Fungal Assemblages in Different Habitats in an Erman's Birch Forest

Abstract

Recent meta-analyses of fungal diversity using deeply sequenced marker genes suggest that most fungal taxa are locally distributed. However, little is known about the extent of overlap and niche partitions in total fungal communities or functional guilds within distinct habitats on a local forest scale. Here, we compared fungal communities in endosphere (leaf interior), phyllosphere (leaf interior and associated surface area) and soil samples from an Erman's birch forest in Changbai Mountain, China. Community structures were significantly differentiated in terms of habitat, with soil having the highest fungal richness and phylogenetic diversity. Endophytic and phyllosphere fungi of Betula ermanii were more phylogenetically clustered compared with the corresponding soil fungi, indicating the ability of that host plants to filter and select their fungal partners. Furthermore, the majority of soil fungal taxa were soil specialists, while the dominant endosphere and phyllosphere taxa were aboveground generalists, with soil and plant foliage only sharing <8.2% fungal taxa. Most of the fungal taxa could be assigned to different functional guilds; however, the assigned guilds showed significant habitat specificity with variation in relative abundance. Collectively, the fungal assemblages in this Erman's birch forest were strictly niche specialized and constrained by weak migration among habitats. The findings suggest that phylogenetic relatedness and functional guilds' assignment can effectively interpret the certain ecological processes.

Keywords: Erman’s birch forest; functional guild; fungal assemblages; habitat filtering; overlap and partition; phylogenetic relatedness.

FIGURE 1

Fungal class distribution among habitats. Any classes accounting for less than 1% abundance are classified as “Others.”

FIGURE 2

Hierarchical clustering dendrogram of all samples. (A) Bary–Curtis dissimilarity based on incidence data, (B) Bary–Curtis dissimilarity based on abundance data, and (C) MNTD dissimilarity based on phylogenetic data. The sequences were subsampled to 1733 sequences per sample. Fungi 21∼30 belong to the endosphere, and are represented by light green leaves; Fungi 61∼66 belong to the phyllosphere, and are represented by dark green leaves; and Fungi 67∼70 belong to the soil, and are represented by dark soil.

FIGURE 3

Variation in standardized effect sizes of the mean nearest taxon distance (ses.MNTD) among habitats. (A) Total fungi, (B) Ascomycota fungi, and (C) Basidiomycota fungi. “Not significant”: P > 0.05 and “significant”: P < 0.05.

FIGURE 4

Operational taxonomic unit (OTU) network map and venn diagrams of the different habitats (OTU numbers). (A) OTU network map of total fungi communities across habitats. Each pink point represents an independent fungal OTU (479 OTUs; sequences were subsampled to 35,328 per habitat). Edges show which OTUs belonged to which Habitat. Edge thickness was weighted by the abundances from the OTU to the specific habitat. Different shaped points represent core or satellite OTUs in each habitat. (B) Venn diagrams of total fungal communities and five functional guilds among habitats. Green, blue and khaki represent samples belonging to the endosphere, phyllosphere and soil, respectively.