Differential Gene Retention in Plastids of Common Recent
Origin
Adrian Reyes-Prieto, à,1,2 Hwan Su Yoon,à,3 Ahmed Moustafa,1,2 Eun Chan Yang,3
Robert A. Andersen,3 Sung Min Boo,4 Takuro Nakayama,5 Ken-ichiro Ishida,5 and
Debashish Bhattacharya*,1,2
1
Department of Ecology, Evolution and Natural Resources, Rutgers University
Institute of Marine and Coastal Sciences, Rutgers University
3
Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, ME
4
Department of Biology, Chungnam National University, Daejeon, Korea
5
Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki, Japan
Present address: Department of Biology, University of New Brunswick, Fredericton, New Brunswick, Canada
àThese authors contributed equally to this work.
*Corresponding author: E-mail: [email protected].
Associate editor: Martin Embely
2
The cyanobacterium-derived plastids of algae and plants have supported the diversification of much of extant eukaryotic
life. Inferences about early events in plastid evolution must rely on reconstructing events that occurred over a billion years
ago. In contrast, the photosynthetic amoeba Paulinella chromatophora provides an exceptional model to study organelle
evolution in a prokaryote–eukaryote (primary) endosymbiosis that occurred approximately 60 mya. Here we sequenced
the plastid genome (0.977 Mb) from the recently described Paulinella FK01 and compared the sequence with the existing
data from the sister taxon Paulinella M0880/a. Alignment of the two plastid genomes shows significant conservation of
gene order and only a handful of minor gene rearrangements. Analysis of gene content reveals 66 differential gene losses
that appear to be outright gene deletions rather than endosymbiotic gene transfers to the host nuclear genome.
Phylogenomic analysis validates the plastid ancestor as a member of the Synechococcus–Prochlorococcus group, and the
cyanobacterial provenance of all plastid genes suggests that these organelles were not targets of interphylum gene transfers
after endosymbiosis. Inspection of 681 DNA alignments of protein-encoding genes shows that the vast majority have dN/
dS ratios ,,1, providing evidence for purifying selection. Our study demonstrates that plastid genomes in sister taxa are
strongly constrained by selection but follow distinct trajectories during the earlier phases of organelle evolution.
Key words: endosymbiosis, primary plastids, photosynthesis, Paulinella.
Introduction
The ancient origins of mitochondria and plastids explain the
fundamentally chimeric nature of eukaryotes (Sagan 1967).
Photosynthesis entered the eukaryotic domain via primary
endosymbiosis, whereby a cyanobacterium was captured by
a heterotrophic protist and converted into a photosynthetic
organelle. This pivotal event occurred more than a billion
years ago and laid the foundation for many food webs on
our planet (Falkowski et al. 2004; Reyes-Prieto et al. 2007).
The host lineage for the endosymbiosis is the putative ancestor of the Plantae that subsequently splits into the glaucophyte, red, and green algae (including land plants;
Cavalier-Smith 1992; Bhattacharya and Medlin 1995; Palmer
2003). The canonical Plantae plastid spread via secondary
and tertiary endosymbiosis to other lineages such as chromalveolates and euglenids (Palmer 2003; Bhattacharya et al.
2004; Reyes-Prieto et al. 2007). All extant plastids, whether of
primary, secondary, or tertiary origin, are specialized organelles with highly reduced genomes (100–200 kb), leaving us
to speculate about the pattern and process of gene loss early
in their evolution. Given this situation, there is much interest
in identifying a more recent case of organelle establishment
via primary endosymbiosis. This need appears to have been
recently fulfilled with molecular studies of the thecate
amoeba Paulinella chromatophora M0880/a (Marin et al.
2005, 2007; Yoon et al. 2006; Nowack et al. 2008). Paulinella
is a member of the supergroup Rhizaria, yet it contains two
blue-green ‘‘chromatophores’’ (fig. 1A and B) that resulted
from a novel plastid acquisition (Marin et al. 2005; Yoon et al.
2006; Nowack et al. 2008) about 60 mya (Nowack et al.
2008). Several lines of evidence support the hypothesis that
Paulinella contains bona fide photosynthetic organelles.
These include a constant plastid number (i.e., two per cell)
following each round of coordinated cell division (Kies 1974),
a plastid genome that is one-third the size of chromosomal
DNA in putative free-living cyanobacterial donors (Nowack
et al. 2008), and evidence for gene transfer to the amoeba
nuclear genome via endosymbiotic gene transfer (EGT) (i.e.,
the cyanobacterium-derived psaE gene; Nakayama and
Ishida 2009).
