A survey of flowering genes reveals the role of gibberellins in floral control in rose.

Exhaustive studies on flowering control in annual plants have provided a framework for exploring this process in other plant species, especially in perennials for which little molecular data are currently available. Rose is a woody perennial plant with a particular flowering strategy--recurrent blooming, which is controlled by a recessive locus (RB). Gibberellins (GA) inhibit flowering only in non-recurrent roses. Moreover, the GA content varies during the flowering process and between recurrent and non-recurrent rose. Only a few rose genes potentially involved in flowering have been described, i.e. homologues of ABC model genes and floral genes from EST screening. In this study, we gained new information on the molecular basis of rose flowering: date of flowering and recurrent blooming. Based on a candidate gene strategy, we isolated genes that have similarities with genes known to be involved in floral control in Arabidopsis (GA pathway, floral repressors and integrators). Candidate genes were mapped on a segregating population, gene expression was studied in different organs and transcript abundance was monitored in growing shoot apices. Twenty-five genes were studied. RoFT, RoAP1 and RoLFY are proposed to be good floral markers. RoSPY and RB co-localized in our segregating population. GA metabolism genes were found to be regulated during floral transition. Furthermore, GA signalling genes were differentially regulated between a non-recurrent rose and its recurrent mutant. We propose that flowering gene networks are conserved between Arabidopsis and rose. The GA pathway appears to be a key regulator of flowering in rose. We postulate that GA metabolism is involved in floral initiation and GA signalling might be responsible for the recurrent flowering character.