Abstract

The phylogenetic relationships of the biflagellate protist group Apusomonadidae have been unclear despite the availability of some molecular data. We analyzed sequences from 6 nuclear encoded genes—small-subunit rRNA, large-subunit rRNA, α-tubulin, β-tubulin, actin, and heat shock protein 90—to infer the phylogenetic position of Apusomonas proboscidea Aléxéieff 1924. To increase the taxon richness of the study, we also obtained new sequences from representatives of several other major eukaryotic groups: Chrysochromulina sp. National Institute for Environmental Studies 1333 (Haptophyta), Cyanophora paradoxa (Glaucophyta), Goniomonas truncata (Cryptophyceae), Leucocryptos marina (Kathablepharidae), Mesostigma viride (Streptophyta, Viridiplantae), Peridinium limbatum (Alveolata), Pterosperma cristatum (Prasinophytae, Viridiplantae), Synura sphagnicola (Stramenopiles), and Thaumatomonas sp. (Rhizaria). In most individual gene phylogenies, Apusomonas branched close to either of the 2 related taxa—Opisthokonta (including animals, fungi, and choanoflagellates) or Amoebozoa. Combined analyses of all 4 protein-coding genes or all 6 studied genes strongly supported the hypothesis that Apusomonadidae is closely related to Opisthokonta (or to all other eukaryotic groups except Opisthokonta, depending on the position of the eukaryotic root). Alternative hypotheses were rejected in approximately unbiased tests at the 5% level. However, the strong phylogenetic signal supporting a specific affiliation between Apusomonadidae and Opisthokonta largely originated from the α-tubulin data. If α-tubulin is not considered, topologies in which Apusomonadidae is sister to Opisthokonta or is sister to Amoebozoa were more or less equally supported. One current model for deep eukaryotic evolution holds that eukaryotes are divided into primary “unikont” and “bikont” clades and are descended from a “uniflagellate” common ancestor. Together with other information, our data suggest instead that unikonts (=Opisthokonta and Amoebozoa) are not strictly monophyletic and are descended from biflagellate ancestors.

Introduction

Our understanding of eukaryotic phylogeny has improved in recent years as the result of increasing sequence data from diverse taxonomic groups. For example, a molecular gene analysis revealed that many morphologically diverse protists form a superclade known as Rhizaria (Nikolaev et al. 2004). Most eukaryotes are now placed into 1 of about 15 major eukaryotic lineages, whose monophyly is generally undisputed. Some authors reduce these lineages to 5–6 “supergroups” (Simpson and Roger 2004; Adl et al. 2005; Keeling et al. 2005), although the monophyly of some of these supergroups is currently under debate. More contentiously still, it has been proposed that all major lineages fall into just 2 primary clades—“unikonts” and “bikonts” (Stechmann and Cavalier-Smith 2003b; Richards and Cavalier-Smith 2005). However, a number of protist taxa cannot be assigned unambiguously to any of these major eukaryotic lineages, despite the presence of electron microscopical data and at least some sequence data (Simpson and Roger 2004; Adl et al. 2005). These unassigned taxa are pivotal for understanding eukaryotic diversification and for testing macroevolutionary hypotheses, such as the unikont/bikont bifurcation (Stechmann and Cavalier-Smith 2003b; Cavalier-Smith et al. 2004). Apusomonadidae, the focus of our study, is one such group.

The Apusomonadidae (“apusomonads”) is a group of free-living heterotrophic biflagellates consisting of 2 genera—Amastigomonas and Apusomonas. Apusomonads glide along surfaces and feed on bacteria, which are usually engulfed using ventral pseudopodia (Vickerman et al. 1974). Their cells are covered with a thickened submembranous “theca” except in the ventral feeding region (Molina and Nerad 1991). The cells possess 2 heterodynamic flagella: 1 anteriorly directed and 1 posteriorly oriented (Vickerman et al. 1974). The anterior flagellum is encased by a membranous sleeve, a trait that is a synapomorphic feature for Apusomonadidae (Ekelund and Patterson 1997). The 2 basal bodies are inserted at right angles, and 3 rootlets are associated with these (Molina and Nerad 1991). One rootlet is a multilayered structure, whereas 2 other rootlets, each consisting of 3–5 and 5–12 microtubules, underline each side of the ventral groove (Karpoff and Zhukov 1986; Molina and Nerad 1991). Mitochondria have tubular cristae (Karpoff and Zhukov 1986; Molina and Nerad 1991). A distinguishing feature of Apusomonas is the mastigophore, a long extension of the 2 ventral grooves, from which the 2 flagella originate (Ekelund and Patterson 1997). Small-subunit (SSU) rRNA phylogenies clearly established the monophyly of Apusomonadidae (Cavalier-Smith and Chao 2003b).

Thus far, 2 nuclear encoded genes—SSU rRNA and heat shock protein (Hsp) 90—have been used to infer the phylogenetic relationships of Apusomonadidae (Cavalier-Smith and Chao 1995; Stechmann and Cavalier-Smith 2003a). Based on SSU rRNA gene phylogenies, Cavalier-Smith and Chao (1995) initially suggested that Apusomonas is closely related to Opisthokonta, the clade that includes animals, fungi, and choanoflagellates, with this hypothesis receiving moderate bootstrap support. The same relationship was also recovered in later SSU rRNA studies but with lower bootstrap support (e.g., Atkins et al. 2000; Cavalier-Smith and Chao 2003b, Figure 1; Cavalier-Smith et al. 2004, Figure 4). However, other analyses have not recovered a close association between Apusomonadidae and Opisthokonta (e.g., Cavalier-Smith 2002, Figure 2; Simpson et al. 2002; Berney et al. 2004; Cavalier-Smith et al. 2004, Figure 2). Analyses based on Hsp90 gene sequences including that of Amastigomonas marina did not strongly support any particular phylogenetic position for apusomonads (Stechmann and Cavalier-Smith 2003a). Furthermore, the dihydrofolate reductase–thymidylate synthase (DHFR–TS) gene fusion was identified in Amastigomonas debruynei, suggesting a placement of apuosomonads in the large bikont clade supposedly identified by this fusion character and, therefore, remote from the unikont clade (Opisthokonta and Amoebozoa), whose members lack this gene fusion (Stechmann and Cavalier-Smith 2002, 2003b).

FIG. 1.—

ML tree (JTT + Γ + I, 8 rate categories) inferred from 4 nuclear encoded protein-coding gene sequences. The data set included 26 taxa and 1,594 amino acid positions. The root was arbitrarily placed between the Apusomonadidae–Opisthokonta–Amoebozoa clade and the other taxa. Bayesian analysis (Whelen and Goldman + Γ + I, 8 rate categories) also found a similar topology. ML bootstrap values and Bayesian posterior probabilities are indicated at the corresponding nodes. Bootstrap values ≥50% and posterior probabilities ≥0.5 are shown. Dashes represent bootstrap values <50%. Taxa from which new sequences were obtained in our study are labeled in boldface.

FIG. 1.—

ML tree (JTT + Γ + I, 8 rate categories) inferred from 4 nuclear encoded protein-coding gene sequences. The data set included 26 taxa and 1,594 amino acid positions. The root was arbitrarily placed between the Apusomonadidae–Opisthokonta–Amoebozoa clade and the other taxa. Bayesian analysis (Whelen and Goldman + Γ + I, 8 rate categories) also found a similar topology. ML bootstrap values and Bayesian posterior probabilities are indicated at the corresponding nodes. Bootstrap values ≥50% and posterior probabilities ≥0.5 are shown. Dashes represent bootstrap values <50%. Taxa from which new sequences were obtained in our study are labeled in boldface.

FIG. 2.—

ML tree (general time reversible + Γ + I, 8 rate categories) was inferred from the combined nuclear encoded SSU rRNA and LSU rRNA gene data set. The data set included 48 taxa and 3,287 nucleotide positions—1,283 from SSU rRNA and 2,004 from LSU rRNA. Statistical support values are listed in the same way as figure 1. The only exception is for the node leading to Apusomonas. The SSU rRNA gene sequence of Mastigamoeba balamuthi was substituted with that of Vexillifera armata to reduce the overall branch length of Amoebozoa. When the SSU rRNA sequence of M. balamuthi was used, Mastigamoeba branched with the Cryptophyceae + Kathablepharidae clade, whereas Apusomonas remained branching close to Opisthokonta.

