Phylogenetic history of plastid-targeted proteins in the peridinin-containing dinoflagellate Heterocapsa triquetra

Abstract

The evolutionary history and relationship between plastids of dinoflagellate algae and apicomplexan parasites have been controversial both because the organelles are unusual and because their genomes contain few comparable genes. However, most plastid proteins are encoded in the host nucleus and targeted to the organelle, and several of these genes have proved to have interesting and informative evolutionary histories. We have used expressed sequence tag (EST) sequencing to generate gene sequence data from the nuclear genome of the dinoflagellate Heterocapsa triquetra and inferred phylogenies for the complete set of identified plastid-targeted proteins. Overall, dinoflagellate plastid proteins are most consistently related to homologues from the red algal plastid lineage (not green) and, in many of the most robust cases, they branch with other chromalveolate algae. In resolved phylogenies where apicomplexan data are available, dinoflagellates and apicomplexans are related. We also identified two cases of apparent lateral, or horizontal, gene transfer, one from the green plastid lineage and one from a bacterial lineage unrelated to plastids or cyanobacteria.

Full text

Phylogenetic history of plastid-targeted proteins in
the peridinin-containing dinoflagellate Heterocapsa
triquetra
Ross F. Waller,3Nicola J. Patron and Patrick J. Keeling
Correspondence
Patrick J. Keeling
pkeeling@interchange.ubc.ca
Canadian Institute for Advanced Research, Department of Botany, University of British
Columbia, 3529-6270 University Boulevard, Vancouver, BC, V6T 1Z4, Canada
The evolutionary history and relationship between plastids of dinoflagellate algae and apicomplexan
parasites have been controversial both because the organelles are unusual and because their
genomes contain few comparable genes. However, most plastid proteins are encoded in the
host nucleus and targeted to the organelle, and several of these genes have proved to have
interesting and informative evolutionary histories. We have used expressed sequence tag (EST)
sequencing to generate gene sequence data from the nuclear genome of the dinoflagellate
Heterocapsa triquetra and inferred phylogenies for the complete set of identified plastid-targeted
proteins. Overall, dinoflagellate plastid proteins are most consistently related to homologues
from the red algal plastid lineage (not green) and, in many of the most robust cases, they branch with
other chromalveolate algae. In resolved phylogenies where apicomplexan data are available,
dinoflagellates and apicomplexans are related. We also identified two cases of apparent lateral,
or horizontal, gene transfer, one from the green plastid lineage and one from a bacterial lineage
unrelated to plastids or cyanobacteria.
INTRODUCTION
Plastids originated by the endosymbiotic uptake of a
cyanobacterium and the subsequent conversion of this
endosymbiont into the highly reduced and specialized
double-membrane-bound primary plastid found today in
land plants and some algae. Most algal groups, however,
acquired their plastids by an additional step called secondary
endosymbiosis. Here, an alga containing a primary plastid
is itself taken up and converted into an organelle within its
new eukaryotic host (Archibald & Keeling, 2002). These
secondary plastids are bounded by either three or four
membranes and are found in chlorarachniophytes, eugle-
nids, cryptomonads, heterokonts, haptophytes, dinoflagel-
lates and apicomplexans. In all of these groups, the plastid
retains only a small genome: most proteins are nuclear-
encoded and are targeted post-translationally to the plastid
using specific N-terminal peptides that are characteristic for
either primary or secondary plastids (McFadden, 2001).
Dinoflagellates are closely related to apicomplexans and,
together with ciliates and a handful of other protists, make
up the alveolates (Fast et al., 2002; Gajadhar et al., 1991).
Since many members of both dinoflagellates and apicom-
plexans contain secondary plastids, the most parsimonious
explanation is that they share a common origin, but the
evolutionary history of both plastids has proved conten-
tious, in part because they are divergent, making them
difficult to compare.
While some dinoflagellates have undergone plastid replace-
ments through further endosymbiotic events (Delwiche,
1999; Keeling, 2004), the plastid found in the majority of
photosynthetic dinoflagellates is a secondary plastid con-
taining the distinctive pigment peridinin. This plastid is
further distinguished in many ways, not least in that it has
three membranes rather than four, which appears to have
a significant affect on how proteins are targeted to the
organelle (Nassoury et al., 2003; Patron et al., 2005). The
genomes of dinoflagellate plastids are also the most reduced
known. All but a few of their genes have moved to the
nucleus (Bachvaroff et al., 2004; Hackett et al., 2004a), and
the remaining genome (16 genes have been found so far,
nearly all related to photosynthesis) has been broken up
into mini-circles each generally encoding a single gene
(Zhang et al., 1999).
3Present address: Botany School, University of Melbourne, Parkville,
VIC 3010, Australia.
The GenBank/EMBL/DDBJ accession numbers for the completed
cDNA sequences reported in this study are AY826826–AY826947
and AY884246–AY884255. The dbEST accession numbers for the
EST sequences reported in this study are DT379484–DT386290.
Protein maximum-likelihood trees for a further 23 plastid-targeted
proteins are available as supplementary material in IJSEM Online.
Abbreviations: EST, expressed sequence tag; FBA, fructose-1,6-
bisphosphate aldolase; GAPDH, glyceraldehyde-3-phosphate dehydro-
genase.
