History shapes regulatory and evolutionary responses to tigecycline in strains of Acinetobacter baumannii from the pre- and post-antibiotic eras

Evolutionary history encompasses genetic and phenotypic bacterial differences, but the extent to which history influences drug response and antimicrobial resistance (AMR) adaptation is unclear. Historical contingencies arise when elements from an organisms past leave lasting effects on the genome, altering the paths available for adaptation. We utilize strains isolated before and after widespread antibiotic use to study the impact of deep historical differences shaped by decades of evolution in varying antibiotic and host pressures. We evaluated these effects by comparing immediate and adaptive responses of two strains of Acinetobacter baumannii to the last-resort antibiotic, tigecycline (TGC). When grown in subinhibitory TGC, the two strains demonstrated divergent transcriptional responses suggesting that baseline transcript levels may dictate global responses to drug and their subsequent evolutionary trajectories. Experimental evolution in TGC revealed clear differences in population-genetic dynamics - with hard sweeps in populations founded by one strain and no mutations reaching fixation in the other strain. Transcriptomes of evolved populations no longer showed signatures of drug response, as was seen in the ancestors, suggesting that genetic adaptation may outweigh preexisting differences in transcriptional networks. Genetically, AMR was acquired through predictable mechanisms of increased efflux and drug target modification; however, the two strains adapted by mutations in different efflux regulators. Fitness tradeoffs of AMR were only observed in lineages evolved from the pre-antibiotic era strain, suggesting that decades of adaptation to antibiotics resulted in preexisting compensatory mechanisms in the more contemporary isolate, an important example of a beneficial effect of historical contingencies. SIGNIFICANCE STATEMENTAcinetobacter baumannii is a high priority pathogen often causing multidrug resistant nosocomial infections. Many healthcare systems experience clonal outbreaks of A. baumannii infections, yet treatment strategies are often strain-agnostic, ignoring the importance of strain differences. We show that historical differences between two strains, one isolated prior to widespread antibiotic use and the other following decades of selection to clinical conditions, dictate transcriptional patterns and response to a last-resort antibiotic and influence the genetic and phenotypic routes of resistance adaptation. While our study focuses on two reference strains of A. baumannii, these findings can be more broadly applicable to other pathogenic organisms in which a better understanding of the forces influencing resistance adaptation is essential for combating the antimicrobial resistance crisis.