Eukaryotic Evolution: Getting to the Dispatch Root of the Problem

Home , Centrohelid, Cercozoa, Euglenozoa, Opisthokont, Phalansterium, Retortamonad

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Alastair G.B. Simpson and Andrew J. Roger diatoms), cryptomonads and probably haptophyte algae. A major grouping of flagellates and amoebae, the ‘cercozoa’, emerged from improved taxon sampling of Comparative analyses of multiple genes suggest small subunit rRNAs. Recent evidence indicates that most known eukaryotes can be classified into half a the Foraminifera, and perhaps the Radiolaria, may dozen ‘super-groups’. A new investigation of the dis- belong with this assemblage [5–7]. Most ‘typical’ tribution of a fused gene pair amongst these ‘super- amoebae (such as Amoeba and Acanthamoeba), myce- groups’ has greatly narrowed the possible positions tozoan slime moulds, and the mitochondrion-lacking of the root of the eukaryote tree, clarifying the broad pelobionts and entamoebae form the ‘amoebozoa’. outlines of early eukaryote evolution. Most controversially, morphological data suggest a grouping, the ‘excavates’, which includes diplomonads and parabasalids (previously thought to be early- A decade ago, phylogenies based on small subunit diverging) together with Euglenozoa, Heterolobosea, ribosomal (r)RNA sequences provided an intuitively jakobids and several other protists [8]. appealing evolutionary tree of eukaryotes. Complex But where does the root lie? Philippe et al. [2] noted eukaryotes, including animals, fungi, plants and most that land plants (plantae), alveolates (chromalveolates) algae, emerged as a broad radiation usually called and some Euglenozoa (excavates) have fused genes the ‘eukaryotic crown’ [1]. Below this ‘crown’, more encoding a protein with both dihydrofolate reductase bizarre, and generally simpler, organisms diverged in a (DHFR) and thymidylate synthase (TS) activities. In ladder-like succession. The small subunit rRNA tree animals and fungi, these genes encode separate was ‘rooted’ with mitochondrion-lacking unicellular proteins, as in prokaryotes. Might this gene fusion be a eukaryotes such as diplomonads, parabasalids and derived feature within eukaryotes? This possibility microsporidia forming the basal branches (Figure 1a). prompted Stechmann and Cavalier-Smith [3] to examine But several studies now indicate that this rooting other groups for evidence of fused DHFR–TS. They was patterned more by methodological artefact than determined partial sequences of fused DHFR–TS genes historical signal, temporarily encouraging a view of in two chromalveolates (a stramenopile and a ciliate), eukaryote phylogeny as a large unresolved radiation [2]. another euglenozoan, and, most interestingly, a cerco- Recent years have seen tremendous progress in zoan plus two unplaced unicellular heterotrophs — an resolving this ‘radiation’. A wealth of single and apusomonad, and a centrohelid heliozoan. Apparently, concatenated gene phylogenies, improved taxon just two of the six super-groups may lack the gene sampling and examination of strong shared-derived fusion — the opisthokonts and amoebozoa. characters have revealed several novel eukaryote Under the simplest interpretation, the DHFR–TS ‘super-groups’. The hardest question of all — the fusion character drastically reduces the possible placement of the root in the eukaryote tree — has now locations for the eukaryote root. Stechmann and been clarified by Stechmann and Cavalier-Smith’s [3] Cavalier-Smith [3] argue that just three clusters need work on the phylogenetic distribution of a gene fusion consideration: opisthokonts, amoebozoa and the in eukaryotes. ‘fusion-bearing cluster’ of plantae, chromalveolates, Most known eukaryotes seem to fall into six ‘super- cercozoa (with Foraminifera and Radiolaria) and exca- groups’. The longest recognised super-group is the vates. Of these, only amoebozoa could ‘include’ the ‘opisthokonts’, including animals, fungi and a variety of root, because of the putatively derived nature of the unicellular relatives. As reported by Lang et al. [4] in this DHFR–TS fusion, and the opisthokont-specific EF-1α issue, the relationships within the opisthokonts — sequence insertion (see Figure 1 legend). Stechmann specifically the identification of the protistan sisters of and Cavalier-Smith [3] tentatively favour a rooting animals — have now been addressed by concatenated between amoebozoa plus the fusion cluster on one mitochondrial protein phylogenies. The ‘plantae’ side, and the opisthokonts alone on the other. This includes land plants and green algae, as well as red would place animals, fungi and their relatives in the algae and the obscure glaucocystophyte algae. Most ‘basal’ position in the eukaroytic tree: a humbling other eukaryote algae, and many heterotrophs, belong reversal for humans when compared to our previous to the ‘chromalveolate’ assemblage, which unites alve- lofty ‘crown’ position under the small subunit rRNA- olates (dinoflagellates, ciliates, Apicomplexa), stra- based model. menopiles/heterokonts (including brown algae and The DHFR–TS fusion provides the best estimate to date for the placement of the eukaryotic root. The Canadian Institute for Advanced Research, Program in DHFR–TS story could be misleading, however, if fused Evolutionary Biology, Genome Atlantic, Department of genes were acquired more than once in eukaryotic Biochemistry and Molecular Biology, Dalhousie University, evolution, either by multiple fusion events, or by Halifax, Nova Scotia B3H 4H7, Canada. eukaryote-to-eukaryote lateral gene transfers and E-mail: [email protected] and replacements. Alternatively the fused gene may be [email protected] ancestral for extant eukaryotes, with the separated Dispatch R692

