From the rhizosphere to the biosphere: Mapping the form and function of fungal communities across North American pine forests

Background/Question/Methods

Fungi are a critical component of the diversity and function of terrestrial ecosystems. They regulate decomposition rates, facilitate plant nutrient uptake and have a profound impact on agriculture and economics. Understanding the forces that structure fungal communities thus has important theoretical and practical implications. While ecologists have long recognized the importance of scale on ecological processes, fungal communities have primarily been studied at small-scales, focusing on deterministic processes. Understanding how macroecological processes shape fungal communities is in part hindered by the lack of datasets spanning large spatial-scales and the absence of distributional data for most fungi. To rectify this knowledge gap we use next generation sequencing of the internal transcribed spacer region of the nrRNA genes to survey soil fungi across North American pine forests, spanning a diverse range of climates from Florida to Alaska.  At each site we took 26 soil samples arranged in a nested grid and stratified into organic and mineral soil horizons.  This design allowed us to examine fungal community structure spanning small to large spatial scales and across steep local environmental gradients. We also conducted extracellular enzyme assays to test for relationships between form and function of the fungal community across the North American continent.

Results/Conclusions

We obtained fungal community data from over a million DNA sequences derived from 609 soil samples taken at 25 plots. Soil fungal communities in North American pine forests are highly diverse, with 10,576 species observed at a 97% sequence similarity cutoff.  Fungal communities were highly variable, but showed strong evidence of regional geographic structure, with both site identity (r2 = 0.23, P < 0.001) and geographic region (r2= 0.10, P < 0.001) explaining significant proportions of community variation.  Because of high turnover in community composition across sites and regions, core level soil environmental variables, such as soil horizon, explained < 1% of variation in the whole fungal community. In contrast, measures of potential soil enzyme activity appear determined by within sample resource availability. For example, %C concentration explained 23% of variation in cellobiohydrolase activity (P < 0.001) across sites separated by >1000 km.  These results suggest that macroecological processes are a major determinant of fungal community structure and that there is high functional redundancy in soil fungal communities across North America. This work demonstrates that increasing the scale of observation is critical to a complete understanding of the ecological dynamics of soil fungi and the ecosystem processes they mediate.