Ascidian molecular phylogeny inferred from mtDNA data with emphasis on the Aplousobranchiata
Introduction
Tunicates are a key group in studies of deuterostome evolution, particularly those addressing the origin of chordates. Recently, molecular tools have been applied to the study of topics such as deuterostome phylogeny (Cameron et al., 2000; Holland, 1991; Turbeville et al., 1994; Wada and Satoh, 1994; Winchell et al., 2002), ancestral chordate lifestyle (Wada, 2000), the origin of ascidian coloniality (Jacobs et al., 2000; Wada et al., 1992), and the origin of anural development (Hadfield et al., 1995; Jeffery et al., 1999). Molecular data have also been used to shed light on relationships between major Tunicate groups (Christen and Braconnot, 1998; Stach and Turbeville, 2002; Swalla et al., 2000). In general, these studies have supported the monophyly of the Tunicata, but not of the class Ascidiacea, as the Thaliacea appear related to the Phlebobranchiata. The position of the Appendicularia remains controversial (Swalla et al., 2000), although a recent study suggests that they are the sister group to the aplousobranch ascidians (Stach and Turbeville, 2002). Phylogenetic relationships within the ascidians, on the other hand, have received little attention in molecular studies.
Since the seminal works of Seeliger (1885) and Lahille (1890), and other studies in the last century (e.g., Berrill, 1936, Berrill, 1955; Garstang, 1928; Millar, 1966; Tokioka, 1971) the debate on the characteristics of the ascidian ancestor and the main evolutionary lines within the group has relied on non-cladistic schemes. Taxonomically oriented studies currently classify the ascidians according to two main characters, one considering the structure of the branchial basket and the other the position of the gonads. Lahille, 1886, Lahille, 1887, Lahille, 1890 established the groups Aplousobranchiata, Phlebobranchiata, and Stolidobranchiata on the basis of a progressive structural complication of the branchial sac. Aplousobranch ascidians have simple branchial walls with only transverse vessels between rows of stigmata; phlebobranch pharynges have distinct papillae projecting from the transverse vessels, and these papillae are in many cases connected by longitudinal vessels. In stolidobranch species the branchial wall features longitudinal vessels, but it is plicated, forming internal longitudinal folds. In addition, Perrier (1898) and later Garstang (1928), developed a classification scheme based on gonad position, either attached to the body wall (Pleurogona) or associated with the digestive system (Enterogona). In fact, it is common to find a combination of the two schemes, in which ascidians are classified into the orders Enterogona and Pleurogona, the former comprising the suborders Aplousobranchiata and Phlebobranchiata and the latter the suborder Stolidobranchiata (Berrill, 1950; Kott, 1985; Millar, 1970). However, other authors use only Lahille’s classification, and Aplousobranchiata, Phlebobranchiata, and Stolidobranchiata are ranked as orders (Harant and Vernières, 1933; Monniot et al., 1991; Van Name, 1945).
The relationships between the members of Aplousobranchiata are unclear (Kott, 1990), a fact that is reflected in the poor agreement in its internal classification. Some recent, taxonomically oriented studies have tended to accept a reduced number of large groupings; for instance, the Aplousobranchiata would comprise only the families Polycitoridae, Polyclinidae, and Didemnidae (Monniot et al., 1991; Nishikawa, 1990), or the families Clavelinidae, Holozoidae, Polycitoridae, Polyclinidae, and Didemnidae (Monniot and Monniot, 2001). In contrast, Kott, 1985, Kott, 1990, Kott, 1992, Kott, 2001, in her detailed monographs on the Australian ascidians, considered up to 14 families within the Aplousobranchiata on the basis of careful morphological observations. The placement of the Cionidae and Diazonidae within the Phlebobranchiata (the traditional view) or the Aplousobranchiata (Kott, 1969, Kott, 1990) is another point on which no general agreement has been reached (see Stach and Turbeville, 2002).
