Patience is a virtue for tolerant bacteria

13 March 2015
By Kate Baker

Researchers produced three strains of <em>E. coli</em> bacteria tolerant to antibiotics. Credit: David Gregory and Debbie Marshall, Wellcome Images

Researchers produced three strains of E. coli bacteria tolerant to antibiotics.
Credit: David Gregory and Debbie Marshall, Wellcome Images

Treatment failure of antibiotics for bacterial infections represents a global public health crisis. The most well-known issue is antimicrobial resistance, which is identifiable by an elevation in the amount of antibiotic required to stop bacterial growth. Another problem that can result in antibiotic treatment failure is that of antibiotic tolerance.

Tolerant bacteria can survive exposure to elevated concentrations of antibiotics for a limited period of time, with no overall change to the concentration of antibiotics needed to stop the bacteria growing.

Antibiotic tolerance has been linked with treatment failure in those infected with Pseudomonas aeruginosa, a bacterium that often causes blood infections, pneumonia and other complications in hospital patients with weakened immune systems, and Candida albicans, a type of yeast that causes infections such as thrush. Although both the resistance and tolerance of these bacteria to antibiotics are important factors in treatment, the latter is much less studied.

To meet the challenge of safeguarding antibiotics as a therapeutic option, it is important to study antibiotic tolerance and the mechanisms that underpin it. Antibiotic tolerance has previously been linked to small subpopulations of dormant microbial cells, known as persisters, that are not active at the time of antibiotic exposure. However, a recent study identified a further evolutionary mechanism behind tolerance and used genomics to identify the genetic changes responsible.

In the study, researchers from The Hebrew University and the Broad Institute produced three strains of E. coli bacteria tolerant to antibiotics by cyclically exposing them to high concentrations of the beta-lactamase antibiotic ampicillin for periods of three, five and eight hours every 24 hours. After eight to ten cycles, the bacteria were tolerant to ampicillin, as well as an antibiotic from a different antibiotic class, the quinolone norfloxacin.

Researchers could tell that the bacteria were tolerant because the time it took to kill 99 per cent of the bacterial cells increased, whereas the effective concentration remained unchanged.

The altered time period was not associated with changes in cell-doubling time during exponential growth or an increased proportion of persister cells. Using a recently developed automated microscopy monitor (ScanLag1), the authors were able to attribute the enhanced survival to the increased amount of time it took for cells to begin to grow vigorously after exposure to antibiotics, the lag time.

Remarkably, the increase in lag-time was dependent on the length of time each strain was exposed to antibiotics, indicating that the tolerance was customised to the original selection pressure.

The authors named this mechanism tolerance-by-lag and went on to explore the genetic mechanisms behind the phenotype. Performing whole-genome sequencing of individual clones from each of the evolved strains, the authors identified eight new mutations in six genes.

By introducing the genetic changes they saw in the tolerant strains into the original strains, the researchers confirmed that mutations in three of the six genes conferred tolerance-by-lag. The systems of two of the genes (the antitoxin homolog vapB and tRNA-synthetase metG) have been previously associated with persisters but the authors also identified a further pathway involving the essential prs gene, which encodes ribose-phosphate diphosphokinase, an enzyme involved in forming genetic letter bases. In this way, the authors identified and confirmed the genetic changes responsible for tolerance-by-lag.

The clinical significance of tolerance-by-lag will undoubtedly be investigated in subsequent studies. In the interim however, we should consider that antibiotic tolerance goes largely undetected in routine clinical microbiological testing, and that tolerance-by-lag, would be readily identifiable.

As well as uncovering another potential basis for antibiotic treatment failure, this study highlights the benefits of using whole-genome sequencing of bacteria to investigate antibiotic tolerance and resistance within bacterial populations.

Kate Baker is a Postdoctoral Fellow in the Pathogen Genomics group at the Wellcome Trust Sanger Institute, working on the global molecular epidemiology of enteric pathogens and their antimicrobial resistance determinants


  • Lewis K (2010). Persister cells. Annual review of microbiologyDOI:10.1146/annurev.micro.112408.134306
  • Fridman O, et al (2014). Optimization of lag time underlies antibiotic tolerance in evolved bacterial populations. NatureDOI:10.1038/nature13469
  • Levin-Reisman I, et al (2010). Automated imaging with ScanLag reveals previously undetectable bacterial growth phenotypes. Nature MethodsDOI:10.1038/nmeth.1485
  • Kaspy I, et al (2013). HipA-mediated antibiotic persistence via phosphorylation of the glutamyl-tRNA-synthetase. Nature CommunicationsDOI:10.1038/ncomms4001

Related Links: