Basins of attraction of microbiome structure and soil ecosystem functions

Abstract

Archaea, bacteria, and fungi in the soil are increasingly recognized as major determinants of agricultural productivity and sustainability. Although our knowledge of soil biodiversity has been greatly enhanced for the last decade, defining key characteristics of stable microbiome structure and their ecosystem functions remains a major challenge. We here show that stable states of soil microbiomes can be defined with their disease-suppressive effects on crop plants. We integrated agricultural field metadata with microbial community analyses by targeting > 2,000 agricultural soil samples collected across the Japanese Archipelago, profiling microbiomes constituted, in total, by 579 archaeal, 26,640 bacterial, and 6,306 fungal species/strains. A DNA-metabarcoding-based evaluation of balance between prokaryote and fungal abundance indicated that fungus-dominated soil is more resistant to crop plant disease. A statistical physics approach for reconstructing the energy landscape of community structure indicated that disease-suppressive and disease-susceptible soils could be distinguished as alternative stable states of fungal communities. A network theoretical analysis further indicated that the entire microbiomes were partitioned into several substructures differing in potential impacts on crop disease level. We then highlighted potential keystone prokaryotes and fungi governing ecological processes within/across such substructures. These results suggest that agroecosystem performance and sustainability can be managed by identifying key structure and species within microbiomes.