We investigated evolutionary relationships among deuterostome subgroups by obtaining nearly complete large-subunit ribosomal RNA (LSU rRNA)-gene sequences for 14 deuterostomes and 3 protostomes and complete small-subunit (SSU) rRNA-gene sequences for five of these animals. With the addition of previously published sequences, we compared 28 taxa using three different data sets (LSU only, SSU only, and combined LSU + SSU) under minimum evolution (with LogDet distances), maximum likelihood, and maximum parsimony optimality criteria. Additionally, we analyzed the combined LSU + SSU sequences with spectral analysis of LogDet distances, a technique that measures the amount of support and conflict within the data for every possible grouping of taxa. Overall, we found that (1) the LSU genes produced a tree very similar to the SSU gene tree, (2) adding LSU to SSU sequences strengthened the bootstrap support for many groups above the SSU-only values (e.g., hemichordates plus echinoderms as Ambulacraria; lancelets as the sister group to vertebrates), (3) LSU sequences did not support SSU-based hypotheses of pterobranchs evolving from enteropneusts and thaliaceans evolving from ascidians, and (4) the combined LSU + SSU data are ambiguous about the monophyly of chordates. No tree-building algorithm united urochordates conclusively with other chordates, although spectral analysis did so, providing our only evidence for chordate monophyly. With spectral analysis, we also evaluated several major hypotheses of deuterostome phylogeny that were constructed from morphological, embryological, and paleontological evidence. Our rRNA-gene analysis refutes most of these hypotheses and thus advocates a rethinking of chordate and vertebrate origins.

No abstract

No abstract

Although the small-subunit ribosomal RNA (SSU rRNA) gene is widely used in the molecular systematics, few large-subunit (LSU) rRNA gene sequences are known from protostome animals, and the value of the LSU gene for invertebrate systematics has not been explored. The goal of this study is to test whether combined LSU and SSU rRNA gene sequences support the division of protostomes into Ecdysozoa (molting forms) and Lophotrochozoa, as was proposed by Aguinaldo et al. (1997) (Nature 387:489) based on SSU rRNA sequences alone. Nearly complete LSU gene sequences were obtained, and combined LSU ϩ SSU sequences were assembled, for 15 distantly related protostome taxa plus five deuterostome outgroups. When the aligned LSU ϩ SSU sequences were analyzed by tree-building methods (minimum evolution analysis of LogDet-transformed distances, maximum likelihood, and maximum parsimony) and by spectral analysis of LogDet distances, both Ecdysozoa and Lophotrochozoa were indeed strongly supported (e.g., bootstrap values Ͼ90%), with higher support than from the SSU sequences alone. Furthermore, with the LogDet-based methods, the LSU ϩ SSU sequences resolved some accepted subgroups within Ecdysozoa and Lophotrochozoa (e.g., the polychaete sequence grouped with the echiuran, and the annelid sequences grouped with the mollusc and lophophorates)-subgroups that SSU-based studies do not reveal. Also, the mollusc sequence grouped with the sequences from lophophorates (brachiopod and phoronid). Like SSU sequences, our LSU ϩ SSU sequences contradict older hypotheses that grouped annelids with arthropods as Articulata, that said flatworms and nematodes were basal bilateralians, and considered lophophorates, nemerteans, and chaetognaths to be deuterostomes. The position of chaetognaths within protostomes remains uncertain: our chaetognath sequence associated with that of an onychophoran, but this was unstable and probably artifactual. Finally, the benefits of combining LSU with SSU sequences for phylogenetic analyses are discussed: LSU adds signal, it can be used at lower taxonomic levels, and its core region is easy to align across distant taxa-but its base frequencies tend to be nonstationary across such taxa. We conclude that molecular systematists should use combined LSU ϩ SSU rRNA genes rather than SSU alone.

