Copyright Ó 2008 by the Genetics Society of America

DOI: 10.1534/genetics.107.086090

Green Transcription Factors: A Chlamydomonas Overview

Diego Mauricio Riaño-Pachón,*,†,1 Luiz Gustavo Guedes Corrêa,*,‡,1

Raúl Trejos-Espinosa*,‡ and Bernd Mueller-Roeber*,‡,2
*Department of Molecular Biology, University of Potsdam, 14476 Potsdam-Golm, Germany, †Bioinformatics
Research Group, Max-Planck Institute of Molecular Plant Physiology, GabiPD Team, 14476
Potsdam-Golm, Germany and ‡Cooperative Research Group, Max-Planck Institute of
Molecular Plant Physiology, 14476 Potsdam-Golm, Germany
Manuscript received December 15, 2007
Accepted for publication January 29, 2008

ABSTRACT
Transcription factors (TFs) control gene expression by interacting with cis-elements in target gene
promoters. Transcription regulators (TRs) assist in controlling gene expression through interaction with
TFs, chromatin remodeling, or other mechanisms. Both types of proteins thus constitute master controllers
of dynamic transcriptional networks. To uncover such control elements in the photosynthetic green alga
Chlamydomonas reinhardtii, we performed a comprehensive analysis of its genome sequence. In total, we
identified 234 genes encoding 147 TFs and 87 TRs of 40 families. The set of putative TFs and TRs, including
their transcript and protein sequences, domain architectures, and supporting information about putative
orthologs, is available at http://plntfdb.bio.uni-potsdam.de/v2.0/. Twelve of 34 plant-specific TF families
were found in at least one algal species, indicating their early evolutionary origin. Twenty-two plant-specific
TF families and one plant-specific TR family were not observed in algae, suggesting their specific association
with developmental or physiological processes characteristic to multicellular plants. We also analyzed the
occurrence of proteins that constitute the light-regulated transcriptional network in angiosperms and found
putative algal orthologs for most of them. Our analysis provides a solid ground for future experimental
studies aiming at deciphering the transcriptional regulatory networks in green algae.

T HE regulation of growth and development and the

coordination of these processes in response to
hormonal or environmental stimuli, including adverse
structure similarities in the DNA-binding and multimeriza-
tion domains. TF genes represent a considerable fraction
of the genomes of all eukaryotic organisms, including
conditions, requires a dynamic control of the expression angiosperms (Riechmann et al. 2000; Goff et al. 2002). In
of hundreds to thousands of genes in each organism Oryza sativa (rice), for example, 2.6% of the identified
(Lemon and Tjian 2000; Chen et al. 2002; Li et al. 2007). genes encode TFs (Goff et al. 2002). Currently, the genome
Transcription factors (TFs) are master control pro- sequences of four angiosperms (Arabidopsis thaliana, O.
teins that regulate gene expression levels by binding to sativa, Populus trichocarpa, and Vitis vinifera) are in the public
specific DNA sequences, so-called cis-acting elements, in domain (Arabidopsis Genome Initiative 2000; Goff
the promoters of target genes, thereby enhancing or et al. 2002; Yu et al. 2002; Tuskan et al. 2006; Jaillon et al.
repressing their transcriptional rates. The genomewide 2007). Additionally, the genomes of various algae, in-
identification of TF genes through computational meth- cluding the red alga Cyanidioschyzon merolae (Nozaki et al.
ods, and genomewide comparative studies, are impor- 2007), the green algae Ostreococcus tauri (Derelle et al.
tant tasks that not only provide an insight into existing TF 2006), Chlamydomonas reinhardtii (Merchant et al. 2007),
families within individual species or organism lineages and the moss Physcomitrella patens (Rensing et al. 2008) have
but also help to understand how evolution shaped become available.
developmental and physiological diversification. TFs, as To facilitate the analysis of plant TFs and TRs, we
well as other transcriptional regulators (TRs) that generally have recently established the Plant Transcription Factor
do not directly bind DNA but assist in gene expression Database (PlnTFDB) (Riano-Pachon et al. 2007) and
regulation through interaction with cis-element-binding updated it by including additional plant species (avail-
proteins, can be grouped into different protein families able at http://plntfdb.bio.uni-potsdam.de/v2.0). Here
according to their primary and/or three-dimensional we report about the occurrence of putative transcrip-
tional regulators in Chlamydomonas. We identified 147
1
putative TFs that belong to 29 different protein families
These authors contributed equally to this work.
2
and 87 putative TRs that are members of 10 families. Of
Corresponding author: Department of Molecular Biology, University of
Potsdam, Karl-Liebknecht-Strasse 24-25, Haus 20, 14476 Potsdam-Golm, 34 plant-specific families, 3 (G2-like, PLATZ, RWP-RK)
Germany. E-mail: bmr@uni-potsdam.de predate the split between green and red algae. Nine

Genetics 179: 31–39 (May 2008)

32 D. M. Riaño-Pachón et al.

additional families, i.e., ABI3/VP1, AP2-EREBP, ARR-B, PlnTFDB (http://plntfdb.bio.uni-potsdam.de/v2.0). Pairs of

