Inbreeding depression changes with stress response over time in flooded Mimulus guttatus

Inbreeding usually reduces offspring fitness (‘inbreeding depression’, ID), and may affect the plasticity of functional traits involved in the response to stress. ID is often found to increase under stress, but there are also reports of no effects or even a reduction of ID under stress. One reason for this variation that has received little attention may be related to different concepts of stress. In particular, the magnitude of ID may be unrelated to the effect of an environment on fitness (evolutionary stress concept), but increase particularly during the ‘alarm phase’ after a stress has been initiated (physiological stress concept). We clonally replicated inbred and outbred Mimulus guttatus plants, for which ID was known to increase under flooding. We exposed the clonal replicates to control and flooding conditions and harvested replicates of each genotype after two, six and 11 weeks of growth. As functional traits related to stress response we measured chlorophyll fluorescence, root mass and the production of stolons and adventitious roots. As fitness estimates we measured biomass and flower number, and we pollinated a subset of plants and grew a second generation of plants under control and flooding conditions to calculate multiplicative fitness. Overall, M. guttatus proved to be very flooding-tolerant. Chlorophyll fluorescence (F_v/F_m) was not influenced by flooding, but decreased with leaf age and increased after fertilization. At the end of the experiment, biomass and flower number (F₁ generation) as well as multiplicative fitness (including performance in the F₂ generation) were even higher under flooding than under control conditions. Flooding reduced the root mass in the pots, but increased the production of stolons and floating roots. Plasticity in these traits can be regarded as beneficial, although selection gradient analysis failed to identify plasticity in stolon number as adaptive. Only two functional traits were influenced by an interaction between flooding and inbreeding, early stolon length (suggesting a reduced flooding escape response of inbred plants) and root tissue density of floating adventitious roots (suggesting a reduced aeration of the roots of inbred offspring). ID in fitness-related traits was higher under flooding, but its magnitude changed strongly over the course of the experiment. ID under flooding was particularly high after two weeks (δ = 0.42 vs. 0.05 in the control), suggesting sensitivity of inbred plants to the initiation of flooding (‘alarm phase’ of stress response). This effect had disappeared after 6 weeks when plants had acclimated to ongoing flooding. However, under flooding ID increased again after 11 weeks, this time because outbred plants grew much better under flooded than control conditions, and the same pattern was found for the multiplicative fitness function (δ = 0.68 under flooding vs. 0.36 in the control). Our results suggest that ID was higher under flooding, but not because this environment was generally more stressful. Instead, at an early stage ID increased because inbred offspring was more sensitive to the physiological stress after the initiation of flooding (‘alarm phase’), whereas at a late stage ID increased because outbred offspring was more capable of exploiting the favourable flooding conditions. In general, our results show that phenotypic plasticity may often be robust against the effects of inbreeding. Moreover, ID may increase in particular under conditions of physiological stress, during which many stress-specific genes are expressed, whereas ID may not necessarily increase under constant poor conditions that reduce fitness, even if the stress is very strong in the evolutionary sense.