Speaker
Description
The evolution of antibiotic resistance is a serious and growing problem. The ability to predict a pathogen’s capacity to evolve resistance is therefore a critical public-health goal. In previous work, we found that differences between genetic backgrounds can sometimes lead to unpredictable responses in phenotypic resistance and influence its genetic basis by channeling evolution down particular mutational paths. However, it is still not clear how background integrates with demographic and other factors to influence resistance evolution. For example, rare pathways leading to high-level resistance may become more accessible with larger population sizes and increased mutation rates, but can this increase in evolutionary potential outweigh, or be outweighed by, any potentiating or constraining effects of changes in genetic background? We are addressing these questions using stochastic simulations. We first evolve populations of varying sizes and mutation rates from different genotypes under the influence of genetic drift. We then transfer these evolved populations into a simulated “drug” environment to measure resistance potential. We found that, on balance, large populations with high mutation rates are more evolvable in these drug environments, as expected. However, the maximum effect due to background is more pronounced when the populations are small, and the number of possible initial genotypes is large. In future work, we will examine whether resistance potential is conserved among drug environments that are more similar (i.e., within-class drugs) relative to environments that are less similar (i.e., between-class drugs), and then empirically validate our results using a time-series of E. coli strains isolated from one population that evolved increases in both population size and mutation rate during the Lenski long-term evolution experiment (LTEE). In summary, our work underscores the need to consider the multifactorial nature of resistance evolution when predicting this phenotype.