New, independent sets of data such as sequence information, in combination with formal evolutionary analyses, may be particularly useful in ascertaining ascidian evolution and establishing a sound classification scheme. To date, however, molecular studies have been applied only to a small fraction of ascidian species. The main groups (at the order/suborder level) have generally been recovered in molecular analyses (Stach and Turbeville, 2002; Swalla et al., 2000), although many traditional families appeared as paraphyletic or polyphyletic. In addition, limited data are available on aplousobranch ascidians. In a thorough study combining molecular data and morphology, Stach and Turbeville (2002) were the first to generate aplousobranch sequences of the genes 18S rRNA (three species) and cytochrome c oxidase subunit I (COI, five species). In addition, Kakuda (2001) and Kurabayashi et al. (2003) utilized mitochondrial DNA data to address some phylogenetic problems in ascidians. The mitochondrial COI gene, due to its high variability, has been the molecule of choice in studies of population genetics and phylogeography (Avise, 2000), and has been used in ascidians to address cryptic speciation and invasions (López-Legentil and Turon, 2004; Tarjuelo et al., 2001; Turon et al., 2003). However, the COI gene may provide useful information at higher taxonomic levels (Hebert et al., 2003; Remigio and Hebert, 2003). Problems such as incomplete lineage sorting and introgression that hinder analyses at the species level using mitochondrial data (Ballard and Whitlock, 2004) may not be relevant over the larger timescales of evolutionary processes.
In this study, we present novel partial COI sequences for 28 ascidian species, of which 21 belong to the Aplousobranchiata sensu Kott (1990). We added nine sequences (five of them from aplousobranchs) published by other authors to obtain a representative database of 11 families sensu Kott, 1985, Kott, 1990, Kott, 1992, Kott, 2001. Although our study is limited to a single gene, thus necessitating more data sources before definitive conclusions can be reached, our aim was to explore the contribution of COI sequence data to the phylogeny of ascidians, particularly the aplousobranchs. Given the paucity of formal phylogenetic studies (molecular or morphological) in ascidians, we believe that this new database will represent a step forward in the unravelling of the evolutionary pattern in this group.
Section snippets
Ascidian samples
Twenty-eight species of ascidians, 21 of the Aplousobranchiata (following the assignment of Kott, 1990), three Phlebobranchiata and four Stolidobranchiata were collected from sites in the Balearic Islands and NE Spain (Western Mediterranean) by SCUBA diving (Table 1). Two samples were from the Eastern Atlantic (Clavelina oblonga from Azores and Archidistoma aggregatum from Galicia, NW Spain). Taxonomic identification was performed following Turon (1987), with the exception of Pycnoclavella sp.,
Sequence saturation
A fragment of 617 bp from the mitochondrial COI gene was compared for 37 ascidian species. Four hundred and thirteen variable sites were found in the data set, of which 361 were parsimony informative. Of the nucleotide substitutions, 31.4% occurred at first codon positions, 11.6% at second codon positions, and 56.9% occurred at third positions (Fig. 1), with an overall transition/transversion ratio of 0.91. To check the possibility of this gene being saturated at the phylogenetic level
Discussion
The combination of diverse analytical approaches, each with a different underlying philosophy, may be particularly useful for testing the robustness of the phylogenetic signal recovered from the data. The results obtained using MP, ML, and BI methods differed in some topological aspects, but the main groups obtained were similar. Stach and Turbeville (2002) reported a lack of congruence between results obtained with COI and 18S rRNA data in ascidians, and attributed this either to a high
Acknowledgments
Dr. Thomas Stach (Smithsonian Institution) kindly provided the alignment of his sequences and helpful comments on a draft of the manuscript. Two anomymous reviewers improved greatly the submitted version. Dr. Peter Wirtz (University of Madeira) collected and sent the specimens of Clavelina oblonga. Dr. Isabel Tarjuelo and Dr. Sandra Duran (University of Barcelona) did the sequencing work for Clavelina lepadiformis, C. dellavallei, C. oblonga, and Pycnoclavella sp. Rocı́o Pérez-Portela kindly
References (68)
- et al.
Molecular phylogeny of eastern Pacific sea cucumbers (Echinodermata: Holothuroidea) based on mitochondrial DNA sequence
Mol. Phylogenet. Evol.
(1996) - et al.
Mitochondrial pseudogenes: evolution’s misplaced witnesses
TREE
(2001) - et al.
Vanadium content and oxidation state related to ascidian phylogeny
Comp. Biochem. Physiol.
(1983) Cloning and evolutionary analysis of msh-like homeobox genes from mouse, zebrafish and ascidian
Gene
(1991)- et al.
Testing the utility of partial COI sequences for phylogenetic estimates of gastropod relationships
Mol. Phylogenet. Evol.
(2003) - et al.
The general stochastic model of nucleotide substitution
J. Theor. Biol.
(1990) - et al.
Phylogeny of Tunicata inferred from molecular and morphological characters
Mol. Phylogenet. Evol.
(2002) - et al.
Model of amino acid substitution in proteins encoded by mitochondrial DNA
J. Mol. Evol.
(1996) Phylogeography. The History and Formation of Species
(2000)- et al.
The incomplete natural history of mitochondria
Mol. Ecol.
(2004)