Hydrobiologia 383: 147–154, 1998. E. Schockaert, N. Watson & J.-L. Justine (eds), Biology of the Turbellaria. ©1998 Kluwer Academic Publishers. Printed in the Netherlands.
147
Flatworm phylogeny from partial 18S rDNA sequences Ulf Jondelius Swedish Museum of Natural History, POB 50007, S-104 05 Stockholm, Sweden. E-mail:
[email protected] Key words: Platyhelminthes, Acoelomorpha, phylogeny, 18S rDNA, cladistic, decay index, jack-knife
Abstract Partial 18S rDNA sequences from 29 flatworms and 2 outgroup taxa were used in a cladistic analysis of the Platyhelminthes. Support for the clades in the resulting single most parsimonious tree was estimated through bootstrap analysis, jack-knife analysis and decay indices. The Acoelomorpha (Acoela and Nemertodermatida) were absent from the most parsimonious tree. The Acoela and the Fecampiidae form a strongly supported clade, the sister group of which may be the Tricladida. There is some support for monophyly of the rhabdocoel taxon Dalyellioida, previously regarded as paraphyletic. The sister group of the Neodermata remains unresolved.
Introduction The phylogeny of the phylum Platyhelminthes has received a lot of attention in recent years with the possibility of non-monophyly of the flatworms as one principal point of interest (Smith et al., 1986; Haszprunar, 1996). The relationship of the parasitic taxon Neodermata to free-living flatworms was discussed by numerous authors (e.g. Ax, 1984; Brooks, 1989). Studies of flatworm evolutionary history utilising morphological characters have, however, yielded conflicting hypotheses regarding these questions (Ehlers 1984a, 1985, 1986; Rohde, 1990; Jondelius & Thollesson, 1993). The morphological studies suffered from lack of suitable characters to fully resolve trees, and from the difficulties inherent in homologising structures in anatomically disparate groups. More data on morphology, especially ultrastructure will undoubtedly prove helpful in this context, particularly in combination with nucleotide sequence data. 18S rDNA sequences are suitable for phylogeny reconstruction of high ranking taxa in a number of animal groups and for the whole Metazoa (e.g. Field et al., 1988; Kobayashi et al., 1993; Winnepenninckx et al., 1995; Halanych et al., 1995; Telford & Holland, 1993). In a study of platyhelminth phylogeny Katayama et al. (1993) sequenced 18S rDNA from ten representatives of the Acoela, Polycladida and Tricladida, used neighbour-joining to estimate phylogeny,
and concluded that acoels branched off earlier than other flatworms. Part of the 18S rDNA gene in 16 flatworm species was sequenced and analysed together with literature data by Rohde et al. (1993, 1995). Their resulting cladograms suggested a large taxon consisting of ‘the Proseriata and some other turbellarians’ as the sister group of the Neodermata and a separate high-ranking taxon was proposed for the Fecampiidae. Taxonomic sampling was, however, scattered with sequences from numerous high-ranking taxa missing. Several methods for evaluating the support for a cladistic hypothesis have been developed (e.g. Bremer, 1988; Faith & Cranston, 1991; Bremer, 1994, for a review). The bootstrap has been somewhat controversial, but a study using known phylogenies indicated that bootstrap proportions above 50% are conservative estimates of accuracy under conditions thought to be typical of phylogenetic analyses, and bootstrap proportions >70% represent true clades at least 95% of the time (Hillis & Bull, 1993). The jack-knife is a resampling technique less error prone than the bootstrap (Farris, 1996). In the present paper emphasis is put on testing support for cladistic hypotheses using jack-knife, bootstrap and decay analyses. New partial 18S rDNA sequences from 11 flatworm species are combined with 20 literature sequences in an attempt to improve on taxonomic sampling and provide a tentative phylogenetic analysis of the phylum.
