Precipitation, Not Land Use, Primarily Determines the Composition of Both Plant and Phyllosphere Fungal Communities

Abstract

Plant communities and fungi inhabiting their phyllospheres change along precipitation gradients and often respond to changes in land use. Many studies have focused on the changes in foliar fungal communities on specific plant species, however, few have addressed the association between whole plant communities and their phyllosphere fungi. We sampled plant communities and associated phyllosphere fungal communities in native prairie remnants and post-agricultural sites across the steep precipitation gradient in the central plains in Kansas, USA. Plant community cover data and MiSeq ITS2 metabarcode data of the phyllosphere fungal communities indicated that both plant and fungal community composition respond strongly to mean annual precipitation (MAP), but less so to land use (native prairie remnants vs. post-agricultural sites). However, plant and fungal diversity were greater in the native remnant prairies than in post-agricultural sites. Overall, both plant and fungal diversity increased with MAP and the communities in the arid and mesic parts of the gradient were distinct. Analyses of the linkages between plant and fungal communities (Mantel and Procrustes tests) identified strong correlations between the composition of the two. However, despite the strong correlations, regression models with plant richness, diversity, or composition (ordination axis scores) and land use as explanatory variables for fungal diversity and evenness did not improve the models compared to those with precipitation and land use (ΔAIC < 2), even though the explanatory power of some plant variables was greater than that of MAP as measured by R2. Indicator taxon analyses suggest that grass species are the primary taxa that differ in the plant communities. Similar analyses of the phyllosphere fungi indicated that many plant pathogens are disproportionately abundant either in the arid or mesic environments. Although decoupling the drivers of fungal communities and their composition – whether abiotic or host-dependent – remains a challenge, our study highlights the distinct community responses to precipitation and the tight tracking of the plant communities by their associated fungal symbionts.

Amorpha canescens and Andropogon gerardii Recruit Comparable Foliar Fungal Communities Across the Steep Precipitation Gradient in Kansas

“…A lack of distinction between the foliar communities of the two hosts was surprising, particularly because differences have even been reported between co-occurring Dalea L. (prairie clover) species in Minnesota prairies (DeMers and May 2021). Equally surprising was the lack of any relationship between fungal diversity and MAP, because others (e.g., DeMers and May 2021; Dea et al 2022) have documented strong differences based on sampling locations and their position along the MAP gradient in the Midwest. Despite the lack of differences in fungal communities between our plant hosts or among sampling sites, fungi represent a significant source of biodiversity, and affect ecosystem function (Perreault and Laforest-Lapointe 2022).…”

“…We aimed to evaluate how foliar fungal communities associated with two host species representing two functional groups (a grass and a legume) would respond to MAP along the steep precipitation gradient in Kansas. Previous research has highlighted that edaphic (Glynou et al 2016; Rudgers et al 2021) and climatic (Dea et al 2022; Oita et al 2021) variables may control the assembly of host-associated fungal communities. Similarly, host-associated communities may correlate with host communities (Dea et al 2022) or may differ among the host species (U'Ren et al 2012; Kembel et al 2014).…”

“…Previous research has highlighted that edaphic (Glynou et al 2016; Rudgers et al 2021) and climatic (Dea et al 2022; Oita et al 2021) variables may control the assembly of host-associated fungal communities. Similarly, host-associated communities may correlate with host communities (Dea et al 2022) or may differ among the host species (U'Ren et al 2012; Kembel et al 2014). Foliar fungal communities can be particularly sensitive to climatic drivers and buffered against edaphic factors (Bowman and Arnold 2021; Oita et al 2021) as the foliar communities do not directly interact with the soil matrix and are thus more susceptible to temporal and diurnal oscillations in the environment.…”

“…These studies differed from ours in several ways, however. For example, Carroll and Carroll (1978) compared fungal communities on gymnosperms at both high (dry) and low (wet) elevation sites; Zimmerman and Vitousek (2012) compared communities across a strong hydrologic and elevational gradient in Hawaii on Mauna Loa Volcano; U'Ren et al (2012) compared fungal communities along a precipitation gradient, but at continental scale; Oita et al (2021) compared communities at a landscape scale, but this was in the tropics and also spanned an elevation gradient; and finally, Dea et al (2022) studied leaf fungal communities of prairies in Kansas, but their work spanned the entire precipitation gradient in Kansas sampling whole plant communities rather than a single target plant.…”

“…Climate factors like precipitation can affect fungal community assembly and alter fungal diversity and community composition on plants and in soils (House and Bever 2018; Lagueux et al 2020; Rudgers et al 2021; Dea et al 2022). For example, using Bray-Curtis dissimilarity matrices, DeMers and May (2021) documented a potential fungal community-level endophyte gradient along a precipitation gradient in Minnesota prairies.…”