Ehd4 Encodes a Novel and Oryza-Genus-Specific Regulator of Photoperiodic Flowering in Rice
et al. (2013) Ehd4 Encodes a Novel and Oryza-Genus-Specific Regulator of Photoperiodic Flowering in Rice. PLoS
Genet 9(2): e1003281. doi:10.1371/journal.pgen.1003281
Ehd4 Encodes a Novel and Oryza -Genus-Specific Regulator of Photoperiodic Flowering in Rice
He Gao 0
Xiao-Ming Zheng 0
Guilin Fei 0
Jun Chen 0
Mingna Jin 0
Yulong Ren 0
Weixun Wu 0
Kunneng Zhou 0
Peike Sheng 0
Feng Zhou 0
Ling Jiang 0
Jie Wang 0
Xin Zhang 0
Xiuping Guo 0
Jiu- Lin Wang 0
Zhijun Cheng 0
Chuanyin Wu 0
Haiyang Wang 0
Jian-Min Wan 0
Li-Jia Qu, Peking University, China
0 1 National Key Laboratory for Crop Genetics and Germplasm Enhancement, Jiangsu Plant Gene Engineering Research Center, Nanjing Agricultural University , Nanjing , China , 2 National Key Facility for Crop Gene Resources and Genetic Improvement, Institute of Crop Science, Chinese Academy of Agricultural Sciences , Beijing , China
Land plants have evolved increasingly complex regulatory modes of their flowering time (or heading date in crops). Rice (Oryza sativa L.) is a short-day plant that flowers more rapidly in short-day but delays under long-day conditions. Previous studies have shown that the CO-FT module initially identified in long-day plants (Arabidopsis) is evolutionary conserved in short-day plants (Hd1-Hd3a in rice). However, in rice, there is a unique Ehd1-dependent flowering pathway that is Hd1independent. Here, we report isolation and characterization of a positive regulator of Ehd1, Early heading date 4 (Ehd4). ehd4 mutants showed a never flowering phenotype under natural long-day conditions. Map-based cloning revealed that Ehd4 encodes a novel CCCH-type zinc finger protein, which is localized to the nucleus and is able to bind to nucleic acids in vitro and transactivate transcription in yeast, suggesting that it likely functions as a transcriptional regulator. Ehd4 expression is most active in young leaves with a diurnal expression pattern similar to that of Ehd1 under both short-day and long-day conditions. We show that Ehd4 up-regulates the expression of the ''florigen'' genes Hd3a and RFT1 through Ehd1, but it acts independently of other known Ehd1 regulators. Strikingly, Ehd4 is highly conserved in the Oryza genus including wild and cultivated rice, but has no homologs in other species, suggesting that Ehd4 is originated along with the diversification of the Oryza genus from the grass family during evolution. We conclude that Ehd4 is a novel Oryza-genus-specific regulator of Ehd1, and it plays an essential role in photoperiodic control of flowering time in rice.
Funding: This work was supported by grants from the 864 Program of China (grants 2012AA10A301 and 2012AA100101), the National Natural Science
Foundation of China (grant 31000534), Jiangsu Cultivar Development Program (grant BE2009301-3), and PAPD. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
Flowering is a profound transition from vegetative to
reproductive development in plants, and is largely determined by genetic
pathways that integrate endogenous and environmental signals
. Plants control flowering by perceiving their surroundings, such
as day-length (photoperiod) and temperature that is synchronized
with seasonal changes, in order to maximize their reproductive
fitness . Flowering time or heading date in crops is also a critical
agronomic trait that determines the cropping season and regional
adaptability of plants. Thus, control of flowering time has been
extensively studied by plant breeders and scientists for more than
100 years .
Photoperiod control of flowering refers to the ability of plants to
measure day-length and use it as an indicator to initiate flowering.
Extensive studies in a model long-day plant (LDP), Arabidopsis
thaliana, have revealed that light regulation of the GIGANTEA
(GI)CONSTANT (CO)-FLOWERING LOCUS T (FT) pathway is
essential for integrating cellular signals from light signaling
transduction and the circadian clock to promote flowering under
long-day conditions (LDs) . Phytochrome A (phyA),
phytochrome B (phyB) and cryptochrome 2 (cry2) regulate FT
expression by post-translationally regulating CO protein [7,8].
In addition, blue light promotes CO expression by stabilizing the
FLAVIN-binding KELCH DOMAIN F BOX PROTEIN1
(FKF1)-GI protein complex [9,10]. CO, a zinc finger transcription
factor, promotes FT expression under LDs by directly binding to
its promoter [11,12]. FT, a small mobile protein functioning as the
florigen, is synthesized in the phloem of leaves, and is then
transported to the apical meristem where it initiates flowering by
inducing the expression of the floral meristem identity genes, such
as AP1 .
Rice (Oryza sativa L.) is an important source of calories for
mankind and a model short-day plant (SDP) that flowers more
rapidly in short-day conditions (SDs) but delays under LDs with a
critical day-length response [16,17]. Previous studies have revealed
that rice flowering is regulated both by a SD-activation pathway
and a LD-suppression pathway. OsGIGANTEA (OsGI), Heading
date 1 (Hd1) and Heading date 3a (Hd3a) have been identified as the
counterpart of GI, CO and FT, respectively . Hd1 executes
dual function that promotes flowering by regulating Hd3a (the
major SD florigen) expression under SDs, but suppresses it
through unknown mechanisms under LDs [19,21,22]. However,
the OsGI-Hd1-Hd3a pathway only plays a limited role in flowering
Rice is an important source of calories for mankind.
Flowering time determines cropping seasons and regional
adaptability of crops. Rice is originated from its wild
progenitor, O. rufipogon, which is mainly distributed at
tropical latitudes with a northern limit about 28 uN, more
than 10,000 years ago. However, cultivated rice is now
grown widely in Asia, with a northern limit of nearly 53 uN.
The northward expansion of cultivated rice must be
accompanied by human selection of the flowering time
trait during domestication and breeding, to secure a
harvest before cold weather approaches. By identifying a
rice mutant that never flowers under natural long-day
conditions (NLDs), we cloned Ehd4 as a novel
transcriptional regulator that promotes flowering through
activation of two florigen genes, the signals for flowering
initiation. We found that Ehd4 has two major haplotypes:
Hap_2 is the major haplotype in indica accessions mostly
distributed in lower latitude and elevation zones, whereas
Hap_3 is the major haplotype in japonica accessions
mostly distributed in higher latitudes and elevation zones.
Genetic studies showed that Hap_3 is functionally more
potent in promoting flowering under NLDs, implying that
Ehd4 may have contributed to the northward expansion
and regional adaptability of cultivated rice into higher
time control in rice because there is a high degree of
polymorphism in Hd1 and non-functional alleles of Hd1 are
associated with only moderate phenotypic changes .