C2C2-CO-like, C2C2-Dof, PBF-2-like/Whirly, Pseudo ARR- orthologs (and direct paralogs) allowed us to identify possible
clusters of orthologs (and paralogs) comprising genes from
B, SBP, and WRKY, predate the split between chlo-
more than two species. This was achieved using a graph-
rophyta (green algae) and streptophyta (land plants theoretic approach as follows: (i) only InParanoid clusters
and charophycean algae). In total, 12 families were containing at least one protein annotated as a TF in PlnTFDB
identified from algal groups onward. Interestingly, 22 were kept; (ii) all proteins identified in this way were
plant-specific TF families and one TR family are not represented as nodes in a network of orthologous relation-
ships; edges were drawn between nodes when the InParanoid
present in algae, indicating their particular importance
confidence score for the orthologous relationship was $0.9;
for plant multicellularity and tissue organization. (iii) connected components were extracted from the network;
by definition, a connected component is a subgraph in which
every node can be reached from every other node. The
connected components (subgraphs) represent putative clus-
ters of orthologs. Network visualization and analysis were
MATERIALS AND METHODS
carried out using the software package Pajek (De Nooy et al.
Identification of transcription factors: Putative complete 2005). The identification of orthologs through BLAST
sets of transcription factors of the following species were searches can lead to false positives; consequently, we made use
retrieved from the Plant Transcription Factor Database v2.0, of a phylogenetic approach to largely compensate for this fact. In
PlnTFDB (http://plntfdb.bio.uni-potsdam.de/v2.0; Riano- addition to that, and affecting both approaches for ortholog
Pachon et al. 2007): the red alga C. merolae (Nozaki et al. detection (phylogenetics and InParanoid), false negatives can
2007), the green algae O. tauri (Derelle et al. 2006) and C. arise due to incomplete genome sequence information (gaps in
reinhardtii (Merchant et al. 2007), the moss P. patens (Rensing the sequence) or misannotated genes.
et al. 2008), and the angiosperms A. thaliana (Arabidopsis As mentioned above, in addition to the BLAST approach,
Genome Initiative 2000), P. trichocarpa (black cottonwood) we performed phylogenetic analyses of each family, which
(Tuskan et al. 2006), and O. sativa (rice) (Goff et al. 2002). allowed the identification of possible groups of orthologs
PlnTFDB has two divisions: one providing information about (PoGOs). A PoGO is defined by the following criteria: (i)
transcription factors, defined as proteins that directly bind to members of a PoGO have a monophyletic origin, indicated by
DNA and affect the level of transcription (called TFs here), a bootstrap support of .50%; (ii) a PoGO conserved in all
and the other providing information about transcriptional green plants possesses at least one representative gene of each
regulators that, for example, exert regulatory control through of the main lineages analyzed here, including algae, bryo-
interaction with TFs or through chromatin remodeling (called phytes, and angiosperms, assuming that the putative complete
TRs). Additionally, we identified TF and TR families common to sets of TF genes of these organisms were identified and no
all eukaryotes using the following model organisms: the pro- selective gene loss had occurred; (iii) the inferred phylogeny
tozoan Giardia lamblia (Best et al. 2004), the yeast Saccharomyces is consistent with the known phylogeny of plant species
cerevisiae (Goffeau et al. 1996), the nematode Caenorhabditis (Vincentz et al. 2003).
elegans (C. elegans Sequencing Consortium 1998), the insect We evaluated the overlap between the clusters of orthologs
Drosophila melanogaster (Adams et al. 2000), and Homo sapiens identified by InParanoid and by phylogenetic analysis using
(International Human Genome Sequencing Consortium the Adjusted Rand Index (Rand 1971; Hubert and Arabie
2004). For the identification of nonplant TFs, we used the 1985), implemented in the statistical package R (R Development
procedure described by Riano-Pachon et al. (2007), using Core Team 2007).
PFAM (Finn et al. 2006) release 20.0 for domain identification.
Sequences were downloaded from Integr8 (http://www.ebi.
ac.uk/integr8/; Kersey et al. 2005), except for G. lamblia
sequences, which were downloaded from GiardiaDB (http:// RESULTS AND DISCUSSION
www.giardiadb.org). Transcription factors in eukaryotes: We identified
Phylogenetic analysis: Protein sequences corresponding to
the defining conserved domain of each TF and TR family the putatively complete nonredundant sets of TFs
were extracted from whole-protein sequences of the photosyn- and TRs in the algae Chlamydomonas and Ostreococ-
thetic eukaryotes using the domain coordinates identified by cus, the moss Physcomitrella, and the angiosperms
the PFAM search described above. Alignment of protein se- Arabidopsis, black cottonwood, and rice (Table 1). The
quences was performed employing ClustalX (Thompson et al. genes were grouped into 66 gene families according to
1997), using default parameters. Phylogenetic analyses based
on amino acid sequences were conducted using MEGA v3.1 their characteristic conserved domains, as described by
(Kumar et al. 2004). Unrooted phylogenetic tree topologies Riano-Pachon et al. (2007). We identified the putatively
were reconstructed by neighbor-joining (NJ), the distances complete sets of genes for the same families in G. lamblia
were obtained using p-distances (Nei and Kumar 2000), and (protozoa), S. cerevisiae (yeast), C. elegans (nematodes), D.
the resampling of the original protein set was a 1000-bootstrap melanogaster (fruit flies), and H. sapiens (humans). Twenty
repetition. These NJ analyses provide an overview of the
general patterns of TF and TR evolution. All sequences and TF and 11 TR families were also present in nonphotosyn-
alignments used in this study are available upon request. thetic eukaryotes. In contrast to the previous report by
Identification of orthologs among green plants: We identi- Riechmann et al. (2000), we observed that the Trihelix
fied orthologs through pairwise comparisons of protein se- family is not restricted to the plant kingdom (Table 1).
quences in whole-protein sets of the green plants Chlamydomonas, G2-like and WRKY TFs are generally regarded as plant
Ostreococcus, Physcomitrella, rice, Arabidopsis, and black
cottonwood, using a variation of the best BLAST bidirectional specific; our analysis largely confirms this view. However,
hit approach implemented in the program InParanoid (Remm we also identified genes encoding putative members
et al. 2001). Orthologs identified in this way are presented in of these families in the nonplant species G. lamblia, a
 Transcription Factors in Chlamydomonas 33

TABLE 1
Transcription factors and regulators present in different eukaryotic species

