Chlamydiae has contributed at least 55 genes to Plantae with predominantly plastid functions.

Chlamydiae Has Contributed at Least 55 Genes to
Plantae with Predominantly Plastid Functions
Ahmed Moustafa1,2., Adrian Reyes-Prieto2., Debashish Bhattacharya1,2*
1 Interdisciplinary Program in Genetics, University of Iowa, Iowa City, Iowa, United States of America, 2Department of Biological Sciences and Roy J. Carver Center for
Comparative Genomics, University of Iowa, Iowa City, Iowa, United States of America
Abstract
Background: The photosynthetic organelle (plastid) originated via primary endosymbiosis in which a phagotrophic protist
captured and harnessed a cyanobacterium. The plastid was inherited by the common ancestor of the red, green (including
land plants), and glaucophyte algae (together, the Plantae). Despite the critical importance of primary plastid
endosymbiosis, its ancient derivation has left behind very few ‘‘footprints’’ of early key events in organelle genesis.
Methodology/Principal Findings: To gain insights into this process, we conducted an in-depth phylogenomic analysis of
genomic data (nuclear proteins) from 17 Plantae species to identify genes of a surprising provenance in these taxa,
Chlamydiae bacteria. Previous studies show that Chlamydiae contributed many genes (at least 21 in one study) to Plantae
that primarily have plastid functions and were postulated to have played a fundamental role in organelle evolution. Using
our comprehensive approach, we identify at least 55 Chlamydiae-derived genes in algae and plants, of which 67% (37/55)
are putatively plastid targeted and at least 3 have mitochondrial functions. The remainder of the proteins does not contain a
bioinformatically predicted organelle import signal although one has an N-terminal extension in comparison to the
Chlamydiae homolog. Our data suggest that environmental Chlamydiae were significant contributors to early Plantae
genomes that extend beyond plastid metabolism. The chlamydial gene distribution and protein tree topologies provide
evidence for both endosymbiotic gene transfer and a horizontal gene transfer ratchet driven by recurrent endoparasitism as
explanations for gene origin.
Conclusions/Significance: Our findings paint a more complex picture of gene origin than can easily be explained by
endosymbiotic gene transfer from an organelle-like point source. These data significantly extend the genomic impact of
Chlamydiae on Plantae and show that about one-half (30/55) of the transferred genes are most closely related to sequences
emanating from the genome of the only environmental isolate that is currently available. This strain, Candidatus
Protochlamydia amoebophila UWE25 is an endosymbiont of Acanthamoeba and likely represents the type of endoparasite
that contributed the genes to Plantae.
Citation: Moustafa A, Reyes-Prieto A, Bhattacharya D (2008) Chlamydiae Has Contributed at Least 55 Genes to Plantae with Predominantly Plastid Functions. PLoS
ONE 3(5): e2205. doi:10.1371/journal.pone.0002205
Editor: Robert DeSalle, American Museum of Natural History, United States of America
Received January 18, 2008; Accepted April 7, 2008; Published May 21, 2008
Copyright: � 2008 Moustafa et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was supported by the National Aeronautics and Space Administration, the National Science Foundation, and the National Institutes of
Health in grants awarded to DB (NNG04GM17G, EF 04-31117, R01ES013679, respectively).
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: debashi-bhattacharya@uiowa.edu
. These authors contributed equally to this work.
Introduction
The origin of the photosynthetic organelle (plastid) in eukaryotes
occurred via the capture and enslavement of a cyanobacterium
(primary endosymbiosis) [1,2]. This ‘‘primary’’ plastid is shared by
the red, green (including land plants), and glaucophyte algae
(together the Plantae) [3,4]. Under the most parsimonious
scenario, the Plantae share a unique common branch that defines
the point of entry of the primary endosymbiont [1,5,6,7], although
the monophyly of this group remains to be unambiguously
demonstrated using phylogenetic analysis of nuclear genes [8,9].
The descendants of the first algae came to dominate many aquatic
environments and ultimately gave rise to land plants [10]. Plastid
characters shared by the Plantae include a complex protein import
system (TIC-TOC translocons) and a similar genome architecture
and gene content [2,5,11,12,13,14].
