Effect of Ppd-1 on the expression of flowering-time genes in vegetative and reproductive growth stages of wheat

ABSTRACT

The photoperiod sensitivity gene Ppd-1 influences the timing of flowering in temperate cereals such as wheat and barley. The effect of Ppd-1 on the expression of flowering-time genes was assessed by examining the expression levels of the vernalization genes VRN1 and VRN3/WFT and of two CONSTANS-like genes, WCO1 and TaHd1, during vegetative and reproductive growth stages. Two near-isogenic lines (NILs) were used: the first carried a photoperiod-insensitive allele of Ppd-1 (Ppd-1a-NIL), the other, a photoperiod-sensitive allele (Ppd-1b-NIL). We found that the expression pattern of VRN1 was similar in Ppd-1a-NIL and Ppd-1b-NIL plants, suggesting that VRN1 is not regulated by Ppd-1. Under long day conditions, VRN3/WFT showed similar expression patterns in Ppd-1a-NIL and Ppd-1b-NIL plants. However, expression differed greatly under short day conditions: VRN3/WFT expression was detected in Ppd-1a-NIL plants at the 5-leaf stage when they transited from vegetative to reproductive growth; very low expression was present in Ppd-1b-NIL throughout all growth stages. Thus, the Ppd-1b allele acts to down-regulate VRN3/WFT under short day conditions. WCO1 showed high levels of expression at the vegetative stage, which decreased during the phase transition and reproductive growth stages in both Ppd-1a-NIL and Ppd-1b-NIL plants under short day conditions. By contrast to WCO1, TaHd1 was up-regulated during the reproductive stage. The level of TaHd1 expression was much higher in Ppd-1a-NIL than the Ppd-1b-NIL plants, suggesting that the Ppd-1b allele down-regulates TaHd1 under short day conditions. The present study indicates that down-regulation of VRN3/WFT together with TaHd1 is the cause of late flowering in the Ppd-1b-NIL plants under short day conditions.

INTRODUCTION

The transition from vegetative to reproductive growth (flowering) is associated with heading time, one of the most important traits in cereal crops. In bread wheat (Triticum aestivum, 2n=6x=42, genome constitution AABBDD), heading time is genetically determined by three characteristic components, i.e., vernalization requirement, photoperiod sensitivity and narrow-sense earliness (earliness per se), which compose the autonomous promoting pathway (Worland and Snape, 2001). Of these characteristics, photoperiod sensitivity is the most important for determining heading time in autumn sown temperate cereals.

Photoperiod sensitivity is determined by the major genes Ppd-A1, Ppd-B1 and Ppd-D1 located on chromosomes 2A, 2B and 2D, respectively (Laurie, 1997). A barley (Hordeum vulgare, 2n=2x=14, HH) ortholog of the Ppd-1 genes, Ppd-H1, was identified as a pseudo-response regulator (PRR) gene with a CCT domain that showed similarity to Arabidopsis PRR7 (Turner et al., 2005). In Arabidopsis, the PRR family consists of five members (PRR9, PRR7, PRR5, PRR3 and PRR1/TOC1) that are involved in circadian clock function together with CIRCADIAN CLOCK-ASSOCIATED 1 (CCA1)/LATE ELONGATED HYPOCOTYL (LHY) and GIGANTEA (GI) (Nakamichi, 2011). Outputs from the circadian clock control the expression of CONSTANS (CO), a key gene of the photoperiod pathway, and up-regulate the mobile florigen gene FLOWERING LOCUS T (FT) (Imaizumi and Kay, 2006). The high sequence similarity of Ppd-H1 with PRR7 suggests that it is likely to be a circadian clock-associated gene. Comparative mapping indicated that wheat Ppd-A1, Ppd-B1 and Ppd-D1 are orthologous to barley Ppd-H1, a conclusion that is supported by sequence and expression analyses (Beales et al., 2007). Ppd-D1 is located on chromosome 2D and confers a stronger effect on heading time than other Ppd-1 alleles located on chromosome 2 homoeologs (Law et al., 1978). A semi-dominant photoperiod insensitive allele, Ppd-1a, was identified along with its photoperiod sensitive counterpart, Ppd-1b. A 2,089 bp deletion upstream of the coding region of Ppd-D1a, which causes mis-expression of the gene, is associated with photoperiod insensitivity in wheat (Beales et al., 2007). In the Ppd-A1a allele, a 1,027 bp or 1,117 bp deletion upstream of the coding region is associated with photoperiod insensitivity (Wilhelm et al., 2009). In the A and D genome, Ppd-1a alleles are more highly expressed than Ppd-1b alleles (Shaw et al., 2012). Contrary to Ppd-A1 and Ppd-D1, there are no sequence differences (insertion/deletion) between Ppd-B1a and Ppd-B1b (Beales et al., 2007). Among three photoperiod sensitive alleles, Ppd-A1b, Ppd-B1b, and Ppd-D1b, the B genome was predominantly expressed (Shaw et al., 2012). In all genomes, Ppd-1a alleles confer photoperiod insensitivity on wheat plant, but the mechanism is still unknown.