Rice has a unique, Hd1-independent flowering pathway that is
mediated by Early heading date 1 (Ehd1). Ehd1 encodes a B-type
response regulator that is highly conserved in cultivated rice, but
has no homolog in Arabidopsis [23,24]. It has been shown that
Ehd1 positively regulates the expression of Hd3a and RICE
FLOWERING LOCUS T 1 (RFT1), the closest paralog of Hd3a
that works as a LD florigen [22,24,25]. Circumstantial evidence
suggests that Ehd1 is a critical convergence point of regulation by
multiple signaling pathways. Among them, OsphyB inhibits
flowering under both SDs and LDs by suppressing Hd3a
expression through posttranslational modification of HD1 protein
function and transcriptional suppression of Ehd1 expression [25
27]. The OsphyB-mediated suppression of Ehd1 is regulated by
OsCOL4, which encodes a protein containing two B-box zinc
finger domains and one CCT domain and it also acts as a
constitutive suppressor of flowering in rice under both SD and LD
conditions [26,27]. In addition, both Ghd7 (Grain number, plant height
and heading date 7), encoding a CCT domain protein , and
DTH8 (Days to heading 8), encoding a putative HAP3 subunit of the
CCAAT-box-binding transcription factor, down-regulate Ehd1
expression and delay flowering under LDs . On the other
hand, Ehd1 expression is promoted by a number of positive
regulators. Among them, OsMADS51 encodes a type I MADS-box
protein and induces Ehd1 expression under SDs , whereas a
rice homolog of Arabidopsis SOC1 (Suppressor of Overexpression of
Constant1), OsMADS50, was identified as a promoter of Ehd1
expression under LDs . Recently, it was shown that Ehd1
expression could be independently up-regulated by Early heading
date 2/Rice Indeterminate 1/Oryza sativa Indeterminate 1 (referred to as
Ehd2 hereafter) and Early heading date 3 (Ehd3) under both SDs and
LDs . The former encodes a Cys2/His2-type zinc finger
protein with high homology to maize Indeterminate1 , while the
latter encodes a putative plant homeodomain (PHD)
fingercontaining protein. Notably, loss of function of OsMADS50, Ehd2
and Ehd3 showed a never-flowering phenotype under LDs
[25,32,34,36]. Thus, it appears that OsMAD50, Ehd2, Ehd3 and
Ehd1 may constitute a LD-activation pathway in rice. Although
these studies have revealed much insight into the photoperiodic
flowering of rice, the underlying molecular mechanisms are still
not well understood.
Here we report the identification of Early heading date 4 (Ehd4)
using a mutagenesis approach and its positional cloning. Ehd4
encodes a novel CCCH (C-X7-C-X5-C-X3-H)-type zinc finger
protein and it acts as a critical regulator promoting flowering
under both SDs and LDs, particularly under LDs. Mutation in
Ehd4 causes a never-flowering phenotype under natural long-day
conditions (NLDs). EHD4 protein is localized to the nucleus and it
has nucleic acid-binding and transcriptional activation properties,
consistent with a plausible function as a transcription factor. We
show that Ehd4 promotes flowering by up-regulating the
expression of Hd3a and RFT1 through stimulation of Ehd1 expression.
Interestingly, Ehd4 is highly conserved in the Oryza genus and it
has no homologs in other plant species. Thus, our findings
identified a novel, highly conserved rice-specific regulator of
Characterization of the late flowering mutant ehd4
In an effort to isolate genes that are essential for promoting
flowering time in rice, we generated a large T-DNA population in
a day-length neutral, early flowering variety Kita-ake (O. sativa ssp.
japonica). Kita-ake (Kit) has been widely used in rice transformation
experiments because of its short life cycle. Kit flowers about two
months after germination under both SDs (10 h light/14 h dark)
and LDs (14.5 h light/9.5 h dark) conditions in the controlled
growth chamber, as well as under natural long-day field conditions
(NLDs) in Beijing (39u549N, 116u239E), North China (Figure 1A
and 1B). To understand the day-length neutral nature of Kita-ake,
we cloned ten genes reported to have significant effect on
flowering time in rice, including seven genes that promote
flowering (Ehd 1 to 3, OsMADS50, OsMADS51, Hd3a and RFT1)
and three genes that suppress flowering under LDs (Hd6, Hd1 and
Ghd7), and compared them with the corresponding genes in
Nipponbare (Nip), a japonica variety that is sensitive to day-length.
Those flowering-promoting genes are identical in Kit and Nip
varieties, except OsMADS51 that contains one amino acid
variation (Figure S1). In contrast, Kit has an immature stop in
Ghd7 and a 36-bp insertion and two amino acid changes in Hd1
(Figure S1). Although there is no difference in Hd6 sequences
between Kit and Nip, both of them have an early stop compared
with the allele of the indica variety Kasalash (Figure S1), which
delays flowering in Nip background under LDs [37,38].
Therefore, complete or partial loss of function of those three genes in Kit
could at least partially explain its insensitivity to day-length.
We screened our T-DNA population in the NLDs and
identified a mutant that failed to flower during the 160 days of
growing season (from late April to early October, 2006), whereas
the wild-type (WT) Kit flowered 55 days after germination
(Figure 1A and 1B). We were able to produce seeds by moving this
mutant plant to a controlled SDs. Plants from an F2 population
derived from a cross of the mutant and WT segregated in field
conditions into three categories based on their flowering time (days
after germination): 56.961.8, 70.861.8 and never flowering
mutants in a ratio of 1:2:1 (x2[1:2:1] = 0.415,x20.05,2 = 5.99,
n = 200). This result indicates that the mutation is semidominant
and is controlled by a single gene. We named this locus Ehd4 (Early
heading date 4). Compared with WT, ehd4 delayed flowering time by
Figure 1. Characterization of Ehd4. (A) Never-flowering phenotype of ehd4 mutants in field (Top). WT, Kita-ake wild-type plants (Bottom). (B)
Flowering time of ehd4, heterozygote (HETE) and WT plants under different day length conditions in Kita-ake (day-length neutral) and Nipponbare
(day-length sensitive) backgrounds (n = 12). ND, natural-day; SD, short-day; LD, long-day. (C) ehd4 plants had the same leaf emergence rate as WT
(Kita-ake) under both SDs and LDs (n = 8). Arrow indicates the flowering time of WT plants. (D) Panicle morphology of WT and ehd4 plants. (E) to (H)
Comparisons of grain number per panicle (E), 1000-grain weight (F), plant height (G) and fertility (H) between WT and ehd4 plants. Values are
means6s.d. (standard deviations) (n = 15). **Significant at 1% level; n.s., not significant.
49 d and 106 d under SDs and LDs, respectively (Figure 1B).
Consistent with field observations, flowering time of the
heterozygotes was also delayed under both SDs and LDs (Figure 1B).
Notably, ehd4 had a similar leaf emergence rate to WT under both
SDs and LDs (Figure 1C), indicating that the late flowering
phenotype is not caused by retardation in growth rate. The mature
ehd4 plants were taller, producing more but smaller seeds. The
fertility of ehd4 plants was similar to that of WT (Figure 1D1H).
To test whether the delayed flowering phenotype is genetic
background-dependent, we introduced the ehd4 locus into Nip by
backcrossing five times, followed by selfing. The ehd4-NIP plants
(BC5F3) delayed flowering by 23 d under SDs compared to the
WT NIP, but did not flower in NLDs or LDs (Figure 1B).
Flowering time of the heterozygotes was also significantly delayed
(Figure 1B). Thus, ehd4 has a profound effect on flowering time,
especially under LDs, in both a day-length neutral and a
daylength sensitive genetic backgrounds.
