Molecular Phylogeny of Unikonts: New Insights into the Position of Apusomonads and Ancyromonads and...

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Protist, Vol. 164, 2–12, January 2013http://www.elsevier.de/protisPublished online date 18 October 2012

ORIGINAL PAPER

Molecular Phylogeny of Unikonts: New Insightsinto the Position of Apusomonads andAncyromonads and the InternalRelationships of Opisthokonts

Jordi Papsa,1,2, Luis A. Medina-Chacóna, Wyth Marshallb,Hiroshi Sugaa,c, and Inaki Ruiz-Trilloa,c,d,2

aDepartament de Genètica, Universitat de Barcelona, Av. Diagonal, 645, 08028 Barcelona, SpainbB.C. Centre for Aquatic Health, 871A Island Hwy, Campbell River, B.C., V9W 5B1, CanadacInstitut de Biologia Evolutiva (UPF-CSIC), Passeig Martim de la Barceloneta 37-49, 08003 Barcelona, SpaindInstitució Catalana per a la Recerca i Estudis Avancats (ICREA), Spain

Submitted June 2, 2011; Accepted September 10, 2012Monitoring Editor: Sandra L. Baldauf

The eukaryotic supergroup Opisthokonta includes animals (Metazoa), fungi, and choanoflagellates,as well as the lesser known unicellular lineages Nucleariidae, Fonticula alba, Ichthyosporea, Filastereaand Corallochytrium limacisporum. Whereas the evolutionary positions of the well-known opisthokontsare mostly resolved, the phylogenetic relationships among the more obscure lineages are not. Withinthe Unikonta (Opisthokonta and Amoebozoa), it has not been determined whether the Apusozoa(apusomonads and ancyromonads) or the Amoebozoa form the sister group to opisthokonts, nor towhich side of the hypothesized unikont/bikont divide the Apusozoa belong. Aiming at elucidating theevolutionary tree of the unikonts, we have assembled a dataset with a large sampling of both organismsand genes, including representatives from all known opisthokont lineages. In addition, we includenew molecular data from an additional ichthyosporean (Creolimax fragrantissima) and choanoflag-ellate (Codosiga botrytis). Our analyses show the Apusozoa as a paraphyletic assemblage withinthe unikonts, with the Apusomonadida forming a sister group to the opisthokonts. Within the Holozoa,the Ichthyosporea diverge first, followed by C. limacisporum, the Filasterea, the Choanoflagellata,and the Metazoa. With our data-enriched tree, it is possible to pinpoint the origin and evolution ofmorphological characters. As an example, we discuss the evolution of the unikont kinetid.© 2012 Elsevier GmbH. All rights reserved.

Key words: Cilia; Corallochytrea; eukaryotic phylogeny; Filasterea; opisthokonts; Unikonta.

1Current address: Department of Zoology, University of Oxford,Tinbergen Building, South Parks Road, Oxford OX1 3PS,United Kingdom2Corresponding authors; fax +34 93 221 10 11e-mail [email protected] (J. Paps),[email protected], [email protected] (I. Ruiz-Trillo).

Introduction

Our understanding of organismal evolution hasimproved significantly in recent decades, thankslargely to the contributions of improved moleculartechniques and new microscopy data. Molecu-lar phylogenies have consistently improved the

© 2012 Elsevier GmbH. All rights reserved.http://dx.doi.org/10.1016/j.protis.2012.09.002


Multigene Analysis of the Unikonts. 3

tree of eukaryotes, now divided into five or sixsupergroups (Adl et al. 2005; Keeling et al. 2005;Roger and Simpson 2009; Simpson and Roger2004). However, the root of the eukaryotes and therelationships among these eukaryotic supergroupsremain uncertain. One of the most widespreadhypotheses roots the eukaryote tree between theUnikonta, unicellular organisms whose ancestralmode of locomotion appears to have been basedon a single cilium and basal body, and the Bikonta(Cavalier-Smith 2002). Other authors have pro-posed alternative hypotheses, placing the root ofthe eukaryotes in the Excavata (Cavalier-Smith2010) or the Plantae (Rogozin et al. 2009); how-ever, these hypotheses do not generally contradicta monophyletic union of the unikonts (except seeKatz et al. 2012).

The unikonts, comprised of the Amoebozoaand Opisthokonta, have a striking diversity offorms. For example, several types of multicellularityhave independently emerged, including the distinc-tive and well known metazoan and fungal bodyplans, the colonial stages of choanoflagellates andichthyosporeans, and the aggregative fruiting bod-ies of Fonticula alba and dictyostelid slime molds(Paps and Ruiz-Trillo 2010 and references within).A sister group relationship between the Amoebo-zoa and the Opisthokonta has been supportedby several molecular phylogenies (Burki et al.2007; Rodriguez-Ezpeleta et al. 2007; Ruiz-Trilloet al. 2006, 2008) and molecular synapomor-phies (Richards and Cavalier-Smith 2005; Rogozinet al. 2009; Stechmann and Cavalier-Smith 2002).However, the Apusozoa, which consists of apu-somonads and ancyromonads, has a contentiousrelationship with the unikonts. In some studiesthe Apusomonadida appear to belong within theunikonts and to be related to the opisthokonts(Cavalier-Smith and Chao 1995; Katz et al. 2011;Kim et al. 2006; Torruella et al. 2012). Otherdata, however, suggest a closer relationship to thebikonts due to their bi-flagellated form and the pres-ence of a bikont-specific molecular gene fusion(Stechmann and Cavalier-Smith 2002, 2003).

The monophyletic grouping of Opisthokonta iswell supported by both molecular trees (Baldaufet al. 2000; Lang et al. 2002; Medina et al.2003; Ruiz-Trillo et al. 2004, 2008; Steenkamp andBaldauf 2004; Steenkamp et al. 2006; Torruellaet al. 2012) and molecular synapomorphies, suchas a 9-17 amino acid insertion in elongation 1 alpha(EF1-alpha, Baldauf and Palmer 1993; Steenkampand Baldauf 2004) and a haloarchaeal-type tyro-syl tRNA synthetase (Huang et al. 2005, but seeShadwick and Ruiz-Trillo 2012). These molecular

analyses also tend to divide the opisthokonts intotwo clades: the Holozoa (Lang et al. 2002), whichincludes the Metazoa and their unicellular relatives,and the Holomycota (Liu et al. 2009, also namedNucletmycea (Brown et al. 2009), which containsthe fungi, nucleariids, and F. alba. Within the Holo-zoa, the sister-group relationship of Metazoa andchoanoflagellates is also well supported (Carr et al.2008; Lang et al. 2002; Medina et al. 2001; Ruiz-Trillo et al. 2006, 2008; Torruella et al. 2012).

