Differential Regulation of FLOWERING LOCUS C Expression by Vernalization in Cabbage and Arabidopsis1
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Differential Regulation of FLOWERING LOCUS C
Expression by Vernalization in Cabbage and Arabidopsis1
Shu-I Lin, Jhy-Gong Wang, Suk-Yean Poon, Chun-lin Su2, Shyh-Shyan Wang, and Tzyy-Jen Chiou*
Institute of BioAgricultural Sciences, Academia Sinica, Taipei 115, Taiwan R.O.C. (S.-I.L., J.-G.W.,
S.-Y.P., C.-l.S., T.-J.C.); and Tainan District Agricultural Research and Extension Station,
Tainan 701, Taiwan R.O.C. (S.-S.W.)
Vernalization is required to induce flowering in cabbage (Brassica oleracea var Capitata L.). Since FLOWERING LOCUS C (FLC)
was identified as a major repressor of flowering in the vernalization pathway in Arabidopsis (Arabidopsis thaliana), two
homologs of AtFLC, BoFLC3-2 and BoFLC4-1, were isolated from cabbage to investigate the molecular mechanism of
vernalization in cabbage flowering. In addition to the sequence homology, the genomic organization of cabbage FLC is similar
to that of AtFLC, except that BoFLC has a relatively smaller intron 1 compared to that of AtFLC. A vernalization-mediated
decrease in FLC transcript level was correlated with an increase in FT transcript level in the apex of cabbage. This observation
is in agreement with the down-regulation of FT by FLC in Arabidopsis. Yet, unlike that in Arabidopsis, the accumulation of
cabbage FLC transcript decreased after cold treatment of leafy plants but not imbibed seeds, which is consistent with the
promotion of cabbage flowering by vernalizing adult plants rather than seeds. To further dissect the different regulation of FLC
expression between seed-vernalization-responsive species (e.g. Arabidopsis) and plant-vernalization-responsive species (e.g.
cabbage), the pBoFLC4-1::BoFLC4-1::GUS construct was introduced into Arabidopsis to examine its vernalization response.
Down-regulation of the BoFLC4-1::GUS construct by seed vernalization was unstable and incomplete; in addition, the
expression of BoFLC4-1::GUS was not suppressed by vernalization of transgenic rosette-stage Arabidopsis plants. We propose
a hypothesis to illustrate the distinct mechanism by which vernalization regulates the expression of FLC in cabbage and
Arabidopsis.
The transition from the vegetative to reproductive flowering occurs only after winter, in order for flowers
phase is essential for completion of the life cycle of and seeds to develop under favorable conditions.
flowering plants. This transition is particularly impor- During the past few years, two major loci, FRIGIDA
tant in agriculture in terms of productivity. The timing (FRI) and FLOWERING LOCUS C (FLC), have estab-
of the reproductive transition is determined by de- lished a vernalization requirement in Arabidopsis
velopmental status and environmental conditions. A (Arabidopsis thaliana; Burn et al., 1993; Lee et al., 1993;
combination of these two factors ensures that flower- Clarke and Dean, 1994; Koornneef et al., 1994; Johanson
ing occurs at appropriate times, with an accumulation et al., 2000). FRI acts upstream of FLC to positively
of sufficient nutrients and favorable environmental regulate FLC expression (Michaels and Amasino, 2000).
conditions (Levy and Dean, 1998; Reeves and Coup- In addition to FRI, other floral repressors, FRL1 and
land, 2000; Mouradov et al., 2002; Simpson and Dean, FRL2 (Michaels et al., 2004), VIP3 and VIP4 (Zhang and
2002). Genetic analysis has revealed that multiple van Nocker, 2002; Zhang et al., 2003), ESD4 (Reeves
pathways, such as the photoperiod and autonomous et al., 2002), HOS1 (Lee et al., 2001), ART1 (Poduska
pathways, are involved in regulating this transition et al., 2003), and PIE1 (Noh and Amasino, 2003), were
(Koornneef et al., 1998; Levy and Dean, 1998; Simpson subsequently identified as promoting the expression
et al., 1999). In addition, exposure to an extended of FLC. By contrast, several genes in the autonomous
period of low temperature is needed to promote pathway, such as LD, FCA, FY, FLK, FPA, FLD, and
flowering in many species (Chourad, 1960). This pro- FVE, have been shown to repress FLC expression
cess is known as vernalization (Napp-Zinn, 1987). (Michaels and Amasino, 1999; Sheldon et al., 1999,
Vernalization is a natural adaptation ensuring that 2000; Simpson, 2004).
FLC encodes a MADS-box transcription factor
that functions as a repressor of the floral transition
1
(Michaels and Amasino, 1999; Sheldon et al., 1999).
This work was supported by grants from Academia Sinica and
In different vernalization-responsive ecotypes and
the National Science Council of the Republic of China (grant nos.
NSC–90–2313–B–001–022 and NSC–91–2313–B–001–032 to T.-J.C.).
flowering-time mutants of Arabidopsis, the levels of
2
Present address: Vita Genomics, Inc., Taipei 248, Taiwan R.O.C. FLC mRNA and protein correlate well with flowering
* Corresponding author; e-mail tjchiou@gate.sinica.edu.tw; fax time in response to cold treatment (Michaels and
886–2–26515600. Amasino, 2000; Sheldon et al., 2000). Also, the duration
Article, publication date, and citation information can be found at of vernalization has been shown to be proportional to
www.plantphysiol.org/cgi/doi/10.1104/pp.104.058974. the degree of down-regulation of FLC (Sheldon et al.,
Plant Physiology, March 2005, Vol. 137, pp. 1037–1048, www.plantphysiol.org Ó 2005 American Society of Plant Biologists 1037
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
2000). These results suggest that FLC is a major de- responsive type, in which plants can sense low tem-
terminant of the vernalization response. Two genes, FT peratures during seed germination; the other is the
and SUPPRESSOR OF CO 1 (SOC1, or AGAMOUS- plant-vernalization-responsive type, in which plants
LIKE 20), functioning downstream of CONSTANS need to reach a certain developmental stage before they
(CO), a flowering activator in the photoperiod path- become sensitive to low temperatures. Biennial plants
way, were identified as being negatively regulated by that grow vegetatively in the first year and flower in the
FLC (Lee et al., 2000; Onouchi et al., 2000; Samach et al., following year after winter usually belong to the plant-
2000). Evidence supports the direct binding of the vernalization-responsive type. Some species in the
FLC protein to the CArG box of the SOC1 promoter, Brassicaceae family, such as Arabidopsis and B. rapa,
which results in the suppression of SOC1 expression are the seed-responsive type, but several varieties in
(Hepworth et al., 2002). This regulation by FLC is B. oleracea are plant responsive (Friend, 1985).
antagonistic to the positive regulation of CO in the Cabbage (B. oleracea var Capitata L.) is one of the most
expression of SOC1. important and popular vegetable crops in the Brassi-
The repression of AtFLC by vernalization is mitot- caceae family and is a plant-vernalization-responsive
ically stable, which means that the effect caused by type (Ito et al., 1966; Friend, 1985). Cabbage normally
cold temperature can be maintained even if treated requires 6 to 8 weeks of low temperature (5°C) to induce
plants are returned to a warm temperature, which flower initiation at the stage of 7 to 9 leaves or when the
suggests an epigenetic repression (Wellensiek, 1964; stem diameter reaches 6 mm (Ito et al., 1966; Friend,
Michaels and Amasino, 2000; Sheldon et al., 2000; 1985). It is of interest to understand the mechanisms
Henderson et al., 2003; Henderson and Dean, 2004). involved in flowering in seed- versus plant-vernaliza-
However, this effect is reset in the next generation after tion-responsive types. Hossain et al. tried to transfer the
meiosis (Michaels and Amasino, 2000; Sheldon et al., seed-vernalization character from Chinese cabbage
2000). Modification of chromatin structure, such as (B. rapa) into cabbage via ovule culture (Hossain et al.,
deacetylation and methylation of histone 3, recently 1990). The F2 progeny of this interspecies hybrid
was demonstrated to be involved in the epigenetic showed seed-vernalized characteristics. In our study,
repression of AtFLC during vernalization (Bastow we cloned the FLC homologs from cabbage and char-
et al., 2004; Sung and Amasino, 2004). VIN3 is in- acterized their expression in response to vernalization.