Photosynthetic species Nonphotosynthetic species

Family CME OTA CRE PPA OSAJ ATH PTR HSA DME CEL SCE GLA
ABI3VP1 1 30 59 59 81
Alfin-like 7 12 8 9
AP2-EREBP 8 11 150 174 160 206
ARF 13 42 32 36
ARR-B 1 1 5 9 15 15
BBR/BPC 3 10 15
BES1 6 6 10 12
bHLH 1 1 4 100 175 160 159 154 65 50 7
bZIP 3 7 7 37 113 93 84 59 32 39 12 1
C2C2-CO-like 3 1 11 21 19 14
C2C2-Dof 2 1 20 33 42 41
C2C2-GATA 6 4 6 12 37 30 36 15 7 14 10
C2C2-YABBY 13 7 13
C2H2 7 4 5 56 103 104 113 644 312 150 39 5
CAMTA 1 7 6 7 2 1 2
CCAAT 6 8 8 27 58 53 59 25 11 12 10 3
CPP 2 2 1 6 16 9 12 4 2 2
CSD 4 1 3 3 4 7 16 4 5
DBP 7 5 10
E2F-DP 5 3 6 10 12 11 10 18 3 7 1
EIL 2 7 6 6
FHA 2 7 12 15 19 17 18 44 22 12 14 2
G2-likea 1 2 4 41 52 48 66 1
GeBP 6 20 7
GRAS 39 56 35 97
GRF 2 17 9 9
HB 5 6 1 42 124 97 129 299 114 107 7
HRT 7 1 2 1
HSF 3 1 2 8 36 23 31 6 1 1 5
LFY 2 1 1 1
LIM 1 1 3 7 6 13 ND ND ND ND ND
MADS 1 1 2 22 82 122 108 9 3 2 4
MYB 11 10 11 61 129 161 210 19 5 7 3 2
MYB-related 21 17 14 44 99 90 100 36 16 12 12 2
NAC 32 140 115 163
NOZZLE 1
PBF-2 like/Whirly 1 1 3 4 3
PLATZ 1 1 3 13 18 13 20
Pseudo ARR-B 1 2 2 7 5 7
RWP-RKb 1 4 14 8 13 14 18
S1Fa-like 1 2 3 2
SAP 1 1
SBP 21 13 21 17 29
Sigma70-likec 4 1 1 5 9 6 9
SRS 2 5 11 10
TAZ 2 5 9 9 7 4 1 6
TCP 6 22 26 33
Trihelix 25 24 27 43 8 1
TUB 1 3 6 17 12 11 6 3 2
ULT 2 2 2
VOZ 2 2 2 4
WRKYa 2 1 37 114 84 101 1
zf-HD 7 15 17 22
ZIM 12 22 22 16
ARIDd 4 1 2 7 6 10 13 21 6 5 2
AUX/IAAd 2 43 34 32
(continued )
34 D. M. Riaño-Pachón et al.

TABLE 1
(Continued)

Photosynthetic species Nonphotosynthetic species

Family CME OTA CRE PPA OSAJ ATH PTR HSA DME CEL SCE GLA
C3Hd 7 18 15 44 97 75 96 85 33 40 7 5
DDTd 1 1 2 7 5 5 5 3 2 2
HMGd 5 7 8 17 19 12 90 26 22 7 3
Jumonjid 3 5 7 10 17 19 20 38 11 15 3
LUGd 1 12 3 5 6 1 1 1
MBF1d 1 3 4 3 3 1 1 1 1
PHDd 7 11 12 50 55 53 70 118 44 23 14 2
RBd 1 1 1 2 4 1 1 4 2 1
SETd 6 10 22 26 32 38 44 49 17 29 7 3
SNF2d 13 20 19 35 44 43 48 48 22 23 17 6
Plant-specific TF and TR families are in italics; all other families are in roman. We also highlight TF families in italics that, in
addition to plants, have members in early branching eukaryotes. Numbers represent distinct protein sequences. CME, C. merolae;
OTA, O. tauri; CRE, C. reinhardtii; PPA, P. patens; OSAJ, O. sativa ssp. japonica; ATH, A. thaliana; PTR, P. trichocarpa; HAS, H. sapiens;
DME, D. melanogaster; CEL, C. elegans; SCE, S. cerevisiae; GLA, G. lamblia. ND, not determined.
a
Present in G. lamblia.
b
Present in D. discoideum and E. histolytica (according to PFAM website).
c
Present in bacteria.
d
Transcription regulators (TRs).