Within the unikonts the relationships betweenseveral lineages and placement of certainenigmatic taxa remain uncertain. For exam-ple, within the Holozoa, the position of C.limacisporum, relative to the Filasterea (Min-isteria vibrans + Capsaspora owczarzaki) andIchthyosporea is still contentious (reviewed byPaps and Ruiz-Trillo 2010), as are the specificrelationships between the Holomycota lineagesF. alba, Nucleariidae and Fungi. The affiliation ofBreviata anathema (Walker et al. 2006), previouslyknown as Mastigamoeba invertens (Minge et al.2008), with the Amoebozoa is also uncertain.Some studies with high numbers of genes (Mingeet al. 2008; Ruiz-Trillo et al. 2008; Shalchian-Tabriziet al. 2008; Torruella et al. 2012) and others withfewer genes but broader taxonomic sampling(Brown et al. 2009; Ruiz-Trillo et al. 2004, 2006;Steenkamp et al. 2006) have been able to hintat the positions of some of these organisms.However, no single molecular study so far hasincluded multiple representatives from each of theknown opisthokont lineages.

In this study we have built a dataset that bal-ances both the sampling of taxa and markers withthe aim of solving the position of the lesser-knowntaxa and clades in the unikont tree. Our alignmentcontains representatives of all the main opisthokontclades, several representatives of the Amoebo-zoa and the Apusozoa, and a taxonomically broadselection of bikonts. We have collected data foreight molecular markers, the two ribosomal sub-unit rDNA genes (18S and 28S) and six proteincoding genes: actin, elongation factor 1 alpha (EF1alpha), alpha and beta tubulins, and the heat shockproteins 70 (hsp 70) and 90 (hsp 90). We havealso produced new sequences by a PCR-basedsurvey and by mining genome projects and ESTs(expressed sequence tag) collections. Specifically,we have amplified and sequenced five molecularmarkers from the ichthyosporean C. fragrantis-sima (Marshall et al. 2008) and the freshwaterchoanoflagellate C. botrytis. The final alignmentcontains 73 taxa and was analyzed using proba-bilistic methods.


4 J. Paps et al.

Results

Datasets

The main molecular dataset contains 492 dif-ferent sequences, and has 84% occupancy (i.e.84% of all possible sequences for all taxa areincluded) and 30% missing data. Taxa used, theirpercentage of missing data, and their accessionnumbers are listed in Supplementary Table S1.There are 12 taxa whose missing data percent-age is above 50%, and they are evenly spreadthrough the groups sampled. The data consist ofnucleotides from the two ribosomal RNA genesand amino acids from the 6 protein coding genes,totalling 6,110 characters. To test the effect oflong branched Amoebozoa on the position of theApusozoa, two additional datasets were generated,one excluding the longest-branched amoebozoansand another with no amoebozoan representa-tives.

Phylogenetic Analyses

Figure 1 shows the combined results of themaximum likelihood (ML) and Bayesian infer-ence (BI) analyses of the full dataset. Bothtrees show almost the same branching pat-tern and display high support for several nodes.Notable exceptions include the weak associationof B. anathema with Amoebozoa, a lack of supportfor a relationship between the Choanoflagellataand Filasterea, and poor resolution of some ofthe internal branches within the choanoflagellates.The datasets which either exclude all amoebo-zoans, including B. anathema (SupplementaryFig. S1), or exclude the fastest evolving Amoe-bozoa (Supplementary Fig. S2), show topologiesmostly congruent with the main dataset, exceptthat B. anathema groups with the bikonts in theanalysis without the fastest evolving Amoebozoa(Supplementary Fig. S2).

The tree shows the division of Unikonta, includ-ing Apusozoa, from the Bikonta (Fig. 1), albeitwith weak support. However, when amoebo-zoans are completely removed from the analysis(Supplementary Fig. S1), the support for theunikont-bikont partition increases to a boot-strap support (BS) of 94%. This may becaused by the unstable phylogenetic position ofB. anathema, which has 57% missing data, and/orthe long branches present in most amoebozoanrepresentatives. With the full data set, B. anathemais positioned as the sister group to all other amoe-bozoans, but with low support (BS=54%, Fig. 1).

The monophyletic relationship and internal nodesof the other conventionally accepted amoebozoangroups are well supported.

The Apusozoa appear as a paraphyletic assem-blage laddered between the amoebozoans andthe opisthokonts (Fig. 1). The apusozoan Ancy-romonas micra (Order Ancyromonadida, for-merly known as Planomonas micra belongingto Order Planomonadida, see Heiss et al.2010) diverges first while a clade consist-ing of Apusomonas proboscidea and Theca-monas trahens (Order Apusomonadida, T. trahenswas formerly known as Amastigomonas sp.ATCC50062, see Cavalier-Smith and Chao 2010)is shown as sister group to the opisthokonts.Both the relationship between the two apuso-zoan lineages and the opisthokonts and the splitbetween the ancyromonads and apusomonadsare further reinforced when the amoebozoansare removed from the analyses (SupplementaryFig. S1).

