Both abundant and rare fungi colonizing Fagus sylvatica ectomycorrhizal root-tips shape associated bacterial communities

Abstract

Ectomycorrhizal fungi live in close association with their host plants and form complex interactions with bacterial/archaeal communities in soil. We investigated whether abundant or rare ectomycorrhizal fungi on root-tips of young beech trees (Fagus sylvatica) shape bacterial/archaeal communities. We sequenced 16S rRNA genes and fungal internal transcribed spacer regions of individual root-tips and used ecological networks to detect the tendency of certain assemblies of fungal and bacterial/archaeal taxa to inhabit the same root-tip (i.e. modularity). Individual ectomycorrhizal root-tips hosted distinct fungal communities associated with unique bacterial/archaeal communities. The structure of the fungal-bacterial/archaeal association was determined by both, dominant and rare fungi. Integrating our data in a conceptual framework suggests that the effect of rare fungi on the bacterial/archaeal communities of ectomycorrhizal root-tips contributes to assemblages of bacteria/archaea on root-tips. This highlights the potential impact of complex fine-scale interactions between root-tip associated fungi and other soil microorganisms for the ectomycorrhizal symbiosis.

Fig. 1. Visualizations of morphological characteristics of abundant morphotypes on mycorrhizal root-tips and their bacterial colonization.

Stereoscope images of most abundant morphotypes (a1-3) and corresponding scanning electron microscopy (SEM) images (a4-6) offer detailed insights in morphological structures. Coherent series of SEM images, starting from roots with root-tips, zoomed in on one mycorrhizal root-tip (see white boxes) that shows microbial colonization on the ectomycorrhizal fungal mantle enclosing the root-tip (b). Our sequencing results revealed that the root-tip depicted in a1 and a2 encompassed a diverse fungal community. The classification of the most abundant OTUs (>50% of the reads) revealed that a1 belonged to Thelephoraceae/Tomentella and a2 to Agaricales/Hebelomataceae. The root-tip depicted in a3 was classified as Cenococcum geophilum based on its distinct morphology and sequence data.

Fig. 2. Distribution of most abundant fungal and bacterial/archaeal OTUs among bulk soil, rhizosphere and mycorrhizal root-tip habitats.

Venn diagrams depict shared and unique OTUs of sequenced bacterial/archaeal (a) and fungal (b) communities across investigated bulk soil (grey), rhizosphere (red) and mycorrhizal root-tip (yellow) microenvironments. Ternary plots depicting the distribution of 100 most abundant OTUs of fungal (c, d) and bacterial/archaeal (e) sequences in investigated bulk soil, rhizosphere and mycorrhizal root-tip habitats. The dot size corresponds to the average relative abundance of the OTUs across all samples. OTUs are colored by class (c, e) or fungal lifestyles (d). The position of the dots is determined by the occurrence and relative abundance of the OTUs in the different habitats. Points close to the corners indicate that the OTU has elevated relative abundance in this habitat compared to other habitats. Pie charts depict the taxonomic composition of the 100 most abundant OTUs in all habitats.

Fig. 3. Taxonomic assignment of fungal OTUs on individual mycorrhizal root-tips.

Relative read abundance (%) of fungal OTUs associated to the investigated 62 mycorrhizal root-tips samples. The 60 most abundant fungal OTUs are depicted in color; all other (less abundant) OTUs are depicted in white. The dendrogram indicates clustering of root-tips based on fungal community composition (Bray-Curtis dissimilarity). Each bar represents one mycorrhizal root-tip sample, the number underneath refers to the tree from which the root-tip originated. Same trees are highlighted in the same color.

Fig. 4. Structure of the modules in the investigated networks.

Ring graphs visualizing (a) network “10%”, (b) network “5%” and (c) network “0.1%”. Each network shows the structure of its modules (modules are numbered) and the surrounding rings correspond to the taxonomic composition of the respective module. The inner ring represents the composition of fungal orders in each module, ectomycorrhizal lifestyles are depicted striped. The outer ring represents the distribution of bacterial phyla. All taxa with a relative abundance <1% are condensed and depicted in white. The strength of the connection between the modules of each network corresponds to the thickness of the connecting black line.

Fig. 5. Chord diagrams comparing the consistency of modules’ composition across networks with different fungal OTU abundance cut-offs.

The fungal (a, c) and bacterial/archaeal taxa (b, d) assignments into modules of different networks are shown. Modules of network “0.1%” (encompassing fungal OTUs >0.1% relative abundance) are depicted in the upper part of each chord diagram and were compared to modules of network “10%” (encompassing fungal OTUs >10% rel. abundance, panel a and b) and network “5%” (encompassing fungal OTUs >5% rel. abundance, panel c and d). Only those fungi present in both, network “0.1%” and “5%” or “10%” (i.e. more abundant fungi) are compared. All modules are presented, but those composed exclusively of rare fungi (present only in network “0.1%”) appear empty (i.e. M1 and M4 in a).

Fig. 6. Conceptual framework on processes shaping the effect of fungi on the bacterial/archaeal communities associated with ectomycorrhizal root-tips.

Morphological features of ectomycorrhizal root-tips provide distinct microenvironments for smaller microorganisms such as bacteria and archaea. Such morphological features, which select for certain bacterial/archeal communities, can be shaped by only the most abundant fungi, but may also be influenced by the entire fungal community, including the less abundant ones. Depending on the influence of rare fungal taxa, fungal-bacterial/archaeal bipartite networks may show different structures if only highly abundant fungi or the whole fungal community are considered. Here we present a conceptual framework for interpreting such emerging differences in network structure: We assume that a shift in network structure after considering rare fungi can result from two non-exclusive processes: a) the emergence of specific bacterial associations to the rare fungi, and b) changes in the similarity patterns of bacterial associations with the highly abundant fungi. If changes are induced by the emergence of specific bacterial associations to rare fungi, these fungi will tend to be grouped into certain (new) modules in the network that considers all fungi (1.1 and 1.2, notice the grey module only present in the most complete network). If changes are induced because rare fungi alter the relationship between bacteria/archea and abundant fungi (f.e. by a split of the group of bacteria/archaea that were previously associated to the same highly abundant fungi) we expect to see that fungi which shared the same module in the ‘abundant fungi only’ network will be in separate modules in the ‚all fungi’ network (1.1 and 2.1, notice the green module consisting of a mix of colors (red and blue) in the lower part of the chord diagrams). These two processes can also simultanously affect the shifts in the network (1.1), and if none of them take place the shifts will be minimized (2.2). Networks for ‘only abundant’ and ‘all fungi’ are represented in the example chord diagrams in the middle of each panel (upper half shows only abundant fungi, lower half shows all fungi, colors depict different modules, as in Fig. 5), which visualize a potential difference in module structure. The elipses are colored according to the colors in the chord diagram: blue and red refers to non-changing modules in networks; grey refers to new modules that emerge due to the consideration of rare fungi and only contain rare fungi; green refers to modules that emerge due to changes in the highly-abundant fungi. Black triangles: abundant fungi, grey triangles: rare fungi, black circles: bacteria/archaea.