SEARCH

SEARCH BY CITATION

Keywords:

  • transcription factor;
  • flowering;
  • development;
  • NF–Y;
  • CONSTANS

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

NF–Y transcription factors represent a complex of three proteins called NF–YA, NF–YB and NF–YC. Each protein is highly conserved in eukaryotes, and in the plant lineage has undergone numerous rounds of duplication. Individual NF–Y are emerging as important regulators of several essential plant processes, including embryogenesis, drought resistance, maintenance of meristems in nitrogen-fixing nodules and photoperiod-dependent flowering time. Building on the recent finding that NF–YB2 and NF–YB3 have overlapping functionality in Arabidopsis photoperiod-dependent flowering (Kumimoto et al., 2008), we have identified three NF–YC (NF–YC3, NF–YC4, and NF–YC9) that are also required for flowering, and physically interact in vivo with both NF–YB2 and NF–YB3. Furthermore, NF–YC3, NF–YC4 and NF–YC9 can physically interact with full-length CONSTANS (CO), and are genetically required for CO-mediated floral promotion. Collectively, the present data greatly strengthens and extends the argument that CO utilizes NF–Y transcription factor complexes for the activation of FLOWERING LOCUS T (FT) during photoperiod-dependent floral initiation.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Many plant species have evolved to initiate flowering under specific photoperiods (Imaizumi and Kay, 2006; Kobayashi and Weigel, 2007). For example, flowering in the model plant Arabidopsis thaliana is rapidly induced in long photoperiods: i.e. Arabidopsis is a long-day (LD) plant. Other photoperiod-responsive species, such as the monocot rice (Oryza sativa), are induced to flower in short photoperiods (SD plants). In Arabidopsis, a key regulator of photoperiod-dependent flowering time is the zinc-finger type transcriptional activator CONSTANS (CO; Redei, 1962; Koornneef et al., 1991; Putterill et al., 1995). CO mRNA levels are controlled by the circadian clock and oscillate on a daily basis: CO expression peaks during the day in LD conditions, and peaks during the night in SD conditions (Suarez-Lopez et al., 2001). Peak CO expression during the day is essential for CO activity because the protein is rapidly degraded in the dark (Valverde et al., 2004; Jang et al., 2008; Liu et al., 2008). Under LD conditions, CO protein accumulates and rapidly induces the expression of FLOWERING LOCUS T (FT) (Samach et al., 2000). FT protein subsequently translocates to the meristem and triggers its conversion to a floral meristem (Corbesier et al., 2007; Jaeger and Wigge, 2007; Mathieu et al., 2007; Tamaki et al., 2007).

Although CO is thought to directly control FT expression (Samach et al., 2000), precisely how CO might integrate with the FT promoter remains unclear. Potentially addressing this question, several recent publications demonstrated that NF–Y transcription factors are intimately involved in photoperiod-dependent flowering. Research groups studying flowering time in both tomato (Solanum lycopersicum) and Arabidopsis identified NF–Y subunits as CO-interacting proteins (Ben-Naim et al., 2006; Wenkel et al., 2006). Following these initial descriptions of NF–Y/CO interactions, two independent research groups described the same nf–yb2 loss-of-function allele as causing delays in flowering time (Cai et al., 2007; Chen et al., 2007). Conversely, overexpression of NFYB2 resulted in the opposite phenotype: significantly more rapid flowering. Finally, nf–yb2 nf–yb3 double mutants phenocopied co mutants: i.e. the late flowering and strong downregulation of FT expression (Kumimoto et al., 2008). Because mammalian and yeast NF–Ys are well-characterized as transcriptional-activating, DNA-binding complexes, these results strongly suggested that CO interacts with DNA via an NF–Y platform.

Genes encoding NF–Y transcription factors are found in all eukaryotes (Edwards et al., 1998; Maity and de Crombrugghe, 1998; Mantovani, 1999; Matuoka and Chen, 2002; Siefers et al., 2009). In mammals, NF–Y functions as a heterotrimer consisting of the single gene-encoded subunits NF–YA, NF–YB and NF–YC; individual NF–Y subunits do not appear to have DNA binding or transcriptional activation properties (Maity and de Crombrugghe, 1998). Mammalian NF–Y subunits assemble in a strict, stepwise fashion to form the mature, DNA-binding transcription factor (Maity et al., 1992; Sinha et al., 1996). Initially, the two histone fold-containing proteins, NF–YB and NF–YC, form a dimer that translocates to the nucleus (Frontini et al., 2004; Steidl et al., 2004; Goda et al., 2005; Tuncher et al., 2005). Once in the nucleus, NF–YA is recruited, and the resulting mature NF–Y transcription factor is competent to bind promoters at CCAAT nucleotide sequences. Recent bioinformatic analyses suggest that ∼7.6% of all human promoters have functional CCAAT binding sites (FitzGerald et al., 2004).

In the plant lineage, the discovery of a complete NF–YA/B/C complex controlling the expression of a particular gene or process has never been described. Single plant NFYA and NFYB genes are known to have functions in embryogenesis, drought resistance, ABA signaling, nitrogen-fixing nodule development and flowering time (Lotan et al., 1998; Kwong et al., 2003; Combier et al., 2006; Nelson et al., 2007; Suzuki et al., 2007; Warpeha et al., 2007; Li et al., 2008). To our knowledge, only one NFYC loss-of-function phenotype has been reported in Arabidopsis: nf–yc4 mutants have increased sensitivity to ABA in plate germination assays (Warpeha et al., 2007). The current lack of knowledge regarding complete NF–Y complexes in the plant lineage is most likely the result of overlapping functionality: in Arabidopsis there are 10 NF–YA, 13 NF–YB and 13 NF–YC (Gusmaroli et al., 2001, 2002; Siefers et al., 2009), and this expansion appears to be consistent in all sequenced monocots and dicots (Stephenson et al., 2007; Thirumurugan et al., 2008; Siefers et al., 2009).

In this paper we identify and describe three Arabidopsis NFYC genes with overlapping functionality in flowering time. Furthermore, we demonstrate that CO function in floral promotion is, at least, partially dependent on these genes. We present yeast two-hybrid (Y2H) and in vivo interaction data demonstrating that these three NF–YC proteins physically interact with the known floral-promoting NF–YB2 and NF–YB3 proteins. Further, as previously demonstrated for tomato COL1 (TCOL1; Ben-Naim et al., 2006), our data supports the general concept that CO and CO-LIKE (COL) proteins are recruited to NF–Y complexes via the NF–YC subunit.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Identification of NFYC candidate genes in flowering-time regulation

In Arabidopsis, NFYC represents a multigene family containing thirteen members (Figure 1a) (Siefers et al., 2009). Using yeast two-hybrid (Y2H) analyses, it was previously reported that eight different NF–YC (NF–YC1–NF–YC7 and NF–YC9) can physically interact with the CCT (CO, CO-LIKE and TIMING OF CAB1) domains of CO and/or COL15 (Wenkel et al., 2006). Although this data clearly indicated the potential for NF–YC/CCT-domain interactions, we expected a more limited set of interactions in vivo– an expectation supported by the highly variable, tissue-specific expression patterns of NFYC genes (Siefers et al., 2009). To identify likely candidates for involvement in the CO-dependent induction of flowering, we performed tissue-specific gene expression analyses on all 13 members of the NFYC family. Because CO induces FT expression in the phloem tissue of young, LD-grown leaves (An et al., 2004), we inferred that candidate NFYC genes would also be expressed in the leaf vasculature at this developmental stage.

image

Figure 1.  A subset of NF–YC have consistently strong vascular expression. (a) Phylogenetic relationships of 13 Arabidopsis NF–YC genes. Vascular-expressed genes are highlighted by the red box. Panel adapted from Siefers et al. (2009). (b) Promoter:GUS expression patterns for vascular-expressed NF–YCs. Only NF–YC3, NF–YC4 and NF–YC9 had strong and consistent vascular expression in 10-day-old plants. (c) Protein blot analysis using native IG-Y antibodies raised against peptides specific for NF–YC3 and NF–YC4.

