Fine Spatial Scale Variation of Soil Microbial Communities under European Beech and Norway Spruce

Abstract

The complex interactions between trees and soil microbes in forests as well as their inherent seasonal and spatial variations are poorly understood. In this study, we analyzed the effects of major European tree species (Fagus sylvatica L. and Picea abies (L.) Karst) on soil bacterial and fungal communities. Mineral soil samples were collected from different depths (0-10, 10-20 cm) and at different horizontal distances from beech or spruce trunks (0.5, 1.5, 2.5, 3.5 m) in early summer and autumn. We assessed the composition of soil bacterial and fungal communities based on 16S rRNA gene and ITS DNA sequences. Community composition of bacteria and fungi was most strongly affected by soil pH and tree species. Different ectomycorrhizal fungi (e.g., Tylospora) known to establish mutualistic associations with plant roots showed a tree species preference. Moreover, bacterial and fungal community composition showed spatial and seasonal shifts in soil surrounding beech and spruce. The relative abundance of saprotrophic fungi was higher at a depth of 0-10 vs. 10-20 cm depth. This was presumably a result of changes in nutrient availability, as litter input and organic carbon content decreased with soil depth. Overall bacterial community composition showed strong variations under spruce with increasing distance from the tree trunks, which might be attributed in part to higher fine root biomass near spruce trunks. Furthermore, overall bacterial community composition was strongly affected by season under deciduous trees.

Keywords: bacterial 16S rRNA gene; fungal ITS DNA; horizontal distance from tree trunk; seasons; soil depth; soil microbial community structure; soil properties; tree species.

Figure 1

Sampling design: In early summer and autumn 2012 samples were taken at a distance of 0.5, 1.5, 2.5, and 3.5 m from the tree trunks of four European beech and four Norway spruce trees. At all sampling points soil samples from 0–10 cm and 10–20 cm were taken.

Figure 2

Box-and-whiskers plot showing relative abundances of bacterial and fungal phyla as well as proteobacterial classes detected in each of the analyzed 128 soil samples. Relative abundances of taxa across all samples (gray color) as well as separately with respect to soil surrounding beech (brown color) and spruce (green color) are depicted. The dashed line separates relative abundances of bacterial and fungal taxa.

Figure 3

Relative abundance of dominant bacterial and fungal genera detected in the analyzed soil samples. The data represent mean values and standard errors of relative abundance for the 20 most abundant bacterial and fungal genera, respectively. Acidobacteria were analyzed at the subgroup level and therefore not considered within this figure. Relative abundances of taxa across all samples (gray color) as well as separately with respect to soil surrounding beech (brown color) and spruce (green color) are depicted. Asterisks indicate taxa showing an at least five-fold difference in mean relative abundance between spruce and beech (P < 0.001 for the Mann-Whitney U test). Underlined taxa: saprotrophic fungi (all other depicted fungal genera represent ectomycorrhizal fungi).

Figure 4

Principal coordinates analysis plots based on weighted UniFrac distances calculated at 3% genetic distance. Brown circles represent samples derived from beech surrounding soil and samples derived from spruce surrounding soil are depicted as green circles. Vectors represent response variables pH, estimated clay content, C:N ratio, organic carbon (OC), and inorganic carbon (IC). Significant values (P < 0.05) according to “envfit” calculations are indicated by asterisks.

Figure 5

Box-and-whiskers plots showing relative abundance of ectomycorrhizal and saprotrophic fungi under beech and spruce in relation to soil depths. The asterisks indicate significant differences between soil depths for each ecological group determined by ANOVA; ^*significant (P < 0.05), ^***highly significant (P < 0.001).