, performed under minimum evolution (ME), maximum median¼78, Wmax¼21, W min¼14, W median¼18); USNM parsimony (MP) and maximum likelihood (ML) optim- 1025525(3specimens,TLmax¼92, TLmin¼67, ality criteria using PAUP 4.0b8a (PPC) (Swo¡ord, 1998). median¼76, Wmax¼30, Wmin¼16, W median¼22); ZMC, Starting trees were obtained by neighbour joining for ME HOL 00156 (3 specimens, TLmax¼80, TLmin¼66, and ML and via stepwise addition with random sequence median¼74, Wmax¼24, Wmin¼22, Wmedian¼23); ICML- addition for the MP analysis. In the MP analysis, gaps UNAM 5.125.0 (3 specimens, TLmax¼97, TLmin¼67, were treated as missing and of the 597 characters 144 median¼80, Wmax¼24, Wmin¼17, W median¼20); SOC-DC were parsimony-informative. For all optimality criteria (3 specimens, TLmax¼81, TLmin¼61, median¼73, a heuristic search was employed using the tree-bisection- Wmax¼24, Wmin¼17, W median¼20). reconnection (TBR) branch-swapping algorithm, in addition to an exhaustive ML search. Diagnosis The best model of evolution for ME and ML analysis A medium-sized species up to 146 mm long. Body was determined by hierarchical likelihood ratio tests cylindrical, slightly £attened ventrally. Mouth subterm- (hLRTs) using Modeltest v3.0 (Posada & Crandall, 1998). inal, anal opening in a vertical furrow; dorsal papillae The general time reversible model with I (GTR+I) was conspicuous, arranged in six parallel series. Papillae chosen above others as the most suitable model of evolu- slightly longer and more abundant in the posterior part of tion for the data set. The estimations from the hLRTs the body.Twenty tentacles. Calcareous ring stout, and well were used to re¢ne the optimality criteria settings as developed. Radial pieces of di¡erent robustness and shape, follows; the proportion of invariable sites (I) was set to depending on the position in the calcareous ring. Inter- 0.5618, gamma distribution (the distribution of rates at radial pieces of similar aspect and size. Polian vesicle variable sites) was assumed equal, the substitution model single, ventral. Gonad unbranched, with two tufts at both was based on the rate matrix (Table 4) and, base frequen- sides of the dorsal mesentery. Ossicles of the tentacles:

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh 1080 F.A. Sol|¤s-Mar|¤n et al. DNA sequence of a new Pseudostichopus species

Figure 1. Pseudostichopus aemulatus Sol|¤s-Mar|¤n and Billett. (ICML-UNAM 5.125.0). (A) Entire animal, dorsal side; (B) entire animal, ventral side; (C) tentacular ossicles; (D) gonad ossicles; (E) irregular calcareous bodies from tentacles; (F) Polian vesicle; (G) one half of left genital gland; h1^3. Calcareous ring, Figure h1. Ventral pieces, Figure h2. Median pieces, Figure h3. Dorsal pieces; (I) di¡erent morphologies of the posterior anal lobules. Scale bars: A, B, 20 mm; C, 5 m; D, 7 m; E, 15 m; F, G, 5 mm; H, 10 mm.

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh DNA sequence of a new Pseudostichopus species F.A. Sol|¤s-Mar|¤n et al. 1081

Table 5. List of material examined.

Total length (mm) Width (mm) Station No. of number specimens maximum minimum median maximum minimum median

52701#42 85 115 30 72 19 9 18 12930#46 53 114 30 75 33 8 25 12930#78 82 116 34 74 20 11 20 13078#29 183 117 29 71 27 10 20 13627#10 68 109 40 72 27 10 17 54901#5 97 123 39 75 35 11 21 54901#7 2 95 89 23 20 54901#9 111 107 38 69 32 11 20 54903#1 545 100 20 58 43 5 17 54905#172957071271618

