Fitness costs of antibiotic resistance mutations in rich media do not reliably predict those in infection-like environments
Knowles, J.; Pittaccio, L. M.; Lindon, S.; Woods, C.; Gifford, D. R.; Kahl, L. J.; MacLean, R. C.; Cameron, S. J.; Wheatley, R. M.
Show abstract
The acquisition of antibiotic resistance is typically associated with a fitness cost, reducing the competitive ability of cells in the absence of antibiotics. Our ability to predict these costs informs our ability to predict how likely it is that resistance will subside in the absence of treatment. Fitness costs are typically measured in laboratory cultures using a nutritionally rich medium for growth. Here, we tested the impact of infection-like environments on the prevalence and magnitude of fitness costs associated with antibiotic resistance mutations in the opportunistic pathogen Pseudomonas aeruginosa. We assessed two different clinically relevant infection niches, the lung and urinary tract, through the use of infection-niche mimicking media that models the chemical compositions of these niches. The fitness costs of resistance mutations in rich media did not reliably predict those in infection-like environments, on average costs were increased in the lung-mimicking environment and costly mutations were less prevalent in the urinary tract-mimicking environment. Synthetic human urine significantly enhanced the fitness of P. aeruginosa with resistance mutations in nfxB, encoding a multidrug efflux pump transcriptional repressor, and we identified oxalate as the likely component driving this effect. Our findings highlight the importance of environmental chemical composition for shaping the fitness costs of resistance in P. aeruginosa and suggest that nfxB mutants may have a selective advantage in the urinary tract, mediated by the presence of oxalate.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Frequency of quorum sensing mutations in Pseudomonas aeruginosa strains isolated from different environments 93%
- Biofilm hydrophobicity in environmental isolates of Bacillus subtilis 92%
- ChpC controls twitching motility-mediated expansion of Pseudomonas aeruginosa biofilms in response to serum albumin, mucin and oligopeptides 91%
Similar papers in this journal
- When does a Lotka-Volterra model represent microbial interactions? Insights from in-vitro nasal bacterial communities 94%
- Collateral sensitivity interactions between antibiotics depend on local abiotic conditions 92%
- A model Roseobacter employs a diffusible killing mechanism to eliminate competitors 92%
Similar papers in this journal
- Spheroplast-mediated carbapenem tolerance in Gram-negative pathogens 94%
- Weakest link dynamics predict apparent antibiotic interactions in a model cross-feeding community. 94%
- Evolution of β-lactam resistance causes fitness reductions and several cases of collateral sensitivities in the human pathogen Haemophilus influenzae 93%
Similar papers in this journal
- Critical role of growth medium for detecting drug interactions in Gram-negative bacteria that model in vivo responses 94%
- Analysis of the oxidative stress regulon identifies soxS as a genetic target for resistance reversal in multi-drug resistant Klebsiella pneumoniae 93%
- Parallel evolution of tobramycin resistance across species and environments 93%
Similar papers in this journal
- Strain background, species frequency and environmental conditions are important in determining population dynamics and species co-existence between Pseudomonas aeruginosa and Staphylococcus aureus 93%
- Long-term Culturing of Pseudomonas aeruginosa in Static, Minimal Nutrient Medium Results in Increased Pyocyanin Production, Reduced Biofilm Production, and Loss of Motility 93%
- Benzoate-Tolerant Escherichia coli Strains from Experimental Evolution Lose the Gad Regulon, Multi-Drug Efflux Pumps, and Hydrogenases 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.