© The Author 2010. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved. For permissions, please
e-mail: [email protected]
1530
Mol. Biol. Evol. 27(7):1530–1537. 2010 doi:10.1093/molbev/msq032
Advance Access publication February 1, 2010
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
Research article
Abstract
Paulinella Plastid Evolution · doi:10.1093/molbev/msq032
MBE
Here we generated the plastid genome sequence from a sister taxon of Paulinella M0880/a that was recently isolated in
Japan (Yoon et al. 2009). This second isolate, Paulinella FK01,
provides anidealtool tounderstand plastidgenome evolution
using a homologous organelle of recent origin. A strategy utilizing fluorescence-activated cell sorting (FACS) to isolate organelles was followed by single-cell genomics and ‘‘454’’
pyrosequencing to generate a draft genome of FK01 that
was closed using targeted polymerase chain reaction (PCR).
Materials and Methods
Cell Isolates
Paulinella FK01 was established from a single cell collected
at Daigo-machi, Ibaraki prefecture, Japan (Yoon et al. 2009).
Paulinella chromatophora M0880/a was kindly provided
by Michael Melkonian (University of Cologne, Germany).
Both isolates are maintained at the Bigelow Laboratory
for Ocean Sciences using DY-V medium at 20 °C with a
14/10 h light/dark cycle.
Plastid Isolation and Whole-Genome
Amplification
After cell disruption with glass-bead beating, single plastids were isolated using FACS (see supplementary fig. S1,
Supplementary Material online). DNA derived from 50
isolated plastids was used for genome amplification using
the Repli-G Mini Kit (Qiagen), which applies multiple displacement amplification (MDA) methods (Stepanauskas
and Sieracki 2007). This resulted in approximately 10 lg
of DNA per reaction with an A260/280 ratio of 1.85. After the de-branching step with S1 nuclease to reduce
chimeric sequences during MDA, a PCR survey was done
using several gene markers for nuclear and plastid (pt)
DNA (16S ribosomal DNA [rDNA], 18S rDNA, rbcL, ftsZ,
beta-tubulin) to validate the source of the nucleic acids,
that is, the nuclear gene amplifications acted as negative
controls for this procedure. The PCR products were
sent to the DNA Facility at the University of Iowa for
Sanger sequencing, whereas the amplified ptDNA
was sent to Macrogen (Seoul, Korea) and to the University of Iowa for 454 (Roche Diagnostics Corporation)
pyrosequencing.
Genome Sequencing, Assembly, and Annotation
Combination of two separate runs of a ¼ plate each from
Macrogen and the University of Iowa using the Genome
Sequencer FLX standard chemistry generated a total of
;2.4 million sequences with an average length of 230 bases,
providing .55 theoretical coverage of the FK01 plastid
1531
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
FIG. 1. (A) Maximum likelihood phylogenetic tree inferred using 16S rDNA sequences from plastids, cyanobacteria, and other bacteria. Numbers
above the branch nodes denote support values (when 50%) from RAxML bootstrap replicates, and thick branches indicate Bayesian posterior
probabilities .0.95. Branch lengths are proportional to the number of substitutions per site (see scale bar). (B) Light microscopy and scanning
electron microscopy images from Paulinella FK01: A, oral scales; C, chromatophores (plastids); and S, scales.
MBE
Reyes-Prieto et al. · doi:10.1093/molbev/msq032
genome. The filtered reads were assembled using the GS De
Novo Assembler software (Roche Diagnostics Corporation), resulting in 11 contigs. Using as reference both the
M0880/a plastid genome and the termini of the FK01 contigs, we designed specific PCR primers to complete the sequence of the missing regions in FK01. The final assembly
was done using Sequencher. A total of 888 protein-coding
genes were predicted in FK01 ptDNA using GeneMarkS
(Besemer et al. 2001) and 912 proteins using RAST (Aziz
et al. 2008). Visual inspection and manual refinement substantiated a total of 841 likely protein-encoding gene models. After annotation, alignment of the Paulinella M0880/
a and FK01 ptDNA was done under Mauve (Darling et al.
2004) using the progressive algorithm with default parameters. The FK01 plastid genome sequence is available from
the National Center for Biotechnology Information under
the Genome Project ID 45921.
