Drug-target binding quantitatively predicts optimal antibiotic dose levels

Abstract

Antibiotic treatment may be compromised by the sporadic appearance and selection of drug resistant mutants during therapy. Identifying optimal dosing strategies for treating bacterial infections that minimize the risk of resistance is difficult, and improving dosing guidelines usually requires long and costly in vitro and in vivo experimentation. Previously, we proposed a mathematical model that links bacterial population biology with chemical reaction kinetics and demonstrated that this model had high predictive and explanatory power for antibiotic pharmacodynamics. Here, we extend our model to incorporate several distinct molecular mechanisms of resistance, e.g. altered drug-target affinity or target molecule content, to explore the how these mechanisms affect the risk of acquired resistance during treatment.Our model is based on a system of partial and ordinary differential equations. The central assumption encoded in the model is that bacterial growth declines (and/or bacterial kill rate increases) as the fraction of bound antibiotic target molecules within the bacterium increases. We use experimental data from E. coli exposed to ciprofloxacin and ampicillin to estimate the dependence of bacterial growth and death on the fraction of bound target molecules and to calibrate our model. We then predict how several alternative molecular mechanisms of resistance should affect bacterial growth and compare these predictions with data from experimentallymanipulated bacteria with varying target molecule content. Finally, we use our model to estimate antibiotic susceptibility for different antibiotics that share similar mechanisms of action but differ in their drug-target affinity.Our extended model offers a novel approach for predicting how resistance will evolve during therapy and provides a framework for generating new testable hypotheses about how the risk of resistance depends on the molecular mechanism of resistance.All rights reserved. No reuse allowed without permission.

Mechanisms of antibiotic action shape the fitness landscapes of resistance mutations

“…The affinity K D of the drug is 125 thus the ratio of off-rate to on-rate, K D = k R /k F . The model assumes that the growth and death 126 rates of a bacterial cell depend on its drug-target occupancy (that is, the number of inactivated 127 drug-target complexes it contains) (Clarelli et al, 2019;Wiesch et al, 2015). We denote drug-128 target occupancy with the index i, which ranges from 0 to N. Cells harboring successively 129 larger numbers of inactivated drug-target complexes have successively faster death rates curves derived from the model for a wild-type (light green) and drug-resistant (dark green) 152 bacterial strain.…”

vCOMBAT: a Novel Tool to Create and Visualize a COmputational Model of Bacterial Antibiotic Target-binding

“…Based on the extended model, we have developed an interactive web-based tool, namely vCOMBAT, to allow non-quantitative scientists to create and visualize their own computational models of bacterial antibiotic target-binding. In contrast to our previously developed COMBAT modeling framework [18], this tool allows to incorporate antibiotic time-concentration profiles measured in patients. The tool can inform optimal dosing strategies based on antibiotic and bacteria data provided by the users.…”