sacciform rods with one or two knobs and £at or knobbed mid-ventral radial piece was slightly narrower on its calcareous bodies of di¡erent complexity. Ossicles in the posterior margin than on the anterior margin; 6.5 mm gonad: branched and unbranched arched rods with and 5 mm respectively. The posterior margin was gently pointed ends, with one, two or three spiny tips. Some curved (Figure 1H1). The lateral radial pieces were rods with a single knobbed centre. Body wall, papillae narrower on their anterior margin than on the posterior and respiratory trees devoid of any ossicles. margin; 15mm as opposed to 10 mm respectively and had a strong posterior indentation (Figure 1H2). The dorsal Etymology radial pieces were narrower on the anterior side than on The speci¢c epithet aemulatus in Latin means to emulate the posterior one; 5 mm as opposed to 7.5 mm respectively or imitate. It is here used as a noun in apposition and (Figure 1H3). The interradial pieces had a wider anterior refers to the fact that externally, the adults of the new margin than posterior margin; 5 mm vs 3 mm respectively. species are very similar to those of the congeneric species. The interradial pieces were in direct contact with the radial pieces over the entire length. The posterior edges of Description the interradial pieces were gently concave in all cases Of the 1298 specimens examined (Table 5), the largest (Figure 1H1^3). measured 146 mm in total length (TL) and 22 mm wide The stone canal was attached to the body wall and was (W). Generally the body was cylindrical in shape (Figure di⁄cult to detect. Two well-developed respiratory trees 1A,B). The body surface was sometimes encrusted with were present arising from a common base at the cloaca. foraminiferan shells and sand. Some specimens (i.e. from The respiratory trees reached forward about two-thirds Station 13078#29), small (520 mm) or large (4100 mm) of the way up the body. The gonad was split into two were completely covered by foraminiferans. The tegument parts one each side of the dorsal mesentery.The two parts was faint, pale, yellowish white and opaque. It was about are slender and long, approximately 12 mm in length 2 mm thick and of moderate hardness. (Figure 1G). The Polian vesicle was single. It was long The tentacles, 20 in number, were small, shield-shaped and slender, about 2 to 3 mm in diameter and 13 to and coloured faint, pale yellowish. The mouth was situated 15mm long (Figure 1F). ventrally on the anterior part of the body. The anus was Ossicles: ossicles were exclusively rods in the tentacles situated in the shallow vertical furrow on the posterior and the gonad. No ossicles were present in the body wall part of the body. Small papillae were arranged in two or papillae. The tentacular rods (Figure 1C) were very double rows and two single rows along the dorsal and numerous and were of two types: slender, spiny-bifurcated lateral ambulacra, respectively (Figure 1A). They varied (from 3 to 11mm) and smooth^slender with a central knob in number between 10 and 23 along each row of the (from 3.1 to 17 mm). Some rods were slightly bent. It was dorsal radii and between 6 and 13 in the lateral radii, very common to ¢nd irregular £at or knobbed bodies, up depending on specimen size. The ventral surface was to 8 mm, of di¡erent complexity in the tentacles (Figure completely naked (Figure 1B). In some individuals greater 1E). The rods in the gonad were smaller (from 6 to than 100 mm long, the specimens appeared to lack any 15 mm) and thinner, with short processes near their extre- dorsal papillae. The posterior anal lobules exhibited mities (Figure 1D). A central knob was present in some of di¡erent morphology depending on the size of the the smaller ones (10 mm). posterior dorsal papillae that converged in this area of the body (Figure 1I1^3). Distribution The calcareous ring consisted of ¢ve radial pieces and North-east Atlantic, Porcupine Abyssal Plain (Billett et ¢ve interradials. The radial pieces were wide and stout, al., 2001). bearing di¡erent shapes depending on their position in the calcareous ring (Figure 1H1^3). They had two central Bathymetric distribution anterior processes and two antero-lateral ones. The From 4350 to 4850 m (Billett et al., 2001).

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh 1082 F.A. Sol|¤s-Mar|¤n et al. DNA sequence of a new Pseudostichopus species

Table 6. Holothurian pairwise distances matrix1.