volved in the initiation of histone deacetylation in Moreover, the cabbage FLC gene was introduced into
AtFLC chromatin during prolonged cold treatment, Arabidopsis to investigate the different regulatory
and VIN3, VRN1, and VRN2 are required for histone mechanisms involved in the seed- and plant-vernali-
methylation and the creation of a stable inactive state zation responses. Our results support a distinct
(Henderson and Dean, 2004; Sung and Amasino, machinery in regulating FLC expression being respon-
2004). Interestingly, the DNA regions associated with sible for the different vernalization responses between
these changes in histone modification are located seed- and plant-responsive types.
within the first intron of AtFLC and its promoter in
Arabidopsis (He et al., 2003; Bastow et al., 2004; Sung
and Amasino, 2004), in agreement with the previous
observation that intron 1 is required for the mainte- RESULTS
nance of FLC repression (Sheldon et al., 2002). Molecular Cloning and Characterization of
In addition to Arabidopsis, other species in the Cabbage FLC Genes
family of Brassicaceae rely on vernalization to promote
flowering (Friend, 1985). Vernalization-responsive Cabbage, like Arabidopsis, is a vernalization-
flowering-time loci of Brassica species segregate as dependent plant in the family Brassicaceae. Since
two major quantitative trait loci that are colinear with AtFLC was identified as a major flowering repressor
the regions of FRI and FLC in the Arabidopsis genome in the vernalization pathway in Arabidopsis (Michaels
(Osborn et al., 1997). This observation indicates that and Amasino, 1999; Sheldon et al., 1999), an FLC
homologs of FRI and FLC genes are likely to be homolog was hypothesized to exist in cabbage and
important in the control of flowering time through function similarly. Primers were designed according to
vernalization in other Brassica species. In fact, several the C-terminal coding sequence of AtFLC without the
FLC homologs have been isolated from Brassica spe- MADS-box domain to amplify the cabbage FLC cDNA.
cies, such as Brassica napus (Tadege et al., 2001) and An inbred cabbage line, TNSS42-12, was chosen for
doubled haploid lines of Brassica rapa and Brassica RNA isolation because of its high degree of vernaliza-
oleracea (Schranz et al., 2002). Moreover, genetic ma- tion requirement (S.-S. Wang, unpublished data). A
nipulation of FLC expression has been proven to partial cDNA of one FLC homolog was cloned from
modify flowering time in both Arabidopsis (Michaels cabbage via reverse transcription-PCR. A full-length
and Amasino, 1999; Sheldon et al., 1999) and B. napus cDNA clone was subsequently isolated after screening
(Tadege et al., 2001). a cabbage cDNA library constructed from young
Vernalization can be classified into two types leaves with a probe generated from the partial cDNA
according to the age of the plant that senses low tem- sequence. This clone was designated as BoFLC4-1
perature (Friend, 1985). One is the seed-vernalization- (GenBank accession no. AY306122). With use of
1038 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Cabbage FLOWERING LOCUS C
genomic walking to isolate the BoFLC4-1 promoter, winter season in the lowlands of Taiwan, where the
another FLC homolog was found. This second FLC average winter temperature is 17°C to 20°C from
gene was named BoFLC3-2 (GenBank accession no. December to February (http://www.cwb.gov.tw/V4;
AY306123). The nomenclatures of BoFLC4-1 and S.-S. Wang, unpublished data). BoFLC transcripts were
BoFLC3-2 were based on their deduced amino acid analyzed in different tissues (apex, young leaf, mature
sequence similarity to BnFLCs from B. napus (Tadege leaf, stem, and root) before and after cold treatment
et al., 2001). BoFLC4-1 and BoFLC3-2 share 88% and with use of a BoFLC4-1 cDNA probe lacking the MADS
83% identity in nucleotide and deduced amino acid box (Fig. 2). In all three varieties, FLC transcripts were
sequences, respectively. The deduced amino acid se- most abundant in the apex before cold treatment. The
quences of BoFLC4-1 and BoFLC3-2 show 79% to 81% FLC transcript level was decreased mainly in the apex,
identity with AtFLC and up to 98% identity with the young and mature leaves, and to a lesser extent in the
BnFLCs from B. napus. In addition, sequence compar- stem and root during cold treatment. This observation
ison with the partial sequences of BrFLCs and BoFLCs is consistent with the decrease in AtFLC level in all
from B. rapa and B. oleracea, respectively (Schranz et al., tissues after vernalization in Arabidopsis (Sheldon
2002), showed high similarity at the amino acid level. et al., 2000). Nevertheless, the expression of FLC in
The phylogenetic relationship among them is illus- the root and stem was not as strong as that in the apex
trated in Figure 1. BoFLC4-1 is most similar to BnFLC4, and leaf. Low expression levels of BoFLC transcripts in
whereas BoFLC3-2 has a closer relationship to BoFLC3, the roots of Yehsen and YSL-0 and a lesser response in
BrFLC3, and BnFLC3. The genomic sequences of both the roots of TNSS42-12 indicated that roots may not be
genes were subsequently obtained by PCR and geno- involved in the vernalization process. This is in agree-
mic walking (GenBank accession nos. AY306124 for ment with the previous observation in pie1 mutants, in
BoFLC4-1 and AY306125 for BoFLC3-2). The genomic which root expression of FLC did not affect flowering
organization of BoFLC4-1 and BoFLC3-2 is similar to (Noh and Amasino, 2003).
that of AtFLC and the partial genomic sequences of The initial transcript levels of FLC in the apex,
BrFLCs and BoFLCs (Schranz et al., 2002). They all young leaf, and mature leaf before cold treatment are
consist of seven exons in similar sizes. However, similar in these three varieties (Fig. 2D); however, the
BoFLC4-1 and BoFLC3-2 have a relatively small intron down-regulation of FLC in YSL-0 was more rapid than
1 (approximately 1.1 kb) compared to AtFLC (approx- in TNSS42-12 and Yehsen after cold treatment. The
imately 3.5 kb; see Fig. 5 for details). FLC transcript level was significantly reduced in YSL-0
after 2-week treatment and barely detectable after
4-week treatment but remained substantial in the other
Repression of BoFLC Expression by Vernalization 2 varieties during the same period of cold treatment
(Fig. 2, A–C), which suggests that the repression of
To determine whether cabbage FLCs are involved in FLC in YSL-0 is more sensitive to cold temperature
controlling flowering via the vernalization process, compared to the other 2 varieties. This observation
RNA gel-blot analysis was carried out to investigate may explain the shorter vernalization requirement for
the FLC expression pattern in response to vernaliza- YSL-0 to flower.
tion. Eight-week-old cabbage plants from 3 different Because in Arabidopsis FT is down-regulated by
varieties, TNSS42-12, Yehsen, and YSL-0, were vernal- FLC (Samach et al., 2000), we examined the expression
ized at 4°C for 2 to 6 weeks. Vernalization require- of the FT homolog in parallel with FLC expression
ments for these three varieties are different. TNSS42-12 in cabbage. A distinct band that cross-hybridized with
requires at least 45 d at 5°C and Yehsen needs about the Arabidopsis FT probe was detected in cabbage
30 d at 5°C, whereas YSL-0 can flower after the local tissues, and we refer to this putative FT homolog as
Figure 1. Neighbor-joining tree showing phylo-
genetic relationships between cabbage FLC
(BoFLC3-2 and BoFLC4-1) and other FLC pro-
teins. Partial amino acid sequences without exon
1 were analyzed by use of the Phylip program.