PCR Reactions
Phylogenomic Analysis
Genome Sequencing
We prepared a local database that included more than
500 genome sequences and expressed sequence tag (EST)
libraries, with more than 6,000,000 protein sequences (e.g.,
Moustafa et al. 2009). This database comprised all completely sequenced bacterial and cyanobacterial genomes,
representatives of the six major eukaryotic supergroups,
and organelle-encoded proteins including the M0880/a
plastid data. The 841 predicted proteins from the FK01
plastid genome were used as Blast queries against the
genome database with e-value thresholds of 1 10 5
and 1 10 10 to address nonconserved and conserved
queries, respectively (see Moustafa et al. 2009). Each query
sequence and its significant homologs were aligned using
Multiple Alignment using Fast Fourier Transform (MAFFT)
(Katoh et al. 2002). Sequence alignments were used to infer
phylogenetic trees using PhyML (Guindon and Gascuel
2003) with the approximate likelihood ratio test (aLRT;
Anisimova and Gascuel 2006). The phylogenetic trees were
grouped based on their topological patterns using PhyloSort (Moustafa and Bhattacharya 2008) with a minimum
threshold set at 75% aLRT support values for groups of
interest.
We used flow cytometric single-cell sorting followed by
whole-genome amplification and pyrosequencing (Yoon
et al. 2009) to determine the plastid genome sequence
from the recently described taxon Paulinella FK01 (Yoon
et al. 2009; fig. 1B). We then compared its genome structure
and gene composition to the previously reported ptDNA
from Paulinella M0880/a (Nowack et al. 2008). The ptDNA
in FK01 is a circular molecule of 977,329 bases that encodes
841 predicted proteins (supplementary table S2, Supplementary Material online) and 48 structural RNAs. Alignment of FK01 and M0880/a ptDNA (fig. 2) reveals
overall conservation of gene order, but there are five genome inversions involving fragments of sizes 110.3 kb,
24.9 kb, 3.9 kb, 2.1 kb, and 569 bp and a single 9.2-kb translocation. Phylogenomic analysis of the predicted plastid
proteins indicates that they are all derived from cyanobacteria, and in most trees (820/841, 97.5%), Paulinella genes
branch as sister to the cyanobacterial clade containing
Synechococcus–Prochlorococcus species (Badger et al. 2002;
Marin et al. 2005, 2007; Yoon et al. 2006, 2009). The majority (814/841, 96.7%) of FK01 protein-encoding genes
have orthologs in M0880/a, and in most cases (.98%),
these sequences branch as sisters in the phylogenetic trees
(results not shown; all trees and alignments are available
from http://dblab.rutgers.edu/home/downloads/).
A total of 33 genes were apparently acquired by the cyanobacterial donor of the Paulinella plastid via horizontal
gene transfer (HGT) from other bacterial sources, prior to
endosymbiosis (supplementary table S3, Supplementary
Material online). This group comprises 26 genes (including
eight that were previously reported; Marin et al. 2007) of
proteobacterial origin and 7 of unresolved affiliation.
Therefore, as observed in Plantae plastids (e.g., Rice and
Palmer 2006), the Paulinella photosynthetic organelle is
protected from HGT, that is, gene loss, not gain characterizes this earlier phase in organelle evolution. Estimations of
nonsynonymous (dN) and synonymous substitution (dS)
rates for 681 ortholog pairs from both Paulinella isolates
reveal that all dN/dS ratios are ,1 with the vast majority
We used the 841 predicted nucleotide-coding regions from
FK01 as BlastN (e-value threshold 1 10 5) queries to
identify the corresponding orthologous DNA regions from
M0880/a ptDNA. Using this threshold, we identified 681
orthologous pairs (75% of the total predicted proteincoding genes in FK01) that could be reliably aligned using
DNA data. Each ortholog pair was aligned with MAFFT
(Katoh et al. 2002) using a maximum of 1,000 iterations
for alignment refinement. Individual alignments were visually inspected and edited to conserve the codon structure
encoding the corresponding predicted protein. We estimated the ratio of nonsynonymous (dN) to synonymous
(dS) substitutions for all 681 codon-based DNA alignments
using CONSEL that is included in the PAML suite (Yang
2007).
1532
Results and Discussion
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
Nucleotide Substitution Rate Estimation
Total genome DNA from both Paulinella species was extracted using the DNeasy Plant Mini Kit (Qiagen). PCR reactions to determine possible cases of EGT were done using
specific primer pairs (see supplementary table S1, Supplementary Material online) for 27 genes encoded only in the
FK01 plastid genome and for 19 genes exclusively present
in M0880/a ptDNA. In all 46 cases, total DNA from each
species was used as the template in independent PCR reactions. All reactions were done with an initial denaturation step at 94 °C for 10 min, followed by 35 cycles of
94 °C for 1 min, 53–57 °C for 1 min, and 72 °C for 2
min, concluding with a 10-min extension at 72 °C. PCR
products were purified and either directly sequenced or
cloned prior to sequencing.