FIG. 2.—

ML tree (general time reversible + Γ + I, 8 rate categories) was inferred from the combined nuclear encoded SSU rRNA and LSU rRNA gene data set. The data set included 48 taxa and 3,287 nucleotide positions—1,283 from SSU rRNA and 2,004 from LSU rRNA. Statistical support values are listed in the same way as figure 1. The only exception is for the node leading to Apusomonas. The SSU rRNA gene sequence of Mastigamoeba balamuthi was substituted with that of Vexillifera armata to reduce the overall branch length of Amoebozoa. When the SSU rRNA sequence of M. balamuthi was used, Mastigamoeba branched with the Cryptophyceae + Kathablepharidae clade, whereas Apusomonas remained branching close to Opisthokonta.

In this study, we sequenced 5 nuclear encoded genes from Apusomonas proboscidea (large-subunit [LSU] rRNA, α-tubulin, β-tubulin, actin, and Hsp90) and performed various phylogenetic analyses with the goal of resolving the phylogenetic position of Apusomonadidae. We also analyzed existing SSU rRNA gene data from apusomonads. New sequences were also obtained from several distantly related protists in known eukaryotic groups, in order to increase the taxon richness of our study. Organisms included were Chrysochromulina sp. National Institute for Environmental Studies (NIES) 1333 (Haptophyta), Cyanophora paradoxa (Glaucophyta), Goniomonas truncata (Cryptophyceae), Leucocryptos marina (Kathablepharidae), Mesostigma viride (Streptophyta, Viridiplantae), Pterosperma cristatum (Prasinophytae, Viridiplantae), Peridinium limbatum (Alveolata), Synura sphagnicola (Stramenopiles), and Thaumatomonas sp. (Rhizaria). We carefully chose taxa with sequences that are not particularly long branched and/or that are early diverging members of some major eukaryotic lineages and performed both maximum likelihood (ML) and Bayesian analyses of individual and combined gene data sets.

Materials and Methods

Cultures

Apusomonas proboscidea Aléxéieff (CCAP 1905/1) was purchased from the Culture Collection of Algae and Protozoa (CCAP), Argyll, Scotland. Chrysochromulina sp. (NIES 1333), L. marina (Braarud) Butcher (NIES 1335), M. viride Lauterborn (NIES 296), and P. cristatum Schiller (NIES 936) were obtained from the Microbial Culture Collection at the NIES, Ibaraki, Japan. Cyanophora paradoxa Korshikov and Glaucocystis nostochinearum Itzigsohn var. nostochinearum were obtained from Carolina Biological Supply Company, Burlington, North Carolina. Goniomonas truncata (Fresenius) Stein, P. limbatum (Stokes) Lemmermann, S. sphagnicola (Korshikov) Korshikov, and Thaumatomonas sp. were cultured from various lakes in Wisconsin, United States following single-cell isolation (Stein 1973) and were maintained in appropriate culture media—in most cases, a mixture of sterilized filtered lake water and soil extract. Cultures were maintained at 15 °C. Cultures were identified both by light microscopic features and SSU rRNA gene sequences. The isolate identified as Thaumatomonas sp. had SSU rRNA gene sequence 99.8% similar to database sequences for Thaumatomonas sp. (South Africa), based on a National Center for Biotechnology Information Blast sequence similarity search.

DNA/RNA Preparation

Cells were concentrated by centrifugation, and genomic DNA was extracted using the DNeasy Plant Mini Kit (Qiagen, Valencia, CA), according to the manufacturer's suggested protocols. Cells of Glaucocystis, Mesostigma, Peridinium, Pterosperma, and Synura were disrupted in liquid nitrogen with a plastic pestle. Vigorous vortexing in the lysis solution was sufficient for colorless protist cells. In some cases, use of degenerate polymerase chain reaction (PCR) primers for amplifications of protein-coding genes from genomic DNA resulted in multiple bands presumably representing nonspecific amplification, and reverse transcriptase (RT)–PCR was necessary. Total RNA was purified from Apusomonas, Chrysochromulina, Goniomonas, Leucocryptos, and Thaumatomonas, using the RNeasy Mini Kit (Qiagen, Valencia, CA), and cDNA was synthesized from oligo dT primers using the Access RT–PCR kit (Promega, Madison, WI).

PCR Amplification, Cloning, and Sequencing

PCR primers for amplifying SSU rRNA, LSU rRNA, α-tubulin, β-tubulin, actin, and Hsp90 gene sequences were designed based on sequence alignments as well as previous studies (table 1) (Simpson et al. 2002, 2006). For some protein-coding genes, a 2-step nested PCR technique was applied. The standard 50-μl reaction mixture consisted of 2.5 unit of Takara Ex Taq (Takara, Tokyo, Japan), 1× Ex Taq buffer, 0.2 mM of each dNTP, 0.6 μM of each primer, and 5% glycerol. When PCR primers were degenerated at several positions, the standard PCR cyclic reactions consisted of a denaturation step at 95 °C for 3 min; 13 cycles of 1 min at 95 °C, 1 min at 58 °C (1 °C decrease each cycle), and 1.5 min at 68 °C; 20 cycles of 30 sec at 94 °C, 1 min at 45 °C, and 1.5 min at 68 °C; and a final 10 min at 68 °C. For SSU rRNA and LSU rRNA gene amplifications, the standard PCR cyclic reactions consisted of a denaturation step at 95 °C for 3 min; 30 cycles of 1 min at 95 °C, 1 min at 45 °C, and 1–3 min at 72 °C; and a final 15 min at 72 °C. For Hsp90 from Apusomonas, a 330-bp fragment near the 5′ end of the coding region was amplified by nested PCR and used to design an exact-match primer. A near-complete coding region was then amplified by a nested PCR with this exact-match primer. The 20-μl reaction mixtures consisted of 1 unit of Taq polymerase (Sigma-Aldrich, St. Louis, MO), 1× buffer (1.5mM MgCl2), 0.2 mM of each dNTP, and 1 μM of each primer. The cycling for the final PCR consisted of a denaturation step at 94 °C for 2 min; 35 cycles of 20s at 94 °C, 1 min at 54 °C, and 2.5 min at 72 °C; and a final 5 min at 72 °C. PCR-amplified fragments were either gel purified, or were cleaned using Wizard® SV Gel and PCR Clean-Up System (Promega, Madison, WI), and then were cloned into pCR 4-TOPO or pCR 2.1-TOPO vector (Invitrogen, Carlsbad, CA) or pGEM®-T Easy vector (Promega, Madison, WI). Plasmids were isolated from multiple positive bacterial clones using the QIAquick miniprep kit (Qiagen, Valencia, CA) or Sigma miniprep kit (Sigma-Aldrich, St. Louis, MO). Multiple clones were partially sequenced, and at least 1 clone from each reaction was selected for complete sequencing. To eliminate the possibility of contamination, the identities of all gene sequences were verified as described in Supplementary Material online (Method S1). GenBank accession numbers of new sequences obtained in this study are listed in Supplementary Material online (table S1).

Table 1

Primers for PCR Amplifying and Sequencing Nuclear Encoded SSU rRNA, LSU rRNA, α-Tubulin, β-Tubulin, Actin, and Hsp90 Genes. Primer Positions Are Relative to Location within Arabidopsis thaliana (LSU rRNA) or Chlamydomonas reinhardtii (all others)