A Eukaryote tree, B Eukaryote tree, circa 1993 2002 Cercozoa "Fusion (w/ Foraminifera, cluster" Diplomonads, w/ Radiolaria??) "Eukaryotic Retortamonads, crown" Carpediemonas, Plants + Plantae Excavates Parabasalia Green algae Stramenopiles, Animals Euglenozoa, Red algae Alveolates, Heterolobosea Glaucocystophytes Red algae, Fungi Malawimonas Chrom- etc. Plants Amoebozoa Oxymonads, Alveolates alveolates Trimastix Cryptophytes Euglenozoa, "Typical" amoebae Jakobids Stramenopiles Heterolobosea, ? Haptophytes Entamoebae, Mycetozoan slime moulds ? Mycetozoa, Pelobionts + Entamoebae Centrohelid Heliozoa Various amoebae Animals Apusomonads Parabasalia Choanoflagellates Diplomonads Mitochondria Ichthyosporea DHFR-TS fusion Microsporidia Fungi (w/ microsporidia) Nucleariid amoebae Others: Eukaryotic Mitochondria Eukaryotic Phalansterium Opisthokonts root Collodictyonids root Spironemids Eukaryotes Eukaryotes Kathablepharids Prokaryotes Telonema Prokaryotes Stephanopogon Multicilia etc.