64061 G2006 IUMS Printed in Great Britain 1439
International Journal of Systematic and Evolutionary Microbiology (2006), 56, 1439–1447 DOI 10.1099/ijs.0.64061-0

Apicomplexans, on the other hand, are obligate parasites,
so their plastid (the apicoplast) is non-photosynthetic and
generally reduced. Its relict genome contains few genes,
typically highly divergent at the sequence level, and mostly
encoding housekeeping functions (Wilson et al., 1996). The
evolutionary origin of this plastid has been debated since its
discovery, with some data interpreted as showing a green
algal origin (Cai et al., 2003; Funes et al., 2002; Ko
¨hler et al.,
1997) and other data interpreted as showing a red algal
origin (Blanchard & Hicks, 1999; Fast et al., 2001; McFadden
& Waller, 1997; Patron et al., 2004; Waller et al., 2003). Since
nearly all of the genes remaining in the dinoflagellate plastid
are related to photosynthesis, there are few plastid-encoded
genes that can be compared directly between apicomplexans
and dinoflagellates. Those that have been (primarily encod-
ing small- and large-subunit rRNA) are highly divergent in
both groups, making phylogenies difficult to interpret
(Zhang et al., 2000).
The disputes over the origin of dinoflagellate and apicom-
plexan plastids widened with the suggestion that both plas-
tids originated in the ancestor of all chromalveolates. The
chromalveolates are a hypothetical grouping of alveolates
and chromists (cryptomonads, haptophytes and hetero-
konts): all eukaryotes hypothesized to have red algal-derived
plastids (Cavalier-Smith, 1998). Plastid-encoded gene trees
have given variable results, but multigene analyses weakly
unite chromist plastids (Yoon et al., 2002, 2004). Nuclear
gene trees provide strong support for the alveolates and their
relationship to heterokonts (Baldauf et al., 2000; Harper
et al., 2005). Taken together, plastid and cytosolic data are
therefore consistent with the chromalveolate hypothesis, but
neither support the whole group: the strongest support for
this comes from two nucleus-encoded genes for plastid-
targeted proteins: glyceraldehyde-3-phosphate dehydro-
genase (GAPDH) and fructose-1,6-bisphosphate aldolase
(FBA) (Fast et al., 2001; Harper & Keeling, 2003; Patron
et al., 2004). In both cases, the chromalveolate plastid-
targeted protein appears to have evolved in a unique way
relative to other plastids, and these deviations suggest that
plastids of chromalveolates share a common origin.
Phylogenies of GAPDH and FBA demonstrate the usefulness
of nucleus-encoded plastid-targeted proteins for studying
plastid evolution, but they are poorly sampled because the
nuclear genome is practically less accessible than that of the
plastid. In dinoflagellates, this problem is aggravated by
their unusually large genome size, so expressed sequence tag
(EST) surveys have been used to generate genomic data from
several species of dinoflagellate (Bachvaroff et al., 2004;
Hackett et al., 2004a; Patron et al., 2005). Phylogenetic
analyses for some of these proteins have been carried out,
some showing a substantial conflict in gene trees indicat-
ing either a red or green algal origin, or high levels of lateral
gene transfer (Hackett et al., 2004a). Here, we have phylo-
genetically analysed all the plastid-targeted proteins identi-
fied from an EST survey of the dinoflagellate Heterocapsa
triquetra (Patron et al., 2005). We have inferred phylogenies
for 52 proteins, eight of which have apicomplexan homo-
logues. Overall, the dinoflagellate plastid proteins tend to
branch with red algae and chromists (with red algal second-
ary plastids) with stronger consistency than previously
observed. Of the phylogenies containing apicomplexan
homologues that are resolved with reasonable support, the
apicomplexans group within the red algal plastid clade and
a specific relationship with the dinoflagellate homologues
was evident in some of these. Many dinoflagellate plastid-
targeted proteins are relatively divergent and contain unique
oddities [such as a tandem fusion of translation elongation
factor Ts (EF-Ts)], and a few well-resolved phylogenies
appear to support lateral gene transfer, including genes
derived from prokaryotes.
METHODS
EST sequencing and protein identification. H. triquetra CCMP
449, cultivated in Guillard’s f2-Si medium at 16 uC with a 12 h: 12 h
light/dark cycle, was harvested in batches and total RNA was used to
construct a cDNA library as described by Patron et al. (2005). ESTs
were 59-sequenced and gene identification was performed at the
PEPdb (http://amoebidia.bcm.umontreal.ca/pepdb/searches/welcome.
php). Plastid-targeted proteins were identified as part of an analysis
of the nature of plastid-targeting leader sequences in dinoflagellates
reported in Patron et al. (2005). Briefly, EST annotation was
searched for genes with known function in the plastid and the
sequence database was searched using known plastid-targeted pro-
teins from other organisms. In cases where candidate genes were
determined to be full-length, they were analysed for the presence of
an N-terminal leader with characteristics expected of a dinoflagellate
plastid-targeting peptide (in particular the presence of a predicted
signal peptide). In addition, phylogenetic analysis was carried out
for all candidate genes (see below), revealing some to be related to
cytosolic or mitochondrial homologues. In these cases, unless the
gene encoded a leader warranting further investigation, they were no
longer considered. In cases where cDNAs were truncated and the
presence of a leader could not be verified, genes were considered to
encode plastid-targeted proteins if they were phylogenetically related
to other plastid-targeted homologues and to cyanobacterial homo-
logues. Identification of putatively plastid-targeted proteins in api-
complexans followed previous annotation, which is based on the
well-characterized leaders of annotated proteins (Ralph et al., 2004)
and direct localization (e.g. Jomaa et al., 1999; Waller et al., 1998).
