SummaryPhenotypically-switched, antibiotic-refractory persisters may prevent pathogen eradication. Although how triggered persistence via starvation-induced (p)ppGpp is well characterized, generation of persisters without starvation are poorly understood. Here we visualized the formation of spontaneous persisters in a small fraction of cells from growing wild type bacteria, revealing a striking single cell rapid switch from growth to dormancy. This switch-like entrance is triggered by GTP dropping beneath a threshold due to stochastic production and self-amplification of (p)ppGpp via allosteric enzyme activation. In addition, persisters are induced by lethal and sublethal concentrations of cell wall antibiotics by inducing (p)ppGpp via cell wall stress response. Thus spontaneous, triggered and antibiotic-induced persisters can all stem from a common metabolic switch: GTP depletion by (p)ppGpp induction, and each pathway of persister formation is activated by different (p)ppGpp synthetases. These persistence pathways are likely conserved in pathogens which may be exploited to potentiate antibiotic efficacy.

Antibiotics combat bacteria through their bacteriostatic (by growth inhibition) or bactericidal (by killing bacteria) action. Mechanistically, it has been proposed that bactericidal antibiotics trigger cellular damage, while bacteriostatic antibiotics suppress cellular metabolism. Here, we demonstrate how the difference between bacteriostatic and bactericidal activities of the antibiotic chloramphenicol can be attributed to an antibiotic-induced bacterial protective response: the stringent response. Chloramphenicol targets the ribosome to inhibit the growth of the Gram-positive bacterium Bacillus subtilis. Intriguingly, we found that chloramphenicol becomes bactericidal in B. subtilis mutants unable to produce (p)ppGpp. We observed a similar (p)ppGpp-dependent bactericidal effect of chloramphenicol in the Gram-positive pathogen Enterococcus faecalis. In B. subtilis, chloramphenicol treatment induces (p)ppGpp accumulation through the action of the (p)ppGpp synthetase RelA. (p)ppGpp subsequently depletes the intracellular concentration of GTP and antagonizes GTP action. This GTP regulation is critical for preventing chloramphenicol from killing B. subtilis, as bypassing (p)ppGpp-dependent GTP regulation potentiates chloramphenicol killing, while reducing GTP synthesis increases survival. Finally, chloramphenicol treatment protects cells from the classical bactericidal antibiotic vancomycin, reminiscent of the clinical phenomenon of antibiotic antagonism. Taken together, our findings suggest a role of (p)ppGpp in the control of the bacteriostatic and bactericidal activity of antibiotics in Gram-positive bacteria, which can be exploited to potentiate the efficacy of existing antibiotics.