Molecular evidence that phoronids are a subtaxon of brachiopods (Brachiopoda : Phoronata) and that genetic divergence of metazoan phyla began long before the early Cambrian

摘要

Concatenated SSU (18S) and partial LSU (28S) sequences (similar to 2 kb) from 12 ingroup taxa, comprising 2 phoronids, 2 members of each of the craniid, discinid, and lingulid inarticulate brachiopod lineages, and 4 rhynchonellate, articulate brachiopods (2 rhynchonellides, 1 terebratulide and 1 terebratellide) were aligned with homologous sequences from 6 protostome, deuterostome and sponge outgroups (3964 sites). Regions of potentially ambiguous alignment were removed, and the resulting data (3275 sites, of which 377 were parsimony-informative and 635 variable) were analysed by parsimony, and by maximum and Bayesian likelihood using objectively selected models. There was no base composition heterogeneity. Relative rate tests led to the exclusion (from most analyses) of the more distant outgroups, with retention of the closer pectinid and polyplacophoran (chiton). Parsimony and likelihood bootstrap and Bayesian clade support values were generally high, but only likelihood analyses recovered all brachiopod indicator clades designated a priori. All analyses confirmed the monophyly of (brachiopods + phoronids) and identified phoronids as the sister-group of the three inarticulate brachiopod lineages. Consequently, a revised Linnean classification is proposed in which the subphylum Linguliformea comprises three classes: Lingulata, 'Phoronata' (the phoronids), and 'Craniata' (the current subphylum Craniiformea). Divergence times of all nodes were estimated by regression from node depths in non-parametrically rate-smoothed and other chronograms, calibrated against palaeontological data, with probable errors not less than 50 My. Only three predicted brachiopod divergence times disagree with palaeontological ages by more than the probable error, and a reasonable explanation exists for at least two. Pruning long-branched ingroups made scant difference to predicted divergence time estimates. The palaeontological age calibration and the existence of Lower Cambrian fossils of both main brachiopod clades together indicate that initial genetic divergence between brachiopod and molluscan (chiton) lineages occurred well before the Lower Cambrian, suggesting that much divergence between metazoan phyla took place in the Proterozoic. (c) 2005 Gesellschaft fur Biologische Systematik. Published by Elsevier GmbH. All rights reserved.