Evolutionary Relationships Among ‘‘Jakobid’’ Flagellates as Indicated by
Alpha- and Beta-Tubulin Phylogenies
Virginia P. Edgcomb,* Andrew J. Roger,*1 Alastair G. B. Simpson,† David T. Kysela,* and
Mitchell L. Sogin*
*Josephine Bay-Paul Center for Comparative Molecular Biology and Evolution, Marine Biological Laboratory, Woods Hole,
Massachusetts; and †School of Biological Sciences, University of Sydney, Sydney, New South Wales, Australia
Jakobids are free-living, heterotrophic flagellates that might represent early-diverging mitochondrial protists. They
share ultrastructural similarities with eukaryotes that occupy basal positions in molecular phylogenies, and their
mitochondrial genome architecture is eubacterial-like, suggesting a close affinity with the ancestral alpha-proteobacterial symbiont that gave rise to mitochondria and hydrogenosomes. To elucidate relationships among jakobids
and other early-diverging eukaryotic lineages, we characterized alpha- and beta-tubulin genes from four jakobids:
Jakoba libera, Jakoba incarcerata, Reclinomonas americana (the ‘‘core jakobids’’), and Malawimonas jakobiformis.
These are the first reports of nuclear genes from these organisms. Phylogenies based on alpha-, beta-, and combined
alpha- plus beta-tubulin protein data sets do not support the monophyly of the jakobids. While beta-tubulin and
combined alpha- plus beta-tubulin phylogenies showed a sister group relationship between J. libera and R. americana, the two other jakobids, M. jakobiformis and J. incarcerata, had unclear affinities. In all three analyses, J.
libera, R. americana, and M. jakobiformis emerged from within a well-supported large ‘‘plant-protist’’ clade that
included plants, green algae, cryptophytes, stramenopiles, alveolates, Euglenozoa, Heterolobosea, and several other
protist groups, but not animals, fungi, microsporidia, parabasalids, or diplomonads. A preferred branching order
within the plant-protist clade was not identified, but there was a tendency for the J. libera–R. americana lineage to
group with a clade made up of the heteroloboseid amoeboflagellates and euglenozoan protists. Jakoba incarcerata
branched within the plant-protist clade in the beta- and the combined alpha- plus beta-tubulin phylogenies. In alphatubulin trees, J. incarcerata occupied an unresolved position, weakly grouping with the animal/fungal/microsporidian group or with amitochondriate parabasalid and diplomonad lineages, depending on the phylogenetic method
employed. Tubulin gene phylogenies were in general agreement with mitochondrial gene phylogenies and ultrastructural data in indicating that the ‘‘jakobids’’ may be polyphyletic. Relationships with the putatively deep-branching
amitochondriate diplomonads remain uncertain.
Introduction
The ‘‘jakobids’’ are a collection of free-living, biflagellate protists that possess a conspicuous ventral
groove used for suspension feeding (O’Kelly 1993).
Currently, this group encompasses three named groups
given the rank of family: Histionidae (Histiona, Voight
1901; Reclinomonas, Flavin and Nerad 1993), Jakobidae
(Jakoba, Patterson 1990), and Malawimonadidae (Malawimonas, O’Kelly and Nerad 1999). Until recently,
attempts to understand the evolutionary affinities of the
jakobid flagellates have relied on ultrastructural data.
These data indicate that the jakobids are most similar to
an interesting collection of other flagellate groups that
possess suspension feeding grooves. These groups include diplomonads, retortamonads, heteroloboseids, and
the recently studied taxa Trimastix and Carpediemonas
(O’Kelly 1993, 1997; O’Kelly, Farmer, and Nerad 1999;
O’Kelly and Nerad 1999; Simpson and Patterson 1999;
Simpson, Bernard, and Patterson 2000). Due to their
common possession of a suspension feeding groove,
these organisms, together with jakobids, have been in1 Present address: Department of Biochemistry and Molecular Biology, Dalhousie University, Halifax, Nova Scotia, Canada.
Key words: jakobid, alpha-tubulin, beta-tubulin, phylogeny,
flagellates.
Address for correspondence and reprints: Mitchell L. Sogin, Josephine Bay-Paul Center of Comparative Molecular Biology and Evolution, Marine Biological Laboratory, 7 MBL Street, Woods Hole,
Massachusetts 02543. E-mail: [email protected].
Mol. Biol. Evol. 18(4):514–522. 2001
q 2001 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038
514
formally named the ‘‘excavate taxa’’ (Simpson and Patterson 1999). Ultrastructural data strongly suggest that
most or all excavate taxa have a common excavate ancestor; that is, the excavate taxa form either a clade or
a grade (Simpson and Patterson 1999).
Several lines of evidence argue that jakobids are
among the earliest diverging of eukaryotic lineages.
First, they were implicated to be key organisms in the
establishment of the mitochondrial symbiosis because of
their ultrastructural similarity to the retortamonads and,
indirectly, diplomonads. Diplomonads and retortamonads lack mitochondria (Brugerolle 1991), consistent
with a divergence prior to the establishment of the mitochondrial symbiosis (Cavalier-Smith 1983, 1987;
Sleigh 1989; Sogin et al. 1989; Sogin 1991; Roger, Morrison, and Sogin 1999). This hypothesis was supported
by the early divergence of the diplomonads in molecular
phylogenies of SSU rDNA and several protein families
(Hashimoto et al. 1994; Sogin et al. 1989). On the basis
of these data, O’Kelly suggests that the ancestral mitochondrial symbiont was first engulfed by a sister taxon
to the retortamonads, producing a ‘‘jakobid-like’’ ancestor from which all mitochondriate eukaryotes evolved
(O’Kelly 1993). This argument has been weakened by
the discovery of genes of mitochondrial origin in the
diplomonad Giardia (Hashimoto et al. 1998; Roger et
al. 1998; Roger 1999), indicating that the symbiosis took
place prior to the divergence of diplomonads and parabasalids. Nevertheless, a deeply diverging position of
diplomonads among eukaryotes remains a plausible sce-
Evolutionary Relationships Among “Jakobid” Flagellates
nario and implies that mitochondriate excavate taxa such
as jakobids may similarly be early-diverging eukaryotes.
Second, O’Kelly (1993) noted that the jakobids encompass all three of the basic forms of mitochondrial
cristae known in eukaryotes. Histionidae have tubular
cristae, while Jakoba libera has flat cristae, and the sole
studied member of Malawimonadidae, Malawimonas jakobiformis, has discoidal cristae (Patterson 1990;
O’Kelly 1993; O’Kelly and Nerad 1999). Mitochondrial
cristae shape has been seen as a strongly conserved
character and has been used to delimit the deepest evolutionary divisions within eukaryotes (Taylor 1976;
Sleigh 1989; Patterson 1994). Although the presence of
all three forms in jakobids implies that jakobids are not
monophyletic, this was also consistent with a model in
which jakobid flagellates were a stem group from which
all extant mitochondriate eukaryotes emerged in at least
three separate radiations (O’Kelly 1993).
