<i>Tergipes tergipes</i>(Förskal, 1775) (Gastropoda: Nudibranchia) is an amphiatlantic species

Nearly 13% of ‘opisthobranch’ gastropods in the Atlantic Ocean are believed to show amphiatlantic distributions (García & Bertsch, 2009). This assumption, which has been based mainly on morpho-anatomical similarities, has rarely been tested within a molecular phylogenetic framework. During the past decade, molecular methods have frequently shown that cosmopolitan or widely distributed heterobranch ‘species’ consist of taxonomic complexes of multiple linages (Malaquias & Reid, 2008, 2009; Carmona et al., 2011, 2014a, b; Jörger et al., 2012; Ornelas-Gatdula et al., 2012; Pola et al., 2012; Krug et al., 2013). At present, the aeolid genus Tergipes Cuvier, 1805 contains six species (Bouchet & Gofas, 2013): T. antarcticus Pelseneer, 1903; T. brochiRisso, 1818; T. dicquemariRisso, 1818; T. edwardsii Nordmann, 1844; T. tergipes (Förskal, 1775) and T. valentini (Elliot, 1907). However, the two species described by Risso (1818) have not been encountered in the past century. Regarding T. valentini, this species was transferred to Cuthona by Schrödl (2003) because it has more than a single ceras per row. However, this reassignment has been overlooked by later researchers (e.g. Valdés et al., 2012). Tergipes tergipes is regarded as an amphiatlantic boreal species (Franz, 1970; Thompson & Brown, 1984; Templado et al., 1987; Bleakney, 1996). It has been reported from Iceland (Lemche, 1938; Platts, 1985) to the southern Iberian Peninsula, including Sweden, Denmark, The Netherlands, the British Isles and Atlantic coast of France (Schmekel & Portmann, 1982; Thompson & Brown, 1984; Cervera et al., 2004). Schmekel & Portman (1982), Betti (2011) and Martynov & Korshunova (2011) reported specimens of T. tergipes from the Mediterranean, Adriatic, and the Barents and the Black Seas respectively. The western Atlantic distribution of this species ranges from Newfoundland (Canada) to New Jersey (USA) (Bleakney, 1996). In addition, a single specimen of T. tergipes has been reported from Brazil (Marcus, 1957). Although the amphiatlantic distribution of this species has been generally accepted, Bleakney (1996) expressed doubt about the conspecificity of populations on either side of the Atlantic. In this study we use molecular phylogenetic methods to test the hypothesis that the populations of T. tergipes on the eastern and western coastlines of the Atlantic Ocean are conspecific. Samples were collected by SCUBA diving with standard sampling techniques for heterobranchs and obtained from wet collections housed at several museums, namely California Academy of Sciences, San Francisco, USA (CASIZ), Museo Nacional de Ciencias Naturales, Madrid, Spain (MNCN) and Zoological Museum of Moscow State University (ZMMU). Nineteen specimens of T. tergipes were studied. Thirty-four additional sequences were obtained from GenBank (see Table 1 for full list of samples, localities and vouchers), including one sequence of T. antarcticus. Tritonia challengeriana Bergh, 1884 was chosen as a distant outgroup because of its basal phylogenetic position within Cladobranchia (Pola & Gosliner, 2010). Tissue samples were taken from the foot. Extraction, amplification, purification and sequencing of portions of the COI, 16S rRNA and H3 genes followed the methods of Carmona et al. (2013, 2014b). Sequence reactions were run on a 3730XL DNA sequencer (Applied Biosystems). Sequences were verified by forward and reverse comparisons and have been deposited in GenBank (Table 1). List of specimens used for phylogenetic analyses. EA, Eastern Atlantic; EP, Eastern Pacific; GB, GenBank. List of specimens used for phylogenetic analyses. EA, Eastern Atlantic; EP, Eastern Pacific; GB, GenBank. Sequences were assembled and edited with Geneious Pro v. 4.7.6 (Drummond et al., 2009), aligned in MAFFT (Katoh et al., 2009) and further checked using MacClade v. 4.06 (Maddison & Maddison, 2005). Uncorrected pairwise p-distance values between each taxon were