MBE
Paulinella Plastid Evolution · doi:10.1093/molbev/msq032
kb
100
200
300
400
500
600
700
800
900
1000
M0880/a
FK01
FIG. 2. Alignment of the Paulinella M0880/a and FK01 plastid genomes (see Materials and Methods). The putative origin of replication of
M0880/a ptDNA (Nowack et al. 2008) was used to align the genomes. The diagonal lines indicate genome rearrangements.
Differential Gene Loss
Gene-by-gene comparison of FK01 and M0880/a ptDNA
reveals 27 genes encoded in FK01 that are absent from
M0880/a, whereas 39 genes in M0880/a are absent from
FK01 (supplementary table S5, Supplementary Material online). These 66 genes trace their origin to the Synechococcus–Prochlorococcus group of cyanobacteria suggesting
that they were present in the ancestor of these photosynthetic Paulinella and are examples of lineage-specific losses.
We investigated whether these plastid gene losses may be
explained by EGT to the host nucleus (e.g., as for psaE) or by
outright loss. A total of 46 lineage-specific genes (27 specific
to FK01 and 19 specific to M0880/a) were targeted by PCR
with gene-specific primers and total amoeba DNA. Here if
the gene was absent in ptDNA but provided a product
when analyzing total DNA, we interpreted this as a putative
case of EGT. Positive controls were provided by total DNA
from the lineage that we knew contained a copy of the
gene in the plastid (although additional nuclear copies
could be present in these cases). This approach provided
unambiguous results suggesting that none of the 46 lineage-specific plastid genes are present in total DNA when
absent from the plastid but result in PCR fragments from
positive control DNA (results not shown), that is, they are
not candidates for EGT and likely constitute complete losses. It is of course possible that high sequence divergence or
the insertion of spliceosomal introns that interrupt the PCR
primer sites explains some of these results. However, given
the overall trend, we postulate that gene loss, not EGT, ex-
plains much of the differential plastid gene loss observed in
these Paulinella isolates. We expect, however, that many
endosymbiont genes that encode critical plastid functions
have already been transferred to the nucleus of these photosynthetic amoebae and would only be uncovered by
analysis of the nuclear genome sequence.
Furthermore, when we consider genome data from
many endosymbiotic bacteria (Moran 1996; Hosokawa
et al. 2006; Pérez-Brocal et al. 2006; Moran et al. 2009),
we can assume that during organellogenesis the effective
population size of the endosymbiont is small (e.g., two plastids per amoeba), entailing restricted (or absence of) DNA
recombination. This would lead to the accumulation of deleterious mutations and shifts in base composition (Moran
1996). As a consequence, it is plausible that genetic drift
(Moran 2002; Marais et al. 2008) has been driving reduction
of the Paulinella plastid genome. Consistent with this
model (Moran 2002), both the AT content (.60%) and
the nucleotide substitution rate of the Paulinella plastid
genomes are relatively elevated (Nowack et al. 2008; Yoon
et al. 2009). The assumption that a subset of endosymbiont
genes have little or no consequences for host fitness would
explain why the size of endosymbiont genomes is significantly reduced over short evolutionary periods (Marais
et al. 2008; Moran et al. 2009). In Paulinella, we observe
that the relative distribution of plastid gene losses in
FK01 and M0880/a (i.e., 39/27, respectively, vs. the null expectation of 33/33) is consistent with genetic drift as the
underlying explanation (Fisher’s exact test, P 5 0.382). The
independent trajectory of gene retention (vs. loss) apparent in FK01 and M0880/a is also likely to be explained by
selective forces operating under local ecological conditions
although it should be noted that taxa closely related to
both Paulinella isolates analyzed here have been collected
from the same pond (Yoon HS, Yang EC, Ishida K-i,
Nakayama T, Bhattacharya D, unpublished results).
Coordinated Plastid Gene Loss in Paulinella
A striking example of plastid gene loss in Paulinella is ferredoxin-dependent glutamate synthase (Fd-GOGAT; glsF/
gltS, 4.6 kb) that is present in FK01 but absent in M0880/a.
Fd-GOGAT is involved in ammonium assimilation (Kameya
et al. 2007), and its absence results in nitrogen-deficiency
phenotypes in some cyanobacteria (see Okuhara et al.
1999). The genome alignment indicates that the two loci
(lipoyl synthase, lipA, and hypothetical protein PCC_0558)
flanking the glsF/gltS gene in FK01 ptDNA are contiguous
1533
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
(629) having a dN/dS ,0.1 (supplementary table S4, Supplementary Material online). Reduced dN/dS values reflect
an elevation in the dS value (i.e., in 97% of gene pairs, dS .