Primer Other Names 5′ End 3′ End Primer Sequence Reference 
nu-SSU primers      
    nu-SSU-0024-5′ NSF4/21 0004 0024 CTG GTT GAT CCT GCC AGT AGT This study 
    nu-SSU-0033-5′ NSF13/21 0013 0033 CCT GCC AGT AGT CAT AYG CTT This study 
    nu-SSU-0977-5′ NSF963/18 0960 0977 TTR ATC AAG AAC GAA AGT This study 
    nu-SSU-1173-3′ NSR1197/24 1196 1173 CCC GTG TTG AGT CAA ATT AAG CCG This study 
    nu-SSU-1757-3′ NSR1784/21 1777 1757 CAG GTT CAC CTA CGG AAA CCT This study 
    nu-SSU-1768-3′ NSR1795/21 1788 1768 TGA TCC TTC YGC AGG TTC ACC This study 
nu-LSU primers      
    nu-LSU-0046-5′ NLF184/23 0024 0046 ACC CGC TGA AYT TAA GCA TAT CA This study 
    nu-LSU-0058-5′ NLF196/23 0036 0058 TAA GCA TAT CAM TAA GCG GAG GA This study 
    nu-LSU-1152-5′ NLF1280/23 1130 1152 TTT GGT AAG CAG AAC TGG CGA TG This study 
    nu-LSU-1262-3′ NLR1431/23 1284 1262 AGT TGT TAC ACA CTC CTT AGC GG This study 
    nu-LSU-2199-5′ NLF2343/24 2176 2199 TGA TTT CTG CCC AGT GCT CTG AAT This study 
    nu-LSU-2383-3′ NLR2571/22 2404 2383 CTC AAC AGG GTC TTC TTT CCC C This study 
    nu-LSU-3100-3′ NLR3287/23 3122 3100 GGA TTC TGR CTT AGA GGC GTT CA This study 
α-Tubulin primers      
    nu-αTUB-0044-5′ TUAF22/23 0022 0044 CAC ATC GGN CAR GCC GGN RTC CA This study 
    nu-αTUB-0083-5′ TUAF58/26 0058 0083 TGC TGG GAG CTN TAC TGC CTN GAG CA This study 
    nu-αTUB-1219-3′ TUAR1248/26 1244 1219 TCC TCC ATN CCY TCN CCN ACR TAC CA This study 
    nu-αTUB-1237-3′ TUAR1268/26 1262 1237 GCY TCR GAR AAY TCN CCY TCC TCC AT This study 
β-Tubulin primers      
    nu-βTUB-0050-5′ TUBF28/23 0028 0050 GGN CAG TGY GGN AAC CAG ATY GG This study 
    nu-βTUB-0065-5′ funiv 0040 0065 AAY CAR ATY GGY KC/ideoxyI/AAR TTY TGG GA This study 
    nu-βTUB-1207-3′ buniv 1232 1207 GCY TC/ideoxyI/GWR AAY TCC AWY TCG TCC AT This study 
    nu-βTUB-1261-3′ TUBR1294/23 1283 1261 GCN TCC TGG TAC TGY TGR TAC TC This study 
Actin primers      
    nu-ACTIN-0056a-5′ ACTf-12 0034 0056 TGC GAC AAY GGN TCN GGM ATG GT This study 
    nu-ACTIN-0056b-5′ ACTf-13(s) 0037 0056 GAC AAY GGN WCN GGM ATG TG This study 
    nu-ACTIN-0060-5′ ACTf-13 0037 0060 GAC AAY GGN TCN GGM ATG GTS AAG This study 
    nu-ACTIN-0071-5′ ACTf-17 0049 0071 GGM ATG TGY AAG GCN GGN TTY GC This study 
    nu-ACTIN-0377-5′ ACT120f 0355 0377 GAR AAR ATG ACN CAR ATH ATG TT This study 
    nu-ACTIN-0400-3′ ACT139b 0419 0400 GCY TGD ATN GCN ACR TAC AT This study 
    nu-ACTIN-1108-3′ ACTb-376 1132 1108 AGA AGC AYT T/ideoxyI/C KGT GNA CRA TNG A This study 
    nu-ACTIN-1111-3′ ACTb-377 1133 1111 TAG AAG CAY TTN CKG TGN ACR AT This study 
HSP90 primers      
    nu-HSP90-0041-5′ HspFA 0022 0041 GAR ACN TTY GCN TTY CAR GC This study 
    nu-HSP90-0083-5′ 100XF 0052 0083 CAG CTG ATG TCC CTG ATC ATY AAY ACN TTY TA Simpson et al. (2002) 
    nu-HSP90-0389-5′ HspFB 0367 0389 CAR TTY GGT GTB GGY TTY TAC TC This study 
    nu-HSP90-0602-5′ HspFC 0581 0602 TSA AGG ACC TSR TCA AGA AGC A This study 
    nu-HSP90-1390-3′ HspRD 1410 1390 CTC NCC RGT GAT GWA GTA GAT This study 
    nu-HSP90-1732-3′ HspRB2 1754 1732 CGY TCC ATR TTN GCN GAC CAN CC This study 
    nu-HSP90-1741-3′ HspRA 1761 1741 CAT GAT NCG YTC CAT RTT NGC This study 
    nu-HSP90-1759-3′ 880XR 1781 1759 TCG CGC AGR GCY TGN GCR TTC AT Simpson et al. (2006) 
    nu-HSP90-1809-3′ HspRC 1833 1809 GGG GTT GAT TTC CAT NGT YTT CTT G This study 
    nu-HSP90-1813-3′ 910XR 1835 1813 TCG GGG TTG ATY TCC ATN GTY TT Simpson et al. (2006) 
    nu-HSP90-1990-3′ 970XR 2015 1990 TCG AGG GAG AGR CCN ARC TTR ATC AT Simpson et al. (2002) 
Primer Other Names 5′ End 3′ End Primer Sequence Reference 
nu-SSU primers      
    nu-SSU-0024-5′ NSF4/21 0004 0024 CTG GTT GAT CCT GCC AGT AGT This study 
    nu-SSU-0033-5′ NSF13/21 0013 0033 CCT GCC AGT AGT CAT AYG CTT This study 
    nu-SSU-0977-5′ NSF963/18 0960 0977 TTR ATC AAG AAC GAA AGT This study 
    nu-SSU-1173-3′ NSR1197/24 1196 1173 CCC GTG TTG AGT CAA ATT AAG CCG This study 
    nu-SSU-1757-3′ NSR1784/21 1777 1757 CAG GTT CAC CTA CGG AAA CCT This study 
    nu-SSU-1768-3′ NSR1795/21 1788 1768 TGA TCC TTC YGC AGG TTC ACC This study 
nu-LSU primers      
    nu-LSU-0046-5′ NLF184/23 0024 0046 ACC CGC TGA AYT TAA GCA TAT CA This study 
    nu-LSU-0058-5′ NLF196/23 0036 0058 TAA GCA TAT CAM TAA GCG GAG GA This study 
    nu-LSU-1152-5′ NLF1280/23 1130 1152 TTT GGT AAG CAG AAC TGG CGA TG This study 
    nu-LSU-1262-3′ NLR1431/23 1284 1262 AGT TGT TAC ACA CTC CTT AGC GG This study 
    nu-LSU-2199-5′ NLF2343/24 2176 2199 TGA TTT CTG CCC AGT GCT CTG AAT This study 
    nu-LSU-2383-3′ NLR2571/22 2404 2383 CTC AAC AGG GTC TTC TTT CCC C This study 
    nu-LSU-3100-3′ NLR3287/23 3122 3100 GGA TTC TGR CTT AGA GGC GTT CA This study 
α-Tubulin primers      
    nu-αTUB-0044-5′ TUAF22/23 0022 0044 CAC ATC GGN CAR GCC GGN RTC CA This study 
    nu-αTUB-0083-5′ TUAF58/26 0058 0083 TGC TGG GAG CTN TAC TGC CTN GAG CA This study 
    nu-αTUB-1219-3′ TUAR1248/26 1244 1219 TCC TCC ATN CCY TCN CCN ACR TAC CA This study 
    nu-αTUB-1237-3′ TUAR1268/26 1262 1237 GCY TCR GAR AAY TCN CCY TCC TCC AT This study 
β-Tubulin primers      
    nu-βTUB-0050-5′ TUBF28/23 0028 0050 GGN CAG TGY GGN AAC CAG ATY GG This study 
    nu-βTUB-0065-5′ funiv 0040 0065 AAY CAR ATY GGY KC/ideoxyI/AAR TTY TGG GA This study 
    nu-βTUB-1207-3′ buniv 1232 1207 GCY TC/ideoxyI/GWR AAY TCC AWY TCG TCC AT This study 
    nu-βTUB-1261-3′ TUBR1294/23 1283 1261 GCN TCC TGG TAC TGY TGR TAC TC This study 
Actin primers      
    nu-ACTIN-0056a-5′ ACTf-12 0034 0056 TGC GAC AAY GGN TCN GGM ATG GT This study 
    nu-ACTIN-0056b-5′ ACTf-13(s) 0037 0056 GAC AAY GGN WCN GGM ATG TG This study 
    nu-ACTIN-0060-5′ ACTf-13 0037 0060 GAC AAY GGN TCN GGM ATG GTS AAG This study 
    nu-ACTIN-0071-5′ ACTf-17 0049 0071 GGM ATG TGY AAG GCN GGN TTY GC This study 
    nu-ACTIN-0377-5′ ACT120f 0355 0377 GAR AAR ATG ACN CAR ATH ATG TT This study 
    nu-ACTIN-0400-3′ ACT139b 0419 0400 GCY TGD ATN GCN ACR TAC AT This study 
    nu-ACTIN-1108-3′ ACTb-376 1132 1108 AGA AGC AYT T/ideoxyI/C KGT GNA CRA TNG A This study 
    nu-ACTIN-1111-3′ ACTb-377 1133 1111 TAG AAG CAY TTN CKG TGN ACR AT This study 
HSP90 primers      
    nu-HSP90-0041-5′ HspFA 0022 0041 GAR ACN TTY GCN TTY CAR GC This study 
    nu-HSP90-0083-5′ 100XF 0052 0083 CAG CTG ATG TCC CTG ATC ATY AAY ACN TTY TA Simpson et al. (2002) 
    nu-HSP90-0389-5′ HspFB 0367 0389 CAR TTY GGT GTB GGY TTY TAC TC This study 
    nu-HSP90-0602-5′ HspFC 0581 0602 TSA AGG ACC TSR TCA AGA AGC A This study 
    nu-HSP90-1390-3′ HspRD 1410 1390 CTC NCC RGT GAT GWA GTA GAT This study 
    nu-HSP90-1732-3′ HspRB2 1754 1732 CGY TCC ATR TTN GCN GAC CAN CC This study 
    nu-HSP90-1741-3′ HspRA 1761 1741 CAT GAT NCG YTC CAT RTT NGC This study 
    nu-HSP90-1759-3′ 880XR 1781 1759 TCG CGC AGR GCY TGN GCR TTC AT Simpson et al. (2006) 
    nu-HSP90-1809-3′ HspRC 1833 1809 GGG GTT GAT TTC CAT NGT YTT CTT G This study 
    nu-HSP90-1813-3′ 910XR 1835 1813 TCG GGG TTG ATY TCC ATN GTY TT Simpson et al. (2006) 
    nu-HSP90-1990-3′ 970XR 2015 1990 TCG AGG GAG AGR CCN ARC TTR ATC AT Simpson et al. (2002) 

