MOLECULAR PHYLOGENETICS AND EVOLUTION Vol. 9, No. 1, February, pp. 55-63, 1998 article no FY970439 Details of Gastropod Phylogeny Inferred from 18S rRNA Sequences Birgitta Winnepenninckx ,*'1 Gerhard Steiner,t Thierry Backeljaui,* and Rupert De Wachter* * Departement Biochemie, Universiteit Antwerpen (UIA), Universiteitsplein 1, B-2610 Antwerpen, Belgium; i Institute o f Zoology, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria; and tRoyal Belgian Institute of Natural Sciences, Vautierstraat 29, B -1000 Brussel, Belgium Received February 4, 1997; revised June 6, 1997 molecular data (e.g., Tillier et al., 1992, 1994, 1996; Some generally accepted viewpoints on the phyloge­ Rosenberg et al., 1994; Winnepenninckxet al., 1996). A netic relationships within the molluscan class Gas­ recent 18S rRNA analysis of molluscan relationships tropoda are reassessed by comparing complete 18S suggested that this molecule might be suitable to rRNA sequences. Phylogenetic analyses were per­ resolve phylogenetic problems at infraclass levels (Win­ formed using the neighbor-joining and maximum par­ nepenninckxet al., 1996). In the present paper we simony methods. The previously suggested basal posi­ further explore this issue by analyzing a number of tion of Archaeogastropoda, including Neritimorpha generally accepted ideas on the infraclass phylogeny of and Vetigastropoda, in the gastropod clade is con­ Gastropoda using 11 new and 7 published (Winnepen­ firmed. The present study also provides new molecular ninckx et al., 1992, 1994, 1996) complete gastropod 18S evidence for the monophyly of both Caenogastropoda rRNA sequences. The points dealt with are the position and Euthyneura (Pulmonata and Opisthobranchia), and suggested paraphyly of Prosobranchia and Archaeo­ making Prosobranchia paraphyletic. The relation­ ships within Caenogastropoda and Euthyneura data gastropoda, as well as the monophyly of taxa such as turn out to be very unstable on the basis of the present Caenogastropoda, Neotaenioglossa, Muricacea, Euthy­ 18S rRNA sequences. The present 18S rRNA data ques­ neura, Pulmonata, and Stylommatophora. In this con­ tion, but are insufficient to decide on, muricacean text, particular attention will' be paid to the monophyly (Neogastropoda), neotaenioglossan, pulmonate, or sty- and position of the Systellommatophora, a group which lommatophoran monophyly. The analyses also focus includes the families Veronicellidae, Onchidiidae, and on two systellommatophoran families, namely, Veroni­ Rathousiidae, and according to some authors also the cellidae and Onchidiidae. It is suggested that Systellom­ Rhodopidae (von Salvini-Plawen, 1970) and the matophora are not a monophyletic unit but, due to the Smeagolidae (Climo, 1980; Tillier and Ponder, 1992). lack of stability in the euthyneuran clade, their affinity Systellommatophora are considered to be either pulmo- to either Opisthobranchia or Pulmonata could not be nates (e.g., Van Mol, 1974; Solem, 1979; Tillier, 1984; determined. i 1998 Academic Press Haszprunar, 1988b; Haszprunar and Huber, 1990; Tillier and Ponder, 1992) or opisthobranchs (e.g., Boett- ger, 1955), although von Salvini-Plawen (1970) consid­ INTRODUCTION ered them as a proper subclass, the Gymnomorpha, related to the opisthobranchs. Their status as a sepa­ Gastropoda is the largest molluscan class and in­ rate group was confirmed by von Salvini-Plawen and cludes the common terrestrial, freshwater, and marine Steiner (1996), who related them to the pulmonates. snails and slugs. It has an excellent fossil record going However, systellommatophoran monophyly (e.g., von back to 550 MYA (Runnegar and Pojeta, 1985). The Salvini-Plawen, 1970) is still debated (e.g., Climo, class is traditionally divided into three subclasses; 1980; Tillier, 1984; Haszprunar and Huber, 1990). Prosobranchia (Streptoneura), Opisthobranchia, and Pulmonata (together the latter constitute the Euthy­ MATERIALS AND METHODS neura). The phylogenetic relationships between and within these subclasses are longstanding problems Amplification and Sequencing of the 18S rRNA Genes (e.g.. Ponder and Lindberg, 1996, 1997; von Salvini- The taxonomy of the gastropod species used in this Plawen and Steiner, 1996; for a review of earlier work study is given in Table 1. Sampling locations are listed see Bieler, 1992), which are increasingly studied with in Table 2. The species were frozen alive. After dissec­ » tion, DNA was extracted (Winnepenninckxet al., 1993) ‘ Present address: Royal Belgian Institute for Natural Sciences, from the tissues indicated in Table 2. The 18S rRNA Vautierstraat 29, B-1000 Brussel, Belgium. genes were PCR-amplified, cloned, and sequenced as 55 1055-7903/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved. 