Bacterial defiance as a form of prodrug failure

Brandon Alexander Holt; Gabriel A. Kwong

To fit this model to our system, we measured the values for relevant parameters, including 20 enzymatic efficiency (e.g., kcat, Km), bacterial growth (e.g., r, Bmax), and prodrug activity (e.g., a, . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/556951 doi: bioRxiv preprint a * ) (Fig. S1 , Table S2 ). This allowed us to predict bacteria-prodrug response curves (red and blue dashed lines; Fig. 2B ) across nine distinct combinations of kcat and r values, which we experimentally controlled by altering the ambient temperature and concentration of broth (conditions labeled A1-3, B1-3, and C1-3; Table S3 ). Strikingly, our model anticipated a binary bacterial response to prodrug treatment that would be evident within 24 hours; bacteria were 5 predicted to be either susceptible to the prodrug and die (red dashed lines) or to exhibit a druginvariant phenotype and proliferate to saturating levels (blue dashed lines) (Fig. 2B, C) . To experimentally validate our model predictions, we incubated bacteria with prodrugs under the defined nine conditions, and quantified the number of living bacteria longitudinally over the course of a 24 hour treatment window. Quantified bacterial counts taken during the course of treatment 10 as well as at endpoint closely matched the values predicted by our model (R 2 = 0.84, red and blue dots; Fig. 2B , C) and likewise revealed a binary bacterial response to prodrugs leading to antibiotic failure. We therefore called this form of resistance to prodrugs as bacterial "defiance."