Sequence Alignments

Protein-coding genes were translated to amino acids, which were manually aligned using MacClade ver. 4.05 (Maddison DR and Maddison WP 2001). For the alignment of SSU rRNA genes, ClustalX (Thompson et al. 1997) was used to produce an approximate alignment, after which the sequences were aligned more accurately based on eukaryotic SSU rRNA secondary structure models (Wuyts et al. 2002) from the European database. Yves Van de Peer (Ghent University) kindly provided the LSU rRNA gene sequence alignment used in a previous study (Ben Ali et al. 2001), which incorporated both primary and secondary structure information. We added our new sequences and other sequences available from GenBank. Ambiguously aligned positions were excluded for all analyses. Initial individual gene alignments included more sequences than the final versions. This allowed detection of possible paralogy and selection of short-branched homologous copies. Through various preliminary phylogenetic analyses, mostly using Neighbor-Joining and maximum parsimony methods, long-branched or potentially nonorthologous sequences were identified and excluded. For example, sequences of most animals and embryophytes and some fungi were removed from our α-tubulin, β-tubulin, or actin gene alignments because multiple paralogs were present in these multicellular organisms. Multiple gene copies, present in some of the included taxa, were closely related to each other, to the exclusion of all other sequences analyzed. We carefully chose taxa included in the final alignments to increase overall taxonomic representation of the study, yet minimize potential problems associated with long-branch attraction artifacts (Philippe 2000). In some cases, gene sequences from closely related taxa were concatenated for the combined gene analyses. No alignment position or taxon was included in the analysis, if more than 20% of the total data were missing. In addition, in the combined gene alignments, no taxon was included if more than 40% of the individual gene data were missing. Sequence alignments are deposited at TreeBASE (http://www.treebase.org).

Because we restricted sequence length and percentage of included positions across taxa, missing data did not exceed 3% in any one alignment. Only 1.6% of sequence data were missing in the combined 4 protein-coding gene sequence analysis. All included amino acid sequences passed a chi-square test of compositional homogeneity at the 5% level, as implemented in Tree-Puzzle. However, some LSU rRNA gene sequences (Chrysochromulina, Cryptosporidium, Cyanidioschyzon, Mesostigma, Phaeocystis, and Prymnesium) failed compositional homogeneity tests.

Phylogenetic Analysis

Protein-coding gene alignments were analyzed at the amino acid level. ML analysis of amino acid sequences was performed using PROML in the PHYLIP package ver. 3.7 (Felsenstein 2004). A JTT + Γ + I model of protein evolution was applied with the user-defined hidden Markov model (HMM) option for modeling among-site rate variation. The rates and probabilities for the HMM were estimated from the Neighbor-Joining trees using Tree-Puzzle ver. 5.2 (Schmidt et al. 2002). For each ML tree search, the input order of sequences was randomized and the process was repeated 100 times with “global rearrangements.” Bootstrap values were obtained from 100 resamplings, each search with 1 round of random taxon addition followed by global rearrangements.

PAUP* 4.0b (Swofford 2002) was utilized for the ML analysis of SSU and LSU rRNA gene sequences. Modeltest ver. 3.7 (Posada and Crandall 1998) was used to find the best-fitting model of nucleotide evolution and to estimate substitution rates, base frequencies, Γ distribution parameter (α), and proportion of invariable sites. For each ML tree search, the input order of sequences was randomized, and the process was repeated 100 times with the tree bisection and reconnection branch-swapping algorithm. Bootstrap values were obtained from 100 resamplings, each search with 1 round of random taxon addition and the nearest-neighbor interchange branch-swapping algorithm.

Bayesian inference of phylogeny was performed using MrBayes ver. 3.1 (Huelsenbeck and Ronquist 2001). For DNA sequence analyses, the general time reversible + Γ + I model of evolution was applied. For protein sequence analyses, the Whelan and Goldman + Γ + I model of evolution was used. Preliminary Markov chain Monte Carlo (MCMC) runs with about 10,000 generations of trees were used to find the optimal temperature values, which seemed to be an important factor in chain mixing (data not shown). At least 2 independent MCMC runs were then completed and were compared to assess the reliability of each run. A total of 1,000,000–2,000,000 generations of trees were selected and evaluated, and every hundredth tree was sampled for further analysis. The burn-in period was evaluated using Gnuplot ver. 4.0 (Williams and Kelly 1998).

The approximately unbiased (AU) test was performed to compare 3 competing hypotheses related to the phylogenetic position of A. proboscidea relative to Opisthokonta and Amoebozoa. Tree topologies reflecting these hypotheses were generated by rearrangement of the ML tree for the data set (if required). Site likelihoods for each topology were calculated using Tree-Puzzle. The AU test was performed using CONSEL ver. 0.1h (Shimodaira and Hasegawa 2001). The output file from Tree-Puzzle was converted to a CONSEL-compatible format using a Python script kindly provided by Jessica Leigh (Department of Biochemistry and Molecular Biology at Dalhousie University).

Results

Evolutionary Relationships of A. proboscidea

The phylogenetic position of Apusomonas was well resolved in the ML analysis of the combined 4 protein-coding genes (fig. 1). The clade comprising Apusomonas and Opisthokonta was recovered, and strongly supported, with an ML bootstrap value of 99%. Amoebozoa was sister to the Apusomonadidae + Opisthokonta clade with 99% ML bootstrap support.

The same relationship between Apusomonas and Opisthokonta was found in the ML trees based on individual gene sequence analyses of α-tubulin and β-tubulin with 90% and 41% bootstrap values, respectively (see Supplementary Material online). The specific and strongly supported relationship between Apusomonas and Opisthokonta is still recovered if diplomonads, Carpediemonas, parabasalids, and Andalucia incarcerata are added to the analysis (these sequences are normally the closest relatives to Opisthokonta in α-tubulin phylogenies—data not shown). Monophyly of Opisthokonta was not recovered in the actin analysis, but Apusomonas weakly formed a clade with the opisthokont Amoebidium (27% ML bootstrap value).

In the analysis of Hsp90 gene sequences, both Amastigomonas and Apusomonas weakly allied with Stramenopiles (13% ML bootstrap value). In this case, neither an Opisthokonta–Apusomonadidae–Amoebozoa clade nor an Opisthokonta–Amoebozoa clade was found (see Supplementary Material online). When slightly longer-branched sequences were removed, Apusomonadidae instead formed a weak clade with Amoebozoa and Opisthokonta (for more information about sequences removed, see Supplementary Material online). Apusomonadidae were sister to Dictyostelium in the ML tree, making Amoebozoa paraphyletic; but this relationship was not significantly supported (17% ML bootstrap value).

In the combined SSU–LSU rRNA gene phylogeny, Opisthokonta, Apusomonadidae, and Amoebozoa formed a clade in both the ML and Bayesian analyses (fig. 2). Apusomonas branched weakly with the concatenated “amoebozoan” sequence (44% ML bootstrap value). However, if additional (long-branched) amoebozoan sequences were included, Apusomonas formed a very weak clade with Opisthokonta instead (data not shown). SSU rRNA gene analysis also suggested that Apusomonadidae were related to the naked lobose amoebozoan Vexillifera but without significant support. In the LSU rRNA gene phylogenies, Apusomonas weakly branched within unresolved clades of biflagellates, not closely related to the included amoebozoan (Mastigamoeba).

