coli K12, the majority of persister studies have focused on three bacterial taxa: Mycobacterium tuberculosis, Pseudomonas

aeruginosa, and Staphylococcus aureus. M. tuberculosis is known for its recalcitrance to antibiotic LB-100 supplier treatment [14–16], and genetic studies have shown that toxin overexpression exhibits drug-specific effects: toxins that increase persistence in one antibiotic do not necessarily increase persistence in other antibiotics [15]. This contrasts with results in E. coli K12 outlined above, in which persistence is generally characterized buy DMXAA by multidrug tolerance [9, 11]. In clinical settings, P. aeruginosa mutants that produce increased persister fractions (up to 100-fold above wildtype) have been isolated [4]; however, the genetic mechanisms causing increased persister fractions are not well understood. Finally, in S. aureus, although some research on the influence of metabolism on persister formation [17], genetic studies Metabolism inhibitor are lacking. Most studies on persister formation have focused on strains

harboring mutations that increase or decrease persister frequency. However, one recent study [18] tested how persister formation differs among strains of bacteria. In this study, mammalian commensal and pathogenic E. coli isolates were found to exhibit substantial variation in the fraction of persisters that are present in exponentially growing populations of cells. In addition, it was found that the fraction of persisters that survived treatment in one antibiotic was uncorrelated with the fraction surviving in a second antibiotic. However, without Inositol monophosphatase 1 a quantitative model of persistence, this result cannot unambiguously exclude other explanations, such as differences in the death rates of cells between isolates. Here, using a collection of environmental isolates of E. coli, we examine

variation in the frequency of persister cells in naturally occurring strains. In order to consistently measure persister fractions, we use a mathematical model to derive quantitative and reliable estimates of the fraction of persisters in each population. Our quantitative set of data corroborates the results of the previous study on commensal and pathogenic E. coli isolates [18], showing that there is substantial variation in the fraction of persister cells among environmental isolates of E. coli. In addition, we show that the fraction of cells that survive drug treatment in one drug is uncorrelated with the fraction surviving in a second drug. Importantly, we show that this lack of correlation extends to drugs have nearly identical modes of action. Finally, by using combinations of antibiotics, we provide evidence that for any single strain, there may be a subset of persister cells that are recalcitrant to treatment with any antibiotic.