Download figure to PowerPoint

The promoter region (1000 bp upstream of the ATG) for each NFYC was used to drive in planta expression of a β-glucuoronidase (GUS) reporter gene (Siefers et al., 2009). Stable transgenic lines were grown for 10 days under standard LD conditions (16-h light/8-h dark; 22°C), and were then microscopically assayed for GUS expression in the vascular tissue. NFYC3, NFYC4 and NFYC9 were consistently and strongly expressed in the leaf vasculature (Figure 1b). These three NFYC are also expressed in the leaf mesophyll and other tissues, where they probably participate in the transcriptional control of genes unrelated to flowering-time regulation (Siefers et al., 2009). Phylogenetic analysis revealed that NF–YC3, NF–YC4 and NF–YC9 are very closely related (Figure 1a); in fact, NF–YC3 and NF–YC9 are 100% identical throughout their conserved histone fold motifs. Therefore, we considered NFYC3, NFYC4 and NFYC9 to be our best candidates for further investigations.

Arabidopsis requires NFYC3, NFYC4 and NFYC9 for the proper timing of photoperiod-induced flowering

We isolated stable, homozygous T–DNA insertion lines for NFYC3 (SALK_034838, nf–yc3-1), NFYC4 (SALK_032163, nf–yc4-1) and NFYC9 (SALK_058903, nf–yc9-1). The T–DNA insertions in nf–yc3-1, nf–yc4-1 and nf–yc9-1 are located ∼299 base pairs (bp) upstream of the translational start, ∼512 bp after the translational start in an annotated exon, and ∼102 bp upstream of the translational start, respectively. To examine the effects of these T-DNA insertions, we developed native antibodies for NF–YC3 and NF–YC4 (Figure 1c, we do not have an antibody for NF–YC9). Based on protein blot analyses, nf–yc3-1 and nf–yc4-1 represent very strong knock-down alleles (we did occasionally detect NF–YC3 protein, but it never exceeded ∼5% normal accumulation). For the T-DNA insertion in nf–yc9-1, we examined mRNA accumulation by quantitative, real-time (qRT)-PCR and microarray analysis (Table S1). We estimate that nf–yc9-1 retains ∼30% normal NFYC9 expression. These are currently the strongest mutant alleles we have obtained for each gene, and the only alleles discussed in this paper (hereafter discussed without allele designation).

Under standard LD conditions, none of the single mutants showed significant changes in flowering time compared with the parental ecotype control [Columbia (Col–0); Fig. 2b]. Because NF–YC3, NF–YC4 and NF–YC9 have high amino acid identity, we hypothesized that functional overlap could be masking roles in flowering-time regulation. To test this hypothesis, we created all three possible double mutant combinations. Under LD conditions, all three double mutants showed a mild delay in flowering. The double mutant for the closely related paralogs NFYC3 and NFYC9 was consistently later flowering than the other combinations (Figure 2a–b). This apparent functional overlap was fully confirmed by the creation of the nf–yc3 nf–yc4 nf–yc9 triple mutants (hereafter ‘nfyc triple). By comparison with parental Col–0 (mean = 15.6 leaves at bolting), the nfyc triple produced almost twice as many total leaves before flowering (mean = 28.6).

image

Figure 2. nf–yc mutants are delayed in flowering under longh-day (LD) conditions. (a) Mutant plant phenotypes (day 32) compared with parental Col-0. (b) Comparison of leaf number at flowering for all possible nf–yc3, nf–yc4 and nf–yc9 mutant combinations. (c) Complementation of the nf–yc triple mutant by NF–YC3, NF–YC4 and NF–YC9 genomic constructs in LD conditions. Three independent, basta-selected T2 populations were examined for each genomic construct. Bars in all panels represent means for at least 10 plants per genetic background (±95% confidence interval), and asterisks represent significantly different comparisons with the Col-0 parental ecotype [anova< 0.0001; Bonferroni’s multiple comparison test (BMCP), < 0.001].

Download figure to PowerPoint

Although nfyc triple mutants had significantly delayed flowering, they did not flower as late as either cosail (Col–0 mutant allele of CO; mean = 51 total leaves in our LD conditions, data not shown) or nf–yb2 nf–yb3 mutants (mean = 48.3; Kumimoto et al., 2008). The simplest explanation is that the nfyc triple still had residual expression of NFYC9, generating enough functional NF–YC9 to prevent greater delays. Nevertheless, we cannot exclude the possibility that other NFYC genes are involved in flowering-time regulation.

To confirm that the observed flowering-time phenotypes were caused by the respective NFYC T–DNA insertions, we performed complementation experiments. Constructs that expressed NFYC3, NFYC4 and NFYC9 from their native promoters were intergrated into the nfyc triple by Agrobacterium-mediated transformation (Bechtold et al., 1993). Transformants were selected by resistance to glufosinate ammonium (BASTA). Each construct was able to rescue the nfyc triple late-flowering phenotype (Figure 2c; T2 generation plants shown). For all three genes, >50% of the original BASTA resistant T1 lines rescued the late-flowering phenotype.

NFYC3, NFYC4 and NFYC9 are primarily involved in photoperiod-dependent flowering

Previous research demonstrated that the roles for NFYB2 and NFYB3 in flowering time appear to be primarily confined to the photoperiod-dependent pathway (Kumimoto et al., 2008). To determine if NFYC3, NFYC4 and NFYC9 are required in the autonomous and vernalization pathways, flowering time for nfyc mutants was quantified under SD conditions (8–h light/16–h dark), and after extended cold treatments (4°C for 10 weeks). Under SD conditions, nfyc triple mutants sometimes flowered a few leaves earlier than parental Col–0 (Figure 3a), but these differences were not consistently reproducible (e.g. see mock-treated Col–0 versus nfyc triple comparison in Figure 3c). In the vernalization experiments, nfyc mutants flower later than Col–0, but are similarly responsive to the vernalization treatments: e.g. Col–0 flowered 21% earlier after vernalization compared with 29% earlier for the nfyc triple mutant (Figure 3b).

image

Figure 3.  Alternative flowering pathways are largely unaffected in the nf–yc triple mutant. Mean leaf number at flowering under (a) SD conditions (8-h light/16-h dark), (b) vernalization treatment (10 weeks at 4°C, moist growth media, ∼100% humidity), and (c) GA3 treatment [short day (SD) and long day (LD); see text]. Asterisks represent significantly different comparisons with the appropriate Col-0 control (anova, < 0.0001, BMCP, < 0.001). Error bars ± 95% confidence interval.

Download figure to PowerPoint

We additionally determined if NFYC3, NFYC4 and NFYC9 were involved in the gibberellic acid (GA) flowering pathway. Col–0 plants and nfyc triple mutants were either mock or GA3 treated in both SD and LD photoperiods. For SD experiments, plants were sprayed weekly with 100 μm GA3 until bolting (Wilson et al., 1992). For LD experiments, plants were germinated and grown on 10–μm GA3 plates for 6 days, transferred to soil and sprayed one additional time with 100 μm GA3 on day 8. In both SD and LD conditions (Figure 3c), nfyc triple mutants were highly responsive to GA treatment (similar to Col–0). In SD conditions, nfyc triple mutants flowered earlier than Col–0 after treatment with GA: this difference was statistically significant in some experiments (shown), but only represented a trend towards early flowering in other experiments. This suggests that nfyc triple mutants may have slightly enhanced sensitivity to externally applied GA. Overall, the flowering-time experiments are consistent with a predominant NFY role in the photoperiod-dependant flowering pathway.

