Scientists crack the code of bacterial evolution across species
Researchers have demonstrated that evolutionary predictions proven in one bacterial species can reliably forecast how different species will adapt—a breakthrough with implications for predicting antibiotic resistance and designing industrial bioprocesses. The finding suggests that evolution follows consistent patterns across diverse organisms, enabling companies and health authorities to anticipate microbial behavior rather than simply react to it.
Originaltitel: Extending evolutionary forecasts across bacterial species
<p>Improving evolutionary forecasting requires progressing from studying repeated evolution of a single genotype under identical conditions to formulating broad principles. These principles should enable predictions of how similar species will adapt to similar selective pressures. Evolve-and-resequence experiments with multiple species allow testing forecasts on different biological levels and elucidating the causes for failed predictions. Here, we show that forecasts for adaptation to static culture conditions can be extended to multiple species by testing previous predictions for Pseudomonas syringae and Pseudomonas savastanoi. In addition to sequence divergence, these species differ in their repertoire of biofilm regulatory genes and structural components. Consistent with predictions, both species repeatedly produced biofilm mutants with a wrinkly spreader phenotype. Predominantly, mutations occurred in the wsp operon, with less frequent promoter mutations near uncharacterized diguanylate cyclases. However, mutational patterns differed on the gene level, which was explained by a lack of conservation in relative fitness of mutants between more divergent species. The same mutation was the most frequent for both species suggesting that conserved mutation hotspots can increase parallel evolution. This study shows that evolutionary forecasts can be extended across species, but that differences in the genotype-phenotype-fitness map and mutational biases limit predictability on a detailed molecular level.</p>