The plant-soil feedback framework has been used to simplify the complexity of plant-rhizosphere interactions and identify mechanisms that can lead to density dependent outcomes among competing plant species. In recent work by our lab, we demonstrated a parallel approach for understanding plant-phyllosphere interactions termed “plant-phyllosphere feedbacks”. Here, we present a follow-up study exploring context-dependent outcomes in plant-phyllosphere feedbacks in the presence or absence of a foliar pathogen. Using a full factorial experimental design, greenhouse grown plants (three Asteraceae species: Liatris spicata, Parthenium integrifolium, Ratibida pinnata) were exposed to either conspecific or heterospecific microbial communities derived from leaf litter and fresh leaves, or a sterile control. The phyllosphere inoculum was obtained from mature plants grown in a common garden. After 11 weeks, half of the plants were inoculated with a bacterial pathogen (Pseudomonas syringae pv. tomato DC3000). Plant growth responses were measured throughout the experiment (height, leaf number) and at harvest (biomass). In addition, we collected samples from all inocula sources (e.g., litter, leaves) and samples from the exposed experimental plants (e.g., leaves) to determine patterns of microbial colonization based on Illumina sequencing of the fungal communities.
Results/Conclusions
Overall, we found that the presence of the pathogen shifted average pairwise feedback from net negative to net positive for L. spicata – indicating a benefit to growth in the presence of conspecific phyllosphere microbiota when dually exposed to a pathogenic enemy. However, the plant phyllosphere feedbacks were not significant for the other two plant species and were not context dependent with pathogen presence. In concordance with previous plant-phyllosphere research, we found that live-sterile comparisons for phyllosphere treatments were not significantly different from zero. There were differences among plant species in response to live phyllosphere treatments. Specifically, L. spicata performed worse with either conspecific or heterospecific microbiota relative to sterile controls, while P. integrifolium performed better. Illumina sequencing data revealed differences among host species in the structure of their fungal endophyte communities and these varied by the type of phyllosphere treatment (e.g., conspecific, heterospecific). Fungi identified in the leaf litter and fresh leaf inocula showed partial overlap with the recipient plant endophyte communities, indicating successful exposure and infection of phyllosphere microbiota. Our results suggest that phyllosphere microbiota can be beneficial for some species under conditions of pathogen stress and reveal mechanisms of assembly for fungal taxa in the phyllosphere microbiome.