Completed cDNA sequences have been deposited in GenBank
(accession numbers AY826826–AY826947, AY884246–AY884255)
and EST sequences have been deposited in dbEST (DT379484–
DT386290).
Phylogenetic analysis. Protein alignments were constructed using
CLUSTAL X (Thompson et al., 1997) and edited manually. All ambi-
guous sites of the alignments were removed from the dataset for
phylogenetic analyses. The alignment data are available on request.
Protein maximum-likelihood analyses used PhyML (Guindon &
Gascuel, 2003) with input trees generated by BIONJ, the JTT model
of amino acids substitution, proportion of variable rates estimated
from the data and nine categories of substitution rates (eight
variable and one invariable). One hundred bootstrap trees were cal-
culated with PhyML initially without gamma correction categories;
if the resulting trees showed resolution, the analysis was repeated
with four rate categories. For distance analyses, gamma-corrected
distances were calculated by TREE-PUZZLE 5.2 (Schmidt et al., 2002)
using the WAG substitution matrix with eight variable rate
categories and invariable sites. Trees were inferred by weighted
1440 International Journal of Systematic and Evolutionary Microbiology 56
R. F. Waller, N. J. Patron and P. J. Keeling

neighbour-joining using WEIGHBOR 1.0.1a (Bruno et al., 2000).
Bootstrap resampling was performed using PUZZLEBOOT (shell script
by A. Roger and M. Holder; http://www.tree-puzzle.de) with rates
and frequencies estimated using TREE-PUZZLE 5.2.
RESULTS AND DISCUSSION
Red algal origin of Heterocapsa plastid-
targeted proteins
A previous analysis of plastid-targeting leaders in H. trique-
tra (Patron et al., 2005) identified a total of 63 distinct
genes from 2022 EST clusters as likely being plastid-targeted
protein-coding genes. Of these, 11 represented multiple
copies of certain genes; hence there were 52 distinct plas-
tid proteins identified in total (including several distinct
lineages within the light-harvesting complex superfamily).
The majority of these proteins are involved in the light
or dark reactions of photosynthesis, but other activities
such as transcription, translation or synthesis of fatty acids
and isoprenoids were also represented, and eight proteins
representing such functions had identifiable homologues
in apicomplexans. These plastid proteins were each sub-
jected to phylogenetic analyses, with the exceptions of
form II ribulose-1,5-bisphosphate carboxylase oxygenase
(RuBisCO), which has already been analysed in detail and
is known to be unique to dinoflagellates and photosyn-
thetic proteobacteria (Whitney et al., 1995), and the light-
harvesting complex proteins, which we found to form a
poorly resolved protein family in dinoflagellates. Further-
more, dinoflagellate GAPDH, FBA and PsbO have been
analysed previously (Fast et al., 2001; Harper & Keeling,
2003; Ishida & Green, 2002; Patron et al., 2004).
Of the 34 discrete proteins analysed (Table 1), many of
the H. triquetra genes were divergent compared with other
plastid homologues, and the position of H. triquetra was
completely unresolved in two cases. The other 32 proteins
supported H. triquetra branching with other plastid homo-
logues, as expected. Of these, H. triquetra branched with
neither the red nor green algal lineages (i.e. the position was
unresolved within plastids) or, in 18 cases, plastids were
polyphyletic. H. triquetra branched with the red plastid
lineage (red and chromistan algae) in another 13 cases,
eight with moderate to strong support, as shown in three
examples in Fig. 1. In the phosphoglycerate kinase phylo-
geny (Fig. 1a), the H. triquetra genes for the plastid and
cytosolic enzymes are both shown, and the plastid gene
branches within a relatively well supported (89 %) clade of
red algal and chromists genes, as well as Bigelowiella natans,
which has a green algal secondary plastid and whose phos-
phoglycerate kinase has been proposed to be derived
from a red alga by lateral gene transfer (Archibald et al.,
2003). Similarly, H. triquetra phosphoribulokinase (Fig. 1b)
branches specifically with chromist homologues, and most
closely with the haptophyte Isochrysis (94–96 %). Lastly, H.
triquetra and Alexandrium tamarense (Hackett et al., 2004a)
possess a gene for photosystem II extrinsic protein (Fig. 1c),
a protein apparently lost from green algae altogether and
otherwise only known in cyanobacteria, red algal plastids
and their chromist derivatives.
As a whole, the phylogenies of H. triquetra plastid-targeted
proteins are most consistent with the peridinin-containing
plastid being derived from a red algal plastid. This is in line
with results from plastid-encoded genes (Zhang et al., 2000)
and a few plastid-targeted genes (Bachvaroff et al., 2004;
Hackett et al., 2004a). However, in some analyses, several
proteins have shown a green algal origin (Hackett et al.,
2004a) and, in one multigene analysis, the red origin was
not significantly better supported than a green algal origin
(Yoon et al., 2005). We see no strong evidence for a green
algal origin and only a few cases that might suggest lateral
gene transfer (see below). In nearly all resolved cases, H.
triquetra branches specifically with heterokonts, hapto-
phytes or cryptomonads, while only three proteins show
H. triquetra branching with a red alga to the exclusion of
these taxa; none of these are statistically supported. Among
the more strongly supported phylogenies, the data are
more consistent with the chromalveolate hypothesis than
with independent plastid origins; however, broader repre-
sentation of red algal taxa is required in order to test this
hypothesis more thoroughly.