Third, and most significantly, the mitochondrial genomes of Reclinomonas americana and Jakoba libera
preserve ancestral bacterial features not found in any
other eukaryotes studied to date. Mitochondrial DNAs
(mtDNAs) from R. americana (Lang et al. 1997) and J.
libera (http://megasun.bch.umontreal.ca/ogmp/projects/
individual.html) possess gene clusters that closely resemble ribosomal protein operons of eubacteria. They
also code for RNase P RNAs with bacterial minimum
consensus structures, as well as subunits of bacterialtype RNA polymerase genes that were ancestrally present in the proto-mitochondrial symbiont genome (Lang
et al. 1997, 1999). Indeed, all other mitochondrial genomes sequenced to date encode a subset of the proteins
described in jakobid mtDNAs (Burger et al. 1999; Gray,
Burger, and Lang 1999). Nonetheless, jakobid mtDNAs
share several unique gene arrangements with mitochondrial genomes from other eukaryotic lineages to the exclusion of all known bacterial genomes. These gene arrangements, together with the monophyly of all
mtDNAs (including those of jakobids) recovered in phylogenies of single and concatenated mitochondrial gene
data sets, indicate that all known mitochondria evolved
from a single endosymbiosis (Gray, Burger, and Lang
1999; Lang et al. 1999). The unique retention of many
ancestral alpha-proteobacterial genes in the jakobid
mtDNAs is most parsimoniously explained by ‘‘jakobids’’ being among the earliest diverging mitochondrion-containing taxa.
For all of their potential evolutionary significance,
our phylogenetic understanding of ‘‘jakobids’’ is incomplete. Ultrastructural data indicate that the Jakobidae
(Jakoba) and the Histionidae (Reclinomonas and Histiona) are closely related but also suggest that Malawimonas may have evolved independently from within a
broader radiation of excavate taxa (O’Kelly and Nerad
1999; Simpson and Patterson 1999). Although phylogenies of concatenated mitochondrial genes also indicate
that J. libera and R. americana form a clade, this group
is not clearly allied with Malawimonas (Burger et al.
1999). In light of the possibility of jakobid polyphyly,
we will refer to Jakobidae and Histionidae collectively
as the ‘‘core jakobids’’ (Simpson and Patterson 1999)
515
and treat Malawimonas as a separate lineage. Our understanding of the phylogenetic placement of both
groups remains hampered by the absence of nuclear
gene sequences.
Over the last decade, phylogenetic reconstructions
of small-subunit ribosomal RNA (SSU rRNA) gene sequences have provided a molecular phylogenetic framework for understanding the course of early eukaryotic
evolution. In these phylogenies, the amitochondriate diplomonads, microsporidia, and parabasalids consistently
branch before any mitochondrion-bearing lineages (Sogin 1991; Cavalier-Smith 1993; Leipe et al. 1993). However, there is doubt as to whether SSU rRNA correctly
identifies the deepest diverging groups and their relative
branching order (Leipe et al. 1993; Hirt et al. 1999).
Large disparities in the rates of evolution (leading to
long-branch attraction) and extreme biases in base compositions of different SSU rRNA sequences, especially
among ‘‘early-branching’’ eukaryotic taxa, may confound accurate phylogenetic reconstruction (Loomis and
Smith 1990; Hasegawa et al. 1993; Hasegawa and Hashimoto 1993; Galtier and Gouy 1995; Roger et al. 1999;
Stiller and Hall 1999; Philippe and Germot 2000). Highly conserved protein-coding genes might be less sensitive to some of these problems for several of the taxa
identified as ‘‘deep-branches’’ in SSU rRNA trees (Hasegawa et al. 1993; Hashimoto et al. 1994; Roger et al.
1996; Hirt et al. 1999). Therefore, the development of
these markers in combination with SSU rRNA should
help in determining a more robust global eukaryotic
phylogeny.
To this end, we focused on the tubulin gene family
as a phylogenetic marker for early eukaryotic evolution.
This gene family consists of three highly conserved subfamilies, alpha-, beta-, and gamma-tubulin, that arose
from a series of gene duplications in early eukaryotic
evolution (Edlind et al. 1996; Keeling and Doolittle
1996). Currently, alpha- and beta-tubulin genes have the
widest taxonomic representation and have been identified
in the putatively early diverging eukaryotic lineages
(Kirk-Mason, Turner, and Chakraborty 1988; Burns 1991;
Katiyar and Edlind 1994, 1996; Edlind et al. 1996; Keeling and Doolittle 1996). In order to study the relationships among the ‘‘jakobid’’ flagellates and their relationship to other ‘‘excavate’’ protist lineages, we characterized the alpha- and beta-tubulin genes from four jakobid
species: Jakoba libera, Jakoba incarcerata, Reclinomonas americana, and Malawimonas jakobiformis.
Materials and Methods
Sources of DNA
Genomic DNA from J. libera (ATCC 50422), M.
jakobiformis (ATCC 50310), and R. americana (ATCC
50284) were kindly provided by Dr. B. Franz Lang,
Université de Montreal, Canada. Genomic DNA from R.
americana (ATCC 50283) was provided by Tom Nerad,
American Type Culture Collection. Genomic DNA was
extracted from the ‘‘type culture line’’ of J. incarcerata
(Bernard, Simpson, and Patterson 2000) using the Puregene DNA Isolation Kit (Gentra Systems).
516
Edgcomb et al.
PCR Amplification, Cloning, and Sequencing
Alpha- and beta-tubulin genes were amplified from
genomic DNA. Degenerate primers were previously designed based on highly conserved N- and C-terminal
regions of the alpha- and beta-tubulin gene families
(Roger 1996). BtubA was based on the amino acid motif
GQCGNQ and had the sequence 59-GCAGGNCARTGYGGNAAYCA-39; BtubB was based on MDEMEFT
and had the sequence 59-AGTRAAYTCCATYTCRTCCAT-39; AtubA was based on QVGNCWE with
the sequence 59-RGTNGGNAAYGCNTGYTGGGA-39;
and AtubB was based on WYVGEGM with the sequence 59-CCATNCCYTCNCCNACRTACCA-39.
These primer combinations amplified .90% of the coding region of each gene. Amplification reactions were
performed with 50–100 ng of total genomic DNA under
standard conditions both in the presence and in the absence of 5%–6% acetamide (final concentration). Thermal cycling consisted of 35 cycles with an annealing
temperature of 48–558C for 1 min, an extension temperature of 728C for 1 min, and a denaturing temperature
of 948C for 30 s.
