Towards model-informed design of antibiotic therapies: understanding the impact of antibiotics on bacterial physiology & growth.
Elena Pascual-García (1), Charlotte Kloft (2), Andrea Weiße (3), Wilhelm Huisinga (1,4).
(1) Institute of Biochemistry and Biology, Potsdam Universität, Germany; (2) Institute of Pharmacy, Freie Universität Berlin, Germany; (3) School of Biological Sciences & School of Informatics, University of Edinburgh, UK; (4) Institute of Mathematics, University of Potsdam, Potsdam, Germany
Introduction and objectives:
Antimicrobial resistance is on the rise globally. Increased levels of resistance have been reported across bacterial strains and antibiotic compounds, threatening the effectiveness of current treatment options [1]. In order to better target resistant bacteria, understanding how bacterial physiology changes in the presence of antibiotics is critical. Variations in a cell’s environmental conditions, such as the antibiotic concentration or the nutrient quantity and quality of the growth medium, affect the internal allocation of resources and result in changes in growth. Here we present a mechanistic modelling framework that quantitatively predicts antibiotic effect on bacterial growth dynamics in different environmental conditions. We focus on protein synthesis inhibitors, a family drugs that constitutes more than half of the antimicrobials used to treat infections [2].
Methods:
An established coarse-grained model of bacterial growth [3] was used as initial framework to describe baseline bacterial physiology and growth dynamics. A mechanistic description of the mode of action of ribosome-binding antibiotics was combined with the initial model. Inspired by the work described in [4], the combined model accounts for passive diffusion of the drugs through the cell membrane and the interaction of the antibiotics with their cytoplasmic target, translating ribosomes. We used the same model architecture to describe four different drugs (chloramphenicol, streptomycin, tetracycline and kanamycin), varying only in their parameter values of diffusion and binding kinetics. We distinguish between reversibly binding antibiotics (chloramphenicol, tetracycline) and irreversibly binding antibiotics (streptomycin, kanamycin). We used data on growth rates of E. coli cultures across different growth media and antibiotic concentrations [4] in the estimation of antibiotic-associated parameters. We followed a Bayesian approach using the bayesopt function package in MATLAB.
Results:
We successfully developed a model of ordinary differential equations that couples gene expression and bacterial growth with a mechanistic description of drug uptake and action of four ribosome-targeting antibiotics. The model accurately describes the growth inhibition responses of all four studied drugs and predicts that differences in the binding strength of the antibiotics to their targets affect the shape of the concentration-response curves. Furthermore, our model predicts the presence of bistability in the concentration-response curves of antibiotics that irreversibly bind mRNA-ribosome complexes (streptomycin, kanamycin). This bistability depends upon previous exposure of a cell to the drug: cells previously exposed to high antibiotic concentrations remain in a low-growth state when exposed to intermediate doses, whereas cells that have been exposed to lower antibiotic concentrations remain in a high-growth state for the same intermediate concentration. Bistability has been predicted and analyzed previously [4], but the mechanistic nature of our model allows us to study the physiological underpinnings behind this bistable behaviour and how environmental conditions affect it. Our analysis suggests that the region of bistability is sensitive to the nutrient conditions in the cell media, being more prominent in conditions that allow fast cell growth.
Conclusion:
Bistable growth dynamics may be an important aspect in the formation of heterogeneous bacterial populations such as persister or tolerant cells. Our framework can serve as a starting point to study in more detail the levels of phenotypic heterogeneity in isogenic populations of bacteria caused by the described bistable response.
References:
[1] O’Neill, J. Tackling Drug-Resistant Infections Globally: Final Report and Recommendations. Review on Antimicrobial Resistance. Wellcome Trust and HM Government; 2016.
[2] Lin, J., Zhou, D., Steitz, T. A., Polikanov, Y. S. & Gagnon, M. G. Ribosome-Targeting Antibiotics: Modes of Action, Mechanisms of Resistance, and Implications for Drug Design. Annu Rev Biochem 87, 451–478 (2018).
[3] Weiße, A. Y., Oyarzún, D. A., Danos, V. & Swain, P. S. Mechanistic links between cellular trade-offs, gene expression, and growth. Proceedings of the National Academy of Sciences of the United States of America;2015.
[4] Greulich P, Scott M, Evans MR, Allen RJ. Growth-dependent bacterial susceptibility to ribosome-targeting antibiotics. Mol Syst Biol; 2015.