The Opisthokonta are recovered with maximumsupport with all the alignments and algorithms(Fig. 1, Supplementary Figs S1 and S2). Theirsubdivision into the Holozoa and the Holomycotais also well supported. Within the Holomycota,two clades with high support are found, one lead-ing to F. alba and Nuclearia simplex and theother to Fungi. Within the Holozoa, the treesshow the Ichthyosporea (Amoebidium parasiticum,Sphaeroforma arctica and C. fragrantissima) asthe first branch, with C. fragrantissima as a sis-ter taxon to S. arctica (with maximum support byall analyses, Fig. 1, Supplementary Figs S1 andS2). The node that positions the ichthyospore-ans at the base of the Holozoa and groupsC. limacisporum with the remaining Holozoaholds moderately high support (BS=87%), whichincreases to 98% BS in the tree without Amoebo-zoa (Supplementary Fig. S1). While it seems clearthat Corallochytrea, Filasterea and Choanoflagel-lates + Metazoa form a clade, the exact branchingorder among them is not resolved. Also, whilethe monophyly of Filasterea (M. vibrans andC. owczarzaki, Shalchian-Tabrizi et al. 2008)is recovered, the support value is not high(BS=61%). The Metazoa and the Choanoflagellataare strongly recovered as sister groups (BS=90%),and the internal phylogeny of the choanoflag-ellates recovers two clades present in previousstudies (Nitsche et al. 2011): the Acanthoecidaeand the Salpingoecidae (Fig. 1). The freshwaterchoanoflagellate C. botrytis groups with the Acan-thoecidae but only with moderate Bayesian support(Fig. 1).



Trichoplax sp.Nematostella vectensis

Amphimedon queenslandica

Anemonia viridis

Clytia hemisphaericaHydra magnipapillata

Oscarella carmelaMnemiopsis leidyi

Heterochone calyx

67

58

Codosiga botrytis

Stephanoeca diplocostata

Didymoeca costata

Acanthoeca spectabilisSavillea micropora

Helgoeca nana

Salpingoeca rosettaSalpingoeca pyxidiumSalpingoeca urceolata

Diaphanoeca grandis

Choanoeca perplexaMonosiga brevicollis

Salpingoeca infusionum

Salpingoeca napiformis

Monosiga ovata

Ministeria vibransCapsaspora owczarzaki

Corallochytrium limacisporum

Salpingoeca amphoridium

88

94

53/0.93

6891-

79

90

70

0.99

85

53

55

Amoebidium parasiticumSphaeroforma arctica

Creolimax fragrantissima

Neurospora crassa

Blastocladiella emersoniiAllomyces macrogynus

Cryptococcus neoformans Mucor circinelloides

Glomus intraradicesRhizopus stolonifer

Nuclearia simplexFonticula alba

Apusomonas proboscideaThecamonas trahens

Ancyromonas micraBreviata anathema

Physarum polycephalum

Spizellomyces punctatus

Batrachochytrium dendrobatidisChytriomyces confervae

Neocallimastix sp. GE13

Rhizophlyctis rosea

Bigelowiella natans

Phytophthora palmivora

Toxoplasma gondii

Ostreococcus tauriChlamydomonas reinhardtii

Oryza sativaArabidopsis thaliana

Dictyostelium discoideumDictyostelium purpureum

Malawimonas jakobiformis

Acanthamoeba castellaniiHartmannella vermiformis

Reclinomonas americana

Paulinella chromatophora

Thalassiosira pseudonana

Tetrahymena thermophila

Telonema subtilisLeucocryptos marina

Pavlova lutheri

Guillardia thetaCryptomonas paramecium

87

74

61

0.84

7271/0.99

93 76

96

82

61

93

92

54

96

54

94

62

61

48

0.1

Jakoba bahamiensis Euglena gracilis

Metazoa

Choanoflagellata

Filasterea

Ichthyosporea

Fungi

Holozoa

Holomycota

Amoebozoa

Bikonta

ApusozoaOpisthokonta

1

1

11

1

10.98

0.9

0.76

0.8

1

-

-

0.96

0.95

1

1

0.9711

1

1

1

1

1

1

1

0,7

-

1

0.51

0.95

-

-

-0.67


6 J. Paps et al.

Table 1. Comparison of topologies using the approx-imately unbiased test.

Topologies AU Test(p-values)

Test 1Best tree 0.87Corallochytrea + Choanoflagellata 0.0054Corallochytrea + Ichthyosporea 0.013Ichthyosporea + Filasterea 0.076

Test 2Best tree 0.99Apusozoa monophyletic 0.0056Apusozoa with Bikonta 2.7e-04Apusozoa as a sister group to

Unikonta4.0e-04

Topological Constraint Tests

Different phylogenetic hypotheses were statis-tically tested (Table 1). Most of the constrainedtopologies were based upon results from previouspublications. Three hypothetical positions forthe Apusozoa were tested: 1) the ‘Apusozoa asmonophyletic’ hypothesis (Cavalier-Smith andChao 2003), 2) the ‘Apusozoa as members ofthe Bikonta’ (Stechmann and Cavalier-Smith2003), and 3) the ‘Apusozoa as a sister groupto the Unikonta’ (where the Bikonta are alsoconstrained as a monophyletic lineage). Within theopisthokonts several alternative branching orderswere tested including: C. limacisporum as a sistertaxon to the Choanoflagellata (Cavalier-Smith andChao 2003; Jostensen et al. 2002; Mendoza et al.2002; Ruiz-Trillo et al. 2006), a C. limacisporumsister taxon relationship to the Ichthyosporea(Carr et al. 2008; Ruiz-Trillo et al. 2004, 2006;Steenkamp et al. 2006), and an ichthyosporeansister group relationship to the Filasterea (similarto Ruiz-Trillo et al. (2008) where C. owczarzakirepresents the Filasterea). The ApproximatelyUnbiased test (AU test) results (Table 1) showthat all these alternative hypotheses were

statistically rejected (values under 0.05), exceptfor the sister-group relationship of Filasterea andIchthyosporea.

Discussion

Unikonta Phylogeny

Our results recover the Amoebozoa, including B.anathema, as one of two major unikont clades(Fig. 1), although this arrangement is not robustlysupported. Nonetheless, this result has also beenseen in previous ribosomal RNA (18S) and phy-logenomic analyses (Minge et al. 2008; Nikolaevet al. 2006). Recent phylogenomics analyses havealso suggested that Breviatea (Cavalier-Smith et al.2004) is the sister group to all other amoebozoans(Minge et al. 2008), but other analyses have shownB. anathema to be related to the Apusozoa (Katzet al. 2011; Walker et al. 2006). Thus it remainsunclear whether B. anathema branches within oroutside the Amoebozoa.

