Microbiome single-cell transcriptomics reveal functional heterogeneity of metabolic niches covering more than 2,500 species in the rumen

Deciphering the activity of individual microbes within complex communities and environments remains a challenge. Here, we developed microbiome single-cell transcriptomics (MscT) using droplet-based single-cell RNA sequencing and pangenome-based computational analysis to characterize the functional heterogeneity of the rumen microbiome. We generated a microbial genome database (Bovine Gastro Microbial Genome Map, BGMGM) as a functional reference map for the construction of a single-cell transcriptomic atlas of the rumen microbiome. The atlas includes 174,531 microbial cells and 2,534 species, of which 172 were core active species grouped into 12 functional clusters. We detected single-cell level functional roles, including Basfia succiniciproducens, which has a key role in the carbohydrate metabolic niche of the rumen microbiome. Furthermore, we explored functional heterogeneity and revealed metabolic niche trajectories driven by biofilm formation pathway genes within B. succiniciproducens. Our results provide a resource for studying the rumen microbiome and illustrate the diverse functions of individual microbial cells that drive their ecological niche stability or adaptation within the ecosystem.