Here we use phylogenomics to search for genes contributed to
Plantae by a surprising source, Chlamydiae bacteria. These
prokaryotes are well known as obligate intracellular vertebrate
pathogens and encode a unique gene, the ADP/ATP translocase
to parasitize energy from the host. This gene is shared by
Chlamydiae, Rickettsiales, microsporidians, and photosynthetic
eukaryotes. Phylogenetic analysis has demonstrated a chlamydial
origin of the plastid-targeted ADP/ATP translocator in algae and
plants (e.g., [15,16,17]). Interest in the Chlamydiae-plant connec-
tion was originally raised by the finding of an affinity for several
genes in the sequenced genomes of Chlamydia trachomatis and
UWE25 (i.e., Candidatus Protochlamydia amoebophila) to plant
homologs [18,19]. Later analyses showed that many of these
proteins (not just the ADP/ATP translocator) are plastid-targeted
in plants [15]. These findings led to a number of different
hypotheses to explain chlamydial gene origin in photosynthetic
PLoS ONE | www.plosone.org 1 May 2008 | Volume 3 | Issue 5 | e2205

eukaryotes including an ancient, unappreciated relationship
between Chlamydiae and the cyanobacterial donor of the plastid
[15], the possibility that infected insects may have been the vectors
for introducing Chlamydiae genes into plants [20], and ancient
horizontal gene transfer from Chlamydiae to the Plantae ancestor
that may have played a role in plastid establishment (e.g., [17,21]).
The most complete analysis to date of the Chlamydiae-Plantae
connection is a phylogenomic study that, as reported, found at
least 21 genes of chlamydial origin among the 4,771 predicted
proteins in the genome of the extremophilic red alga Cyanidioschy-
zon merolae [22]. Virtually all of these Chlamydiae genes encode a
plastid targeting signal, are involved in different plastid associated
tasks such as fatty acid biosynthesis and ion transport, and are
therefore postulated to have played a key role in the establishment
of essential plastid functions [22]. Given the large number
chlamydial genes that were uncovered, Huang and Gogarten
[22] postulated an ancient symbiosis between an environmental
chlamydial cell and the Plantae ancestor to account for gene origin
rather than repeated horizontal gene transfers (HGTs) from
different Chlamydiae. Under their view, the chlamydial endosym-
biosis could have been a mutualistic, parasitic, or a commensal
relationship but was long-term and stable in the Plantae host. This
three-way partnership between the host, the cyanobacterial plastid
ancestor, and an environmental Chlamydia-like cell was thought
to have played a fundamental role in plastid evolution [22].
Here we reexamine the evolutionary relationship between
Chlamydiae and Plantae genes using a phylogenomic approach
that incorporates predicted proteins from 17 Plantae genomes to
query .500 eukaryotic and prokaryotic genomes in a local
database. We use our recently developed automated tree-sorting
tool PhyloSort [23] to identify candidate trees (genes). Unlike the
previous study [22] however, about two-thirds of the Chlamydiae
genes we found are clearly of plastid function, whereas the
remainder serve a diversity of potential functions including three
that encode a putative mitochondrial targeting sequence. These
data provide strong evidence for a long-term symbiotic association
vis a vis Huang and Gogarten [22] of Chlamydia-like cells with the
Plantae ancestor that extends beyond plastid establishment. The
association may have been one of recurrent infections by one or
more specific endoparasite(s) of the Plantae host, as occurs in
modern-day environmental Chlamydiae and their eukaryotic hosts
[16]. Another possibility (e.g., [22]) is an endosymbiotic, organelle-
like association. Although it is currently unknown which (or both)
of these explanations may be correct, a ratchet (e.g., [24,25])
driven by horizontal gene transfer (HGT) from the parasite(s)
using a type IV secretion system could readily explain the
movement of many Chlamydiae genes into the host nucleus. In
either case, the cyanobacterium provided the critical function
(photosynthesis) and was retained as a compartment, whereas the
Chlamydiae provided key genes through endosymbiotic gene
transfer (EGT) and/or HGT. These latter cells were however
eliminated by the host, due perhaps to costs they placed on host
cell fitness (i.e., as a result of energy-parasitism, [17,26]).