Our analyses show the Apusozoa as a para-phyletic assemblage, with the Ancyromonadida(Heiss et al. 2010) splitting off first, followed bya well-supported Apusomonadida (Karpov andMylnikov 1989) and Opisthokonta. Although thisresult was reinforced following removal of theamoebozoan taxa (Supplementary Fig. S2), it isbased on a single representative of the Ancy-romonadida. A recent article by Cavalier-Smithand Chao (2010) based on the 18S ribosomalRNA gene also suggests, although with low sup-port, that the apusozoans are paraphyletic andthat they are more closely related to opisthokontsthan to amoebozoans. In addition, Torruella et al.(2012) recovered a well supported sister grouprelationship between the opisthokonts and a sin-gle apusomonad representative (T. trahens), usingphylogenomic analysis. Finally, our topologicalconstraint tests reject all alternative hypothe-ses enforcing either a monophyletic Apusozoa ortheir placement elsewhere in the eukaryote tree(Table 1).

➛

Figure 1. Maximum Likelihood phylogenetic tree of the Unikonta from eight genes.The figure shows a phylogenetic tree of the Unikonta inferred from the eight molecular markers by maximumlikelihood. A black dot indicates a clade with Bayesian inference posterior probabilities (PP) of 1.0 and bootstrapsupport (BS) of 100%. A grey dot indicates a clade with PP values of 1.0 and BS of 97-99%. Values aboveand below the branches correspond to BS values (if greater than 50%), and PP values, respectively. Dashesspecify non recovered nodes. The scale bar indicates the number of changes per site. For accession numberscorresponding to the markers for each terminal see Supplementary Table S1.



The Opisthokonta

The Opisthokonta and their division into two majorsubgroups, the Holomycota and the Holozoa, arestrongly supported by all our analyses (Fig. 1,Supplementary Figs S1 and S2). While thosetwo subgroups have been recovered in previ-ous trees (Lang et al. 2002; Medina et al. 2003;Ruiz-Trillo et al. 2004, 2008; Steenkamp andBaldauf 2004), this is the first multigenic analysisto include such high numbers of choanoflagel-lates and ichthyosporeans, as well as generoustaxon sampling from close outgroup taxa. Withinthe Holomycota, N. simplex and F. alba forma robust clade as a sister group to the Fungi,in agreement with the only other study withcomparable taxon sampling (Brown et al. 2009)(Fig. 1).

Within the Holozoa, our results support the well-known relationship between the Metazoa and theChoanoflagellata (Carr et al. 2008; Lang et al.2002; Medina et al. 2001; Ruiz-Trillo et al. 2006,2008; Torruella et al. 2012). Otherwise, most of theinternal phylogeny of Choanoflagellata is weaklysupported, likely a consequence of the high per-centage of missing data in most choanoflagellatetaxa. Our trees also support a sister-group rela-tionship between the Ichthyosporea and the restof holozoans rather than placing Ichthyosporeaas a sister group to the Filasterea as shownby Ruiz-Trillo et al. (2008). This result is moreconsistent with the “Filozoa hypothesis” (Shalchian-Tabrizi et al. 2008) as seen in other molecularanalyses (Ruiz-Trillo et al. 2008; Shalchian-Tabriziet al. 2008; Torruella et al. 2012), but an exclusiverelationship between Ichthyosporea and Filastereacan not be rejected by our AU test (p-value0.076).

Our analysis differsfrom previous ones in thatit also includes C. limacisporum. C. limacisporumhas previously been suggested to be related tothe choanoflagellates (Cavalier-Smith and Allsopp1996), to Fungi (Sumathi et al. 2006) or to theIchthyosporea (Steenkamp et al. 2006). Instead ourresults suggest that C. limacisporum is a sisterlineage to the Filozoa and all previous hypothe-ses were rejected based on the AU test of ourdataset. Nonetheless, the specific position of C.limacisporum remains unresolved, and more dataand a wider taxon sampling are almost certainlyneeded to recover this specific phylogenetic posi-tion. Unfortunately current taxon sampling fromCorallochytrium, Capsaspora, and Ministeria islimited by the number of representatives known toscience.

The Evolution of Basal Body Arrangementand the Unikont/Bikont Condition

Although our tree does not robustly infer the phylo-genetic position of all taxa analyzed, it does bothresolve and confirm many relationships, therebyforming a reasonable starting point to postulateevolutionary scenarios. As an example, Figure 2shows a summary of our main results, accompa-nied by a sketch of the locomotion complex (kinetid)for each group. This scheme can be used to out-line a hypothetical proposal for the evolution of theflagellar apparatus in unikonts. Nonetheless, werecognize that this proposal is limited by boththe number species known to date, as well asthe incomplete morphological data available. Thekinetid is formed by one or two flagella and one ormore basal bodies. Flagella are always attachedto a basal body, but not all the basal bodies withina kinetid have an attached flagellum. Groups thatlack a kinetid tend to utilize amoeboid locomotion(see Fig. 2). The structure of the kinetid in the lastcommon ancestor of all Amoebozoa is importantfor understanding the origins of the eukaryotes, i.e.the validity of unikont/bikont rooting. Although mostamoebozoans do not have flagella, some displayone or more flagella and varying numbers of basalbodies. For example among Amoebozoa, one basalbody is found in some Archamoebae and in theconosean Multicilia marina, while two basal bod-ies are seen in some myxogastrid slime molds, anda mixture of both states is found in different proto-stelids (Minge et al. 2008).

The kinetid configuration consisting of one fla-gellum and one basal body as displayed in someamoebozoans is similar to the ancestral unikont-like eukaryote proposed by Cavalier-Smith (2002);thus an ancestral Amoebozoa with this arrange-ment would support the latter hypothesis. However,those flagellated amoebozoans having a singlebasal body show a more recently derived positionwithin the Amoebozoa, making it difficult to clar-ify the ancestral state of the group. Furthermore,B. anathema, which positions here as the sistergroup to all other amoebozoans, has a single fla-gellum but has at least two basal bodies, a statealso present in the flagellated Opisthokonta (Rogerand Simpson 2009). In contrast to opisthokonts, B.anathema’s kinetid is found in the anterior part ofthe cell while that of opisthokonts’ is in the posteriorpart. This, together with B. anathema’s position asthe earliest known branch of Amoebozoa (Fig. 1,Minge et al. 2008), points to the presence of atleast two basal bodies in the last common ances-tor of unikonts, as opposed to an ancestor with a


8 J. Paps et al.

Figure 2. A schematic tree of the unikonts with an outline of kinetid structure.

single basal body, and further suggests that thekinetid was located in the anterior part of the cell.This hypothesis is further reinforced by the appar-ent paraphyly of the Apusozoa, where each branchis represented by taxa having two flagella and twobasal bodies in the anterior part of the cell (Rogerand Simpson 2009). Therefore the presence ofan anterior kinetid with at least two basal bod-ies in some bikonts, B. anathema, ancyromonadsand apusomonads, suggests that the last commonancestor of eukaryotes may also have had an ante-rior kinetid with at least two basal bodies.