protozoan that arose early in eukaryote evolution. In Similarly, MADS-box TFs are present in only small num-
general, the number of TFs and TRs increases with the bers in Chlamydomonas (two genes) and in all other
number of genes in the genome, following a power law as unicellular organisms. In contrast, in these organisms,
observed before (Van Nimwegen 2003). TFs and TRs members of the C2H2 family are slightly more abundant
were found to be similarly abundant in algae and yeast; than members of the MADS-box family, with five, four,
however, in these lineages they are considerably less fre- and seven genes, respectively, in the algae Chlamydomonas,
quent than in animals. Numbers of TFs and TRs in many Ostreococcus, and Cyanidioschyzon, and 39 members in
cases were similar in mosses and animals, whereas gene Saccharomyces (Table 1). C2H2 TFs contain a zinc-finger
numbers were often greater in angiosperms. In Chlamy- domain. The recruitment of this domain for transcrip-
domonas, we identified 147 putative TF and 87 putative tional regulation occurred in prokaryotes, and mem-
TR coding sequences from 29 and 10 protein families, bers of the Ros family may have been the origin of C2H2
respectively, totaling 234 distinct proteins involved in the in eukaryotes (Bouhouche et al. 2000). In general, TFs
regulation of transcription (Table 1; protein sequences bearing a zinc-finger domain have significantly contrib-
are available at http://plntfdb.bio.uni-potsdam.de/v2.0/ uted to the evolution of eukaryotic organisms (Riechmann
index.php?sp_id¼CRE). A schematic of the transcrip- et al. 2000) either through gene duplication leading to
tional regulatory proteins identified in Chlamydomonas, an increased gene number or through the modulation
including their defining domains, is given in supplemen- of other domains present in these proteins, resulting in
tal Figure 1. To date, however, the biological functions of the formation of new families of TFs.
only a small number of these proteins have been analyzed The acquisition of chloroplasts represents an im-
(supplemental Table 1). portant step in the evolutionary path that separated
Chlamydomonas transcription factors: In animals, plants from animals and fungi. Evidently, new regulatory
TFs of the C2H2 and HB families play important roles in networks had to be established through evolution to
growth-related and development processes (Wu 2002) achieve an optimal integration of photosynthetic func-
and body-plan formation (Deutsch and Mouchel-Vielh tions with other cellular processes. TFs and TRs constitute
2003). These two families are the largest in animals, with important elements of such networks. Three families of
.100 members each in humans, Drosophila, and Caeno- putative TFs predate the split between rhodophyta (red
rhabditis. In animals, HB TFs function as homeotic genes algae) and chlorophyta, i.e., G2-like, PLATZ, and RWP-
that control the formation and differentiation of different RK. These families appear to be of particular importance
body parts (Garcia-Fernandez 2005; Negre and Ruiz for the evolution of eukaryotic photosynthetic organ-
2007). In contrast, in plants homeotic functions are isms, as they are the only plant-specific TFs (with perhaps
carried out by TFs of the MADS-box family (Irish 2003). the exception of G2-like, which might also be present
Typically, angiosperms have 80–120 MADS-box proteins, in Giardia; see above) that are present in both red and
whereas such TFs are largely absent from animals (,10). green algae. Both algal groups derived from the original
 Transcription Factors in Chlamydomonas 35

primary endosymbiotic event that led to the establish- i.e., CC-503 cw92 mt1; Goodenough et al. 2007) is re-
ment of plastids (Reyes-Prieto et al. 2007). Nine addi- quired for expression of minus-specific gamete-specific
tional families, i.e., ABI3/VP1, AP2-EREBP, ARR-B, C2C2- genes in response to nitrogen deprivation (Ferris and
CO-like, C2C2-Dof, PBF-2-like/Whirly, Pseudo ARR-B, Goodenough 1997; Lin and Goodenough 2007). An-
SBP, and WRKY (Table 1), predate the split between other TF in this family, NIT2, is a positively acting regu-
green algae and streptophytes. Plant-specific TF families latory gene of the nitrate assimilation pathway (Camargo
might have important roles in the control of light- et al. 2007) (GenBank accession no. DQ311647; this gene
dependent processes and related biochemical pathways is mutated in the sequenced Chlamydomonas strain;
such as those involved in sugar production or starch Fernandez and Matagne 1984); the most similar entry
accumulation. in PlnTFDB is protein ID 195807).
TFs of the G2-like family (a distinct group within the Information regarding SBP TFs, Jumonji, and SET
GARP superfamily of TFs; Rossini et al. 2001) are present TRs, as well as microRNAs that often control TF genes in
in all plants, including red and green algae, and in angiosperms but appear to be of minor importance in
G. lamblia, suggesting a deep evolutionary origin, but they Chlamydomonas, is given in the supplemental text.
are not found in animals or fungi. G2-like TFs regulate Transcription factors involved in hormone signaling:
chloroplast development in diverse plant species (e.g., Phytohormones coordinate a vast spectrum of develop-
Physcomitrella, Arabidopsis, and Zea mays) through a mental and physiological processes in angiosperms. In
process that requires a close coordination between plas- contrast, knowledge about the occurrence of hormones
tidial and nuclear genomes. More specifically, GOLDEN in algae and their possible functions in cellular signaling
2-like (GLK) TFs are required for correct stacking of is extremely limited. Some evidence indicates that auxins
thylakoids within chloroplasts, although it is not known in and cytokinins are present in algae (Tarakhovskaya
detail how they exert their function in this process. One et al. 2007), indicating their functional importance early
possible model is that GLKs regulate the transcription of in plant evolution. TF families known to participate in
genes encoding thylakoid-stabilizing factor(s) (Yasumura hormone signaling in angiosperms are also found in
et al. 2005). We did not detect the ortholog of GLK in the Chlamydomonas (Table 1). Recent work on ABSCISIC
sequenced Chlamydomonas genome, which is consistent ACID INSENSITIVE 3 (ABI3) from Arabidopsis has
with the fact that chloroplast thylakoid stacking is less indicated a possible role in cross talk of abscisic acid
advanced in this alga as compared to bryophytes and angio- and auxin response pathways (Brady et al. 2003; Rock
sperms, as previously discussed (Yasumura et al. 2005). and Sun 2005). A similar observation was made for VP1
PHOSPHORUS STARVATION RESPONSE1 (PSR1) from from maize, the ortholog of ABI3 (Suzuki et al. 2001). We
Chlamydomonas and its ortholog PHOSPHATE STARVA- observed a single ABI3/VP1 gene in Chlamydomonas,
TION RESPONSE1 (PHR1) from Arabidopsis encode TFs whereas Physcomitrella has 30 ABI3/VP1 genes, and
that control cellular responses to phosphate deprivation angiosperms have 60–80 (Table 1). To our knowledge
(Wykoff et al. 1999; Rubio et al. 2001). Both proteins were the role of the ABI3/VP1 gene in Chlamydomonas has
originally thought to be members of the MYB TF family, not been characterized yet. TFs of the ARR-B and AP2-
but subsequently were placed within the GARP superfam- EREBP families are involved in cytokinin response path-
ily (G2-like) (Fitter et al. 2002). PSR1 targets include ways in angiosperms (Rashotte et al. 2006; Ishida et al.
genes encoding chloroplast-localized proteins involved in 2008). We detected one ARR-B gene and 11 AP2-EREBP
photosynthesis, regulation of gene expression, and other genes in Chlamydomonas (Table 1). The role of these
processes (Moseley et al. 2006). Recently, PSR1 has also TFs has not been analyzed.
been shown to control the accumulation of chloroplast TF families absent from algae: Interestingly, 22
RNA under phosphorous limitation through control of plant-specific TF families and 1 TR family are not pres-
the expression of ribonuclease polynucleotide phosphor- ent in algae (Table 1). These families may be related to
ylase (Yehudai-Resheff et al. 2007). Whether the angio- the acquisition of multicellularity and tissue organiza-
sperm ortholog exerts a similar function is currently tion, invasion of the terrestrial environment, and long-
unknown. distance trafficking. NAC TFs could be identified only
RWP-RK (Figure 1) is a TF family present in all green from bryophytes onward. Functional studies have shown
plants, as well as in red algae. It is also present in the that several NAC genes play an important role in cell
early diverging amoebozoa Dictyostelium discoideum and differentiation (Olsen et al. 2005). As we did not find
Entamoeba histolytica, but not in animals or fungi, sug- any NAC gene in the Volvox carteri genome (not shown),
gesting a deep evolutionary origin. In vascular plants, we assume that TFs of this family were not important for
this family is involved in the regulation of genes in establishing multicellularity in this organism.
response to nitrogen status and nodule development Orthologs across green plants: The green plant
in legumes (Schauser et al. 1995; Borisov et al. 2003). lineage is a monophyletic group, its members having
In Chlamydomonas, the gene minus dominance (MID; split from the red algal lineage 1142 6 167 million
GenBank accession no. U92071; specific to mt strains years ago (Zimmer et al. 2007). Tracing gene orthology
and consequently not present in the sequenced strain, relations across lineages provides a way to assess, to some
36 D. M. Riaño-Pachón et al.