Molecular cloning of Ehd4
Flowering time control in rice is regulated by the interaction of
multiple QTLs and the environments. Generally, flowering time of
F2 population derived from japonica6indica displays a normal
distribution pattern. Therefore, we crossed ehd4 with 93-11, an
indica variety with an available genome sequence , and
generated a BC1F2 population for mapping the ehd4 locus by
backcrossing the F1 with 93-11. The ehd4 locus was initially
mapped to the short arm of chromosome 3 (Figure 2A). Using 871
extremely late flowering plants from approximately 25,000 BC1F2
plants grown in Hainan Island (18u489N, 110u029E, average day
length 11 hours), South China, during the winter of 2008, we
further delimited Ehd4 to a 103 kb region, between the markers
EJ-4 and EJ-5 (Figure 2B). This region contains 16 annotated
ORFs (http://rapdb.dna.affrc.go.jp) (Figure 2C). Sequencing of
the genomic DNA of all these genes revealed that there is a single
nucleotide substitution (G to A) in the first exon of
LOC_Os03g02160, which is predicted to encode a CCCH-Type
zinc finger protein. The nucleotide change creates a premature
stop codon at the very beginning of the predicted coding region
(Figure 2D). Genomic sequence of this gene is identical between
Kit and Nip (Figure S1). Quantitative real-time PCR (qRT-PCR)
assay showed comparable expression of LOC_Os03g02160 in wild
type, heterozygote and ehd4 mutant plants (Figure S2). Transgenic
plants carrying the full-length cDNA of LOC_Os03g02160, driven
by the maize Ubiquitin-1 promoter, fully complemented the ehd4
phenotype under both SDs and LDs. Further, cDNA driven by its
native promoter (2.7 kb upstream from ATG) also partially
rescued the ehd4 phenotype. The phenotypes of these transgenic
lines (days to flowering) appeared to correlate with the expression
level of Ehd4 (Figure 2E and Figure S2). Thus, we concluded that
the LOC_Os03g02160 locus corresponds to Ehd4.
Expression of Ehd4 is constitutive and diurnal
We examined the expression levels of Ehd4 in various tissues and
at different stages of leaf development (Figure 3A) by using
qRTPCR. Ehd4 transcripts were detected in all tissues examined, but
the highest expression was found in emerging young leaves and the
lowest level in fully expanded leaves (Figure 3B). Histochemical
staining of transgenic plants carrying the GUS reporter gene driven
by the Ehd4 promoter indicated that GUS was expressed in all
tissues examined and was most abundant in the vascular tissue and
apical meristem (Figure 3C3I). The expression of Ehd4 showed a
diurnal expression pattern in leaves. It accumulates after dusk,
reaching a peak at dawn, and damping rapidly thereafter under
both SDs and LDs (Figure 3J). Moreover, Ehd4 was expressed
constantly during the vegetative growth from the second week to
the 10th week after germination (Figure 3K).
EHD4 may act as a transcriptional regulator
In higher plants, CCCH-type zinc finger proteins have been
shown to regulate gene expression by binding to DNA or RNA
molecules in the nucleus . We fused Ehd4 with GFP and
transiently expressed the EHD4-GFP fusion protein in rice leaf
protoplasts. EHD4-GFP was exclusively co-localized with the
OsMADS3-mCherry fusion protein (a nuclear marker) in the
nucleus (Figure 4A4C), indicating that EHD4 functions in the
nucleus. We further fused EHD4 and its various deletions with the
GAL4 DNA binding domain and investigated if EHD4 has
transcriptional activation activity in yeast. Full-length wild type
EHD4 and an EHD4 variant with only the CCCH motif removed
were able to activate the reporter gene expression (Figure 4D).
Further deletion of the C terminal region resulted in a dramatic
reduction of the activation activity, whereas deletion of both the
N-terminal and CCCH motif only had mild effects (Figure 4D).
These observations suggest that the activation domain is located in
the middle region close to the C-terminal of EHD4. In addition, a
nucleic acid binding assay demonstrated that the C-terminal
region, but not the N-terminal region, can bind to both
doubleand single-stranded calf thymus DNA and ribohomopolymers in
vitro, and that removal of the CCCH motif from the C-terminal
abolished the binding activity (Figure 4E). These results strongly
support the notion that EHD4 likely functions as a transcriptional
activator and that the CCCH motif is essential for its nucleic acid
Ehd4 regulates expression of the florigen genes
Photoperiodic induction of the floral transition in rice requires
transcriptional activation of Hd3a and RFT1, the two florigen
genes, in leaves . The diurnal expression pattern of Ehd4
implies that it could be involved in photoperiodic control of
flowering. To test this, we examined mRNA abundance of Hd3a
and RFT1 in ehd4 and WT plants by qRT-PCR. The expression
levels of Hd3a and RFT1 were undetectable in ehd4 under both
SDs and LDs at all-time points examined during the 48 h period
(Figure 5A5D). Subsequently, expression of the downstream genes
OsMADS1, OsMADS14 and OsMADS15 (three floral meristem
identity genes; [22,25]) was severely impaired in the ehd4 mutants
Hd1 and Ehd1 are known to regulate Hd3a and RFT1 [19,24].
To investigate whether the activation of florigen genes by Ehd4 is
mediated by Hd1 and/or Ehd1, we compared their mRNA levels
between ehd4 and WT plants. Strikingly, expression of Ehd1, but
not Hd1, was abolished in ehd4 mutants, indicating that Ehd4 is
essential for Ehd1 expression (Figure 5E5H). Ehd4 has a diurnal
expression pattern similar to that of Ehd1, typically peaking at
dawn (Compare Figure 3J with Figure 5E and 5F). Next, we
examined whether Ehd4 affects the expression of other known
regulators of Ehd1. To our surprise, the transcription levels of five
positive regulators (Ehd2, Ehd3, OsMADS50, OsGI and OsMADS51)
and four negative regulators of Ehd1 (OsphyB, OsCOL4, DTH8 and
Ghd7) were not significantly affected in ehd4 (Figure 6A and 6B).
These observations suggest that Ehd4 functions upstream of Ehd1,
but largely independent of other known regulators of Ehd1.
Consistent with this, down regulation of Ehd1, Hd3a and RFT1 in
ehd4 was also seen in the Nipponbare background and constantly
seen at different stages during plant development (Figure 6B and
To investigate whether Ehd4 expression is regulated by other
flowering genes, we examined the expression of Ehd4 in osphyb,
ehd2, ehd3, osmads50 and osmads51 mutants and near-isogenic lines
(NILs) which carrying a deficient Hd1, Ghd7 or DTH8 alleles.
Notably, we detected no significant differences of Ehd4 expression
in these mutants or NILs, as compared to their corresponding WT
plants (Figure 6C). In addition, no significant change of Ehd4
expression was seen in NILs deficient in Ehd1 or Hd3a either
(Figure 6C). These results suggest that Ehd4 acts independent of
other Ehd1 regulators we examined. Together, these observations
suggest that Ehd4 regulates the expression of Hd3a and RFT1
through Ehd1. This notion was also supported by the observation
that over-expression of Ehd1 fully rescued the late flowering
phenotype of ehd4 under SDs (Figure 7).