Bootstrap values are indicated on the branches.
Bo, B. oleracea; Br, B. rapa; Bn, B. napus; At,
Arabidopsis. Accession numbers are as follows
(in parentheses): AtFLC (AAD21249.1), BnFLC1
(AAK70215), BnFLC2 (AAK70216), BnFLC3
(AAK70217), BnFLC4 (AAK70218), BnFLC5
(AAK70219), BrFLC1 (AAO13159), BrFLC2
(AAO86066 and AAO86067), BrFLC3
(AAO13158), BrFLC5 (AAO13157), BoFLC1
(AAN87902), BoFLC3 (AAN87901), and BoFLC5
(AAN87900).
Plant Physiol. Vol. 137, 2005 1039
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
Figure 2. RNA gel-blot analysis of
BoFLC4-1 in response to vernalization.
Three varieties (A, TNSS42-12; B, Yehsen;
C, YSL-0) of 8-week-old cabbage plants
were vernalized at 4°C from 2 to 6 weeks.
The BoFLC4-1 transcript was examined in
different tissues (apex, young leaf, mature
leaf, stem, and root) before and after
indicated weeks of cold treatment. The
BoFLC4-1 transcript levels before vernal-
ization were compared in these three
varieties (D). The BoFT transcript was
also examined in the TNSS42-12 variety.
The methylene blue staining of 25S rRNA
in each blot is shown as a loading control.
BoFT (Fig. 2A). The expression of BoFT in the apex pressed in the shoot of 1-week-old seedlings, whereas
showed an inverse expression pattern compared with FLC expression was barely detectable at the same stage.
that of FLC (Fig. 2A). This observation is in agreement However, in older plants, the expression of FT in the
with the down-regulation of FT by FLC in Arabidopsis apex gradually decreased to an almost undetectable
(Samach et al., 2000). In contrast with the apex, ex- level at the 8-week-old stage, whereas the expression of
pression of BoFT in young and mature leaves, stems, FLC increased. This result again supports the down-
and roots did not increase during vernalization, even regulation of FT by FLC (Samach et al., 2000).
though the expression of BoFLC decreased, which Previous studies had shown that FLC expression was
suggests that other regulators participate in regulating down-regulated and flowering promoted after vernal-
the expression of BoFT in these tissues. izing Arabidopsis seeds (Michaels and Amasino, 1999;
The expression of BoFLC4-1 in TNSS42-12 was mon- Sheldon et al., 2000). Unlike Arabidopsis, in cabbage
itored in different developmental stages by RNA gel- seed vernalization is not able to induce flowering.
blot analysis (Fig. 3A). Since the apex was shown Vernalization is effective only when cold treatment
previously to have the highest expression of BoFLC4-1, is applied to mature cabbage plants (Ito et al., 1966;
apex tissue was harvested from 1- to 8-week-old Friend, 1985). As shown in Figure 2, a decrease in
cabbage plants for further examination. The entire expression of BoFLC was observed in vernalized
aboveground tissue, including apex and cotyledon, 8-week-old cabbage plants. However, BoFLC does not
was collected and indicated as ‘‘shoot’’ for the first respond to seed vernalization: cold treatment of cab-
week of sampling. Cotyledons were collected sepa- bage seeds (4°C for 4 weeks) did not alter the expression
rately from the apex of 2-week-old seedlings but not at pattern of FLC or FT (compare Fig. 3, A and B). This
the 4- and 8-week-old stages since they were senescent. observation is in agreement with seed vernalization
The expression of BoFLC4-1 was very weak in the shoot not being effective in promoting flowering in cabbage,
at the 1-week-old stage but significantly increased at which is in contrast with the down-regulation of
the 2-week-old stage and had a dramatic boost at 4 to AtFLC in Arabidopsis when seeds are vernalized.
8 weeks (Fig. 3A). Thus, FLC expression strongly To determine whether the change of signal intensity
increases with age. The FT homolog was highly ex- shown in these RNA gel blots was attributable to
1040 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Cabbage FLOWERING LOCUS C
a C-terminal truncated BoFLC4-1 construct (Dominant
Negative 2, DN2), lacking the C domain, were driven
by the cauliflower mosaic virus 35S promoter and
introduced into a vernalization-responsive Arabidop-
sis ecotype, C24 (Fig. 4A). Transgenic T2 plants over-
expressing either the DN1 or the DN2 construct
exhibited early flowering phenotypes without vernal-
ization (Fig. 4B). These plants flowered much earlier
(about 5–6 leaves in the DN1 construct and 11–12 leaves
in the DN2 construct) than the untransformed wild
type (45–55 leaves) without cold treatment. The early
flowering phenotype in both transgenic plants suggests
that overexpressing these two truncated BoFLC4-1
proteins may interfere with the normal function of the
Figure 3. RNA gel-blot analysis of BoFLC4-1 and BoFT in response to endogenous Arabidopsis FLC proteins.
different developmental stages and seed vernalization. Cabbage seeds
were subjected without (A) or with (B) seed vernalization treatment.
Seed vernalization was carried out at 4°C for 4 weeks. Various tissues
were harvested at different stages from 1- to 8-week-old cabbage
Comparison of Promoter and Intron 1 Sequences of
plants. The methylene blue staining of 25S rRNA in each blot is shown BoFLCs and AtFLC
as a loading control.
To further study the regulation of BoFLC gene
expression in response to vernalization, the promo-
ters of both BoFLC4-1 and BoFLC3-2 were isolated
BoFLC4-1 or BoFLC3-2, a 3#-untranslated region (UTR) by genome walking. Genomic fragments 2.0 kb and
specific sequence of BoFLC3-2 was used as a probe for 1.5 kb upstream from the translational start sites of
hybridization. No BoFLC3-2 specific signal was de- BoFLC4-1 and BoFLC3-2, respectively, were cloned.
tected in all tissues examined (data not shown). This re- Comparing promoter sequences to that of AtFLC
sult indicated that the expression level of BoFLC3-2 identified three conserved regions (Fig. 5A). The first
was very low, and signals observed in the RNA gel- conserved sequence (P1) includes the 5# UTR and
blot analysis were mainly derived from BoFLC4-1, al- a sequence up to 2200 bp upstream from the trans-
though we cannot exclude cross-hybridization to other lational start site. The second conserved sequence (P2),
unidentified BoFLCs. DNA gel-blot analysis revealed consisting of 51 bp, was found at 2219 to 2269 bp of
that both BoFLC4-1 and BoFLC3-2 exist as single-copy AtFLC. The third conserved sequence (P3) was located
genes in cabbage (data not shown). Nevertheless, more at about 22.4 kb and 22.0 kb of AtFLC and BoFLC4-1,
cabbage FLC genes are likely to exist based on the pat-
tern of multiple bands from low stringency hybridiza-
tion. Previously, Schranz et al. cloned three FLC genes
(BoFLC1, BoFLC3, and BoFLC5) from a doubled haploid
line of B. oleracea (Schranz et al., 2002; Fig. 1), which
supports the existence of at least one additional mem-
ber of the FLC gene family in cabbage.
Alteration of Flowering Time via Heterologous
Expression of Truncated BoFLC4-1 in Arabidopsis
To further understand the biological function of
cabbage FLC in regulating flowering, heterologous
expression of BoFLC4-1 was conducted in Arabidopsis
and flowering time was monitored. We took a dominant-
negative approach to inactivate the function of the en-
dogenous AtFLC gene. Since the FLC protein is a MADS
transcription factor and MADS proteins usually func-
tion as a multimer (Pellegrini et al., 1995; Riechmann
and Meyerowitz, 1997; Favaro et al., 2003), over-
Figure 4. Alteration of flowering time via heterologous expression of
expression of the truncated nonfunctional BoFLC4-1
truncated BoFLC4-1 in Arabidopsis. A, A schematic illustration of the
protein may interfere with the multimerization of the protein domain of the FLC protein and two dominant-negative con-
endogenous AtFLC proteins by competition and con- structs. The DN1 construct was truncated in the MADS-box domain,
sequently result in the alteration of flowering time. An whereas the DN2 construct lacked the C domain. B, Flowering-time
N-terminal truncated BoFLC4-1 construct (Dominant analysis of transgenic T2 Arabidopsis lines carrying the DN1 or DN2
Negative 1, DN1), lacking the MADS-box domain, and construct. Error bar indicates the SE.