Bayesian analysis of the 6 combined gene sequences indicated that Apusomonas is closely related to Opisthokonta, with Amoebozoa falling as the sister group to Apusomonadidae + Opisthokonta (fig. 4). Most deep divergences including these clades were resolved with Bayesian posterior probabilities of 1. The more conservative ML bootstrapping approach could not be applied to this mixed amino acid/nucleotide data set.

Our data were used to compare the following 3 hypotheses. Hypothesis I is that Apusomonadidae is sister to Opisthokonta (the hypothesis most strongly supported by the analyses described above). Hypothesis II is that Apusomonadidae and Amoebozoa are sisters (supported by some of our phylogenies). Hypothesis III is that Opisthokonta and Amoebozoa are sisters to the exclusion of Apusomonadidae (consistent with a unikont clade). In the AU test based on the combined 4 proteins, hypotheses II and III were rejected (P = 2 × 10−4 and 4 × 10−4, respectively) (see Supplementary Material online). The AU test based on α-tubulin alone strongly rejected hypotheses II and III (both P = 1 × 10−7). However, the analysis of the combined β-tubulin, actin, and Hsp90 gene sequences did not reject hypotheses II and III (P = 0.539 and 0.472, respectively). These results suggest that the strong rejection signals in the combined protein-coding gene analysis (fig. 1) mostly originated from the α-tubulin data set. AU tests based on the combined 6 gene sequences also rejected hypotheses II and III (P = 2 × 10−4 and 4 × 10−5, respectively) (see Supplementary Material online).

Evolutionary Relationships of Other Protist Groups

Our combined protein-coding gene sequence analyses found moderate to strong support (88–95%) for a sister relationship between Alveolata and Stramenopiles. Close affinity of Kathablepharidae to Cryptophyceae was recovered in both SSU and LSU gene phylogenies (see Supplementary Material online). The combined SSU and LSU gene phylogeny recovered a Kathablepharidae–Cryptophyceae clade with 100% ML bootstrap support (fig. 2). Interestingly, none of the protein phylogenies suggested monophyly of Kathablepharidae + Cryptophyceae, and the position of Kathablepharidae was not resolved with over 50% ML bootstrap support in any of the protein-gene phylogenies, although the 6-gene analysis placed it in a clade with Cryptophyceae (fig. 4). Two genes—SSU and Hsp90—supported a sister relationship between Kathablepharidae + Cryptophyceae and Glaucophyta, although ML bootstrap was moderate or weak (see Supplementary Material online).

Discussion

The Evolutionary Position of Apusomonadidae

With the goal of clarifying the evolutionary position of Apusomonadidae, we determined 5 new Apusomonas gene sequences—those for LSU rRNA, α-tubulin, β-tubulin, actin, and Hsp90, the first 4 of which are new to Apusomonadidae. Most individual gene analyses suggested that Apusomonadidae is closely related to Opisthokonta or Amoebozoa. α-tubulin, β-tubulin, and actin genes separately provided evidence of an Opisthokonta + Apusomonadidae clade, assuming that the eukaryotic root does not lie between the 2 groups. Hsp90 genes, when analyzed without including some long-branched sequences, supported an Apusomonadidae plus Amoebozoa clade. However, the position of Apusomonadidae based on Hsp90 gene sequences was sensitive to taxon sampling, and Apusomonadidae sometimes appeared as sister to biflagellate clades, as found by Stechmann and Cavalier-Smith (2003a). Our LSU rRNA gene phylogenies likewise did not support the affinity of Apusomonadidae to any particular group, and the position of Apusomonadidae was sensitive to taxon sampling.

The position of Apusomonadidae in SSU rRNA gene phylogenies has been unstable. In most cases, inferred topologies were poorly supported except for the earlier study by Cavalier-Smith and Chao (1995), in which a sister relationship between Apusomonadidae and Opisthokonta was relatively well supported in maximum parsimony analyses (84%) and least-squares analyses of Jukes–Cantor distances (96%). However, in subsequent studies, support for the Apusomonadidae–Opisthokonta clade was weaker (e.g., Cavalier-Smith and Chao 2003b) or alternative topologies were inferred (e.g., Berney et al. 2004; Cavalier-Smith et al. 2004, Figure 2).

When α-tubulin gene sequences were included, our analyses strongly supported an Apusomonadidae–Opisthokonta clade, and AU tests rejected the alternative hypotheses of an Apusomonadidae–Amoebozoa clade or an Opisthokonta–Amoebozoa clade. When α-tubulin sequences were excluded, an Apusomonadidae–Opisthokonta clade was recovered in the ML tree (fig. 3), but 2 alternative hypotheses—an Apusomonadidae–Amoebozoa clade and an Opisthokonta–Amoebozoa clade (with Apusomonadidae as their sister group)—could not be rejected. None of the individual or combined gene data sets analyzed in this study rejected the Apusomonadidae–Opisthokonta clade when AU tests were applied. Overall, an Apusomonadidae–Opisthokonta clade is the best-supported hypothesis.

FIG. 3.—

ML tree inferred from the same data set used in figure 1 without α-tubulin gene sequences. The same position for Apusomonas (hypothesis I) was recovered. Statistical support values for hypothesis II and III are also indicated.

FIG. 3.—

ML tree inferred from the same data set used in figure 1 without α-tubulin gene sequences. The same position for Apusomonas (hypothesis I) was recovered. Statistical support values for hypothesis II and III are also indicated.

FIG. 4.—

Bayesian consensus tree based on 6 genes. A general time reversible + Γ + I model (8 rate categories) of nucleotide evolution was applied to the concatenated SSU and LSU rRNA gene sequences, and a Whelan and Goldman + Γ + I model (8 rate categories) of protein evolution was applied to the combined 4 protein-coding genes. Bayesian posterior probabilities are shown at the corresponding nodes.

FIG. 4.—

Bayesian consensus tree based on 6 genes. A general time reversible + Γ + I model (8 rate categories) of nucleotide evolution was applied to the concatenated SSU and LSU rRNA gene sequences, and a Whelan and Goldman + Γ + I model (8 rate categories) of protein evolution was applied to the combined 4 protein-coding genes. Bayesian posterior probabilities are shown at the corresponding nodes.

Apusomonadidae is not likely to branch within Opisthokonta because members of Opisthokonta included in our study formed a strong clade (figs. 1–323). In addition, Steenkamp et al. (2005) reported that Apusomonadidae lack an amino acid insertion in elongation factor 1-α, a synapomorphic character for Opisthokonta.

Although Hampl et al. (2005) and Simpson et al. (2006) noted that tubulin gene (α-tubulin in particular) phylogenies conflicted with other gene trees with respect to the positions of some excavates, we did not find any significant conflict between the α-tubulin tree and any of the other 5 gene trees with respect to the position of Apusomonadidae. Determination of the utility of α-tubulin or any other genes as phylogenetic markers for analysis of deep eukaryote divergences would require the comparative analysis of many gene sequences from diverse eukaryotic groups.

Some components of our study suggested a close relationship between Amoebozoa and Apusomonadidae. Amoebozoa includes diverse unicellular and multicellular eukaryotic organisms for which the only morphological commonality is amoeboid movement (Walochnik et al. 2004) (the nonamoeboid flagellates Phalansterium and Multicilia being likely exceptions [Nikolaev et al. 2006]). Several molecular analyses have suggested monophyly of Amoebozoa (see Fahrni et al. 2003 for review), however, without robust statistical support, except when taxon sampling was low (Bapteste et al. 2002). Although there are several major subclades in Amoebozoa, so far a significant amount of genomic data have only been obtained for a few amoebozoan taxa, which, like Entamoeba, are typically long branched. This is the reason for the limited taxon sampling for Amoebozoa in our study. Additional sequence data for other amoebozoan taxa, such as Phalansterium, Tubulinea, and Fabellinea, are needed to further explore the relatedness of Apusomonadidae to Amoebozoa.

Richards and Cavalier-Smith (2005) suggested that Opisthokonta and Amoebozoa (unikonts) form a monophyletic group based on 5 putative synapomorphies concerning myosin gene types and sequence features. However, this study did not include several important eukaryotic taxa, such as Rhizaria and Apusomonadidae. Because our multigene analyses place Apusomonadidae cladistically within the unikont clade, testing for the presence of unikont-specific myosin genes and indels in Apusomonadidae would be particularly valuable. Likewise, determination of whether Apusomonadidae undergo flagellar transformation and which flagellum is mature would be useful because flagellar transformation (in which the posterior flagellum is the mature one) is proposed to be a key character distinguishing bikonts from unikonts (Cavalier-Smith 2003).

Cavalier-Smith and Chao (2003b) united apusomonads with the obscure flagellate Ancyromonas in a more inclusive taxon Apusozoa, based on the presence of a thecaelike layer on the dorsal surface of Ancyromonas and a weak affinity between the groups in some SSU rRNA phylogenies. This relationship requires further confirmation and once more data are available from Ancyromonas.