Relationship between dinoflagellate and
apicomplexan plastid-targeted proteins
Most of the H. triquetra plastid-targeted proteins are
involved in photosynthesis and so are not present in api-
complexans. However, eight proteins were identified from
both groups (see asterisks in Table 1), and the phylogenies
of four offer some resolution (Fig. 2). The phylogenies of
both dimethyladenosine synthase (Fig. 2a) and queuine
tRNA ribosyltransferase (Fig. 2b) support dinoflagellates
and apicomplexans as sister taxa within the red plastid clade.
The latter is also of interest because it is not known from
green plastids. Interestingly, however, the dinoflagellate–
apicomplexan clade of queuine tRNA ribosyltransferase also
includes two alphaproteobacterial sequences, the chromist
plastid sequences and a Dictyostelium homologue, and there
is no clear relationship between plastid genes in general
and cyanobacterial homologues. Ultimately, the source of
this protein in plastids is unclear: it is possible the pro-
teobacteria and Dictyostelium each acquired this gene from
plastids, but it is also possible the plastid genes are not
ancestrally cyanobacterial or that there are many paralogues.
In any case, the H. triquetra gene forms a strongly supported
group with Toxoplasma (91–99 %), suggesting that they
are mostly likely closely related, and the presence of this
protein in apicomplexans is not consistent with their
having a plastid of green algal ancestry. 1-Deoxy-D-xylulose-
5-phosphate synthase (Fig. 2c) also places apicomplexans
firmly within the red plastid clade, although the position
of H. triquetra is unresolved within this red group.
Apicomplexans are, nevertheless, moderately strongly
allied to the chromist Thalassiosira (74–82 %) and not the
green plastid lineage.
http://ijs.sgmjournals.org 1441
Phylogeny of dinoflagellate plastid proteins

GAPDH provides further support for dinoflagellates and
apicomplexans grouping in the chromalveolates, although
not as sisters. This gene has been analysed in detail pre-
viously (Fast et al., 2001; Harper & Keeling, 2003; Takishita
et al., 2005) and will therefore not be described here except
to state that the H. triquetra data are consistent with
previous observations that GAPDH supports the origin of
both dinoflagellate and apicomplexan plastids from the
red plastid clade and that the apicomplexan homologues
are specifically related to those of haptophytes, not
Table 1. H. triquetra plastid-targeted proteins predicted from cDNAs sorted by inferred evolutionary origin
The phylogenetic position of H. triquetra protein sequences is indicated where H. triquetra groups within either a clade of red algal and red
algal-derived plastids (Red) or green algal and green algal-derived plastids (Green). If the H. triquetra position is not resolved with either of
these groups but still groups within the plastid clade, this is indicated by ‘Plastid’. The level of support from maximum-likelihood and
weighted neighbour-joining bootstrap for these groups is indicated as follows: +++, greater than 80; ++, 70–80; +, 60–70; no symbol,
below 60. Proteins for which homologues are represented in apicomplexans are indicated by asterisks.
Protein (gene) GenBank accession no. Phylogeny Figure/reference
Red algal/chromalveolate
Phosphoglycerate kinase (pgk) AY826862 Red +++ Fig. 1a
Phosphoribulokinase (prk) AY826860 Red +++ Fig. 1b
Photosystem II extrinsic protein (psbU) AY826889 Red +++ (no green) Fig. 1c
Dimethyladenosine synthase* (ksgA) AY826874 Red Fig. 2a
Queuine tRNA ribosyltransferase* (tgt) AY826892 Red +++ Fig. 2b
1-Deoxy-D-xylulose-5-phosphate synthase* (dxs) AY826876 Red Fig. 2c
Ribose-5-phosphate isomerase (rpiA) AY826893 Red +++ Supplementary Fig. S1
Cytochrome f(petA) AY826881 Red Supplementary Fig. S2
Cytochrome b559 (psbF) AY826887 Red (plus B. natans) Supplementary Fig. S3
Transketolase (tktA) AY826896 Red Supplementary Fig. S4
GAPDH* AY884246, AY884247 Red +++ Takishita et al. (2005)
Fructose-1,6-bisphosphate aldolase (fbaA) AAV71135 Red +++ Patron et al. (2004)
Oxygen-evolving enhancer 1 (psbO) AAM77465 Red +++ Ishida & Green (2002)
Green algal/plant
Oxoglutarate/malate translocator AY826859 Green +++ (no red) Fig. 3a
Protochlorophyllide reductase subunit (chlL) AY826880 Green Supplementary Fig. S5
Photosystem I subunit III (psaF) AY826884 Green Supplementary Fig. S6
Non-plastid
Acetolactate synthase (als) AY826826 Bacteria +++ Fig. 3b
RuBisCO form II AY826897 Bacteria +++ Whitney et al. (1995)
Red/green unresolved
Photosystem II protein L (psbL) AY826888 Plastid Supplementary Fig. S7
Thylakoid 11 kDa protein AY826895 Plastid Supplementary Fig. S8
Translation elongation factor Ts (tsf) AY826878 Plastid Fig. 3c
Ferredoxin* (petF) AY826847, AY826848 Plastid Supplementary Fig. S9
Ferredoxin-NADP
+
reductase* (petH) AY826853 Plastid Supplementary Fig. S10
Geranylgeranyl reductase/hydrogenase AY826855 Plastid Supplementary Fig. S11
Beta-keto-acyl reductase AY826869 Plastid Supplementary Fig. S12
Carbonic anhydrase (yadF) AY826838–AY826840 Plastid Supplementary Fig. S13
ATP synthase subunit gamma (atpC) AY826835 Plastid Supplementary Fig. S14
ATP synthase subunit C (atpH) AY826871, AY884249–
AY884255
Plastid Supplementary Fig. S15
Cytochrome b6 (petC) AY826843 Plastid Supplementary Fig. S16
Cytochrome c6 (petJ) AY826872, AY884248 Plastid Supplementary Fig. S17
Photosystem I protein E (psaE) AY826882 Plastid Supplementary Fig. S18
Ascorbate peroxidase AY826833 Plastid Supplementary Fig. S19
Adenylate kinase (adk) AY826832 Plastid Supplementary Fig. S20
Photosystem I subunit XI (psaL) AY826885 Plastid Supplementary Fig. S21
Acyl carrier protein* (acp) AY826829 Plastid and mitochondrion Supplementary Fig. S22
Lipoate protein ligase* AY826879 Plastid and mitochondrion Supplementary Fig. S23
1442 International Journal of Systematic and Evolutionary Microbiology 56
R. F. Waller, N. J. Patron and P. J. Keeling

dinoflagellates. Plastid-targeted TufA was originally used to
argue for a green algal origin of apicomplexan plastids (in
the absence of data from dinoflagellates) (Ko
¨hler et al.,
1997), but it has now been shown that the apicomplexan and
dinoflagellate homologues are closely related (Hackett et al.,
2004a), in our view ruling this gene out as support for a
green plastid in apicomplexans.
