Broadening our understanding of plant-soil feedback at the community level through mycorrhizal associations of temperate hardwood forests

Andrew C. Eagar, Colleen Cosgrove, Mark W. Kershner and Christopher Blackwood, Department of Biological Sciences, Kent State University, Kent, OH, Ryan M. Mushinski, School of Life Sciences, University of Warwick, Coventry, United Kingdom, Amber L. Horning, Department of Biology, University of Mississippi, University, MS, Kurt A. Smemo, Environmental Studies and Sciences, Skidmore College, Saratoga Springs, NY, Richard P. Phillips, Department of Biology, Indiana University, Bloomington, IN

Background/Question/Methods
Recent work on plant-soil feedback has highlighted the role that mycorrhizal associations play in influencing feedback strength and direction. Here, we examine how the dominant mycorrhizal associations of temperate hardwood forests influence community composition of soil fungi and the consequences for tree community dynamics. Ectomycorrhizal (ECM) and arbuscular mycorrhizal (AM) trees cycle soil nutrients and carbon in distinct ways due to differences in litter chemistry and the metabolic abilities of their mycorrhizal fungi, which should influence the relative abundances of saprotrophic and pathogenic fungi. To test this hypothesis, we investigated forests representing a gradient of mycorrhizal dominance, ranging from entirely ECM tree communities to entirely AM tree communities. We used next-generation sequencing on soil samples to characterize the fungal communities across our mycorrhizal gradient. Additionally, because pathogenic fungi are drivers of negative plant-soil feedback, we tested for community-wide feedback effects in ECM and AM tree communities using point pattern analyses of stem-mapped juvenile and adult tree distributions. We expect that AM tree communities with higher relative abundances of pathogenic fungi will experience stronger negative feedback effects, resulting in overdispersion between adult and juvenile trees irrespective of their species identities.
Results/Conclusions
Our sequencing work revealed significant relationships between tree community mycorrhizal associations and the relative abundance of saprotrophic and pathogenic soil fungi. Increasing relative basal areas of AM trees was positively correlated with increases in both species richness and relative abundance of fungal saprotrophs and pathogens. These increases correspond with decreases in ECM fungal species richness and relative abundance. In addition, within our tree communities, we found the spatial structure between adult and juvenile trees to be dependent on the dominant mycorrhizal association of the community. Adult and juvenile trees in AM-associated communities were overdispersed from one another, suggestive of negative feedback, while ECM-associated communities reflected patterns of positive feedback through clustering between adult and juvenile trees. Taken together, the results of our studies suggest that trees growing in AM-associated communities encounter greater loads of fungal pathogens, which subsequently influence the spatial structure of these stands through negative feedback. Likewise, trees growing in ECM-associated communities experience positive feedback due to reduced loads of pathogens and greater occurrences of mutualists. Our findings indicate that the effects of plant-soil feedback, influenced by mycorrhizal associations, affect entire communities in temperate forests regardless of the species identity of community members.