Several near-isogenic lines (NILs) of the spring wheat cv. Triple Dirk (TD) have been established using a backcrossing method (Pugsley, 1968). The TD cultivar is photoperiod insensitive with Ppd-D1a. The photoperiod sensitive allele Ppd-D1b was introduced from the wheat cv. Selkirk and an NIL developed by backcrossing to TD. The NIL carrying a photoperiod sensitive allele of Ppd-D1b was originally called TD(A) (Pugsley, 1968). In a previous study, we compared flowering-times in a growth chamber of TD and TD(A) plants, in which they were named Ppd-NIL and ppd-NIL, respectively (Murai et al., 2003). This comparison showed that the Ppd-NIL is not completely photoperiod-insensitive and that short day (SD) conditions cause only a slightly delay in heading-time compared to long day (LD) conditions. The ppd-NIL shows early-heading in a similar fashion to the Ppd-NIL under LD conditions, but extremely late heading under SD conditions. In a subsequent study, we compared the diurnal expression patterns of flowering-time genes in the Ppd-NIL and ppd-NIL, in which they were named Ppd-TD and ppd-TD, respectively (Shimada et al., 2009). At the 3-leaf stage of vegetative growth, both Ppd-TD and ppd-TD plants showed a diurnal pattern of VRN1 expression under both LD and SD conditions, with a peak of VRN1 expression at the beginning of the light period. VRN1, a flowering promoter gene, encodes a MADS-box transcription factor that controls vernalization-induced flowering in temperate cereals (Yan et al., 2003; Murai et al., 2003; Trevaskis et al., 2003; Danyluk et al., 2003). In contrast to the expression pattern of VRN1 in the two NILs, we found that diurnal expression of the wheat FT ortholog, VRN3 (also named WFT), differed greatly under LD and SD conditions. In both Ppd-TD and ppd-TD plants, VRN3/WFT showed a diurnal expression pattern under LD conditions and a constant and very low level of expression under SD conditions.

Temperate cereals have two CO-like genes, Wheat CO (WCO1) (Shimada et al., 2009) and Triticum aestivum HEADING DATE 1 (TaHd1) (Nemoto et al., 2003). Hd1 is a rice ortholog of CO (Yano et al, 2000). At the 3-leaf stage, Ppd-TD and ppd-TD plants have similar diurnal expression patterns for these genes, although they do show a slight difference under SD conditions: WCO1 and TaHd1 mRNAs accumulate at higher levels in the dark in Ppd-TD compared to ppd-TD (Shimada et al., 2009). The observations summarized above indicate that the difference in flowering time phenotypes of Ppd-TD and ppd-TD at the 3-leaf stage under SD conditions cannot be ascribed to dissimilarities in the expression patterns of flowering-time genes (Shimada et al., 2009). In this study, we examined the expression patterns of VRN1, VRN3/WFT, and two CO-like genes, WCO1 and TaHd1 in the Ppd-TD and ppd-TD during the vegetative and reproductive growth phases. We found that Ppd-TD and ppd-TD plants had similar patterns of VRN1 and WCO1 expression; however, VRN3/WFT and TaHd1 expression differed between the two genotypes at the phase transition and reproductive growth stages under SD conditions. The present results suggest that down-regulation of VRN3/WFT and TaHd1 is associated with late flowering in ppd-NIL plants under SD conditions. In this study, Ppd-TD and ppd-TD are renamed Ppd-1a-NIL and Ppd-1b-NIL, respectively, to describe their alleles precisely.

MATERIALS AND METHODS

Plant materials

Bread wheat (Triticum aestivum) cv. Triple Dirk (TD) lines that are nearly isogenic for the photoperiod-insensitive (Ppd-1a) or photoperiod-sensitive (Ppd-1b) alleles of Ppd-1 were developed by Pugsley (1968). Ppd-1b-NIL is photoperiod-sensitive; short day (SD) conditions cause an extremely delayed heading time compared with long day (LD) conditions (Murai et al., 2003). By contrast, Ppd-1a-NIL is not completely photoperiod-insensitive, and SD conditions only cause a slight delay in heading time compared with LD conditions. Both NILs carry the vernalization-insensitive (spring habit) genes Vrn-A1 and Vrn-B1.

Growth conditions

Sprouted seeds, which had been placed on wet filter paper for 2 days at 20°C, were sown in small soil-filled containers spaced 2 cm apart. Tsuchitaro (Sumirin, Japan) was used as soil, which contains optimized amount of fertilizers. For the phenotypic study, non-vernalized plants were grown in a growth chamber, LH-350S (NK system, Japan), under continuous light (24 h light), long day (16 h light/8 h dark) or short day (10 h light/14 h dark) conditions at 20°C (100 μE m^–2 s^–1). For the expression studies, non-vernalized plants were grown in a growth chamber under long day or short day conditions at 20°C (100 μE m^–2 s^–1).