Evidence for lateral gene transfer and gene
fusions in dinoflagellate plastid-targeted
proteins
Previously it has been shown that B. natans, a chlorar-
achniophyte alga with a green algal secondary plastid,
acquired several plastid-targeted protein genes from other
Fig. 1. Relationship of dinoflagellate plastid-targeted proteins phosphoglycerate kinase (a), queuine tRNA ribosyltransferase
(b) and photosystem II extrinsic protein (c) to homologues from red algal and chromist plastids. Numbers at nodes indicate
bootstrap support (ML top/left; NJ bottom/right) for major nodes over 50 % by at least one method. Major eukaryotic groups
and plastids are indicated to the right.
http://ijs.sgmjournals.org 1443
Phylogeny of dinoflagellate plastid proteins

algae, including red algae, or bacteria (Archibald et al.,
2003). In dinoflagellates, RuBisCO has long been known
to be of a bacterial type and considered to have originated
by lateral gene transfer (Rowan et al., 1996; Whitney et al.,
1995). One additional protein, d-aminolaevulinic acid
dehydratase, has also been suggested to be derived from a
green alga along with other more complex cases (Hackett
et al., 2004a), but the extent to which dinoflagellate plastid
proteins may not originate from a single source is not
well known. Other than the well-studied RuBisCO, it is
similarly unclear whether any bacterial proteins have been
harnessed in the plastid.
In our survey, a handful of proteins branched with the
green algae or plants rather than the red plastid clade, but
in all cases except one this was without support (Table 1).
The one exception, the oxoglutarate/malate translocator
(Fig. 3a), is interesting because, to date, the only eukaryotic
sources from which it has been reported are plastids of
plants and green algae. Since the preponderance of genes
support the red origin of the dinoflagellate plastid, the
presence of this protein in H. triquetra but not in finished
genomes of red algae or diatoms (Armbrust et al., 2004;
Barbier et al., 2005; Matsuzaki et al., 2004) or in any other
known red or red-derived plastids suggests that this dino-
flagellate protein is derived by lateral gene transfer from a
green source.
Even more interesting, the H. triquetra acetolactate synthase
(Fig. 3b) appears to have originated from a non-plastid
source. The protein is known to exist in cyanobacteria
and other plastids, but the H. triquetra gene is related to
neither and instead forms a highly supported group within
alphaproteobacteria (100 % support), specifically related
to a Paracoccus/Rhodobacter subgroup (96–98 % support).
Nested well within a group of related bacteria as this is, it
suggests that this gene is derived relatively recently from an
alphaproteobacterial genome. The gene was represented by
13 ESTs, which indicates that it is highly expressed, and we
also discovered a related homologue in a subsequent EST
Fig. 2. Relationship between dinoflagellate and apicomplexan plastid-targeted proteins dimethyladenosine synthase (a),
queuine tRNA ribosyltransferase (b) and 1-deoxy-D-xylulose-5-phosphate synthase (c). See the legend to Fig. 1 for other
details.
1444 International Journal of Systematic and Evolutionary Microbiology 56
R. F. Waller, N. J. Patron and P. J. Keeling

survey of another species of dinoflagellate, Karlodinium
micrum (Patron et al., 2006), both strongly supporting the
conclusion that this gene is encoded in the dinoflagellate
genome (as opposed to being a bacterial contaminant). It is
not clear whether this gene is common to other chromal-
veolate plastids or indeed even whether all other dino-
flagellates encode this gene, but its apparent recent origin
suggests that its distribution may be relatively restricted in
these plastids.
Lastly, one of the H. triquetra genes that branches weakly
with the green algae is EF-Ts. The phylogeny of this gene
(Fig. 3c) is too weak to conclude much about its origin;
however, it is noteworthy because of its unique structure.