1.1; supplementary table S4, Supplementary Material online), which is generally found for endosymbiont-encoded
genes (Moran 2002; Moran et al. 2009). These results indicate that even under elevated rates of synonymous substitutions in Paulinella plastid genes (Nowack et al. 2008;
Yoon et al. 2009) and the tendency for endosymbiont genomes to accumulate deleterious mutations, purifying selection is strongly constraining amino acid substitutions in
most of the proteins encoded by M0880/a and FK01
ptDNA. A similar complex pattern of genome evolution is
also found in Plantae plastids and differs fundamentally
from insect endosymbionts that may be highly streamlined
(e.g., Carsonella ruddii; Nakabachi et al. 2006) to only a
handful of essential functions (e.g., amino acid synthesis).
MBE
Reyes-Prieto et al. · doi:10.1093/molbev/msq032
A
M0880/a
FK01
fus
rps7
hyp
rps7 rpsL PCC 0558
fus
PCC 0560
lipA
hyp
glsF/ gltS
rpsL hyp
hyp proline-rich
lipA
GOGAT
B
M0880/a
PCC 0158
FK01
hyp
FK01
pstA
dnaK 1
dnaJ
PCC 0392
hyp
pstB
tRNA
SER
PCC 0393 PCC 0394
hyp
hyp
petF
pstS
tRNA
SER
petF
syh
syh
petF
htpG
htpG
rps6
aroE
dnaK 2
dnaK 2
petF
aroE
rps6
FIG. 3. Plastid genome alignment illustrating the loss of genes encoding (A) Fd-GOGAT (glsF/gltS) in Paulinella M0880/a and (B) the four
subunits of the phosphate transporter in Paulinella FK01, pstSACB (pstA, pstB, pstC, pstS). The white blocks indicate the genes that were
differentially lost in each genome. The black blocks and gray diagonal lines indicate gene order conservation of the flanking genome regions.
in M0880/a, separated by 980 bp (fig. 3A). Fd-GOGAT and
glutamine synthetase type I (GSI; plastid encoded in both
Paulinella species) are essential for ammonium assimilation
in cyanobacteria and plastids (i.e., the GS-GOGAT cycle;
Marques et al. 1992; Muro-Pastor et al. 2005; Kameya
et al. 2007). The regulatory protein P-II is encoded in both
plastid genomes suggesting strongly that the Fd-GOGAT
locus has been transferred to the nucleus in M0880/a.
Apart from the absence of this key gene, genes encoding
ammonium transporters (amt), isocitrate dehydrogenases
(icd), transcriptional regulators of the Crp/Fnr family
(ntcA), and glutamate dehydrogenases (gdhA) are absent
from M0880/a ptDNA. These latter genes that are encoded
in the genome of Synechococcus sp. WH 5701 (Marin et al.
2005; Yoon et al. 2006) have surprisingly also been lost from
FK01 ptDNA. Given the importance of the ammonium
assimilation pathway, our working hypothesis is that some
or all of the ‘‘missing’’ genes in FK01 and M0880/a ptDNA
have been differentially transferred to the Paulinella
nucleus, although it is possible that some of these functions
may have been substituted by existing host proteins that
are now plastid targeted.
Another example of gene loss is the four (pstS and a 3gene cluster composed of pstA, pstB, and pstC) members
encoding different subunits of the ABC phosphate transporter pstSACB (Raymond et al. 2001) that are present
in M0880/a ptDNA but coordinately lost in FK01. Conservation of the homologous flanking genes in FK01 demonstrates the precise nature of gene deletions for these
1534
regions (fig. 3B). The question that remains is: how does
the FK01 plastid internalize phosphate after loss of this
high-affinity transporter? There is, for example, no evidence
of genes encoding putative low-affinity phosphate (e.g.,
pitA permeases) or phosphonate (phnCDE) ABC transporters in either plastid genome. Therefore, if all cyanobacterial pstSACB genes have been lost in FK01, as our total
DNA PCR results suggest (although this clearly provisional),
we hypothesize that phosphate (or phosphorous compounds) uptake by this organelle relies on transport mechanisms that evolved after the endosymbiosis, for example,
the co-option of host-derived transporters to support plastid metabolism.
Given the results described above, what evidence exists
thus far for EGT in Paulinella? Analysis of the .3,000 ESTs
available from Paulinella FK01 reveals that in addition to
the previously described psaE gene (Nakayama and Ishida
2009), cyanobacterial psaI, that encodes subunit VIII of
photosystem I (PSI), has been relocated to the amoeba nucleus. This gene is still identifiable in the FK01 plastid genome but has been silenced by two nonsense mutations. In
contrast, plastid-encoded psaI in M0880/a is intact. Using
FK01 total DNA as template, we amplified a genome region
that contains a psaI locus and is interrupted by a spliceosomal intron, demonstrating its nuclear derivation (fig. 4).
