Molecular Phylogenetics and Evolution 66 (2013) 551–557 Contents lists available at SciVerse ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev Review Chasing the urmetazoon: Striking a blow for quality data? ⇑ Hans-Jürgen Osigus a, Michael Eitel a,d, Bernd Schierwater a,b,c, a ITZ, Division of Ecology and Evolution, Stiftung Tieraerztliche Hochschule Hannover, Germany b American Museum of Natural History, New York, USA c Department of Ecology and Evolutionary Biology, Yale University, USA d The Swire Institute of Marine Science, Faculty of Science, School of Biological Sciences, The University of Hong Kong, Hong Kong article info abstract Article history: The ever-lingering question: ‘‘What did the urmetazoan look like?’’ has not lost its charm, appeal or elu- Available online 6 June 2012 siveness for one and a half centuries. A solid amount of organismal data give what some feel is a clear answer (e.g. Placozoa are at the base of the metazoan tree of life (ToL)), but a diversity of modern molec- Keywords: ular data gives almost as many answers as there are exemplars, and even the largest molecular data sets Urmetazoon could not solve the question and sometimes even suggest obvious zoological nonsense. Since the prob- Trichoplax lems involved in this phylogenetic conundrum encompass a wide array of analytical freedom and uncer- Placozoa tainty it seems questionable whether a further increase in molecular data (quantity) can solve this Non-bilaterian animals classical deep phylogeny problem. This review thus strikes a blow for evaluating quality data (including Placula hypothesis Metazoan evolution morphological, molecule morphologies, gene arrangement, and gene loss versus gene gain data) in an appropriate manner. Ó 2012 Elsevier Inc. All rights reserved. Contents 1. Traditional morphological views . ................................................................................... 551 2. Modern molecular views . ................................................................................... 552 2.1. Analyzing gene sequences (so-called ‘‘quantity data’’) . ...................................................... 552 2.2. Analyzing gene presence and structure (so-called ‘‘quality data’’). ...................................................... 553 2.2.1. Hox-/ParaHox genes . .............................................................................. 553 2.2.2. Pax genes .............................................................................................. 554 2.2.3. Dicer genes . .............................................................................. 554 2.2.4. Leucine-rich repeat containing G protein-coupled receptors (LGRs). ........................................ 554 2.2.5. Ribosomal genes . .............................................................................. 554 2.2.6. Mitochondrial genome characteristics . ........................................................... 554 3. Summary quantity versus quality data. ................................................................................... 555 Acknowledgments . ................................................................................... 555 References . ...................................................................................................... 556 1. Traditional morphological views ple morphology, perhaps just complex enough to pass the bridge between a protist and a metazoan. This bridge is the possession Three prominent urmetazoan hypotheses are most often under of more than one somatic cell type (protists may consist of doz- debate, (i) the placula, (ii) the planula and (iii) the gastrea ens or hundreds of cells and may have more than one cell type hypothesis (Bütschli, 1884; Haeckel, 1874; Lankester, 1877, for within a colony but they never have two or more different so- overview see Kaestner, 1980). Common to all hypotheses is that matic cells). Intrasomatic differentiation, which is the invention the hypothetical ‘‘urmetazoon’’ must have had an extremely sim- of the urmetazoon and a synapomorphy for all metazoans, became the motor for radiation of the metazoan bauplans (cf. Boero et al., 2007; Schierwater, B., de Jong, D., Desalle, R., 2009). ⇑ Corresponding author at: Division of Ecology and Evolution, Stiftung Tieraerz- Unfortunately for phylogenetics at the base of the tree, however, tliche Hochschule Hannover, Bünteweg 17d, D-30559 Hannover, Germany. Fax: +49 morphology was ‘‘frozen’’ as very subtle and uninterpretable 511 953 8485. E-mail address: [email protected] (B. Schierwater). anatomical changes occurred, and hence we are left with very 1055-7903/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ympev.2012.05.028 552 H.