Molecular noise accelerates cell division and death under antibiotic treatment

ABSTRACT It is well known that microbial cell populations can exhibit sustained exponential growth. More surprising is the fact that at high antibiotic levels, cell populations exhibit sustained exponential decay over several orders of magnitude. The boundary between growth and decay occurs at the Minimal Inhibitory Concentration (MIC) of the antibiotic, where the density of living cells remains constant over time. These observations suggest that positive (growth) or negative (decay) exponents arise as a difference of cell division and death rates obeying first-order kinetics. Thus for antibiotic concentrations below MIC, division dominates; for concentrations above MIC, death dominates; while MIC itself is a dynamic steady state of balanced division and death, rather than cell stasis. To measure these rates we separately tracked living and dead cells in Escherichia coli populations treated with the ribosome-targeting antibiotic kanamycin. We found that cells divide rapidly even at MIC: inferred division and death rates at MIC are 0.6 times the antibiotic-free division rate. A stochastic model of cells as collections of self-replicating units we term “widgets” reproduces both steady-state and transient features of our experiments, and explains first-order exponential kinetics. In this model cell division and death rates at MIC can be tuned from low to high values by amplifying molecular noise in the synthesis, degradation and partitioning of the widgets. At extremely low noise, cells approach the classic bacteriostatic limit at MIC: neither dividing nor dying. Noise-induced division and death of cells following antibiotic treatment could increase the likelihood of sepsis and antibiotic resistance.