Characterization of FLOWERING LOCUS T1 (FT1) gene in Brachypodium and wheat

Abstract

The phase transition from vegetative to reproductive growth is a critical event in the life cycle of flowering plants. FLOWERING LOCUS T (FT) plays a central role in the regulation of this transition by integrating signals from multiple flowering pathways in the leaves and transmitting them to the shoot apical meristem. In this study, we characterized FT homologs in the temperate grasses Brachypodium distachyon and polyploid wheat using transgenic and mutant approaches. Downregulation of FT1 by RNAi was associated with a significant downregulation of the FT-like genes FT2 and FT4 in Brachypodium and FT2 and FT5 in wheat. In a transgenic wheat line carrying a highly-expressed FT1 allele, FT2 and FT3 were upregulated under both long and short days. Overexpression of FT1 caused extremely early flowering during shoot regeneration in both Brachypodium and hexaploid wheat, and resulted in insufficient vegetative tissue to support the production of viable seeds. Downregulation of FT1 transcripts by RNA interference (RNAi) resulted in non-flowering Brachypodium plants and late flowering plants (2-4 weeks delay) in wheat. A similar delay in heading time was observed in tetraploid wheat plants carrying mutations for both FT-A1 and FT-B1. Plants homozygous only for mutations in FT-B1 flowered later than plants homozygous only for mutations in FT-A1, which corresponded with higher transcript levels of FT-B1 relative to FT-A1 in the early stages of development. Taken together, our data indicate that FT1 plays a critical role in the regulation of flowering in Brachypodium and wheat, and that this role is associated with the simultaneous regulation of other FT-like genes. The differential effects of mutations in FT-A1 and FT-B1 on wheat heading time suggest that different allelic combinations of FT1 homoeologs could be used to adjust wheat heading time to improve adaptation to changing environments.

Figure 1. Schematic diagram of plasmids used in this study.

(A) Ubi::FT1 construct, (B) 35S::FT1:GFP construct. Overexpression constructs were developed in binary vectors pCAMBIA1300 and pGWB5. For Brachypodium FT1, the cDNA was cloned in both vectors. For wheat FT-B1, both the coding and genomic regions were cloned in pCAMBIA1300. (C and D) Ubi::FT1 _RNAi constructs. (C) The Brachypodium FT1 RNAi trigger was cloned in the pCAMBIA1300-based vector. (D) The FT1 RNAi trigger from T. monococcum was cloned in the pANDA-based vector. In both constructs, expression of the selected RNAi trigger was driven by the maize Ubiquitin promoter (Ubi). The black and gray triangles indicate the left (LB) and right (RB) border repeats respectively.

Figure 2. FT1 overexpression promotes floral organogenesis.

(A–C) Brachypodium 35S::BdFT1:GFP, (A) Shoot regeneration of non-transgenic control calli (B) direct spike formation from transformed calli and (C) rudimentary leaves associated with spikelet formation from transformed calli. (D–F) Wheat Ubi::cFT-B1. (D) Cluster of florets surrounded by rudimentary leaves in a transformed callus. (E) Different floral organs: lemma (Le), palea (Pa), pistil (Pi) and stamen (St). The additional organs seem to be glumes but it was difficult to determine because of the close clustering of multiple florets. (F) Anther with regions of non-viable pollen (blue color after pollen staining).

Figure 3. Silencing of FT1 by RNAi delays heading time.

(A) Heading was prevented in transgenic FT1 _RNAi Brachypodium and (B) delayed in FT1 _RNAi transgenic wheat. (C) Wheat transgenic plants at booting stage. (D) Some spikes had difficulty in emerging from the leaf sheath (leaf sheath opened manually in this picture). (E) Complete floral organs from transgenic wheat flowers (stigmas failed to open in some transgenic plants).

Figure 4. ft1 TILLING mutants.