PCR products were purified from agarose gels using the Prep-A-Gene DNA Purification Kit (BioRad)
and cloned directly into pGEM-T-easy vector (Promega)
or TOPO-XL vector (Invitrogen, Inc.). Sequence data
were gathered on a LI-COR 4200L automated sequencing apparatus using infrared dye-labeled T7 and M13R
primers (LI-COR) and a cycle sequencing protocol (Sequitherm, Epicenter, Inc.). Single-stranded sequence of
4–10 clones of each PCR product was initially determined to test for the presence of multiple distinct copies
of tubulin genes within each organism and to determine
internal restriction sites for subcloning. Gene fragments
of 100–900 bp were generated by restriction enzyme
digestion of plasmid templates and subsequently subcloned into pBluescript KS (2) or pUC-19 vectors. Fulllength double-stranded DNA sequences of subcloned
distinct gene copies were determined by combined analysis of both complete gene and subcloned template data
using ALIGNIR sequence assembly software (LI-COR).
Phylogenetic Analysis
Nucleotide sequence data were submitted to NCBI
BLAST (Basic Local Alignment Search Tool) (Altschul
et al. 1990) for comparison with available sequence databases to verify tubulin gene identity and locate introns.
Protein sequences were compiled and manually aligned
in PAUP*, version 4.0b2 (Swofford 1999), along with
tubulin sequences from representatives of diverse taxa
obtained using the NCBI Entrez protein sequence search
program. In total, a data set of 63 alpha tubulins and 85
beta tubulin sequences was compiled, with representatives from all available eukaryotic groups.
The alignment of intron sequences was obtained
using the CLUSTAL W program, version 1.7 (Thompson, Higgins, and Gibson 1994) under default parameter
settings and manually adjusted by eye. Phylogenetic
analyses of the R. americana complete beta-tubulin
DNA sequences including introns were performed using
the maximum-likelihood (ML) method implemented in
PUZZLE, version 4.02 (Strimmer and von Haeseler
1996), employing the Hasegawa, Kishino and Yano
(1985) model with an eight-category discrete gamma
model of among-site rate variation (HKY1G). ML estimation on a neighbor-joining topology was used to derive a transition : transversion parameter of 1.21 and a
gamma shape parameter (a) of 0.02. Bootstrapping
(1,000 resamplings) was accomplished using the SEQBOOT and CONSENSE programs (from PHYLIP
[Felsenstein 1995]) in combination with QBOOT, a unix
shell-script that allows quartet ML bootstrap analysis
with PUZZLE, version 4.02 (available on request from
A.J.R.).
Phylogenetic analyses of the amino acid sequences
were performed using both the protein ML distance
(ML-distance) and the ML quartet puzzling (ML-QP)
methods implemented in PUZZLE, version 4.02 (Strimmer and von Haeseler 1996). Data missing in more than
a single sequence and regions of ambiguous alignment
were removed, yielding 381 sites for the alpha-tubulin
analysis, 387 sites for the beta-tubulin analysis, and 768
sites for a combined (concatenated) alpha- plus betatubulin analysis. The ML model employed in both distance and likelihood analyses was the Jones, Taylor, and
Thornton (1992) amino acid replacement matrix with an
eight-category discrete gamma model of among-sites
rate variation (JTT1G). The gamma shape parameter a
was estimated by ML optimization on a neighbor-joining topology to be 0.72 for the alpha-tubulin data set,
0.52 for the beta-tubulin data set, and 0.57 for the concatenated alpha1beta-tubulin data set. Phylogenetic
trees were inferred from the ML distances using the
Fitch-Margoliash algorithm with global rearrangements
(FITCH, version 3.57c [Felsenstein 1995]). ML distance
bootstrap values for bipartitions were calculated by analysis of 100 resampled data sets generated with the SEQBOOT program and analyzed with the PUZZLEBOOT
script (http://www.tree-puzzle.de/) in conjunction with
PUZZLE, version 4.02, and CONSENSE. Full protein
ML analyses were performed using the quartet puzzling
algorithm in PUZZLE, version 4.02, utilizing 1,000 puzzling steps.
Results and Discussion
Tubulin Genes from Jakobids
Amplification reactions for alpha-tubulin genes for
J. incarcerata, J. libera, R. americana 50283, and M.
jakobiformis all yielded a product of roughly the expected size of an alpha-tubulin gene without introns. A
single product was also obtained for the beta-tubulin
gene of J. incarcerata. Two distinct beta-tubulin gene
products were obtained from R. americana 50283
(99.1% identical at the nucleotide level), R. americana
50284 (99.6% identity), and M. jakobiformis (98.4%
identity). No introns were found in alpha- or beta-tubulins of M. jakobiformis and J. incarcerata or in the
alpha-tubulin genes of R. americana and J. libera. However, the two R. americana 50283 beta tubulin paralogs
contained introns of 70 (clone 22) and 78 (clone 18) bp
Evolutionary Relationships Among “Jakobid” Flagellates
517
Table 1
Codon Usage Bias in Jakobid Flagellates
% G 1 C,
Third
%G1C
Positions
Overall
Reclinomonas americana 50283 alpha . . . . .
R. americana 50283 beta clone 22 . . . . . . . .
R. americana 50283 beta clone 18 . . . . . . . .
R. americana 50284 beta clone 12 . . . . . . . .
R. americana 50284 beta clone 14 . . . . . . . .
Jakoba libera alpha . . . . . . . . . . . . . . . . . . . . .
J. libera beta . . . . . . . . . . . . . . . . . . . . . . . . . .
Jakoba incarcerata alpha . . . . . . . . . . . . . . . .
J. incarcerata beta . . . . . . . . . . . . . . . . . . . . . .
Malawimonas jakobiformis alpha . . . . . . . . . .
M. jakobiformis beta clone 1 . . . . . . . . . . . . .
M. jakobiformis beta clone 7 . . . . . . . . . . . . .
87.8
90.7
90.9
90.9
90.9
95.1
95.3
56.8
61.8
82.3
84.8
85.0
62.7
63.8
63.9
63.9
64.0
66.0
65.2
52.0
50.9
59.4
60.8
61.0
NOTE.—Percentages of G 1 C in the third codon positions of alpha- and
beta-tubulin genes and across all sites within each gene were calculated. When
present, introns were removed prior to the calculation.
in length, while the R. americana 50284 paralogs contained introns of 70 (clone 14) and 75 (clone 12) bp in
length, located in the same position, starting at the 56th
base of the sequences (located between a threonine and
a glycine codon). Phylogenetic analysis of these tubulin
paralogs revealed that the gene duplication that created
the two copies of beta-tubulin in R. americana appears
to have predated the divergence between strains 50283
and 50284 (data not shown). Similarly, initial sequences
from six J. libera beta-tubulin clones revealed three distinct gene copies that were distinguishable by intron sequences of 80 (clone 6), 134 (clone 1), and 157 (clone
2) bp in length, starting at the 116th base of the betatubulin genes (located between a glutamate and an alanine codon).
Another noteworthy property of the tubulin genes
described here is a moderate to extreme codon usage
bias toward codons ending in G or C (table 1). The GCbias of alpha- and beta-tubulin genes from each organism were very similar, confirming that each set of genes
derive from the same genome.
