Prodrug defiance reveals logic-based strategies for treating bacterial resistance

Classifying the mechanisms of antibiotic failure has led to the development of new treatment strategies for killing bacteria. Among the currently described mechanisms, which include resistance, persistence and tolerance, we propose bacterial defiance as a form of antibiotic failure specific to prodrugs. As a prototypic model of a bacteria-activated prodrug, we construct cationic antimicrobial peptides (AMP), which are charge neutralized until activated by a bacterial protease. This construct successfully eliminated the vast majority of bacteria populations, while localizing activity to bacterial membranes and maintaining low active drug concentration. However, we observed defiant bacteria populations, which survive in the presence of identical drug concentration and exposure time. Using a multi-rate kinetic feedback model, we show that bacteria switch between susceptibility and defiance under clinically relevant environmental (e.g., hyperthermia) and genetic (e.g., downregulated protease expression) conditions. From this model, we derive a dimensionless quantity (Bacterial Advantage Heuristic, BAH) - representing the balance between bacterial proliferation and prodrug activation - that perfectly classifies bacteria as defiant or susceptible across a broad range of conditions. To apply this concept to other classes of prodrugs, we expand this model to include both linear and nonlinear terms and use general pharmacokinetic parameters (e.g., half-life, EC50, etc.). Taken together, this model reveals an analogous dimensionless quantity (General Advantage Key, GAK), which can applied to prodrugs with different activation mechanisms. We envision that these studies will enable the development of more effective prodrugs to combat antibiotic resistance.