Unusually hot ambient temperatures (HAT) can cause preanthesis abortion of flowers in many diverse species, limiting crop production. This limitation is becoming more substantial with climate change. Flower primordia of passion fruit (Passiflora edulis Sims) vines exposed to HAT summers, normally abort. Flower abortion can also be triggered by gibberellin application. We screened for, and identified a genotype capable of reaching anthesis during summer as well as controlled HAT conditions, and also more resistant to gibberellin. Leaves of this genotype contained higher levels of endogenous cytokinin. We investigated a possible connection between higher cytokinin levels and response to gibberellin. Indeed, the effects of gibberellin application were partially suppressed in plants pretreated with cytokinin. Can higher cytokinin levels protect flowers from aborting under HAT conditions? In passion fruit, flowers at a specific stage showed more resistance in response to HAT after cytokinin application. We further tested this hypothesis in Arabidopsis. Transgenic lines with high or low cytokinin levels and cytokinin applications to wild-type plants supported a protective role for cytokinin on developing flowers exposed to HAT. Such findings may have important implications in future breeding programmes as well as field application of growth regulators.

The androgynophore column, a distinctive floral feature in passion flowers, is strongly crooked or bent in many Passiflora species pollinated by bats. This is a floral feature that facilitates the adaptation to bat pollination. Crooking or bending of plant organs are generally caused by environmental stimulus (e.g. mechanical barriers) and might involve the differential distribution of auxin. Our aim was to study the role of the perianth organs and the effect of auxin in bending of the androgynophore of the bat-pollinated species Passiflora mucronata. Morpho-anatomical characterisation of the androgynophore, including measurements of curvature angles and cell sizes both at the dorsal (convex) and ventral (concave) sides of the androgynophore, was performed on control flowers, flowers from which perianth organs were partially removed and flowers treated either with auxin (2,4-dichlorophenoxyacetic acid; 2,4-D) or with an inhibitor of auxin polar transport (naphthylphthalamic acid; NPA). Asymmetric growth of the androgynophore column, leading to bending, occurs at a late stage of flower development. Removing the physical constraint exerted by perianth organs or treatment with NPA significantly reduced androgynophore bending. Additionally, the androgynophores of plants treated with 2,4-D were more curved when compared to controls. There was a larger cellular expansion at the dorsal side of the androgynophores of plants treated with 2,4-D and in both sides of the androgynophores of plants treated with NPA. This study suggests that the physical constraint exerted by perianth and auxin redistribution promotes androgynophore bending in P. mucronata and might be related to the evolution of chiropterophily in the genus Passiflora.

Developing reproductive organs within a flower are sensitive to environmental stress. A higher incidence of environmental stress during this stage of a crop plants’ developmental cycle will lead to major breaches in food security. Clearly, we need to understand this sensitivity and try and overcome it, by agricultural practices and/or the breeding of more tolerant cultivars. Although passion fruit vines initiate flowers all year round, flower primordia abort during warm summers. This restricts the season of fruit production in regions with warm summers. Previously, using controlled chambers, stages in flower development that are sensitive to heat were identified. Based on genetic analysis and physiological experiments in controlled environments, gibberellin activity appeared to be a possible point of horticultural intervention. Here, we aimed to shield flowers of a commercial cultivar from end of summer conditions, thus allowing fruit production in new seasons. We conducted experiments over three years in different settings, and our findings consistently show that a single application of an inhibitor of gibberellin biosynthesis to vines in mid-August can cause precocious flowering of ~2–4 weeks, leading to earlier fruit production of ~1 month. In this case, knowledge obtained on phenology, environmental constraints and genetic variation, allowed us to reach a practical solution.