Inferring the number of flagella per kinetid in thelast common eukaryotic ancestor is more specu-lative and may involve several independent gainsand/or losses. Assuming that the ancestor hadtwo flagella (as in bikonts and the two apusozoanbranches), then one flagellum was independentlylost in Breviatea and Opisthokonta. Alternatively, ifthe eukaryotic ancestor had one flagellum, then asecond one was independently gained within theAmoebozoa (i.e. in Physarium) and again, eitherin the last common ancestor of ancyromonads,apusomonads and opisthokonts (then lost again

in opishtokonts) or in both the ancyromonads andapusomonads.

Within the Opisthokonta, the archetypical pos-teriorly flagellated morphology is represented bythe choanoflagellates. Uni-flagellated life stagesare present as zoospores in early-branching fungi(Tanabe et al. 2005) and the ichthyosporean orderDermocystida (Mendoza et al. 2002), and also asmetazoan sperm cells. While it is clear that thekinetid was lost in the ichthyosporean order Ichthyo-phonida and in the derived fungi (Tanabe et al.2005), the evolution of the flagellar apparatus inother opisthokonts remains less clear due to thelimited number of isolated taxa and morphologicalstudies.

If the flagellar apparatus is homologous betweenopisthokonts and apusozoans, then the possibleparaphyly of the apusozoans (Fig. 1) would indi-cate an apusozoan-like ancestor for opisthokonts.This allows for a provisional description of kinetidevolution in the opisthokont lineage. Accordingly,the anterior kinetid seen in apusozoans wouldhave moved to the posterior part of the cell and oneof the two flagella would have been lost, thereby



leaving the opisthokont ancestor with one basalbody attached to a single posterior flagellumattached to one basal body and a second non-flagellated basal body. This means that thishypothetical second flagellum would have beenlost twice within the unikonts: once in the anteriorkinetid of B. anathema and once in a commonancestor of the opisthokonts. Finally, the kinetidwould have been lost on at least five occasionswithin the opisthokonts: 1) in the common ancestorof nucleariids and F. alba as proposed by Brownet al. (2009), 2) in the derived fungi, likely manytimes, consistent with Hibbet et al. (2007), and threemore times within the Holozoa in 3) Ichthyophonida,4) C. limacisporum and 5) Filasterea. In the caseof filastereans, it is noteworthy that according toCavalier-Smith (Cavalier-Smith and Chao 2003)“Ministeria vibrans has a vibratile stalk which hassome ultrastructural similarity to a cilium”, but thereis no evidence of any flagellum/cilium in Ministeriamarisola or C. owczarzaki. In any case, furtherdata on the ontogeny, structure and genetics of thedifferent eukaryotic clades is needed to completelyunderstand flagellar evolution.

Conclusions

Our multigene analysis of the Opisthokonta con-firms some previous phylogenetic relationships,such as the monophyly of the opisthokonts, thedivision of opisthokonts into Holozoa and Holomy-cota, and the monophyly of nucleariids + F. albaand their sister group relationship to Fungi. Ourdata also provide strong support to some con-tentious hypotheses, such as apusozoan paraphylyand their sister group relationship to opisthokonts,and moderate support to the position of Filastereaas sister group to Choanoflagellata and Metazoarather than to Ichthyosporea. Finally, our resultssuggest that C. limacisporum is related to thefilozoan clades, although its specific phylogeneticposition within the Holozoa remains unclear. Ourtree provides a framework on which to discusscharacter evolution as more morphological and bio-chemical data on lesser known taxa and lineagesbecomes available.

Methods

Organisms and molecular methods: The culture of theichthyosporean C. fragrantissima strain CH2 was grown asspecified in Marshall et al. (2008). The culture of thechoanoflagellate C. botrytis was a kind donation from Prof.Hartmut Arndt (University of Cologne) to Hiroshi Suga and

was grown in WC medium (Guillard and Lorenzen 1972). TheRNA extractions were performed from pelleted live culturesusing TRIzol (Amersham Pharmacia Biotech) and cDNA wasobtained by reverse transcription with M-MLV reverse transcrip-tase (Promega). Eight markers were amplified and sequenced:the two ribosomal RNA genes (18S and 28S) and six proteincoding genes: actin, elongation factor 1 alpha (EF1 alpha),alpha- and beta-tubulin, and the heat shock proteins 70 (hsp70) and 90 (hsp 90). The 8 gene fragments were amplifiedby PCR using universal degenerate primers (SupplementaryTable S2). The PCR was performed with 25 �l final volume,using 1 unit of Taq polymerase and 0.5 �l betaine (Sigma)as PCR enhancer. PCR products were purified using Micro-con PCR columns (Millipore) and directly sequenced from bothstrands (Big Dye Terminator V.2.0, Applied Biosystems). Ampli-fication products were ethanol precipitated and run on an ABIPrism 3700 (Applied Biosystems) automated sequencer. Thesequence data were assembled with SeqEd V.1.0.3 (AppliedBiosystems). New sequences have been deposited in GenBankwith accession numbers HQ896013-22.

