Reliable antimicrobial susceptibility testing is essential in informing both clinical antibiotic therapy decisions and the development of new antibiotics. Mammalian cell culture media have been proposed as an alternative to bacteriological media, potentially representing some critical aspects of the infection environment more accurately. Here, we use a combination of NMR metabolomics and electron microscopy to investigate the response of Escherichia coli and Pseudomonas aeruginosa to growth in differing rich media to determine whether and how this determines metabolic strategies, the composition of the cell wall, and consequently susceptibility to membrane active antimicrobials including colistin and tobramycin. The NMR metabolomic approach is first validated by characterizing the expected E. coli acid stress response to fermentation and the accompanying changes in the cell wall composition, when cultured in glucose rich mammalian cell culture media. Glucose is not a major carbon source for P. aeruginosa but is associated with a response to osmotic stress and a modest increase in colistin tolerance. Growth of P. aeruginosa in a range of bacteriological media is supported by consumption of formate, an important electron donor in anaerobic respiration. In mammalian cell culture media, however, the overall metabolic strategy of P. aeruginosa is instead dependent on consumption of glutamine and lactate. Formate doping of mammalian cell culture media does not alter the overall metabolic strategy but is associated with polyamine catabolism, remodelling of both inner and outer membranes, and a modest sensitization of P. aeruginosa PAO1 to colistin. Further, in a panel of P. aeruginosa isolates an increase between 2- and 3-fold in sensitivity to tobramycin is achieved through doping with other organic acids, notably propionate which also similarly enhances the activity of colistin. Organic acids are therefore capable of nonspecifically influencing the potency of membrane active antimicrobials.
Bibliographical noteFunding Information:
NMR experiments described in this paper were carried out using the facilities of the Centre for Biomolecular Spectroscopy, King’s College London, and at the MRC Biomedical NMR Centre at the Francis Crick Institute. The King’s instruments were acquired with a Multiuser Equipment Grant from the Wellcome Trust and an Infra-structure Grant from the British Heart Foundation. The MRC Biomedical NMR Centre is supported by the Francis Crick Institute, which receives its core funding from Cancer Research UK (FC001029), the U.K. Medical Research Council (FC001029), and the Wellcome Trust (FC001029). We thank Dr. Tom Frenkiel and Dr. Alain Oregioni for their assistance with HR-MAS NMR experiments performed at the Francis Crick Institute and Dr. Andrew Atkinson for liquid-state NMR experiments performed at KCL. P.M.F. was supported by a Health Schools Studentship funded by the EPSRC (EP/M50788X/1). A.J.M. and G.M. received funding from the MRC Proximity to Discovery: Industry Engagement Fund (MC_PC_16074) and the King’s Health Partners R&D Challenge rapid fund.
© 2021 The Authors. Published by American Chemical Society.
- NMR metabolomics
- antimicrobial susceptibility testing