Results
Our analyses identified at least 55 Plantae proteins (52 trees at
$75% and 3 trees at $50% RAxML [27] bootstrap support) that
are putatively derived from a Chlamydia-like source (Table 1). Of
these, 37/55 (67%) encoded a putative plastid targeting sequence.
The remainder of the proteins were putatively of non-plastid
function (e.g., involved in protein translation; see Table 1) based
on organelle targeting predictions using TargetP [28], Predotar
[29], ChloroP [30], PSLDoc [31], and WoLF PSORT [32] and
the gene ontology (GO) annotation of the Arabidopsis homolog
when available, or other plants or algae when not. However,
because one of these non-plastid proteins encoded an N-terminal
extension (see Table 1) in comparison to the Chlamydiae and
other prokaryotic homologs, it is possible that it also has an
organellar target or alternatively, is a membrane protein.
Thirty-one of the Chlamydiae genes were present in the green
algae plus plants clade (with or without chromalveolates) and 20
were shared by at least red and green algae, thereby suggesting
their ancient origins in the Plantae (see Figure 1). An expanded list
of protein characteristics is provided in Table S1 and the RAxML
bootstrap trees are presented in Table S2. Our ability to identify a
larger set of Chlamydiae genes than Huang and Gogarten [22]
likely reflected the fact that we used the combined protein set from
17 Plantae genomes, thereby including as large a diversity of query
sequences as possible. As also noted by Huang and Gogarten [22],
C. merolae has a highly reduced nuclear genome (16.5 Mb; 5,331
genes [33]), therefore some genes (e.g., Fig. 1A, 1C) absent from
this species could still be present in the ‘‘normal-sized’’ genomes of
mesophilic green algae (e.g., Chlamydomonas reinhardtii, 120 Mb;
.15,000 genes [34]) and plants. Consistent with this idea, 32 of
the genes we found of Chlamydiae origin were undetected in red
algae. Many of these genes may have been lost from the
Cyanidiales, or diverged beyond detection using our bioinformatic
pipeline, or are independent gains in the green lineage. More
extensive data are needed from mesophilic red algae to address
this issue. Currently we only had available partial EST data from
non-Cyanidiales red algae.
Examples of novel genes we found are shown in Figs. 2, 3, and
4. In Fig. 2A, we present the phylogeny of PFC1 (a plastid-targeted
RNA methylase that is essential for low-temperature development
of chloroplasts [35]) that shows a clear affiliation of green algae to
Chlamydiae (RAxML bootstrap, RB= 80%, PHYML [36]
bootstrap, PB= 84%) and this clade is sister to cyanobacteria
(RB= 97%, PB= 87%). One possible explanation of cases with a
cyanobacteria-Chlamydiae-Plantae connection is that Chlamydiae
may be sister to or in the past exchanged genes with cyanobacteria
(i.e., the plastid donor) and therefore their close relationship is a
reflection of the bacterial tree rather than endosymbiotic gene
transfer (EGT)/HGT from the former group (see [15] and [22] for
a detailed discussion of this scenario). A different sort of topology is
shown in Fig. 2B in which there are two types of queuine tRNA-
ribosyltransferase genes (a tRNA-guanine transglycosylase; puta-
tively mitochondrial targeted in algae and plants) in Plantae, one is
putatively derived from cyanobacteria in chlorophytes (i.e.,
Chlamydomonas and Volvox) and another from Chlamydiae in red
algae, chromalveolates, and prasinophytes (i.e., Ostreococcus spp.;
PB, RB=100%). This tree is likely explained by differential gene
loss in green algae with red and prasinophyte algae retaining the
Chlamydiae gene (that was subsequently transferred to chromal-
veolates via red or green algal secondary endosymbiosis) and green
algae the cyanobacterial copy. A third example of the types of
genes we found is glgA (glycogen synthase) that is shown in Fig. 3.
