Predicted adaptation of microbial population growth deceleration in feast-and-famine environments

A major challenge is to predict the success of a mutant lineage. It is the life-history of the organism that integrates multiple traits into the overall relative fitness. The feast-and-famine lifestyle often found for microbial populations in natural environments can be clearly defined and controlled but measuring the precise direction of selection in multidimensional trait space remains difficult. Here we give a model for co-culture experiments between two strains competing for a single limiting resource where cell growth slows down with nutrient concentration. We derive the selection pressure as a function of growth traits across all frequencies and environmental conditions. This explicit relationship reveals that selection increases growth rate and minimizes deceleration as two separate traits, but does not act on resource efficiency. For evolution under generic feast-and-famine conditions we predict the maximum possible adaption in the rate-limiting resource threshold K. We find that that the evolved trait scales with the effective population size and is expected to adapt to orders of magnitude below the environmental concentration. We apply these results to quantify the necessary environment to enable selection, to optimize selection pressure in serial transfer evolution experiments and show how to detect the degree of adaptation from the shape of growth curves. Our findings on the mechanisms of selection serve to interpret the resource threshold evolution in the LTEE and realign the observed trait variation in natural isolates.