Microbes are ubiquitous, hidden players that can dramatically affect the functioning of plants and animals as well as biological processes involved in maintenance of biodiversity and ecosystem function. In recent work, we found that soil microbiomes often ameliorate environmental stress for imperiled plants in the Florida Scrub ecosystem and the extent of microbial mitigation of stressful conditions explains a significant amount (>80%) of the variation in plant species abundances across the landscape. Here, we determine how a common form of disturbance – fire – affects soil microbiomes and their consequences for native plant communities using a combination of field surveys and experiments, metatranscriptomics, and greenhouse manipulations. Specifically, we surveyed soils from ~36 Florida Scrub sites that varied in their fire history (2-120 since last burn), and then conducted an experiment exposing replicates of soil from each site to a prescribed burn in the field (and other replicates to conditions on an adjacent parcel). We flash froze/archived samples of each soil microbiome for metatranscriptomics to determine how fire legacy and the fire treatment individually and interactively alter expression of functional genes of the soil microbiome. We then grew 12 native plant species in an experiment manipulating fire history of the microbiome (36 site origins), microbiome exposure to the prescribed burn exposure (burn+/burn-), and microbial presence (sterilized/live inoculum) and measured effects on plant performance to determine fire legacy and prescribed burning effects on the host-microbe interaction.
Results/Conclusions
In the performance experiment, data from fast-growing plant species showed that only fire treatment, fire history, or microbial treatment affected plant performance for some species (e.g., Chamaecrista fasciculata, fire treatment: F=15.0, P=0.0002; Lechea cernua, microbial treatment: F=37.3, P<0.0001). However, other species showed a significant interactive effect of the fire and microbial inoculation treatments. For example, for Balduina angustifolia only microbes exposed to fire negatively affecting plant biomass (fire*microbe: F=6.5, P=0.012), suggesting fire shifted the balance of beneficial versus pathogenic microbes in favor of pathogen abundance or function. For the fire-dependent, slow-growing plant taxa (currently being harvesting), we predict fire will increase plant biomass by favoring microbes/microbial function beneficial to those species. If true, this experiment will indicate that differences in plant species’ responses to fire depend on fire’s selection on microbiome composition or function. Our next step is to link functional variation within the microbiome to where the microbe-plant interactions fall out on the mutualism-to-parasitism continuum under different fire regimes using our metatranscriptome and compositional data.