These results provide evidence that two photosystem I subunits are encoded in the nucleus of Paulinella and expressed as messenger RNA. Subunits produced by the
genes psaE and psaI participate, respectively, as accessory
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
M0880/a
pstC
dnaK 1
dnaJ
MBE
Paulinella Plastid Evolution · doi:10.1093/molbev/msq032
protein (PSI-subunit IV or PsaE) modulating binding between the PSI and soluble ferredoxin during electron transport and in the correct oligomerization (PSI-subunit VIII or
PsaI) of photosystem I. Structural interactions of PsaI with
the subunit PsaL (plastid encoded in Paulinella) are critical
to stabilize the trimeric state of PSI (Xu et al. 1995). Absence of PsaI has detrimental effects on the structural organization and activity of PSI in cyanobacteria (Xu et al.
1995). This limited evidence suggests that PsaI is produced
in the FK01 cytoplasm and likely imported into the plastid
to support standard PSI function. In contrast, ten other
genes encoding PSI proteins (Ycf4, Ycf4, psaC, psaA, psaB,
psaL, psaD, psaJ, psaF, psaK) are still encoded in the Paulinella plastid genome. It remains to be addressed how Paulinella PsaE and PsaI, if in fact translated in the cytosol,
enter the plastid because the ESTs do not appear to encode
an organelle transit sequence (Nakayama and Ishida 2009).
Other examples of FK01 ‘‘exclusive’’ genes are proteins
(e.g., QueC, QueD, QueD) presumably involved in queuosine
(a nucleoside) biosynthesis, two subunits of phosphoribosylformylglycinamidine synthase (de novo purine biosynthesis),
the uroporphyrinogen-III synthase (tetrapyrrole biosynthesis), S-adenosylmethionine synthetase (cysteine and methionine metabolism), and a DNA helicase. In the case of the
M0880/a ptDNA, some examples of its set of exclusive proteins are those involved in cell (i.e., organelle) division and
DNA replication, such as two adenosine triphosphate
(ATP)–dependent metalloproteases FtsH (encoded by ftsH2
and ftsH3, with 56% of identity at the protein level), and
chromosome duplication (DnaA ATPase) and segregation
(SMC ATPase). Our results also demonstrate that genes essential for key enzymatic functions such as tetrapyrrole (e.g.,
chlorophyll) biosynthesis and DNA replication have been
lost from ptDNA in both Paulinella species. The upcoming
nuclear genome sequence from FK01 (Yoon HS, Andersen
RA, and Bhattacharya D, unpublished data) will provide definitive evidence for EGT versus outright loss as explanation
for the missing genes in this taxon.
Conclusions
Our data provide important insights into early organelle
evolution that until recently appeared intractable due to
the ancient origins of the mitochondrion and the plastid.
The relatively recently established Paulinella plastid appears to mimic some of the predicted ancestral features
of the canonical Plantae plastid, including differential gene
loss in descendant lineages (e.g., Martin et al. 1998), high
conservation of photosynthetic and informational functions, protection from HGT (e.g., Rice and Palmer 2006),
and EGT to the host nucleus (Martin and Herrmann
1998). The photosynthetic Paulinella lineage may therefore
provide a ‘‘retelling of the tale’’ that took place when the
Plantae ancestor, long ago, diversified into some of the
most important photosynthetic eukaryotes on our planet.
Supplementary Material
Supplementary figure S1 and tables S1–S5 are available at
Molecular Biology and Evolution online (http://www.mbe.
oxfordjournals.org/).
1535
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
FIG. 4. (A) Plastid genome alignment showing the presence of the psaI (subunit VIII of photosystem I, PSI-VIII) gene (white block) in Paulinella
M0880/a and its apparent loss in the homologous region from Paulinella FK01 ptDNA after automated annotation (see text and Materials and
Methods). (B) Protein alignment of PSI-VIII sequences from Plantae plastids, cyanobacteria, M0880/a, and conceptual translation of both the
pseudogenes present in the FK01 plastid genome (FK01 ptDNAtra) and the EST from FK01 (FK01 ESTtra). The nucleotide alignment of the psaI
genome sequence (FK01 genomic) and the EST (FK01 EST) from FK01 indicates the presence of a 198-bp spliceosomal intron in this coding
region.