Eubacteria Archaea ArchaeaEubacteria Current Biology

Figure 1. Contrasting views of eukaryotic evolution. (A) Eukaryotic evolution, as understood circa 1993 from small subunit ribosomal RNA phylogenies (after [1]). The earliest divergences involve amitochondriate protists, with animals ‘remote’ from the root. (B) Current view of eukaryotic phylogeny, with super-groups as determined primarily by multiple gene phylogenies, and with the deepest structure resolved according to the simplest interpretation of the DHFR–TS fusion, as reported by Stechmann and Cavalier-Smith [3]. Opisthokonts (purple), including animals and Fungi are supported by multiple gene phylogenies, and a large insertion in EF1α (see [13,14]). Relationships with opisthokonts are resolved as in [4]. Amoebozoa (light blue) include mycetozoan slime moulds, which are allied to typical amoebae (Euamoebae) by actin trees and a cox1–cox2 gene fusion [2,9], and to amitochondriate pelobionts and entamoebae in large concatenated gene trees [15]. Plantae, including land plants, are united, somewhat weakly, by concatenated gene phylogenies [16]. Chromalveolates (excluding haptophytes) are weakly grouped by concatenated protein phylogenies [14,15], but share a gene replacement of plastid GAPDH by a nuclear-derived copy, implying a common origin of their secondary plastids, where present [17]. Inclusion of haptophytes is widely expected, but not yet demonstrated [18]. Cercozoa are placed together and ‘adjacent to’ Radiolaria in small subunit ribosomal RNA phylogenies [6,7,18]. Some small subunit rRNA trees weakly place Radiolaria in an exclusive group with cercozoa (A.G.B.S., unpublished data). Actin trees imply a cercozoan-formaniferan relationship [5]. Within excavates, diplomonads plus parabasalids and Euglenozoa plus Heterolobosea groupings are recovered in several gene trees [14,19]. All excavates except parabasalids and Euglenozoa share a suite of cytoskeletal similarities, but almost never form a single group in molecular trees [8,9,14,20]. Apusomonads and centrohelid heliozoa have the DHFR–TS gene fusion [3], but no strong evidence suggests that they fall with any particular super-group. Some other taxa with con- temporary identities, but no robust position in tree are listed under ‘others’. Branch lengths are arbitrary, and all multifircations repre- sent uncertainty as to branching order. Branches currently lacking molecular corroboration are indicated with question marks. genes in opisthokonts being the derived condition. conflict with other examinations of similar datasets This ‘reversal’ could arise if multiple copies of the [8,9]. Certain excavates, the jakobids, have been impli- fused gene became specialized for different activities, cated in early eukaryotic diversification because of allowing loss of the other half of the gene. Multiple their ancestral bacterial-type mitochondrial RNA poly- genes could originate through either conventional merases [10]. There are also many eukaryotes that duplication, or lateral transfer. Stechmann and Cava- cannot be placed with confidence with any super- lier-Smith [3] cite sequence similarities between all group. In addition to apusomonads and centrohelid eukaryotic forms to refute transfer of either gene from heliozoa, which have the fusion, these include prokaryotes to opisthokonts, but transfer amongst Phalansterium, Collodictyonids, Spironemids, Kathe- eukaryotes remains possible. Interestingly, ‘eukaryotic blepharids, Stephanopogon, Telonema, Multicilia and type’ DHFR and TS are each present in some viruses. many parasitic and/or other amoeboid organisms that The rooting inference also relies on a reasonably may, or may not, be aberrant members of known correct underlying ‘super-group’ tree. The DHFR–TS groups [11]. Any of these might be critical to resolving fusion is currently known from only one or a few the exact position of the root. isolated taxa in each super-group. If the super-group Other aspects of Stechmann and Cavalier-Smith’s is not a natural group — a ‘clade’ — the inference that [3] view of eukaryote evolution — as illustrated in their all ‘members’ ancestrally had the fusion will be invalid. Figure 1 — are not directly addressed by the DHFR–TS In the case of excavates, the fusion is known from data and are weakly supported. The largely resolved just one subgroup, Euglenozoa; however, molecular branching order they depict within the ‘fusion cluster’ phylogenies generally place Euglenozoa separate from is mostly based on evolutionary scenarios for a few many other excavates, and the latter grouping is not contentious morphological characters, and is not sup- widely accepted. The one analysis where excavates ported by current molecular data. They present three form a natural group, an rRNA tree [7], is in marked arguments to tentatively place amoebozoa on the Current Biology R693