Since EHD4 has a transcriptional activation and nucleic acid
binding activity and it promotes Ehd1 expression, we next carried
out a yeast one-hybrid assay and a transient transcription assay
[43,44] to test whether Ehd1 is a direct downstream target of
EHD4. However, only OsLFL1 , but not EHD4, was able to
interact with the Ehd1 promoter (Figure S5), indicating that Ehd1 is
likely an indirect target of Ehd4. In addition, yeast three-hybrid
assay  also failed to detect a direct binding of EHD4 to the
Ehd1 mRNA (Figure S6). Moreover, neither EHD2 nor EHD3,
directly binds to the Ehd1 promoter (Figure S5). Our yeast
twohybrid assay showed that there was no direct interaction among
the EHD2, EHD3 and EHD4 proteins (Figure S7). Together,
these results suggest that Ehd2, Ehd3 and Ehd4 likely act through
distinct pathways to promote the expression of Ehd1.
Transcriptome analysis of ehd4 plants
To further reveal the molecular basis of the flowering phenotype
of ehd4, we performed a transcriptome analysis of ehd4 and
wildtype plants using RNA-seq to identify genes downstream of Ehd4.
RNA samples were extracted from the penultimate leaves
(collected at dawn) of 30 d-old ehd4 and WT plants (Kita-ake)
grown under LDs. We obtained 2.5 M tags and found a total of
256 genes altered in expression with an estimated false-discovery
rate of 0.1% and the absolute value of log2Ratio at 3.15 under the
Bayesian model (Table S1; ). We found that the transcript
numbers of Hd3a, RFT1, Ehd1, OsMADS1, OsMADS14 and
OsMADS15 reduced dramatically in ehd4 plants (Table S1),
consistent with the qRT-PCR results (Figure 5A5F and Figure
S3). Our qRT-PCR analysis with other four genes (with a log2
Ratio of 211.73, 210.33, 25.80 and 23.15, respectively) also
further confirmed the reliability of the RNA-seq results (Figure
S8). Notably, we found that among the genes down-regulated in
end4, 25 of them are known or putative transcription factors,
including MADS box, Zinc finger, MYB, SBP and B3 proteins
(Table S1). These genes could be potential candidates involved in
the Ehd4-Ehd1-Hd3a/RFT1 pathway to regulate photoperiodic
flowering in rice.
Ehd4 is unique and highly conserved in rice
Ehd4 is a single copy gene in the rice genome. It is predicted to
code for a polypeptide of 832 amino acids long, which contains a
CCCH (C-X7-C-X5-C-X3-H)-type zinc finger motif at the
Cterminus (Figure S9). A blast search (http://www.ncbi.nlm.nih.
gov/) found that EHD4 has no clear homologs in other plant or
animal species. Thus it appears that Ehd4 represents a unique
regulator of flowering time in rice.
To investigate the evolutionary history of the gene in rice, we
analyzed the Ehd4 sequences from 86 rice accessions with wide
geographic distribution and diverse genetic backgrounds,
including 32 wild rice species (O. rufipogon and O. nivara) and 54 cultivated
rice (Table S2; ). Ehd4 appears to be highly conserved across
these accessions (share .99.2% or higher amino acid sequence
identities) (Table S2). Sequence analysis identified 25 haplotypes
among these accessions. Strikingly, 21 haplotypes were identified
in 32 wild rice accessions (O. rufipogon and O. nivara) but only 8
haplotypes were identified in 54 cultivated rice accessions analyzed
in this study. The dramatic reduction in genetic diversity at this
locus suggests that Ehd4 might subject to bottleneck effect .
Notably, 4 haplotypes (Hap_2, 3, 6 and 7) are shared in cultivated
and wild rice (Figure 8A, Table S2), and among them, Hap_2 and
Hap_3 together account 25% of the 32 wild rice and 85% of the
54 cultivated rice (O. sativa) respectively (Figure 8A, 8B and Table
S2). Among the cultivated rice, 20 (77%) indica and 6 (23%) japonica
accessions belong to Hap_2, while 19 (95%) japonica and 1 (5%)
indica accessions belong to Hap_3 (Figure 8A and Table S2). This
result suggests that the Hap_2 and Hap_3 represent the two major
haplotypes at the Ehd4 locus in cultivated rice and that they exist in
wild rice before domestication. The distribution pattern of Hap_2
and Hap_3 in indica (mostly distributed in lower latitude and
elevation zones) and japonica (mostly distributed in higher latitude
and elevation zones) (Figure 8C) implies a likely correlation
between the geographic distribution and the functional differences
of Ehd4 haplotypes among these cultivated accessions analyzed.
To test the possible functional differences of these two
haplotypes, we introduced the full-length Ehd4 cDNA of indica
variety 93-11 (Hap_2) and Kita-ake (a japonica landrace, Hap_3)
driven by the maize Ubiquitin-1 promoter into the ehd4 mutant
(Kita-ake background). Strikingly, over-expression of Hap_3, but
not Hap_2, fully complemented the ehd4 phenotype under NLDs,
although their expression levels are comparable (Figure 8D and
Figure S10). This result suggests that Hap_3 of Ehd4 is functionally
more potent in promoting flowering than Hap_2. As another test
of this notion, we introduced the Hap_3 allele from Kita-ake into
93-11 by backcrossing five times, followed by selfing. Strikingly,
the NIL Ehd4Hap_3 plants (BC5F3) flowered earlier by 19 d under
NLDs compared to the parental 93-11 plants (Figure 8D).
Together, these results suggest that the functional differences of
Ehd4 haplotypes might play a role in geographic adaptation of
In this study, we have uncovered Ehd4, which codes for a
CCCH-type zinc finger protein essential for promoting flowering
under both SD and LD conditions in rice, irrespective of genetic
backgrounds. We demonstrated that Ehd4 promotes flowering by
positively regulating the expression of Hd3a and RFT1 through
Ehd1 but independent of other known important Ehd1 regulators.
We further showed that the late-flowering phenotype of ehd4 is
more profound in Kita-ake (Kit) (a day-length neutral variety) than
in Nipponbare (Nip) (a day-length sensitive variety) under SDs, but
ehd4 plants flowered eventually in Kit (164 days after germination)
but not in Nip under LDs (Figure 1B). It is known that Hd1 acts to
promote flowering under SDs but delay flowering under LDs .
We found that Hd1 has a 36-bp insertion and two SNPs in Kit
compared to Nip (Figure S1), implying that in Kit, the promoting
role under SDs and repressing role under LDs of Hd1 may be
impaired. This may at least partially explains why the
lateflowering phenotype of ehd4-Nip is less severe under SDs but more
severe under LDs. In addition, we found that Kit carries a
truncated allele of Ghd7, a repressor of Ehd1 under LDs, whereas
Nip carries a partially functional Ghd7 allele (Figure S1; ).
Therefore, the never-flowering phenotype of ehd4 in Nip under
LDs could be due to the repressive effect of Hd1 and Ghd7.
Strikingly, even in the absence of the functional Hd1 and Ghd7
alleles in Kit, ehd4 alone delayed flowering time by three folds
under LDs (Figure 1B), suggesting that Ehd4 plays a major role in
promoting flowering in rice, particularly under LDs.
Previous studies revealed that rice Ehd1 is a critical convergence
point of flowering time regulation by multiple signaling pathways
and that Ehd1 acts independently of Hd1. Ehd1 encodes a B-type
response regulator that is highly conserved in cultivated rice, but
has no homolog in Arabidopsis [23,24]. Up to date, 12 genes have
been shown to regulate Ehd1 expression, including 5 positive
regulators (Ehd2, Ehd3, OsGI, OsMADS50 and OsMADS51) and 7
negative regulators (SE5, OsphyB, Ghd7, DTH8, OsLFL1, OsCOL4
and OsMADS56). We demonstrated that Ehd4 promotes flowering
by positively regulating the expression of Hd3a and RFT1 through
Ehd1, but independently of these known Ehd1 regulators.