If2 If10 If15 If1 If4 If7 Pv6 Pv7 Pv1 Pa5 Pa8 Pa2 Pa1 Pm

If2 If10 0.20 If15 0.12 0.16 If1 0.16 0.12 0.12 If4 0.08 0.20 0.12 0.16 If7 0.23 0.12 0.20 0.16 0.23 Pv6 0.36 0.23 0.31 0.28 0.36 0.20 Pv7 0.36 0.23 0.31 0.28 0.36 0.20 0.08 Pv1 0.31 0.20 0.28 0.23 0.31 0.16 0.12 0.12 Pa5 0.40 0.28 0.36 0.31 0.40 0.23 0.28 0.28 0.23 Pa8 0.44 0.31 0.40 0.36 0.44 0.28 0.31 0.31 0.28 0.12 Pa2 0.44 0.31 0.40 0.36 0.44 0.28 0.31 0.31 0.28 0.12 0.08 Pa1 0.36 0.23 0.31 0.28 0.36 0.20 0.23 0.23 0.20 0.12 0.16 0.16 Pm* 0.31 0.20 0.28 0.23 0.31 0.16 0.20 0.20 0.16 0.16 0.20 0.20 0.12

1, Average pairwise general time reversible+site-speci¢c rates distances. *, Distances calculated between consensus sequences and COI GenBank accession nos. U31901, U32210^U32221, U32198 and U32199. If, Isostichopus fuscus;Pv,Pseudostichopus villosus;Pa,P.aemulatus; Pm, P. mollis.

Figure 2. (A) Minimum evolution (ME) and maximum parsimony (MP) 50% majority-rule bootstrap consensus tree topologies. Bootstrap values are given for ME and MP respectively; (B) maximum likelihood exhaustive search tree. If, Isostichopus fuscus; Pv, Pseudostichopus villosus;Pa,P. aemulatus;Pm,P. mollis.

Molecular analysis published by Arndt et al. (1996) (GenBank accession nos. The overall length of the ampli¢ed mitochondrial DNA U31901, U32210^U32221, U32198 and U32199). The products varied between 610 base pairs (Pseudostichopus sequence containing the putative COI gene fragment was villosus) and 690 base pairs (P. aemulatus)atthe30 end of translated using the echinoderm codon table (Himeno et the gene. The consensus DNA sequences of the ampli¢ed al., 1987). The COI gene is assigned ATG as the initiation fragments had 597 base pairs. The sequences were codon. All specimens of the same species had identical submitted to the GenBank data base (accession nos. amino acid sequences. The number of amino acid di¡er- AF486430^AF486433). Full molecular analysis may be ences between the Pseudostichopus species ranged from 4 to found in Sol|¤s-Mar|¤n(2003). 18. The maximum di¡erence (18) occurred between The identity of the partial sequence of the COI gene was P. villosus and P. mollis. There were only 7 amino acid con¢rmed by similarity of the peptide sequence to the di¡erences between Pseudostichopus and Isostichopus. COI gene of the sea urchin Strongylocentrotus purpuratus The average contributions of bases across the sequences (Jacobs et al., 1988) and the COI data from holothurians studied were A¼27%, C¼20%, G¼17%, and T¼34%.

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh DNA sequence of a new Pseudostichopus species F.A. Sol|¤s-Mar|¤n et al. 1083