calculated for the COI gene. The best-fit evolutionary model (GTR+I+G for COI and 16S and GTR+G for H3) was determined in MrModeltest v. 2.3 (Nylander, 2004), using the Akaike information criterion (Akaike, 1974). MrBayes v. 3.1.2 (Ronquist & Huelsenbeck, 2003) was used for Bayesian inference analysis and to estimate posterior probabilities (PP) for node support with two runs of 5,000,000 generations each. Convergence was checked in TRACER v. 1.5 (Drummond & Rambaut, 2007) with a burn-in of 25%. We also applied the Automatic Barcode Gap Discovery (ABGD) method to detect species-level clusters (Puillandre et al., 2012a). ABGD is a distance-based method designed to detect the so-called ‘barcode gap’ in the distribution of pairwise distances calculated in a COI alignment (Puillandre et al., 2012a, b). The web-based ABGD program (available at http://wwwabi.snv.jussieu.fr/public/abgd/) was employed with the default settings to generate a preliminary partition of sequences, using the COI alignment and excluding the outgroup. In addition an unrooted statistical parsimony network was generated for COI using TCS v. 1.21 (Clement et al., 2000) with a 95% connection limit. The combined dataset of three genes yielded a sequence alignment of 1,427 positions. Trees produced by separate analyses of H3, 16S and COI genes (not shown) showed the same topology and similar resolution as the three-gene tree (Fig. 1). All the specimens of T. tergipes clustered together in a single clade (PP = 1.0). The uncorrected p-distance for COI among the specimens ranged from 0.0% to 2.7% (between one specimen from The Netherlands and one from Italy). Applying the ABGD method resulted in a single partition with one group containing all the specimens of T. tergipes included in this study. The prior maximal distance was 0.001. Fourteen haplotypes were identified in the fifteen specimens sequenced for COI (Fig. 2). The haplotype network did not suggest any structure correlated with geographical origin. In addition, we did not find external morphological differences among the specimens studied (Fig. 3). Phylogenetic hypothesis based on the combined dataset (H3+COI+16S) inferred by Bayesian analysis. Numbers above branches are posterior probabilities; values >0.95 indicate strong support. Specimens are listed in Table 1. COI haplotype network for Tergipes tergipes in The Netherlands (N), Wales (W), Italy (I), the Barents Sea (BS), Maine (M) and New Hampshire (NH). Numbers in brackets indicate the haplotype frequency. Tergipes tergipes. A. Wales (photo by David Kipling), 7 mm length. B. The Netherlands (photo by Peter H. van Bragt), 5 mm length. C. Barents Sea, Russia (photo by Tatiana Korshunova), 5 mm length. D. Maine (photo by Terrence M. Gosliner), 4 mm length. These results support the amphiatlantic status of T. tergipes, because the methods used here have not detected any population differentiation or structure. Todd (1981) highlighted the opportunist behaviour of this small aeolid, which preys on a wide range of hydroids. In addition, this species has been reported in estuarine areas of the North Sea (Swennen, 1961), indicating resistance to low salinities. In the Barents Sea this species has been found in fouling communities and on the drifting objects off the coast (Martynov et al., 2006). These characteristics may explain the wide distribution of this species and highlight its ability to survive in a range of habitats. We are deeply grateful to all individuals who helped to collect and provided specimens and images for this study, including T. Korshunova, D. Kliping, P. H. van Bragt, T. D'Onofrio and T.M. Gosliner. This work was supported by the following research grants: Spanish Ministry of Economy and Competitiveness (includes former Ministry of Sciences and Innovation) (CGL2006-05182/BOS and CGL2010-17187), National Science Foundation (DEB 0329054 PEET) and Russian Foundation for Basic Research (No 13-04-01641a). This is CEI·MAR journal publication 52.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.