Implications of an “Apusomonadidae–Opisthokonta” Clade

The 2 most likely relationships of Apusomonadidae suggested in our study—the Apusomonadidae–Opisthokonta clade and the Apusomonadidae–Amoebozoa clade—conflict with 2 existing hypotheses related to deep eukaryotic divergences. The first of these hypotheses is that Opisthokonta and Amoebozoa shared a common unikont ancestor (i.e., with a single flagellum and 1 basal body), whereas other eukaryotic groups were ancestrally bikont (having 2 flagella and 2 basal bodies) (Cavalier-Smith 2002). The second widely cited hypothesis is that bikonts—including Apusomonadidae—are monophyletic, based primarily on the presence or absence of the DHFR–TS gene fusion (Stechmann and Cavalier-Smith 2002). As explained below, neither hypothesis is well supported by existing morphological/genomic data, and both are contraindicated by our results.

Morphological and Our Molecular Phylogenetic Data Conflict with a Hypothesis of Unikont Ancestry for Opisthokonta and/or Amoebozoa

Though flagellate cells of Opisthokonta (those of choanoflagellates, Chytridiomycetes, certain Ichthyosporea [=Mesomycetozoea], and male gametes of animals) possess a single flagellum, such cells typically have a nonflagellated second basal body (Barr 1981; Karpov and Leadbeater 1997), consistent with biflagellate ancestry. The ancestral flagellate condition of Amoebozoa is more ambiguous because flagellate amoebozoans are relatively uncommon, and evolutionary relationships among the major subclades of Amoebozoa are poorly understood (Fahrni et al. 2003; Smirnov et al. 2005). On the one hand, flagellate members of pelobionts, some Protostelia, Phalansterium, and Multicilia possess a single flagellum and 1 basal body per kinetid (a unit consisting of 1 or more flagellar basal bodies and any associated fibers, roots, and cytoskeleton). On the other hand, those of Myxogastria and some Protostelia have 2 basal bodies and usually 2 flagella (Olive 1975). Spiegel (1981) suggested that the common ancestor of Protostelia likely possessed 2 basal bodies and that some members of Protostelia had lost their second basal body. Hence, the unikont hypothesis is not particularly well supported by the available morphological data, even before considering the position of apusomonads.

Because the apusomonads have 2 basal bodies and 2 flagella, their possible positioning within the Opisthokonta–Amoebozoa clade makes a biflagellate common ancestor for this clade more parsimonious. Our molecular study therefore further weakens the hypothesis of unikont ancestry for Opisthokonta and Amoebozoa (fig. 5). In view of available morphological and molecular evidence, categorizing the minimum Opisthokonta + Amoebozoa clade as unikonts seems unjustified on present data.

FIG. 5.—

A simplified unrooted tree showing the main conclusion of this study. Regardless of the position of the eukaryote root, unikonts do not form a monophyletic group.

FIG. 5.—

A simplified unrooted tree showing the main conclusion of this study. Regardless of the position of the eukaryote root, unikonts do not form a monophyletic group.

Conflict between Our Molecular Phylogenetic Results and Interpretations of the DHFR–TS Gene Fusion Data

Philippe et al. (2000) and Stechmann and Cavalier-Smith (2002) proposed that possession of 2 separate, monofunctional DHFR and TS (thyA) genes was the archaic condition for eukaryotes, because bacteria, when they possess these 2 genes, also produce 2 separate proteins. Based on this premise, Stechmann and Cavalier-Smith (2002) used the presence or absence of gene fusion between DHFR and TS (thyA) to help infer the position of the eukaryotic root. Stechmann and Cavalier-Smith (2002, 2003b) did not find the DHFR–TS fusion gene in Opisthokonta and Amoebozoa but noted that studied representatives of Alveolata, Apusomonadidae, Euglenozoa, Rhizaria, Stramenopiles, and Viridiplantae have bifunctional DHFR–TS fusion genes. Therefore, these authors suggested that eukaryotes with this “derived” gene fusion form a monophyletic group (bikonts), within which the eukaryote root cannot lie. However, this concept is questionable for several reasons.

First, as Embley and Martin (2006) noted DHFR–TS fusion data are currently available for relatively few taxonomic groups and cannot be used to infer positions of lineages such as diplomonads and parabasalids that are devoid of these genes. The assumption that the DHFR–TS gene fusion represents the derived condition in eukaryotes is another issue. Because multiple cases of replacement and separation of DHFR and TS genes have been documented, particularly in bacteria (Philip et al. 2005), separate positioning of DHFR and TS (thyA) genes may not represent the archaic condition in eukaryotes. Alternatively, the DHFR–TS fusion gene could be viewed as the archaic condition in eukaryotes because the DHFR–TS fusion gene is transcribed into a single mRNA molecule like bacterial DHFR and TS (thyA) genes, which usually occur in a single operon (although the gene order is reversed). In contrast, DHFR and TS (thyA) genes are separately transcribed in Opisthokonta and the amoebozoan Hartmannella (Stechmann and Cavalier-Smith 2003b). These separate DHFR and TS (thyA) genes may have been derived from refission of the fused gene. Although Stechmann and Cavalier-Smith (2002) suggested that the reversal of the DHFR–TS gene fusion is improbable, several examples of fusion and refission of other genes over evolutionary time have now been documented in eukaryotes (Arisue et al. 2005; Krauss et al. 2006; Waller et al. 2006). Phylogenetic analyses of the DHFR or TS (thyA) gene sequences also did not positively support a hypothesis of recent common origin of the fused genes (Lazar et al. 1993; Schlichtherle et al. 1996). In addition, the DHFR and TS genes may have been subjected to multiple lateral gene transfer events. For example, the amoebozoans Dictyostelium (Leduc et al. 2004) and Physarum have apparently replaced their TS (thyA) genes with nonhomologous TS (thyX) genes. Lastly, there is a lack of strong independent evidence for the bikont clade supposedly identified by DHFR–TS fusion. Stechmann and Cavalier-Smith (2002) propose that the presence of flagellar transformation is a second synapomorphy for the bikont clade (see also Cavalier-Smith 2002); however, this idea is complicated by the unambiguous presence of a form of flagellar transformation in the biflagellate unikont Physarum (Wright et al. 1980). Collectively, these considerations suggest that the proposal that the DHFR–TS gene fusion represents a single derived evolutionary event within the diversification of extant eukaryotes is questionable.

Our study, which supports an Apusomonadidae + Opisthokonta clade, places a fusion-bearing taxon within the only non–fusion-bearing clade of eukaryotes. If our phylogenetic placement of Apusomonadidae is correct, this implies 1 of the 2 possibilities: 1) The DHFR–TS fusion was laterally transferred at least once (or that the fusion event occurred more than once) and, hence, is an unreliable phylogenetic marker, or 2) The DHFR–TS fusion represents an unique evolutionary event, but this took place before the divergence of extant eukaryotes and, hence, is an ancestral character state (plesiomorphy) for all living eukaryotes. Therefore, our study adds substantial additional doubt to the validity of the DHFR–TS fusion as a marker for deep eukaryote diversification and the monophyly of the group identified by the fusion (bikonts).

Evolutionary Relationships of Other Groups

Our new sequences for representatives of Alveolata, Cryptophyceae, Glaucophyta, Haptophyta, Stramenopiles, Kathablepharidae, Rhizaria, and Viridiplantae allowed us to evaluate additional relationships among major eukaryotic lineages. Our LSU rRNA gene phylogeny confirmed the previous result of SSU rRNA phylogeny that Kathablepharidae and Cryptophyceae are sister taxa (Okamoto and Inouye 2005). Alveolata and Stramenopiles were sisters in multiple-protein–gene phylogenies in our analyses (figs. 1 and 3), consistent with previous multiprotein analyses (Baldauf et al. 2000; Harper et al. 2005; Simpson et al. 2006). In the combined SSU and LSU rRNA phylogeny (fig. 2), however, Alveolata branched weakly with Rhizaria, which was also observed in some previous SSU rRNA gene analyses (e.g., Cavalier-Smith and Chao 2003a, Figures 1 and 2).