extent, the forces driving the functional diversification

of multigene families. As reported previously, TF genes
in plants have a higher retention rate after duplication
than other genes (Seoighe and Gehring 2004; De Bodt
et al. 2005). Additionally, genes functionally related to
stress responses tend to undergo a more intense dupli-
cation process (Shiu et al. 2004). Therefore, TF gene
families are well suited to trace back important events in
evolution.
In our NJ analyses (for examples, see supplemental
Figures 2–7), we have identified 120 clusters of ortho-
logs with 1183 genes in total. Seventy-one of them are
conserved in all green plants, and 26 are also common
to red algae (see supplemental Table 2). Clusters to
which functions could be assigned are involved mainly
in light perception/response, control of plastidial gene
expression, regulation of circadian rhythm, and the tran-
sition from the vegetative to the reproductive phase of
growing plants (data not shown). Moreover, 38 of these
clusters were found to have a one-to-one relationship
(they do not possess any paralog inside the same group
of orthologs). Such genes tend to exert key biological
functions (Shiu et al. 2004). The greatest number of clus-
ters, i.e., 20, among all green plants is represented by the
SWI2/SNF2 gene family that encodes proteins involved
in chromatin remodeling and thus the regulation of
transcription, replication, and DNA recombination and
repair. In plants, some Swi2/Snf2 proteins have been
studied (Shaked et al. 2006), but a detailed functional
analysis is missing for most of them. In the RWP-RK
family, we found only one PoGO (Figure 1) in represen-
tatives from all green plants. The position of the C. merolae
sequence is not evident from this analysis. In addition,
groups of paralogs of Chlamydomonas are shown.
We also made a comparison between the clusters of
orthologs obtained by phylogenetic analysis and by the
InParanoid-Graph theoretic approach (see supplemen-
tal Table 3). In total, 446 genes from both classifications
overlap, representing 99 of the 120 clusters obtained by
the NJ analysis, and 98 of 168 clusters identified by the
InParanoid approach (see supplemental Table 4). Thus,
a large number of clusters was identified irrespective of
the detection method used. We computed the Adjusted

Figure 1.—Phylogenetic tree of RWP-RK TFs in plants. We

identified one PoGO (PoGO 1) conserved in all green plants,
which includes the NIT2 TF (CRE 195807), a regulatory factor
of genes involved in the nitrate assimilation pathway. Addition-
ally, there are two possible groups of paralogs (PoGP 1 and
PoGP 2) of Chlamydomonas genes. Red, C. merolae (CME);
violet, O. tauri (OTA); light green, C. reinhardtii (CRE); light blue,
P. patens (PPA); green, A. thaliana (ATH); brown, P. trichocarpa
(PTR); gray, O. sativa ssp. japonica (OSAJ). The first three let-
ters of the sequence name indicate the species (the first four
letters in the case of OSAJ), and the remaining letters or
numbers represent the accession code through which the re-
spective sequence can be retrieved from the PlnTFDB (http://
plntfdb.bio.uni-potsdam.de/v2.0).
 Transcription Factors in Chlamydomonas 37

