Flowering time: From physiology, through genetics to mechanism

Abstract

Plant species have evolved different requirements for environmental/endogenous cues to induce flowering. Originally, these varying requirements were thought to reflect the action of different molecular mechanisms. Thinking changed when genetic and molecular analysis in Arabidopsis thaliana revealed that a network of environmental and endogenous signaling input pathways converge to regulate a common set of "floral pathway integrators." Variation in the predominance of the different input pathways within a network can generate the diversity of requirements observed in different species. Many genes identified by flowering time mutants were found to encode general developmental and gene regulators, with their targets having a specific flowering function. Studies of natural variation in flowering were more successful at identifying genes acting as nodes in the network central to adaptation and domestication. Attention has now turned to mechanistic dissection of flowering time gene function and how that has changed during adaptation. This will inform breeding strategies for climate-proof crops and help define which genes act as critical flowering nodes in many other species.

Figure 1.

The main genetic pathways controlling flowering time in Arabidopsis. Colored boxes highlight different pathways; FRI (purple) and clock components (brown), key integration nodes (FLC, CO, FT, and SOC1), and those with extensive natural variation (FRIGIDA, FLC, and FT) are in bold. Arrows indicate positive and bars represent negative regulatory relationships. Genetic pathways converge on FT, encoding a transmissible signaling molecule transported from the leaves to the SAM. The floral pathway integrators (in a circle) and floral meristem identity genes are shown in the green schematic meristem. The influence of sugar on some pathways is indicated. Different pathways are interconnected, for example, photoperiod and light quality and temperature pathways by COP1/SPA, and circadian and high temperature pathways by ELF3.

Figure A.

Antisense transcription at FLC locus. A) Schematic illustration of FLC gene architecture and COOLAIR transcripts. Black lines represent introns, black boxes represent exons, and grey boxes indicate UTR regions. FLC and COOLAIR transcription start sites are shown by black arrows. The 3′ end of FLC is enlarged below to show binding motifs/regions and currently available mutants that disrupt COOLAIR. B)COOLAIR expression profile during cold treatment, measured using Q-RT-PCR and 2 amplicons shown in (A) at different conditions. The data in condition 1 is from Jeon et al., 2023 while the condition 2 is the same as described in Swiezewski et al. 2009.

Figure B.

Co-transcriptional 3′processing at the FLC locus. A) RNA 3′ sequencing at the FLC locus reveals altered polyadenylation site selection of both the sense (left) and antisense (right) transcripts by fca-9, fld-4 or ColFRI compared with Col-0. B) The proximal to distal polyadenylation ratio of FLC by 3′RNA sequencing matches conventional qPCR analysis of COOLAIR polyadenylation ratio. C) Differentially expressed genes in fca-9, fld-4 and ColFRI compared with Col-0 (padj < 0.05 and Log2FC > 1). Consistently, FLC is one of the top upregulated genes.

Figure 2.

Natural variation in flowering pathways, comparing domesticated crops and natural species. Nodes circled in black in each pathway represent the major nodes with high allelic diversity. Yellow box shows selected photoperiodic regulators in rice (red text), maize (purple text), and barley (blue text). Extensive natural variation also occurs in vernalization regulators in barley (blue box), wheat, and brassica crops. Soybean breeding has predominantly targeted circadian clock components (pale brown box). White box: loci showing natural variation in A. thaliana.