Molecular Phylogenetics and Evolution 57 (2010) 518–527 Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev Evolutionary relationships among marine cercozoans as inferred from combined SSU and LSU rDNA sequences and polyubiquitin insertions Chitchai Chantangsi a,b,*, Mona Hoppenrath a,1, Brian S. Leander a a Canadian Institute for Advanced Research, Program in Integrated Microbial Biodiversity, Departments of Zoology and Botany, University of British Columbia, Vancouver, BC, Canada V6T 1Z4 b Department of Biology, Faculty of Science, Chulalongkorn University, Phayathai Road, Pathumwan, Bangkok 10330, Thailand article info abstract Article history: An insertion of one or two amino acids at the monomer–monomer junctions of polyubiquitin is a distinct Received 23 October 2009 and highly conserved molecular character that is shared by two very diverse clades of microeukaryotes, Revised 15 July 2010 the Cercozoa and the Foraminifera. It has been suggested that an insertion consisting of one amino acid, Accepted 15 July 2010 like that found in foraminiferans and some cercozoans, represents an ancestral state, and an insertion Available online 21 July 2010 consisting of two amino acids represents a derived state. However, the limited number of cercozoan taxa examined so far limits inferences about the number and frequency of state changes associated with this Keywords: character over deep evolutionary time. Cercozoa include a very diverse assemblage of mainly unculti- Cercozoa vated amoeboflagellates, and their tenuous phylogenetic interrelationships have been based largely on Ebria LSU rDNA small subunit (SSU) rDNA sequences. Because concatenated datasets consisting of both SSU and large Molecular phylogeny subunit (LSU) rDNA sequences have been shown to more robustly recover the phylogenetic relationships Polyubiquitin of other major groups of eukaryotes, we employed a similar approach for the Cercozoa. In order to recon- Protaspis struct the evolutionary history of this group, we amplified twelve LSU rDNAs, three SSU rDNAs, and seven polyubiquitin sequences from several different cercozoans, especially uncultured taxa isolated from mar- ine benthic habitats. The distribution of single amino acid insertions and double amino acid insertions on the phylogenetic trees inferred from the concatenated dataset indicates that the gain and loss of amino acid residues between polyubiquitin monomers occurred several times independently. Nonetheless, all of the cercozoans we examined possessed at least one amino acid insertion between the polyubiquitin monomers, which reinforced the significance of this feature as a molecular signature for identifying members of the Cercozoa and the Foraminifera. Our study also showed that analyses combining both SSU and LSU rDNA sequences leads to improved phylogenetic resolution and statistical support for dee- per branches within the Cercozoa. Ó 2010 Elsevier Inc. All rights reserved. 1. Introduction radiolarians), cercozoans are strongly nested within the supergroup Rhizaria (Adl et al., 2005; Burki et al., 2007, 2008; Keeling et al., 2005; Cercozoans comprise a diverse clade of amoeboflagellates that Nikolaev et al., 2004; Polet et al., 2004). Cercozoans, along with was initially recognized and established using molecular phyloge- euglenids and dinoflagellates, are among the most commonly netic analyses of SSU rDNA sequences (Cavalier-Smith, 1998a,b). encountered predatory flagellates in marine benthic environments, An obvious morphological feature that unites the group has yet to and environmental surveys of this diversity have demonstrated a be identified. Nonetheless, phylogenetic analyses of concatenated huge number of cercozoan lineages that have yet to be adequately gene datasets have subsequently demonstrated that along with characterized (Bass and Cavalier-Smith, 2004; Chantangsi et al., the Foraminifera and the Radiozoa (i.e., most of the traditional 2008; Hoppenrath and Leander, 2006a,b). Marine benthic cercozo- ans, especially members of the Cryomonadida (e.g., Cryothecomonas and Protaspis) and their benthic and planktonic relatives (e.g., * Corresponding author at: Department of Biology, Faculty of Science, Chul- Botuliforma, Clautriavia, Ebria, Thaumatomastix, Ventrifissura, and alongkorn University, Phayathai Road, Pathumwan, Bangkok 10330, Thailand. Fax: Verrucomonas), have so far been relatively neglected by the protisto- +66 2 218 5386. logical community (Chantangsi et al., 2008). Improved knowledge of E-mail addresses: [email protected], [email protected] (C. Chantang- these particular lineages is important for gaining a more compre- si), [email protected] (M. Hoppenrath), [email protected] hensive understanding of marine benthic ecosystems and shedding (B.S. Leander). 