Evolutionary landscapes of Pseudomonas aeruginosa towards ribosome-targeting antibiotic resistance depend on selection strength

•The width of the sublethal selection window depends on the selective antibiotic.•Selection strength affects the evolution landscape of tigecycline resistance in Pseudomonas aeruginosa.•Sublethal concentrations of antibiotics select multi-drug-resistant bacteria.•Cross-resistance of antibiotic-resistant mutants is decoupled from selection strength.•Antibiotic pollution can be a risk for selecting multi-drug-resistant bacteria. It is generally accepted that antibiotic-resistant mutants are selected in a range of concentrations ranging from the minimum inhibitory concentration (MIC) to the mutant preventive concentration. More recently, it has been found that antibiotic-resistant mutants can also be selected at concentrations below MIC, which expands the conditions where this selection may occur. Using experimental evolution approaches followed by whole-genome sequencing, the current study compares the evolutionary trajectories of Pseudomonas aeruginosa in the presence of tobramycin or tigecycline at lethal and sublethal concentrations. Mutants were selected at sublethal concentrations of tigecycline (1/10 and 1/50 MIC), whereas no mutants were selected in the case of tobramycin, indicating that the width of sub-MIC selective windows is antibiotic-specific. In addition, the patterns of evolution towards tigecycline resistance depend on selection strength. Sublethal concentrations of tigecycline select mutants with lower tigecycline MICs and higher MICs to other antibiotics belonging to different structural families than mutants selected under lethal concentrations. This indicates that the strength of the cross-resistance phenotype associated with tigecycline resistance is decoupled from selection strength. Accurate information on the sublethal selection window for each antibiotic of clinical value, including the phenotypes of cross-resistance of mutants selected at each antibiotic concentration, is needed to understand the role of ecosystems polluted with different antibiotic concentrations in the selection of antibiotic resistance. Integration of this information into clinical and environmental safety controls may help to tackle the problem of antibiotic resistance.