148 Table 1. Flatworm species sequenced in this study and their collection localities Species
Higher
taxonLocality
Acrorhynchides robustus Anoplodium stichopi Graffilla bucinicola Maerenthalia agilis Meara stichopi Philactinoposthia saliens Plagiostomum vittatum Promesostoma agilis Provortex tubiferus Pterastericola astropectinis Urastoma cyprinae
Kalyptorhynchia Dalyellioida Dalyellioida Rhabdocoela Nemertodermatida Acoela Prolecithophora Typhloplanoida Dalyellioida Dalyellioida Rhabdocoela?
Tjärnö, Swedish west coast Tjärnö, Swedish west coast Kristineberg, Swedish west coast Tjärnö, Swedish west coast Bergen area, Norwegian West coast Kristineberg, Swedish west coast Kristineberg, Swedish west coast Kristineberg, Swedish west coast Kristineberg, Swedish west coast Tjärnö, Swedish west coast Tjärnö, Swedish west coast
Table 2. Taxa and sources for literature sequences used in this study Species
Higher taxon
Reference
Acoela sp. Anoplodiscus cirrusspiralis Arthioposthia sp. Bothromesostoma personatum Coelogynopora sp. Crenobia alpina Dendrocoelum lacteum Diclidophora merlangi Dugesia tigrina Gyratrix sp. Lineus sp. Macrostomum sp. Prorhynchus sp. Pterastericola australis Schistosoma mansoni Siboglinum fiordicum Stenostomum sp. Syndisyrinx punicea Temnocephala sp. Kronborgia isopodicola Suomina sp. Luriculus australiensis
Acoela Neodermata Tricladida Typhloplanoida Proseriata Tricladida Tricladida Neodermata Tricladida Kalyptorhynchia Nemertea (outgroup) Macrostomida Lecithoepitheliata Dalyellioida Neodermata Pogonophora (outgroup) Catenulida Dalyellioida Temnocephalida Fecampiidae Catenulida Dalyellioida
Rohde et al. (1994) Rohde et al. (1993) Rohde et al. (1993) Riutort et al. (1993) Rohde et al. (1994) Riutort et al. (1993) Riutort et al. (1993) Baverstock et al. (1991) Baverstock et al. (1991) Rohde et al. (1993) Winnepenninckx et al. (1995) Rohde et al. (1994) Rohde et al. (1993) Rohde et al. (1993) Ali et al. (1991) Winnepenninckx et al. (1995) Rohde et al. (1993) Rohde et al. (1993) Rohde et al. (1993) Rohde et al. (1994) Rohde et al. (1994) Rohde et al. (1994)
Material and methods
Extraction, PCR and cloning
Included taxa
Total DNA was extracted from ethanol preserved specimens through a conventional phenol-chloroform extraction. PCR was performed using the ExpandTM High Fidelity PCR-kit (Boehringer Mannheim) in a 100 µl reaction volume according to the following protocol: 0.8 mM dNTP-mix, 0.15 mM AP1 and A
Table 1 gives names, higher taxa and collection data for the species sequenced for this study. Literature sequences are listed (with higher taxa and references) in Table 2. The data matrix is in Appendix 1.