1 Present address: Forschungsinstitut Senckenberg, Deutsches Zentrum für Marine light on our view of the evolutionary history of marine benthic Biodiversitätsforschung (DZMB), Südstrand 44, D-26382 Wilhelmshaven, Germany. cercozoans. 1055-7903/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2010.07.007 C. Chantangsi et al. / Molecular Phylogenetics and Evolution 57 (2010) 518–527 519 The Cercozoa and the Foraminifera share a significant molecular 2. Materials and methods character involving the presence of novel amino acid insertion(s) at the junctions between monomers of polyubiquitin genes (Archi- 2.1. Source of samples and light microscopy (LM) bald et al., 2003). Ubiquitin is a small regulatory protein composed of 76 amino acids and functions to mark other proteins for destruc- Fourteen cercozoan taxa were examined in this study (Table 1 tion (Archibald et al., 2003; Bass et al., 2005). This protein is found and Fig. 1); only three of these have been cultivated: Cryotheco- only in eukaryotes and plays essential roles in many biological pro- monas sp. (strain APCC MC5-1Cryo), Gymnochlora stellata (strain cesses, such as cell cycle regulation, DNA repair, transcriptional CCMP 2057), and Lotharella vacuolata (strain CCMP 240). Placocista regulation, signal transduction, endocytosis, embryogenesis, and sp. was isolated from freshwater aquatic moss, and individual cells apoptosis (Hershko and Ciechanover, 1998). Ubiquitin genes are representing the remaining ten taxa were isolated from marine also highly conserved, as indicated by only three residue differ- sand samples collected near Vancouver, British Columbia (details ences between humans and yeast, and can be configured in three in Table 1). Cercozoan cells were extracted from the sand samples main ways: (1) individual genes with single open reading frames; through a 48 lm mesh using a melted seawater–ice method de- (2) genes fused to ribosomal protein genes; and (3) genes orga- scribed by Uhlig (1964). Briefly, 2–3 spoons of sand samples were nized as linear head-to-tail ubiquitin coding region repeats, called placed into an extraction column wrapped with the mesh, and two polyubiquitin genes (Bass et al., 2005). Although the amino acid to three seawater ice cubes were then put on top of the sand sam- sequences of polyubiquitin genes cannot be used directly to con- ples and left to melt over several hours. The cells of interest passed struct phylogenetic trees, the specific amino acids inserted be- through the mesh and were concentrated in a seawater-filled Petri tween the monomers (e.g., serine or theonine) might provide dish that was placed underneath the extraction column. The Petri molecular signatures for specific subclades within the Cercozoa dish containing the cells was then observed using a Leica DMIL in- and the Foraminifera (Archibald et al., 2003; Archibald and Keeling, verted microscope. Cells were individually isolated and placed on a 2004; Bass et al., 2005). slide for imaging and identification using phase contrast and differ- Phylogenetic analyses of multi-gene datasets have proven to be ential interference contrast (DIC) microscopy with a Zeiss Axioplan a powerful approach to better resolving relationships among 2 imaging microscope (Fig. 1). eukaryotes (e.g., Baldauf et al., 2000; Bapteste et al., 2002; Burki et al., 2007, 2008; Burki and Pawlowski, 2006; Harper et al., 2.2. DNA extraction and PCR amplification 2005; Kim et al., 2006; Rokas et al., 2003). However, large-scale data (i.e., hundreds of genes or whole genomes) are only available Cells were individually isolated and washed three times in for a very limited number of eukaryotes, and phylogenetic analyses either autoclaved distilled water or autoclaved filtered seawater, of these datasets are computationally challenging, time consuming depending on the species. DNA was extracted using the protocol (Moreira et al., 2007), and currently impractical for studying uncul- provided in the Total Nucleic Acid Purification kit by EPICENTRE tivated lineages. Phylogenetic analyses of combined SSU and LSU (Madison, WI, USA). Semi-nested polymerase chain reaction rDNA sequences, on the other hand, have been shown to be a prag- (PCR) with a final reaction volume of 25 ll was performed in a matic and effective way to significantly improve the resolution of thermal cycler using puReTaq Ready-To-Go PCR beads (GE Health- distant relationships within the tree of eukaryotes (e.g., Moreira care Bio-Sciences, Inc., Québec, Canada). The first PCR amplification et al., 2007). These molecular markers are, therefore, expected to was conducted using the outermost forward and reverse primers provide