It is also of interest to note that the majority of Ehd1 regulators
uncovered thus far are nuclear proteins and many of them act as
transcriptional regulators, including GHD7, DTH8, OsMAD50,
OsMAD51, EHD2, EHD3, OsLFL1, OsMAD56 and OsCOL4
[26,2834,36,45,50]. Map-based cloning revealed that Ehd4
encodes a novel CCCH-type zinc finger protein also localized to
the nucleus. The CCCH-type zinc finger protein family is defined
as a group of proteins containing 16 copies of the canonical
C-XC-X-C-X-H motif (C-X614-C-X45-C-X3-H, where X is any
amino acid) . This type of proteins has been found in
organisms ranging from human to yeast and many of them have
been shown to have either an RNA binding function involved in
RNA processing or DNA binding activity . There are at
least 68 CCCH-type genes in Arabidopsis and 67 in rice,
respectively . However, only a few plant CCCH proteins
have been functionally characterized. EHD4 is the first
CCCHtype protein found to regulate photoperiodic flowering. We found
that EHD4 is capable of binding to nucleic acids in vitro and
transactivate transcription in yeast, suggesting that it likely
functions as a transcription factor. Further, our transcriptome
analysis revealed that a significant portion of Ehd4-regulated
downstream genes are also transcription factors, including several
previously identified flowering regulators, such as Ehd1, OsLFL1
OsMADS1, OsMADS14 and OsMADS15, and other putative
transcription factors, including MADS box, Zinc finger, MYB,
SBP and B3 proteins (Table S1). These findings together suggest
that transcriptional regulation plays a critical role in photoperiodic
regulation of flowering in rice. However, despite we have
demonstrated that EHD4 has double-stranded DNA and
ribohomopolymer binding activity and transactivation activity in yeast,
we have not been able to identify the direct target genes of EHD4
in this study. It is also possible that EHD4 may bind to RNA
molecules and degrades transcripts of unknown Ehd1 repressors.
Further studies are required to elucidate the biochemical function
of EHD4 and its functional relationship with other nuclear
regulators of flowering time.
Rice is known as a short day plant. However, cultivated rice (O.
sativa) is grown widely in Asia, with a northern limit of nearly 53uN
in northern Asia (Northern provinces of China and Korea, where
natural day length during rice cultivation is nearly 15 hours light;
), whereas O. rufipogon, a wild rice that is the most relative
ancestor of O. sativa, is mainly distributed at tropical latitudes with
a northern limit about 28uN . The northward expansion of
cultivated rice into higher latitudes must be accompanied by
human selection of the flowering time trait during rice
domestication and breeding, to secure a harvest before cold weather
approaches. Strikingly, Ehd4 has no obvious homologs in other
plant species including Arabidopsis, maize and sorghum,
suggesting that Ehd4 originated along with the diversification of the Oryza
genus from the grass family during evolution. Amino acid
sequence comparison of EHD4 showed identities at 99.2% or
higher among a core collection of rice germplasm with wide
geographic distribution and diverse genetic backgrounds,
including wild rice species (; Table S2). Interestingly, we found two
major haplotypes of Ehd4, Hap_2 (the major haplotype in indica)
and Hap_3 (the major haplotype in japonica) and that Hap_3 is
functional more potent in promoting flowering under NLDs. Since
indica rice is known to distribute mostly in lower latitude and
elevation zones (between the latitude 3uS-35uN), while japonica
varieties are mostly distributed in higher latitude and elevation
zones (between the latitude 15uN-53uN), our findings suggest that
Ehd4 may have contributed to the northward expansion and
regional adaptation of cultivated rice into higher latitudes.
Materials and Methods
Plant material and growth conditions
The ehd4 mutant was initially identified from a tissue
culturederived population of rice cv Kita-ake (japonica) under natural-day
conditions in a paddy field in Beijing (39u549N, 116u239E), China
(2006). To generate ehd4-nip plants, the mutant locus was
introgressed into Nipponbare (japonica) background by crossing
and backcrossing for five generations (BC5), where ehd4 mutant is
the donor parent and Nipponbare is the recurrent parent, by using
marker assisted selection (MAS).
Plants were grown in controlled-growth chambers (Conviron)
under SDs (10 h light at 30uC/14 h dark at 25uC) or LDs (14.5 h
light at 30uC/9.5 h dark at 25uC) with a relative humidity of
,70%. The light intensity was ,800 mmol m22 s21.
To map the ehd4 locus, the mutant was crossed with the indica cv
93-11 and then the F1 plants were backcrossed with 93-11 to
produce a BC1F2 population. We used two DNA pools generated
from 15 BC1F2 late-flowering and 15 normal plants, respectively,
for rough mapping. For fine mapping, 871 never-flowering plants
segregated in the BC1F2 population were used.
Vector construction and plant transformation
For the complementary test, the Ehd4 full-length cDNA driven
by its native (2.7-kb) or the maize Ubiquitin-1 (Ubi) promoter were
cloned into the binary vector pCAMBIA1390 by using In-Fusion
Advantage PCR Cloning Kits (Clontech) to create pEhd4::Ehd4
and pUbi::Ehd4, respectively. The 2.7-kb long promoter was also
cloned into pCAMBIA-1305.1 to create pEhd4::GUS. The
resultant plasmids were transformed into the Agrobacterium
tumefaciens strain EHA105 and then introduced into ehd4 (for
complementary test) or Kita-ake WT plants (pEhd4::GUS). At least
15 transgenic events were produced for each construct.
Subcellular localization of EHD4
GFP was fused to the C-terminus of EHD4 under the control of
the 35S CaMV promoter in the pA7 vector. The EHD4-GFP
fusion and the nucleus marker OsMADS3-mCherry were
transiently co-expressed in rice leaf protoplasts by PEG
(polyethylene glycol) treatment . Fluorescence was observed using a
Leica TCS-SP4 confocal microscope.
Transactivation activity assay
Transactivation activity assay was performed using the
Matchmaker GAL4 Two-Hybrid System 3 (Clontech). Plasmids
containing GAL4 DNA binding domain fused with EHD4
deletions were transformed into the yeast strain AH109. The
substrate chlorophenol red-b-D-galactopyranoside (CPRG; Roche
Biochemicals) was used to measure the b-galactosidase activity
according to the Yeast Protocols Handbook (Clontech).
In vitro nucleic acid binding assay
EHD4 deletions were cloned into the pMAL-c2x (NEB) and
expressed in E. coli. 0.5 mg purified protein was incubated with
20 mL of poly rG, poly rC, or poly rU attached to agarose beads
or double- or single-stranded calf thymus DNA attached to
cellulose beads (Sigma) in 500 mL of RHPA binding buffer
(10 mM Tris, pH 7.4, 2.5 mM MgCl2, 0.5% Triton X-100, NaCl
at various concentrations) with 1 mg/mL heparin. After
incubation at 4uC for 10 min, the beads were washed five times in the
RHPA buffer and then boiled in the SDS loading buffer. Binding
of fusion proteins to RNA or DNA was confirmed by protein gel
blot using anti-MBP antibodies (NEB).