(A) Interaction plot showing the effect of ft-A1 and ft-B1 ₂₆₃ mutations on heading time in a BC₁F₂ population segregating for both genes. Differences in heading time between the FT-A1 alleles were significant only for the homozygous ft-B1 ₂₆₃ mutant plants. (B) Transcript levels of FT-A1 and FT-B1 homoeologs in three-week-old wild type Kronos. The expression level of FT-B1 was significantly higher than that of the FT-A1 (P = 0.017). ACTIN was used as an internal control. Samples were harvested at 10∶00 a.m. Asterisks indicate P values of Student’s t-tests: * = P<0.05.

Figure 5. Phylogenetic analysis of FT-like genes in temperate grasses.

Phylogenetic analysis was performed using the full-length proteins. A neighbor-joining tree was constructed using pairwise deletions and 1,000 bootstrap iterations with the program MEGA 5.0 . The scale bar 0.05 represents 5% base substitution. Bootstrap numbers larger than 50 are shown in the respective nodes. To simplify the tree, only the wheat A-genome homoeologs of wheat were included. Accession information: BdFT1 (Bradi1g48830), BdFT2 (Bradi2g07070), BdFT3 (Bradi2g49795), BdFT4 (Bradi1g38150), BdFT5 (Bradi2g19670), BdFT6 (Bradi3g08890), HvFT1 (DQ100327), HvFT2 (DQ297407), HvFT3 (DQ411319), HvFT4 (DQ411320), HvFT5 (EF012202), HvFT6 (morex_contig_54196), OsFTL1 (Os01g11940), OsFTL2/Hd3a (Os06g06320), OsFTL3/RFT1 (Os06g06300), OsFTL8 (Os01g10590), OsFTL10 (Os05g44180), OsFTL12 (Os06g35940), OsFTL13 (Os02g13830), TaFT-A1 (CD881060), TaFT-A2 (BT009051), TaFT-A3 (IWGSC_1AL_913428), TaFT-A4 (IWGSC_2AS_5252557), TaFT-A5 (IWGSC_5AL_2803506), TaFT-A6 (IWGSC_6AS_4388307). HvFT6 sequence is from the International Barley Sequencing Consortium (IBSC, http://webblast.ipk-gatersleben.de/barley/).

Figure 6. Transcript levels of target genes in T₀ Brachypodium FT1 _RNAi lines.

(A) Comparison between the average of four transgenic plants (RNAi) and the wild type control using contrasts. (B–G) Comparison between individual transgenic lines and wild type using Dunnett’s test. (B) FT1, (C) FT2, (D) FT3, (E) FT4, (F) FT5, (G) VRN1 (gene regulated by FT1). ACTIN was used as the internal control. Samples were harvested at 12∶00 p.m., approximately one week after wild type control plants began to flower. Asterisks indicate P values: * = P<0.05; ** = P<0.01; *** = P<0.001.

Figure 7. Transcript levels of target genes in transgenic T₁ wheat RNAi plants and the non-transgenic wheat control.

(A) Comparison between the average of five transgenic plants (RNAi) and the wild type control using contrasts. (B–G) Comparison between individual transgenic lines and wild type using Dunnett’s test. (B) FT1, (C) FT2, (D) FT3, (E) FT4, (F) FT5, (G) VRN1 (gene regulated by FT1). ACTIN was used as the internal control. Samples were harvested at 4∶00 p.m., when the wild type controls began to flower. Asterisks indicate P values: * = P<0.05, ** = P<0.01, *** = P<0.001.

Figure 8. Expression of FT-like genes in FT1 _HOPE transgenic wheat.

The experiment was performed on transgenic lines (FT1 _HOPE) and the non-transgenic control Jagger. Leaf tissues were collected from two independent experiments, including plants grown under a (A) LD photoperiod for five weeks and (B) plants grown under a SD photoperiod for six weeks. ACTIN was used as the internal control. Samples were harvested at 10∶00 a.m. Asterisks indicate P values: * = P<0.05, ** = P<0.01, *** = P<0.001.