Drug detoxification dynamics explain the postantibiotic effect

Abstract The postantibiotic effect (PAE) refers to the temporary suppression of bacterial growth following transient antibiotic treatment. This effect has been observed for decades for a wide variety of antibiotics and microbial species. However, despite empirical observations, a mechanistic understanding of this phenomenon is lacking. Using a combination of modeling and quantitative experiments, we show that the PAE can be explained by the temporal dynamics of drug detoxification in individual cells after an antibiotic is removed from the extracellular environment. These dynamics are dictated by both the export of the antibiotic and the intracellular titration of the antibiotic by its target. This mechanism is generally applicable for antibiotics with different modes of action. We further show that efflux inhibition is effective against certain antibiotic motifs, which may help explain mixed cotreatment success.

.  Figure S4. Correlation between RT pop and RT cell is robust to ribosome degradation rate .. 9 Appendix Figure S5.
Protein synthesis inhibitor, binds to 50S ribosomal subunit Appendix Figure S1. Fluorescence luminescence reporters accurately reflect density following antibiotic treatment.
A. The total GFP signal of a population increases approximately linearly with the cell density (as measured by CFU counts) after 2-hour antibiotic treatment. Here, colors indicate increasing streptomycin concentrations (0, 2, 4, 6, 8, 10, 12 μg/mL). Dashed line indicates linear fit; all error bars show the standard deviation of four replicates.
B. The total luminescence of a population increases approximately linearly with the cell density (as measured by CFU counts) after 2-hour antibiotic treatment. A. Concentration dependent streptomycin killing rates were determined in the absence of population growth. Overnight cultures were diluted 10-fold and grown in M9 media to a high density. Cells were then incubated in minimal media with the indicated antibiotic concentration for increasing durations. Population viability decreased exponentially with dose duration. The control population showed negligible growth/death in these conditions, indicating that decrease in viability is due to cell killing. All data points show mean and standard deviation of 6 replicates; legend shows streptomycin concentrations in μg/mL. B. Correcting for partial population death results in a lower effective recovery time (purple data points). However, the log-linear dependence of the recovery time on the total antibiotic exposure remains.  A. Periodic dosing regimens can be specified by four key parameters: the antibiotic concentration , the dosing interval ! , the time between doses ! , and the total number of doses, !"#$ . We denote the recovery time following n doses as ! . B. Dosing parameters dictate treatment efficacy. Here, the heat map colors indicate recovery time after !"#$ = 10, for each combination of ! and ! ; yellow corresponds to large recovery time (i.e. a successful treatment) and blue corresponds to a low recovery time (i.e. treatment failure). C.
! predicts the efficacy of -dose treatments. Here, each trajectory corresponds to one value of ! , and colors correspond to increasing values of !"#$ ; each data point corresponds to a single multi-dose treatment at a fixed antibiotic concentration . There is a sharp transition in treatment efficacy (as measured by ! ) at roughly ! ! !! ! = 1; that is, the recovery time in response to a single dose corresponds to the maximum value of ! that results in a successful multi-dose treatment. D. RTn is a function of total antibiotic exposure. Multi-dose recovery exhibits a biphasic response to dosing frequency, wherein populations are able to recover faster at intermediate frequencies, regardless of the total number of doses (black arrow). Here, we assume that ! = ! for all doses, and thus frequency = A. Dose response for carbonyl cyanide m-chlorophenyl hydrazine (CCCP). Seven logarithmically spaced concentrations were chosen, in addition to 0. Growth rates were calculated by log-transforming time series data and calculating the slope of the longest linear segment. Data points are the average of three replicates; black line shows the best fit Hill function (IC50 = 3.87µg/mL).
B. CCCP increases ethidium bromide (EtBr) accumulation, and inhibits antibiotic efflux, in a dosedependent fashion. A previously published protocol was used to measure the effect of CCCP on efflux pump activity, in the absence of growth. Intracellular EtBr, which acts as a general efflux pump substrate, can be measured via population fluorescence; accumulation is increased in the presence of increasing CCCP concentrations.
C. Efflux rate decreases with the addition of CCCP. Rates were calculated by fitting fluorescence time series data to Equation (3). Data points are the average of eight replicates. RT(k eff,1 ) Appendix Figure S8. Determining IC50 values for various antibiotics. IC50 values were calculated from long-term dose responses in 96-well plates. Antibiotics were serially diluted to create a concentration gradient; cells were initially diluted 100-fold from overnight culture (~2e7 CFU/mL). Growth curves were collected using a Tecan M200 plate reader measuring OD600 every ten minutes. Each data point corresponds to the average of three technical replicates. IC50 values are shown in parentheses and were determined by fitting growth rates to a Michaelis-Menten equation (