Abstract:
The transition to flowering in plants is under multifactorial regulation. Environmental cues, such as light, temperature, and endogenous factors are integrated by a complex genetic network to ensure the correct timing of this transition. In Arabidopsis thaliana, the plant hormone gibberellic acid (GA) is an important endogenous element involved in the regulation of flowering under inductive long-day (LD) and non-inductive short-day (SD) photoperiod. However, important questions regarding the spatial organization of the flowering response, and the relative contribution of factors involved in the GA-mediated control of flowering such as DELLA proteins require a more detailed analysis.
In the first part of this thesis I present evidences indicating that the abundance of DELLA proteins in leaves and/or the shoot meristem is an important factor affecting flowering transition, depending on the photoperiodic conditions. Under LD, GA controls flowering by promoting the expression of the florigen FLOWERING LOCUS T (FT) and TWIN SISTER OF FT (TSF) independently of CONSTANS (CO) and GIGANTEA (GI) in the phloem companion cells of the leaf vasculature. In addition, GA positively regulates the expression of several SQUAMOSA PROMOTER BINDING PROTEIN-LIKE (SPL) transcription factors in both leaf vasculature and shoot meristem, which are themselves promoting flowering. In contrast to the spatial separation observed in LD, the control of flowering time by GA is restricted to the shoot meristem in SD. The data presented in chapter 2 integrate the well-known effect of GA on the expression of SUPPRESSOR OF OVEREXPRESSION OF CONSTANS (SOC1) and FRUITFUL (FUL) and the recently discovered age flowering pathway, which ensures flowering under SD conditions. Under SD, GA controls flowering through SPL-mediated control of SOC1 and FUL expression at the shoot meristem. In agreement with a recent GA signaling model, my data suggest that in addition to the transcriptional control of SPL genes, GA may regulate flowering through direct SPL-DELLA interaction.
Finally, I describe a new method to map mutations in Arabidopsis thaliana, which relies on high coverage sequencing of polymorphic regions captured with specific probes. Using this method we were able to accurately estimate the confidence interval harboring an unknown causal mutation isolated in a suppressor screen of the strong flowering promoting gene FT. Interestingly, we performed mapping-by-sequencing without the use of a reference genome using only syntheny information between Arabidopsis and Brassica rapa. This method represents an interesting alternative for mapping mutations in species without a reference genome or genetic map.