Effects of flower bacteria on floral scent and pollination

Microorganisms that live on flowers have the potential to modify some floral traits that play relevant roles in pollinator attraction. For example, they can cause quantitative and qualitative changes in floral scent. In the project entitled Effects of flower bacteria on floral scent and pollination, we aim to study the ecological impacts of flower microorganisms on plant- pollinator communication and plant reproduction through their effects on floral scent. We have designed two experiments 1) to measure the effects of bacterial microbiota on floral scents and pollinator attraction to flowers and 2) to assess how pollinators contribute to the dispersal of bacteria among flowers. We hypothesize that floral microbiota growing on flowers can cause significant changes in floral scent composition and emission rates, possibly evoking different responses on pollinators and affecting pollinator attraction to flowers and plant reproduction. Furthermore, we also hypothesize that pollinators, which normally conduct numerous and regular visits between flowers, are an important vector of dispersal of bacteria across the flower community, thus becoming an important factor determining the composition of bacterial communities that colonize flowers. The project will include the use of powerful and novel techniques such as gas-chromatography mass-spectrometry (GC-MS) for the identification and measurement of floral volatiles, the use of green fluorescent protein (GFP) labelled bacteria to measure bacteria dispersal among flowers, next-generation sequencing (NGS) for the characterization of bacterial communities, and radio frequency identification system (RFID) to track and record pollinator visits to flowers. The project is very innovative because it addresses two questions that combine research topics such as microbiology, floral scent and plant-pollinator ecology, deepening our knowledge in a highly unexplored research field. Pollination is a key process for agriculture and wild life all around the world. The results of this project can provide information of relevant interest for agricultural productivity, food security and the conservation of natural plant communities, by revealing unknown potential impacts of bacteria on plant-pollinator communication mediated by floral scents and finally on plant reproductive success.

The main objective of this project was to improve our understanding of the ecological impacts that non-pathogenic bacteria growing on flowers have on plant-pollinator interactions and on plant reproductive success, while paying special attention to their effects on floral scents that play very relevant roles in pollinator attraction. Flowers are colonized by specific bacterial communities that have been shown to affect floral scent emission and pollinator behavior. Whether these effects translate into changes in the plants' reproductive success remains unknown. In our experiments, we inoculated rapeseed (Brassica napus) flowers with two bacterial strains commonly found in flowers (Pantoea agglomerans and Pseudomonas syringae) to measure their effects on floral scent composition, on pollinator attraction to flowers and on the reproductive success of plants. The inoculation of B. napus flowers with P. agglomerans and P. syringae resulted in successful colonization of flowers by the inoculated strains and also modified the resident bacterial communities on flowers, according to next-generation sequencing (NGS) of flower microbial communities. The results from floral scent measurements analysed with gas-chromatography-mass spectrometry (GC-MS) showed that the inoculation of rapeseed flowers with these two bacterial strains modified floral scent composition. Further experiments in the lab showed that the bacterial strains emit volatiles from their own metabolism and have the potential to consume floral scent compounds, supporting their cappacity to produce changes in floral scent composition. Visitation frequencies by pollinators to the flowers in the field were higher for inoculated plants, suggesting that the changes in floral scent composition produced by bacteria made flowers more attractive to pollinators. Olfactometer tests conducted with bumblees (Bombus terrestris) also support that bumblebees preferred floral scents from inoculated plants. Increased pollinator visits resulted in increased pollination efficiency, and as a result we observed higher seed set and tendentially increased seed viability in plants inoculated with bacteria. Overall, the results of this project highlight the importance of complex tri-trophic interactions among plants, pollinators and bacteria by demonstrating that floral bacteria have significant effects on the pollination and reproduction of entomophilous plants that is at least partly mediated by their effects on floral chemical signaling. The project results also reveal the potential of some non-pathogenic bacteria from flowers to be used in agriculture to enhance the pollination of insect-pollinated plants and improve crop production.