Rand Index (ar) on the subset of common genes. The that 22 plant-specific TF and 1 plant-specific TR family
obtained value (ar ¼ 0.912) indicates that the compo- were not present in algae, highlighting their impor-
sition of the common clusters (gene membership) tance for the evolution of multicellular plants. Many of
obtained by both methods is similar. the elements of light-regulated transcriptional networks
Evolution of photosynthetic networks: A recent re- known from bryophytes and angiosperms are also pre-
view by Jiao et al. (2007) provides a good backbone for sent in Chlamydomonas, indicating an early evolutionary
comparison of the light-regulated transcriptional net- origin. Exceptions are elements of the phytochrome-
works of angiosperms and Chlamydomonas. The per- mediated signaling pathways that are missing in algae.
ception of light signals in dicots occurs through three Our analysis provides a basis for further experimental
cryptochromes and two phototropins, for which we found studies on Chlamydomonas transcriptional regulators.
orthologs in Chlamydomonas (see supplemental Table We thank the three anonymous reviewers for comments that helped
5). In contrast, phytochromes involved in the absorption to improve our manuscript. We are grateful to the Department of
of red and far-red light do not have homologs in green Energy Joint Genome Institute and the Chlamydomonas research
algae, consistent with previous findings (Mittag et al. community for sequencing and annotating the Chlamydomonas
genome. L.G.G.C. and B.M.-R. thank the Interdisciplinary Research
2005). One putative ortholog of the angiosperm bZIP
Centre, Advanced Protein Technologies, of the University of Potsdam
protein COMMON PLANT REGULATORY FACTORS 1 and the International Ph.D. Programme, Integrative Plant Science
(CPRF1), CMJ034C, was found in the red alga Cyanidio- ½supported by the Deutscher Akademischer Austauschdienst and the
schyzon, although with a low InParanoid confidence Deutsche Forschungsgemeinschaft (DAAD), no. DAAD D/04/01336
score. The same Cyanidioschyzon protein is also orthol- for financial support. L.G.G.C. thanks the DAAD for providing a
scholarship (no. A/04/34814). B.M.-R. thanks the Fonds der Chem-
ogous to G-BOX BINDING FACTOR 1 (GBF1), suggest- ischen Industrie for financial support (no. 0164389). D.M.R.-P. acknowl-
ing subfunctionalization of the original multifunctional edges financial support by the Bundesministerium fuer Bildung und
algal gene during angiosperm evolution; however, more Forschung (BMBF) (GABI-future grant 0315046). B.M.-R. and R.T.-E.
detailed analyses are required to substantiate this hy- thank the BMBF for funding of the systems biology research unit
pothesis. GBF1 is phosphorylated by CASEIN KINASE II GoFORSYS–Potsdam-Golm BMBF Forschungseinrichtung zur System-
biologie ½(Photosynthesis and Growth: A Systems Biology Based Ap-
(CKII), which allows it to bind to target promoters con-
proach (FKZ 0313924).
taining the G-box, a well-defined light-response element.
We found a putative CKII ortholog in Chlamydomonas,
suggesting that light-dependent post-translational protein
modification of the GBF1 ortholog was established early in LITERATURE CITED
plant evolution. Another important regulatory mecha- Adams, M. D., S. E. Celniker, R. A. Holt, C. A. Evans, J. D. Gocayne
nism is the ubiquitin-mediated degradation of the bZIP et al., 2000 The genome sequence of Drosophila melanogaster.
Science 287: 2185–2195.
TF ELONGATED HYPOCOTYL 5 (HY5) that is triggered Arabidopsis Genome Initiative, 2000 Analysis of the genome se-
by CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1) quence of the flowering plant Arabidopsis thaliana. Nature 408:
and its associated protein PHYTOCHROME A SUP- 796–815.
Best, A. A., H. G. Morrison, A. G. McArthur, M. L. Sogin and G. J.
PRESSOR 1 (SPA1). Both HY5 and SPA1 orthologs were Olsen, 2004 Evolution of eukaryotic transcription: insights
found in green algae (see supplemental Table 5), from the genome of Giardia lamblia. Genome Res. 14: 1537–1547.
whereas COP1 has so far been found only in red algae. Borisov, A. Y., L. H. Madsen, V. E. Tsyganov, Y. Umehara, V. A.
Voroshilova et al., 2003 The Sym35 gene required for root
As the bZIP degradation mechanism triggered by COP1 nodule development in pea is an ortholog of Nin from Lotus
is conserved throughout the plant kingdom (Yi and japonicus. Plant Physiol. 131: 1009–1017.
Deng 2005), one might speculate that COP1 is also Bouhouche, N., M. Syvanen and C. I. Kado, 2000 The origin of
prokaryotic C2H2 zinc finger regulators. Trends Microbiol. 8:
present in Chlamydomonas. In summary, most of the 77–81.
components of the light-regulated transcriptional net- Brady, S. M., S. F. Sarkar, D. Bonetta and P. McCourt, 2003 The
works are shared between Chlamydomonas and seed ABSCISIC ACID INSENSITIVE 3 (ABI3) gene is modulated by
farnesylation and is involved in auxin signaling and lateral root
plants, although phytochromes are missing in green development in Arabidopsis. Plant J. 34: 67–75.
algae. PHYTOCHROME INTERACTING FACTOR (PIF) Camargo, A., A. Llamas, R. A. Schnell, J. J. Higuera, D. Gonzalez-
TFs represent a subgroup of the bHLH family in an- Ballester et al., 2007 Nitrate signaling by the regulatory gene
NIT2 in Chlamydomonas. Plant Cell 19: 3491–3503.
giosperms. No PIF ortholog could be found in Chlamy- C. elegans Sequencing Consortium, 1998 Genome sequence of
domonas. In addition, most other factors that are the nematode C. elegans: a platform for investigating biology. Sci-
directly involved in phytochrome activity in land plants, ence 282: 2012–2018.
Chen, W., N. J. Provart, J. Glazebrook, F. Katagiri, H. S. Chang
such as FHY3, FAR1, HFR1, ATHB4, and PAR1, are ab- et al., 2002 Expression profile matrix of Arabidopsis transcrip-
sent from Chlamydomonas (see supplemental Table 5). tion factor genes suggests their putative functions in response
Conclusions: We have identified 147 putative TFs and to environmental stresses. Plant Cell 14: 559–574.
De Bodt, S., S. Maere and Y. Van de Peer, 2005 Genome duplication
87 putative TRs in Chlamydomonas. Three TF families and the origin of angiosperms. Trends Ecol. Evol. 20: 591–597.
predate the rhodophyta–viridiplantae divide, while nine de Nooy, W., A. Mrvar and A. Batagelj, 2005 Exploratory Social Net-
more of the TF families predate the chlorophyta– work Analysis With Pajek (Structural Analysis in the Social Sciences).
Cambridge University Press, New York.
streptophyta divide and diversified further in bryo- Derelle, E., C. Ferraz, S. Rombauts, P. Rouze, A. Z. Worden et al.,
phytes and angiosperms. However, we also observed 2006 Genome analysis of the smallest free-living eukaryote
38 D. M. Riaño-Pachón et al.