The H. triquetra gene encodes a tandem duplication of
EF-Ts and the phylogeny shows that the two halves of the
duplication branch together with 100 % support. This close
relationship between the two halves shows that the dupli-
cation took place relatively recently, which has interesting
implications for the function of this protein, which recycles
GDP from elongation factor-Tu during translation elonga-
tion. Several other dinoflagellate proteins are expressed as
polymers, some of which are processed and some are
functional fusion proteins (Hiller et al., 1995; Liu et al., 2004;
Rowan et al., 1996). It is not clear whether the EF-Ts is
processed or not. Despite this protein having a house-
keeping role in translation, we could not identify a plastid
version in apicomplexans, but we did find non-duplicated
plastid homologues of EF-Ts in a diatom (Phaeodactylum)
and a cryptomonad (Guillardia). The distribution of this
Fig. 3. Phylogenies of three dinoflagellate
plastid-targeted proteins with unusual evolu-
tionary histories. (a) The oxoglutarate/malate
translocator is only known in green algal
plastids and now also dinoflagellates. (b)
The H. triquetra acetolactate synthase is clo-
sely related to those of alphaproteobacteria
and is suggested to have originated by lat-
eral gene transfer. (c) The H. triquetra EF-Ts
is a unique tandem duplication where the
two halves (labelled C and N) are closely
related, suggesting a recent duplication. See
the legend to Fig. 1 for other details.
http://ijs.sgmjournals.org 1445
Phylogeny of dinoflagellate plastid proteins

character is relatively restricted, therefore, but seeking this
protein in other apicomplexans might be of interest.
Concluding remarks
Overall, the phylogeny of dinoflagellate plastid-targeted
proteins supports the origin of this organelle from the red
plastid lineage. This is consistent with previous suggestions
based on pigmentation (chlorophyll cis found in dino-
flagellates and chromists) and some molecular data, but the
analyses described here provide more consistency than
previously observed with molecular phylogenetic surveys.
In general, dinoflagellate proteins also tend to branch with
homologues from other chromalveolates, although no single
protein shows this relationship unambiguously and the best
evidence for this remains the unusual evolutionary histories
of GAPDH and FBA. The few plastid proteins available from
both dinoflagellates and apicomplexans tend to support a
common origin of these two plastids and add further
evidence for the red ancestry of apicomplexan plastids.
Altogether there is very little evidence to support a green
origin for either dinoflagellate or apicomplexan plastids.
Confirming that the dinoflagellate plastid is related to those
of apicomplexans and the chromalveolates as a whole is
significant for several reasons, in particular because of the
many differences between dinoflagellate and apicomplexan
plastids. Distinctions in membrane number and protein
targeting between dinoflagellates and other chromalveolates
mean that dinoflagellates must have undergone a dramatic
transformation necessitating changes to the targeting system
and also to the transit peptides of perhaps hundreds of
proteins (Nassoury et al., 2003; Patron et al., 2005). In
addition, a plastid-containing ancestor of apicomplexans
and dinoflagellates is interesting because deep-branching
members of these two lineages, Colpodella and Perkinsus,
respectively (Goggin & Barker, 1993; Kuvardina et al., 2002),
potentially enable us to reconstruct many characteristics of
this ancestor and the subsequent evolution of both groups in
detail (Leander & Keeling, 2003).
Analyses of plastid-targeted protein genes also continue to
show the evolutionary complexity of dinoflagellate plastids,
and indeed plastids as a whole. There are now a handful of
dinoflagellate plastid proteins that do not seem to fit with
the overall picture of the plastid’s evolution from the red
plastid lineage: this work and other studies (Hackett et al.,
2003) have shown some proteins being more akin to green
homologues and others being only distantly related to
plastid homologues in general. This is similar to observa-
tions from another myxotrophic alga, B. natans (Archibald
et al., 2003), raising interesting questions about the genetic
content of such genomes in general. In addition, dino-
flagellates seem to be prone to developing novelty at the
molecular level (Hackett et al., 2004b), as evident from
their plastid proteins. Complexes appear to form between
proteins from distantly related sources and new structures
are found, making dinoflagellates an interesting case study
of molecular diversity in microbial eukaryotes.
ACKNOWLEDGEMENTS
This work was supported by the Protist EST Program of Genome
Canada/Genome Atlantic and by a grant (MOP-42517) from the
Canadian Institutes for Health Research (CIHR). R. F. W. is supported
by Fellowships from CIHR and the Michael Smith Foundation for
Heath Research (MSFHR) and P. J. K. is a Fellow of the CIAR and a new
investigator of the CIHR and MSFHR.
REFERENCES
Archibald, J. M. & Keeling, P. J. (2002). Recycled plastids: a ‘green
movement’ in eukaryotic evolution. Trends Genet 18, 577–584.
Archibald, J. M., Rogers, M. B., Toop, M., Ishida, K. & Keeling, P. J.
(2003). Lateral gene transfer and the evolution of plastid-targeted
proteins in the secondary plastid-containing alga Bigelowiella natans.
Proc Natl Acad Sci U S A 100, 7678–7683.
Armbrust, E. V., Berges, J. A., Bowler, C. & 42 other authors (2004).
The genome of the diatom Thalassiosira pseudonana: ecology,
evolution, and metabolism. Science 306, 79–86.
Bachvaroff, T. R., Concepcion, G. T., Rogers, C. R., Herman, E. M. &
Delwiche, C. F. (2004). Dinoflagellate expressed sequence tags data
indicate massive transfer of chloroplast genes to the nuclear genome.
Protist 155, 65–78.
Baldauf, S. L., Roger, A. J., Wenk-Siefert, I. & Doolittle, W. F. (2000).
A kingdom-level phylogeny of eukaryotes based on combined
protein data. Science 290, 972–977.
Barbier, G., Oesterhelt, C., Larson, M. D., Halgren, R. G.,
Wilkerson, C., Garavito, R. M., Benning, C. & Weber, A. P. (2005).