Examination of heading-time

Nineteen to 28 plants of each line were grown under each of the day length conditions. The number of days from unfolding of the first leaf to flag leaf unfolding (D1f) was scored for each plant. The D1fs were used as a measure of “days to heading”. The difference of D1f between Ppd-1a-NIL and Ppd-1b-NIL were statistically analyzed by ANOVA. Photoperiod sensitivity was estimated as the ratios of the mean D1fs of plants under continuous light or LD conditions to those under SD conditions. The method for estimation of photoperiod sensitivity was originally described by Kato and Yamashita (1991), where D0f (days from sprouting to flag leaf unfolding) was used instead of D1f.

Estimation of growth phase

Identification of the growth stage in the wheat plants was based on leaf stage. For example, the 2-leaf stage is the growth stage at which the second leaf has just unfolded, so that the plants have unfolded first and second leaves and a folded third leaf. Phase transition from vegetative to reproductive growth was defined as the time when the first node became visible and started to elongate.

Real-time PCR analysis

Expression analyses were performed using RNAs extracted from leaves of three biological replicates each of one plant at each growth stage. The leaves were collected from plants at the 1-leaf to flag-leaf stages. Newly unfolded leaves were sampled from each plant: the first leaf from 1-leaf stage plants, the second leaf from 2-leaf stage plants, and the flag leaf from flag-leaf stage plants. The leaves were collected one hour after the beginning of the light period for the analysis of VRN1 and VRN3 expression, and three hours before the beginning of the light period for WCO1 and TaHd1. All these genes are known to show diurnal expression patterns (Shimada et al., 2009) and the sampling times were aimed at collecting leaves at approximately the times of peak expression. Total RNAs were extracted using ISOGEN (Nippon-gene, Japan) and cDNAs were subsequently synthesized with oligo-dT primer in accordance with the protocol for the Ready-To-Go T-Primed First-Strand Kit (GE Healthcare Life Sciences). Real-time PCR analyses were performed using a LightCycler 2.0 (Roche Diagnostics GmbH) with gene specific primer sets for VRN1, VRN3, WCO1 and TaHd1. The relative quantities of transcripts were determined by comparison to SYBR Green fluorescence of the Ubiquitin gene. For comparison of expression levels between WCO1 and TaHd1, we examined the amplification efficiency of each primer set using four 2-fold gene-specific plasmid dilutions and compared their amplification efficiency each other. Expression level of TaHd1 was rectified against that of WCO1, and then normalized by Ubiquitin gene. The sequences of the primer sets, the annealing temperatures, and the extension times for PCR are shown in Table 1.

Table 1. Sequences of primers used for real-time PCR analyses

Gene	Primer name	Sequence (5’-3’)	Annealing temperature (°C)	Extension time (sec.)
VRN1	TaMADS#11-545L	GGAGAGGTCACTGCAGGAGGA	65	10
	TaMADS#11-698R	GCCGCTGGATGAATGCTG
VRN3/WFT	WFT-F4	CAGGCCGGTCGATCTATACTA	58	15
	WFT-R4	TCCTGTTCCCGAAGGTCA
WCO1	CO1-LC	GCACCACTTGTAGGGGCAGA	63	16
	CO1-RC	TTGATCCTTGGCCGTGCTT
TaHd1	CO2-2L	CCAGTACCTACACAGCTTCCA	63	16
	CO2-2R	GCCTGCTTCTTCTCCTTGT
Ubiquitin	Ubi-1L	GCATGCAGATATTTGTGA	58	15
	Ubi-1R	GGAGCTTACTGGCCAC

RESULTS

Difference in heading times between Ppd-1a-NIL and Ppd-1b-NIL plants

We previously reported that the near-isogenic line (NIL) of bread wheat cv. Triple Dirk carrying the photoperiod-insensitive Ppd-1a allele (Ppd-1a-NIL) exhibits earlier heading than the NIL with the photoperiod-sensitive Ppd-1b allele (Ppd-1b-NIL) under short day (SD) conditions (Murai et al., 2003). Here, we first re-examined heading times in the two NILs under three different light regimes (Table 2). Under continuous light, no significant difference in heading time was found between Ppd-1a-NIL and Ppd-1b-NIL; both genotypes showed early heading. Under LD conditions (16 h light/8 h dark), heading time in Ppd-1a-NIL was significantly shorter than in Ppd-1b-NIL, although the difference was only three days on average. Under SD conditions, both NILs showed delayed heading but the effect was much greater in the Ppd-1b-NIL than in Ppd-1a-NIL. Ppd-1a-NIL plants transited from the vegetative to reproductive growth phase at the 4-leaf stage under LD conditions and at the 5-leaf stage under SD conditions. In comparison, Ppd-1b-NIL plants transited from the vegetative to reproductive growth phase at the 4-leaf stage under LD conditions, and at the 6-leaf stage under SD conditions (Fig. 1, 2, 3).

Table 2. Days to heading (mean ± SE) under short day (SD, 10 h light), long day (LD, 16 h light) and continuous light (24 h light) conditions

NILs	SD	LD
	10 h light (A)	16 h light (B)	24 h light (C)	(A)/(B)	(A)/(C)
Ppd-1a	67.1 ± 0.7	43.6 ± 0.2	35.8 ± 0.2	1.54	1.87
Ppd-1b	109.1 ± 1.0	46.8 ± 0.3	35.2 ± 0.4	2.33	3.10
F-value	1039.1**	71.8**	2.1ns

** : Significant difference at the 1% level with ANOVA.

^ns : Not significant.