NFYC regulates FT transcript expression

Previous research demonstrated that NFYB2 and NFYB3 have additive roles in the positive regulation of FT expression (Kumimoto et al., 2008). To confirm a similar role for NFYC3, NFYC4 and NFYC9, we examined the comparative expression of FT and CO in Col–0, nfyc triple and co-sail genetic backgrounds over a 24–h time course. CO expression was similar between Col–0 and nfyc triple, although we note that slightly higher CO levels in late-day time points for nfyc triple mutants were a consistent trend (Figure 4a). This data demonstrates that reductions in CO were not responsible for the measured delays in nfyc triple flowering time. As consistently reported for LD-grown Arabidopsis (Imaizumi and Kay, 2006), FT expression was maximal approximately 16 h after lights were switched on (Figure 4b). FT expression was significantly repressed in both the nfyc triple (∼twofold) and co-sail (essentially ‘off’) mutants. Intermediate levels of FT expression (between the wild type and co-sail) in the nfyc triple mutants were consistent with both residual NFYC9 expression (Figure 4e) and the intermediate flowering delays.

image

Figure 4.  Relative expression levels of CO, FT, NF–YC3, NF–YC4 and NF–YC9 in flowering-time mutants by qRT-PCR over a time course. (a) CO, (b) FT, (c) NF–YC3, (d) NF–YC4 and (e) NF–YC9 expression in nf–yc triple, co-sail and wild type (WT) genetic backgrounds. Expression levels presented relative to reference gene (±SE, see Experimental procedures).

Download figure to PowerPoint

To test the possibility that NFYC genes are regulated by CO, we also measured transcript levels of NFYC3, NFYC4 and NFYC9 in Col–0 compared with nfyc triple and co-sail mutants over the time course. By comparison with parental Col–0, none of the NFYCs were significantly misregulated in co-sail, and all three NFYCs were strongly downregulated in nfyc triple plants (Figure 4c–e). All three NFYCs were most highly expressed in late-day to night-time points, with sharp increases beginning between 12 and 16 h after lights were switched on (NFYC4 had an additional peak at 4 h). It is important to note that these NFYCs are expressed both in and outside of the vascular tissue; therefore, this data represents the entire aboveground plants, and may or may not accurately represent what happens in the vascular tissue. Collectively, the qRT-PCR data supports the hypothesis that NFYC3, NFYC4 and NFYC9 act as upstream regulators of FT expression. This data supports the hypothesis that NF–YC3, NF–YC4 and NF–YC9 act in a complex with NF–YB2, NF–YC3 and CO to positively regulate FT expression.

CO requires NFYC3, NFYC4 and NFYC9 to drive early flowering

All available evidence indicates that specific NFYs are required for normal FT expression and photoperiod–dependent flowering. The molecular and developmental phenotypes of various nfy mutants appear to be virtually identical to those previously published for various co alleles (Robson et al., 2001). Nevertheless, there is no available genetic evidence that CO function in floral promotion requires NFY. To provide this genetic evidence, we transformed Col–0 and nfyc triple with a CO overexpression construct (35S:CO-YFP/HA), and measured the total number of leaves at flowering. We hypothesized that constitutively overexpressed CO would drive early flowering when transformed into Col–0, but not into the nfyc triple mutant.

To establish an unbiased view of the range of possible phenotypes, we initially examined phenotypes of first-generation (T1) transformants. Col–0 and nfyc triple mutant control plants flowered at 13.4 (±0.79) and 23.2 (±2.27) total leaves, respectively (Figure 5a–b). When 35S:CO-YFP/HA was introduced into Col–0, we observed a bimodal distribution of flowering times. As expected, the majority of lines (2/3) flowered significantly earlier than wild-type controls, whereas a smaller subset flowered later than Col–0 (Figure 5c). After repeating this analysis several times, we consistently measured this bimodal distribution (we suspect that late flowering lines are caused by gene silencing, but this has not been confirmed in T1 plants; however, we have confirmed high rates for 35S:CO silencing in subsequent generations). Significantly, the same 35S:CO-YFP/HA construct was unable to drive early-flowering phenotypes in the nfyc triple mutant (Figure 5d, early-flowering classes are completely missing from the frequency distribution). On average, 35S:CO-YFP/HA nfyc triple mutants did flower slightly earlier than nfyc triple mutants (mean = 21.2 ± 5.4 leaves), but this difference was not statistically significant (P > 0.05; anova). This non-significant, but probably biologically relevant decrease in total leaf number is consistent with residual NFYC9 accumulation in the nfyc triple mutant.

image

Figure 5. CO requires NF–YC3, NF–YC4 and NF–YC9 to drive early flowering. Data represents plant frequency distributions based on total leaf number at flowering (bins = 3 leaves; i.e. ‘12’ = range of 11–13). (a) Col-0 (n = 12); (b) nf–yc triple (n = 11); (c) 35S:CO (n = 26 T1 individuals); and (d) 35S:CO nf–yc triple (n = 42 T1 individuals). The gray bar extending through each panel represents the range of flowering-time responses in Col-0.

Download figure to PowerPoint

To extend the T1 results, we generated two independent, homozygous (T3), single-insertion 35S:CO-YFP/HA nfyc triple lines (35S:CO-YFP/HA nfyc triple-16 and -100). We compared these two lines with stably overexpressed CO in the parental background (35S:CO-YFP/HA-1 and 35S:CO-YFP/HA-202) for three variables: CO mRNA expression, FT mRNA expression and flowering time. Regardless of genetic background, all chosen 35S:CO lines had elevated CO expression (Figure 6a). For 35S:CO in the wild-type background, elevated CO levels translated to the expected elevated FT levels (Figure 6b). For both 35S:CO nfyc triple lines, FT levels did not significantly change relative to the depressed levels in the nfyc triple parental line – in fact, expression of FT remained below the levels of non-transgenic, parental Col–0 plants. Predictably, these 35S:CO nfyc triple mutants also maintained the late-flowering phenotype of the parental nfyc triple mutant (Figure 6c–d).

image

Figure 6. 35S:CO does not significantly alter FT expression in the nf–yc triple background. (a–c) The ‘0’ line represents the Col-0 mean. (a) Mean CO and (b) mean FT mRNA expression measured by qRT-PCR (Log2 values, ±SD). (c,d) Mean (±95% confidence interval) total leaf number at flowering for stable, homozygous T3 lines relative to Col-0 and pictures of representative plants for each genotype. Col-0 flowered at 12.8 leaves in this experiment.

Download figure to PowerPoint

Because 35S:CO-YFP/HA nfyc triple-16 and 35S:CO-YFP/HA nfyc triple-100 had lower CO overexpression than 35S:CO-YFP/HA-1 and 35S:CO-YFP/HA nfyc triple-202, we additionally performed test crosses to parental Col–0 to demonstrate that these transgene insertions could drive early flowering in the context of functional NF–YC3, NF–YC4 and NF–YC9. Early flowering was readily measured in the resulting 35S:CO-YFP/HA/-NFYC3/nf–yc3-1 NFYC4/nf–yc4-1 NFYC9/nf–yc9-1 genotype plants (Figure S1). Thus, 35S:CO overexpression could not rescue either low FT expression or the late-flowering phenotypes of nfyc triple mutants. This data strongly supports the conclusion that NFYC3, NFYC4 and NFYC9 are necessary for CO function.

NF–YC3, NF–YC4 and NF–YC9 physically interact with NF–YB2, NF–YB3 and CO

Yeast two-hybrid studies previously revealed interactions between the CCT domain of CO and various NF–YB and NF–YC subunits. Relative to NF–Y with genetically demonstrated roles in flowering time, Wenkel et al. (2006) reported that NF–YB2, but not NF–YB3, could interact with the CCT domain of CO. Additionally, the CCT domain of CO interacted with NF–YC3 and NF–YC9, but not with NF–YC4. Ben-Naim et al. (2006) also found that THAP5c (tomato homolog of NF–YC9) interacted with full-length tomato COL1 and Arabidopsis CO. We wished to confirm and extend these findings to full-length Arabidopsis CO for the entire suite of NF–Ys with genetically demonstrated roles in photoperiod-dependent flowering: NF–YB2 and NF–YB3, and NF–YC3, NF–YC4 and NF–YC9. We also tested in vivo interactions between the NF–YB and NF–YC proteins.

