Importance of microbial communities at the root-soil interface for extracellular polymeric substances and soil aggregation in semiarid grasslands

In the past years, extracellular polymeric substances (EPS) produced by soil microorganisms received an increasing interest, as they not only protect microbes against environmental stresses, but seem to play a pivotal role in soil structure formation as well. Within soils, root deposits provide an important source of easily accessible energy and nutrients, stimulating microbial growth to produce EPS. Especially under semiarid climates, where a full vegetation cover cannot be sustained, large gradients in living conditions for microbes can be found between the root-soil interface and barren intercanopy spaces. In this study, we aimed to elucidate the plant-specific effects on microbes, EPS production and soil aggregation. At two sites in southern Spain, differing in carbonate and graphite content, legume shrubs of Anthyllis cytisoides and grass tussocks of Macrochloa tenacissima were selected. Soil samples were taken in the adjacent bare interspace, under the canopy and of the rhizosphere. From these samples the microbial community (here bacteria and archaea), EPS(-saccharide) content and soil aggregation (<1 mm) were analysed. DNA extracted from the microbial cells detached from the surface of the sampled roots (rhizoplane), was subjected to 16S rRNA gene amplicon sequencing. The rhizoplane microbial communities differed strongly between plant species and sites, whereby site was the most important factor shaping the communities. The plant species effect on microbial communities diminished strongly with distance to the root surface. At the carbonate-poor Rambla Honda site (site 1), plant species-specific effects were observed in the rhizoplane and rhizosphere, whereas in the carbonate-rich Alboloduy site (site 2) almost no plant species-specific effects were found at the genus level. The larger heterogeneity in microbial communities at site 1 was reflected in EPS-saccharide contents and subsequent soil aggregation, while no difference in soil aggregation was found at site 2. Both parameters increased strongest in the Anthyllis cytisoides rhizosphere at site 1. Despite the lack of a strong gradient with distance from the root at the carbonate-rich site 2, microbial taxa were found by network analysis that positively correlated to EPS-saccharide contents and/or soil aggregation. The relationship between the identified taxa and EPS and/or aggregation relationships were clearest at the root-soil interface, while several other taxa were found to be widely occurring in the other soil compartments too. In conclusion, we found in all compartments potential EPS producers, which could have influenced soil aggregation. Nevertheless, microbes with higher relative abundance in the rhizoplane were linked to higher EPS contents, especially in conjunction with legume shrubs, and subsequently related to soil aggregation. The spatial extent of the root effect was only governed by carbonate contents, as higher carbonate content diminished the observed root effects on the microbial community and subsequent soil aggregation.