Structure and specialization of mycorrhizal networks in phylogenetically diverse tropical communities

Abstract

Background: The root mycobiome plays a fundamental role in plant nutrition and protection against biotic and abiotic stresses. In temperate forests or meadows dominated by angiosperms, the numerous fungi involved in root symbioses are often shared between neighboring plants, thus forming complex plant-fungus interaction networks of weak specialization. Whether this weak specialization also holds in rich tropical communities with more phylogenetically diverse sets of plant lineages remains unknown. We collected roots of 30 plant species in semi-natural tropical communities including angiosperms, ferns, and lycophytes, in three different habitat types on La Réunion island: a recent lava flow, a wet thicket, and an ericoid shrubland. We identified root-inhabiting fungi by sequencing both the 18S rRNA and the ITS2 variable regions. We assessed the diversity of mycorrhizal fungal taxa according to plant species and lineages, as well as the structure and specialization of the resulting plant-fungus networks.

Results: The 18S and ITS2 datasets are highly complementary at revealing the root mycobiota. According to 18S, Glomeromycotina colonize all plant groups in all habitats forming the least specialized interactions, resulting in nested network structures, while Mucoromycotina (Endogonales) are more abundant in the wetland and show higher specialization and modularity compared to the former. According to ITS2, mycorrhizal fungi of Ericaceae and Orchidaceae, namely Helotiales, Sebacinales, and Cantharellales, also colonize the roots of most plant lineages, confirming that they are frequent endophytes. While Helotiales and Sebacinales present intermediate levels of specialization, Cantharellales are more specialized and more sporadic in their interactions with plants, resulting in highly modular networks.

Conclusions: This study of the root mycobiome in tropical environments reinforces the idea that mycorrhizal fungal taxa are locally shared between co-occurring plants, including phylogenetically distant plants (e.g. lycophytes and angiosperms), where they may form functional mycorrhizae or establish endophytic colonization. Yet, we demonstrate that, irrespectively of the environmental variations, the level of specialization significantly varies according to the fungal lineages, probably reflecting the different evolutionary origins of these plant-fungus symbioses. Frequent fungal sharing between plants questions the roles of the different fungi in community functioning and highlights the importance of considering networks of interactions rather than isolated hosts.

Keywords: Common mycorrhizal networks; Early-diverging plants; Endomycorrhiza; Fungal metabarcoding; Mucoromycotina fine root endophytes; Root endophytism.

Fig. 1

The three sampled communities correspond to habitats with contrasted environmental conditions. A map of La Réunion island indicating the three sampled communities in this study. The sampling sites were characterized by different vegetations and abiotic conditions with different elevations, levels of disturbance, humidity, and soil conditions: (right) Grand brûlé (recent lava flow close to the ocean on the wet East coast), (middle) Plaine-des-Palmistes (wet thicket on old lava flows in the central valley, elevation 900 m), and (left) Dimitile (ericoid shrubland on old lava flows in the dry crests of Cilaos cirque on the West side dominated by ericoid vegetation, elevation 2000 m). In each community, we replicated the sampling in three plots distant from 50 to 250 m. The photos illustrate the overall vegetation in each sampled community and the gradients at the bottom resume the main variations in the environments. The raw map in the background was generated by Eric Gaba (Wikimedia Commons user: Sting)

Fig. 2

The composition of the root mycobiomes varies according to the plant species and the habitats. For each plant species, the different root samples were merged and the relative abundances of the root fungi are indicated according to the 18S rRNA (left) or ITS2 (right) markers. We only retained the fungal lineages that may form mycorrhizal interactions with at least one plant species. Plant species are separated according to the sampled community (Grand brûlé, Plaine-des-Palmistes, or Dimitile) covering contrasted habitats. For each species, the number of individual root systems sampled is indicated in brackets. One species of Ipomoea remains taxonomically unidentified at the species level in Plaines-des-Palmistes. In each sampled community, a phylogenetic tree of the plants is represented on the left, with branch colors indicating the main plant taxonomic groups we considered in our study. The bar plots represent in colors the class and the order of each fungus. Rare taxa, representing less than 0.5% of the data, are plotted in dark grey

