Keystone interaction: Mutualistic fungal endophytes in a non-dominant grass cause large changes in community diversity at range-wide scales

Background/Question/Methods

The keystone species concept -- i.e., species that have a disproportionate effect on their community relative to their biomass -- has played an important role in understanding the maintenance of community diversity and alternative stable states. Further, faced with shifting species distributions under environmental change, knowledge of which keystone species are involved in the maintenance of community diversity and ecosystem functioning is of critical importance. Most classical studies of keystone species have focused on visible macro-organisms, however “hidden” microbes could also act as community keystones. Here, we show that fungal endophytes (Epichloë spp.), which associate with a non-dominant, California-native grass (Bromus laevipes), are keystone species that significantly affect plant and arthropod diversity across their distribution. First, in a 3-year field experiment, we created plots of B. laevipes with either naturally high or experimentally-reduced levels of fungal endophytes and measured abundance, diversity, and composition of the plant and arthropod communities. Second, we studied the effect of endophytes on plant community diversity within 5 common gardens spanning ~1/3 of the grass’ range, uniquely testing how common these patterns are across ecologically diverse habitats and climates. 

Results/Conclusions

In the 3-year field experiment, we documented for the first time that fungal endophytes can dramatically increase plant community diversity. In some years, we observed a ~80% greater increase in plant diversity in plots with high levels of endophyte (compared to plots with low endophyte) and a ~45% increase across the entire experiment. While the endophytes marginally increased richness, the strongest effect was on evenness which endophytes increased by suppressing a dominant exotic species. Endophyte also caused a significant shift in communities, such that plant communities with high endophyte shared a unique composition. We also detected an endophyte-generated decline in arthropod diversity and richness, an effect as strong as that previously documented for predation, plant genetic/species diversity, etc. Further, endophyte enhanced plant community richness across all 5 of the common gardens that spanned a wide range of habitats and ~1/3 of the host plant range. Taken together, we show substantial endophyte effects on both plant and arthropod communities in a native grass-fungal endophyte association, and further that the effect is a range-wide phenomenon. Our results illustrate the importance of considering microbial keystone species for understanding and preserving diverse communities and ecosystems.