COS 17-2 - Dimensions of ectomycorrhizal diversity: Seasonality and diversity of host tree genera determine species richness of fungal communities

Monday, August 7, 2017: 1:50 PM
E141, Oregon Convention Center
Brian Steidinger, Biology, Stanford University, Kabir G. Peay, Department of Biology, Stanford University, Stanford, CA, Thomas D. Bruns, Department of Plant and Microbial Biology, University of California, Berkeley, CA, Rytas Vilgalys, Department of Biology, Duke University, Durham, NC, John Taylor, Plant & Microbial Biology, University of California, Berkeley, Berkeley, CA and Jennifer M. Talbot, Department of Biology, Boston University, Boston, MA

Ectomycorrhizal (ECM) fungi are a functionally diverse guild of plant symbionts that promote nutrient cycling, carbon sequestration and plant productivity. However, it is currently unknown how many ECM fungal species occur at different scales and what factors influence this diversity. This information is critical for determining the causes of ECM fungal coexistence and the diversity-function relationships that govern ecosystem processes.

To remedy this gap in our understanding, we collected 1540 soil cores taken from 70 sites along a climactic/latitudinal gradient across North America. Each site was dominated by trees from the Pinaceae, reducing variability in ECM fungal abundance and species richness caused by availability of compatible host trees. We used next-gen environmental sequencing (454 and Illumina) to determine the number of OTU's associated with each site. We assigned OTU's to the ECM functional guild using FunGuild.

Next, we fitted linear models to compare the number of ECM fungal species associated with individual forest sites vs. soil chemical variables (e.g., pH, % carbon, % nitrogen), host tree diversity (e.g., # of ECM tree species, genera, families), and climatic variables derived from WorldClim (e.g., seasonal precipitation, variability in temperature). We selected the best model based on AIC values.


After rarefying to even depth, we sequenced 9955 fungal species, 1,665 of which were ECM fungi. At the scale of individual forest sites, we found 90 ECM species (511 total fungal species); at the scale of soil cores we found an average of 16 ECM species (90 total fungal species). Our model accounted for ~70% of the variability in ECM species richness across sites. The model predicts increased ECM richness: i) in the mineral vs. organic horizon; ii) with a greater # of ECM tree genera; iii) associated with lower mean annual precipitation; and iv) associated with seasonality in temperature. These results suggest that ECM species may form niches around seasons and host tree genera.