No signi¢cant di¡erences in base composition were tentacle ossicles in the two species are di¡erent. Moreover, detected across the four taxa (w2 ¼10.16; df ¼39; P40.999). the calcareous ring is short and stout in P. aemulatus and In the synallactid group, the GC content of the sea tall and thin in P. occultatus. cucumber COI gene fragments ranged from a low of 30% Pseudostichopus depressus and P. lapidus lack ossicles in any in P. mollis to a high of 37% in P. aemulatus and P. villosus, internal organ, unlike P. aemulatus. Moreover, the general with an average of 35%. As reported for numerous organ- body shape of P. depressus is completely di¡erent to isms, including a number of other echinoderms (Jacobs et P. aemulatus.InP. depressus the body is £at with a marginal al., 1988; Cantatore et al., 1989; Asakawa et al., 1991; rim while in P. aemulatus it is more cylindrical, although it Arndt et al., 1996), the proportion of guanine (G) in the does have a £at ventral surface. third position was low (6^11%) with a corresponding Pseudostichopus lapidus di¡ers from P. aemulatus in having increase in adenine (A) (average of 45%). a row of 25 ventral tubefeet along the edge of the body. There was no signi¢cant intraspeci¢c divergence These are restricted to the middle part of the body. More- between the regions of the COI gene (0.06%) For the over, P. lapidus possesses one row of papillae along each COI gene, the observed directional mutation pressure dorsal radius, with four papillae in each row. results in a restricted substitution pattern where C?T tran- Pseudostichopus aemulatus di¡ers from P. v i l l o s u s in the sitions and transversions involving A predominate. Table 6 general external body shape appearance. Pseudostichopus presents the matrix of the average pairwise GTR+site- aemulatus possesses a well de¢ned £at ventral region, speci¢c rates distances for the COI sequences. while P. villosus is much more rounded, and does not have Pseudostichopus aemulatus shows a degree of genetic diver- a well-de¢ned ventral region. Pseudostichopus aemulatus has gence from P. v i l l o s u s , comparable with the pairwise well-developed dorsal papillae arranged in double rows, distances between P. mollis and P. v i l l o s u s . but these are absent in P. villosus. Minimum evolution, maximum parsimony and The distinguishing diagnostic characters of P. aemulatus maximum likelihood (heuristic search) analyses produced are: (1) the presence of dorsal papillae arranged in two identical 50% majority-rule bootstrap consensus tree double rows along the dorsal radii and two single rows topologies, with varying bootstrap values (Figure 2A). along the lateral radii where the dorsal and ventral The ML exhaustive search produced a very similar phylo- surfaces meet; (2) the shape of the ossicles in the gonad genetic tree di¡ering only in the intraspeci¢c relationships irregular bodies and rods with short processes near the of P.aemulatus. Of the 135135 trees evaluated in the exhaus- ends; and (3) the shape of the ossicles in the tentacles tive search, a single tree was retained with the best score of slender, spiny-bifurcated and smooth^slender rods with a lnL 1834.16714 (Figure 2B). central knob. The nucleotide distance between P. aemulatus and P. villosus and P. mollis is COI: d¼39%, su⁄cient to DISCUSSION support distinct species status. In sea cucumbers, a COI: Mortensen (1927) recorded seven species of d¼15% is enough to identify a distinct species status, as Pseudostichopus from the north-east Atlantic: P. atlanticus argued by Arndt et al., (1996). This is also compatible R. Perrier, from 42819 0N238360W, 4 0 6 0 ^4068 m; with recent studies using COI sequences as a species P. depressus He¤rouard, 398540N20817 0W, 4360 m (Princesse ‘barcode’ in which the majority of echinoderm pairwise Alice; only one specimen known); P. g l o b i g e r i n a e He¤rouard, species comparisons showed per cent sequence divergences Bay of Biscay, 4380 m (Princesse Alice); P. lapidus of 4^16% with a mean of 10.9% (Hebert et al., 2003). He¤rouard, o¡ Azores, 4020 m (Princesse Alice; only one Therefore, based on partial sequences of the COI gene specimen known); P. marenzelleri He¤rouard, 368540N from sympatric specimens and ossicle morphology, we 208460W, 4400 m (Princesse Alice); P. occultatus conclude that P.villosus and P.aemulatus are not conspeci¢c. Marenzeller, o¡ north-west Spain, 500 m (Hirondelle), We found that there are only seven amino acid di¡er- also found in the Mediterranean, 415^1445 m and ences in the DNA sequences of the mitochondrial gene P. villosus The¤el, 37816 0N20811 0W, 4275 m (Princesse Alice), COI gene between the aspidochirotid families Synallac- otherwise of cosmopolitan distribution, 2600^5300 m. tidae (Pseudostichopus spp.) and Stichopodidae (Isostichopus The list of Pseudostichopus species known from the North fuscus). Other pairwise amino acid comparisons between Atlantic, based on this revision with the new species added families and orders of sea cucumbers show di¡erences is as follows (in alphabetic order): ranging from 11 to 31 amino acids (Arndt et al., 1996). The di¡erence in the number of amino acids within the Pseudostichopus aemulatus Sol|¤s-Mar|¤n & Billett Pseudostichopus species ranged from 4 to 18. This is higher Pseudostichopus depressus He¤rouard than that found by Arndt et al., (1996) within the family Pseudostichopus lapidus He¤rouard Cucumariidae, were it ranged from 1 to 8. Pseudostichopus marenzelleri He¤rouard The large variation in amino acids in the species of Pseudostichopus occultatus Marenzeller Pseudostichopus shows that the intergeneric relationships Pseudostichopus villosus The¤el within Synallactidae are probably not fully resolved. This Pseudostichopus aemulatus di¡ers from P. marenzelleri in the is re£ected in the trichotomy of the unrooted tree (Figure lack of small thread-like papillae over whole body. Also, 2B). However, in relation to P. villosus and P. aemulatus the in P. marenzelleri, ventral podia form irregular groups phylogenetic analysis clearly supports separate species around the mouth and in the inferior third of the body. status of these sympatric morphotypes (Figure 2B), as indi- These were not apparent in P. aemulatus. cated by the morphological analysis. Pseudostichopus villosus Pseudostichopus aemulatus has ossicles in the gonad wall. forms a distinct and separate clade to P. aemulatus and This is unlike P. occultatus. In addition, the shapes of the P.mollis in all tree topologies with bootstrap values of 100%.