149 1985 primer, 1 × PCR buffer, 1.5 mM MgCl2 , 1 µl Expand High FidelityTM enzyme mix, 3 µl DNA, and one drop mineral oil was added. A Perkin-Elmer thermal cycler was used with the following cycling profile: 95 ◦ C for 1 min 30 s; 94 ◦ C/30 s; 45 ◦ C/1 min; 72 ◦ C/3 min ∗ 30 cycles; 72 ◦ C/8 min. The products were purified with the QIAquick Spin PCR Purification Kit (Qiagen) and stored at –20 ◦ C until sequencing. Due to problems with reamplification, PCR-products had to be cloned with a Blue-script vector according to the pCR-ScriptTM SK(+) cloning kit protocol (Stratagene). Plasmids were purified with the Qia-quick plasmid miniprep kit (Qiagen). Sequencing Denatured plasmid DNA was sequenced via the cyclic dideoxy chain termination method using the protocol of Amersham’s Thermo Sequenase Fluorescent Labelled Primer Cycle Sequencing Kit (Amersham) with 5–10 µl template on an automatic ALFexpressTM DNA Sequencer (Pharmacia-Biotech AB). The primers of Rohde et al. (1993) (WM1: 5-Cy5-ttcaagtgtctgacctatca-3 and D-comp: 5-Cy5agtccgagggagaggcc-3) marked with Cy5 were used for sequencing reactions. D-comp is the complement of the D primer. Nucleotide sequences data reported in this paper are available in the GenBank data base under the accession numbers AF100187–AF100197. Data analysis Sequences were aligned with ClustalW 1.5 (Thompson et al., 1993) with the default parameters, except that transitions were unweighted (the full alignment is available on the world wide web at: www.nrm.se/ev). The aligned sequences were analysed with PAUP 3.1.1 (Swofford, 1993) at the following heuristic search settings: random addition sequence/100 replicates, Tree Bisection and Reconnection (TBR). A bootstrap analysis was performed in PAUP using the heuristic search option, simple addition sequence and TBR with 1000 replicates. Decay index (also known as Bremer support) (Bremer, 1988) was calculated using the HyperCard stack AutoDecay (Eriksson, 1996) in conjunction with PAUP. Total support index (Bremer, 1994) was then calculated for the most parsimonious tree. The above analyses were carried out on an Apple Power Macintoshþ 9500 computer. A jack-knife analysis was carried out with the software JAC (Farris, 1995) on a 586 PC.
Results Most parsimonious tree The heuristic search resulted in a single most parsimonious tree of 2172 steps (Figure 1) consisting of two primary clades. Within the first clade, Promesostoma and Meara form the sister group of a clade consisting of the Tricladida (Artioposthia, Dendrocoelum, Crenobia, Dugesia) as sister group of the two Acoela (Philactinoposthia, Acoela sp.) and Kronborgia (Fecampiidae–Acoela clade). The second primary clade includes Macrostomum and Prorhynchus (Macrostomida–Lecithoepitheliata clade) as the sister group of the remaining taxa. Among the latter the Neodermata (Schistosoma, Diclidophora, Anoplodiscus) together with Gyratrix (Kalyptorhynchia) and Coelogynopora (Proseriata) constitute the sister group of a clade that consists of rhabdocoels and the prolecithophoran Plagiostomum (Rhabdocoela–Prolecithophora clade) as the sister group of a clade formed by Suomina and Stenostomum (Catenulida clade). Within the Rhabdocoela–Prolecithophora clade, the two pairs Urastoma + Maerenthalia and Acrorhynchides + Plagiostomum make up the sister group of the dalyellioids and Bothromesostoma (dalyellioid clade). The two Pterastericola species, are the sister group of Graffilla (Pterastericola–Graffilla clade). Temnocephala and Bothromesostoma (Temnocephalida– Typhloplanidae clade) form the sister group of a clade composed of the two Umagillidae Syndisyrinx and Anoplodium + Luriculus (Luridae) and Provortex (Provorticidae) (Umagillidae–Provorticidae clade). Decay index, bootstrap and jack-knife analyses 50% majority-rule consensus cladograms from the Bootstrap and jack-knife analyses are shown in Figures 2 and 3. Decay indices for the taxa in the single most parsimonious cladogram are shown in Figure 1. The following clades are present both in the bootstrap and jack-knife cladograms: Tricladida (decay index, d = 8), Fecampiidae–Acoela (d = 15), Temnocephalida–Typhloplanidae (d = 12), Macrostomida–Lecithoepitheliata (d = 6), Catenulida (d = 13), Umagillidae (d = 4), Neodermata (d = 8), Pterastericola–Graffilla (d = 12). Some clades are present in the Bootstrap consensus, but not in the jackknife tree: Fecampiidae–Acoela–Tricladida (d = 6), Umagillidae–Provorticidae (d = 7), dalyellioid (d = 8),
150
Figure 1. Most parsimonious tree resulting from heuristic search (random addition sequence, 100 replicates), 2172 steps. Decay index values (d) are shown at nodes.