Plant Physiol. Vol. 137, 2005 1041
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
Figure 5. Sequence comparison of AtFLC, BoFLC4-1, and BoFLC3-2 sequences. A, Three conserved segments (P1, P2, and P3)
were identified among the 5# upstream sequences of three FLC genes. The organization and size (bp) of exon and intron (shown
as a solid triangle) are shown. The arrow indicates the translational start site. The arrow bar shown in the upstream sequence of
BoFLC3-2 indicates an additional open reading frame encoding a 3-keto-acyl-ACP dehydratase. B, Comparison of intron 1
sequence between BoFLC4-1 and AtFLC revealed several conserved regions: I1, I2, I3, I4, I5, and I6. The arrow bar in the I3
region represents a direct repeat (see text for detail). Lines a and b in A and lines c and d in B indicate the regions previously
identified to be associated with chromatin modification in the epigenetic repression of FLC by vernalization in Arabidopsis.
respectively, but at about 20.4 kb upstream of BoFLC3-2 Vernalization Responses of BoFLC4-1 in
from the translational start site. A partial open read- Transgenic Arabidopsis
ing frame encoding a 3-keto-acyl-ACP dehydratase
was identified in this 1.5-kb upstream sequence of the Since the promoter and intragenic sequences were
BoFLC3-2 gene (Fig. 5A), which indicates that the pro- reported to be crucial for the regulation of AtFLC in
moter of BoFLC3-2 could be truncated and this may ac- response to vernalization (Sheldon et al., 2002), the
count for the lack of detectable BoFLC3-2 mRNA. complete genomic sequence of BoFLC4-1, including the
As mentioned earlier, the introns 1 of BoFLC4-1 and 2-kb upstream promoter and complete sequences of 7
BoFLC 3-2 (approximately 1.1 kb) are relatively smaller exons and 6 introns, was fused to the b-glucuronidase
than that of AtFLC (approximately 3.5 kb), and the (GUS) reporter gene (pBoFLC4-1::BoFLC4-1::GUS) and
intron 1 sequence of AtFLC was identified to be transformed into Arabidopsis to examine its expression
important in the epigenetic repression of AtFLC pattern and vernalization responses in Arabidopsis.
through chromatin modification (Sheldon et al., 2002; GUS staining was observed in the apex, young leaves,
He et al., 2003; Bastow et al., 2004; Sung and Amasino, and roots of transgenic Arabidopsis seedlings (Fig. 6A).
2004). Thus, the intron 1 sequences of cabbage and Interestingly, pollen grains and germinating pollen
Arabidopsis were compared. The sequence compari- showed strong GUS staining (Fig. 6, B–D).
son revealed that several segments in the AtFLC intron The transgenic T2 lines were examined for the re-
1 could be aligned with the sequences in the BoFLC4-1 sponse to seed or plant vernalization. Seeds or 10-d-old
intron 1 (I1–I6 in Fig. 5B). An approximately 50-bp rosette plants were treated at 4°C for 2, 4, 6, and 8 weeks.
sequence, indicated by an arrow in the conserved Vernalized seedlings were returned to the normal
segment of I3 (Fig. 5B), was found to be organized as growth temperature for another 10 d before harvest.
an imperfect direct repeat in the intron 1 of BoFLC4-1. Vernalized rosette plants were harvested immediately
Intriguingly, a continuous 2.6-kb segment of the AtFLC after treatment. Total RNAs from seedlings without
intron 1 could not identify any homologous sequence vernalization or with seed-vernalization (Fig. 7A) or
in the BoFLC4-1 intron 1 (Fig. 5B). A similar result was plant-vernalization treatment (Fig. 7B) were isolated,
obtained after comparing the intron 1 sequences of and the transcript levels of the endogenous AtFLC and
BoFLC3-2 and AtFLC, except there was no 50-bp repeat BoFLC4-1::GUS transgenes were inspected.
in the BoFLC3-2 intron 1. Thus, size differences be- Upon seed vernalization, the expression of endog-
tween introns 1 of AtFLC and BoFLC is mainly due to enous AtFLC deceased significantly and was not
an additional 2.6-kb segment close to the 3# end of the detectable after a 4-week cold treatment (Fig. 7A).
AtFLC intron 1. Endogenous AtFLC in plant-vernalized seedlings were
1042 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Cabbage FLOWERING LOCUS C
lines exhibited reduced BoFLC::GUS expression after
plant-vernalization treatment, even though the level of
endogenous AtFLC transcript was gradually reduced
during the treatment (Fig. 7B). The instability of
repression by seed vernalization and lack of repres-
sion of BoFLC::GUS after plant vernalization in trans-
genic Arabidopsis demonstrate that the repression of
BoFLC4-1 by vernalization cannot be properly regu-
lated in Arabidopsis. This result suggests that the
regulatory machinery controlling FLC expression by
vernalization in cabbage is different from that in
Arabidopsis.
DISCUSSION
In this study, two FLC homologs, BoFLC4-1 and
BoFLC3-2, were isolated from cabbage (Fig. 1). The
expression of BoFLC4-1 and its regulation by vernal-
ization were analyzed. The FLC transcripts were high-
est in the apex, as was observed in Arabidopsis
(Michaels and Amasino, 1999). Steady-state levels of
the BoFLC transcript decreased during vernalization of
cabbage plants, mainly in the apex and leaf (Fig. 2, A–
C). It is of interest to note that the three cabbage
varieties (TNSS42-12, Yehsen, and YSL-0) with differ-
ent vernalization requirements showed similar tran-
Figure 6. GUS staining in transgenic Arabidopsis harboring the script levels of FLC before vernalization (Fig. 2D) but
pBoFLC4-1::BoFLC4-1::GUS construct. Staining was observed in the a different degree of reduction of FLC expression after
apex, young leaf, and root of the seedling (A). In addition, strong GUS vernalization (Fig. 2, A–C). Suppression of FLC in YSL-0,
staining was observed in pollen grains (B and C) and germinating a short-vernalization requirement variety, occurred
pollens and pollen tubes (D). Scale bar 5 1 mm in A and B and 0.1 mm
much more rapidly than that of the other two varieties,
in C and D.
which require longer exposure to cold.
FT and SOC1 are recognized as key integrators of
also down-regulated during cold treatment. Thus, multiple floral pathways in Arabidopsis (Simpson and
both seed- and plant-vernalization treatments were Dean, 2002). FT and SOC1 expression is repressed by
effective, although the repression during plant vernal- FLC (Lee et al., 2000; Onouchi et al., 2000; Samach et al.,
ization was not as complete as that during seed 2000). When Arabidopsis FT and SOC1 cDNAs were
vernalization (Fig. 7B). However, the down-regulation used as probes to search for homologous genes in
of the BoFLC4-1::GUS transcript by seed vernalization cabbage, the SOC1 homolog was not detected in
showed heterogeneity among the 11 transgenic T2 cabbage tissues tested, but an FT homolog was detec-
lines examined. Three groups of T2 lines were classi- ted. The reduction of BoFLC expression by vernaliza-
fied, and one independent line from each group was tion corresponded to the increased level of BoFT in the
selected for representation in Figure 7. In the 4 in- apex (Fig. 2A). Moreover, BoFT also showed an ex-
dependent lines in group 1, the transcript levels of pression pattern opposite to that of FLC in the apex
BoFLC4-1::GUS did not change significantly during from different developmental stages of nonvernalized
seed vernalization (line 1 in Fig. 7A). In the 3 indepen- cabbage (Fig. 3). This observation implies that Arabi-
dent lines of group 2, transcripts of BoFLC4-1::GUS dopsis and cabbage may regulate flowering similarly
were undetectable after 4- and 6-week treatments in terms of the correlation between FT and FLC.