Tubulin Phylogeny
The inferred amino acid sequences for alpha-tubulin genes from J. libera, J. incarcerata, R. americana
50283, and M. jakobiformis, as well as the beta-tubulin
genes from these organisms, plus R. americana 50284,
were added to our alignments of alpha-tubulin and betatubulin sequences. Preliminary phylogenetic analyses
using all sequences in our alignment were performed.
From these analyses, we identified and selectively removed partial sequences and sequences that were extremely divergent paralogs or orthologs (to avoid problems with long-branch attraction artifacts) and pared
down representation of major eukaryotic groups to yield
final alignments of 42 alpha-tubulin and 39 beta-tubulin
sequences that were amenable to more rigorous phylogenetic analyses. Separate analyses were performed on
the alpha- and beta-tubulin (fig. 1) data sets and for a
combined alpha- and beta-tubulin data set that included
organisms for which both sequences (or for which re-
FIG. 1.—Phylogeny of eukaryotes inferred from (A) alpha-tubulin
and (B) beta-tubulin protein sequences. Phylogenetic trees were inferred with the Fitch-Margoliash algorithm from maximum-likelihood
distance matrices calculated with the JTT1G model of amino acid
replacement. The gamma distribution shape parameter alpha was 0.72
for the alpha-tubulin analysis (A) and 0.52 for the beta-tubulin analysis
(B). Support values shown near branches were computed using ML
distance/Fitch-Margoliash bootstrap analysis (100 replicates) and ML
quartet puzzling (1,000 puzzling steps) and are displayed in that order.
Jakobid sequences (determined in this study) are highlighted and identified by the letters ‘‘A’’ (alpha-tubulin) and ‘‘B’’ (beta-tubulin) and
the number of the clone used for the analysis. In cases in which there
were significant (more than six amino acids) differences between
clones sequenced, more than one clone is included in the analysis. The
scale bar indicates expected sequence divergence per unit branch
length, expressed as substitutions per site. The question mark indicates
the uncertain origin of the beta-tubulin sequence ascribed to the rhodophyte Porphyra purpurea. This sequence appears to branch within
the stramenopiles, and its identity is therefore questionable (see also
Keeling et al. [1999] for discussion of this point).
518
Edgcomb et al.
FIG. 2.—Phylogeny of eukaryotes inferred from concatenated alpha1beta-tubulin protein sequences. The tree topology was inferred
with the Fitch-Margoliash algorithm from a maximum-likelihood (ML)
distance matrix calculated with the JTT1G model of amino acid replacement. The gamma distribution shape parameter alpha was 0.57.
Support values shown near branches were computed using ML distance/Fitch-Margoliash distance bootstrap analysis (100 replicates) and
ML quartet puzzling (1,000 puzzling steps) and are displayed in that
order. Jakobid sequences are highlighted, and the genes are identified
by the letters ‘‘A’’ (alpha-tubulin) and ‘‘B’’ (beta-tubulin) and the number of the clone used for the combined analysis in each case. The scale
bar indicates expected sequence divergence per unit branch length,
expressed as substitutions per site.
lated sequences) were represented (fig. 2). We did not
attempt to reciprocally root the alpha- and beta-tubulin
trees on each other. The two families are extremely divergent from one another, and previous analyses showed
that when used as reciprocal outgroups, they artifactually attract fast-evolving sequences to the base of each
of the paralog subtrees (Keeling, Deane, and McFadden
1998; Keeling et al. 1999). Instead, sequences from diplomonads were arbitrarily drawn as the outgroup for
each of the three phylogenies, as these amitochondriate
protists consistently occupy basal positions in rooted
trees based on ribosomal RNA, EF-1a, EF-2, and RPB1
(Sogin 1991; Leipe et al. 1993; Hashimoto et al. 1994;
Hirt et al. 1999).
A variety of traditional eukaryotic groups that are
well established on the basis of morphology or molecular phylogeny are also recovered in either alpha- or
beta-tubulin trees (Edlind et al. 1996; Keeling and Doolittle 1996) (figs. 1 and 2). In addition, several ‘‘higherlevel’’ eukaryotic groupings have been observed, including fungi plus microsporidia, animals plus fungi
(this clade is likely to include taxa which are neither
animals nor fungi, e.g., choanoflagellates), and ciliates
plus apicomplexa (alveolates). Phylogenies based on
SSU rRNA, alpha-tubulin, and beta-tubulin support the
animals-plus-fungi assemblage (represented in this data
set by metazoa, fungi, and microsporidia) and the grouping of the alveolates (Sogin 1991; Wainright et al. 1993;
Edlind et al. 1996; Keeling and Doolittle 1996; Keeling,
Deane, and McFadden 1998). In addition, elongation
factors and actins demonstrate a common evolutionary
history of animals plus fungi exclusive of most other
eukaryotes (Baldauf and Palmer 1993). However, alphaand beta-tubulin phylogenies also display attributes not
always supported by other gene-based trees (Edlind et
al. 1996; Keeling et al. 1999). In agreement with other
studies (Edlind et al. 1996; Keeling et al. 1999), our
analyses of single tubulin data sets show that microsporidia group strongly with the fungi (figs. 1 and 2).
This is in contrast to their deep placement in SSU rRNA
trees (Leipe et al. 1993) but in agreement with other
molecular phylogenies, including mitochondrial HSP70
(Germot, Philippe, and Le Guyader 1997; Hirt et al.
1997), TATA-box binding protein (Fast, Logsdon, and
Doolittle 1999), elongation factor-2, RPB1 (Hirt et al.
1999), ValtRS (Weiss et al. 1999), and, most recently,
large-subunit ribosomal RNA (LSU rRNA) (Van de
Peer, Ben Ali, and Meyer 2000). Also unlike rRNA
trees, tubulin phylogenies do not show many independent deep-branching lineages leading to a ‘‘crown’’ radiation of eukaryotes if diplomonads are considered the
outgroup. Instead, a deep split occurs within eukaryotes
separating two robust groupings, a plant-protist ‘‘superclade’’ and an animals-plus-fungi group, with several
protist lineages intervening (fig. 1). Although the exact
composition of the plant-protist ‘‘superclade’’ varies between our alpha- and beta-tubulin phylogenies and other
published analyses, the group is consistently composed
of Viridiplantae (land plants and chlorophyte algae),
stramenopiles, cercozoans, cryptomonads, Euglenozoa,
and heteroloboseids (fig. 1) (Edlind et al. 1996; Keeling,
Deane, and McFadden 1998; Keeling et al. 1999). Intervening taxa between the two major clades include
Eumycetozoa, parabasalids, and diplomonads. The Eumycetozoa weakly associate with animals plus fungi in
beta-tubulin analysis; however, in both alpha-tubulin and
combined alpha- plus beta-tubulin phylogenies, they
strongly associate with the plant protist superclade (figs.
1 and 2). Similarly, parabasalids and diplomonads are
sister taxa in beta-tubulin phylogenies (fig. 1B), yet parabasalids, not diplomonads, show some affinity to the
animals-plus-fungi group in alpha-tubulin trees (fig. 1A).