-J. Osigus et al. / Molecular Phylogenetics and Evolution 66 (2013) 551–557 few anatomical characters. In sharp contrast, the more derived choanocytes, and not vice versa (Clark, 1868; Kent, 1878; Maldo- root of the tree invented a third germ layer, the mesoderm, which nado, 2004). In sum, there are currently two reasonable candidates fueled an explosion of bauplan radiation and new anatomical for the closest living relative of the urmetazoon, Placozoa and characters in the Bilateria. Porifera. Placozoans meet the expectations of a primitive but general urmetazoan bauplan (see Fig. 1); they possess the simplest bau- plan among extant metazoans (Grell and Benwitz, 1971; Grell 2. Modern molecular views and Ruthmann, 1991; Schierwater, 2005; Schulze, 1883, 1891). Only five somatic cell types (Guidi et al., 2011; Jakob et al., 2.1. Analyzing gene sequences (so-called ‘‘quantity data’’) 2004) have been recognized. The simplicity of placozoans is fur- ther highlighted by their lack of any kind of axis of symmetry, or- Early molecular systematic studies used single gene data sets gans, nerve and muscle cells, basal lamina, and an ultrastructurally (mainly 18S or 28S rRNA) to resolve conflicts at the base of the identifiable extracellular matrix (note however, several cell–cell metazoan tree of life. The resulting mixture of trees mostly sup- contact genes have been identified in Trichoplax; Chapman et al., ports the traditional view of early branching Porifera (for overview 2010). In contrast to placozoans, Porifera usually possess more see Schierwater et al., 2010a), sometimes with the phylum Placo- than a dozen somatic cell types (see Jiang and Xu, 2010; Valentine zoa jumping inside a class of Cnidaria (Bridge et al., 1995; Siddall et al., 1994, for an overview of cell types in animals) and normally et al., 1995). Analysis of mitochondrial protein coding sequence also develop an extracellular matrix (ECM) and basal membrane data have promised to lead to a more reasonable picture of early (e.g. Boury-Esnault et al., 2003) or even complete sealing epithelia animal evolution supporting a split between Bilateria and (Adams et al., 2010). In Cnidaria and Ctenophora morphological non-bilaterian animals with Placozoa branching first within the complexity has increased. Thus many zoologists have been consid- non-Bilateria (Dellaporta et al., 2006; Signorovitch et al., 2007). ering Placozoa as the closest living relative to the urmetazoon (e.g. Nevertheless, analyses of recently sequenced mitochondrial gen- Grell, 1971, 1982; Schulze, 1891). A larger group of researchers has omes from Ctenophora (Kohn et al., 2012; Pett et al., 2011) seen the sponges, Porifera, closest to the base of the metazoan tree highlight limitations of analyses based on mitochondrial protein of life. The main argument here has been a single-character argu- sequence data. One of several problems relates to the observation ment, the possession of the eye-catching character choanocytes, that mitochondrial sequences in Ctenophora have evolved several i.e. the morphological similarity between the choanocytes in the times faster than in other animal phyla. sponge gastrodermis and the single-celled choanoflagellates. But The rapid progress in next generation sequencing techniques there also are other explanations for this observation than a and computational data processing opens the door for phyloge- non-simple evolutionary transformation of a choanoflagellate col- netic analyses using large whole genome and EST data sets includ- ony into a sponge bauplan. Some authors are in favor of a conver- ing several hundred or thousands of genes. As a consequence gent evolution of collar structures and metazoan choanocytes phylogenomic approaches to elucidate adaptive evolution in genes (Maldonado, 2004) or even claim that the choanoflagellates are de- and genomes will become an important field in future research (cf. rived sponges. For example, several genes in the choanoflagellate Goodman and Sterner, 2010; Shinzato et al., 2011). At present we genome (including genes for metazoan style cell adhesion and cell are mostly limited to non-causal analyses of descriptive characters, signaling; King et al., 2003; Manning et al., 2008) might be seen like gene sequences. At the base of the metazoan ToL, the outcome as indicators that evolution here went the opposite of the of such analyses has been highly contradictory and can be summa- obvious way, i.e. that choanoflagellates originated from sponge rized in three main scenarios: Fig. 1. (A) Photograph of Trichoplax adhaerens, Schulze (1883). For additional images of placozoan specimens see www.trichoplax.com. (B) Modern placula hypothesis of metazoan origin (for details see Schierwater et al., 2009a). (from