opposite side of the root to opisthokonts. Two are 9. Silberman, J.D., Simpson, A.G.B., Kulda, J., Cepicka, I., Hampl, V., structural similarities — tubular mitochondrial cristae Johnson, P.J. and Roger, A.J. (2002). Retortamonad flagellates are closely related to diplomonads – implications for the history of mito- and an anterior flagellum — shared by amoebozoa and chondrial function in eukaryote evolution. Mol. Biol. Evol. 19, the ‘fusion cluster’, which could easily be ancestral 777–786. features for all extant eukaryotes, rather than shared- 10. Gray, M.W., Burger, G. and Lang, B.F. (1999). Mitochondrial evolution. Science 283, 1476–1481. derived characters. Their third argument derives from 11. Patterson, D.J. (1999). The diversity of eukaryotes. Am. Nat. 65, the strong support for the split between opisthokonts S96–S124. and other eukaryotes in many gene trees, but this 12. Brocks, J.J., Logan, G.A., Buick, R. and Summons, R.E. (1999). could equally be interpreted as support for a relatively Archean molecular fossils and the early rise of eukaryotes. Science 285, 1033–1036. recent radiation of opisthokonts irrespective of where 13. Baldauf, S.L. and Palmer, J.D. (1993). Animals and fungi are each the root may lie. A more widely accepted depiction of other’s closest relatives: congruent evidence from multiple proteins. our current state of knowledge is shown in Figure 1b. Proc. Natl. Acad. Sci. USA 90, 11558–11562. So where do we go from here? The DHFR–TS fusion 14. Baldauf, S.L., Roger, A.J., Wenk-Siefert, I. and Doolittle, W.F. (2000). A kingdom-level phylogeny of eukaryotes based on combined gene should be sought in a wider range of eukaryotes, protein data. Science 290, 972–977. especially a greater diversity of excavates, amoebo- 15. Bapteste, E., Brinkmann, H., Lee, J.A., Moore, D.V., Sensen, C.W., zoa, and currently unplaced organisms. When more Gordon, P., Durufle, L., Gaasterland, T., Lopez, P., Müller, M., et al. (2002). The analysis of 100 genes supports the grouping of three complete DHFR and TS sequences are available, highly divergent amoebae: Dictyostelium, Entamoeba, and Mastig- actual phylogenies of these genes might reveal any amoeba. Proc. Natl. Acad. Sci. USA 99, 1414–1419. confounding eukaryote–eukaryote lateral transfers. 16. Moreira, D., Le Guyader, H. and Philippe, H. (2000). The origin of red algae and the evolution of chloroplasts. Nature 405, 69–72. Most importantly, other strong markers are needed to 17. Fast, N.M., Kissinger, J.C., Roos, D.S. and Keeling, P.J. (2001). confidently establish the monophyly of the super- Nuclear-encoded, plastid-targeted genes suggest a single common groups and test the possible roots implied by the origin for apicomplexan and dinoflagellate plastids. Mol. Biol. Evol. 18, 418–426. DHFR–TS fusion. 18. Cavalier-Smith, T. (2000). Flagellate megaevolution: the basis for Regardless of the precise position of the root, many eukaryote diversification. In The flagellates; unity, diversity and evo- other questions regarding early eukaryote evolution lution B.S.C. Leadbeater and J.C. Green, pp. 361–390. (London: persist. Does the difficulty in resolving the highest-level Taylor and Francis), pp. 361–390. 19. Embley, T.M. and Hirt, R.P. (1998). Early branching eukaryotes? branchings stem from a ‘big bang’ radiation of eukary- Curr. Opin. Genet. Dev. 8, 624–629. ote super-groups or does it reflect a ‘saturation’ of 20. Simpson, A.G.B., Radek, R., Dacks, J.B. and O’Kelly, C.J. (2002). phylogenetic signal? How old are eukaryotes anyway? How oxymonads lost their groove: An ultrastructural comparison of Monocercomonoides and excavate taxa. J. Eukaryot. Microbiol. 49, Did they originate in the Archean, as suggested by 2.7 239–248. billion year old eukaryotic-like sterane biomarkers [12], or after a ‘snowball Earth’ glaciation only 850 million years ago, as argued by Cavalier-Smith [7]? Are there any living eukaryotes that are simpler in their genetic or cellular makeup than the common ancestor of animals, fungi and plants, or have all ancestrally simple eukaryotes gone extinct [2]? With the wealth of data emerg- ing from comparative protistan genomics efforts and renewed intensive study of the ancient paleontological and geochemical record, we may finally be able to answer these questions.

References 1. Sogin, M.L. (1991). Early evolution and the origin of eukaryotes. Curr. Opin. Genet. Dev. 1, 457–463. 2. Philippe, H., Lopez, P., Brinkmann, H., Budin, K., Germot, A., Laurent, J., Moreira, D., Müller, M. and Le Guyader, H. (2000). Early- branching or fast-evolving eukaryotes? An answer based on slowly evolving positions. Proc. R. Soc. Lond. B. 267, 1213–1221. 3. Stechmann, A. and Cavalier-Smith, T. (2002). Rooting the eukaryote tree by using a derived gene fusion. Science 297, 89–91. 4. Lang, B.F., O’ Kelly, C.J., Nerad, T.A., Gray, M.W. and Burger, G. (2002). The closest unicellular relatives of animals. Curr. Biol., this issue. 5. Keeling, P.J. (2001). Foraminifera and Cercozoa are related in actin phylogeny: two orphans find a home? Mol. Biol. Evol. 18, 1551–1557. 6. Lopez-Garcia, P., Rodriguez-Valera, F. and Moreira, D. (2002). Toward the monophyly of Haeckel’s Radiolaria: 18S rRNA environ- mental data support the sisterhood of Polycystinea and Acan- tharea. Mol. Biol. Evol. 19, 118–121. 7. Cavalier-Smith, T. (2002). The phagotrophic origin of eukaryotes and phylogenetic classification of Protozoa. Int. J. Syst. Evol. Micro- biol. 52, 297–354. 8. Simpson, A.G.B., Roger, A.J., Silberman, J.D., Leipe, D., Edgcomb, V.P., Jermiin, L.S., Patterson, D.J. and Sogin, M.L. (2002). Evolution- ary history of ‘early diverging’ eukaryotes: The excavate taxon Car- pediemonas is closely related to Giardia. Mol. Biol. Evol. in press.