Fig. 1.

VRN1 expression levels during vegetative and reproductive growth stages. The growth stages are defined by leaf stage, from the 1-leaf stage (1L) to the flag-leaf stage (FL). Horizontal bars below the leaf stages indicate the time of transition from vegetative to reproductive growth. Expression levels are normalized and relative to the Ubiquitin gene. Error bats represent standard error. (A) Expression levels in photoperiod-insensitive Ppd-1a-NIL plants grown under long day (LD) or short day (SD) conditions. (B) Expression levels in photoperiod-sensitive Ppd-1b-NIL plants grown under long day (LD) or short day (SD) conditions.

Fig. 2.

VRN3/WFT expression levels during the vegetative and reproductive growth stages. The growth stages are defined by leaf stage, from the 1-leaf stage (1L) to flag-leaf stage (FL). Horizontal bars below the leaf stages indicate the time of transition from vegetative to reproductive growth. Expression levels are normalized and relative to the Ubiquitin gene. Error bats represent standard error. (A) Expression levels in photoperiod-insensitive Ppd-1a-NIL plants grown under long day (LD) or short day (SD) conditions. (B) Expression levels in photoperiod-sensitive Ppd-1b-NIL plants grown under long day (LD) or short day (SD) conditions.

Fig. 3.

Levels of WCO1 (blue) and TaHd1 (red) expression during the vegetative and reproductive growth stages. The growth stages are defined by leaf stage, from the 1-leaf stage (1L) to flag-leaf stage (FL). Horizontal bars below the leaf stages indicate the time of transition from vegetative to reproductive growth. Expression levels are normalized and relative to the Ubiquitin gene. Error bats represent standard error. (A) WCO1 and TaHd1 expression levels in photoperiod-insensitive Ppd-1a-NIL plants grown under long day (LD) or short day (SD) conditions. (B) WCO1 and TaHd1 expression levels in photoperiod-sensitive Ppd-1b-NIL plants grown under long day (LD) or short day (SD) conditions.

Similar VRN1 expression patterns in the two NILs

We used real-time PCR to compare the levels of VRN1 expression in the two NILs during the vegetative and reproductive growth stages (Fig. 1). The cDNAs were obtained from leaves of 1-leaf to flag-leaf stage non-vernalized plants. Under LD conditions, both NILs transited from the vegetative to reproductive phase at the 4-leaf stage, and showed similar VRN1 expression patterns. Expression of VRN1 was low at the 1-leaf stage and increased as vegetative growth progressed. A higher level of expression was observed at the 4-leaf stage when the plants transited from the vegetative to reproductive phase. VRN1 expression was low at the beginning of the reproductive phase and increased as reproductive growth progressed. Under SD conditions, a low level of VRN1 expression was maintained during the vegetative to reproductive stages in both NILs. Despite the similar levels of VRN1 expression, the timing of phase transition differed between Ppd-1a-NIL and Ppd-1b-NIL plants under SD conditions: the Ppd-1a-NIL transited to the reproductive growth phase at the 5-leaf stage, while the Ppd-1b-NIL transited at the 6-leaf stage. The occurrence of similar patterns of VRN1 expression in Ppd-1a-NIL and Ppd-1b-NIL plants suggests that VRN1 is not regulated by the photoperiod pathway controlled by Ppd-1.

VRN3/WFT expression under SD conditions differs in Ppd-1a-NIL and Ppd-1b-NIL plants

The expression of VRN3/WFT was similar in the two NILs, with abundant transcripts during the vegetative to reproductive stages (Fig. 2). Under LD conditions, expression showed two peaks: the first occurred at the 4-leaf stage when the plants transited to the reproductive phase; the second peak occurred at the 7-leaf stage in Ppd-1a-NIL and at the flag-leaf stage in Ppd-1b-NIL. By contrast, the pattern of VRN3/WFT expression differed considerably under SD conditions between the two NILs. VRN3/WFT was highly expressed at the 5-leaf stage in Ppd-1a-NIL plants when they transited from the vegetative to reproductive phase, but a very low level of expression was present in Ppd-1b-NIL plants. Thus, the Ppd-1b allele down-regulated VRN3/WFT expression under SD conditions.

The expression patterns of WCO1 and TaHd1 vary with Ppd-1 allele

Under LD conditions, WCO1 expression was at a relatively higher level than that of TaHd1 in both NILs during the vegetative growth stage; TaHd1 expression increased as reproductive growth progressed (Fig. 3). Under SD conditions, WCO1 was highly expressed in the vegetative stage but expression decreased during the phase transition and reproductive growth stages in both Ppd-1a-NIL and Ppd-1b-NIL. The level of WCO1 expression in the vegetative stage under SD conditions was much higher than that under LD conditions, suggesting that WCO1 is associated with vegetative growth under SD conditions, possibly through suppression of the phase transition. By contrast to WCO1, TaHd1 was up-regulated during the reproductive stage. The level of TaHd1 expression was much higher in Ppd-1a-NIL than Ppd-1b-NIL plants, suggesting that the Ppd-1b allele down-regulates TaHd1 under SD conditions. The lower level of TaHd1 expression might be associated with the delayed heading time in Ppd-1b-NIL plants.

