Fungi structure ecosystems through interactions that ameliorate or exacerbate plant stress. Despite the reality that plants engage in complex interactions with diverse fungal communities, much of our understanding of how plants and fungi interact comes from studies of pairwise interactions. Simplifying plant-fungal interactions ignores the potential for direct and indirect interactions between fungi to generate non-additive effects on plant performance. Here, we performed the first study manipulating foliar fungal community diversity to determine how it impacts host productivity. Using 20 fungal species isolated from field-collected morning glory (Ipomoea hederifolia) leaves, we applied 16 combinations of 3 fungi (low diversity) and 16 combinations of 10 fungi (high diversity) to understand if different diversity levels led to additive, synergistic (greater than the additive outcome), or antagonistic (less than the additive outcome) effects on plant growth. Plants inoculated with single fungal species and fungicide-treated control plants were compared to determine if individual fungal species were mutualistic, commensal, or pathogenic. We used Monte Carlo simulations to compare high and low diversity fungal communities’ effects on plant performance to predicted plant performance based on single inocula effects, to determine if varying levels of diversity varies the amount of non-additivity (synergism or antagonism) on plant performance.
Results/Conclusions
Increasing foliar fungal diversity does improve plant performance compared to single inocula (F48,158=1.28, p-value=0.0286), but low and high diversity treatments were comparable. While foliar fungal community impacts on plant performance were equal between both high and low diversity communities, the level of fungal diversity impacted the strength of non-additivity, with low diversity communities having a greater magnitude of non-additive outcomes than high diversity communities (F1,112=9.727, p-value=0.002). Non-additive effects were equally common in frequency in plants grown with low or high diversity communities (9/16 communities for both) indicating that 1) non-additivity is common and 2) the dampening of non-additive effects with increasing diversity is independent of how common they are. Interestingly, we found that all single strain inoculations acted as commensals (species that do not affect plant performance). When grown in more diverse fungal communities these ‘commensals’ interacted in ways that often led to non-additive effects on plant performance further demonstrating that our understanding of plant-fungal interactions in natural environments must be interpreted under a framework that incorporates non-additive outcomes of multi-species interactions. Overall, our results highlight how interactions within host-associated microbiomes can strongly and non-additively affect the interaction between plants and microbes.