Placement of ‘‘Jakobid’’ Flagellates in Tubulin
Phylogenies
ML distance/Fitch and ML-QP trees for each of the
three analyses (figs. 1A, 1B, and 2) show that the ‘‘jakobids’’ in the broad sense (Reclinomonas, Jakoba, and
Malawimonas) (O’Kelly 1993) and the ‘‘excavate taxa’’
(in this analysis, jakobids, heteroloboseids, and diplomonads) do not form clades. However, our analyses do
show a close grouping between beta-tubulins from J.
libera, R. americana 50283, and R. americana 50284
(99% bootstrap support in a distance analysis, and quartet puzzling support of 83%) (fig. 2B). The extremely
close relationship between R. americana and J. libera
beta-tubulins (fig. 1B) is particularly interesting because
these two sequences each harbor a single nonhomologous intron that occupies different positions in the gene
Evolutionary Relationships Among “Jakobid” Flagellates
(data not shown). Neither intron position is found in any
other beta-tubulin gene studied to date (J. M. Logsdon
Jr., personal communication), thus indicating recent intron turnover in beta-tubulin in the R. americana–J. libera lineage.
Interestingly, the J. libera–R. americana grouping
is not recovered in alpha-tubulin phylogenies by either
method of analysis (fig. 1A). Lack of support for this
grouping, which is well supported by other data (discussed below), is puzzling and may have several causes.
For instance, placement of R. americana and J. libera
alpha-tubulin sequences may be obscured by the overall
poorer support for monophyletic groupings observed in
the alpha-tubulin tree, especially within the plant-protist
superclade (fig. 1A). It is also possible that one of these
sequences represents an aberrant paralog of alpha-tubulin which could have life-cycle-specific expression
and has an altered pattern of substitution, a frequently
observed phenomenon in tubulins (Kube-Granderath
and Schliwa 1998). If so, then the orthologous alphatubulin gene(s) from one or another of these organisms
may not have been detected in our PCR approach, perhaps due to introns interrupting the primer binding sites
in the genes. However, the J. libera–R. americana relationship was well supported in the combined alphaand beta-tubulin analysis (91% bootstrap support and
90% QP support; fig. 2), indicating that the alpha-tubulin data do not significantly conflict with the betatubulin data, but instead lack strong enough phylogenetic signal to resolve the relationships between these
taxa.
Also unexpected was the positioning of J. incarcerata in tubulin trees. Jakoba incarcerata was assigned
to Jakoba primarily on the basis of light microscope
observations (Bernard, Simpson, and Patterson 2000).
Ultrastructural data support assignation of J. incarcerata
to the ‘‘core jakobids’’ (Bernard, Simpson, and Patterson
2000; unpublished data) but also suggest that J. incarcerata may be basal to other known ‘‘core jakobids’’
(i.e., Jakoba may not be monophyletic). Jakoba incarcerata would therefore be expected to group with J. libera and R. americana in a ‘‘core jakobid’’ clade (Simpson and Patterson 1999), but not necessarily to group
specifically with J. libera. However, in our beta-tubulin
analyses, J. incarcerata did not cluster with these taxa,
emerging separately within the plant-protist superclade
in beta-tubulin trees. In alpha-tubulin phylogeny, it
joined the animals-plus-fungi–parabasalid grouping with
very weak support (45% distance bootstrap and 59% QP
support; fig. 1A). Interestingly, this sequence robustly
grouped with the plant-protist superclade in the combined alpha- plus beta-tubulin analysis (100% distance
bootstrap and 74% QP support) and emerged with moderate support as the earliest-diverging lineage within this
clade (70% distance bootstrap and QP support). Similarly, M. jakobiformis occupied a relatively basal position in the plant-protist superclade in all three analyses,
varying in position only slightly (figs. 1 and 2).
To evaluate the stability of the position of J. incarcerata in the combined analysis in light of the weakly supported affinity of J. incarcerata for the animals-
519
plus-fungi–parabasalid grouping in the alpha-tubulin
analysis, we performed a separate combined alpha- plus
beta-tubulin analysis that included the alpha- and betatubulin genes from a second diplomonad, Giardia intestinalis, and a composite trichomonad sequence composed of Trichomonas vaginalis beta-tubulin and Monocercomonas sp. alpha-tubulin (data not shown). In this
analysis, there were only minor changes in support values at several nodes, and the position of the jakobids,
including J. incarcerata, did not change from the combined alpha- plus beta-tubulin analysis presented here.
There is little, if any, resolution in the branching
order among the major groups in the plant-protist superclade in our tubulin trees, and, in general, the jakobid
lineages show inconsistent affinities to other taxa in the
various analyses. One exception is the tendency in betatubulin trees for the J. libera–R. americana group to
branch with the heteroloboseid Naegleria and the euglenozoan Euglena (fig. 1B). This tendency is amplified,
albeit weakly (44% distance bootstrap support and 56%
QP support), in the combined alpha- plus beta-tubulin
analysis, in which the J. libera–R. americana group became a sister group of a Heterolobosea-Euglenozoa
clade. The best resolution of the major phyla and their
relative branching order was obtained in our combined
alpha- plus beta-tubulin analyses, yet support values for
some groups (e.g., Viridiplantae) remained low. It is
likely that the extreme conservation of both tubulin
genes leave few phylogenetically informative sites, especially for groups in the plant-protist superclade. In this
respect, ancient phylogenetic signal appears to be retained in tubulin sequences, but its overall amount is
low.
Jakobid Polyphyly
Our observation of a J. libera–R. americana grouping is in agreement with concatenated mitochondrial
gene phylogenies (Burger et al. 1999; Gray, Burger, and
Lang 1999; Burger et al. 2000), SSU rRNA phylogenies
(unpublished data), and shared unique mtDNA gene
composition between these taxa (e.g., their common
possession of bacterial DNA-dependent RNA polymerase genes) (Lang et al. 1999). Concatenated mitochondrial gene phylogenies also support the distinctness of
Malawimonas from ‘‘core jakobids’’ and show them
emerging from within a strikingly similar plant-protist
superclade (although the monophyly of this clade is not
well supported). These results are consistent with recent
evaluations of ultrastructural features of excavate taxa
(O’Kelly, Farmer, and Nerad 1999; O’Kelly and Nerad
1999; unpublished data) which indicate that ‘‘core jakobids’’ (including J. libera and R. americana) share
several characters not found in other excavate taxa but
that only one or two features of debatable importance
are unique to Malawimonas and ‘‘core jakobids.’’ Given
the conflict between ultrastructural and tubulin data in
the placement of J. incarcerata, further data on this taxon would be desirable. Unfortunately, mitochondrial or
nuclear gene data have not yet been published.
520
Edgcomb et al.
