Population genetic bottleneck has only marginal effects on fitness evolution and its repeatability in dioecious C. elegans

Evolution is a key process by which populations can adapt to novel conditions, but it is not well understood how predictable this process is. Predictability is expected to depend on the ratio of deterministic and stochastic processes that contribute to evolutionary change and this ratio is modulated by the effective population size. Smaller effective populations harbor less genetic diversity and stochastic processes are generally expected to play a larger role, leading to less repeatable evolutionary trajectories. Empirical insight into the relationship between effective population size and repeatability is limited and biased towards asexual unicellular organisms. Here, we used populations of obligately outcrossing Caenorhabditis elegans to test whether fitness increase and selection response were more heterogeneous after a moderate or strong population bottleneck compared to a scenario without bottleneck. Nematodes were exposed to a novel bacterial prey and lower temperature. Population sizes after one week of growth (as a proxy of fitness) were measured before and after 15 generations of evolution. We found that replicates across all (no/moderate/strong bottleneck) treatments evolved higher fitness and no significant difference in average or maximum fitness was found among treatments. Partitioning fitness variance among effects from selection and effects from chance showed that a strong (but not a moderate) bottleneck reduced the relative contribution of selection effects to fitness variation. However, the reduced contribution from selection did not translate to a significant reduction in the repeatability of fitness evolution. Thus, although a strong bottleneck reduced the contribution of deterministic evolutionary change, we found only marginal effects on quantitative measurements of repeatability in evolution. We conclude that the extent to which evolution is predictable may not universally depend on effective population size. IMPACT SUMMARYEvolution is a key process by which populations can adapt to novel conditions. The predictability of evolution is elusive, but the extent to which evolution is predictable is central to our understanding of evolutionary processes and to emergent applications in medicine, agriculture, and conservation. Predictability is expected to be high when evolutionary change is mainly driven by selection, because this leads to repeatable evolutionary trajectories. However, chance effects, e.g. from genetic drift, reduce repeatability. In populations with smaller effective sizes (and thus reduced genetic diversity), chance effects are generally thought to be more prevalent than in large effective populations, suggesting that effective population size influences evolutionary repeatability. Recent theoretical insights cast doubt on the universality of this relationship between evolutionary repeatability and effective population size. Thus far, empirical work testing these theoretical insights is limited to asexual species, leaving a gap in our knowledge. Therefore, we performed an evolutionary experiment using sexually reproducing populations of the nematode Caenorhabditis elegans. Populations either went through a strong or moderate bottleneck or through no bottleneck at all to create populations with different degrees of genetic variation. These populations with different histories (but similar census size) were then exposed to novel conditions. Our results showed that neither average fitness in the novel environment or the repeatability of fitness evolution were significantly lower following a bottleneck. This was despite a reduced contribution from selection and increased contribution from chance to fitness variation in populations that experienced a strong bottleneck. We therefore show that in sexually reproducing species, increased contributions from chance in populations with lower effective sizes do not necessarily affect evolutionary repeatability. This is an important novel insight that is relevant both for our fundamental understanding as well as for evolutionary forecasting applications in agriculture and conservation, which are primarily targeted at sexual species.