but reappeared after 8-week treatment of seed ver- Nevertheless, BoFT was strongly expressed in both
nalization (line 2 in Fig. 7A). In the remaining 4 inde- nonvernalized young and mature leaves, and the
pendent lines in group 3, transcripts of BoFLC4-1:: expression of BoFT was not up-regulated in leaves
GUS were undetectable at the 4- or 6-week time after vernalization, even though the expression of FLC
points, but the expression of BoFLC4-1::GUS was re- was already down-regulated (Fig. 2A). This observa-
stored after that (line 3 in Fig. 7A). The heterogeneity tion suggests that the expression of BoFT in the leaf is
among several transgenic lines and fluctuation of the not sufficient to promote flowering, and the increase in
BoFLC4-1::GUS transcript during vernalization indi- BoFT expression in the apex during vernalization is
cate that the suppression of BoFLC4-1::GUS was in- a prerequisite for the induction of flowering in cab-
complete and unstable when transgenic Arabidopsis bage. This observation is in conflict with the recent
seeds were vernalized. By contrast, none of transgenic study that misexpression of FT in a wide range of
Plant Physiol. Vol. 137, 2005 1043
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
Figure 7. Analysis of the vernalization response
of BoFLC4-1 expression in transgenic Arabidop-
sis. Transgenic T2 lines were vernalized at the
seed stage (A) or rosette stage (B) at 4°C for 2 to
8 weeks. The control without vernalization is
indicated as 0 weeks. The transcript level of
BoFLC4-1::GUS transgene genes and endoge-
nous AtFLC in three representative transgenic
lines (line 1–3) was revealed by RNA gel-blot
analysis. The change in level of transcript in
response to vernalization is indicated below the
image. The methylene blue staining of 25S rRNA
in each blot is shown as a loading control.
tissues activates flowering in Arabidopsis (An et al., alternative approach to regulating flowering time in
2004). Decreased FLC expression accompanies in- vernalization-dependent plant species.
creased expression of FT in the apex, which suggests Comparison of promoter sequences upstream of
that the apex is the main location mediating flowering BoFLC4-1, BoFLC3-2, and AtFLC revealed three con-
by vernalization in cabbage. This suggestion is in agree- served regions (P1, P2, and P3; Fig. 5A). The alignment
ment with early observations that the apex is sen- of these conserved regions suggests that a large se-
sitive to low temperatures and must be exposed to quence deletion between the P2 and P3 regions oc-
cold for vernalization to occur (Chourad, 1960; Ito curred in the promoter of BoFLC3-2. The truncated
et al., 1966). promoter of BoFLC3-2 may explain its low level of
MADS-box proteins are thought to form dimers transcripts in cabbage. Positive and negative regulatory
through the interaction of the K domain at the center elements between the P2 and P3 regions were identified
and bind to DNA through the MADS-box domain in the promoter of AtFLC (Sheldon et al., 2002). Whether
at the N terminus (Schwarz-Sommer et al., 1992; there are any cis-acting elements located between the
Mizukami et al., 1996; Riechmann et al., 1996). Recent P2 and P3 regions that are responsible for the expres-
evidence indicates that certain MADS-box proteins sion of BoFLC3-2 remains to be studied. The P2 con-
may associate as tetramers (Favaro et al., 2003). ag and served region (51 bp) is located within the 75-bp
ap3 mutant phenotypes were generated by transform- promoter sequence of AtFLC that was reported to be
ing the wild-type Arabidopsis with mutant forms of essential for AtFLC expression in nonvernalized plants
AG or AP3 lacking the C- and N-terminal regions, (Sheldon et al., 2002). About 500 bp upstream to the P3
respectively (Mizukami et al., 1996; Krizek et al., 1999). region of BoFLC3-2, a gene encoding 3-keto-acyl-ACP
Moreover, a dominant-negative mutation was success- dehydratase was identified (Fig. 5A). Coincidently, this
fully created using heterologous AP3 genes from two gene can be aligned with a similar gene (At5g10160.1)
plant species (Tzeng and Yang, 2001). Here, we dem- that is one locus away from the AtFLC (At5g10140.1) in
onstrate another example of generating a dominant- the Arabidopsis genome. This suggests that this geno-
negative mutation by use of genes from different plant mic organization had been preserved between cabbage
species. Both DN1 (lack of MADS box) and DN2 (lack and Arabidopsis during evolution.
of C domain) constructs (Fig. 4) retain the K domain pre- Cabbage flowering can be promoted by vernalizing
sumably involved in protein-protein interaction. Trans- adult plants but not seeds (Friend, 1985). Cabbage
genic Arabidopsis overexpressing either N-terminal must pass through the juvenile phase to be able to
truncated (DN1) or C-terminal truncated (DN2) sense cold and induce flowering. We showed that
BoFLC4-1 flowered earlier than wild-type plants (Fig. vernalization of 8-week-old cabbage plants was effec-
4). The early flowering phenotype could result from tive in repressing the expression of BoFLC4-1 (Fig. 2);
the formation of a nonfunctional multimer between however, cold treatment of cabbage seeds did not alter
the truncated BoFLC4-1 and endogenous AtFLC pro- the expression level of BoFLC4-1 (Fig. 3). The regula-
teins. Overexpression of the truncated FLC protein tion of BoFLC4-1 expression by vernalization of adult
through a dominant-negative effect may provide an cabbage plants but not seeds is correlated with the
1044 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Cabbage FLOWERING LOCUS C
effectiveness of vernalization. This situation is different When transgenic Arabidopsis seeds were subjected
from that in Arabidopsis in which FLC is down- to vernalization, the BoFLC4-1::GUS transcript did not
regulated and flowering is promoted by seed vernal- show significant changes in some of the transgenic
ization, which suggests that the regulatory mechanism lines (e.g. line 1 in Fig. 7A), but its transcript was
of FLC expression in plant-vernalization-responsive undetectable at one or two time points of the treatment
species is different from that in seed-vernalization- in other lines (e.g. lines 2 and 3 in Fig. 7A). However,
responsive species. To further characterize the differ- the down-regulation of the BoFLC4-1::GUS transcript
ences in vernalization responses between the 2 species was not sustained after prolonged cold treatment,
types, the whole genomic sequence of BoFLC4-1 con- which indicated that down-regulation was not stable.
taining the 2-kb promoter and 7 exons was fused to the This observation is inconsistent with the quantitative
GUS reporter gene (pBoFLC4-1::BoFLC4-1::GUS) to relationship between the duration of cold treatment
examine its localization and vernalization response and the extent of down-regulation of AtFLC, as ob-
in Arabidopsis. It is of interest to note that GUS served in Figure 7 and in a previous report (Sheldon
staining was strongly detected in pollen grains and et al., 2000). Because a 2-kb promoter region and
germinating pollen of the transgenic plants (Fig. 6, C complete exon and intron sequences were included
and D). Indeed, the BoFLC4-1 transcript also was in the transgene construct, it is not likely that critical
detected in cabbage flowers, and transgenic cabbage sequences for vernalization responses were left out of
overexpressing a double-stranded RNAi construct of this construct. Nevertheless, we cannot rule out that
BoFLC4-1 produced very few seeds (J.-G. Wang, S.-Y. the BoFLC4-1::GUS construct would not be properly
Poon, S.-S. Wang, and T.-J. Chiou, unpublished data). regulated in cabbage.