The Cristal Hypothesis
O’Kelly’s (1993) original suggestion for jakobids
as a stem group from which all three basic cristal morphologies arose requires the jakobids to be paraphyletic,
giving rise to at least three clades of nonjakobid organisms. While it could be argued that this degree of paraphyly of jakobids is reconcilable with our results, it
appears safe to reject O’Kelly’s hypothesis from the evidence at hand. First, O’Kelly’s hypothesis requires that
the J. libera–R. americana grouping be interrupted by
at least two nonjakobid clades (one essentially platycristate, the other tubulocristate). Second, it would predict that Malawimonas would fall adjacent to Heterolobosea and Euglenozoa to form a clade of discicristate
organisms. In fact, it is the J. libera–R. americana clade
which shows some affinity for Heterolobosea and Euglenozoa (figs. 1 and 2).
Jakobids and Deep-Branching Excavate Taxa
Tubulin genes are the first molecules to be used to
evaluate the affinities of jakobids to amitochondriate
‘‘excavate’’ protist lineages. We did not observe a close
relationship between the diplomonads and any of the
jakobid lineages in this study, although sequences from
Malawimonas and J. incarcerata branched near the base
of the plant-protist superclade and therefore emerged in
a region of the tree close to the diplomonad branch.
Currently, the ‘‘excavate hypothesis’’ can be rationalized
in terms of tubulin phylogenies in the following manner.
If some jakobid lineages (e.g., J. incarcerata and Malawimonas) are among the deepest branches in a plantprotist superclade, and diplomonads are the outgroup to
these and all other eukaryotes, then the common ancestor of the plant-protist superclade and the common ancestor of extant eukaryotes could have had an ‘‘excavate’’ phenotype. Paradoxically, it is precisely R. americana and J. libera that display the most primitive mitochondrial genome organization (Lang et al. 1999), yet
they do not seem to occupy a particularly deep branch
in tubulin trees. Clearly, more molecular data will be
required from jakobids and other excavates, such as retortamonads, Trimastix, and Carpediemonas, to decide
where these protists fit in the large-scale phylogeny of
eukaryotes.
Perhaps even more problematic will be understanding the basis of conflicts between large-scale eukaryotic
phylogenies inferred with the various nuclear gene
markers (e.g., tubulin vs. SSU rRNA phylogenies). Substitution rate calibration analyses of ‘‘crown’’ eukaryotic
rRNAs support the division between an animals-plusfungi group and a large heterogeneous plant-protist
clade of eukaryotes (Van de Peer et al. 1996), as do
concatenated mitochondrial genes (Burger et al. 1999).
However, the most conspicuous disagreement between
SSU rRNA trees and tubulin trees is the fact that ‘‘deep’’
lineages in SSU rRNA trees (Leipe et al. 1993), such as
Microsporidia, Eumycetozoa, Euglenozoa, and Heterolobosea, emerge from within the these two major eukaryotic clades in tubulin trees. Several authors have
suggested that the deep-branching position of these and
other taxa within SSU rRNA trees could result from
long-branch attraction artifacts (Embley and Hirt 1998;
Keeling, Deane, and McFadden 1998; Philippe and
Adoutte 1998; Roger, Morrison, and Sogin 1999) and
that the alternative positioning of these taxa within the
tubulin tree may be correct (Embley and Hirt 1998;
Keeling, Deane, and McFadden 1998; Dacks and Roger
1999). Although this seems likely to be the case for
Microsporidia (Keeling and McFadden 1998), we
should also be cautious in placing too much confidence
in tubulin phylogenies. Most taxa that lack flagella or
centrioles, such as Microsporidia, higher fungi, Entamoeba, Dictyostelium, and red algae, tend to form extremely long branches in tubulin trees, and often group
together (Edlind et al. 1996; Keeling and Doolittle 1996;
Keeling et al. 1999; Keeling, Luker, and Palmer 2000;
unpublished data), as do aberrant developmentally regulated tubulin isoforms (data not shown). Therefore,
much of the structure of tubulin-based trees could also
result from long-branch attraction effects.
Supplementary Material
GenBank accession numbers for newly deposited
sequences are as follows: J. incarcerata clone 22 alphatubulin, AF267179; J. libera alpha-tubulin, AF267180;
M. jakobiformis alpha-tubulin, AF267181; R. americana
sp. clone 44 alpha-tubulin, AF267182; J. incarcerata
beta-tubulin, AF267183; J. libera clone 3 beta-tubulin,
AF267184; M. jakobiformis clone 1 beta-tubulin,
AF267185; M. jakobiformis clone 7 beta-tubulin,
AF267186; R. americana 50283 clone 22 beta-tubulin,
AF267190; R. americana 50283 clone 18 beta-tubulin,
AF267189; R. americana 50284 clone 12 beta-tubulin,
AF267187; R. americana 50284 clone 14 beta-tubulin,
AF267188.
Acknowledgments
We thank John M. Archibald and Joel Dacks for
helpful comments on early drafts of the manuscript; B.
Franz Lang, G. Burger, and C. O’Kelly for kindly providing genomic DNAs; and John M. Logsdon Jr. for the
tubulin intron information. This work was supported by
NASA grant NAG5-4895 to M.L.S., NASA Astrobiology Cooperative Agreement NCC2-1054, and continuing support from the Unger G. Vetlesen Foundation to
M.L.S. V.P.E. and A.J.R. both contributed equally to this
work.
LITERATURE CITED
ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D.
J. LIPMAN. 1990. Basic local alignment search tool. J. Mol.
Biol. 215:403–410.
BALDAUF, S. L., and J. D. PALMER. 1993. Animals and fungi
are each other’s closest relatives: congruent evidence from
multiple proteins. Proc. Natl. Acad. Sci. USA 90:11558–
11562.
BERNARD, C., A. G. B. SIMPSON, and D. J. PATTERSON. 2000.
Some free-living flagellates from anoxic sediments. Ophelia
52:113–142.
Evolutionary Relationships Among “Jakobid” Flagellates
BRUGEROLLE, G. 1991. Flagellar and cytoskeletal systems in
amitochondrial flagellates: Archamoeba, Metamonada and
Parabasala. Protoplasma 164:70–90.
BURGER, G., D. SAINT-LOUIS, M. W. GRAY, and B. F. LANG.
1999. Complete sequence of the mitochondrial DNA of the
red alga Porphyra purpurea: cyanobacterial introns and
shared ancestry of red and green algae. Plant Cell 11:1675–
1694.
BURGER, G., Y. ZHU, T. G. LITTLEJOHN, S. J. GREENWOOD, M.
N. SCHNARE, B. F. LANG, and M. W. GRAY. 2000. Complete
sequence of the mitochondrial genome of Tetrahymena pyriformis and comparison with Paramecium aurelia mitochondrial DNA. J. Mol. Biol. 297:365–380.
BURNS, R. G. 1991. Alpha-, beta-, and gamma-tubulins: sequence comparisons and structural constraints. Cell Motil.
Cytoskeleton 20:181–189.