We initially examined Y2H interactions between NF–YB2/3 and NF–YC3/4/9. Both NF–YB2 and NF–YB3 interacted strongly with NF–YC3, NF–YC4 and NF–YC9 (Figure 7a). Y2H interactions worked equally well, regardless of which subunit was fused to the Gal4 DNA-binding (DBD) or activation domain (AD), although DBD:NF–YC3 fusions had low to moderate autoactivation (data not shown). To test whether these interactions occur in vivo, we performed co-immunoprecipitation experiments. Initially, we transformed Col–0 with 35S:NFYB2- or 35SNFYB3-YFP/HA constructs. In agreement with previous studies, both constructs were able to drive early flowering under LD conditions (data not shown; Cai et al., 2007; Kumimoto et al., 2008). Total protein extracts were then collected from 10-day-old seedlings and NF–YB2 or NF–YB3 were isolated by immunoprecipitation with HA antibodies. Co-immunoprecipitation of NF–YC3 and NF–YC4 was assayed by western blot analysis using native antibodies. NF–YB2 and NF–YB3 were readily able to co-immunoprecipitate NF–YC3 and NF–YC4 from total plant protein extracts (Figure 7b). Although we do not currently have an antibody for NF–YC9, all available genetic and Y2H data strongly suggests it should also be a component of these floral promoting, in vivo complexes.

image

Figure 7.  NF–YC3, NF–YC4 and NF–YC9 physically interact with NF–YB2, NF–YC3 and CO. (a) Yeast two-hybrid (Y2H) analysis of interactions between the Gal4 DNA binding domain (DBD):NF–YB2 and DBD:NF–YB3, and Gal4 activation domain (AD):NF–YC3, AD:NF–YC4 and AD:NF–YC9 fusions: yeast growth shown on uracil-deficient plates and additionally confirmed with both His and LacZ reporter genes. EV, empty vector control shows that neither NF–YB2 nor NF–YB3 autoactivate the reporter gene. (b) Protein blot of in vivo interaction between NF–YB2, NF–YB3, NF–YC3 and NF–YC4. (c) Y2H control plate for DBD:NF–YC3, DBD:NF–YC4 and DBD:NF–YC9 constructs: none autoactivate the reporter. + and – lines represent the manufacturer’s controls (ProQuest; Invitrogen). (d) Y2H interaction tests between NF–YB2, NF–YB3, NF–YC3, NF–YC4 and NF–YC9 against full-length CO: six independent colonies tested per interaction.

Download figure to PowerPoint

To examine NF–Y by CO interactions, we created both full-length DBD:CO and AD:CO constructs. Because DBD:CO constructs strongly activated all reporter genes in our Y2H system (ProQuest; Invitrogen, http://www.invitrogen.com), we examined DBD:NF–Y by AD:CO interactions (Figure 7c–d). In the context of a URA3 reporter gene, the previously described, moderate NF–YC3 autoactivation was not a problem for this analysis (Figure 7c). Using full-length CO, our results differed from Wenkel et al. (2006) in that we consistently detected no interaction with either NF–YB2 or NF–YB3 (Figure 7d). This was true for three different reporter genes (Figure S2). Additionally, we found that full-length CO interacted strongly and consistently with all three NF–YC proteins (Figure 7d).

Because we could readily detect NF–YB2/3 by NF–YC3/4/9 interactions (Figure 7a), our DBD:NF–YB constructs appeared to be functioning properly. To further examine possible NF–YB and NF–YC interactions with CO/COL proteins, we performed fairly extensive library screening with DBD:NF–YB2, DBD:NF–YB3 and DBD:NF–YC9. For NF–YC9 we sequenced 167 putative interacting clones and found that 135 (81%) represented various NF–YB-expressing clones. Of the remaining 32 interactors, 22 (13%) were COL family members. Identical, albeit less extensive, library screening with NF–YB2 and NF–YB3 did not identify any COL family members, although 24/51 (47%) putative interacting proteins represented various NF–YCs. Therefore, from the pool of NF–YC-interacting COL proteins, we chose three (COL3, COL5 and COL13) to directly test for Y2H interactions with NF–YB2 and NF–YB3 (Figure S2). Once again, we did not detect any NF–YB2 or NF–YB3 interaction with CO or these related family members, although we readily detected interactions with appropriate controls. Collectively, our data suggest that all of the floral-promoting NF–YB and NF–YC subunits can form complexes in vivo, and that CO is most likely recruited into these complexes by direct interaction with the NF–YC components. Nevertheless, we cannot rule out the possibility that, in the context of a plant cell, NF–YB2 and NF–YB3 might interact with COL family members.

NF–YC3, NF–YC4 and NF–YC9 interactions with CO primarily involve FT regulation

NF–Y and CO physically interact and have clear genetic roles in FT transcriptional control. To identify other possible targets of the NF–YC3/4/9-CO transcriptional complex, we conducted Affymetrix microarray experiments and compared the expression profiles of 10-day-old, whole, aboveground, LD-grown seedlings for Col–0, nfyc triple and co-sail genetic backgrounds (triplicate biological replicates for each background).

Compared with parental controls (Col–0), the nfyc triple mutant had five upregulated and six downregulated genes (Table S1; misregulation defined as expression difference >1.5-fold, Benjamini–Hochberg false discovery rate-corrected anova, P ≤ 0.05). Using the Benjamini–Hochberg corrected P–value calculations, only one gene was simultaneously downregulated in nfyc triple and co-sail mutants: FT. No genes were found to overlap in the nfyc triple and co-sail upregulated sets. We note that the Benjamini–Hochberg correction is fairly conservative, potentially resulting in false negatives. Furthermore, tissue-specific transcriptional profiling might reveal additional common targets that are diluted when examining whole seedlings. It is interesting to note that co-sail had many more significantly upregulated (37 in total) than downregulated genes (three in total; Table S1). This data suggests that CO may also have undiscovered negative regulatory roles. We conclude that the simplest explanation for the nfyc and co-sail flowering phenotype is coordinated control of FT by an NF–YC/CO complex.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We have provided evidence that NFYC3, NFYC4 and NFYC9 are additively necessary for the proper photoperiod-dependant induction of flowering in Arabidopsis thaliana. Furthermore, CO function in the transcriptional activation of FT requires these NFYC. It was previously shown that NFYB2 and NFYB3 additively regulate FT expression, and are required for photoperiod-dependent initiation of flowering (Cai et al., 2007; Kumimoto et al., 2008). All three NF–YC proteins described here can physically interact with both floral-promoting NF–YB proteins, establishing at least six possible complexes for CO-mediated activation of FT expression.

Perhaps as a response to various environmental conditions, such as water and nutrient availability, NF–Y combinatorial diversity could provide unique platforms for the fine-tuning of flowering time and other processes. The fact that NF–Y complexes are also involved in drought resistance (Nelson et al., 2007; Li et al., 2008) further hints at a potential interaction between various environmental conditions and developmental outcomes. In addition to subtle changes in the positive activation of flowering time, we propose that NF–Y subunit heterogeneity at a given promoter might also provide antagonistic gene regulation: i.e. there may be both positive and negative NF–Y complexes competing for regulation of the same promoter.

In the plant lineage, there is extensive NF–Y heterogeneity, and, by extension, many possibilities for differential regulation at a single DNA binding location. For example, although reported loss-of-function phenotypes suggest that NF–Ys are positive regulators of FT expression (this work and in Kumimoto et al., 2008), it was also previously demonstrated that overexpression of NF–YA1, NF–YA4 and NF–YB1 reduced FT expression and delayed flowering (Wenkel et al., 2006; Nelson et al., 2007). Therefore, a subset of NF–Y are predicted to positively regulate FT expression, whereas another subset can act as negative regulators. One trivial explanation is that overexpression of a subset of NF–Y leads to dominant negative phenotypes: i.e. these particular NF–Y family members are not normally involved in flowering, and, when ectopically expressed, can interfere with floral-promoting complexes. Whether or not these NF–Y-overexpression phenotypes are biologically relevant, it is clear that NF–Ys have the ability to both positively and negatively regulate FT. What role CO plays in these complexes remains unclear.