This gene catalyzes starch synthesis and there are two gene copies
in plants, one that is derived from Chlamydiae (specifically
Candidatus Protochlamydia amoebophila UWE25; RB=69%,
PB=75%) and another that is shared by many greens and is
derived from an unknown prokaryotic source. Both genes function
in the chloroplast. This is the second gene of chlamydial origin
that is involved in a carbohydrate metabolic process (i.e., the first is
the starch debranching enzyme ATISA3; Table 1). An interesting
point about Fig. 3 is that it shows a specific relationship between
UWE25 and Plantae (see also [22]). This environmental
Chlamydiae species (symbiont in Acanthamoeba, [18]) was found
Chlamydiae Genes in Plantae
PLoS ONE | www.plosone.org 2 May 2008 | Volume 3 | Issue 5 | e2205

to be sister to Plantae in 30/55 trees (Table S1) identifying it as the
closest living relative in our data set of the endoparasite that
donated the genes. Complete genomes from other environmental
Chlamydiae are needed to more comprehensively address the issue
of the potential gene donor(s) in Plantae. UWE25 is a member of
ECL V clade of Chlamydiae [20] with a genome (2.4 Mb) that is
twice the size of the more highly derived vertebrate pathogens and
in contrast to the latter, retains a complete TCA cycle. UWE25
however, still encodes an ADP/ATP translocator and is
presumably an endoparasite of Acanthamoeba [16]).
Table 1. The 55 nuclear genes of chlamydial origin that we found in genome data from 17 Plantae species, and the putative
cellular localizations of the encoded proteins.
Plantae gene annotation Localization Plantae gene annotation Localization
Dimethyladenosine transferase (PFC1) Chloroplast 4-diphosphocytidyl-2C-methyl-D-erythritol kinase Chloroplast
Unknown protein (contains nucleotide-diphospho-sugar
transferases domain)
Chloroplast Na+/H+ antiporter, putative Chloroplast
Phosphate transporter 2;1 (PHT2;1) Chloroplast Anthranilate phosphoribosyl transferase Chloroplast
Phosphoglycerate/bisphosphoglycerate mutase family protein Chloroplast LL-diaminopimelate aminotransferase (AGD2) Chloroplast
Exonuclease family protein Chloroplast Heavy metal ATPase 1 (HMA1) Chloroplast
Pseudouridine synthase family protein Chloroplast Oligoendopeptidase F Chloroplast
Malate dehydrogenase (NADP) Chloroplast Conserved hypothetical protein Chloroplast
Phosphoribosylanthranilate isomerase (PAI2) Chloroplast Copper/Zinc superoxide dismutase family protein Chloroplast
Granule-bound starch synthase I (Glycosyl transferase) Chloroplast Carbonic anhydrase 2 (CA2) Chloroplast
D-alanine-D-alanine ligase B Chloroplast 50S ribosomal protein-related Mitochondrion
Plastidic ATP/ADP transporter Chloroplast Queuine tRNA-ribosyltransferase Mitochondrion
Putative SAM dependent methyltransferases Chloroplast Manganese and iron superoxide dismutase Mitochondrion
Cytidylyltransferase family Chloroplast Plasma membrane intrinsic protein 1c (PIP1C) Membrane
tRNA/rRNA methyltransferase (SpoU) family protein Chloroplast Glycerol-3-phosphate transporter Membrane
Unknown protein (S-adenosyl-L-methionine-dependent
methyltransferases domain)
Chloroplast Prolyl 4-hydroxylase, alpha subunit N-terminal ext
Enoyl-[acyl-carrier-protein] reductase (MOD1) Chloroplast Unknown protein
Rhodanese-like domain containing protein Chloroplast Sugar isomerase (SIS) domain-containing protein
4-hydroxy-3-methylbut-2-en-1-yl diphosphate synthase (GcpE) Chloroplast Unknown protein (similar to zinc finger family protein)
Pyrophosphate-dependent phosphofructokinase PfpB Chloroplast Dihydrouridine synthase, DuS
UDP-glucuronate 4-epimerase 4 (GAE4) Chloroplast RNA-binding region containing protein
3-oxoacyl-(acyl-carrier-protein) synthase I (KAS I) Chloroplast Lipoate protein ligase-like protein
Isoamylase 3 (ISA3) Chloroplast Leucine rich repeat proteins
Aminoacyl-tRNA synthetase, class Ib Chloroplast 39(29),59-bisphosphate nucleotidase (SAL2)
(phosphatidylinositol phosphatase)
