Persister Escherichia coli Cells Have a Lower Intracellular pH than Susceptible Cells but Maintain Their pH in Response to Antibiotic Treatment.

Persister and viable but non-culturable (VBNC) cells are two clonal subpopulations that can survive multidrug exposure via a plethora of putative molecular mechanisms. Here, we combine microfluidics, time-lapse microscopy, and a plasmid-encoded fluorescent pH reporter to measure the dynamics of the intracellular pH of individual persister, VBNC, and susceptible Escherichia coli cells in response to ampicillin treatment. We found that even before antibiotic exposure, persisters have a lower intracellular pH than those of VBNC and susceptible cells. We then investigated the molecular mechanisms underlying the observed differential pH regulation in persister E. coli cells and found that this is linked to the activity of the enzyme tryptophanase, which is encoded by tnaA. In fact, in a ΔtnaA strain, we found no difference in intracellular pH between persister, VBNC, and susceptible E. coli cells. Whole-genome transcriptomic analysis revealed that, besides downregulating tryptophan metabolism, the ΔtnaA strain downregulated key pH homeostasis pathways, including the response to pH, oxidation reduction, and several carboxylic acid catabolism processes, compared to levels of expression in the parental strain. Our study sheds light on pH homeostasis, proving that the regulation of intracellular pH is not homogeneous within a clonal population, with a subset of cells displaying a differential pH regulation to perform dedicated functions, including survival after antibiotic treatment. IMPORTANCE Persister and VBNC cells can phenotypically survive environmental stressors, such as antibiotic treatment, limitation of nutrients, and acid stress, and have been linked to chronic infections and antimicrobial resistance. It has recently been suggested that pH regulation might play a role in an organism's phenotypic survival to antibiotics; however, this hypothesis remains to be tested. Here, we demonstrate that even before antibiotic treatment, cells that will become persisters have a more acidic intracellular pH than clonal cells that will be either susceptible or VBNC upon antibiotic treatment. Moreover, after antibiotic treatment, persisters become more alkaline than VBNC and susceptible E. coli cells. This newly found phenotypic feature is remarkable because it distinguishes persister and VBNC cells that have often been thought to display the same dormant phenotype. We then show that this differential pH regulation is abolished in the absence of the enzyme tryptophanase via a major remodeling of bacterial metabolism and pH homeostasis. These new whole-genome transcriptome data should be taken into account when modeling bacterial metabolism at the crucial transition from exponential to stationary phase. Overall, our findings indicate that the manipulation of the intracellular pH represents a bacterial strategy for surviving antibiotic treatment. In turn, this suggests a strategy for developing persister-targeting antibiotics by interfering with cellular components, such as tryptophanase, that play a major role in pH homeostasis.

Inhibition of indole production increases the activity of quinolone antibiotics against E. coli persisters.

Persisters are a sub-population of genetically sensitive bacteria that survive antibiotic treatment by entering a dormant state. The emergence of persisters from dormancy after antibiotic withdrawal leads to recurrent infection. Indole is an aromatic molecule with diverse signalling roles, including a role in persister formation. Here we demonstrate that indole stimulates the formation of Escherichia coli persisters against quinolone antibiotics which target the GyrA subunit of DNA gyrase. However, indole has no effect on the formation of E. coli persisters against an aminocoumarin, novobiocin, which targets the GyrB subunit of DNA gyrase. Two modes of indole signalling have been described: persistent and pulse. The latter refers to the brief but intense elevation of intracellular indole during stationary phase entry. We show that the stimulation of quinolone persisters is due to indole pulse, rather than persistent, signalling. In silico docking of indole on DNA gyrase predicts that indole docks perfectly to the ATP binding site of the GyrB subunit. We propose that the inhibition of indole production offers a potential route to enhance the activity of quinolones against E. coli persisters.