The Effects of Population Size on Adaptation and Trade-offs: Insights from Experimental Evolution with Escherichia coli and Individual-based Models

Abstract

Population size plays a major role in shaping the adaptive dynamics of asexual microbes. Here we study how it influences adaptation and fitness trade-offs in such populations.

Most studies dealing with the average fitness in periodically-bottlenecked asexual systems assume that for any given bottleneck size (N0) and number of generations between bottlenecks (g), the harmonic mean size (HM=N0g) will predict the ensuing dynamics. However, there is no theoretical or empirical validation of average fitness being predicted by HM. Using experimental-evolution with Escherichia coli and individual-based simulations, we show that HM fails to predict the average fitness (i.e., higher N0g does not lead to higher fitness). This is because although higher g allows populations to arrive at superior benefits by entailing increased variation, it also reduces the efficacy of selection, ultimately lowering EoA. We demonstrate that average fitness can be maximized by either maximizing N0 and/or minimizing g. We also demonstrate that N0/g is a better predictor of fitness than N0g.

Using a follow-up experiment and agent-based simulations, we also demonstrate the novel and unintuitive result that quantitative differences in population size can lead to qualitative differences (decay/enhancement) in the fates of fitness-associated character during adaptation to the same environment.

Adaptation to the selection-environment can cause maladaptation to other (novel) environments via cross-environment fitness trade-offs. However, such trade-offs are difficult to predict because their dependence on population-genetic parameters is largely unknown. We tested how population size affects such trade-offs (and the consequent maladaptation to novel-environments) by conducting multiple independent bacterial evolution experiments at different population sizes in several selection-environments. We found a novel link between population size and fitness trade-offs, which was contingent on the stability of the environment. Specifically, larger asexual populations evolved bigger fitness trade-offs when the environment was stable but avoided such trade-offs when the environment fluctuated. Furthermore, if the selection-environment was constant over time, larger populations evolved greater extent of ecological specialization. These results were independent of environmental identities and were largely explained by correlated responses to selection and not by genetic drift.

Our findings call for a re-evaluation of the role of population size in predicting adaptation, maladaptation and ecological specialization, and thus have significant implications in evolutionary biology.