Transgenerational plasticity and selection shape the adaptive potential of sticklebacks to salinity change.

In marine climate change research, salinity shifts have been widely overlooked. While widespread desalination effects are expected in higher latitudes, salinity is predicted to increase closer to the equator. We took advantage of the steep salinity gradient of the Baltic Sea as a space-for-time design to address effects of salinity change on populations. Additionally, genetic diversity, a prerequisite for adaptive responses, is reduced in Baltic compared to Atlantic populations. On the one hand, adaptive transgenerational plasticity (TGP) might buffer the effects of environmental change, which may be of particular importance under reduced genetic variation. On the other hand, physiological trade-offs due to environmental stress may hamper parental provisioning to offspring thereby intensifying the impact of climate change across generations (nonadaptive TGP). Here, we studied both hypothesis of adaptive and nonadaptive TGP in the three-spined stickleback () fish model along the strong salinity gradient of the Baltic Sea in a space-for-time experiment. Each population tolerated desalination well, which was not altered by parental exposure to low salinity. Despite a common marine ancestor, populations locally adapted to low salinity lost their ability to cope with fully marine conditions, resulting in lower survival and reduced relative fitness. Negative transgenerational effects were evident in early life stages, but disappeared after selection via mortality occurred during the first 12-30 days posthatch. Modeling various strengths of selection, we showed that nonadaptive transgenerational plasticity accelerated evolution by increasing directional selection within the offspring generation. Qualitatively, when genetic diversity is large, we predict that such effects will facilitate rapid adaptation and population persistence, while below a certain threshold populations suffer a higher risk of local extinction. Overall, our results suggest that transgenerational plasticity and selection are not independent of each other and thereby highlight a current gap in TGP studies.