The composition of rhizosphere microbial communities can affect plant responses to environmental stress. Previous research has demonstrated that plants grown under droughted or watered conditions are most fit when their contemporary watering environment matches the historical watering environment of their soil microbial community. This suggests that watering history selects for microbes with traits that maximize plant fitness under those specific watering conditions. Do plants also participate in this process by actively selecting microbes that help them better deal with drought stress? To address this question, we conditioned soil microbial communities for droughted or watered conditions in a greenhouse experiment using pots with plants and without them. After six months of soil conditioning, we conducted a full reciprocal transplant of each soil community into droughted or watered conditions, and we estimated growth and fitness of newly planted target plants in each pot. Comparing the fitness and phenotype of plants in these treatments allowed us to identify the relative contribution of plants to the soil microbe conditioning process.
Results/Conclusions
Similar to previous studies, we found that plants had the highest fitness when inoculated with soil microbial communities that had the same historical watering treatment as the plant’s contemporary watering treatment. However, these results were found in all soil treatments, regardless of whether the soil microbial inoculum used had interacted with a plant during the 6-month conditioning period. These results strongly suggest that direct selection from the watering environment is more important than plant-microbe interactions for conditioning a microbial community that can positively influence plant growth under the appropriate watering environment. We speculate that these results may be due to the watering environment selecting for mutualistic microbes that can effectively function in those watering environments. In contrast, we found that other changes in the plant phenotype, including flowering timing and root-shoot ratio, are tied to plant presence in the conditioning phase of the experiment. This suggests that at least some plant responses to changes in the watering environment are dependent on plant-microbe interactions over the conditioning period.