CO belongs to a plant-specific, approximately 35-member family characterized by the CCT domain. Recent studies suggested that CO can physically interact with both NF–YB and NF–YC proteins via the CCT domain (Ben-Naim et al., 2006; Wenkel et al., 2006). In the present experiments, we were unable to replicate the CO–NF–YB2 interaction (Wenkel et al., 2006), but we were also examining full-length proteins (as opposed to the CCT domain of CO). Thus, it remains possible that we do not detect interactions between NF–YB2 and full-length CO in our Y2H system because appropriate conformational changes in CO only take place in the plant cell. Regardless of this possibility, we clearly and consistently measured interactions between full-length CO and NF–YC3, NF–YC4 and NF–YC9. In yeast chromatin immunoprecipitation experiments, Ben-Naim et al. (2006) demonstrate that HAP5 (from tomato or yeast) is required to recruit TCOL1 to canonical yeast CCAAT-containing promoters. The simplest, although not only, explanation for these data is that CO is recruited to NF–Y trimers via the NF–YC subunit.

Interestingly, CCT domains show amino acid sequence similarity to the DNA binding domain of NF–YA proteins (Wenkel et al., 2006; Distelfeld et al., 2009a,b; Siefers et al., 2009). This sequence similarity, coupled with the finding that NF–YA overexpression can cause late flowering, led Wenkel et al. (2006) to speculate that CO and NF–YA compete for the occupancy of NF–YB/C dimers. In theory, this would suggest floral-inhibiting NF–YA/B/C complexes as antagonists of floral-promoting CO/NF–YB/C complexes.

Co-opting NF–YB/C dimers for novel functions might be a common outcome in the plant lineage. For example, the rice (O. sativa) MADS box domain protein OsMADS18 interacts with OsNF–YB1 and OsNF–YB1/NF–YC dimers (Masiero et al., 2002). Because of a naturally occurring point mutation, OsNF–YB1-containing dimers cannot interact with NF–YA proteins. Thus, OsMADS18/OsNF–YB1/NF–YC trimeric complexes also cannot bind CCAAT box sequences (Masiero et al., 2002). Likewise, CO/NF–YB/C trimeric complexes would not be expected to bind CCAAT sequences in the FT promoter. This is because the essential histidine residues necessary for NF–Y complex binding to DNA are located within the NF–YA subunit (Xing et al., 1993), and these residues are not shared by CO (Siefers et al., 2009).

It was recently reported that bZIP67, LEAFY COTYLEDON 1 (LEC1, NF–YB9), LEC1-LIKE (NF–YB6) and NF–YC2 can interact to promote the expression of two abcisic acid-responsive promoters (Yamamoto et al., 2009). Consistent with the idea that NF–YA proteins can act as negative regulatory factors, inclusion of NF–YA4, NF–YA5, NF–YA7 and NF–YA9 in the same reporter assays interfered with gene activation. Thus, NF–Y complexes that extend beyond the canonical NF–YA/B/C trimer established in animal systems may be commonplace in the plant lineage.

An alternative hypothesis to antagonism between CO and NF–YA is that CO interacts with the entire NF–Y heterotrimer (NF–YA subunit included), and functions as an activation domain. In our Y2H experiments, we found that full-length CO was a very strong transcriptional activator when bound to a DNA binding domain. CO interaction with complete NF–Y complexes would be analogous to yeast where an orthologous NF–Y trimer (HAP2, HAP3 and HAP5) binds DNA, whereas an additional subunit, called HAP4, provides the transcriptional activation domain (Forsburg and Guarente, 1989). If this scenario were correct, one would predict that loss-of-function mutants in appropriate NF–YA genes would show late-flowering phenotypes. Because of the high sequence similarity between Arabidopsis NF–YA proteins, higher order mutants are also likely to be necessary to test this scenario and elucidate their function in flowering (Li et al., 2008; Siefers et al., 2009).

NF–Ys contribute to many aspects of plant growth and development, but no complete NF–Y complex has been identified for a specific process. Because of the high level of similarity and evolutionary conservation between NF–Y proteins in human and plants, we assume that they typically function in an analogous manner. Nevertheless, because of the numerical expansion of NF–Y in the plant lineage, many interesting functional nuances are possible. In our opinion, describing NF–YA functions is currently a particularly important area for understanding these transcription factors. Now that NF–YB and NF–YC subunits required for CO function in photoperiod-dependent flowering have been discovered, we can proceed to this next step with increased confidence.

Experimental Procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Phylogenetic analysis

The phylogenetic tree shown in Figure 1 was generated using the publicly available software package mega 4 (Tamura et al., 2007). This tree was adapted from Siefers et al. (2009), and is included here again for ease of visualizing the NF–YC relationships.

Plant cultivation, GUS staining and flowering-time experiments

All plants were Col-0 ecotype and were grown at 23°C in standard LD (16-h light/8-h dark) and SD (8-h light/16-h dark) conditions in Conviron (model ATC13; http://www.conviron.com) growth chambers or a custom walk-in chamber. Plants were grown in media containing equal parts Farfard C2 Mix and Metromix 200 supplemented with 40 g Marathon pesticide and dilute Peters fertilizer (NPK 20:20:20). Plants were watered throughout with dilute fertilizer (Peters NPK 20:20:20 at ∼1/10th of the recommended feeding levels).

GUS staining was performed on the first set of true leaves of 10-day-old plants as previously described (Malamy and Benfey, 1997). Leaf images were visualized on an Olympus BX41 microscope (http:http://www.olympusamerica.com) and recorded with a Diaginic Insight 2 Megapixel Color Mosiac CCD camera and spot software (v4.6; http://www.diaginc.com).

Leaf number at flowering was determined by counting all primary rosette leaves and cauline leaves just after the first flowers opened (Onouchi et al., 2000). None of the mutant lines exhibited enhanced developmental rates.

DNA manipulations and transgenic plants

Unless otherwise noted, all clones were generated by PCR using Pfu Ultra II (cat#600670-51; Agilent Technologies, http://www.genomics.agilent.com) and ligation into the GATEWAY™ entry vector pENTR/D-TOPO (cat#45-0218; Invitrogen, http://www.invitrogen.com). Full-length (minus the stop codon) NF–YB2, NF–YB3, NF–YC3, NF–YC4, and NF–YC9 were derived from Col-0 cDNA populations by standard methods (cloning primers available upon request). The CO cDNA construct used in Y2H analyses was obtained from a pENTR223 construct available from ABRC (stock#GC105432; http://abrc.osu.edu). Full-length CO, used for in planta overexpression, was amplified from genomic DNA and includes introns. The complementation constructs for NF–YC3, NF–YC4 and NF–YC9 contain the genomic region corresponding to −1000 bp from the ATG through the genomic coding region (minus the stop codon). All constructs were sequenced and found to be identical to the expected sequences found at The Arabidopsis Information Resource (Swarbreck et al., 2008, http://www.arabidopsis.org). Subsequent clones into plant expression and Y2H vectors were created using the GATEWAY™ LR Clonease II reaction kit (cat#56485; Invitrogen). Plant expression complementation constructs were cloned into pEarleyGate301 (stock#CD3-692; ABRC; Earley et al., 2006). Epitope tagged, overexpression constructs of NF–YB2, NF–YB3 and CO were cloned into pEarleyGate101 (stock#CD3-683; ABRC; Earley et al., 2006). Promoter GUS fusion lines were previously described by Siefers et al. (2009). All primer sequences are available upon request. All plant transformations were performed by the Agrobacterium-mediated floral-dipping method, as previously described (Bechtold et al., 1993; Clough and Bent, 1998).

qRT-PCR analyses

Total RNA was isolated from 10-day-old seedlings grown under LD conditions using the Qiagen RNeasy Plant Mini Kit (cat#74904; http://www.qiagen.com). Samples were collected either over a diurnal time course or 15 h after the lights were switched on. Isolated RNA was DNase treated (cat#AM2222; Ambion, http://www.ambion.com) and first-strand cDNA synthesis was performed using the Ambion RETROscript kit (cat#1710) with supplied oligo dT primers. qRT-PCR was performed as previously described, except we used an Applied Biosystems (http://www.AppliedBiosystems.com) Prism 7500 analyzer and the Fermentas Maxima SYBR Green qPCR Master Mix (cat#K0222; http://www.fermentas.com; Kumimoto et al., 2008). For each genotype, we analyzed three or four independent, biological replicates. Ten plants were combined for each biological replicate. All samples were normalized to the constitutively expressed gene At2g32170 as previously described (Czechowski et al., 2005). Sample comparisons were performed by the 2(−ΔΔCT) method (Livak and Schmittgen, 2001), and errors (standard deviation) were computed as previously described (Nordgard et al., 2006).