Umagillidae (d = 4), Acrorhynchides–Plagiostomum (d = 5). Total support index (ti) for the single most parsimonious cladogram is 0.11. This is a low value compared with typical morphological studies, but is in the normal range for molecular studies (Bremer, 1994). Bootstrap, jack-knife and decay index results are summarised in Table 3.
Discussion Whereas the single most parsimonious tree is fully resolved, it is evident from the jack-knife and bootstrap analyses that the present data do not support a
fully dichotomous hypothesis of flatworm phylogeny. Species in the same genus (Pterastericola) or family (Umagillidae) group together with strong support, indicating that the fragment sequenced here is suitable for low-level phylogeny. Some suprafamilial clades are also retained in both the bootstrap and jackknife tests and attain high decay indices: the Tricladida and Fecampiidae–Acoela clades, the Neodermata clade, the Macrostomida–Lecithoepitheliata clade, the Catenulida clade, the Pterastericola–Graffilla clade, the Umagillidae and the Temnocephalida– Typhloplanidae clade. The Fecampiidae–Acoela clade is supported by the bootstrap and jack-knife analyses and has a high decay
151
Figure 2. 50% majority rule consensus tree summarising jack-knife analysis (1000 replicates). Jack-knife index on branches.
index. The branches in the clade are the longest in the cladogram (66–72 steps), but since the shortest branch in the cladogram is 11 steps (Gyratrix–Coelogynopora clade) the branch-length differences are not likely to be long enough to cause parsimony to reconstruct the wrong phylogeny due to ‘long branch attraction’ (Felsenstein, 1978; Hendy & Penny, 1989). Thus, the Fecampiidae–Acoela group is one of the best supported clades in the whole cladogram, and on the basis of evidence presented here there is no foundation for the erection of a new ‘class’ for the Fecampiidae as proposed by Rohde et al. (1994). Since the Nemertodermatid Meara did not appear within the group, it is not compatible with a sister group relationship between the Nemertodermatida and the Acoela (i.e. the taxon Acoelomorpha) or poly-
phyly of the Acoelomorpha and other Platyhelminthes as suggested by Smith et al. (1986) and Haszprunar (1996). Retaining the taxon Acoelomorpha would require at least the inclusion of the Fecampiidae, which do not possess an interconnecting ciliary rootlet system or the distally shelved cilia (Køie and Bresciani, 1973; Williams, 1990) proposed as apomorphies of the Acoelomorpha (Ehlers, 1984). The possible sister group relationship of the Tricladida and the Fecampiidae–Acoela clade (retained in the bootstrap consensus), would require inclusion also of the Tricladida in the Acoelomorpha. The results thus indicate monophyly of the Platyhelminthes, even though a critical test of the polyphyly hypothesis should involve a larger number of non-platyhelminth taxa, since Smith et al. (1986) did not formulate a cladistic hypothesis
152
Figure 3. 50% majority rule consensus tree summarising bootstrap analysis (1000 replicates).
