Chemogenomic model identifies synergistic drug combinations robust to the pathogen microenvironment

Area of Science:

• Microbiology
• Computational Biology
• Pharmacology

Background:

• Antibiotic efficacy is crucial in diverse in vivo environments.
• Pathogen microenvironments significantly impact antibiotic potency.
• The effect of microenvironments on antibiotic combination efficacy remains largely unknown.

Purpose of the Study:

• To develop a computational framework, MAGENTA, to predict drug interactions in various microenvironments.
• To explore the impact of diverse microenvironments on antibiotic combination efficacy.
• To identify effective antibiotic combinations tailored to specific pathogen microenvironments.

Main Methods:

• Developed the Metabolism And GENomics-based Tailoring of Antibiotic regimens (MAGENTA) computational framework.
• Utilized chemogenomic profiles and metabolic perturbations to predict drug interactions.
• Screened 2556 drug combinations across nine distinct microenvironments against E. coli and A. baumannii.

Main Results:

• Identified antibiotic combinations with robust synergy across nine environments against E. coli and A. baumannii.
• MAGENTA accurately predicted efficacy changes of drug combinations in glycerol media, confirmed experimentally.
• Discovered genes in glycolysis and glyoxylate pathways as key predictors of synergy and antagonism.

Conclusions:

• MAGENTA enables prediction of drug interactions within specific microenvironments.
• The framework facilitates tailoring antibiotic therapies based on pathogen microenvironment.
• This systems approach enhances the development of effective antibiotic combination strategies.