2-C-methyl-D-erythritol 4-phosphate cytidyltransferase (ISPD) Chloroplast tRNA isopentenyltransferase (ATIPT9)
Methylase-related Chloroplast FOG: PPR repeat
Conserved hypothetical protein Chloroplast Cytidine/deoxycytidylate deaminase
Glycerol-3-phosphate acyltransferase Chloroplast Predicted nucleic acid-binding protein ASMTL
Polyribonucleotide phophorylase Chloroplast
doi:10.1371/journal.pone.0002205.t001
Reds or Reds/Chromalveolates
Greens
Greens/Reds and any other algal group
Greens/Chromalveolates and
Green/Chromalveolates/Glaucophytes
Glaucophytes
Figure 1. Pie chart showing the distribution of Chlamydiae-like genes among Plantae and chromalveolates.
doi:10.1371/journal.pone.0002205.g001
Chlamydiae Genes in Plantae
PLoS ONE | www.plosone.org 3 May 2008 | Volume 3 | Issue 5 | e2205

An important insight from our study is the identification of 18
genes of putative non-plastid functions that were contributed by
Chlamydiae to Plantae. As described above, these are bioinfor-
matic predictions for the cellular location that await verification
using proteomic methods and many contain a N-terminal
extension potentially indicating a plastid target. In spite of these
caveats, our data suggest that the relationship between Chamydiae
and their Plantae hosts was likely to not have been limited to
plastid functions but rather affected the mitochondrion and other
aspects of the nuclear genome. In addition, we found two proteins
(cytidine/deoxycytidylate deaminase family protein and an
unknown protein, similar to zinc finger family protein) in which
only the C-terminal domain was of Chlamydiae origin. This
suggests the fusion of a eukaryotic and a prokaryotic sequence
gave rise to these genes. Finally, a number of proteins were
identified (acid phosphatase survival protein, SurE, gi:15218620;
embryo defective 2394 gi:15220443; oxoglutarate:malate anti-
porter, DIT1, gi:30684152; RAN GTPase activator, RANGAP2,
gi:15239712; peptide deformylase, PDF1B, gi:15241461; ubiqui-
tin-protein ligase, EBF1, gi:18400846; mechanosensitive ion
channel domain-containing protein, gi:22328173; and ribosome
recycling factor family protein, gi:42563413) that showed Plantae-
Chlamydiae monophyly but fell below the 50% bootstrap
threshold. These trees were not counted in our final tally but
may in the future turn out to also be of chlamydial origin.
Discussion
Although the origin of the 55 Chlamydiae-like genes in Plantae
may appear to be most easily explained by a long-term
endosymbiotic (e.g., organelle-like) association between these
prokaryotes and the host, it is also worth considering whether
these sequences may have arisen from multiple Chlamydiae
sources and could simply reflect a long history of endoparasitism.
For example, even though many (20) of the Chlamydiae-like genes
were present in both red and green algae (and chromalveolates via
EGT), others were detected only in red algae (3) and glaucophytes
(1). This pattern could be explained either by wholesale gene loss
in Plantae lineages or multiple HGTs in these taxa. Speaking
against the latter scenario is the apparent absence of Chlamydiae
endoparasites in extant algae and plants, although this certainly
may not have been the case ca. 1 billion years ago when Plantae
0.1 substitutions/site
0.1 substitutions/site
Ostreococcus tauri
Ostreococcus lucimarinus
Aureococcus anophagefferens
Pavlova lutheri
Cyanidioschyzon merolae
Thalassiosira pseudonana 36917
Phaeodactylum tricornutum
Thalassiosira pseudonana 1283
Galdieria sulphuraria
Chlamydia muridarum
Chlamydia trachomatis
Chlamydophila caviae
Chlamydophila abortus
Chlamydophila felis
Chlamydophila pneumoniae
Wolbachia endosymbiont
Ehrlichia canis
Anaplasma marginale
Neorickettsia sennetsu
Chlamydomonas reinhardtii
Volvox carteri
Synechococcus elongatus PCC6301
LyngbyaSpPCC8106
Nostoc sp. PCC7120
Anabaena variabilis ATCC29413
Nostoc punctiformis PCC73102
Nodularia spumigena CCY9414
Deinococcus radiodurans
Halothermothrix orenii
Thermosinus carboxydivorans
Clostridium perfringens
Magnetococcus sp.