Comparative genomics of two closely related unicellular thermo-
acidophilic red algae, Galdieria sulphuraria and Cyanidioschyzon
merolae, reveals the molecular basis of the metabolic flexibility of
Galdieria sulphuraria and significant differences in carbohydrate
metabolism of both algae. Plant Physiol 137, 460–474.
Blanchard, J. L. & Hicks, J. S. (1999). The non-photosynthetic
plastid in malarial parasites and other apicomplexans is derived from
outside the green plastid lineage. J Eukaryot Microbiol 46, 367–375.
Bruno, W. J., Socci, N. D. & Halpern, A. L. (2000). Weighted neighbor
joining: a likelihood-based approach to distance-based phylogeny
reconstruction. Mol Biol Evol 17, 189–197.
Cai, X., Fuller, A. L., McDougald, L. R. & Zhu, G. (2003). Apicoplast
genome of the coccidian Eimeria tenella.Gene 321, 39–46.
Cavalier-Smith, T. (1998). A revised six-kingdom system of life. Biol
Rev Camb Philos Soc 73, 203–266.
Delwiche, C. F. (1999). Tracing the thread of plastid diversity
through the tapestry of life. Am Nat 154 (Suppl. 4), S164–S177.
Fast, N. M., Kissinger, J. C., Roos, D. S. & Keeling, P. J. (2001).
Nuclear-encoded, plastid-targeted genes suggest a single common
origin for apicomplexan and dinoflagellate plastids. Mol Biol Evol 18,
418–426.
Fast, N. M., Xue, L., Bingham, S. & Keeling, P. J. (2002). Re-
examining alveolate evolution using multiple protein molecular
phylogenies. J Eukaryot Microbiol 49, 30–37.
Funes, S., Davidson, E., Reyes-Prieto, A., Magallo
´n, S., Herion, P.,
King, M. P. & Gonzalez-Halphen, D. (2002). A green algal apicoplast
ancestor. Science 298, 2155.
Gajadhar, A. A., Marquardt, W. C., Hall, R., Gunderson, J., Ariztia-
Carmona, E. V. & Sogin, M. L. (1991). Ribosomal RNA sequences of
Sarcocystis muris,Theileria annulata and Crypthecodinium cohnii
reveal evolutionary relationships among apicomplexans, dinoflagel-
lates, and ciliates. Mol Biochem Parasitol 45, 147–154.
1446 International Journal of Systematic and Evolutionary Microbiology 56
R. F. Waller, N. J. Patron and P. J. Keeling

Goggin, C. L. & Barker, S. C. (1993). Phylogenetic position of the
genus Perkinsus (Protista, Apicomplexa) based on small subunit
ribosomal RNA. Mol Biochem Parasitol 60, 65–70.
Guindon, S. & Gascuel, O. (2003). A simple, fast, and accurate
algorithm to estimate large phylogenies by maximum likelihood. Syst
Biol 52, 696–704.
Hackett, J. D., Maranda, L., Yoon, H. S. & Bhattacharya, D. (2003).
Phylogenetic evidence for the cryptophyte origin of the plastid of
Dinophysis (Dinophysiales, Dinophyceae). J Phycol 39, 440–448.
Hackett, J. D., Yoon, H. S., Soares, M. B., Bonaldo, M. F., Casavant,
T. L., Scheetz, T. E., Nosenko, T. & Bhattacharya, D. (2004a).
Migration of the plastid genome to the nucleus in a peridinin
dinoflagellate. Curr Biol 14, 213–218.
Hackett, J. D., Anderson, D. M., Erdner, D. L. & Bhattacharya, D.
(2004b). Dinoflagellates: a remarkable evolutionary experiment. Am
J Bot 91, 1523–1534.
Harper, J. T. & Keeling, P. J. (2003). Nucleus-encoded, plastid-targeted
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) indicates a
single origin for chromalveolate plastids. Mol Biol Evol 20, 1730–1735.
Harper, J. T., Waanders, E. & Keeling, P. J. (2005). On the
monophyly of chromalveolates using a six-protein phylogeny of
eukaryotes. Int J Syst Evol Microbiol 55, 487–496.
Hiller, R. G., Wrench, P. M. & Sharples, F. P. (1995). The light-
harvesting chlorophyll a-c-binding protein of dinoflagellates: a
putative polyprotein. FEBS Lett 363, 175–178.
Ishida, K. & Green, B. R. (2002). Second- and third-hand chloro-
plasts in dinoflagellates: phylogeny of oxygen-evolving enhancer 1
(PsbO) protein reveals replacement of a nuclear-encoded plastid
gene by that of a haptophyte tertiary endosymbiont. Proc Natl Acad
SciUSA99, 9294–9299.
Jomaa, H., Wiesner, J., Sanderbrand, S. & 9 other authors (1999).
Inhibitors of the nonmevalonate pathway of isoprenoid biosynthesis
as antimalarial drugs. Science 285, 1573–1576.
Keeling, P. J. (2004). Diversity and evolutionary history of plastids
and their hosts. Am J Bot 91, 1481–1493.
Ko
¨hler, S., Delwiche, C. F., Denny, P. W., Tilney, L. G., Webster, P.,
Wilson, R. J. M., Palmer, J. D. & Roos, D. S. (1997). A plastid of
probable green algal origin in apicomplexan parasites. Science 275,
1485–1489.