Moreover, transgenic Arabidopsis expressing an The mechanism involved in inactivation of AtFLC
AtFLC antisense construct showed reduced fertility during vernalization recently was demonstrated to be
(Sheldon et al., 1999), and overexpressed AtFLC was associated with chromatin modification, such as his-
defective in floral morphology and produced less than tone deacetylation and methylation (Bastow et al., 2004;
normal quantities of pollen or no pollen (Hepworth Sung and Amasino, 2004). Because chromatin immu-
et al., 2002). These results indicate that FLC may noprecipitation with a specific antibody recognized the
participate in pollen development or pollination in acetylated or methylated histone H3, several regions in
addition to controlling flowering time. However, the the AtFLC gene associated with the modified histone
observation of normal seed setting in flc null mutants H3 were identified. Those sequences were highlighted
(Michaels and Amasino, 2001) disagrees this hypoth- and designated as the four regions, a, b, c, and d, in
esis. Whether the difference between transgenic ma- Figure 5, A and B. Regions a and b are located in the
nipulation and mutation in endogenous gene gives promoter, while regions c and d are located in intron 1.
rise to this contradiction remains to be resolved. The sequence of region a was previously identified to
Figure 8. A working hypothesis of the vernalization responses in cabbage (A) and Arabidopsis (B). The solid-bold and dashed
lines indicate vernalization of plants and seeds, respectively. The triangle indicates the increase or decrease of FLC or FT
expression after vernalization. See text for description.
Plant Physiol. Vol. 137, 2005 1045
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
consist of a negative regulatory element of AtFLC of FT shows an opposite pattern to that of FLC, and
(Sheldon et al., 2002). Region b is across the conserved increased FT expression in the apex after plant vernal-
segments of P1 and P2. Region c overlaps with the ization may be critical to induce cabbage flowering (Fig.
conserved segments I1 and I2 in intron 1, and region d is 2A). However, seed vernalization is more effective than
located within the 2.6-kb region that is missing in the plant vernalization in down-regulating FLC expression
BoFLC intron 1. In BoFLC4-1, similar b and c sequences in Arabidopsis (Fig. 7). This observation is consistent
were identified, but no homologous sequences of a and with a previous report that cold treatment at the rosette
d were found (Fig. 5). We suspect that lack of counter- stage is less effective in promoting flowering than
part sequences to these regions a and d in BoFLC4-1 seed cold treatment in Arabidopsis (Nordborg and
may cause the unstable repression of BoFLC4-1::GUS Bergelson, 1999). We hypothesized that fewer vernaliza-
after seed vernalization in transgenic Arabidopsis; tion activators or the presence of vernalization repres-
however, we do not exclude the possibility that other sors in the rosette stage may reduce the competence of
unidentified cis-regulatory elements in AtFLC, which the vernalization response and result in later flowering
may not be present in BoFLC4-1, are also important for (Fig. 8B). Nevertheless, vernalized Arabidopsis rosette
responding to seed vernalization in Arabidopsis. plants show accelerated flowering compared to non-
The expression of BoFLC4-1::GUS was not inactiva- vernalized plants.
ted when transgenic rosette plants were subjected to In summary, we show that cabbage BoFLC4-1 plays
vernalization treatment. We postulate that certain a similar role to Arabidopsis FLC in controlling flower-
cabbage-specific trans-regulatory factors required for ing. However, AtFLC responds to seed vernalization
the suppression of BoFLC4-1 during cabbage vernali- but BoFLC4-1 does not. Comparison of promoter and
zation do not exist in Arabidopsis. The different re- intragenic sequences between BoFLC4-1 and AtFLC
sponses in terms of seed or plant vernalization are revealed differences that may be specifically involved
likely due to the distinct interactions between specific in either the seed or plant vernalization response.
cis- and trans-regulatory elements in these two differ- Further characterization of these sequences may pro-
ent vernalization-responsive types. Nevertheless, it is vide an opportunity to dissect different regulatory
also possible that the homologs of upstream regulators mechanisms of the vernalization response in seed-
of AtFLC such as VIN3 (Sung and Amasino, 2004) may versus plant-responsive species.
be involved in the differential responses of seed versus
plant vernalization in cabbage. For example, a cabbage
VIN3 homolog may only be induced during vernali-
MATERIALS AND METHODS
zation of mature cabbage plants but not in vernalized
seedlings to down-regulate the expression of FLC. Plant Material and Treatment
However, the interaction between FLC and its up-
Cabbage (Brassica oleracea var Capitata L.) was grown in growth chambers
stream regulators in Arabidopsis and cabbage may not with a 16-h photoperiod (with cool fluorescence light at 150–200 mE m22 s21)
be identical since the expression of BoFLC4-1 in trans- and a 24°C/20°C light/dark temperature cycle. Three cabbage varieties,
genic Arabidopsis was not suppressed when endoge- TNSS42-12 and YSL-0, two inbred lines, and Yehsen, an open-pollinated local
nous AtFLC was completely inactivated (Fig. 7A). It variety, provided by the Tainan District Agricultural Research and Extension
Station, were used for vernalization treatment (Wang et al., 2000). Seed
was surprising to find that the BoFLC4-1::GUS tran- vernalization was conducted by treating seeds at 4°C under dim light for
script was increased after plant vernalization treat- 4 weeks after they were sown in the soil. The soil was kept moist during the
ment in most of the transgenic lines (Fig. 7B), but we treatment period to ensure seed germination. Plant vernalization was con-
have no explanation for this observation. ducted by transferring 8-week-old plants to 4°C for 2 to 8 weeks, as described
by the research group in the Tainan District Agricultural Research and
In Figure 8, a working hypothesis is proposed for the Extension Station (Wang et al., 2000). Arabidopsis (Arabidopsis thaliana)
different mechanisms in controlling cabbage and Ara- ecotype C24, purchased from Lehle Seeds (Round Rock, TX), was used for
bidopsis flowering by vernalization. The BoFLC tran- all experiments. Seed vernalization of Arabidopsis was carried out by
script gradually accumulates along with a decreased sterilizing seeds and placing them on Murashige and Skoog medium at 4°C
level of BoFT in the apex during cabbage growth. for 2, 4, 6, and 8 weeks, while plant vernalization was carried out by treating
the 10-d-old Arabidopsis rosette plants on plates at 4°C for the same period of
Vernalization of cabbage seeds does not alter the time. The transgenic lines were grown in same condition with the addition of
expression of FLC and FT. The presence of vernalization hygromycin for selection.
repressors or the absence of vernalization activators
could be responsible for the ineffectiveness of seed Cloning of BoFLC Genes and Sequence Analysis
vernalization in cabbage. The expression of repressors
or activators may depend on age and be regulated The cDNA of BoFLC4-1, lacking the MADS-box domain, was cloned by
reverse transcription-PCR using primers designed from the AtFLC sequence
coordinately in an opposite manner. During develop-
(forward primer 5#-ggcgataacctggtcaagat-3# and reverse primer 5#-tac-
ment, the loss of repressors or accumulation of activa- aaacgctcgcccttatc-3#). The full-length BoFLC4-1 gene was subsequently cloned
tors may be associated with increased competence to after screening a cabbage cDNA library constructed by the SMART cDNA
sense cold temperature. The accumulated vernaliza- library construction kit (CLONTECH, Palo Alto, CA). PCR and genomic
tion activators, together with cold temperature, in walking (GenomeWalker kit; CLONTECH) were carried out to obtain the
genomic and cDNA sequences of BoFLC4-1 and BoFLC3-2 genes. The promot-
mature cabbage plants can efficiently down-regulate ers of both BoFLC4-1 and BoFLC3-2 also were cloned with use of the same
the expression of FLC (Fig. 8A). Without vernalization, genomic walking approach. The phylogenetic relationship of the partial amino
cabbage remains in vegetative growth. The expression acid sequences of BoFLC3-2 and BoFLC4-1 and other FLC proteins from
1046 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Cabbage FLOWERING LOCUS C
Arabidopsis, Brassica napus, Brassica rapa, and Brassica oleracea was analyzed by Received December 26, 2004; returned for revision December 27, 2004;
the Phylip program (http://evolution.genetics.washington.edu/phylip.html). accepted December 27, 2004.