DISCUSSION

In temperate cereals, the Ppd-1 gene is the main regulator of photoperiod (day length) sensitivity in flowering. VRN1 is up-regulated by a long photoperiod (Murai et al., 2003; Danyluk et al., 2003; Dubcovsky et al., 2006; Sasani et al., 2009), and, moreover, shows a diurnal pattern of expression under both LD and SD conditions. Expression of VRN1 peaks at the beginning of the light period under both photoperiod conditions (Shimada et al., 2009). These observations suggest that VRN1 expression is controlled by photoperiod. To examine the relationship between Ppd-1 and VRN1, we examined VRN1 expression patterns in NILs carrying either the photoperiod-insensitive Ppd-1a allele (Ppd-1a-NIL) or the photoperiod-sensitive Ppd-1b allele (Ppd-1b-NIL). We found that VRN1 expression did not vary in plants with different Ppd-1 alleles (Fig. 1), suggesting that expression of this gene is not regulated by the photoperiod pathway mediated by Ppd-1. The flowering repressor gene, VRN2, reportedly shows a diurnal expression pattern under LD conditions (Dubcovsky et al., 2006). It is possible that VRN1 expression is directly or indirectly affected by VRN2 expression (Shimada et al., 2009; Distelfeld et al., 2009; Trevaskis, 2010). In barley, analysis of a F2 population segregating for Ppd-H1 alleles indicated that variation at Ppd-H1 did not affect VRN-H1/HvVRN1 expression (Campoli et al., 2012). This supports the conclusion that VRN1 is not downstream of Ppd-1.

Comparison of the expression patterns of VRN1 and VRN3/WFT under LD conditions showed that the VRN3/WFT expression pattern was correlated with that of VRN1. As a consequence, VRN3/WFT transcripts accumulated following the up-regulation of VRN1 in both the Ppd-1a-NIL and Ppd-1b-NIL (Figs. 1 and 2). The fact that this occurs in both genotypes indicates that the association of VRN3/WFT expression with VRN1 expression under LD conditions is not influenced by Ppd-1. Our observations support the idea that VRN1 either directly or indirectly up-regulates VRN3/WFT (Shimada et al., 2009; Trevaskis, 2010).

The expression of VRN3/WFT was similar in Ppd-1b-NIL and Ppd-1a-NIL plants under LD conditions. However, under SD conditions, Ppd-1b-NIL plants had a very low level of VRN3/WFT expression, whereas Ppd-1a-NIL plants had a moderately high level of expression. Recently, Shaw et al. (2012) also reported that Ppd-1b allele down-regulates VRN3/WFT (identical with TaFT1 in their study) using spring wheat cv. Paragon NILs. These results, together with the present flowering-time data (Table 2), strongly indicate that Ppd-1b down-regulates VRN3/WFT; alternatively, Ppd-1a up-regulates VRN3/WFT under SD conditions, resulting in delayed heading-time in Ppd-1b-NIL compared with Ppd-1a-NIL. Beales et al. (2007) reported that in vernalized seedlings of a photoperiod-insensitive wheat cv. Mercia NIL, the Ppd-1a allele was associated with high levels of VRN3/WFT expression. In this study, we used non-vernalized plants to avoid any possible effects of treatment on VRN3/WFT expression. Furthermore, we examined expression levels throughout the vegetative to reproductive growth stages in this study, contrary to the previous study (Beales et al., 2007; Shaw et al., 2012), in which expression analysis was performed in plants at seedling stage. We found that peak expression was at the 5-leaf stage in Ppd-1a-NIL under SD conditions, at the time when the plants are transiting from the vegetative to reproductive stages. Our observations indicate that up-regulation of VRN3/WFT confers the early heading phenotype in the Ppd-1a-NIL compared to the Ppd-1b-NIL.

Although Ppd-1b-NIL plants had low levels of VRN3/WFT expression in all growth stages, they initiated reproductive growth at the 6-leaf stage. It is not clear what mechanism is responsible for the eventual induction of reproductive growth in Pdp-1b-NIL plants under SD conditions, despite their low levels of expression of VRN3/WFT. The rice genome contains thirteen FT-like genes (Chardon and Damerval, 2005). One of these, RICE FLOWERING LOCUS 1 (RFT1), plays a compensatory role in phase transition in plants with RNAi mediated knockdown of Hd3a, the rice FT ortholog (Komiya et al., 2008). Furthermore, analyses of plants with RNAi mediated knockdown of RFT1 revealed that this gene regulates flowering in rice under LD conditions, which is not normally the photoperiod in which flowering occurs in this short day species (Komiya et al., 2009). With regard to temperate cereals, the barley genome contains five FT-like genes, one of which, HvFT3, is highly expressed under SD conditions (Faure et al., 2007). It is therefore possible that an FT-like gene other than VRN3/WFT is involved in the phase transition of Ppd-1b-NIL plants under SD conditions in wheat.