Yeast one-hybrid assay
Yeast one-hybrid assay was performed according to the method
described in . To generate GAD-EHD4, GAD-EHD2,
GADEHD3 and GAD-OsLFL1, their full-length cDNAs were cloned
into pJG4-5 vector. To generate the Ehd1p::LacZ reporter gene, a
3.2 kb fragment of Ehd1 promoter (including the 59-UTR) was
haplotypes of Ehd4 driven by the maize Ubiquitin-1 promoter in ehd4 (Kita-ake background) and NIL carrying Ehd4Hap3 compared with the 93-11
parental plants. T2 plants of two pUbi::Ehd4Hap3 (#18 and #24) and two pUbi::Ehd4Hap2 (#12 and #16) lines were measured (n = 15). All plants were
grown in the natural long day field conditions. Values are means6s.d. (standard deviations) (n = 15). **Significant at 1% level; n.s., not significant.
amplified from Nipponbare genomic DNA and inserted into the
corresponding sites of the reporter plasmid pLacZi2m. Plasmids
were co-transformed into the yeast strain EGY48. Transformants
were grown onto SD/Trp-/Ura plates for 48 hours and then
transferred onto X-gal
(5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside) plates for blue color development.
Yeast three-hybrid assay
Yeast three-hybrid assay was performed according to the
method described in . Full-length Ehd4 cDNA was cloned
into pACTII vector to generate GAD-EHD4. A series DNA
fragments (about 140 bp) that transcripting Ehd1 mRNA sequence
were cloned into pMS2-1 vector. Plasmids were co-transformed
into the yeast strain YBZ-1. Transformants were grown on plates
containing selective media (SD/Ura-/Leu-/His-+0, 2 or 8 mM
3aminotriazole) for 48 hours before assay.
Isolation of rice leaf protoplast and PEG-mediated transfection
were performed as described . The reporter construct
pGreenEhd1p-LUC and effector plasmids (pEGAD-MycEHD4, EHD2,
EHD3 or OsLFL1) were co-transformed into protoplasts. After
transformation, the protoplasts were incubated in darkness for 12
16 h. Bioluminescence assay was performed according to the
method described in .
Quantitative real-time RT-PCR
Total RNA was extracted using the RNeasy Plant Mini Kit
(QIAGEN). For quantitative real-time RT-PCR, the first-strand
cDNA was synthesized using the QuantiTect Reverse Transcription
Kit (QIAGEN) and then PCR was performed using gene-specific
primers and SYBR Premix ExTaq reagent (Takara) with an ABI
Prism 7900 HT Sequence Detection System (Applied Biosystems)
according to the manufacturers instructions. PCR reactions were
carried out in triplicate for each sample from two independent
biological replicates and the rice Ubiquitin-1 gene was used as the
RNAseq (next-generation sequencing of RNA)
We used the Illumina HiSeq 2000 Genome Analyzer to get tags
with CATG site, in which the adapter sequences are 2*100 bp. With
Illuminas digital gene expression assay, we obtained 11.7 million
sequence tags per sample. After removing low quality reads and low
quality bases of quality value, clean reads were mapped to the O. ssp.
japonica reference sequences using SOAPaligner/soap2. Mismatches
of not more than one base were allowed in the alignment and we
generated 8.9 million perfect match tags (76.08%) for each sample.
Initially, we determine 27020 genes of significant differences in
expression between the groups of wild-type and mutants by a
Students t-test. With a dedicated Bayesian model, we found 256
transcripts of differential expression with an estimated false-discovery
rate of 0.1% and the absolute value of log2Ratio is more than 3.15.
All primers used in this study are listed in Table S3.
Data deposition: The Ehd4 sequence reported in this paper has
been deposited in the GenBank database accession no. JQ828863
Figure S2 Transcript levels of Ehd4 in WT (Kita-ake),
heterozygote (HETE), ehd4 and transgenic plants. Penultimate
leaves were harvested at dawn from 28 d-old (SDs) and 35 d-old
(LDs) plants. The rice Ubiquitin-1 (UBQ) gene was used as the
internal control. Values are shown as mean6s.d (standard
deviations) of three independent experiments and two biological
Figure S3 Transcript levels of OsMADS14, OsMADS15 and
OsMADS1 in WT (Kita-ake) and ehd4 plants. Penultimate leaves
were harvested around reported peak expression level of each gene
during the 24 hrs photoperiod - at dawn from 28 d-old (SDs) and
35 d-old (LDs) plants. The rice Ubiquitin-1 (UBQ) gene was used as
the internal control. Values are shown as mean6s.d (standard
deviations) of three independent experiments and two biological
Figure S4 Developmental expression pattern of Ehd1, Hd3a and
RFT1 in WT (Kita-ake) and ehd4 plants under SDs (A, C and E)
and LDs (B, D and F). The rice Ubiquitin-1 (UBQ) gene was used as
the internal control in the quantitative RT-PCR analysis. Values
are shown as mean6s.d. (standard deviations) of three
independent experiments and two biological replicates.
Figure S5 Yeast One-Hybrid and Bioluminescence assays. (A)
GAD-LFL1 (positive control; ), but not GAD-EHD2,
GADEHD3, GAD-EHD4 or GAD itself (negative control), strongly
activate expression of the LacZ reporter genes driven by the Ehd1
promoter (3.2 kb upstream of the ATG start codon) in yeast
onehybrid assay. (B) Structure of the vector used for transient
expression. 35S, 35S CaMV promoter; REN, renilla luciferase;
35S mini, 35S CaMV minimum promoter; LUC, luciferase gene;
35S term, 35S CaMV terminator. (C) Relative reporter activity
(LUC/REN) in rice protoplasts. Bioluminescence assays showing
that expression of Ehd1::LUC reporter was not induced by EHD4,
EHD2, EHD3 or GAD (empty vector) itself but strongly repressed
by LFL1 (positive control; ) in rice leaf protoplasts. The
relative LUC activities normalized to the REN activity are shown
(LUC/REN, n = 3).
Figure S6 Yeast Three-Hybrid assay. (A) Genomic structure of
the Ehd1 locus. Regions used for assays in (B) are underlined and
numbered in order. (B) Yeast Three-Hybrid assays showing that
EHD4 did not interact with any part of Ehd1 mRNA. The plasmid
pIIIA/IRE-MS2 expressing 59 IRE-MS2 39 hybrid RNA from the
yeast RNAseP promoter and the plasmid pAD-IRP expressing the
rabbit Iron Regulatory Protein fused to the Gal 4 Activation
Domain was used as the positive control  and the
corresponding empty vectors were used as the negative control.
Figure S7 Yeast Two-Hybrid Assay. Yeast two-hybrid assays
showing that EHD2, EHD3 and EHD4 did not interact with each
other. BD-OsMADS50 and AD-OsMADS56 were used as the
positive control  and the empty vector were used as the
negative control. ND, not determined.
