The fitness costs of antibiotic resistance mutations

Abstract

Antibiotic resistance is increasing in pathogenic microbial populations and is thus a major threat to public health. The fate of a resistance mutation in pathogen populations is determined in part by its fitness. Mutations that suffer little or no fitness cost are more likely to persist in the absence of antibiotic treatment. In this review, we performed a meta-analysis to investigate the fitness costs associated with single mutational events that confer resistance. Generally, these mutations were costly, although several drug classes and species of bacteria on average did not show a cost. Further investigations into the rate and fitness values of compensatory mutations that alleviate the costs of resistance will help us to better understand both the emergence and management of antibiotic resistance in clinical settings.

Keywords: adaptation; antibiotic resistance; fitness.

Figure 1

Estimation of selection coefficients using the Lenski and Dykhuisen estimators. Competition experiments were simulated for two strains, with the wild-type strain doubling in each time unit, and the growth rate of the focal strain reduced compared to wild type by 0 to 0.25. Competition lasted six generations, starting with a 50:50 ratio of the two strains and an initial population size of 1 million. Each data point represents the mean values of s for 100 simulations. For each replicate, an average of 100 individuals were sampled and used to calculate s_l (y-axis) and s_d (x-axis). The dashed line represents a 1:1 relationship, and the solid line gives the linear regression of s_l on s_d.

Figure 2

The mean relative fitness and 95% confidence intervals of antibiotic resistance mutations associated with a given antibiotic, grouped by class of antibiotic (from top to bottom alpha-pyrone, aminoglycoside, coumarin, dihydrofolate reductase inhibitor, fusidane, macrolide, quinolone and rifamycin). A fitness value of <1 indicates a fitness cost in the absence of the antibiotic.

Figure 3

The mean relative fitness and 95% confidence intervals of antibiotic resistance mutations associated with a given species. A fitness value of <1 indicates a fitness cost in the absence of the antibiotic.

Figure 4

Species–antibiotic comparisons of the mean relative fitness of resistance mutations in the absence of the antibiotic. Numbers indicate antibiotic class: 1 – alpha-pyrone, 2 – aminoglycoside, 3 – coumarin, 4 – dihydrofolate reductase inhibitor (DHRI), 5 – fusidane, 6 – macrolide, 7 – quinolone and 8 – rifamycin.

Figure 5

The level of resistance conferred by a mutation is negatively correlated with its’ fitness in the absence of the antibiotic (r² = 0.228). Different symbols are associated with different classes of antibiotic.