In barley, VRN-H1/HvVRN1, HvFT1, and Ppd-H1 have been identified as orthologs of wheat VRN1, VRN3, and Ppd-1 respectively. Hemming et al. (2008) reported that VRN-H1/HvVRN1 and HvFT1 were up-regulated by vernalization in doubled haploid lines regardless of Ppd-H1 genotype. This suggests that the vernalization pathway involving VRN1 and VRN3 is independent of the photoperiod pathway of Ppd-1. These results support our interpretation that the interaction between VRN1 and VRN3 is not mediated by the photoperiod pathway related to Ppd-1.

The sequence similarity of Ppd-1 to Arabidopsis PRR7 has led to the suggestion that Ppd-1 may function in a similar manner in the circadian clock, and act upstream of FT through CO (Shimada et al., 2009). In our earlier study (Shimada et al., 2009), we identified wheat CO (WCO1) by PCR using primers based on the sequence of barley HvCO1, which is located on chromosome 7H (Griffiths et al., 2003). Nine CO homologs have been identified in the barley genome. On the basis of sequence similarity and chromosomal synteny, HvCO1 may be the counterpart of rice Hd1. A mapping study showed that WCO1 was located on homoeologous group 7 (7A, 7B and 7D), indicating an orthologous relationship between WCO1 and HvCO1. Wheat chromosomes 7A, 7B and 7D are syntenic to rice chromosome 6 (the location of the Hd1 gene), suggesting that WCO1 and HvCO1 are orthologs of Hd1. It is known that TaHd1, previously identified as a wheat CO-like gene, is located on group 6 homoeologous chromosomes (Nemoto et al., 2003), indicating that TaHd1 is not an ortholog of Hd1. Nevertheless, an analysis in transgenic rice indicated that TaHd1 has some functions in flowering (Nemoto et al., 2003). Our previous diurnal expression studies using plants at the 3-leaf stage showed that WCO1 and TaHd1 had similar expression patterns (Shimada et al., 2009). Under LD conditions, WCO1 and TaHd1 mRNAs accumulate during the dark period; VRN3/WFT mRNA accumulates from the beginning of the light phase in the Ppd-1a-NIL. Under SD conditions, WCO1 and TaHd1 expression showed similar patterns as under LD conditions. In the Ppd-1b-NIL, WCO1 and TaHd1 also showed similar expression patterns under LD conditions in leaves at the 3-leaf stage. However, under SD conditions, WCO1 and TaHd1 had lower expression during the dark period in the Ppd-1b-NIL compared with the Ppd-1a-NIL.

In this study, we examined heading time and phase transition timing in wheat NILs (Table 2, Figs. 1, 2, 3). Under SD conditions, Ppd-1a-NIL plants transited from the vegetative to reproductive phase around the 4-leaf stage, whereas Ppd-1b-NIL plants transited at the 6-leaf stage. To investigate the relationship between delayed phase transition in Ppd-1b-NIL under SD conditions and CO gene expression, we compared the patterns of expression of WCO1 and TaHd1 in Ppd-1a-NIL and the Ppd-1b-NIL plants (Fig. 3). Under SD conditions, WCO1 was highly expressed in the vegetative stage; the level of expression decreased during phase transition and reproductive growth stages in both Ppd-1a-NIL and Ppd-1b-NIL plants. Higher expression level of WCO1 in the early stages of Ppd-1a lines under SD conditions was also reported by Shaw et al. (2012). In the present study, we observed that the level of expression in the vegetative phase under SD conditions is much higher than under LD conditions, suggesting that WCO1 functions in the suppression of phase transition under SD conditions. In barley, over-expression of HvCO1 accelerated flowering time and caused up-regulation of HvFT1 under LD conditions (Campoli et al., 2012). This suggests the possibility that under LD conditions WCO1 activates VRN3/WFT while under SD conditions WCO1 suppresses VRN3/WFT. This would parallel the case in short day plant rice, where Hd1 (rice CO) suppresses Hd3a (rice FT) in flowering non-inducible LD conditions (Yano et al., 2000).

By contrast to WCO1, TaHd1 is up-regulated during the reproductive stage in both Ppd-1a-NIL and Ppd-1b-NIL plants (Fig. 3). However, the expression level of TaHd1 was much higher in Ppd-1a-NIL than Ppd-1b-NIL plants, suggesting that the Ppd-1b allele down-regulates TaHd1 under SD conditions.