Reyes-Prieto et al. · doi:10.1093/molbev/msq032
Acknowledgments
This project was partially supported by grants from the National Science Foundation Microbial Genome Sequencing
program (EF 08-27023) that was awarded to H.S.Y., R.A.A.,
and D.B., from the Assembling the Tree of Life program (EF
04-31117) that was awarded to D.B., and from the Korean
Science and Engineering Foundation (Ministry of Science
and Technology; R01-2006-000-10207-0, KRF-2008-357C00148) that were awarded to S.M.B. and E.C.Y. A.M.
was supported by an Institutional National Research Service Award (T32 GM98629) from the National Institutes
of Health. T.N. was supported by a Japan Society for the
Promotion of Science Research Fellowship for Young
Scientists (DC1).
Anisimova M, Gascuel O. 2006. Approximate likelihood-ratio test
for branches: a fast, accurate, and powerful alternative. Syst Biol.
55:539–552.
Aziz RK, Bartels D, Best AA, et al. (26 co-authors). 2008. The RAST
Server: rapid annotations using subsystems technology. BMC
Genomics. 9:75.
Badger MR, Hanson D, Price GD. 2002. Evolution and diversity of
CO2 concentrating mechanisms in cyanobacteria. Funct Plant
Biol. 29:161–173.
Besemer J, Lomsadze A, Borodovsky M. 2001. GeneMarkS: a selftraining method for prediction of gene starts in microbial
genomes. Implications for finding sequence motifs in regulatory
regions. Nucleic Acids Res. 29:2607–2618.
Bhattacharya D, Medlin L. 1995. The phylogeny of plastids: a review
based on comparisons of small-subunit ribosomal RNA coding
regions. J Phycol. 31:489–498.
Bhattacharya D, Yoon HS, Hackett JD. 2004. Photosynthetic
eukaryotes unite: endosymbiosis connects the dots. Bioessays
26:50–60.
Cavalier-Smith T. 1992. The number of symbiotic origins of
organelles. Biosystems. 28:91–106; discussion 107–108.
Darling AC, Mau B, Blattner FR, Perna NT. 2004. Mauve: multiple
alignment of conserved genomic sequence with rearrangements.
Genome Res. 14:1394–1403.
Falkowski PG, Katz ME, Knoll AH, Quigg A, Raven JA, Schofield O,
Taylor FJ. 2004. The evolution of modern eukaryotic phytoplankton. Science 305:354–360.
Guindon S, Gascuel O. 2003. A simple, fast, and accurate algorithm
to estimate large phylogenies by maximum likelihood. Syst Biol.
52:696–704.
Hosokawa T, Kikuchi Y, Nikoh N, Shimada M, Fukatsu T. 2006. Strict
host-symbiont cospeciation and reductive genome evolution in
insect gut bacteria. PLoS Biol. 4:e33.
Kameya M, Ikeda T, Nakamura M, Arai H, Ishii M, Igarashi Y. 2007. A
novel ferredoxin-dependent glutamate synthase from the
hydrogen-oxidizing chemoautotrophic bacterium Hydrogenobacter thermophilus TK-6. J Bacteriol. 189:2805–2812.
Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a novel
method for rapid multiple sequence alignment based on fast
Fourier transform. Nucleic Acids Res. 30:3059–3066.
Kies L. 1974. Elektronenmikroskopische Untersuchungen an
Paulinella chromatophora Lauterborn, einer Thekamöbe mit
blau-grünen Endosymbionten (Cyanellen). Protoplasma 80:69–89.
Marais GA, Calteau A, Tenaillon O. 2008. Mutation rate and genome
reduction in endosymbiotic and free-living bacteria. Genetica
134:205–210.
1536
Marin B, Nowack EC, Glockner G, Melkonian M. 2007. The ancestor
of the Paulinella chromatophore obtained a carboxysomal
operon by horizontal gene transfer from a Nitrococcus-like
gamma-proteobacterium. BMC Evol Biol. 7:85.
Marin B, Nowack EC, Melkonian M. 2005. A plastid in the making:
evidence for a second primary endosymbiosis. Protist 156:
425–432.
Marques S, Florencio FJ, Candau P. 1992. Purification and
characterization of the ferredoxin-glutamate synthase from the
unicellular cyanobacterium Synechococcus sp. PCC 6301. Eur J
Biochem. 206:69–77.
Martin W, Herrmann RG. 1998. Gene transfer from organelles to the
nucleus: how much, what happens, and why? Plant Physiol.
118:9–17.
Martin W, Stoebe B, Goremykin V, Hapsmann S, Hasegawa M,
Kowallik KV. 1998. Gene transfer to the nucleus and the
evolution of chloroplasts. Nature 393:162–165.