Microarray analysis

All microarray data was collected and recorded in compliance with the minimum information about a microarray experiment (MIAME) specification (Brazma et al., 2001). Unprocessed microarray data (.cel and.chp files) and detailed experimental conditions are publicly available at the NASCarrays website (http://affymetrix.arabidopsis.info/narrays/experimentbrowse.pl) using the experiment identifier NASCARRAYS-440. Briefly, plants for microarray analysis were grown for 10 days under standard LD conditions; aboveground tissue was harvested 14 h after the chamber lights were switched on. Plant density and spacing were equivalent for all biological replicates performed. For each plant genotype – Col-0 parental control, nf–yc triple and co-sail– total RNA was collected from 10 plants per biological replicate using the Qiagen RNeasy Plant Mini Kit. Total RNA from three biological replicates was collected for a total of nine independent RNA preparations and microarray hybridizations. Isolated total RNA was sent to the Nottingham Arabidopsis Stock Center (NASC) for quality control and hybridization to Affymetrix ATH-1 microarray chips (http://www.affymetrix.com). The resulting data were analyzed with the GeneSpringGX 7.3.1 software package (Agilent Technologies, http://www.agilent.com). Data were GCRMA normalized to the 50th percentile during data importation into GeneSpring. Variances were calculated using the cross-gene error model with replicates. With the cross-gene error model active, we performed anova with Benjamini–Hochberg false discovery rate-corrected P values (P < 0.05). Significantly misregulated genes, as determined by anova, were further filtered by volcano plot analysis (< 0.05, fold change >1.5) in pairwise comparisons between the parental control and each mutant genotype.

Protein blots, antibodies and co-immunoprecipitation

In Figure 1d, total protein was extracted from 10–12-day-old seedlings by grinding fresh tissue in sucrose buffer (20 mm Tris, pH 8.0, 0.33 m sucrose, 1 mm EDTA, pH 8.0; add fresh 5 mm DTT and 1X Sigma Protease Inhibitor Cocktail; cat#P9599; http://www.sigmaaldrich.com). Proteins were visualized by standard SDS-PAGE methods (15% gel). Western blotting was also performed by standard methods (as described in the ECL Plus Reagent protocol; cat#RPN2132; GE Healthcare/Amersham, http://www.gelifesciences.com). Briefly, we used polyvinylidene fluoride (PVDF) membranes (Hybond-P; cat#RPN2020F; GE Healthcare/Amersham Biosciences) and blocked with tris-buffered saline containing 0.05% Triton X-100 (TBS-T) plus 5% skimmed milk. NF–YC3 and NF–YC4 primary antibodies (described below) in TBS-T plus 1% skimmed milk were followed by hybridization to goat anti-IG-Y secondary antibodies (Santa Cruz Biotechnology, cat#SC-2428, http://www.scbt.com). After each antibody incubation (2-h primary and 1-h secondary), blots were washed three times (15 min each) with excess TBS-T. Protein blots were visualized using the horseradish-peroxidase based ECL Plus reagent on a GE Storm 960 Phosphorimager.

IG-Y antibodies specific for NF–YC3 and NF–YC4 were generated by injecting chickens with pairs of protein-specific synthetic peptide antigens. Peptide design and the production of initial, non-affinity purified IG-Y antibodies was provided as a service by Aves Labs (http://www.aveslab.com). For NF–YC3, chickens were injected with the synthetic peptides CZTTTPTGSDHPAYHQIHQQ and CZQQPGPEQQDPDN. For NF–YC4, chickens were injected with the synthetic peptides CZEEIKEEEDAASA and CZTSVYPPGSAVTTV.

IG-Y antibodies from Aves Labs were purified with the Thermo Scientific/Pierce SulfoLink Immobilization Kit for Peptides (cat#44999; http://www.piercenet.com), as described by the manufacturer. Antibodies were eluted from the columns by 0.2 m glycine-HCL (pH 2.5) directly into 1.0 m Tris-HCL (pH 8.0). Equal volumes of 100% saturated, cold ammonium sulfate (pH 7.2) were added and antibodies were allowed to precipitate overnight at 4°C. The resulting precipitates were concentrated by centrifugation (2000 g for 30 min) and resuspended in 1X PBS. Resuspended antibodies were dialyzed against large volumes of cold PBS for ∼24 h with three buffer changes.

NF–YC3 and NF–YC4 antibodies detect their targets at 1:5000–10 000 and 1:2000 dilutions, respectively. Goat anti IG-Y was used at 1:5000 in all experiments. NF–YC4 antibodies require longer incubations for effective signal visualization (2 h at room temperature (24°C) or overnight at 4°C). We will accommodate requests for samples of these affinity purified antibodies, but we reserve the right to limit their future distribution such that we retain sufficient working stocks.

For the co-immunoprecipitation in Figure 5, we used HEPES lysis buffer [50 mm HEPES, pH 7.5, 150 mm NaCl, 0.5% NP-40, 10% glycerol, add fresh 1 mm phenylmethylsulfonyl fluoride (PMSF), 5 μg μl−1 aproptinin, and 5 μg μl−1 pepstatin] to extract total proteins. For each sample, 1 mg of total protein was brought to a final volume of 500 μl in a 1.5-ml Eppendorf tube, and 5 μl of anti-HA antibody was added (cat#AB9110; ABCAM, http://www.abcam.com). Proteins and antibodies were incubated on a rotator at 4°C for 4 h. Pre-equilibrated (with HEPES lysis buffer) Sepharose A agarose beads (70 μl bead volume in ∼100 μl) were then incubated with the protein/antibody mixture for 4 h. Sepharose A agarose beads were washed by three rounds of full-speed centrifugation at 4°C, supernatant removal and resuspension in 1 ml wash buffer (HEPES buffer above, except 50 mm NaCl, 0.1% NP-40 and no glycerol). The final pellet was incubated with 100 μl of standard 2X SDS sample buffer for 5 min at 90°C, and then stored at −80°C until western analysis (as described above). NF–YB/NF–YC co-immunoprecipitation was repeated three times with similar results.