Conclusions

The findings from the present study, together with those of previous studies, suggest a model of gene interaction for flowering as illustrated in Fig. 4. In this model, VRN3/WFT functions as an integrator of the VRN1 related and Ppd-1 related pathways. VRN1 directly and/or indirectly up-regulates VRN3/WFT, resulting in phase transitions from vegetative to reproductive growth (Shimada et al., 2009; Trevaskis, 2010). Photoperiod (light-dark cycle) regulation of VRN1 is probably induced directly and/or indirectly by VRN2 expression (Shimada et al., 2009; Dubcovsky et al., 2006; Trevaskis, 2010). Under LD conditions, the effect of the photoperiod signal is not moderated by the influence of the Ppd-1 allele on WCO1 and TaHd1 expression (Fig. 4, A and C). Under SD conditions, WCO1 is up-regulated regardless of the Ppd-1 allele (Fig. 4, B and D). In this model, WCO1 shows inverse function in LD and SD; up-regulation of VRN3/WFT under LD conditions and down-regulation of VRN3/WFT under SD conditions. Therefore, the up-regulation of WCO1 and the down-regulation of VRN1 result in the down-regulation of VRN3/WFT and delayed flowering under SD conditions. Furthermore, photoperiod signaling via the Ppd-1b allele affects TaHd1 expression: TaHd1 is down-regulated, resulting in down-regulation of VRN3/WFT (Fig. 4D). The up-regulation of WCO1 and the down-regulation of TaHd1 together with VRN1 might strongly induce down-regulation of VRN3/WFT, resulting in an extreme delay in flowering in plants with the Ppd-1b allele.

Fig. 4.

Model for the interactions between the genetic pathways controlling flowering in wheat. VRN3/WFT acts as an integrator of all pathways. VRN1 directly or indirectly induces VRN3/WFT expression, leading to flowering. The photoperiod signals involving Ppd-1 are transmitted to WCO1 and TaHd1, and regulate VRN3/WFT expression. The present study suggests that WCO1 negatively regulates and TaHd1 positively regulates VRN3/WFT expression under short day (SD) conditions. The VRN3/WFT protein is postulated to be a mobile florigen that induces floral meristem determination and promotes phase transition from vegetative (V) to reproductive (R) growth. Arrows and T-bars mean promotive and suppressive effect, respectively. (A and B) Flowering pathway interactions in photoperiod-insensitive Ppd-1a-NIL plants grown under long day (LD) or short day (SD) conditions. Arrows indicated by bold lines show stronger effects. (C and D) Flowering pathway interactions in photoperiod-insensitive Ppd-1b-NIL plants grown under long day (LD) or short day (SD) conditions. Arrows indicated by bold lines show stronger effects. Arrows indicated by dotted lines indicate weaker effects.

ACKNOWLEDGMENT

We are grateful to the National Bioresource Project – Wheat (NBRP-KOMUGI) for providing wheat materials and the EST database. This work was supported in part by a Grant-in-Aid for Scientific Research (B) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (No. 21380009) to K. Murai.