Ostreococcus tauri unveils many unique features. Proc. Natl. Acad. Olsen, A. N., H. A. Ernst, L. L. Leggio and K. Skriver, 2005 NAC
Sci. USA 103: 11647–11652. transcription factors: structurally distinct, functionally diverse.
Deutsch, J. S., and E. Mouchel-Vielh, 2003 Hox genes and the Trends Plant Sci. 10: 79–87.
crustacean body plan. BioEssays 25: 878–887. Rand, M. W., 1971 Objective criteria for the evaluation of clustering
Fernandez, E., and R. F. Matagne, 1984 Genetic analysis of nitrate methods. J. Am. Stat. Assoc. 66: 846–850.
reductase-deficient mutants in Chlamydomonas reinhardtii. Curr. Rashotte, A. M., M. G. Mason, C. E. Hutchison, F. J. Ferreira,
Genet. 8: 635–640. G. E. Schaller et al., 2006 A subset of Arabidopsis AP2 tran-
Ferris, P. J., and U. W. Goodenough, 1997 Mating type in Chlamy- scription factors mediates cytokinin responses in concert with
domonas is specified by mid, the minus-dominance gene. Genet- a two-component pathway. Proc. Natl. Acad. Sci. USA 103:
ics 146: 859–869. 11081–11085.
Finn, R. D., J. Mistry, B. Schuster-Bockler, S. Griffiths-Jones, R Development Core Team, 2007 A Language and Environment for Sta-
V. Hollich et al., 2006 Pfam: clans, web tools and services. tistical Computing. R Foundation for Statistical Computing, Vienna.
Nucleic Acids Res. 34: D247–D251. Remm, M., C. E. Storm and E. L. Sonnhammer, 2001 Automatic
Fitter, D. W., D. J. Martin, M. J. Copley, R. W. Scotland and J. A. clustering of orthologs and in-paralogs from pairwise species
Langdale, 2002 GLK gene pairs regulate chloroplast develop- comparisons. J. Mol. Biol. 314: 1041–1052.
ment in diverse plant species. Plant J. 31: 713–727. Rensing, S. A., D. Lang, A. D. Zimmer, A. Terry, A. Salamov et al.,
Garcia-Fernandez, J., 2005 The genesis and evolution of homeo- 2008 The Physcomitrella genome reveals evolutionary insights
box gene clusters. Nat. Rev. Genet. 6: 881–892. into the conquest of land by plants. Science 319: 64–69.
Goff, S. A., D. Ricke, T. H. Lan, G. Presting, R. Wang et al., 2002 A Reyes-Prieto, A., A. P. Weber and D. Bhattacharya, 2007 The or-
draft sequence of the rice genome (Oryza sativa L. ssp. japonica). igin and establishment of the plastid in algae and plants. Annu.
Science 296: 92–100. Rev. Genet. 41: 147–168.
Goffeau, A., B. G. Barrell, H. Bussey, R. W. Davis, B. Dujon et al. Riano-Pachon, D. M., S. Ruzicic, I. Dreyer and B. Mueller-Roeber,
1996 Life with 6000 genes. Science 274: 546, 563–567. 2007 PlnTFDB: an integrative plant transcription factor data-
Goodenough, U., H. Lin and J. H. Lee, 2007 Sex determination in base. BMC Bioinformatics 8: 42.
Chlamydomonas. Semin. Cell Dev. Biol. 18: 350–361. Riechmann, J. L., J. Heard, G. Martin, L. Reuber, C. Jiang et al.,
Hubert, L., and P. Arabie, 1985 Comparing partitions. J. Classifica- 2000 Arabidopsis transcription factors: genome-wide compara-
tion 2: 193–218. tive analysis among eukaryotes. Science 290: 2105–2110.
International Human Genome Sequencing Consortium, Rock, C. D., and X. Sun, 2005 Crosstalk between ABA and auxin
2004 Finishing the euchromatic sequence of the human ge- signaling pathways in roots of Arabidopsis thaliana (L.) Heynh.
nome. Nature 431: 931–945. Planta 222: 98–106.
Irish, V. F., 2003 The evolution of floral homeotic gene function. Rossini, L., L. Cribb, D. J. Martin and J. A. Langdale, 2001 The
BioEssays 25: 637–646. maize golden2 gene defines a novel class of transcriptional reg-
Ishida, K., T. Yamashino, A. Yokoyama and T. Mizuno, 2008 Three ulators in plants. Plant Cell 13: 1231–1244.
type-B response regulators, ARR1, ARR10 and ARR12, play essen- Rubio, V., F. Linhares, R. Solano, A. C. Martin, J. Iglesias et al.,
tial but redundant roles in cytokinin signal transduction 2001 A conserved MYB transcription factor involved in phos-
throughout the life cycle of Arabidopsis thaliana. Plant Cell Phys- phate starvation signaling both in vascular plants and in unicel-
iol. 49: 47–57. lular algae. Genes Dev. 15: 2122–2133.
Jaillon, O., J. M. Aury, B. Noel, A. Policriti, C. Clepet et al., Schauser, L., L. Christensen, S. Borg and C. Poulsen, 1995 PZF,
2007 The grapevine genome sequence suggests ancestral hex- a cDNA isolated from Lotus japonicus and soybean root nodule
aploidization in major angiosperm phyla. Nature 449: 463–467. libraries, encodes a new plant member of the RING-finger family
Jiao, Y., O. S. Lau and X. W. Deng, 2007 Light-regulated transcrip- of zinc-binding proteins. Plant Physiol. 107: 1457–1458.
tional networks in higher plants. Nat. Rev. Genet. 8: 217–230. Seoighe, C., and C. Gehring, 2004 Genome duplication led to