None of our analyses suggested the monophyly of “chromalveolates,” including Alveolata, Cryptophyceae (+Kathablepharidae), Haptophyta, and Stramenopiles (Cavalier-Smith 1999), particularly with respect to the Kathablepharidae–Cryptophyceae clade. The “chromalveolate hypothesis,” advocating a single red algal plastid origin for chromalveolates and the monophyly of these groups, is currently hotly debated (Falkowski et al. 2004; Grzebyk et al. 2004; Keeling et al. 2004). Because our individual gene analyses of SSU rRNA and Hsp90 as well as previous Hsp70 phylogeny (Rensing et al. 1997) suggested, albeit without strong support, close affinity of the Kathablepharidae–Cryptophyceae clade (or Cryptophyceae) to Glaucophyta, additional genomic data from Kathablepharidae and Cryptophyceae would be useful to further evaluate their phylogenetic relationship to other putative chromalveolate groups.

Supplementary Material

Supplementary figures S1–S6 representing ML trees based on analyses of individual gene sequences, supplementary tables S1 and S2 showing GenBank accession numbers of newly obtained sequences in this study and the AU test results, respectively, and Method S1, which includes the supplementary method information, are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).

This research was supported by grant MCB-9977903 from the National Science Foundation, a Davis Summer Research Fellowship (Department of Botany at the University of Wisconsin–Madison), an Anna Grant Birge Memorial Award (University of Wisconsin–Madison), and Natural Sciences and Engineering Research Council grant 298366-04 to A.G.B.S. Y. Van de Peer at Ghent University kindly provided LSU rRNA sequence alignments. The authors also thank B. Larget (University of Wisconsin–Madison) for access to a computation facility, J. Graham (University of Wisconsin–Madison) for the P. limbatum culture, L. Wilcox (University of Wisconsin–Madison) for obtaining samples from aquatic habitats, and J. Leigh (Dalhousie University) for a Python script.

References

Adl
SM
Simpson
GB
Farmer
MA
, et al.  , 
(28 co-authors)
The new higher level classification of eukaryotes with emphasis on the taxonomy of protists
J Eukaryot Microbiol
 , 
2005
, vol. 
52
 (pg. 
399
-
451
)
Arisue
N
Hasegawa
M
Hashimoto
T
Root of the Eukaryota tree as inferred from combined likelihood analyses of multiple molecular sequence data
Mol Biol Evol
 , 
2005
, vol. 
22
 (pg. 
409
-
420
)
Atkins
MS
McArthur
AG
Teske
AP
Ancyromonadida: a new phylogenetic lineage among the protozoa closely related to the common ancestor of metazoans, fungi, and choanoflagellates (Opisthokonta)
J Mol Evol
 , 
2000
, vol. 
51
 (pg. 
278
-
285
)
Baldauf
SL
Roger
AJ
Wenk-Siefert
I
Doolittle
WF
A kingdom-level phylogeny of eukaryotes based on combined protein data
Science
 , 
2000
, vol. 
290
 (pg. 
972
-
977
)
Bapteste
E
Brinkmann
H
Lee
JA
, et al.  , 
(11 co-authors)
The analysis of 100 genes supports the grouping of three highly divergent amoebae; Dictyostelium, Entamoeba, and Mastigamoeba
Proc Natl Acad Sci USA
 , 
2002
, vol. 
99
 (pg. 
1414
-
1419
)
Barr
DJS
The phylogenetic and taxonomic implications of flagellar rootlet morphology among zoosporic fungi
Biosystems
 , 
1981
, vol. 
14
 (pg. 
359
-
370
)
Ben Ali
A
De Baere
R
Van der Auwera
G
De Wachter
R
Van de Peer
Y
Phylogenetic relationships among algae based on complete large-subunit rRNA sequences
Int J Syst Evol Microbiol
 , 
2001
, vol. 
51
 (pg. 
737
-
749
)
Berney
C
Fahrni
J
Pawlowski
J
How many novel eukaryotic ‘kingdoms’? Pitfalls and limitations of environmental DNA surveys
BMC Biol
 , 
2004
, vol. 
2
 pg. 
13
 
Cavalier-Smith
T
Principles of protein and lipid targeting in secondary symbiogenesis: euglenoid, dinoflagellate, and sporozoan plastid origins and the eukaryote family tree
J Eukaryot Microbiol
 , 
1999
, vol. 
46
 (pg. 
347
-
366
)
Cavalier-Smith
T
The phagotrophic origin of eukaryotes and phylogenetic classification of Protozoa
Int J Syst Evol Microbiol
 , 
2002
, vol. 
52
 (pg. 
297
-
354
)
Cavalier-Smith
T
Protist phylogeny and the high-level classification of Protozoa
Eur J Protistol
 , 
2003
, vol. 
39
 (pg. 
338
-
348
)
Cavalier-Smith
T
Chao
EE
The opalozoan Apusomonas is related to the common ancestor of animals, fungi, and choanoflagellates
Proc R Soc Lond B Biol Sci
 , 
1995
, vol. 
261
 (pg. 
1
-
6
)
Cavalier-Smith
T
Chao
EE
Molecular phylogeny of centrohelid Heliozoa, a novel lineage of bikont eukaryotes that arose by ciliary loss
J Mol Evol
 , 
2003
, vol. 
56
 (pg. 
387
-
396
)
Cavalier-Smith
T
Chao
EE
Phylogeny of Choanozoa, Apusozoa, and other Protozoa and early eukaryote megaevolution
J Mol Evol
 , 
2003
, vol. 
56
 (pg. 
540
-
563
)
Cavalier-Smith
T
Chao
EE
Oates
B
Molecular phylogeny of Amoebozoa and the evolutionary significance of the unikont Phalansterium
Eur J Protistol
 , 
2004
, vol. 
40
 (pg. 
21
-
48
)
Ekelund
F
Patterson
DJ
Some heterotrophic flagellates from a cultivated garden soil in Australia
Arch Protistenkd
 , 
1997
, vol. 
148
 (pg. 
461
-
478
)
Embley
TM
Martin
W
Eukaryotic evolution, changes and challenges
Nature
 , 
2006
, vol. 
440
 (pg. 
623
-
630
)
Fahrni
JF
Bolivar
I
Berney
C
Nassonova
E
Smirnov
A
Pawlowski
J
Phylogeny of lobose amoebae based on actin and small-subunit ribosomal RNA genes
Mol Biol Evol
 , 
2003
, vol. 
20
 (pg. 
1881
-
1886
)
Falkowski
PG
Katz
ME
Knoll
AH
Quigg
A
Raven
JA
Schofield
O
Taylor
FJR
The evolution of modern eukaryotic phytoplankton
Science
 , 
2004
, vol. 
305
 (pg. 
354
-
360
)
Felsenstein
J
PHYLIP (phylogeny inference package) version 3.6
Distributed by the author
 , 
2004
Seattle (WA)
Department of Genome Sciences, University of Washington
Grzebyk
D
Katz
ME
Knoll
AH
Quigg
A
Raven
JA
Schofield
O
Taylor
FJR
Falkowski
PG
Response to comment on “The evolution of modern eukaryotic phytoplankton”
Science
 , 
2004
, vol. 
306
 pg. 
2191c
 
Hampl
V
Horner
DS
Dyal
P
Kulda
J
Flegr
J
Foster
PG
Embley
TM
Inference of the phylogenetic position of oxymonads based on nine genes: support for Metamonada and Excavata
Mol Biol Evol
 , 
2005
, vol. 
22
 (pg. 
2508
-
2518
)
Harper
JT
Waanders
E
Keeling
PJ
On the monophyly of chromalveolates using a six-protein phylogeny of eukaryotes
Int J Syst Evol Microbiol
 , 
2005
, vol. 
55
 (pg. 
487
-
496
)
Huelsenbeck
JP
Ronquist
F
MrBayes: Bayesian inference of phylogenetic trees
Bioinformatics
 , 
2001
, vol. 
17
 (pg. 
754
-
755
)
Karpov
SA
Leadbeater
BSC
Cell and nuclear division in a freshwater choanoflagellate, Monosiga ovata Kent
Eur J Protistol
 , 
1997
, vol. 
33
 (pg. 
323
-
334
)
Karpoff
SA
Zhukov
BF
Ultrastructure and taxonomic position of Apusomonas proboscidea Alexeieff
Arch Protistenkd
 , 
1986
, vol. 
131
 (pg. 
13
-
26
)
Keeling
PJ
Archibald
JM
Fast
NM
Palmer
JD
Comment on “The evolution of modern eukaryotic phytoplankton”
Science
 , 
2004
, vol. 
306
 pg. 
2191b
 
Keeling
PJ
Burger
G
Durnford
DG
Lang
BF
Lee
RW
Pearlman
RE
Roger
AJ
Gray
MW
The tree of eukaryotes
Trends Ecol Evol
 , 
2005
, vol. 
20
 (pg. 
670
-
676
)
Krauss
V
Fassl
A
Fiebig
P
Patties
I
Sass
H
The evolution of the histone methyltransferase gene Su(var)3-9 in metazoans includes a fusion with and a re-fission from a functionally unrelated gene
BMC Evol Biol
 , 
2006
, vol. 
6
 pg. 
18
 
