Abstract:
Onset of flowering is critical to the plant reproductive success. In plants such as Arabidopsis thaliana, in which flowering is promoted under long days, inductive conditions lead to the production of FLOWERING LOCUS T (FT) protein in leaves. This protein then serves as a florigen, which travels through the phloem to the shoot apex. There, it interacts with the FD transcription factor to directly activate different flower-identity genes. The floral promoting activity of FT is, however, antagonized by floral repressing TERMINAL FLOWER 1 (TFL1), expressed specifically at the shoot apex. Though opposite in function, both FT and TFL1 are members of the PEBP (phosphatidylethanolamine binding protein) family and share a high degree of structural and topological similarities.
Previous studies showed that TFL1 can function similarly to the floral promoting FT via a single amino acid change (H88Y), but the corresponding mutation on the FT backbone (Y85H) cannot convert FT into a floral repressing molecule. Later, exonic swapping between FT and TFL1 demonstrated that the motif encoded by exon 4B and 4C is required for FT function, while either exon 4B or 4C is sufficient for TFL1 activity. However, swapping of exon 4B between the FT paralogs in sugar beet, namely the floral repressing BvFT1 and floral promoting BvFT2, can only convert BvFT1 into BvFT2. Data supporting the conversion from BvFT2 to BvFT1 was not conclusive. Moreover, recent findings postulated the existence of a coactivator of FT, how the functionally critical residues of FT related to the coactivating mechanism remains elusive.
In my study, by combination of mutagenesis, in silico protein analysis and genetic studies, I showed that amino acid residues ultraconserved in Plantae are not necessarily crucial for the robust FT function, while most of the canonical positions do not determine the florigenic identity of FT. Instead, I identified six residues essential for the maintenance of a specific charge and bondings of two localized surfaces. Disturbance of either of these properties is sufficient to convert FT into a complete TFL1 mimic. Further, I demonstrated that the dominant negative phenotype is not caused by interrupting the interaction with previously identified components of FT floral activating complex, specific members of a plant-specific transcription factor family are likely to play a key role in differentiating between FT and TFL1 action. Altogether, my work revealed novel and surprising insights into both FT function and mode-of-action, thereby increased our understanding on the mechanism mediating the critical developmental shift.