CAVALIER-SMITH, T. 1983. A 6-kingdom classification and a
unified phylogeny. Pp. 1027–1034 in W. SCHWEMMLER and
H. E. A. SCHENK, eds. Endocytobiology. Vol. 2. de Gruyter,
Berlin.
———. 1987. Eukaryotes with no mitochondria. Nature 326:
332–333.
———. 1993. Kingdom Protozoa and its 18 phyla. Microbiol.
Rev. 57:953–994.
DACKS, J., and A. J. ROGER. 1999. The first sexual lineage and
the relevance of facultative sex. J. Mol. Evol. 48:779–783.
EDLIND, T. D., J. LI, G. S. VISVESVARA, M. H. VODKIN, G. L.
MCLAUGHLIN, and S. K. KATIYAR. 1996. Phylogenetic analysis of beta-tubulin sequences from amitochondrial protozoa. Mol. Phylogenet. Evol. 5:359–367.
EMBLEY, T. M., and R. P. HIRT. 1998. Early branching eukaryotes? Curr. Opin. Genet. Dev. 8:624–629.
FAST, N. M., J. M. LOGSDON JR., and W. F. DOOLITTLE. 1999.
Phylogenetic analysis of the TATA box binding protein
(TBP) gene from Nosema locustae: evidence for a microsporidia-fungi relationship and spliceosomal intron loss.
Mol. Biol. Evol. 16:1415–1419.
FELSENSTEIN, J. 1995. PHYLIP (phylogeny inference package).
Version 3.57c. Distributed by the author, Department of Genetics, University of Washington, Seattle.
FLAVIN, M., and T. A. NERAD. 1993. Reclinomonas americana
N. G., N. Sp., a new freshwater heterotrophic flagellate. J.
Eukaryot. Microbiol. 40:172–179.
GALTIER, N., and M. GOUY. 1995. Inferring phylogenies from
DNA sequences of unequal base compositions. Proc. Natl.
Acad. Sci. USA 92:11317–11321.
GERMOT, A., H. PHILIPPE, and H. LE GUYADER. 1997. Evidence for loss of mitochondria in Microsporidia from a mitochondrial-type HSP70 in Nosema locustae. Mol. Biochem. Parasitol. 87:159–168.
GRAY, M. W., G. BURGER, and B. F. LANG. 1999. Mitochondrial evolution. Science 283:1476–1477.
HASEGAWA, M., and T. HASHIMOTO. 1993. Ribosomal RNA
trees misleading? Nature 361:23.
HASEGAWA, M., T. HASHIMOTO, J. ADACHI, N. IWABE, and T.
MIYATA. 1993. Early branchings in the evolution of eukaryotes: ancient divergence of entamoeba that lacks mitochondria revealed by protein sequence data. J. Mol. Evol.
36:380–388.
HASEGAWA, M., H. KISHINO, and T. YANO. 1985. Dating the
human-ape splitting by a molecular clock of mitochondrial
DNA. J. Mol. Evol. 22:160–174.
HASHIMOTO, T., Y. NAKAMURA, F. NAKAMURA, T. SHIRAKURA,
J. ADACHI, N. GOTO, K. OKAMOTO, and M. HASEGAWA.
1994. Protein phylogeny gives a robust estimation for early
divergences of eukaryotes: phylogenetic place of a mito-
521
chondria-lacking protozoan, Giardia lamblia. Mol. Biol.
Evol. 11:65–71.
HASHIMOTO, T., L. B. SANCHEZ, T. SHIRAKURA, M. MULLER,
and M. HASEGAWA. 1998. Secondary absence of mitochondria in Giardia lamblia and Trichomonas vaginalis revealed
by valyl-tRNA synthetase phylogeny. Proc. Natl. Acad. Sci.
USA 95:6860–6865.
HIRT, R. P., B. HEALY, C. R. VOSSBRINCK, E. U. CANNING, and
T. M. EMBLEY. 1997. A mitochondrial Hsp70 orthologue in
Vairimorpha necatrix: molecular evidence that microsporidia once contained mitochondria. Curr. Biol. 7:995–998.
HIRT, R. P., J. M. LOGSDON, B. HEALY, M. W. DOREY, W. F.
DOOLITTLE, and T. M. EMBLEY. 1999. Microsporidia are
related to fungi: evidence from the largest subunit of RNA
polymerase II and other proteins. Proc. Natl. Acad. Sci.
USA 96:580–585.
JONES, D. T., W. R. TAYLOR, and J. M. THORNTON. 1992. The
rapid generation of mutation data matrices from protein sequences. CABIOS 8:275–282.
KATIYAR, S. K., and T. D. EDLIND. 1994. Beta-tubulin genes
of Trichomonas vaginalis. Mol. Biochem. Parasitol. 64:33–
42.
———. 1996. Entamoeba histolytica encodes a highly divergent beta-tubulin. J. Eukaryot. Microbiol. 43:31–34.
KEELING, P. J., J. A. DEANE, C. HINK-SCHAUER, S. E. DOUGLAS, U.-G. MAIER, and G. I. MCFADDEN. 1999. The secondary endosymbiont of the cryptomonad Guillardia theta
contains alpha-, beta-, and gamma-tubulin genes. Mol. Biol.
Evol. 16:1308–1313.
KEELING, P. J., J. A. DEANE, and G. I. MCFADDEN. 1998. The
phylogenetic position of alpha- and beta-tubulins from the
Chlorarachnion host and Cercomonas (Cercozoa). J. Eukaryot. Microbiol. 45:561–570.
KEELING, P. J., and W. F. DOOLITTLE. 1996. Alpha-tubulin from
early-diverging eukaryotic lineages and the evolution of the
tubulin family. Mol. Biol. Evol. 13:1297–1305.
KEELING, P. J., M. A. LUKER, and J. D. PALMER. 2000. Evidence from beta-tubulin phylogeny that microsporidia
evolved from within the fungi. Mol. Biol. Evol. 17:23–31.
KEELING, P. J., and G. I. MCFADDEN. 1998. Origins of microsporidia. Trends Microbiol. 6:19–23.
KIRK-MASON, K. E., M. J. TURNER, and P. R. CHAKRABORTY.
1988. Cloning and sequence of beta tubulin cDNA from
Giardia lamblia. Nucleic Acids Res. 16:2733.
KUBE-GRANDERATH, E., and M. SCHLIWA. 1998. Unusual tubulins for unusual cells. Protist 149:123–126.
LANG, B. F., G. BURGER, C. J. O’KELLY, R. CEDERGREN, G.
B. LEMIEUX, D. SANKOFF, M. TURMEL, and M. W. GRAY.
1997. An ancestral mitochondrial DNA resembling a eubacterial genome in miniature. Nature 387:493–497.
LANG, B. F., E. SEIF, M. W. GRAY, C. J. O’KELLY, and G.
BURGER. 1999. A comparative genomics approach to the
evolution of eukaryotes and their mitochondria. J. Eukaryot. Microbiol. 46:320–326.