Journal of the Marine Biological Association of the United Kingdom (2004)

Downloaded from https:/www.cambridge.org/core. Open University Library, on 28 Jan 2017 at 14:21:44, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/S002531540401046Xh 1084 F.A. Sol|¤s-Mar|¤n et al. DNA sequence of a new Pseudostichopus species

Pseudostichhopus aemulatus is also well supported as a mono- Cantatore, P., Roberti, M., Renaldi, G., Gadaleta, M. & phyletic clade by 100% bootstrap values. Pseudostichopus Saccone, C., 1989. The complete nucleotid sequence, gene mollis appears as the sister group to the P. aemulatus clade organization, and genetic code of the mitochondrial genome with bootstrap values of 98% and 100% (Figure 2A). The of Paracentrotus lividus. Journal of Biology and Chemistry, 264, trees are all rooted with the outgroup, Isostichopus fuscus.No 10965^10975. Grube, A.E., 1840. Actinien, Echinodermen und Wu«rmer des further assumptions can be made on the intrageneric Adriatischen und Mittelmeers,Ko«nigsberg, 33^43 pp. relationships of P. v i l l o s u s , P. mollis and P. aemulatus.These Hebert, P.D.N., Ratnasingham, S. & Waard, J.R. de, 2003 preliminary ¢ndings suggest that molecular analysis on the Barcoding animal life: cytochrome c oxidase subunit 1 diver- inclusive species of the Pseudostichopus genus is required to gences among closely related species. Proceedings of the Royal clarify the current controversial taxonomic classi¢cation of Society B, (Supplement Biology Letters) DOI 10.1098/ the genus, and possibly as a result of these initial results, of rsbl.2003.0025. the inclusive families of the Aspidochirotida. Himeno, H., Masaki, H., Kawai, T., Kumagai, I., Miura, I. & The genus Pseudostichopus is a taxon where external Watanabe, K., 1987. Unusual genetic codes and novel Ser morphology and ossicle shape simplicities have led to gene structure for tRNAAGY in star¢sh mtDNA. Gene, 56, uncertainties in the taxonomy at the species level. This 219^230. study demonstrates that recognition of new species using Jacobs, H.T., Elliot, D.J., Math, V.B. & Farquharson, A., 1988. Nucleotide sequence and gene organization of sea urchin mito- a combination of morphological and molecular characters chondrial DNA. Journal of Molecular Biology, 202,185^217. together will bring a better understanding on the systema- Ludwig, H.L., 1894. Reports on an exploration o¡ the west coast tics of this group. It also demonstrates the concept of the of Mexico, Central and South America, and o¡ Galapagos COI partial sequence as a molecular barcode for species Island, in Charge of Alexander Agassiz. by the US Fish level taxonomy in holothurians. Commission Steamer ‘‘Albatross’’ during 1891, Lieut. Commander Z. L. Tanner, U.S.N. Commanding. 12. The Holothurioidea. Memoirs of the Museum of Comparative Zoology at We thank Nadia Ameziane at the Muse¤um National d’Histoire Harvard College, 17,1^183. Naturelle, Paris; Dr Claus Nielsen and Margit Jensen at the Maddison, W.P. & Maddison, D.R., 1989. Interactive analysis of Zoologiske Museum in Copenhagen; David Pawson and Cynthia phylogeny and character evolution using the computer Ahearn at the United States National Museum of Natural program MacClade. Folia Primatologica, 53,190^202. History for hosting the principal author at their respective Maddison, W.P. & Maddison, D.R., 1992. MacClade: analysis of institutions and for opening to us the Synallactidae collection for phylogeny and character evolution. Version 3.0. Sunderland, Mass: morphological and molecular analysis. We are very grateful to Sinauer Associates. Arturo Nun‹ o Hermosillo at CUCBA, Universidad de Mortensen, T., 