Kuvardina, O. N., Leander, B. S., Aleshin, V. V., Myl’nikov, A. P.,
Keeling, P. J. & Simdyanov, T. G. (2002). The phylogeny of
colpodellids (Alveolata) using small subunit rRNA gene sequences
suggests they are the free-living sister group to apicomplexans.
J Eukaryot Microbiol 49, 498–504.
Leander, B. S. & Keeling, P. J. (2003). Morphostasis in alveolate
evolution. Trends Ecol Evol 18, 395–402.
Liu, L., Wilson, T. & Hastings, J. W. (2004). Molecular evolution of
dinoflagellate luciferases, enzymes with three catalytic domains in a
single polypeptide. Proc Natl Acad Sci U S A 101, 16555–16560.
Matsuzaki, M., Misumi, O., Shin-i, T. & 38 other authors (2004).
Genome sequence of the ultrasmall unicellular red alga Cyani-
dioschyzon merolae 10D. Nature 428, 653–657.
McFadden, G. I. (2001). Primary and secondary endosymbiosis and
the origin of plastids. J Phycol 37, 951–959.
McFadden, G. I. & Waller, R. F. (1997). Plastids in parasites of
humans. Bioessays 19, 1033–1040.
Nassoury, N., Cappadocia, M. & Morse, D. (2003). Plastid ultra-
structure defines the protein import pathway in dinoflagellates. J Cell
Sci 116, 2867–2874.
Patron, N. J., Rogers, M. B. & Keeling, P. J. (2004). Gene replacement of
fructose-1,6-bisphosphate aldolase supports the hypothesis of a single
photosynthetic ancestor of chromalveolates. Eukaryot Cell 3,1169–1175.
Patron, N. J., Waller, R. F., Archibald, J. M. & Keeling, P. J. (2005).
Complex protein targeting to dinoflagellate plastids. J Mol Biol 348,
1015–1024.
Patron, N. J., Waller, R. F. & Keeling, P. J. (2006). A tertiary plastid
uses genes from two endosymbionts. J Mol Biol 357, 1373–1382.
Ralph, S. A., van Dooren, G. G., Waller, R. F., Crawford, M. J.,
Fraunholz, M. J., Foth, B. J., Tonkin, C. J., Roos, D. S. & McFadden,
G. I. (2004). Tropical infectious diseases: metabolic maps and functions
of the Plasmodium falciparum apicoplast. Nat Rev Microbiol 2,203–216.
Rowan, R., Whitney, S. M., Fowler, A. & Yellowlees, D. (1996).
Rubisco in marine symbiotic dinoflagellates: form II enzymes in
eukaryotic oxygenic phototrophs encoded by a nuclear multigene
family. Plant Cell 8, 539–553.
Schmidt, H. A., Strimmer, K., Vingron, M. & von Haeseler, A. (2002).
TREE-PUZZLE: maximum likelihood phylogenetic analysis using
quartets and parallel computing. Bioinformatics 18, 502–504.
Takishita, K., Patron, N. J., Ishida, K., Maruyama, T. & Keeling, P. J.
(2005). A transcriptional fusion of genes encoding glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) and enolase in dinoflagellates.
J Eukaryot Microbiol 52, 343–348.
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. &
Higgins, D. G. (1997). The CLUSTAL_Xwindows interface: flexible
strategies for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Res 25, 4876–4882.
Waller, R. F., Keeling, P. J., Donald, R. G. & 7 other authors (1998).
Nuclear-encoded proteins target to the plastid in Toxoplasma gondii
and Plasmodium falciparum.Proc Natl Acad Sci U S A 95, 12352–12357.
Waller, R. F., Keeling, P. J., van Dooren, G. G. & McFadden, G. I.
(2003). Comment on ‘‘A green algal apicoplast ancestor’’. Science
301, 49.
Whitney, S. M., Shaw, D. C. & Yellowlees, D. (1995). Evidence that
some dinoflagellates contain a ribulose-1,5-bisphosphate carboxy-
lase/oxygenase related to that of the alpha-proteobacteria. Proc Biol
Sci 259, 271–275.
Wilson, I. R. J. M., Denny, P. W., Preiser, P. R. & 8 other authors
(1996). Complete gene map of the plastid-like DNA of the malaria
parasite Plasmodium falciparum.J Mol Biol 261, 155–172.
Yoon, H. S., Hackett, J. D., Pinto, G. & Bhattacharya, D. (2002). The
single, ancient origin of chromist plastids. Proc Natl Acad Sci U S A
99, 15507–15512.
Yoon, H. S., Hackett, J. D., Ciniglia, C., Pinto, G. & Bhattacharya, D.
(2004). A molecular timeline for the origin of photosynthetic
eukaryotes. Mol Biol Evol 21, 809–818.
Yoon, H. S., Hackett, J. D., Van Dolah, F. M., Nosenko, T., Lidie, K. L.
& Bhattacharya, D. (2005). Tertiary endosymbiosis driven genome
evolution in dinoflagellate algae. Mol Biol Evol 22, 1299–1308.
Zhang, Z., Green, B. R. & Cavalier-Smith, T. (1999). Single gene
circles in dinoflagellate chloroplast genomes. Nature 400, 155–159.
Zhang, Z., Green, B. R. & Cavalier-Smith, T. (2000). Phylogeny of ultra-
rapidly evolving dinoflagellate chloroplast genes: a possible common
origin for sporozoan and dinoflagellate plastids. J Mol Evol 51,26–40.
http://ijs.sgmjournals.org 1447
Phylogeny of dinoflagellate plastid proteins