REFERENCES

•  Beales,  J.,  Turner,  A.,  Griffiths,  S.,  Snape,  J. W., and  Laurie,  D. A. (2007) A Pseudo-Response Regulator is misexpressed in the photoperiod insensitive Ppd-D1a mutant of wheat (Triticum aestivum L.). Theor. Appl. Genet. 115, 721–733.
•  Campoli,  C.,  Drosse,  B.,  Searle,  I.,  Coupland,  G., and  von Korff,  M. (2012) Functional characterisation of HvCO1, the barley (Hordeum vulgare) flowering time ortholog of CONSTANS. Plant J. 69, 868–880.
•  Chardon,  F., and  Damerval,  C. (2005) Phylogenomic analysis of the PEBP gene family in cereals. J. Mol. Evol. 61, 579–590.
•  Danyluk,  J.,  Kane,  N. A.,  Breton,  G.,  Limin,  A. E.,  Fowler,  D. B., and  Sarhan,  F. (2003) TaVRT-1, a putative transcription factor associated with vegetative to reproductive transition in cereals. Plant Physiol. 132, 1849–1860.
•  Distelfeld,  A.,  Li,  C., and  Dubcovsky,  J. (2009) Regulation of flowering in temperate cereals. Curr. Opin. Plant Biol. 12, 178–184.
•  Dubcovsky,  J.,  Loukoianov,  A.,  Fu,  D.,  Valarik,  M.,  Sanchez,  A., and  Yan,  L. (2006) Effect of photoperiod on the regulation of wheat vernalization genes VRN1 and VRN2. Plant Mol. Biol. 60, 469–480.
•  Faure,  S.,  Higgins,  J.,  Turner,  A., and  Laurie,  D. A. (2007) The FLOWERING LOCUS T-like gene family in barley (Hordeum vulgare). Genetics 176, 599–609.
•  Griffiths,  S.,  Dunford,  R. P.,  Coupland,  G., and  Laurie,  D. A. (2003) The evolution of CONSTANS-like gene families in barley, rice and Arabidopsis. Plant Physiol. 131, 1855–1867.
•  Hemming,  M.N.,  Peacock,  W.J.,  Dennis,  E. S., and  Trevaskis,  B. (2008) Low-temperature and daylength cues are integrated to regulate FLOWERING LOCUS T in barley. Plant Physiol. 147, 355–366.
•  Imaizumi,  T., and  Kay,  S. A. (2006) Photoperiodic control of flowering: not only by coincidence. Trends Plant Sci. 11, 550–558.
•  Kato,  K., and  Yamashita,  S. (1991) Varietal variation in photoperiodic response, chilling requirement and narrow-sense earliness and their relation to heading time in wheat (Triticum aestivum L.). Jpn. J. Breed. 41, 475–484.
•  Komiya,  R.,  Ikegami,  A.,  Tamaki,  S.,  Yokoi,  S., and  Shimamoto,  K. (2008) Hd3a and RFT1 are essential for flowering in rice. Development 135, 767–774.
•  Komiya,  R.,  Yokoi,  S., and  Shimamoto,  K. (2009) A gene network for long-day flowering activates RFT1 encoding a mobile flowering signal in rice. Development 136, 3443–3450.
•  Laurie,  D. A. (1997) Comparative genetics of flowering time. Plant Mol. Biol. 35, 167–177.
•  Law,  C. N.,  Sutka,  J., and  Worland,  A. J. (1978) A genetic study of day-length response in wheat. Heredity 41, 185–191.
•  Murai,  K.,  Miyamae,  M.,  Kato,  H.,  Takumi,  S., and  Ogihara,  Y. (2003) WAP1, a wheat APETALA1 homolog, plays a central role in the phase transition from vegetative to reproductive growth. Plant Cell Physiol. 44, 1255–1265.
•  Nakamichi,  N. (2011) Molecular mechanisms underlying the Arabidopsis circadian clock. Plant Cell Physiol. 52, 1709–1718.
•  Nemoto,  Y.,  Kisaka,  M.,  Fuse,  T.,  Yano,  M., and  Ogihara,  Y. (2003) Characterization and functional analysis of three wheat genes with homology to the CONSTANS flowering time gene in transgenic rice. Plant J. 36, 82–93.
•  Pugsley,  A. T. (1968) Genetic studies of phasic development and their application to wheat breeding. In: Proceedings of the 3^rd International Genetics Symposium, (eds.: K. W. Findlay, and K. W. Shepherd), pp. 288–293. Australian Academy of Science, Canberra.
•  Sasani,  S.,  Hemming,  M. N.,  Oliver,  S. N.,  Greenup,  A.,  Tavakkol-Afshari,  R.,  Mahfoozi,  S.,  Poustini,  K.,  Sharifi,  H.-R.,  Dennis,  E. S.,  Peacock,  W. J., and  Trevaskis,  B. (2009) The influence of vernalization and daylength on expression of flowering-time genes in the shoot apex and leaves of barley (Hordeum vulgare). J. Exp. Bot. 60, 2169–2178.
•  Shaw,  L. M.,  Turner,  A. S., and  Laurie,  D. A. (2012) The impact of photoperiod insensitive Ppd-1a mutations on the photoperiod pathway across the three genomes of hexaploid wheat (Triticum aestivum). Plant J. DOI: 10.1111/j.1365-313X.2012.04971.x
•  Shimada,  S.,  Ogawa,  T.,  Kitagawa,  S.,  Suzuki,  T.,  Ikari,  C.,  Shitsukawa,  N.,  Abe,  T.,  Kawahigashi,  H.,  Kikuchi,  R.,  Handa,  H., and  Murai,  K. (2009) A genetic network of flowering time genes in wheat leaves, in which an APETALA1/FRUITFULL-like gene, VRN1, is upstream of FLOWERING LOCUS T. Plant J. 58, 668–681.
•  Trevaskis,  B. (2010) The central role of the VERNALIZATION1 gene in the vernalization response of cereals. Funct. Plant Biol. 37, 479–487.
•  Trevaskis,  B.,  Bagnall,  D. J.,  Ellis,  M. H.,  Peacock,  W. J., and  Dennis,  E. S. (2003) MADS box genes control vernalization-induced flowering in cereals. Proc. Natl. Acad. Sci. USA 100, 13099–13104.
•  Turner,  A.,  Beales,  J.,  Faure,  S.,  Dunford,  R. P., and  Laurie,  D. A. (2005) The pseudo-response regulator Ppd-H1 provides adaptation to photoperiod in barley. Science 310, 1031–1034.
•  Wilhelm,  E. P.,  Turner,  A. S., and  Laurie,  D. A. (2009) Photoperiod insensitive Ppd-A1a mutations in tetraploid wheat (Triticum durum Desf.). Theor. Appl. Genet. 118, 285–294.
•  Worland,  T., and  Snape,  J. W. (2001) Genetic basis of worldwide wheat varietal improvement. In: The World Wheat Book, A History of Wheat Breeding, (eds.: A. P. Bonjean, and W. J. Angus), pp. 59–100. Lavoisier Publishing, Londres.
•  Yan,  L.,  Loukoianov,  A.,  Tranquilli,  G.,  Helguera,  M.,  Fahima,  T., and  Dubcovsky,  J. (2003) Positional cloning of the wheat vernalization gene VRN1. Proc. Natl. Acad. Sci. USA 100, 6263–6268.
•  Yano,  M.,  Katayose,  Y.,  Ashikari,  M.,  Yamanouchi,  U.,  Monna,  L.,  Fuse,  T.,  Baba,  T.,  Yamamoto,  K.,  Umehara,  Y.,  Nagamura,  Y., and  Sasaki,  T. (2000) Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12, 2473–2484.