1927. Handbook of the Echinoderms of the British Isles. Guadalajara, Mexico and Dinorah Herrero Perezrul at the London: Oxford University Press. Centro Interdisciplinario de Ciencias Marinas, La Paz, Baja Posada, D. & Crandall, K.A., 1998. MODELTEST: testing the California, Mexico for making available specimens of Isostichopus Bioinformatics 14 fuscus. Dr Paul Tyler (SOC), Rodrigo Maggioni (SOES, SOC) model of DNA substitution. , ,817^818. Rice, A.L., 1992. Benthic biology at the European community and two anonymous referees made helpful remarks on the manu- 0 0 script. The ¢rst author was supported by a PhD grant from station (48850 N16830 W) and in the Porcupine Seabight. DGAPA (Direccio¤n General de Asuntos del Personal RRS Challenger cruise 79. 12 May^03 June 1991. Institute of Acade¤mico), UNAM. The ¢rst and second authors were sup- Oceanographic Sciences Deacon Laboratory Cruise Report no. 231, ported by a grant from the European Commission, Copenhagen 52 pp. Biosystematics Centre (COBICE). The sample collection was Rice, A.L., 1996. BENGAL. High resolution, temporal and funded, in part, by the contract 950018 under the EU MAST 3 spatial study of the BENthic biology and Geochemistry of a programme. north-eastern Atlantic abyssal Locality. RRS Discovery cruise 222, Leg 2. 29 August^24 September 1996. Southampton REFERENCES Oceanography Centre Cruise Report no. 4, 86 pp. Rice, A.L., 1997. BENGAL. High resolution, temporal and Arndt, A., Ma¤rquez, C., Lambert, P. & Smith, M.J., 1996. spatial study of the BENthic biology and Geochemistry of a Molecular phylogeny of eastern Paci¢c sea cucumbers north-eastern Atlantic abyssal Locality. RRS Discovery cruise (Echinodermata: Holothuroidea) based on mitochondrial 226. 12 March^10 April 1997. Southampton Oceanography Centre DNA sequence. Molecular Phylogenetics and Evolution, 6,425^437. Cruise Report no. 13, 76 pp. Asakawa, S., Kumasawa, Y., Araki, T., Himeno, H., Miura, K. Sol|¤s-Mar|¤n, F.A., 2003. Molecular phylogeny, systematics and biology & Watanabe, K., 1991. Strand-speci¢c nucleotid composition of the holothurian family Synallactidae. PhD thesis, University of bias in echinoderm and vertebrate mitochondrial genomes. Southampton, UK. Journal of Molecular Evolution, 32,511^520. Swo¡ord, D.L., 1998. PAUP*: phylogenetic analysis using parsimony Billett, D.S.M., 1988. The ecology of deep-sea holothurians.PhD (*and other methods), version 4.0. Sunderland, Mass: Sinaeur thesis, University of Southampton, UK. Associates. Billett, D.S.M., 2000. Temporal and spatial variability of benthic The¤el, H., 1886. Report on the Holothurioidea dredged by the communities on the Porcupine Abyssal Plain and in the HMS Challenger during the years 1873^1876. Part II. Report of Porcupine Seabight. RRS Challenger cruise142.19 April^19 May the Scienti¢c Results of the Voyage of HMS Challenger 1873^1876, 1999. Southampton Oceanography Centre Cruise Report no. 30, 63 pp. Zoology, 14,1^290. Billett, D.S.M., Bett, B.J., Rice, A.L., Thurson, M.H., Gale¤ron, Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. & J., Sibuet, M. & Wol¡, G.A., 2001. Long term change in the Higgins, D.G., 1997. The ClustalX windows interface: £exible megabenthos of the Porcupine Abyssal Plain (NE Atlantic). strategies for multiple sequence alignment aided by quality Progress in Oceanography, 50,325^348. analysis tools. Nucleic Acids Research, 24,4876^4882. Billett, D.S.M. & Rice, A.L., 2001. The BENGAL programme: introduction and overview. Progress in Oceanography, 50,13^25. Submitted 5 September 2003. Accepted 24 June 2004.

Journal of the Marine Biological Association of the United Kingdom (2004)