Metagenomic signatures of balancing selection in the human gut

Bacterial lineages that populate the human gut microbiota contend with spatial and temporal fluctuations in numerous environmental variables, including bouts of extreme selective agents such as antibiotics. Oscillations in the adaptive landscape can impose balancing selection on populations, leaving characteristic signatures in the sequence variation of functionally significant genomic loci. Despite their potential importance for gut bacterial adaptation, the metagenomic targets of balancing selection have not been identified. Here, I present population genetic evidence that balancing selection maintains allelic diversity in multidrug efflux pumps of multiple predominant gut bacterial species. Metagenome-wide scans of 566,958 genes from 287 bacterial species represented by 118,617 metagenome-assembled genomes indicated that most genes have been conserved by purifying selection. However, dozens of core open reading frames (CORFs) displayed positive Tajima's D values that deviated significantly from their species' genomic backgrounds, indicating the action of balancing selection. Multidrug efflux pumps (MEPs) from a diversity of bacterial species were significantly enriched among the CORFs with Tajima's D values >3 in industrialized, but not nonindustrialized, human populations. The AcrB subunit of an MEP from Bacteroides dorei displayed the highest Tajima's D of any CORF. Divergent haplotypes of this CORF displayed evidence of positive selection and homology to an Escherichia coli AcrB subunit that binds tetracycline and macrolide antibiotics, suggesting functional significance and implicating medical antibiotics as an agent of selection acting on this locus. Other proteins identified as targets of balancing selection included peptidoglycan/LPS O-acetylases and ion transporters. Intriguingly, the degree of balancing selection acting on gut bacterial species was associated with species abundance in the gut based on metagenomic data, further suggesting fitness benefits of the allelic variation identified.