Synchronisation of Arabidopsis flowering time and whole-plant senescence in seasonal environments

Abstract

Synchronisation of flowering phenology has often been observed between individuals within plant species. We expected that a critical role of flowering-time control under natural conditions is a phenological synchronisation. However, no studies have quantified the level of synchronisation of reproductive timing relative to germination timing under natural conditions. In a sequential seeding experiment (SSE) in which we manipulated the germination timing of Arabidopsis thaliana accessions, we developed a quantification index to evaluate reproductive synchrony in annual plants. In the SSE, we identified a novel phenomenon of reproductive synchrony: senescence synchrony. The role of vernalisation in realising flowering synchrony between plants of different ages under natural conditions was demonstrated by synchronisation and de-synchronisation of flowering initiation in vernalisation-sensitive and less-vernalisation-sensitive accessions, respectively. We also observed up-regulation of senescence-related genes at corresponding times. The approach we developed in this study provides a set of concepts and procedures that can be used to study reproductive synchrony experimentally under natural conditions.

Figure 1

Design of the sequential seeding experiment (SSE) and synchronisation index (SI_g). (a) Daily mean temperatures (red) and photoperiods (blue) during the experiment, (b) design of the SSE, and (c) formula for calculation and interpretation of the SI_g values. In SSE, seven cohorts (C1–C7) were prepared by sowing seeds at one-week intervals from the beginning of October to early December. Triangles indicate transfer to the outside garden at 21 days after seed sowing (a).

Figure 2

Results of SSE for four accessions of Arabidopsis thaliana. Phenological responses of the seven cohorts for (a) C24, (b) Ler-1, (c) Lov-5, and (d) Tamm-2 are indicated by bars representing the periods between germination and four successive reproductive timings. The SI_g values of bolting, flowering initiation, flowering termination, and whole-plant senescence are presented for (e) C24, (f) Ler-1, (g) Lov-5, and (h) Tamm-2. In (a–d), the colours correspond with those in the bars at the bottom of the figure. In C1 and C2 of Ler-1, flowering initiation took place before bolting.

Figure 3

Comparison of the vegetative periods and PTU across cohorts in the four accessions and the leaf number at bolting and flowering period compared to the vegetative period. Duration of vegetative periods (green bars) and PTU values during the corresponding periods (lines) are presented for seven cohorts of (a) C24, (b) Ler-1, (c) Lov-5, and (d) Tamm-2. (e) The dependency of the rosette leaf number at bolting on the vegetative period, and (f) the relationship between vegetative and flowering periods. The duration of the vegetative periods was calculated as the number of days from germination to flower initiation. In (a–d), the means and standard deviations (SD) are presented. Different letters at the bottom of the bars and next to the lines indicate significant differences (P < 0.05) in periods and PTU values between cohorts. N. S. represents no significant difference detected between all combinations of cohorts. In (e,f), the cohort means are plotted with different symbols for four accessions. Correlation coefficients (r) are also listed for each accessions (^***, ^**, N.S.; P < 0.001, P < 0.01, no significance at P < 0.05, respectively). Standard deviations for the number of rosette leaves are represented by vertical bars (e).

Figure 4

Time-series changes in gene expression of two flowering-time genes (FLC and FT) and four senescence-related genes (ORE1, NAC016, NYC1, and SAG12) before flowering termination. Patterns in C1, C3, C5, and C7 are indicated by thick red, blue, green, and orange lines, respectively, for (a) C24 and (b) Lov-5. Gene expression (log₂ relative to those of ACT2) of the least-senesced leaves of plants was measured weekly from early March. The median values of two replicates at each sampling day are indicated. The average timings of flowering termination are represented by vertical dashed lines with corresponding colours. The top diagrams represent the timing of bolting, flowering initiation, flowering termination, and whole-plant senescence (boundaries between light-grey – grey – dark-grey – black, and end of the black bars, respectively) during the period of March–June in the SSE. The results including all examined genes and accessions are listed in Supplementary Fig. S2.