with alternative sister groups for the Acoelomorpha, Catenulida and Rhabditophora. Haszprunar (1996)’s explicit hypothesis of the Catenulida as the sister group of remaining (non-platyhelminth) Bilateria is not compatible with the present data as the two outgroup taxa (Lineus and Siboglinum) appear outside the Platyhelminthes. Since this paper was submitted, Carranza et al. (1997) using distance, maximum likelihood and parsimony methods, published a study of platyhelminth phylogeny based on 13 complete 18S rDNA sequences. In their parsimony analysis a clade consisting of representatives of the Neoophora (Tricladida, Prolecithophora, Rhabdocoela, Lecithoepitheliata, Proseriata) and the Nemertodermatida is present after bootstrapping. Outside this clade representatives
of Polycladida and Macrostomida are members of a monophyletic Rhabditophora clade. The single representative of the Catenulida does not group with other flatworms, but as the sister taxon of the remaining Bilateria. Acoela were not included in the parsimony analysis. A direct comparison with the bootstrap tree in the present study is difficult, since the two data sets differ in taxonomic composition and deep branches are poorly supported in both analyses. Nonetheless, neither of the studies is consistent with the taxon Acoelomorpha. There is support for two of the subtaxa within the Dalyellioida: the Pterastericola–Graffilla and the Umagillidae. However, the Dalyellioida is not supported in the jack-knife test (but retained in the bootstrap consensus). The strong support for the
153 Table 3. Support for clades in the most parsimonious cladogram according to jack-knifing, bootstrapping and decay index Clade
Bootstrap Jack-knife Decay (%) (%) Index
Tricladida 100 Acoela 100 Pterastericola 100 Catenulida 99 Kronborgia–Acoela 98 Temnocephalida–Bothromesostoma 98 Graffilla–Pterastericola 97 Neodermata 86 Macrostomida–Lecithoepitheliata 86 Kronborgia–Acoela–Tricladida 72 Umagillidae–Provorticidae 56 dalyellioid 54 umagillid 54 Kalyptorhynchia–Prolecithophora 53
93 100 100 97 77 96 87 71 78 – – – 57 –
8 30 23 13 15 12 12 8 6 6 7 8 4 5
Bothromesostoma–Temnocephalida clade is consistent with the sister group relationship between the Typhloplanidae and the Temnocephalida proposed by Jondelius and Thollesson (1993) on morphological grounds. While the most parsimonious cladogram places the Kalyptorhynchia–Proseriata clade as sister group of the Neodermata, this clade collapsed in the Bootstrap and jack-knife analyses. The present data do not support a larger taxon composed of ‘proseriates and some other turbellarians’ as their sister group (Rohde et al., 1995). No conclusions can be drawn regarding the sister group of the Neodermata on the basis of the present data; previous hypotheses of various rhabdocoel subtaxa as sister groups (e.g. Ehlers, 1985, Jondelius and Thollesson, 1993) remain untested. Clearly the results should be regarded as highly tentative as they are based on a fragment of a single gene. Moreover, several higher flatworm taxa were not included in the analysis, e.g Polycladida and rhabdocoel subgroups. A robust hypothesis of flatworm phylogeny should be constructed from a combination of morphological data and nucleotide sequences from several genes.
Acknowledgements Thanks are extended to Drs Mari Källersjö and Gullevi Berqvist for help with sequencing, to Dr Steve Far-
ris for help with the jack-knife and to Kennet Lundin for the supply of Meara stichopi. Financial support was received from the Swedish Natural Science Research Council (NFR), Magn. Bergvalls stiftelse, and the Swedish Museum of Natural History.