Yeast two-hybrid analysis

Gateway™ entry clones (above) were transferred into the ProQuest Two-Hybrid System (Invitrogen) vectors pDEST22 and pDEST32. All interaction tests between full-length proteins were performed exactly as described in the ProQuest manual. The Y2H library we screened was an equal parts mixture of four previously described, Gateway™-ready libraries made from both hormone-treated and untreated seedlings, flowers, developing seeds and primary leaves (Burkle et al., 2005).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The authors thank Zack Nimchuk and Jeff Dangl for their critical reading of the original manuscript, Valarie McMurtry and Jennifer McAvoy for technical support, Franziska Turck and Alexander Heyl for providing the co-sail seeds and Y2H libraries, respectively, and anonymous reviewers for their helpful comments. Funding was provided by the University of Oklahoma Research Council (122748300), the Oklahoma Center for the Advancement of Science and Technology (PSB07-012) and the National Science Foundation (IOS-0920258).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • An, H., Roussot, C., Suarez-Lopez, P. et al. (2004) CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development, 131, 36153626.
  • Bechtold, N., Ellis, J. and Pelletier, G. (1993) In planta Agrobacterium-mediated gene transfer by infiltration of Arabidopsis thaliana plants. C. R. Acad. Sci. Paris, 316, 11941199.
  • Ben-Naim, O., Eshed, R., Parnis, A., Teper-Bamnolker, P., Shalit, A., Coupland, G., Samach, A. and Lifschitz, E. (2006) The CCAAT binding factor can mediate interactions between CONSTANS-like proteins and DNA. Plant J. 46, 462476.
  • Brazma, A., Hingamp, P., Quackenbush, J. et al. (2001) Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365371.
  • Burkle, L., Meyer, S., Dortay, H., Lehrach, H. and Heyl, A. (2005) In vitro recombination cloning of entire cDNA libraries in Arabidopsis thaliana and its application to the yeast two-hybrid system. Funct. Integr. Genomics, 5, 175183.
  • Cai, X., Ballif, J., Endo, S., Davis, E., Liang, M., Chen, D., DeWald, D., Kreps, J., Zhu, T. and Wu, Y. (2007) A putative CCAAT-binding transcription factor is a regulator of flowering timing in Arabidopsis. Plant Physiol. 145, 98105.
  • Chen, N.Z., Zhang, X.Q., Wei, P.C., Chen, Q.J., Ren, F., Chen, J. and Wang, X.C. (2007) AtHAP3b plays a crucial role in the regulation of flowering time in Arabidopsis during osmotic stress. J. Biochem. Mol. Biol. 40, 10831089.
  • Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735743.
  • Combier, J.P., Frugier, F., de Billy, F. et al. (2006) MtHAP2-1 is a key transcriptional regulator of symbiotic nodule development regulated by microRNA169 in Medicago truncatula. Genes Dev. 20, 30843088.
  • Corbesier, L., Vincent, C., Jang, S. et al. (2007) FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis. Science, 316, 10301033.
  • Czechowski, T., Stitt, M., Altmann, T., Udvardi, M.K. and Scheible, W.R. (2005) Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol. 139, 517.
  • Distelfeld, A., Li, C. and Dubcovsky, J. (2009a) Regulation of flowering in temperate cereals. Curr. Opin. Plant Biol. 12, 178184.
  • Distelfeld, A., Tranquilli, G., Li, C., Yan, L. and Dubcovsky, J. (2009b) Genetic and molecular characterization of the VRN2 loci in tetraploid wheat. Plant Physiol. 149, 245257.
  • Earley, K.W., Haag, J.R., Pontes, O., Opper, K., Juehne, T., Song, K. and Pikaard, C.S. (2006) Gateway-compatible vectors for plant functional genomics and proteomics. Plant J. 45, 616629.
  • Edwards, D., Murray, J.A. and Smith, A.G. (1998) Multiple genes encoding the conserved CCAAT-box transcription factor complex are expressed in Arabidopsis. Plant Physiol. 117, 10151022.
  • FitzGerald, P.C., Shlyakhtenko, A., Mir, A.A. and Vinson, C. (2004) Clustering of DNA sequences in human promoters. Genome Res. 14, 15621574.
  • Forsburg, S.L. and Guarente, L. (1989) Identification and characterization of HAP4: a third component of the CCAAT-bound HAP2/HAP3 heteromer. Genes Dev. 3, 11661178.
  • Frontini, M., Imbriano, C., Manni, I. and Mantovani, R. (2004) Cell cycle regulation of NF–YC nuclear localization. Cell Cycle, 3, 217222.
  • Goda, H., Nagase, T., Tanoue, S., Sugiyama, J., Steidl, S., Tuncher, A., Kobayashi, T., Tsukagoshi, N., Brakhage, A.A. and Kato, M. (2005) Nuclear translocation of the heterotrimeric CCAAT binding factor of Aspergillus oryzae is dependent on two redundant localising signals in a single subunit. Arch. Microbiol. 184, 93100.
  • Gusmaroli, G., Tonelli, C. and Mantovani, R. (2001) Regulation of the CCAAT-Binding NF–Y subunits in Arabidopsis thaliana. Gene, 264, 173185.
  • Gusmaroli, G., Tonelli, C. and Mantovani, R. (2002) Regulation of novel members of the Arabidopsis thaliana CCAAT-binding nuclear factor Y subunits. Gene, 283, 4148.
  • Imaizumi, T. and Kay, S.A. (2006) Photoperiodic control of flowering: not only by coincidence. Trends Plant Sci. 11, 550558.
  • Jaeger, K.E. and Wigge, P.A. (2007) FT protein acts as a long-range signal in Arabidopsis. Curr. Biol. 17, 10501054.
  • Jang, S., Marchal, V., Panigrahi, K.C., Wenkel, S., Soppe, W., Deng, X.W., Valverde, F. and Coupland, G. (2008) Arabidopsis COP1 shapes the temporal pattern of CO accumulation conferring a photoperiodic flowering response. EMBO J. 27, 12771288.
  • Kobayashi, Y. and Weigel, D. (2007) Move on up, it’s time for change--mobile signals controlling photoperiod-dependent flowering. Genes Dev. 21, 23712384.
  • Koornneef, M., Hanhart, C.J. and van der Veen, J.H. (1991) A genetic and physiological analysis of late flowering mutants in Arabidopsis thaliana. Mol. Gen. Genet. 229, 5766.
  • Kumimoto, R.W., Adam, L., Hymus, G.J., Repetti, P.P., Reuber, T.L., Marion, C.M., Hempel, F.D. and Ratcliffe, O.J. (2008) The Nuclear Factor Y subunits NF–YB2 and NF–YB3 play additive roles in the promotion of flowering by inductive long-day photoperiods in Arabidopsis. Planta, 228, 709723.
  • Kwong, R.W., Bui, A.Q., Lee, H., Kwong, L.W., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (2003) LEAFY COTYLEDON1-LIKE defines a class of regulators essential for embryo development. Plant Cell, 15, 518.
  • Li, W.X., Oono, Y., Zhu, J. et al. (2008) The Arabidopsis NFYA5 Transcription Factor Is Regulated Transcriptionally and Posttranscriptionally to Promote Drought Resistance. Plant Cell, 20, 22382251.
  • Liu, L.J., Zhang, Y.C., Li, Q.H., Sang, Y., Mao, J., Lian, H.L., Wang, L. and Yang, H.Q. (2008) COP1-mediated ubiquitination of CONSTANS is implicated in cryptochrome regulation of flowering in Arabidopsis. Plant Cell, 20, 292306.
  • Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 25, 402408.
  • Lotan, T., Ohto, M., Yee, K.M., West, M.A., Lo, R., Kwong, R.W., Yamagishi, K., Fischer, R.L., Goldberg, R.B. and Harada, J.J. (1998) Arabidopsis LEAFY COTYLEDON1 is sufficient to induce embryo development in vegetative cells. Cell, 93, 11951205.
  • Maity, S.N. and de Crombrugghe, B. (1998) Role of the CCAAT-binding protein CBF/NF–Y in transcription. Trends Biochem. Sci. 23, 174178.
  • Maity, S.N., Sinha, S., Ruteshouser, E.C. and de Crombrugghe, B. (1992) Three different polypeptides are necessary for DNA binding of the mammalian heteromeric CCAAT binding factor. J. Biol. Chem. 267, 1657416580.
  • Malamy, J.E. and Benfey, P.N. (1997) Organization and cell differentiation in lateral roots of Arabidopsis thaliana. Development, 124, 3344.
  • Mantovani, R. (1999) The molecular biology of the CCAAT-binding factor NF–Y. Gene, 239, 1527.
  • Masiero, S., Imbriano, C., Ravasio, F., Favaro, R., Pelucchi, N., Gorla, M.S., Mantovani, R., Colombo, L. and Kater, M.M. (2002) Ternary complex formation between MADS-box transcription factors and the histone fold protein NF–YB. J. Biol. Chem. 277, 2642926435.
  • Mathieu, J., Warthmann, N., Kuttner, F. and Schmid, M. (2007) Export of FT protein from phloem companion cells is sufficient for floral induction in Arabidopsis. Curr. Biol. 17, 10551060.
  • Matuoka, K. and Chen, K.Y. (2002) Transcriptional regulation of cellular ageing by the CCAAT box-binding factor CBF/NF–Y. Ageing Res. Rev. 1, 639651.
  • Nelson, D.E., Repetti, P.P., Adams, T.R. et al. (2007) Plant nuclear factor Y (NF–Y) B subunits confer drought tolerance and lead to improved corn yields on water-limited acres. Proc. Natl Acad. Sci. USA, 104, 1645016455.
  • Nordgard, O., Kvaloy, J.T., Farmen, R.K. and Heikkila, R. (2006) Error propagation in relative real-time reverse transcription polymerase chain reaction quantification models: the balance between accuracy and precision. Anal. Biochem. 356, 182193.
  • Onouchi, H., Igeno, M.I., Perilleux, C., Graves, K. and Coupland, G. (2000) Mutagenesis of plants overexpressing CONSTANS demonstrates novel interactions among Arabidopsis flowering-time genes. Plant Cell, 12, 885900.
  • Putterill, J., Robson, F., Lee, K., Simon, R. and Coupland, G. (1995) The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell, 80, 847857.
  • Redei, G.P. (1962) Supervital Mutants of Arabidopsis. Genetics, 47, 443460.
  • Robson, F., Costa, M.M., Hepworth, S.R., Vizir, I., Pineiro, M., Reeves, P.H., Putterill, J. and Coupland, G. (2001) Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants. Plant J. 28, 619631.
  • Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F. and Coupland, G. (2000) Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science, 288, 16131616.
  • Siefers, N., Dang, K.K., Kumimoto, R.W., Bynum, W.E.T., Tayrose, G. and Holt, B.F. 3rd. (2009) Tissue-specific expression patterns of Arabidopsis NF–Y transcription factors suggest potential for extensive combinatorial complexity. Plant Physiol. 149, 625641.
  • Sinha, S., Kim, I.S., Sohn, K.Y., de Crombrugghe, B. and Maity, S.N. (1996) Three classes of mutations in the A subunit of the CCAAT-binding factor CBF delineate functional domains involved in the three-step assembly of the CBF-DNA complex. Mol. Cell. Biol. 16, 328337.
  • Steidl, S., Tuncher, A., Goda, H., Guder, C., Papadopoulou, N., Kobayashi, T., Tsukagoshi, N., Kato, M. and Brakhage, A.A. (2004) A single subunit of a heterotrimeric CCAAT-binding complex carries a nuclear localization signal: piggy back transport of the pre-assembled complex to the nucleus. J. Mol. Biol. 342, 515524.
  • Stephenson, T.J., McIntyre, C.L., Collet, C. and Xue, G.P. (2007) Genome-wide identification and expression analysis of the NF–Y family of transcription factors in Triticum aestivum. Plant Mol. Biol. 65, 7792.
  • Suarez-Lopez, P., Wheatley, K., Robson, F., Onouchi, H., Valverde, F. and Coupland, G. (2001) CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature, 410, 11161120.
  • Suzuki, M., Wang, H.H. and McCarty, D.R. (2007) Repression of the LEAFY COTYLEDON 1/B3 regulatory network in plant embryo development by VP1/ABSCISIC ACID INSENSITIVE 3-LIKE B3 genes. Plant Physiol. 143, 902911.
  • Swarbreck, D., Wilks, C., Lamesch, P. et al. (2008) The Arabidopsis Information Resource (TAIR): gene structure and function annotation. Nucleic Acids Res. 36, D1009D1014.
  • Tamaki, S., Matsuo, S., Wong, H.L., Yokoi, S. and Shimamoto, K. (2007) Hd3a protein is a mobile flowering signal in rice. Science, 316, 10331036.
  • Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 15961599.
  • Thirumurugan, T., Ito, Y., Kubo, T., Serizawa, A. and Kurata, N. (2008) Identification, characterization and interaction of HAP family genes in rice. Mol. Genet. Genomics, 279, 279289.
  • Tuncher, A., Sprote, P., Gehrke, A. and Brakhage, A.A. (2005) The CCAAT-binding complex of eukaryotes: evolution of a second NLS in the HapB subunit of the filamentous fungus Aspergillus nidulans despite functional conservation at the molecular level between yeast, A.nidulans and human. J. Mol. Biol. 352, 517533.
  • Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach, A. and Coupland, G. (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science, 303, 10031006.
  • Warpeha, K.M., Upadhyay, S., Yeh, J., Adamiak, J., Hawkins, S.I., Lapik, Y.R., Anderson, M.B. and Kaufman, L.S. (2007) The GCR1, GPA1, PRN1, NF–Y signal chain mediates both blue light and abscisic acid responses in Arabidopsis. Plant Physiol. 143, 15901600.
  • Wenkel, S., Turck, F., Singer, K., Gissot, L., Le Gourrierec, J., Samach, A. and Coupland, G. (2006) CONSTANS and the CCAAT box binding complex share a functionally important domain and interact to regulate flowering of Arabidopsis. Plant Cell, 18, 29712984.
  • Wilson, R.N., Heckman, J.W. and Somerville, C.R. (1992) Gibberellin Is Required for Flowering in Arabidopsis thaliana under Short Days. Plant Physiol. 100, 403408.
  • Xing, Y., Fikes, J.D. and Guarente, L. (1993) Mutations in yeast HAP2/HAP3 define a hybrid CCAAT box binding domain. EMBO J. 12, 46474655.
  • Yamamoto, A., Kagaya, Y., Toyoshima, R., Kagaya, M., Takeda, S. and Hattori, T. (2009) Arabidopsis NF–YB subunits LEC1 and LEC1-LIKE activate transcription by interacting with seed-specific ABRE-binding factors. Plant J. 58, 843856.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1.35S:CO-YFP/HA Lines 16 and 100 can drive early flowering in the context of functional NF-YC3, 4, and 9. Numbers in parentheticals after genotypes represent the line numbers described in the main text and Figure 6. 35S:CO/- 3/- 4/- 9/- is shorthand for hemizygous 35S:CO-YFP/HA transgene in the NF-YC3/nf-yc3-1 NF-YC4/nf-yc4-1 NF-YC9/nf-yc9-1 triple heterozygous background. Prior to statistical analysis, data were transformed (1/(Y)) to correct for unequal standard deviations between groups. ***represents significant differences (P < 0.001) in pairwise comparisons to Col-0 (Bonferroni multiple comparison test) after significant anova (P < 0.0001).