Alignments and phylogenetic analyses: Gene sequenceswere downloaded from GenBank or from genome and ESTprojects hosted in different public institutes (Broad Institute,Joint Genome Institute, etc). See Supplementary Table S1 forsequences origins and accession numbers. Each gene wasaligned independently using MAFFT (Katoh et al. 2005), andthe resulting alignments were checked by eye with Bioedit v.7.0.9.0 (Hall 1999). Regions of ambiguous alignment wereremoved using the online version of Gblocks (Castresana 2000)with the “less stringent selection” options selected. The finalalignments had 1390 positions for 18S, 2206 for 28S, 371for actin, 404 for EF1-alpha, 508 for hsp70, 534 for, hsp90,392 for tubulin alpha, and 398 for tubulin beta. For eachalignment, gene orthology was assessed with single-genephylogenies. The final curated alignments were concate-nated using Bioedit, producing a 6,110 positions matrix for73 taxa. Alignments can be downloaded from the webpagehttp://www.multicellgenome.com and have been deposited atDryad Repository: http://dx.doi.org/10.5061/dryad.v3p3j.

Bayesian inference analyses were conducted with a par-allelized version of MrBayes software v.3.1.2 (Ronquist andHuelsenbeck 2003), using a partitioned dataset (one partitionfor each gene, unlinking parameter estimation for each partition)and running 3,000,000 generations in 2 independent sets of4 chains (with a sample frequency of 1 in 1000). MrBayeswas run using the model GTR + � + I (4 gamma categories + 1invariable) for the nucleotide markers (ribosomal genes) andWAG + � + I (4 gamma categories + 1 invariable) for the pro-tein sequences; all the 8 partitions had an independentlyapplied co-varion correction. Before obtaining consensus treeand posterior probabilities, all trees sampled before the like-lihood values reached a plateau were removed, resulting ina burnin of a 1,000,000 generations (all the PRSF valuesapproaching 1 to the second decimal point). Maximum likeli-hood trees were inferred with RAxML v.7.2.6 (Stamatakis 2006),using a partitioned dataset (8 partitions, one for each gene),using the model GTR + � + I (4 gamma categories + 1 invariable)for the nucleotide partitions (ribosomal genes) and LG + � + I(4 gamma categories + 1 invariable) for the amino acid ones(protein-coding genes). A random topology was used as start-ing tree and 1,000 bootstrap replicates were obtained (options-f i -b 1999 -#1000). Four independent analyses were run tocheck that the resulting topology was not the result of a singlerun being trapped in a local maximum.

In order to evaluate the competing topologies, we performedapproximately unbiased tests (Shimodaira and Hasegawa


10 J. Paps et al.

2001) on two topology sets. First (Test 1), the Holozoa phy-logeny was analysed. All possible topologies were exploredby combining seven subtrees (Metazoa, Choanoflagellata,Filasterea, Corallochytrea, Ichthyosporea, Fungi, and Apuso-zoa+Amoebozoa+Bikonta) from the ML tree, using all possiblecombinations of these sub-trees (745 topologies in total). Sec-ond (Test 2), the Apusozoa phylogeny was analysed. We testedall the possible positions, except inside the Holozoa, of theT. trahens + A. proboscidea lineage and the A. micra lineage onthe ML toplogy, fixing the other branching patterns (2499 topolo-gies in total). CONSEL V0.1i (Shimodaira and Hasegawa 2001)was used to perform the AU tests (Shimodaira 2002).

Acknowledgements

We are grateful to Arnau Sebé, Alex de Mendoza,Romain Derelle and Guifré Torruella for all thelively discussions and stimulating insights on thesubject. We thank Martin Carr’s (Faculty of Biolog-ical Sciences, University of Leeds) for feedback onthe internal choanoflagellate phylogeny, HarmutArndt (Zoological Institute, University of Cologne)for sharing Codosiga botrytis culture, B. FranzLang (Département de Biochimie, Universitéde Montréal) for access to Thecamonas trahensgenome sequences, as well as Alastair G. Simpson(Department of Biology, Dalhousie University) andThomas Cavalier-Smith (Department of Zoology,University of Oxford) for their helpful feedback onflagellar apparatus across different groups. Theauthors thank Prof. Peter Holland (Departmentof Zoology, University of Oxford) for his supportand the two anonymous referees for the greatfeedback provided. We would like to thank MariaJosé Barberà for the graphical support and SaraRojas for the animal drawings included in Figure 2.The authors thankfully acknowledge the computerresources, technical expertise and assistance pro-vided by the Barcelona Supercomputing Center-Centro Nacional de Supercomputación, speciallyto David Vicente (Barcelona SupercomputingCentre) and Luis Cabellos (Instituto de Física deCantabria). H.S. is supported by the Marie CurieIntra-European Fellowship within the 7th EuropeanCommunity Framework Programme. This workwas supported by an ICREA contract, an EuropeanResearch Council Starting Grant (ERC-2007-StG-206883), and two grants (BFU2008-02839/BMCand BFU2011-23434) from Ministerio de Ciencia eInnovación (MICINN) to IR-T.

Appendix A. Supplementary data

Supplementary data associated with this arti-cle can be found, in the online version, athttp://dx.doi.org/10.1016/j.protis.2012.09.002.

References

Adl SM, Simpson AG, Farmer MA, Andersen RA,Anderson OR, Barta JR, Bowser SS, Brugerolle G,Fensome RA, Fredericq S, James TY, Karpov S, KugrensP, Krug J, Lane CE, Lewis LA, Lodge J, Lynn DH, Mann DG,McCourt RM, Mendoza L, Moestrup Ø, Mozley-StandridgeSE, Nerad TA, Shearer CA, Smirnov AV, Spiegel FW,Taylor MF (2005) The new higher level classification of eukary-otes with emphasis on the taxonomy of protists. J EukaryotMicrobiol 52:399–451

Baldauf SL, Palmer JD (1993) Animals and fungi are eachother’s closest relatives: congruent evidence from multiple pro-teins. Proc Natl Acad Sci USA 90:11558–11562

Baldauf SL, Roger AJ, Wenk-Siefert I, Doolittle WF (2000)A kingdom-level phylogeny of eukaryotes based on combinedprotein data. Science 290:972–977

Brown MW, Spiegel FW, Silberman JD (2009) Phylogenyof the “forgotten” cellular slime mold, Fonticula alba, revealsa key evolutionary branch within Opisthokonta. Mol Biol Evol26:2699–2709

Burki F, Shalchian-Tabrizi K, Minge M, Skjaeveland A,Nikolaev SI, Jakobsen KS, Pawlowski J (2007) Phyloge-nomics reshuffles the eukaryotic supergroups. PLoS ONE2:e790