Figure S8 qRT-PCR confirmation of RNA-seq results. Four
genes with reduced expression in ehd4 as determined by RNA-seq,
were chosen for qRT-PCR assay. Independent penultimate leaves
of 28 d-old plants grown under LDs were collected at dawn. The
rice Ubiquitin-1 (UBQ) gene was used as the internal control. Values
are shown as mean6s.d (standard deviations) of three independent
Figure S9 Alignment of the CCCH motif of EHD4 with the zinc
fingers from other CCCH-type proteins. Zinc fingers are from rice
OsLIC (Os06g49080) and OsDOS (Os01g09620), Arabidopsis
FES1 (At2g33835), HUA1 (NP_187874), PEI (S22126),
SOMNUS (At1g03790), SZF1 (At3g55980) and SZF2 (At2g40140),
Cotton GhZFP1 (AY887895), C. elegans MEX-1 (U81043), PIE-1
(AAB17868) and POS-1 (T37246), human TTP (P26651) and
yeast ZFS1(P47979). CCCH motifs from the same gene are shown
as serial numbers. The consensus CCCH residues are shaded with
yellow color. Other identical or similar residues are shaded with
blue or purple color, respectively.
Figure S10 Transcript levels of Ehd4 in WT (Kita-ake), ehd4 and
transgenic plants. Penultimate leaves were harvested at dawn from
35 d-old plants grown under natural long day conditions in
Beijing. The rice Ubiquitin-1 (UBQ) gene was used as the internal
control. Values are shown as mean6s.d (standard deviations) of
three independent experiments and two biological replicates.
Summary of the Ehd4 allele types in cultivated and
Primers used in this study.
We thank Dr. Masahiro Yano (NIAS, Japan) for NILs (Hd1 and Hd3a) and
mutants (ehd2 and ehd3), Dr. Song lim Kim and Gynheung An
(POSTECH, Korea) for mutants (osphyb, osmads50, and osmads51), Dr.
Qifa Zhang (HZAU, China) for NILs (Ghd7), and Dr. Roger I. Pennell
(Ceres Inc., USA) and Dr. Song Ge (IBCAS, China) for their constructive
suggestions and critical reading of the manuscript.
Conceived and designed the experiments: HG CW HW J-MW. Performed
the experiments: HG X-MZ GF JC MJ YR WW KZ PS FZ LJ JW XZ XG
J-LW ZC. Analyzed the data: HG X-MZ CW HW J-MW. Contributed
reagents/materials/analysis tools: CW J-MW. Wrote the paper: HG CW
1. Simpson GG , Dean C ( 2002 ) Arabidopsis, the Rosetta stone of flowering time? Science 296 : 285 - 289 .
2. Izawa T ( 2007 ) Adaptation of flowering-time by natural and artificial selection in Arabidopsis and rice . J Exp Bot 58 : 3091 - 3097 .
3. Srikanth A , Schmid M ( 2011 ) Regulation of flowering time: all roads lead to Rome . Cellular and Molecular Life Sciences 68 : 2013 - 2037 .
4. Kobayashi Y , Weigel D ( 2007 ) Move on up, it's time for change-mobile signals controlling photoperiod-dependent flowering . Genes Dev 21 : 2371 - 2384 .
5. Imaizumi T , Kay SA ( 2006 ) Photoperiodic control of flowering: not only by coincidence . Trends in plant science 11: 550 - 558 .
6. Turck F , Fornara F , Coupland G ( 2008 ) Regulation and identity of florigen: FLOWERING LOCUS T moves center stage . Annu Rev Plant Biol 59 : 573 - 594 .
7. Yanovsky MJ , Kay SA ( 2002 ) Molecular basis of seasonal time measurement in Arabidopsis . Nature 419 : 308 - 312 .
8. Valverde F , Mouradov A , Soppe W , Ravenscroft D , Samach A , et al. ( 2004 ) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering . Science Signalling 303 : 1003 .
9. Sawa M , Nusinow DA , Kay SA , Imaizumi T ( 2007 ) FKF1 and GIGANTEA complex formation is required for day-length measurement in Arabidopsis . Science Signalling 318 : 261 .
10. Fornara F , Panigrahi K , Gissot L , Sauerbrunn N , Ru hl M , et al. ( 2009 ) Arabidopsis DOF Transcription Factors Act Redundantly to Reduce CONSTANS Expression and Are Essential for a Photoperiodic Flowering Response . Developmental cell 17 : 75 - 86 .
11. Putterill J , Robson F , Lee K , Simon R , Coupland G ( 1995 ) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors . Cell 80 : 847 - 857 .
12. Tiwari SB , Shen Y , Chang HC , Hou Y , Harris A , et al. ( 2010 ) The flowering time regulator CONSTANS is recruited to the FLOWERING LOCUS T promoter via a unique cis-element . New Phytologist 187 : 57 - 66 .
13. Abe M , Kobayashi Y , Yamamoto S , Daimon Y , Yamaguchi A , et al. ( 2005 ) FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex . Science Signalling 309 : 1052 .
14. Wigge PA , Kim MC , Jaeger KE , Busch W , Schmid M , et al. ( 2005 ) Integration of spatial and temporal information during floral induction in Arabidopsis . Science Signalling 309 : 1056 .
15. Corbesier L , Vincent C , Jang S , Fornara F , Fan Q , et al. ( 2007 ) FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis . Science Signalling 316 : 1030 .
16. TAKIMOTO A , IKEDA K ( 1961 ) Effect of twilight on photoperiodic induction in some short day plants . Plant and Cell Physiology 2 : 213 - 229 .
17. Itoh H , Nonoue Y , Yano M , Izawa T ( 2010 ) A pair of floral regulators sets critical day length for Hd3a florigen expression in rice . Nature genetics 42 : 635 - 638 .
18. Hayama R , Yokoi S , Tamaki S , Yano M , Shimamoto K ( 2003 ) Adaptation of photoperiodic control pathways produces short-day flowering in rice . Nature 422 : 719 - 722 .
19. Yano M , Katayose Y , Ashikari M , Yamanouchi U , Monna L , et al. ( 2000 ) Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS . The Plant Cell Online 12 : 2473 - 2483 .
20. Kojima S , Takahashi Y , Kobayashi Y , Monna L , Sasaki T , et al. ( 2002 ) Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hd1 under short-day conditions . Plant and Cell Physiology 43 : 1096 - 1105 .
21. Tamaki S , Matsuo S , Wong HL , Yokoi S , Shimamoto K ( 2007 ) Hd3a protein is a mobile flowering signal in rice . Science Signalling 316 : 1033 .
22. Komiya R , Ikegami A , Tamaki S , Yokoi S , Shimamoto K ( 2008 ) Hd3a and RFT1 are essential for flowering in rice . Development 135 : 767 - 774 .
23. Takahashi Y , Teshima KM , Yokoi S , Innan H , Shimamoto K ( 2009 ) Variations in Hd1 proteins, Hd3a promoters, and Ehd1 expression levels contribute to diversity of flowering time in cultivated rice . Proceedings of the National Academy of Sciences 106 : 4555 - 4560 .
24. Doi K , Izawa T , Fuse T , Yamanouchi U , Kubo T , et al. ( 2004 ) Ehd1, a B-type response regulator in rice, confers short-day promotion of flowering and controls FT-like gene expression independently of Hd1 . Genes Dev 18 : 926 - 936 .
25. Komiya R , Yokoi S , Shimamoto K ( 2009 ) A gene network for long-day flowering activates RFT1 encoding a mobile flowering signal in rice . Development 136 : 3443 - 3450 .
26. Lee YS , Jeong DH , Lee DY , Yi J , Ryu CH , et al. ( 2010 ) OsCOL4 is a constitutive flowering repressor upstream of Ehd1 and downstream of OsphyB . The Plant Journal 63 : 18 - 30 .