Rhodobacter sphaeroides
Chloroflexus aggregans
Roseiflexus sp.
Bifidobacterium adolescentis
Arthrobacter aurescens
Tropheryma whipplei
Ostreococcus tauri
Ostreococcus lucimarinus
Candidatus Protochlamydia amoebophila UWE25
Candidatus Protochlamydia amoebophila UWE25
Chlamydia muridarum
Chlamydia trachomatis
Chlamydophila caviae
Volvox carteri
Synechococcus elongatus PCC6301
Lyngbya sp. PCC8106
Nostoc sp. PCC7120
Anabaena variabilis ATCC29413
Nostoc punctiformis PCC73102
Nodularia spumigena CCY9414
Trichodesmium erythraeum IMS101
Cyanothece sp. CCY0110
Crocosphaera watsonii WH8501
Prochlorococcus marinus MIT9301
Medicago truncutula
Arabidopsis thaliana
Zea mays
Oryza sativa
Picea glauca
Lactobacillus
delbrueckii
Bacillus subtilis
Geobacillus thermodenitrifican
Listeria monocytogenes
Streptococcus pyogenes
Lactococcus lactis
Pediococcus
pentosaceus
Lactobacillus brevis
Lactobacillus salivarius
Clostridium sp.
Alkaliphilus metalliredigens
Clostridium novyi
Ruminococcus torques
Eubacterium ventriosum
Bacteroides capillosus
Halothermothrix orenii
Pelotomaculum thermopropionicum
Desulfotomaculum reducens
Moorella thermoacetica
Symbiobacterium thermophilum
Geobacter sp.
Geobacter metallireducens
A B
86
80
95
99
84
61
76
73
95
86
80
79
99
97
77
61
74
64
73
97
100
100
100
100
100
100
100
100
100
87
84
99
90
88
98
71
71
92
89
93
91
85
73
64
100
100
100
100
100
100100
100
100
100
100
100
100
75
100
100
100
100
100
100
100
88
91
100
98
69
95
95
76
72
98
94
94
94
70
70
92
68
84
100
91
99
100
100
100
100
100
100
100
82
62
100
77
70
96
96
67
84
93
95
92
95
98
97
65
67
97
Figure 2. Maximum likelihood (RAxML) trees of Chlamydiae-derived genes in the Plantae. A) The tree of dimethyladenosine transferase
(PFC1). B) The tree of queuine tRNA-ribosyltransferase. The results of a bootstrap analysis using RAxML are shown above the branches, whereas
PHYML bootstrap support values are shown below the branches. Only bootstrap values $60% are shown. Branch lengths are proportional to the
number of substitutions per site (see scale bars). Cyanobacteria are shown in blue text, green algae and land plants in green text, red algae in red,
chromalveolates in brown, and Chamydiae in magenta. All other bacteria are shown in black text. The thick branches unite Chlamydiae and Plantae.