Moran NA. 1996. Accelerated evolution and Muller’s rachet in
endosymbiotic bacteria. Proc Natl Acad Sci U S A. 93:2873–2878.
Moran NA. 2002. Microbial minimalism: genome reduction in
bacterial pathogens. Cell 108:583–586.
Moran NA, McLaughlin HJ, Sorek R. 2009. The dynamics and time
scale of ongoing genomic erosion in symbiotic bacteria. Science
323:379–382.
Moustafa A, Beszteri B, Maier UG, Bowler C, Valentin K,
Bhattacharya D. 2009. Genomic footprints of a cryptic plastid
endosymbiosis in diatoms. Science 324:1724–1726.
Moustafa A, Bhattacharya D. 2008. PhyloSort: a user-friendly
phylogenetic sorting tool and its application to estimating the
cyanobacterial contribution to the nuclear genome of Chlamydomonas. BMC Evol Biol. 8:6.
Muro-Pastor MI, Reyes JC, Florencio FJ. 2005. Ammonium
assimilation in cyanobacteria. Photosynth Res. 83:135–150.
Nakabachi A, Yamashita A, Toh H, Ishikawa H, Dunbar HE,
Moran NA, Hattori M. 2006. The 160-kilobase genome of the
bacterial endosymbiont Carsonella. Science 314:267.
Nakayama T, Ishida K. 2009. Another acquisition of a primary
photosynthetic organelle is underway in Paulinella chromatophora. Curr Biol. 19:R284–R285.
Nowack EC, Melkonian M, Glockner G. 2008. Chromatophore
genome sequence of Paulinella sheds light on acquisition of
photosynthesis by eukaryotes. Curr Biol. 18:410–418.
Okuhara H, Matsumura T, Fujita Y, Hase T. 1999. Cloning and
inactivation of genes encoding ferredoxin- and NADH-dependent
glutamate synthases in the cyanobacterium Plectonema boryanum. Imbalances in nitrogen and carbon assimilations caused by
deficiency of the ferredoxin-dependent enzyme. Plant Physiol.
120:33–42.
Palmer JD. 2003. The symbiotic birth and spread of plastids: how
many times and whodunit? J Phycol. 39:4–12.
Pérez-Brocal V, Gil R, Ramos S, Lamelas A, Postigo M, Michelena JM,
Silva FJ, Moya A, Latorre A. 2006. A small microbial genome: the end
of a long symbiotic relationship? Science 314:312–313.
Raymond JR, Trautman DA, Larkum AWD. 2001. Phosphate limited
cultures of the cyanobacterium Synechococcus are capable of
very rapid, opportunistic uptake of phosphate. New Phytol.
152:189–201.
Reyes-Prieto A, Weber AP, Bhattacharya D. 2007. The origin and
establishment of the plastid in algae and plants. Annu Rev Genet.
41:147–168.
Rice DW, Palmer JD. 2006. An exceptional horizontal gene transfer
in plastids: gene replacement by a distant bacterial paralog and
evidence that haptophyte and cryptophyte plastids are sisters.
BMC Biol. 4:31.
Sagan L. 1967. On the origin of mitosing cells. J Theor Biol. 14:
255–274.
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
References
MBE
Paulinella Plastid Evolution · doi:10.1093/molbev/msq032
Stepanauskas R, Sieracki ME. 2007. Matching phylogeny and
metabolism in the uncultured marine bacteria, one cell at
a time. Proc Natl Acad Sci USA. 104:9052–9057.
Xu Q, Hoppe D, Chitnis VP, Odom WR, Guikema JA, Chitnis PR.
1995. Mutational analysis of photosystem I polypeptides in the
cyanobacterium Synechocystis sp. PCC 6803. Targeted inactivation of psaI reveals the function of psaI in the structural
organization of psaL. J Biol Chem. 270:16243–16250.
MBE
Yang Z. 2007. PAML 4: phylogenetic analysis by maximum
likelihood. Mol Biol Evol. 24:1586–1591.
Yoon HS, Nakayama T, Reyes-Prieto A, Andersen RA, Boo SM, Ishida KI,
Bhattacharya D. 2009. A single origin of the photosynthetic
organelle in different Paulinella lineages. BMC Evol Biol. 9:98.
Yoon HS, Reyes-Prieto A, Melkonian M, Bhattacharya D. 2006.
Minimal plastid genome evolution in the Paulinella endosymbiont. Curr Biol. 16:R670–R672.
Downloaded from http://mbe.oxfordjournals.org/ at Chungnam National University on June 4, 2013
1537