Partial amino acid sequences were used in this analysis because the exon 1
sequences of several BoFLC and BrFLC genes are unavailable (Schranz et al.,
2002). The sequence alignment between BoFLCs and AtFLC was conducted by
use of the BLAST 2 Sequences program at the National Center for Biotechnol-
ogy Information Web site (Tatusova and Madden, 1999).
LITERATURE CITED
An H, Roussot C, Suarez-Lopez P, Corbesier L, Vincent C, Pineiro M,
RNA Extraction and RNA Gel-Blot Analysis Hepworth S, Mouradov A, Justin S, Turnbull C, et al (2004) CON-
STANS acts in the phloem to regulate a systemic signal that induces
Total RNA was isolated from various tissues using TRIzol reagent photoperiodic flowering of Arabidopsis. Development 131: 3615–3626
(Invitrogen, Carlsbad, CA). An amount of 15 mg of total RNA was run on Bastow R, Mylne JS, Lister C, Lippman Z, Martienssen RA, Dean C (2004)
a 2.2 M formaldehyde-agarose gel and blotted onto a nylon membrane (Roche, Vernalization requires epigenetic silencing of FLC by histone methyla-
Mannheim, Germany). The membrane was stained with reversible methylene tion. Nature 427: 164–167
blue (Sigma B-1177; St. Louis) to check the loading and transfer of RNA in each Burn JE, Bagnall DJ, Metzger JD, Dennis ES, Peacock WJ (1993) DNA
sample. AtFLC and BoFLC4-1 cDNA, lacking the MADS box, and AtFT cDNA methylation, vernalization, and the initiation of flowering. Proc Natl
(AV563203, a EST clone from Kazusa DNA Research Institute) were used to Acad Sci USA 90: 287–291
generate antisense DIG-UTP-labeled riboprobes (Roche). Membranes were Chourad P (1960) Vernalization and its relations to dormancy. Annu Rev
hybridized at 62°C in hybridization solution (50% [v/v] formamide, 53 SSC, Plant Physiol 11: 191–238
0.1% [w/v] sodium lauroyl sarcosine, 0.02% [w/v] SDS, 2% [w/v] blocking Clarke JH, Dean C (1994) Mapping FRI, a locus controlling flowering time
reagent) overnight, and then washed in 23 SSC/0.1% (w/v) SDS twice for and vernalization response in Arabidopsis thaliana. Mol Gen Genet 242:
5 min at room temperature, and in 0.13 SSC/0.1% (w/v) SDS twice for 15 min 81–89
at 68°C. Detection was performed by use of the Detection Starter Kit II, Clough SJ, Bent AF (1998) Floral dip: a simplified method for
following the manufacturer’s instructions (Roche). The signal was analyzed Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J
with use of a luminescent image analyzer (LAS-1000 plus; Fujifilm, Tokyo) 16: 735–743
combined with Image Gauge version 3.12 software (Fujifilm). Favaro R, Pinyopich A, Battaglia R, Kooiker M, Borghi L, Ditta G,
Yanofsky MF, Kater MM, Colombo L (2003) MADS-box protein com-
plexes control carpel and ovule development in Arabidopsis. Plant Cell
Construction of Transgene and 15: 2603–2611
Arabidopsis Transformation Friend DJC (1985) Brassica. In AH Halevy, ed, Handbook of Flowering, Vol
II. CRC Press, Boca Raton, FL, pp 48–77
The sequence containing the 2.0-kb promoter and complete 5# UTR, exons, He Y, Michaels SD, Amasino RM (2003) Regulation of flowering time by
and introns of BoFLC4-1 (5.5 kb in total) was placed upstream of the GUS histone acetylation in Arabidopsis. Science 302: 1751–1754
reporter gene in the pCambia 1381Z vector (Cambia, Canberra, Australia). The Henderson IR, Dean C (2004) Control of Arabidopsis flowering: the chill
stop codon of BoFLC4-1 was deleted and immediately fused to the start codon before the bloom. Development 131: 3829–3838
of the GUS gene in the vector. A dominant-negative clone was generated by Henderson IR, Shindo C, Dean C (2003) The need for winter in the switch
overexpressing N-terminal or C-terminal truncated BoFLC4-1 in the pCambia to flowering. Annu Rev Genet 37: 371–392
2300 (Cambia) driven by a cauliflower mosaic virus 35S promoter. The Hepworth SR, Valverde F, Ravenscroft D, Mouradov A, Coupland G (2002)
N-terminal truncated protein lacked the N-terminal MADS-box domain of Antagonistic regulation of flowering-time gene SOC1 by CONSTANS
59 amino acids (DN1), whereas the C-terminal truncated protein lacked the and FLC via separate promoter motifs. EMBO J 21: 4327–4337
C domain of 49 amino acids (DN2). These constructs were sequence verified to Hossain MM, Inden H, Asahira T (1990) Seed vernalization interspecific
rule out any mistakes during cloning. They were subsequently transformed hybrids through in vitro ovule culture in Brassica. Plant Sci 68: 95–102
into Agrobacterium tumefaciens GV3101pMP90. Arabidopsis transformation Ito H, Saito T, Hatayama T (1966) Time and temperature factors for the
was carried out by use of the floral dip method (Clough and Bent, 1998). flower formation in cabbage. II. The site of vernalization and the nature
Transgenic plants were selected under 20 mg mL21 hygromycin (for pCambia of vernalization sensitivity. Tohoku J Agric Res 17: 1–15
1381Z) or 50 mg mL21 kanamycin (for pCambia 2300). Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: b-glucuron-
idase as a sensitive and versatile gene fusion marker in higher plants.
Flowering Time Analysis EMBO J 6: 3901–3907
Johanson U, West J, Lister C, Michaels S, Amasino R, Dean C (2000)
Flowering time was determined by the number of rosette leaves formed Molecular analysis of FRIGIDA, a major determinant of natural varia-
when the inflorescent stem reached approximately 1 to 3 cm long. The tion in Arabidopsis flowering time. Science 290: 344–347
flowering time of 19 and 13 independent T2 lines of DN1 and DN2 dominant- Koornneef M, Alonso-Blanco C, Peeters AJM, Soppe W (1998) Genetic
negative constructs, respectively, were assessed. Three replicate plants of each control of flowering time in Arabidopsis. Annu Rev Plant Physiol Plant
line were analyzed. Mol Biol 49: 345–370
Koornneef M, Blankestijn-de Vries H, Hanhart C, Soppe W, Peeters T
(1994) The phenotype of some late-flowering mutants is enhanced by
GUS Staining a locus on chromosome 5 that is not effective in the Landsberg erecta wild-
type. Plant J 6: 911–919
GUS staining was performed according to Jefferson et al. (1987), with
Krizek BA, Riechmann JL, Meyerowitz EM (1999) Use of the APETALA1
minor modifications. Briefly, transgenic plants were incubated in 100 mM
promoter to assay the in vivo function of chimeric MADS box genes. Sex
NaPO4, pH 7.0, 0.5 mM K-ferricyanide, 0.5 mM K-ferrocyanide, and 1.9 mM
Plant Reprod 12: 14–26
X-Gluc (5-bromo-4-chloro-3-indoyl-b-D-glucuronide). Plant tissues were vac-
Lee H, Suh S-S, Park E, Cho E, Ahn JH, Kim S-G, Lee JS, Kwon YM, Lee I
uum infiltrated briefly, then incubated at 37°C overnight. After staining,
(2000) The AGAMOUS-LIKE 20 MADS domain protein integrates floral
chlorophyll was cleared from the sample by 75% ethanol. Photos were taken
inductive pathways in Arabidopsis. Genes Dev 14: 2366–2376
under an Olympus SZX12 microscope (Tokyo).