doi:10.1371/journal.pone.0002205.g002
Chlamydiae Genes in Plantae
PLoS ONE | www.plosone.org 4 May 2008 | Volume 3 | Issue 5 | e2205

radiated [18]. UWE25 plays a key role in this discussion. This
strain represents an anciently diverged lineage whose ancestors
were likely contemporary with the Plantae ancestor. However the
environmental Chlamydiae are represented by only a single
genome in our analysis (i.e., UWE25). Therefore although many
UWE25 genes had a sister group relationship with Plantae (i.e.,
30/55), in other trees UWE25 was either not present (12 trees) or
when in the tree did not show a specific affiliation with Plantae to
the exclusion of other Chlamydiae (13 cases). An example of the
latter group is shown in Figure 2A. Furthermore, as described
above, because the vertebrate pathogens have highly reduced
genomes, many Chlamydiae genes that are shared uniquely by
UWE25 and Plantae (19 genes) may simply be cases of widespread
gene loss in other endoparasites, giving the misleading impression
of a specific relationship between UWE25 and algae/plants. The
addition of many more environmental Chlamydiae genomes may
help us determine whether the ‘‘big-genomed’’ environmental
UWE25 shares a specific relationship with Plantae or whether
other environmental strains that also contain these genes would
break this relationship. We must however keep in mind the
0.1 substitutions/site
Arabidopsis thaliana 15232051
Oryza sativa 115482170
Oryza sativa 115467234
Oryza sativa 115448649
Physcomitrella patens 222462
Physcomitrella patens 177976
Physcomitrella patens 202950
Volvox carteri 103172
Chlamydomonas reinhardtii 109372
Ostreococcus lucimarinus 2524
Ostreococcus tauri 27007
Ostreococcus tauri 8953
Ostreococcus lucimarinus 41777
Oryza sativa 115466564
Arabidopsis thaliana 15237934
Arabidopsis thaliana 42566924
Triticum aestivum 15717885
Physcomitrella patens 200110
Physcomitrella patens 120571
Physcomitrella patens 70198
Oryza sativa 115471703
Oryza sativa 115466210
Arabidopsis thaliana 15223331
Myxococcus xanthus
Burkholderia phytofirmans
Novosphingobium aromaticivorans
Rhodopseudomonas palustris
Shigella flexneri
Polaromonas naphthalenivorans
Magnetospirillum magneticum
Methylobacillus flagellatus
Dechloromonas aromatica
Nitrosospira multiformis
Nitrosomonas europaea
Azoarcus sp.
Anaeromyxobacter dehalogenans
Physcomitrella patens 111318
Physcomitrella patens 121916
Physcomitrella patens 82584
Physcomitrella patens 149648
Physcomitrella patens 96400
Oryza sativa 115465085
Crocosphaera watsonii WH8501
Synechocystis sp. PCC680
Nostoc sp. PCC7120
Anabaena variabilis ATCC29413
Methylococcus capsulatus
Flavobacterium johnsoniae
Mycoplasma mobile
Geobacter bemidjiensis
Caldicellulosiruptor saccharolyticus
Alkaliphilus metalliredigens
Bacillus subtilis
Halothermothrix orenii
Clostridium cellulolyticum
Hermotoga lettingae
Lactococcus lactis
Streptococcus suis
Vibrio cholerae
Herpetosiphon aurantiacus
Roseiflexus sp.
Treponema denticola
Sulfurovum sp.
Synechococcus sp. CC9605
Prochlorococcus marinus str. NATL1
100
69
95
90
90
100
97
64
99
79
74
69
93
76
80
80
80
68
80
62
100
100
100
100
100
100
92
98
70
89
83
76
66
98
74
67
99
99
87
82
85
79
86
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
75
Candidatus Protochlamydia amoebophila UWE25
Figure 3. Maximum likelihood (RAxML) tree of the Chlamydiae-derived Plantae protein, glycogen synthase. The results of a bootstrap
analysis using RAxML are shown above the branches, whereas PHYML bootstrap support values are shown below the branches. Only bootstrap
values $60% are shown. Branch lengths are proportional to the number of substitutions per site (see scale bar). Cyanobacteria are shown in blue
text, green algae and land plants in green text, and Chamydiae in magenta. All other bacteria are shown in black text. The thick branches unite
Chlamydiae and Plantae.
doi:10.1371/journal.pone.0002205.g003
Chlamydiae Genes in Plantae
PLoS ONE | www.plosone.org 5 May 2008 | Volume 3 | Issue 5 | e2205