Figure S2. NF-YB2 and 3 physically interact with NF-YC3, 4, 9, and At3g25680 by Y2H, but not with full length CO and COL proteins. (a) NF-YC3, 4, and 9 interactions with various full length COL proteins. AD, Activation Domain fusion; DBD, DNA binding domain fusion. Based on our experiences, reporter gene stringencies are aligned from highest (URA) to lowest (HIS). For example, note that NF-YC interactions are generally weaker with COL5 and COL13 as evidenced by the URA plate. +, +/−, and − yeast streaks represent the manufacturer’s (Invitrogen) provided controls and should be compared to the experimental interactions on a plate by plate basis. (b) NF-YB2 and 3 interaction tests with CO and COL proteins. NF-YC3, 4, 9, and At3g25680 represent known positive interaction controls (At3g25680 is a protein of unknown function and was the most abundant non-NF-YC interactor in our library screen with 19/51 (37%) sequenced clones).

Table S1. Microsoft Excel file of misregulated genes in 10 day old, LD grown nf-yc triple and co-sail mutants.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
TPJ_4247_sm_figure-s1.tif16429KSupporting info item
TPJ_4247_sm_figure-s2.tif10059KSupporting info item
TPJ_4247_sm_table-s1.xls32KSupporting info item
TPJ_4247_sm_legends.doc22KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.