Kersey, P., L. Bower, L. Morris, A. Horne, R. Petryszak et al., highly selective expansion of the Arabidopsis thaliana proteome.
2005 Integr8 and Genome Reviews: integrated views of com- Trends Genet. 20: 461–464.
plete genomes and proteomes. Nucleic Acids Res. 33: D297– Shaked, H., N. Avivi-Ragolsky and A. A. Levy, 2006 Involvement
D302. of the Arabidopsis SWI2/SNF2 chromatin remodeling gene fam-
Kumar, S., K. Tamura and M. Nei, 2004 MEGA3: Integrated soft- ily in DNA damage response and recombination. Genetics 173:
ware for molecular evolutionary genetics analysis and sequence 985–994.
alignment. Brief. Bioinform. 5: 150–163. Shiu, S. H., W. M. Karlowski, R. Pan, Y. H. Tzeng, K. F. Mayer et al.,
Lemon, B., and R. Tjian, 2000 Orchestrated response: a symphony of 2004 Comparative analysis of the receptor-like kinase family in
transcription factors for gene control. Genes Dev. 14: 2551–2569. Arabidopsis and rice. Plant Cell 16: 1220–1234.
Li, L., X. Wang, R. Sasidharan, V. Stolc, W. Deng et al., Suzuki, M., C. Y. Kao, S. Cocciolone and D. R. McCarty,
2007 Global identification and characterization of transcrip- 2001 Maize VP1 complements Arabidopsis abi3 and confers a
tionally active regions in the rice genome. PLoS ONE 2: e294. novel ABA/auxin interaction in roots. Plant J. 28: 409–418.
Lin, H., and U. W. Goodenough, 2007 Gametogenesis in the Chla- Tarakhovskaya, E. R., Y. I. Maslov and M. F. Shishova, 2007 Phyto-
mydomonas reinhardtii minus mating type is controlled by two hormones in algae. Russ. J. Plant Physiol. 54: 163–170.
genes, MID and MTD1. Genetics 176: 913–925. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin and D. G.
Merchant, S. S., S. E. Prochnik, O. Vallon, E. H. Harris, S. J. Higgins, 1997 The CLUSTAL_X windows interface: flexible
Karpowicz et al., 2007 The Chlamydomonas genome reveals the strategies for multiple sequence alignment aided by quality anal-
evolution of key animal and plant functions. Science 318: 245–250. ysis tools. Nucleic Acids Res. 25: 4876–4882.
Mittag, M., S. Kiaulehn and C. H. Johnson, 2005 The circadian Tuskan, G. A., S. Difazio, S. Jansson, J. Bohlmann, I. Grigoriev
clock in Chlamydomonas reinhardtii. What is it for? What is it similar et al., 2006 The genome of black cottonwood, Populus trichocar-
to? Plant Physiol. 137: 399–409. pa (Torr. & Gray). Science 313: 1596–1604.
Moseley, J. L., C. W. Chang and A. R. Grossman, 2006 Genome- van Nimwegen, E., 2003 Scaling laws in the functional content of
based approaches to understanding phosphorus deprivation re- genomes. Trends Genet. 19: 479–484.
sponses and PSR1 control in Chlamydomonas reinhardtii. Eukaryot. Vincentz, M., C. Bandeira-Kobarg, L. Gauer, P. Schlogl and A.
Cell 5: 26–44. Leite, 2003 Evolutionary pattern of angiosperm bZIP factors
Negre, B., and A. Ruiz, 2007 HOM-C evolution in Drosophila: Is homologous to the maize Opaque2 regulatory protein. J. Mol.
there a need for Hox gene clustering? Trends Genet. 23: 55–59. Evol. 56: 105–116.
Nei, M., and S. Kumar, 2000 Molecular Evolution and Phylogenetics. Wu, L. C., 2002 ZAS: C2H2 zinc finger proteins involved in growth
Oxford University Press, New York. and development. Gene Expr. 10: 137–152.
Nozaki, H., H. Takano, O. Misumi, K. Terasawa, M. Matsuzaki Wykoff, D. D., A. R. Grossman, D. P. Weeks, H. Usuda and
et al., 2007 A 100%-complete sequence reveals unusually simple K. Shimogawara, 1999 Psr1, a nuclear localized protein that
genomic features in the hot-spring red alga Cyanidioschyzon regulates phosphorus metabolism in Chlamydomonas. Proc.
merolae. BMC Biol. 5: 28. Natl. Acad. Sci. USA 96: 15336–15341.
 Transcription Factors in Chlamydomonas 39

Yasumura, Y., E. C. Moylan and J. A. Langdale, 2005 A conserved Yu, J., S. Hu, J. Wang, G. K. Wong, S. Li et al., 2002 A draft sequence
transcription factor mediates nuclear control of organelle bio- of the rice genome (Oryza sativa L. ssp. indica). Science 296:
genesis in anciently diverged land plants. Plant Cell 17: 1894– 79–92.
1907. Zimmer, A., D. Lang, S. Richardt, W. Frank, R. Reski et al.,
Yehudai-Resheff, S., S. L. Zimmer, Y. Komine and D. B. Stern, 2007 Dating the early evolution of plants: detection and molec-
2007 Integration of chloroplast nucleic acid metabolism into ular clock analyses of orthologs. Mol. Genet. Genomics 278: 393–
the phosphate deprivation response in Chlamydomonas rein- 402.
hardtii. Plant Cell 19: 1023–1038.
Yi, C., and X. W. Deng, 2005 COP1: from plant photomorphogen-
esis to mammalian tumorigenesis. Trends Cell Biol. 15: 618–
625. Communicating editor: S. Dutcher