Since the colonization of land by ancestral plant lineages 450 million years ago, plants and their associated microbes have been interacting with each other, forming an assemblage of species that is often referred to as a “holobiont.” Selective pressure acting on holobiont components has likely shaped plant-associated microbial communities and selected for host-adapted microorganisms that impact plant fitness. However, the high microbial densities detected on plant tissues, together with the fast generation time of microbes and their more ancient origin compared to their host, suggest that microbe-microbe interactions are also important selective forces sculpting complex microbial assemblages in the phyllosphere, rhizosphere, and plant endosphere compartments. Reductionist approaches conducted under laboratory conditions have been critical to decipher the strategies used by specific microbes to cooperate and compete within or outside plant tissues. Nonetheless, our understanding of these microbial interactions in shaping more complex plant-associated microbial communities, along with their relevance for host health in a more natural context, remains sparse. Using examples obtained from reductionist and community-level approaches, we discuss the fundamental role of microbe-microbe interactions (prokaryotes and micro-eukaryotes) for microbial community structure and plant health. We provide a conceptual framework illustrating that interactions among microbiota members are critical for the establishment and the maintenance of host-microbial homeostasis.

SummaryRoots of healthy plants are inhabited by soil-derived bacteria, fungi, and oomycetes that have evolved independently in distinct kingdoms of life. How these microorganisms interact and to what extent those interactions affect plant health are poorly understood. We examined root-associated microbial communities from three Arabidopsis thaliana populations and detected mostly negative correlations between bacteria and filamentous microbial eukaryotes. We established microbial culture collections for reconstitution experiments using germ-free A. thaliana. In plants inoculated with mono- or multi-kingdom synthetic microbial consortia, we observed a profound impact of the bacterial root microbiota on fungal and oomycetal community structure and diversity. We demonstrate that the bacterial microbiota is essential for plant survival and protection against root-derived filamentous eukaryotes. Deconvolution of 2,862 binary bacterial-fungal interactions ex situ, combined with community perturbation experiments in planta, indicate that biocontrol activity of bacterial root commensals is a redundant trait that maintains microbial interkingdom balance for plant health.

Factors that drive continental-scale variation in root microbiota and plant adaptation are poorly understood. We monitored root-associated microbial communities in Arabidopsis thaliana and cooccurring grasses at 17 European sites across three years. Analysis of 5,625 microbial community profiles demonstrated strong geographic structuring of the soil biome, but not of the root microbiota.Remarkable similarity in bacterial community composition in roots of A. thaliana and grasses was explained by the presence of a few diverse and geographically widespread taxa that disproportionately colonize roots across sites. In a reciprocal transplant between two A. thaliana populations in Sweden and Italy, we uncoupled soil from location effects and tested their respective contributions to root . CC-BY-NC-ND 4.

17Roots of healthy plants are inhabited by soil-derived bacteria, fungi, and oomycetes that have 18 evolved independently in distinct kingdoms of life. How these microorganisms interact and to 19 what extent those interactions affect plant health are poorly understood. We examined root-20 associated microbial communities from three Arabidopsis thaliana populations and detected 21 mostly negative correlations between bacteria and filamentous microbial eukaryotes. We 22 established microbial culture collections for reconstitution experiments using germ-free A. 23 thaliana. In plants inoculated with mono-or multi-kingdom synthetic microbial consortia, we 24 observed a profound impact of the bacterial root microbiota on fungal and oomycetal 25 2 community structure and diversity. We demonstrate that the bacterial microbiota is essential 26 for plant survival and protection against root-derived filamentous eukaryotes. Deconvolution 27 of 2,862 binary bacterial-fungal interactions ex situ, combined with community perturbation 28 experiments in planta, indicate that biocontrol activity of bacterial root commensals is a 29 redundant trait that maintains microbial interkingdom balance for plant health. 30 31 reconstitution experiments, we provide community-level evidence that negative interactions 51 between prokaryotic and eukaryotic root microbiota members are critical for plant host 52 survival and maintenance of host-microbiota balance. 53 54 Results 55Root-associated microbial assemblages. We collected A. thaliana plants from natural 56 populations at two neighbouring sites in Germany (Geyen and Pulheim; 5 km apart) and a 57 more distant location in France (Saint-Dié; ~300 km away) ( Figure S1; Table S1). For each 58 population, four replicates, each consisting of four pooled A. thaliana individuals were 59 prepared, together with corresponding bulk soils. Root samples were fractionated into 60 episphere and endosphere compartments, enriching for microbes residing on the root surface 61 or inside roots, respectively ( Figure S2). We characterized the multi-kingdom microbial 62 consortia along the soil-root continuum by simultaneous DNA amplicon sequencing of the 63 bacterial 16S rRNA gene and fungal as well as oomycetal Internal Transcribed Spacer (ITS) 64 regions (Agler et al. 2016) ( Table S2). Alpha diversity indices (within-sample diversity) 65 indicated a gradual decrease of microbial diversity from bulk soil to the root endosphere 66 (Kruskal-Wallis test, p<0.01; Figure S3). Profiles of microbial class abundance between 67 sample-types ( Figure 1A) and Operational Taxonomic Unit (OTU) enrichment tests 68 conducted using a linear model between soil, root episphere and root endosphere samples 69 (p<0.05, Figure 1B) identified 96 bacterial, 24 fungal and one oomycetal OTU that are 70 consistently enriched in plant roots across all three sites. This, together with the reduced alpha 71 diversity, points to a gating role of the root surface for entry into the root interior for each of 72 the three microbial kingdoms ...