Lazar
G
Zhang
H
Goodman
HM
The origin of the bifunctional dihydrofolate reductase-thymidylate synthase isogenes of Arabidopsis thaliana
Plant J
 , 
1993
, vol. 
3
 (pg. 
657
-
668
)
Leduc
D
Graziani
S
Lipowski
G
Marchand
C
Le Maréchal
P
Liebl
U
Myllykallio
H
Functional evidence for active site location of tetrameric thymidylate synthase X at the interphase of three monomers
Proc Natl Acad Sci USA
 , 
2004
, vol. 
101
 (pg. 
7252
-
7257
)
Maddison
DR
Maddison
WP
MacClade 4: analysis of phylogeny and character evolution. Version 4.03
2001
Sunderland (MA)
Sinauer Associates
Molina
FI
Nerad
TA
Ultrastructure of Amastigomonas bermudensis ATCC 50234 sp. nov
Eur J Protistol
 , 
1991
, vol. 
27
 (pg. 
386
-
396
)
Nikolaev
SI
Berney
C
Fahrni
JF
Bolivar
I
Polet
S
Mylnikov
AP
Aleshin
VV
Petrov
NB
Pawlowski
J
The twilight of Heliozoa and rise of Rhizaria, an emerging supergroup of amoeboid eukaryotes
Proc Natl Acad Sci USA
 , 
2004
, vol. 
101
 (pg. 
8066
-
8071
)
Nikolaev
SI
Berney
C
Petrov
NB
Mylnikov
AP
Fahrni
JF
Pawlowski
J
Phylogenetic position of Multicilia marina and the evolution of Amoebozoa
Int J Syst Evol Microbiol
 , 
2006
, vol. 
56
 (pg. 
1449
-
1458
)
Okamoto
N
Inouye
I
The katablepharids are a distant sister group of the Cryptophyta: a proposal for Katablepharidophyta divisio nova/Kathablepharida phylum novum based on SSU rDNA and beta-tubulin phylogeny
Protist
 , 
2005
, vol. 
156
 (pg. 
163
-
179
)
Olive
LS
The mycetozoans
1975
New York
Academic Press
Philip
GK
Creevey
CJ
McInerney
JO
The Opisthokonta and the Ecdysozoa may not be clades: stronger support for the grouping of plant and animal than for animal and fungi and stronger support for the Coelomata than Ecdysozoa
Mol Biol Evol
 , 
2005
, vol. 
22
 (pg. 
1175
-
1184
)
Philippe
H
Opinion: long branch attraction and protist phylogeny
Protist
 , 
2000
, vol. 
151
 (pg. 
307
-
316
)
Philippe
H
Lopez
P
Brinkmann
H
Budin
K
Germot
A
Laurent
J
Moreira
D
Muller
M
Le Guyader
H
Early-branching or fast-evolving eukaryotes? An answer based on slowly evolving positions
Proc R Soc Lond B Biol Sci
 , 
2000
, vol. 
267
 (pg. 
1213
-
1221
)
Posada
D
Crandall
KA
Modeltest: testing the model of DNA substitution
Bioinformatics
 , 
1998
, vol. 
14
 (pg. 
817
-
818
)
Rensing
SA
Obrdik
P
Rober-Kleber
N
Müller
SB
Hofman
CJB
Van de Peer
Y
Maier
U-G
Molecular phylogeny of the stress-70 protein family with reference to algal relationships
Eur J Phycol
 , 
1997
, vol. 
32
 (pg. 
279
-
285
)
Richards
TA
Cavalier-Smith
T
Myosin domain evolution and the primary divergence of eukaryotes
Nature
 , 
2005
, vol. 
436
 (pg. 
1113
-
1118
)
Schlichtherle
IM
Ross
DS
Van Houten
JL
Cloning and molecular analysis of the bifunctional dihydrofolate reductase-thymidylate synthase gene in the ciliated protozoan Paramecium tetraurelia
Mol Gen Genet
 , 
1996
, vol. 
250
 (pg. 
665
-
673
)
Schmidt
HA
Strimmer
K
Vingron
M
Von Haeseler
A
TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets and parallel computing
Bioinformatics
 , 
2002
, vol. 
18
 (pg. 
502
-
504
)
Shimodaira
H
Hasegawa
M
CONSEL: for assessing the confidence of phylogenetic tree selection
Bioinformatics
 , 
2001
, vol. 
17
 (pg. 
1246
-
1247
)
Simpson
AGB
Inagaki
Y
Roger
AJ
Comprehensive multigene phylogenies of excavate protists reveal the evolutionary positions of “primitive” eukaryotes
Mol Biol Evol
 , 
2006
, vol. 
23
 (pg. 
615
-
625
)
Simpson
AGB
Lukes
J
Roger
AJ
The evolutionary history of kinetoplastids and their kinetoplasts
Mol Biol Evol
 , 
2002
, vol. 
19
 (pg. 
2071
-
2083
)
Simpson
AGB
Roger
AJ
The real ‘kingdoms’ of eukaryotes
Curr Biol
 , 
2004
, vol. 
14
 (pg. 
R693
-
R696
)
Simpson
AGB
Roger
AJ
Silberman
JD
Leipe
DD
Edgcomb
VP
Jermiin
LS
Patterson
DJ
Sogin
ML
Evolutionary history of “early-diverging” eukaryotes: the excavate taxon Carpediemonas is a close relative of Giardia
Mol Biol Evol
 , 
2002
, vol. 
19
 (pg. 
1782
-
1791
)
Smirnov
A
Nassonova
E
Berney
C
Fahrni
J
Bolivar
I
Pawlowski
J
Molecular phylogeny and classification of the lobose amoebae
Protist
 , 
2005
, vol. 
156
 (pg. 
129
-
142
)
Spiegel
FW
Phylogenetic significance of the flagellar apparatus in protostelids (Eumycetozoa)
Biosystems
 , 
1981
, vol. 
14
 (pg. 
491
-
499
)
Stechmann
A
Cavalier-Smith
T
Rooting the eukaryote tree by using a derived gene fusion
Science
 , 
2002
, vol. 
297
 (pg. 
89
-
91
)
Stechmann
A
Cavalier-Smith
T
Phylogenetic analysis of eukaryotes using heat-shock protein Hsp90
J Mol Evol
 , 
2003
, vol. 
57
 (pg. 
408
-
419
)
Stechmann
A
Cavalier-Smith
T
The root of the eukaryote tree pinpointed
Curr Biol
 , 
2003
, vol. 
13
 (pg. 
R665
-
R666
)
Steenkamp
ET
Wright
J
Baldauf
SL
The protistan origins of animals and fungi
Mol Biol Evol
 , 
2006
, vol. 
23
 (pg. 
93
-
106
)
Stein
JR
Handbook of phycological methods: culture methods and growth measurements
1973
London
Cambridge University Press
Swofford
DL
PAUP*. Phylogenetic analysis using parsimony (*and other methods). Version 4
2002
Sunderland (MA)
Sinauer Associates
Thompson
JD
Gibson
TJ
Plewniak
F
Jeanmougin
F
Higgins
DG
The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools
Nucleic Acids Res
 , 
1997
, vol. 
24
 (pg. 
4876
-
4882
)
Vickerman
K
Darbyshire
JF
Ogden
CG
Apusomonas proboscidea Aléxéieff 1924, an unusual phagotrophic flagellate from soil
Arch Protistenkd
 , 
1974
, vol. 
116
 (pg. 
254
-
269
)
Waller
RF
Slamovits
CH
Keeling
PJ
Lateral gene transfer of a multigene region from cyanobacteria to dinoflagellates resulting in a novel plastid-targeted fusion protein
Mol Biol Evol
 , 
2006
, vol. 
23
 (pg. 
1437
-
1443
)
Walochnik
J
Michel
R
Aspöck
H
A molecular biological approach to the phylogenetic position of the genus Hyperamoeba
J Eukaryot Microbiol
 , 
2004
, vol. 
51
 (pg. 
433
-
440
)
Williams
T
Kelly
C
Gnuplot: an interactive plotting program
1998
Hanover (Germany)
Dartmouth College
Wright
M
Moisand
A
Mir
L
Centriole maturation in the amoebae of Physarum polycephalum
Protoplasma
 , 
1980
, vol. 
105
 (pg. 
149
-
160
)
Wuyts
J
Van de Peer
Y
Winkelmans
T
De Wachter
R
The European database on small subunit ribosomal RNA
Nucleic Acids Res
 , 
2002
, vol. 
30
 (pg. 
183
-
185
)

Author notes

William Martin, Associate Editor