27. Izawa T , Oikawa T , Sugiyama N , Tanisaka T , Yano M , et al. ( 2002 ) Phytochrome mediates the external light signal to repress FT orthologs in photoperiodic flowering of rice . Genes Dev 16 : 2006 - 2020 .
28. Xue W , Xing Y , Weng X , Zhao Y , Tang W , et al. ( 2008 ) Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice . Nature genetics 40 : 761 - 767 .
29. Wei X , Xu J , Guo H , Jiang L , Chen S , et al. ( 2010 ) DTH8 suppresses flowering in rice, influencing plant height and yield potential simultaneously . Plant Physiology 153 : 1747 - 1758 .
30. Kim SL , Lee S , Kim HJ , Nam HG , An G ( 2007 ) OsMADS51 is a short-day flowering promoter that functions upstream of Ehd1, OsMADS14, and Hd3a . Plant Physiology 145 : 1484 - 1494 .
31. Lee S , Kim J , Han JJ , Han MJ , An G ( 2004 ) Functional analyses of the flowering time gene OsMADS50, the putative SUPPRESSOR OF OVEREXPRESSION OF CO 1/AGAMOUS-LIKE 20 (SOC1/AGL20) ortholog in rice . The Plant Journal 38 : 754 - 764 .
32. Wu C , You C , Li C , Long T , Chen G , et al. ( 2008 ) RID1, encoding a Cys2/ His2-type zinc finger transcription factor, acts as a master switch from vegetative to floral development in rice . Proceedings of the National Academy of Sciences 105 : 12915 - 12920 .
33. Park SJ , Kim SL , Lee S , Je BI , Piao HL , et al. ( 2008 ) Rice Indeterminate 1 (OsId1) is necessary for the expression of Ehd1 (Early heading date 1) regardless of photoperiod . The Plant Journal 56 : 1018 - 1029 .
34. Matsubara K , Yamanouchi U , Wang ZX , Minobe Y , Izawa T , et al. ( 2008 ) Ehd2, a rice ortholog of the maize INDETERMINATE1 gene, promotes flowering by up-regulating Ehd1 . Plant Physiology 148 : 1425 - 1435 .
35. Colasanti J , Yuan Z , Sundaresan V ( 1998 ) The indeterminate Gene Encodes a Zinc Finger Protein and Regulates a Leaf-Generated Signal Required for the Transition to Flowering in Maize . Cell 93 : 593 - 603 .
36. Matsubara K , Yamanouchi U , Nonoue Y , Sugimoto K , Wang ZX , et al. ( 2011 ) Ehd3, encoding a plant homeodomain finger-containing protein, is a critical promoter of rice flowering . The Plant Journal 66 : 603 - 612 .
37. Takahashi Y , Shomura A , Sasaki T , Yano M ( 2001 ) Hd6, a rice quantitative trait locus involved in photoperiod sensitivity, encodes the a subunit of protein kinase CK2 . Proceedings of the National Academy of Sciences 98 : 7922 - 7927 .
38. Ogiso E , Takahashi Y , Sasaki T , Yano M , Izawa T ( 2010 ) The role of casein kinase II in flowering time regulation has diversified during evolution . Plant Physiology 152 : 808 - 820 .
39. Yu J , Hu S , Wang J , Wong GKS , Li S , et al. ( 2002 ) A draft sequence of the rice genome (Oryza sativa L. ssp. indica ). Science 296 : 79 - 92 .
40. Li J , Jia D , Chen X ( 2001 ) HUA1, a regulator of stamen and carpel identities in Arabidopsis, codes for a nuclear RNA binding protein . The Plant Cell Online 13 : 2269 - 2281 .
41. Kim DH , Yamaguchi S , Lim S , Oh E , Park J , et al. ( 2008 ) SOMNUS, a CCCHtype zinc finger protein in Arabidopsis, negatively regulates light-dependent seed germination downstream of PIL5 . The Plant Cell Online 20 : 1260 - 1277 .
42. Hurt JA , Obar RA , Zhai B , Farny NG , Gygi SP , et al. ( 2009 ) A conserved CCCH-type zinc finger protein regulates mRNA nuclear adenylation and export . The Journal of Cell Biology 185 : 265 - 277 .
43. Lin R , Ding L , Casola C , Ripoll DR , Feschotte C , et al. ( 2007 ) Transposasederived transcription factors regulate light signaling in Arabidopsis . Science Signalling 318 : 1302 .
44. Liu H , Yu X , Li K , Klejnot J , Yang H , et al. ( 2008 ) Photoexcited CRY2 interacts with CIB1 to regulate transcription and floral initiation in Arabidopsis . Science Signalling 322 : 1535 .
45. Peng LT , Shi ZY , Li L , Shen GZ , Zhang JL ( 2007 ) Ectopic expression of OsLFL1 in rice represses Ehd1 by binding on its promoter . Biochemical and biophysical research communications 360 : 251 - 256 .
46. Seay D , Hook B , Evans K , Wickens M ( 2006 ) A three-hybrid screen identifies mRNAs controlled by a regulatory protein . Rna 12 : 1594 - 1600 .
47. Chin L , Hahn WC , Getz G , Meyerson M ( 2011 ) Making sense of cancer genomic data . Genes Dev 25 : 534 - 555 .
48. Garris AJ , Tai TH , Coburn J , Kresovich S , McCOUCH S ( 2005 ) Genetic structure and diversity in Oryza sativa L. Genetics 169 : 1631 - 1638 .
49. Zhu Q , Zheng X , Luo J , Gaut BS , Ge S ( 2007 ) Multilocus analysis of nucleotide variation of Oryza sativa and its wild relatives: severe bottleneck during domestication of rice . Molecular Biology and Evolution 24 : 875 - 888 .
50. Ryu CH , Lee S , Cho LH , Kim SL , Lee YS , et al. ( 2009 ) OsMADS50 and OsMADS56 function antagonistically in regulating long day (LD)-dependent flowering in rice . Plant Cell and Environment 32 : 1412 - 1427 .
51. Berg JM , Shi Y ( 1996 ) The galvanization of biology: a growing appreciation for the roles of zinc . Science 271 : 1081 - 1085 .
52. Wang D , Guo Y , Wu C , Yang G , Li Y , et al. ( 2008 ) Genome-wide analysis of CCCH zinc finger family in Arabidopsis and rice . BMC Genomics 9 : 44 .
53. Vaughan D. Collection , conservation, and potential use of the wild relatives of rice in Asia and Australia; 1989 . IRRI.
54. Londo JP , Chiang YC , Hung KH , Chiang TY , Schaal BA ( 2006 ) Phylogeography of Asian wild rice, Oryza rufipogon, reveals multiple independent domestications of cultivated rice, Oryza sativa . Proc Natl Acad Sci U S A 103 : 9578 - 9583 .
55. Bart R , Chern M , Park CJ , Bartley L , Ronald PC ( 2006 ) A novel system for gene silencing using siRNAs in rice leaf and stem-derived protoplasts . Plant Methods 2 : 13 .
56. Huang XY , Chao DY , Gao JP , Zhu MZ , Shi M , et al. ( 2009 ) A previously unknown zinc finger protein, DST, regulates drought and salt tolerance in rice via stomatal aperture control . Genes Dev 23 : 1805 - 1817 .