Fig. 3

Both the habitat and the plant taxonomic group influence mycobiome composition despite frequent fungal sharing. Dendrogram representations of the different root mycobiota across all the sampled communities (a) or within each sampled community (b Grand brûlé, Plaine-des-Palmistes, or Dimitile) based on the 18S rRNA (left) or ITS2 (right) markers. For each community, we only retained the fungal lineages that may form mycorrhizal interactions, computed the dissimilarity between pairs of samples (using Bray–Curtis distances), and reconstructed the dendrogram using neighbor-joining: two plant root samples that are close in the dendrogram tend to have similar fungal compositions. Branches are colored according to the sampled community (a) or to the plant taxonomic group (b). For each dendrogram, we also indicated the results of the PERMANOVA (R² and p-value based on 10,000 permutations) testing the effect of the sampled community (top row) or the plant taxonomic groups (bottom rows) on the Bray–Curtis diversity between root samples

Fig. 4

Plant-fungus network structures vary between fungal lineages, irrespectively of the environmental variations. Species-level network representation in each sampled community (Grand brûlé, Plaine-des-Palmistes, or Dimitile) for the different fungal groups: Glomeromycotina (a), Helotiales (b), Sebacinales (c), Mucoromycotina (d), or Cantharellales (e). Colored round nodes represent plant species (colors indicate the main plant taxonomic groups) and grey squared nodes correspond to fungal OTUs. Grey links represent plant-fungus interactions and their widths are proportional to interaction abundances. The position of the nodes reflects the similarity in species interactions using the Fruchterman-Reingold layout algorithm [100] from the igraph R-package. Fungal lineages (in rows) are ordered according to their network structures: networks that tend to be nested are at the top, whereas networks that tend to be modular are at the bottom (Additional file 1: Tables S6–S8). The sampled communities are in columns and we indicated the environmental gradients on the top. For each network, the total read abundances (L), the connectance (C), and the ratio of interactions within modules (Q) are indicated. Q is computed from the most modular structure according to Beckett’s algorithm for abundance networks, Q close to 0 indicates that most interactions are between modules (i.e. low modularity), while Q close to 1 indicates that most interactions are within modules (i.e. high modularity). Significant connectance and modularity values, evaluated using shuffle-sample null models, are highlighted in bold. We did not report nestedness values as they are not meaningful when compared across networks of different sizes. Details about the fungal taxonomy can be seen in Additional file 1: Figure S18

Fig. 5

Fungal sharing in the plant-fungus networks varies across the different fungal lineages. a Interaction specializations (H₂′) are lower in plant-Glomeromycotina networks than in other plant-fungus networks. For each plant-fungus network (with Glomeromycotina, Mucoromycotina, Sebacinales, Helotiales, or Cantharellales) in each sampled community (Grand brûlé (A), Plaine-des-Palmistes (B), or Dimitile (C)), a colored dot indicates the network-level interaction specialization (H₂′). The significance of the H₂′ values was evaluated using null models maintaining marginal sums or shuffle-sample null models: all the H₂′ values were significant for the marginal sums null models, and asterisks indicate when the H₂′ values are significant, based on the shuffle-sample null models (see Additional file 1: Fig. S11 for details). b Motif frequencies significantly differ between the plant-fungus networks. Principal coordinate analyses (PCoA) of the bipartite motif frequencies (the “building blocks” of the network containing from 2 to 5 species) of each plant-fungus network (Glomeromycotina, Mucoromycotina, Sebacinales, Helotiales, or Cantharellales) in each sampled community (Grand brûlé (A), Plaine-des-Palmistes (B), or Dimitile (C)). The colored triangle areas represent the proximity within the sampled communities for the different groups of fungi