Assessing the Cultivability of Bacteria and Fungi from Arable Crop Residues Using Metabarcoding Data as a Reference

International audience; This study combined culture-dependent (strain isolation plus molecular identification) and culture-independent (metabarcoding) approaches to characterize the diversity of microbiota on wheat and oilseed rape residues. The goal was to develop a methodology to culture microorganisms with the aim of being able to establish synthetic crop residue microbial communities for further study, i.e., testing potential interactions within these communities and characterizing groups of beneficial taxa that could be used as biological control agents against plant pathogens. We generated community-based culture collections. We adapted the isolation strategy to the potential differences in the spatial and temporal distribution of diversity between bacteria and fungi. We performed (i) a high-throughput isolation from few samples with no a priori for bacteria and (ii) a low-throughput isolation from several samples with a priori—i.e., morphotype selection—for fungi. Although isolation using a single medium did not allow us to characterize the microbiome as precisely as metabarcoding, the bacterial diversity (158 ASVs, 36 genera) was relatively higher than the fungal diversity (131 ASVs, 17 genera) known to be limited by competition for growth on non-selective solid media. Isolation and metabarcoding provided consistent and complementary information: they revealed several common but also specific ASVs, leading to close microbial community profiles of the most abundant fungal and bacterial taxa in residues. Finally, by empirically comparing the different profiles, we assessed the cultivability of the most abundant fungal and bacterial taxa obtained in metabarcoding.