LEIPE, D. D., J. H. GUNDERSON, T. A. NERAD, and M. L. SOGIN.
1993. Small subunit ribosomal RNA1 of Hexamita inflata
and the quest for the first branch in the eukaryotic tree. Mol.
Biochem. Parasitol. 59:41–48.
LOOMIS, W. F., and D. W. SMITH. 1990. Molecular phylogeny
of Dictyostelium discoideum by protein sequence comparison. Proc. Natl. Acad. Sci. USA 879093–1097.
O’KELLY, C. J. 1993. The jakobid flagellates: structural features of Jakoba, Reclinomonas and Histiona and implications for the early diversification of eukaryotes. J. Eukaryot.
Microbiol. 40:627–636.
O’KELLY, C. 1997. Ultrastructure of trophozoites, zoospores
and cysts of Reclinomonas americana Flavin and Nerad,
522
Edgcomb et al.
1993 (Protista incertae sedis: Histionidae). Eur. J. Protistol.
33:337–348.
O’KELLY, C. J., M. A. FARMER, and T. A. NERAD. 1999. Ultrastructure of Trimastix pyriformis (Klebs) Bernard et al.:
similarities of Trimastix species with retortamonad and jakobid flagellates. Protist 150:149–62.
O’KELLY, C. J., and T. A. NERAD. 1999. Malawimonas jakobiformis n. gen., n. sp. (Malawimonadidae n. fam.): a Jakoba-like heterotrophic nanoflagellate with discoidal mitochondrial cristae. J. Eukaryot. Microbiol. 46:522–531.
PATTERSON, D. J. 1990. Jakoba libera (Ruinen 1938), a heterotrophic flagellate from deep oceanic sediments. J. Eukaryot. Microbiol. 40:172–179.
———. 1994. Progress in protozoology. IX International Congress of Protozoology, Stuttgart, Germany.
PHILIPPE, H., and A. ADOUTTE. 1998. The molecular phylogeny
of protozoa: solid facts and uncertainties. Pp. 25–52 in G.
H. COOMBS, K. VICKERMAN, M. A. SLEIGH, and A. WARREN, eds. Evolutionary relationships among Protozoa. Kluwer Academic Publishers, London.
PHILIPPE, H., and A. GERMOT. 2000. Phylogeny of eukaryotes
based on ribosomal RNA: long-branch attraction and models of sequence evolution. Mol. Biol. Evol. 17:830–834.
ROGER, A. J. 1996. Studies on the phylogeny and gene structure of early-branching eukaryotes. Ph.D. thesis, Dalhousie
University, Hallifax, Nova Scotia, Canada.
———. 1999. Reconstructing early events in eukaryotic evolution. Am. Nat. 154(Suppl.):S146–S163.
ROGER, A. J., H. G. MORRISON, and M. L. SOGIN. 1999. Primary structure and phylogenetic relationships of a malate
dehydrogenase gene from Giardia lamblia. J. Mol. Evol.
48:750–755.
ROGER, A. J., O. SANDBLOM, W. F. DOOLITTLE, and H. PHILIPPE. 1999. An evaluation of elongation factor 1a as a phylogenetic marker for eukaryotes. Mol. Biol. Evol. 16:218–
233.
ROGER, A. J., M. W. SMITH, R. F. DOOLITTLE, and W. F. DOOLITTLE. 1996. Evidence for the Heterolobosea from phylogenetic analysis of genes encoding glyceraldehyde-3-phosphate dehydrogenase. J. Eukaryot. Microbiol. 43:475–485.
ROGER, A. J., S. G. SVARD, J. TOVAR, C. G. CLARK, M. W.
SMITH, F. D. GILLIN, and M. L. SOGIN. 1998. A mitochondrial-like chaperonin 60 gene in Giardia lamblia: evidence
that diplomonads once harbored an endosymbiont related to
the progenitor of mitochondria. Proc. Natl. Acad. Sci. USA
95:229–234.
SIMPSON, A. G. B., and D. J. PATTERSON. 1999. The ultrastructure of Carpediemonas membranifera: (Eukaryota), with
reference to the ‘excavate hypothesis’. Eur. J. Protistol. 35:
353–370.
SIMPSON, A. G. B., C. BERNARD, and D. J. PATTERSON. 2000.
The ultrastructure of Trimastix marina Kent, 1880 (Eukaryota), an excavate flagellate. Europ. J. Protistol. 36:229–
251.
SLEIGH, M. 1989. Protozoa and other protists. 2nd edition. Edward Arnold, London.
SOGIN, M. L. 1991. Early evolution and the origin of eukaryotes. Curr. Opin. Genet. Dev. 1:457–463.
SOGIN, M. L., J. H. GUNDERSON, H. J. ELWOOD, R. A. ALONSO,
and D. A. PEATTIE. 1989. Phylogenetic meaning of the
kingdom concept: an unusual ribosomal RNA from Giardia
lamblia. Science 243:75–77.
STILLER, J. W., and B. J. HALL. 1999. Long-branch attraction
and the rDNA model of early eukaryotic evolution. Mol.
Biol. Evol. 16:1270–1279.
STRIMMER, K., and A. VON HAESELER. 1996. Quartet puzzling:
a quartet maximum-likelihood method for reconstructing
tree topologies. Mol. Biol. Evol 13:964–969.
SWOFFORD, D. L. 1999. PAUP*. Phylogenetic analysis using
parsimony (*and other methods). Version 4. Sinauer, Sunderland, Mass.
TAYLOR, F. J. R. 1976. Flagellate phylogeny: a study in conflicts. J. Protozool. 23:28–40.
THOMPSON, J. D., D. G. HIGGINS, and T. J. GIBSON. 1994.
CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680.
VAN DE PEER, Y., A. BEN ALI, and A. MEYER. 2000. Microsporidia: accumulating molecular evidence that a group of
amitochondriate and suspectedly primitive eukaryotes are
just curious fungi. Gene 246:1–8.
VAN DE PEER, Y., S. A. RENSING, U. G. MAIER, and R. DE
WACHTER. 1996. Substitution rate calibration of small subunit ribosomal RNA identifies chlorarachniophyte endosymbionts as remnants of green algae. Proc. Natl. Acad. Sci.
USA 93:7732–7736.
VOIGHT, M. 1901. Mitteilungen aus der Biolog. Station Plön,
Holstein-Uber einige bisher unbekannte Süsswasserorganismen. Zool. Az. 24:191–195.
WAINRIGHT, P. O., G. HINKLE, M. L. SOGIN, and S. K. STICKEL.
1993. Monophyletic origins of the metazoa: an evolutionary
link with fungi. Science 260:340–342.
WEISS, L. M., T. D. EDLIND, C. R. VOSSBRINCK, and T. HASHIMOTO. 1999. Microsporidian molecular phylogeny: the fungal connection. J. Eukaryot. Microbiol. 46:17S–18S.
B. FRANZ LANG, reviewing editor
Accepted November 20, 2000