56 WINNEPENNINCKX ET AL. TABLE 1 Taxonomy of the Gastropod Species Used in This Study « Subclass Superorder Order Suborder Family Genus Prosobranchia“ Archaeogastropoda“ Vetigastropoda Trochidae M onodontab Neritimorpha Neritidae N erita Caenogastropoda Neogastropoda Buccinidae P isaniab Nassariidae N a ssa riu sb Fasciolariidae Fasciolariab Muricidae Thais Neotaenioglossa Discopoda Littorinidae L itto rin a b Bursidae B u rsa b Calyptraeidae Crepidula11 Pulmonata Stylommatophora Mesurethra Clausiliidae B aleab Sigmurethra Holopodopes Achatinidae L im ico la n a Aulocopoda Succineidae O xytom ab Holopoda Helicidae H elix Basommatophora Siphonariidae Siphonaria Systellommatorphora Qnchidiacea i Onchidiidae O nchidella Soleolifera Veronicellidae Laevicaulisb Opisthobranchia Anaspidea Aplysidae A p lysia6 Note. The taxonomy of the Archaeogastropoda is based on Haszprunar (1988b, Table 5, p. 428), except for its ranking as superorder; the Caenogastropoda are classified according to Ponder and Warén (1988). The classification of the Neogastropoda is based on Ponder (1973); Pulmonata are classified according to Solem ( 1979), except for the placement of Siphonaria, which follows von Salvini-Plawen (1970). “ Currently considered nonmonophyletic. h Sequence determined in this study. described by Winnepenninckxet al. (1995), using the structure similarity. If necessary, manual adjustments primers published in Winnepenninckxet al. (1994) and were made with the same program. The secondary two M13 universal primers. structure model of Van de Peeret al. (1996a) was used. Data Analysis Sequence regions corresponding to the amplification primer at the 5' end of the gene were removed prior to The new gastropod 18S rRNA sequences were added reconstruction of phylogenetic trees. The 18S rRNA to the alignment of Van de Peer et al. (1996a) using the sequences were analyzed using neighbor-joining (NJ) computer program DCSE (De Rijk and De Wachter, and maximum parsimony (MP) methods. The program 1993), which considers primary as well as secondaryTREECON (Van de Peer and De Wachter, 1993) was used to construct NJ trees based on the formulas of TABLE 2 Jukes and Cantor (1969), Kimura (1980), or Van de Peer et al. (1996b). Gaps were not taken into account. Sources and Tissue Type of Gastropod Species Used Tree stability was assessed via bootstrapping over 1000 for This Study replicates. MP trees were constructed on the phyloge- netically informative sites using either the heuristic or Species Sampling location Tissue the exhaustive search option of PAUP (Swofford, 1993). Aplysia sp. Hong Kong Digestive gland Stability of MP trees was assessed via bootstrapping Balea biplicata Mortsel (Belgium) Complete organism over 1000 replicates and the calculation of decay indi­ Bursa rana Hong Kong Bucal mass * foot ces (Bremer, 1988; Donoghueet al., 1992). As until now muscle there is no consensus about the minimal bootstrap Crepidula adunca Vancouver (Canada ! Albumen gland Fasciolaria lignaria Bahar Ie-Cagnaq Albumen gland value necessary to regard a cluster as firmly supported, (Malta) bootstrap values were arbitrarily considered to reflect Laevicaulis alte Laboratory bred Digestive gland strong support if they exceeded 70% (Hillis and Bull, (Görlitz, Germany i 1993). Littorina obtusata Oosterschelde (Nether- Albumen gland - lands) muscle tissue Monodonta labio Hong Kong Digestive - reproduc­ RESULTS tive gland Nassarius singuin- Hong Kong Penis + foot muscles jarensis New gastropod 18S rRNA sequences were submitted Oxytoma sp. Görlitz (Germany) Digestive gland to the EMBL sequence data library and have the Pisania striata Bahar Ic-Cagnaq Digestive f reproduc- following accession nos.; Aplysia sp., X94268;Balea (Malta) tive gland biplicata, X94278; Bursa rana, X94269; Crepidula GASTROPOD PHYLOGENY 5 7 adunca, X94277;Fasciolaria lignaria, X94275;Laevi- Discopoda) nor Muricacea (represented by Buccinidae, caulis alte, X94273; Littorina obtusata, X94274;Mo­ Fasciolariidae, Muricidae, and Nassariidae) form mono- nodonta labio, X94271; Nassarius singuinj or ensis, phyletic groups. There is 100% bootstrap support for X94273; Oxyloma sp., X94276;Pisania striata, X94272. the monophyly of Euthyneura, which form two unsup­ Figure 1A shows the NJ tree obtained on the basis of ported clades consisting of ( 1 ) the three stylommatopho- the Jukes and Cantor (1969) distances of an alignment rans, Helix aspersa, Ba. biplicata, and Oxyloma sp., of complete 18S rRNA sequences of 18 gastropods. Theand (2) the five remaining euthyneurans. Surprisingly, bivalve Galeomma takii was arbitrarily chosen as the achatinid Limicolana kambeul belongs to this outgroup. The same topology was obtained with Kimuralatter clade instead of to the first. Both