Lee H, Xiong L, Gong Z, Ishitani M, Stevenson B, Zhu JK (2001) The
Arabidopsis HOS1 gene negatively regulates cold signal transduction
and encodes a RING finger protein that displays cold-regulated nucleo-
ACKNOWLEDGMENTS cytoplasmic partitioning. Genes Dev 15: 912–924
Lee I, Bleecker A, Amasino RM (1993) Analysis of naturally occurring late
We thank Drs. Kenrick Jaichard and Ning-Sun Yang and Miss Laura flowering in Arabidopsis thaliana. Mol Gen Genet 237: 171–176
Heraty for critically reading the manuscript, Dr. Pei-Ing Hwang and Miss Levy YY, Dean C (1998) The transition to flowering. Plant Cell 10:
Ho-Ming Chen for sequence analysis, and Dr. Yee-Yung Charng for valu- 1973–1990
able discussion during the process of this study. Michaels SD, Amasino RM (1999) FLOWERING LOCUS C encodes a novel
Plant Physiol. Vol. 137, 2005 1047
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.Lin et al.
MADS domain protein that acts as a repressor of flowering. Plant Cell Samach A, Onouchi H, Gold SE, Ditta GS, Schwarz-Sommer Z, Yanofsky
11: 949–956 MF, Coupland G (2000) Distinct roles of CONSTANS target genes in
Michaels SD, Amasino RM (2000) Memories of winter: vernalization and reproductive development of Arabidopsis. Science 288: 1613–1616
the competence to flower. Plant Cell Environ 23: 1145–1153 Schranz ME, Quijada P, Sung S-B, Lukens L, Amasino R, Osborn TC
Michaels SD, Amasino RM (2001) Loss of FLOWERING LOCUS C activity (2002) Characterization and effects of the replicated flowering time gene
eliminates the late-flowering phenotype of FRIGIDA and autonomous FLC in Brassica rapa. Genetics 162: 1457–1468
pathway mutations but not responsiveness to vernalization. Plant Cell Schwarz-Sommer Z, Hue I, Huijser P, Flor P, Hansen R, Tetens F, Lonnig
13: 935–942 W, Saedler H, Sommer H (1992) Characterization of the Antirrhinum
Michaels SD, Bezerra IC, Amasino RM (2004) FRIGIDA-related genes are floral homeotic MADS-box gene deficiens: evidence for DNA binding
required for the winter-annual habit in Arabidopsis. Proc Natl Acad Sci
and autoregulation of its persistent expression throughout flower de-
USA 101: 3281–3285
velopment. EMBO J 11: 251–263
Mizukami Y, Huang H, Tudor M, Hu Y, Ma H (1996) Functional domains of
Sheldon CC, Burn JE, Perez PP, Metzger J, Edwards JA, Peacock WJ,
the floral regulator AGAMOUS: characterization of the DNA binding
Dennis ES (1999) The FLF MADS box gene: a repressor of flowering in
domain and analysis of dominant negative mutations. Plant Cell 8:
Arabidopsis regulated by vernalization and methylation. Plant Cell 11:
831–845
Mouradov A, Cremer F, Coupland G (2002) Control of flowering time: 445–458
interacting pathways as a basis for diversity. Plant Cell (Suppl) 14: Sheldon CC, Conn AB, Dennis ES, Peacock WJ (2002) Different regula-
S111–S130 tory regions are required for the vernalization-induced repression of
Napp-Zinn K (1987) Vernalization—environmental and genetic regulation. FLOWERING LOCUS C and for the epigenetic maintenance of repres-
In JG Atherton, ed, Manipulation of Flowering. Butterworth, London, sion. Plant Cell 14: 2527–2537
pp 123–132 Sheldon CC, Rouse DT, Finnegan EJ, Peacock WJ, Dennis ES (2000) The
Noh Y-S, Amasino RM (2003) PIE1, an ISWI family gene, is required molecular basis of vernalization: the central role of FLOWERING
for FLC activation and floral repression in Arabidopsis. Plant Cell 15: LOCUS C (FLC). Proc Natl Acad Sci USA 97: 3753–3758
1671–1682 Simpson GG (2004) The autonomous pathway: epigenetic and post-
Nordborg M, Bergelson J (1999) The effect of seed and rosette cold transcriptional gene regulation in the control of Arabidopsis flowering
treatment on germination and flowering time in some Arabidopsis time. Curr Opin Plant Biol 7: 570–574
thaliana (Brassicaceae) ecotypes. Am J Bot 86: 470–475 Simpson GG, Dean C (2002) Arabidopsis, the rosetta stone of flowering
Onouchi H, Igeno MI, Perilleux C, Graves K, Coupland G (2000) time? Science 296: 285–289
Mutagenesis of plants overexpressing CONSTANS demonstrates novel Simpson GG, Gendall AR, Dean C (1999) When to switch to flowering.
interactions among Arabidopsis flowering-time genes. Plant Cell 12: Annu Rev Cell Dev Biol 15: 519–550
885–900 Sung S, Amasino RM (2004) Vernalization in Arabidopsis thaliana is
Osborn TC, Kole C, Parkin IA, Sharpe AG, Kuiper M, Lydiate DJ, Trick M mediated by the PHD finger protein VIN3. Nature 427: 159–164
(1997) Comparison of flowering time genes in Brassica rapa, B. napus and Tadege M, Sheldon CC, Helliwell CA, Stoutjesdijk P, Dennis ES, Peacock
Arabidopsis thaliana. Genetics 146: 1123–1129 WJ (2001) Control of flowering time by FLC orthologues in Brassica
Pellegrini L, Tan S, Richmond TJ (1995) Structure of serum response factor napus. Plant J 28: 545–553
core bound to DNA. Nature 376: 490–497 Tatusova TA, Madden TL (1999) BLAST 2 S, a new tool for comparing
Poduska B, Humphrey T, Redweik A, Grbic V (2003) The synergistic
protein and nucleotide sequences. FEMS Microbiol Lett 174: 247–250
activation of FLOWERING LOCUS C by FRIGIDA and a new flowering
Tzeng T-Y, Yang C-H (2001) A MADS box gene from lily (Lilium longiflorum)
gene AERIAL ROSETTE 1 underlies a novel morphology in Arabidopsis.
is sufficient to generate dominant negative mutation by interacting with
Genetics 163: 1457–1465
PISTILLATA (PI) in Arabidopsis thaliana. Plant Cell Physiol 42: 1156–1168
Reeves PH, Coupland G (2000) Response of plant development to
Wang SS, Chang CG, Lin DL, Yen YF, Wu MT (2000) Studies on cabbage
environment: control of flowering by daylength and temperature.
seed production in the lowland. Res Bull Tainan District Agric Im-
Curr Opin Plant Biol 3: 37–42
Reeves PH, Murtas G, Dash S, Coupland G (2002) early in short days provement Station 37: 56–64
4, a mutation in Arabidopsis that causes early flowering and reduces Wellensiek SJ (1964) Dividing cells as the prerequisite for vernalization.
the mRNA abundance of the floral repressor FLC. Development 129: Plant Physiol 39: 832–835
5349–5361 Zhang H, Ransom C, Ludwig P, van Nocker S (2003) Genetic analysis of
Riechmann JL, Krizek BA, Meyerowitz EM (1996) Dimerization speci- early flowering mutants in Arabidopsis defines a class of pleiotropic
ficity of Arabidopsis MADS domain homeotic proteins APETALA1, developmental regulator required for expression of the flowering-time
APETALA3, PISTILLATA, and AGAMOUS. Proc Natl Acad Sci USA switch Flowering Locus C. Genetics 164: 347–358
93: 4793–4798 Zhang H, van Nocker S (2002) The VERNALIZATION INDEPENDENCE 4
Riechmann JL, Meyerowitz EM (1997) MADS domain proteins in plant gene encodes a novel regulator of FLOWERING LOCUS C. Plant J 31:
development. Biol Chem 378: 1079–1101 663–673
1048 Plant Physiol. Vol. 137, 2005
Downloaded from on March 18, 2020 - Published by www.plantphysiol.org
Copyright © 2005 American Society of Plant Biologists. All rights reserved.You can also read
