First-step mutations reveal a multitude of evolutionary pathways to antibiotic resistance
Bacterial pathogens have evolved a plethora of different adaptations that confer high-level resistance to certain antibiotics. However, it remains unclear whether the observed adaptations are randomly drawn from a much larger set of possibilities or represent the best of only a few options for obtaining resistance. Using a gradient plate platform, we screened 12 strains of bacteria against 22 antibiotics to enumerate the set of single-step genetic mutations capable of conferring antibiotic resistance. Our screening technique allows for the detection of mutants that can survive at antibiotic concentrations only slightly above the minimum inhibitory concentration (MIC) and mutations that may incur a fitness cost. We found that single-step resistant mutants were widespread, occurring in 80% of the antibiotic-strain combinations tested. Resistance occurred more frequently to aminoglycoside and bactericidal antibiotics. We experimentally determined the frequency of resistant mutants in cases where multi-fold changes in resistance level were observed, enabling us to estimate the number of genome positions responsible for beneficial first-step mutations. This was confirmed using whole genome sequencing to identify the mutations directly responsible for imparting resistance. We discovered that the target space for genomic alterations is incredibly large for some antibiotics, while for others, it appears that there are no single-nucleotide mutations able to confer resistance. To our knowledge, this is the first screen to enumerate the repertoire of single-step mutations that impart high-level antibiotic resistance, deepening our understanding of how resistance may begin to emerge in the clinic.