Carr M, Leadbeater BS, Hassan R, Nelson M, Baldauf SL(2008) Molecular phylogeny of choanoflagellates, the sistergroup to Metazoa. Proc Natl Acad Sci USA 105:16641–16646

Castresana J (2000) Selection of conserved blocks from mul-tiple alignments for their use in phylogenetic analysis. Mol BiolEvol 17:540–552

Cavalier-Smith T (2002) The phagotrophic origin of eukary-otes and phylogenetic classification of Protozoa. Int J Syst EvolMicrobiol 52:297–354

Cavalier-Smith T (2010) Kingdoms Protozoa and Chromistaand the eozoan root of the eukaryotic tree. Biol Lett 6:342–345

Cavalier-Smith T, Allsopp MTEP (1996) Corallochytrium, anenigmatic non-flagellate protozoan related to choanoflagel-lates. Eur J Protistol 32:306–310

Cavalier-Smith T, Chao E (1995) The Opalozoan Apusomonasis related to the common ancestor of animals, fungi, andchoanoflagellates. Proc Roy Soc B Biol 261:1–6

Cavalier-Smith T, Chao EE (2003) Phylogeny of Choanozoa,Apusozoa, and other protozoa and early eukaryote megaevo-lution. J Mol Evol 56:540–563

Cavalier-Smith T, Chao EE (2010) Phylogeny and Evolutionof Apusomonadida (Protozoa: Apusozoa): new genera andspecies. Protist 161:549–576

Cavalier-Smith T, Chao E, Oates B (2004) Molecular phy-logeny of Amoebozoa and the evolutionary significance of theunikont Phalansterium. Eur J Protistol 40:21–48

Guillard RRL, Lorenzen CJ (1972) Yellow-green algae withchlorophyllide C1, 2. J Phycol 8:10–14

Hall TA (1999) BioEdit: a user-friendly biological sequencealignment editor and analysis program for Windows 95/98/NT.Nucleic Acids Symp Ser 41:95–98



Heiss AA, Walker G, Simpson AG (2010) Clarifying thetaxonomic identity of a phylogenetically important group ofeukaryotes: Planomonas is a junior synonym of Ancyromonas.J Eukaryot Microbiol 57:285–293

Hibbett DS, Binder M, Bischoff JF, Blackwell M, CannonPF, Eriksson OE, Huhndorf S, James T, Kirk PM, LuckingR, Thorsten Lumbsch H, Lutzoni F, Matheny PB, McLaugh-lin DJ, Powell MJ, Redhead S, Schoch CL, Spatafora JW,Stalpers JA, Vilgalys R, Aime MC, Aptroot A, Bauer R,Begerow D, Benny GL, Castlebury LA, Crous PW, Dai YC,Gams W, Geiser DM, Griffith GW, Gueidan C, HawksworthDL, Hestmark G, Hosaka K, Humber RA, Hyde KD, Iron-side JE, Koljalg U, Kurtzman CP, Larsson KH, LichtwardtR, Longcore J, Miadlikowska J, Miller A, Moncalvo JM,Mozley-Standridge S, Oberwinkler F, Parmasto E, Reeb V,Rogers JD, Roux C, Ryvarden L, Sampaio JP, SchusslerA, Sugiyama J, Thorn RG, Tibell L, Untereiner WA, WalkerC, Wang Z, Weir A, Weiss M, White MM, Winka K, Yao YJ,Zhang N (2007) A higher-level phylogenetic classification of theFungi. Mycol Res 111:509–547

Huang J, Xu Y, Gogarten JP (2005) The presence of a haloar-chaeal type tyrosyl-tRNA synthetase marks the opisthokonts asmonophyletic. Mol Biol Evol 22:2142–2146

Jostensen J-P, Sperstad S, Johansen S, Landfald B (2002)Molecular-phylogenetic, structural and biochemical features ofa cold-adapted, marine ichthyosporean near the animal-fungaldivergence, described from in vitro cultures. Eur J Protistol38:93–104

Karpov SA, Mylnikov AP (1989) Biology and ultrastructureof colourless flagellates Apusomonadida ord. n. ZoologischkeiZhurnal 58:5–17

Katoh K, Kuma K, Toh H, Miyata T (2005) MAFFT version5: improvement in accuracy of multiple sequence alignment.Nucleic Acids Res 33:511–518

Katz LA, Grant JR, Parfrey LW, Burleigh JG (2012) Tur-ning the Crown Upside Down: Gene Tree Parsimony Roots theEukaryotic Tree of Life. Syst Biol 61:653–660

Katz LA, Grant J, Parfrey LW, Gant A, O’Kelly CJ, AndersonOR, Molestina RE, Nerad T (2011) Subulatomonas tetrasporanov. gen. nov. sp. is a member of a previously unrecognizedmajor clade of eukaryotes. Protist 162:762–773

Keeling PJ, Burger G, Durnford DG, Lang BF, Lee RW, Pearl-man RE, Roger AJ, Gray MW (2005) The tree of eukaryotes.Trends Ecol Evol 20:670–676

Kim E, Simpson AG, Graham LE (2006) Evolutionary rela-tionships of apusomonads inferred from taxon-rich analyses of6 nuclear encoded genes. Mol Biol Evol 23:2455–2466

Lang BF, O’Kelly C, Nerad T, Gray MW, Burger G (2002) Theclosest unicellular relatives of animals. Curr Biol 12:1773–1778

Liu Y, Steenkamp ET, Brinkmann H, Forget L,Philippe H, Lang BF (2009) Phylogenomic analysespredict sistergroup relationship of nucleariids and fungi andparaphyly of zygomycetes with significant support. BMC EvolBiol 9:272

Marshall WL, Celio G, McLaughlin DJ, Berbee ML (2008)Multiple isolations of a culturable, motile ichthyosporean(Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n.gen., n. sp., from marine invertebrate digestive tracts. Protist159:415–433

Medina M, Collins AG, Silberman JD, Sogin ML (2001) Eval-uating hypotheses of basal animal phylogeny using completesequences of large and small subunit rRNA. Proc Natl AcadSci USA 98:9707–9712