References Ax, P., 1984. Das phylogenetische System. Systematisierung der lebenden Natur aufgrund ihrer Phylogenese. Gustav Fischer Verlag, Stuttgart, New York. Bremer, K., 1988. The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795–803. Bremer, K., 1994. Branch support and tree stability. Cladistics 10: 295–304. Brooks, D. R., 1989. The phylogeny of the Cercomeria (Platyhelminthes: Rhabdocoela) and general evolutionary principles. J. Parasitol. 75: 606–616. Carranza, S., J. Baguña & M. Riutort, 1997. Are the Platyhelminthes a monophyletic primitive group? An assessment using 18S rDNA sequences. Mol. Biol. Evol. 14: 485–497. Ehlers, U., 1984. Phylogenetisches System der Plathelminthes. Verhandlungen des naturwissenschaftlichen Vereins Hamburg. 27: 291–294. Ehlers, U., 1985. Das Phylogenetische System der Plathelminthes. Gustav Fischer, Stuttgart, New York. Ehlers, U., 1986. Comments on a phylogenetic system of the Platyhelminthes. Hydrobiologia 132: 1–12. Eriksson, T., 1996. AutoDecay HyperCard stack (computer software), available from the author, Department of Botany, Stockholm University. Faith, D. P. & P. S. Cranston, 1991. Could a cladogram this short have arisen by chance alone?: on permutation tests for cladistic structure. Cladistics 7: 1–28. Farris, J. S., 1995. JAC – parsimony jack-knifer (computer software). Swedish Museum of Natural History, Stockholm. Farris, J. S., V. A. Albert, M. Källersjö & D. Lipscomb, 1996. Parsimony jackknifing outperforms neighbour-joining. Cladistics 12: Felsenstein, J., 1978. Cases in which parsimony or compatibility methods will be positively misleading. Syst. Zool 27: 401–410. Field, K. G., G. J. Olsen, D. J. Lane, S. J. Giovannoni, M. T. Ghiselin, E. C. Raff, N. R. Pace & R. A. Raff, 1988. Molecular phylogeny of the animal kingdom. Science 239: 748–753. Halanych, K. M., J. D. Bacheller, A. M. A. Aguinaldo, S. M. Liva, D. M. Hillis & J. A. Lake, 1995. Evidence from 18S ribosomal DNA that the lophophorates are protostome animals. Science 267: 1641–1643. Haszprunar, G., 1996. Plathelminthes and Plathelminthomorpha paraphyletic taxa. J. Zool. Syst. Evol. Res. 34: 41–48. Hendy, M.D. & D. Penny, 1989. A framework for the quantitative study of evolutionary trees. Syst. Zool. 38: 297–309. Hillis, D. M. & J. J. Bull, 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst. Zool. 42: 182–192. Jondelius, U. & M. Thollesson, 1993. Phylogeny of the Rhabdocoela (Platyhelminthes): a working hypothesis. Can. J. Zool. 71: 298–308. Katayama, T., M. Yamamoto, H. Wada & N. Satoh, 1993. Phylogenetic Position of Acoel Turbellarians Inferred from Partial 18S rDNA Sequences. Zool. Sci. 10: 529–536.
154 Kobayashi, M., M. Takahashi, H. Wada & N. Satoh, 1993. Molecular phylogeny inferred from sequences of small subunit ribosomal DNA, supports the monophyly of the Metazoa. Zool. Sci. 10: 827–833. Køie, M. & J. Bresciani, 1973. On the ultrastructure of the larva of Kronborgia amphipodicola Christensen and Kanneworff 1964 (Turbellaria, Neorhabdocoela.) Ophelia 12: 171–203. Riutort, M., K. G. Field, R. A. Raff & J. Baguñà, 1993. 18S rRNA sequences and phylogeny of Platyhelminthes. Biochem. Syst. Ecol. 21: 71–77. Rohde, K., 1990. Phylogeny of Platyhelminthes, with special reference to parasitic groups. Int. J. Parasitol. 20: 979–1007. Rohde, K., A. M. Johnson, P. R. Baverstock & N. A. Watson, 1995. Aspects of the phylogeny of Platyhelminthes based on 18S ribosomal DNA and protonephridial ultrastructure. Hydrobiologia 305: 27–35. Rohde, K., K. Luton, P. R. Baverstock & A. M. Johnson, 1994. The phylogenetic relationships of Kronborgia (Platyhelminthes, Fecampiida) based on comparison of 18S ribosomal DNA sequences. Int. J. Parasitol. 24: 657–669. Rohde, K., C. Hefford, J. T. Ellis, P. R. Baverstock, A. M. Johnson, N. A. Watson & S. Dittmann, 1993. Contributions to the phy-
logeny of Platyhelminthes based on partial sequencing of 18S ribosomal DNA. Int. J. Parasitol. 23: 705–724. Smith, J. P. S., III, S. Tyler & R. M. Rieger, 1986. Is the Turbellaria polyphyletic? Hydrobiologia 132: 13–21. Telford, M. J. & P. W. H. Holland, 1993. The phylogenetic affinities of the chaetognaths: A molecular analysis. Mol. Biol. Evol. 10: 660–676. Thompson, J.D., D. G. Higgins & T. J. Gibson, 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673–4680. Williams, J. B., 1990. Ultrastructural studies on Kronborgia (Platyhelminthes: Fecampiidae): epidermis and subepidermal tissues of the parasitic male K. isopodicola. Int. J. Parasitol. 20: 329– 340. Winnepenninckx, B., T. Backeljau & R. De Wachter, 1995. Phylogeny of protostome worms derived from 18S rRNA sequences. Mol. Biol. Evol. 12: 641–649.
