Scientists uncover new roles for the florigen protein in flowering

Plants start to flower in response to environmental signals such as daylength. This process involves a protein signal, florigen, that is made in the leaves and induces floral development at the shoot tip. Researchers in the groups of George Coupland at the Max-Planck Institute for Plant Breeding Research in Cologne and their collaborators have elegantly shed new light on the existing model for how florigen and two other proteins interact at the shoot apex to form the florigen activation complex (FAC), which is responsible for activating the genes required for flower formation. The findings, published in Nature, show that after movement of florigen into the shoot tip, formation of the FAC occurs on DNA in a distinct sequence of events. Also, the authors show that as well as inducing flowering, florigen has later independent functions during the formation of flowers.

Daylength regulates plant developmental programmes, and in particular causes the transition from vegetative to reproductive development. The perception of daylength involves the synthesis of FLOWERING LOCUS T (FT) florigen protein in the leaves, which is then transported to the shoot apical meristem (SAM) where it becomes integrated into the florigen activation complex (FAC). A current model of FAC assembly proposes that 14-3-3 proteins act as receptors for FT protein at the SAM and then mediate the indirect interaction between FT and a third protein, FD to form the FAC, which subsequently activates the transcription of genes that promote the formation of flowers.

     The group of scientists at the Max-Planck Institute for Plant Breeding Research in Cologne and their collaborators used mutational analysis, and an array of sophisticated biochemical, modelling and gene expression detection techniques to propose a new multifaceted mechanism by which the FAC is formed, and demonstrated how the individual components regulate distinct functions during floral induction and early flower development.

     The authors used biochemical methods to show that ten different 14-3-3 proteins interact with FD. They confirmed that a threonine amino acid at position 282 in the C-terminus of the FD protein is phosphorylated and is essential to mediate interaction of FD with 14-3-3. When it is not phosphorylated, the FD protein forms insoluble and non-functional condensates within the nucleus. Therefore, phosphorylation of FD allows its solubilisation, interaction with 14-3-3 proteins and binding to DNA. Interaction of FD with 14-3-3 proteins also stabilises FD dimers, potentially inducing conformational changes in the tail at the C terminus of FD that enhance DNA binding. Therefore, 14-3-3 proteins function as chaperone-like proteins that enhance FD activity at multiple levels.

     Because interaction between 14-3-3 and FT is weak or unstable, the authors hypothesised that binding of the FD-14-3-3 complex to DNA might create interfaces that are directly recognised by FT, even though FT was not thought until now to contact FD and/or DNA. They modelled the FAC–DNA complex, which suggested direct contact between three positively charged arginine residues in the FT tail and negatively charged DNA (Figure 1). They mutated these residues and showed that indeed, what they termed interface 1 at C-terminal tail of FT is involved in FT recruitment to the DNA-FD-14-3-3 complex by directly interacting with DNA. They also demonstrated that a second interface of amino acids that are crucial for the interaction between rice florigen and 14-3-3 proteins are required in FT for recruitment of FT by the DNA-FD-14-3-3 complex in Arabidopsis. The important conclusion from these experiments is that the FD-14-3-3 complex firstly has to bind DNA in order to recruit FT.

     A detailed quantification of the expression of fluorescently labelled FD and FT proteins at the cellular level in different regions of the SAM was performed to accurately confirm their degree of overlap. In addition, the authors used fluorescence-based, multiplexed RNA in situ hybridizations (RNAscope) to visualize the spatiotemporal accumulation of FD and FT transcripts in different regions of the SAM and floral primordia (Figure 2). Collectively, the protein and mRNA analyses confirmed that FT is present in the same cells as FD and 14-3-3 proteins at different stages of floral induction at the SAM to enable formation of the FAC.

     An unexpected finding was that FT mRNA was also found to accumulate at the boundary of newly formed primordia first on the adaxial side of cauline leaves, and later on the adaxial side of the floral primordium, and later, beneath the domain of expression of APETALA1, which represents the boundary with the suppressed bract. This means it also overlaps with FD and allows formation of the FAC during the first stages of floral development.

     The major importance of the study resides in three major findings: firstly, the demonstration of the importance of 14-3-3 function for the activity of FD at several levels of regulation; secondly, the identification of the two important interfaces that link FT to DNA and the DNA-FD-14-3-3 complex, and thirdly, the dynamic pattern of FT accumulation in the SAM and FT mRNA in lateral organ primordia.

     The finding that local FT transcription occurs within the floral primordium in addition to its transcription in the leaves and subsequent transport to the SAM builds on findings from another recent publication in Development by Romera-Branchat and colleagues also from the group of George Coupland, who analysed the flowers of mutant plants lacking the functions of FD and FD PARALOGUE (FDP), its closely related protein, and TWIN SISTER OF FT (TSF) and FT. The flowers of these fd fdp and ft tsf mutants had defects in the number and identity of floral organs, particularly of the sepals and petals and the shoot apex was also larger. These defects were also seen in short-day conditions, when FT is not produced in the leaves and transported to the apex. This led to the important conclusion that the FT gene is also expressed in developing floral buds and floral organs. This provided genetic evidence that the FT gene is also expressed in developing floral buds and floral organs. The authors identified the genes whose transcription was affected by loss of FD/FDP and FT/TSF function in flowers and concluded that in addition to their function in regulating floral transition, the FAC components FD/FDP and FT/TSF regulate expression of a class of SEPALLATA genes to control floral meristem shape and size, and have effects on floral organ number, development and identity.

     The findings from both these studies provide novel perspectives for understanding the mechanisms through which florigen promotes flowering and floral development in the model plant Arabidopsis. However, the strong conservation of the function of florigen and the FAC in all seed plants suggests that these results will also be important in studying flowering control of major crops.