Medina M, Collins AG, Taylor JW, Valentine JW, Lipps JH,Amaral-Zettler L, Sogin ML (2003) Phylogeny of Opisthokontaand the evolution of multicellularity and complexity in Fungi andMetazoa. Int J Astrobiol 2:203–211

Mendoza L, Taylor JW, Ajello L (2002) The class mesomyce-tozoea: a heterogeneous group of microorganisms at theanimal-fungal boundary. Annu Rev Microbiol 56:315–344

Minge MA, Silberman JD, Orr RJ, Cavalier-Smith T,Shalchian-Tabrizi K, Burki F, Skjaeveland A, Jakobsen KS(2008) Evolutionary position of breviate amoebae and the pri-mary eukaryote divergence. Proc Biol Sci 276:597–604

Nikolaev SI, Berney C, Petrov NB, Mylnikov AP, Fahrni JF,Pawlowski J (2006) Phylogenetic position of Multicilia marinaand the evolution of Amoebozoa. Int J Syst Evol Microbiol56:1449–1458

Nitsche F, Carr M, Arndt H, Leadbeater BS (2011) Higherlevel taxonomy and molecular phylogenetics of the choanoflag-ellatea. J Eukaryot Microbiol 58:452–462

Paps J, Ruiz-Trillo I (2010) Animals and Their UnicellularAncestors. Encyclopedia of Life Sciences (ELS). John Wiley& Sons, Ltd, Chichester

Richards TA, Cavalier-Smith T (2005) Myosin domain evo-lution and the primary divergence of eukaryotes. Nature436:1113–1118

Rodriguez-Ezpeleta N, Brinkmann H, Burger G, Roger AJ,Gray MW, Philippe H, Lang BF (2007) Toward resolving theeukaryotic tree: the phylogenetic positions of jakobids and cer-cozoans. Curr Biol 17:1420–1425

Roger AJ, Simpson AG (2009) Evolution: revisiting the root ofthe eukaryote tree. Curr Biol 19:R165–R167

Rogozin IB, Basu MK, Csuros M, Koonin EV (2009) Analysisof rare genomic changes does not support the unikont-bikontphylogeny and suggests cyanobacterial symbiosis as the pointof primary radiation of eukaryotes. Genome Biol Evol 1:99–113

Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesianphylogenetic inference under mixed models. Bioinformatics(Oxford, England) 19:1572–1574

Ruiz-Trillo I, Lane CE, Archibald JM, Roger AJ (2006)Insights into the evolutionary origin and genome architectureof the unicellular opisthokonts Capsaspora owczarzaki andSphaeroforma arctica. J Eukaryot Microbiol 53:1–6

Ruiz-Trillo I, Roger AJ, Burger G, Gray MW, Lang BF (2008)A phylogenomic investigation into the origin of Metazoa. MolBiol Evol 25:664–672

Ruiz-Trillo I, Inagaki Y, Davis LA, Sperstad S, Landfald B,Roger AJ (2004) Capsaspora owczarzaki is an independentopisthokont lineage. Curr Biol 14:R946–R947

Shadwick JD, Ruiz-Trillo I (2012) A genomic survey showsthat the haloarchaeal type tyrosyl tRNA synthetase is not asynapomorphy of opisthokonts. Eur J Protistol 48:89–93

Shalchian-Tabrizi K, Minge MA, Espelund M, Orr R,Ruden T, Jakobsen KS, Cavalier-Smith T (2008) Multigene


12 J. Paps et al.

phylogeny of Choanozoa and the origin of animals. PLoS ONE3:e2098

Shimodaira H (2002) An approximately unbiased test of phy-logenetic tree selection. Syst Biol 51:492–508

Shimodaira H, Hasegawa M (2001) CONSEL: for assessingthe confidence of phylogenetic tree selection. Bioinformatics(Oxford, England) 17:1246–1247

Simpson AG, Roger AJ (2004) The real ‘kingdoms’ of eukary-otes. Curr Biol 14:R693–R696

Stamatakis A (2006) RAxML-VI-HPC: maximumlikelihood-based phylogenetic analyses with thousandsof taxa and mixed models. Bioinformatics 22:2688–2690

Stechmann A, Cavalier-Smith T (2002) Rooting the eukaryotetree by using a derived gene fusion. Science 297:89–91

Stechmann A, Cavalier-Smith T (2003) The root of the eukary-ote tree pinpointed. Curr Biol 13:R665–R666

Steenkamp ET, Baldauf SL (2004) Origin and Evolution ofAnimals, Fungi and their Unicellular Allies (Opisthokonta). InHirt RP, Horner DS (eds) Organelles, Genomes and Eukaryote

Phylogeny: An Evolutionary Synthesis in the Age of Genomics.CRC Press, Boca Raton, pp 109–129

Steenkamp ET, Wright J, Baldauf SL (2006) The protistanorigins of animals and fungi. Mol Biol Evol 23:93–106

Sumathi JC, Raghukumar S, Kasbekar DP, Raghukumar C(2006) Molecular evidence of fungal signatures in the marineprotist Corallochytrium limacisporum and its implications in theevolution of animals and fungi. Protist 157:363–376

Tanabe Y, Watanabe MM, Sugiyama J (2005) Evolutionaryrelationships among basal fungi (Chytridiomycota and Zygomy-cota): Insights from molecular phylogenetics. J Gen ApplMicrobiol 51:267–276

Torruella G, Derelle R, Paps J, Lang BF, Roger AJ,Shalchian-Tabrizi K, Ruiz-Trillo I (2012) Phylogenetic rela-tionships within the Opisthokonta based on phylogenomicanalyses of conserved single-copy protein domains. Mol BiolEvol 29:531–544

Walker G, Dacks JB, Martin Embley T (2006) Ultrastructuraldescription of Breviata anathema, n. gen., n. sp., the organismpreviously studied as “Mastigamoeba invertens”. J EukaryotMicrobiol 53:65–78

Available online at www.sciencedirect.com

Molecular Phylogeny of Unikonts: New Insights into the Position of Apusomonads and Ancyromonads and...

Documents

Transcript of Molecular Phylogeny of Unikonts: New Insights into the Position of Apusomonads and Ancyromonads and...