The dual impact of bacterial transmission on antimicrobial resistance frequency: linking mathematical models to laboratory experiments

Antimicrobial resistance (AMR) poses a rising threat to global health. Previous studies indicate that reducing antimicrobial use alone does not sufficiently lower AMR in certain populations, and thus, other contributing factors, such as bacterial transmission, need consideration. We previously developed a Lotka-Volterra mathematical model, incorporating within-host and between-host dynamics, which showed that transmission generally promotes increase in AMR frequency. However, under specific conditions, it can lead to AMR decrease. In this study, we evaluated the theoretical hypotheses of that model through in vitro experiments. We employed two E. coli strains, one susceptible and the other resistant to ciprofloxacin, within an experimental setup that mimicked within-host and between-host dynamics. A slight mismatch between our in vitro results and the model's predictions motivated us to refine our model to account for novel mutation emergence. Following these adjustments, our in vitro findings align closely with the in silico simulation outcomes. The model extensions prompted further examination of the strains using whole genome sequencing. This led to the identification of novel mutations, potentially influencing the susceptibility and the fitness of the strains. Our findings confirm that bacterial transmission impacts AMR levels in a dual manner: spreading the resistant strain under certain scenarios while maintaining the susceptible strains in others. This underscores the need for a nuanced understanding of transmission dynamics in combating AMR. Our research also highlights the value of experimental evaluation in corroborating and refining theoretical hypotheses, recognising the power of simulations in exploring a broader range of possible scenarios.