BackgroundMembers of Family Nereididae have complex neural morphology exemplary of errant polychaetes and are leading research models in the investigation of annelid nervous systems. However, few studies focus on the development of their nervous system morphology. Such data are particularly relevant today, as nereidids are the subjects of a growing body of "evo-devo" work concerning bilaterian nervous systems, and detailed knowledge of their developing neuroanatomy facilitates the interpretation of gene expression analyses. In addition, new data are needed to resolve discrepancies between classic studies of nereidid neuroanatomy. We present a neuroanatomical overview based on acetylated α-tubulin labeling and confocal microscopy for post-embryonic stages of Neanthes arenaceodentata, a direct-developing nereidid.ResultsAt hatching (2-3 chaetigers), the nervous system has developed much of the complexity of the adult (large brain, circumesophageal connectives, nerve cords, segmental nerves), and the stomatogastric nervous system is partially formed. By the 5-chaetiger stage, the cephalic appendages and anal cirri are well innervated and have clear connections to the central nervous system. Within one week of hatching (9-chaetigers), cephalic sensory structures (e.g., nuchal organs, Langdon's organs) and brain substructures (e.g., corpora pedunculata, stomatogastric ganglia) are clearly differentiated. Additionally, the segmental-nerve architecture (including interconnections) matches descriptions of other, adult nereidids, and the pharynx has developed longitudinal nerves, nerve rings, and ganglia. All central roots of the stomatogastric nervous system are distinguishable in 12-chaetiger juveniles. Evidence was also found for two previously undescribed peripheral nerve interconnections and aspects of parapodial muscle innervation.ConclusionsN. arenaceodentata has apparently lost all essential trochophore characteristics typical of nereidids. Relative to the polychaete Capitella, brain separation from a distinct epidermis occurs later in N. arenaceodentata, indicating different mechanisms of prostomial development. Our observations of parapodial innervation and the absence of lateral nerves in N. arenaceodentata are similar to a 19th century study of Alitta virens (formerly Nereis/Neanthes virens) but contrast with a more recent study that describes a single parapodial nerve pattern and lateral nerve presence in A. virens and two other genera. The latter study apparently does not account for among-nereidid variation in these major neural features.

No abstract

SUMMARYWe examine terminal addition, the process of addition of serial elements in a posterior subterminal growth zone during animal development, across modern taxa and fossil material. We argue that terminal addition was the basal condition in Bilateria, and that modification of terminal addition was an important component of the rapid Cambrian evolution of novel bilaterian morphology. We categorize the often-convergent modifications of terminal addition from the presumed ancestral condition. Our focus on terminal addition and its modification highlights trends in the history of animal evolution evident in the fossil record. These trends appear to be the product of departure from the initial terminal addition state, as is evident in evolutionary patterns within-fossil groups such as trilobites, but is also more generally related to shifts in types of morphologic change through the early Phanerozoic. Our argument is contingent on dates of metazoan divergence that are roughly convergent with the first appearance of metazoan fossils in the latest Proterozoic and Cambrian, as well as on an inference of homology of terminal addition across bilaterian Metazoa.

BackgroundA number of spider species within the family Theridiidae exhibit a dramatic abdominal (opisthosomal) color polymorphism. The polymorphism is inherited in a broadly Mendelian fashion and in some species consists of dozens of discrete morphs that are convergent across taxa and populations. Few genomic resources exist for spiders. Here, as a first necessary step towards identifying the genetic basis for this trait we present the near complete transcriptomes of two species: the Hawaiian happy-face spider Theridion grallator and Theridion californicum. We mined the gene complement for pigment-pathway genes and examined differential expression (DE) between morphs that are unpatterned (plain yellow) and patterned (yellow with superimposed patches of red, white or very dark brown).ResultsBy deep sequencing both RNA-seq and normalized cDNA libraries from pooled specimens of each species we were able to assemble a comprehensive gene set for both species that we estimate to be 98-99% complete. It is likely that these species express more than 20,000 protein-coding genes, perhaps 4.5% (ca. 870) of which might be unique to spiders. Mining for pigment-associated Drosophila melanogaster genes indicated the presence of all ommochrome pathway genes and most pteridine pathway genes and DE analyses further indicate a possible role for the pteridine pathway in theridiid color patterning.ConclusionsBased upon our estimates, T. grallator and T. californicum express a large inventory of protein-coding genes. Our comprehensive assembly illustrates the continuing value of sequencing normalized cDNA libraries in addition to RNA-seq in order to generate a reference transcriptome for non-model species. The identification of pteridine-related genes and their possible involvement in color patterning is a novel finding in spiders and one that suggests a biochemical link between guanine deposits and the pigments exhibited by these species.