Abstract
Phages shape the structure of natural bacterial communities and can be effective therapeutic agents. Bacterial resistance to phage infection, however, limits the usefulness of phage therapies and could destabilise community structures, especially if individual resistance mutations provide cross-resistance against multiple phages. We currently understand very little about the evolution of cross-resistance in bacteria–phage interactions. Here we show that the network structure of cross-resistance among spontaneous resistance mutants of Pseudomonas aeruginosa evolved against each of 27 phages is highly modular. The cross-resistance network contained both symmetric (reciprocal) and asymmetric (nonreciprocal) cross-resistance, forming two cross-resistance modules defined by high within- but low between-module cross-resistance. Mutations conferring cross-resistance within modules targeted either lipopolysaccharide or type IV pilus biosynthesis, suggesting that the modularity of cross-resistance was structured by distinct phage receptors. In contrast, between-module cross-resistance was provided by mutations affecting the alternative sigma factor, RpoN, which controls many lifestyle-associated functions, including motility, biofilm formation, and quorum sensing. Broader cross-resistance range was not associated with higher fitness costs or weaker resistance against the focal phage used to select resistance. However, mutations in rpoN, providing between-module cross-resistance, were associated with higher fitness costs than mutations associated with within-module cross-resistance, i.e., in genes encoding either lipopolysaccharide or type IV pilus biosynthesis. The observed structure of cross-resistance predicted both the frequency of resistance mutations and the ability of phage combinations to suppress bacterial growth. These findings suggest that the evolution of cross-resistance is common, is likely to play an important role in the dynamic structure of bacteria–phage communities, and could inform the design principles for phage therapy treatments.
Highlights
Natural microbial communities are comprised of complex networks of species interactions, wherein each species may be engaged in ecological interactions with many other species [1,2,3]
Phage therapy is a promising alternative to antibiotics for treating bacterial infections
The extent of cross-resistance provided by a given resistance mutation is likely to depend on the genetic correlations between bacterial resistance traits selected by the different phage species
Summary
Natural microbial communities are comprised of complex networks of species interactions, wherein each species may be engaged in ecological interactions with many other species [1,2,3]. Whereas the statistical structure of interactions in bacteria– phage networks has been well studied [8], the structure and underlying genetic basis of crossresistance networks remain poorly understood. This considerably limits our ability to predict how cross-resistance evolution affects bacteria–phage communities across different environmental, agricultural, and clinical contexts. Even though pleiotropic costs should limit the evolution of generalist resistance, cross-resistance is commonly observed [14,15] Most of this evidence is, based on relatively simple phage communities, and it is less clear how the range of cross-resistance provided by different resistance mutations is related to the magnitude of fitness costs in more complex bacteria–phage networks
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
