Siderophore synthetase-receptor gene coevolution reveals habitat- and pathogen-specific bacterial iron interaction networks

Predicting bacterial social interactions from genome sequences is notoriously difficult. Here, we developed bioinformatic tools to predict whether secreted iron-scavenging siderophores stimulate or inhibit the growth of community members. Siderophores are chemically diverse and can be stimulatory or inhibitory depending on whether bacteria possess or lack corresponding uptake receptors. We focused on 1928 representative Pseudomonas genomes and developed a co-evolution algorithm to match all encoded siderophore synthetases to corresponding receptor gene groups with >90% accuracy based on experimental validation. We derived community-level iron interaction networks to show that selection for siderophore-mediated interactions differs across habitats and lifestyles. Specifically, dense networks of siderophore sharing and competition were observed among environmental (soil/water/plant) strains and non-pathogenic species, while only fragmented networks occurred among human-derived strains and pathogenic species. Altogether, our sequence-to-ecology approach empowers the analyses of social interactions among thousands of bacterial strains and uncovers ways for targeted intervention to microbial communities.