Plant communities exert deterministic control over microbial community structure in coastal wetlands

Jennifer L. Bowen and Joseph A. Vineis, Marine Science Center, Northeastern University, Nahant, MA, Ashley N. Bulseco, Ecosystems Center, Marine Biological Laboratory, John H. Angell, University of South Florida

Background/Question/Methods: Microbes perform important ecosystem services in coastal wetland habitats, including organic matter decomposition, carbon storage, and promoting nutrient retention and removal. Typically, coastal wetlands are characterized by a number of physical gradients (salinity, pH, oxygen saturation) that are influenced by the duration of tidal inundation and that regulate redox chemistry, organic matter content, and nutrient availability with cascading effects on the microbial community. In addition to these physical gradients, vegetation influences biochemical signatures through translocation of oxygen belowground and supplying organic matter to the system in the form of decomposing plant biomass and root exudates. These plant-mediated effects could further drive shifts in microbial community structure and function. In these complex systems, however, it is unclear whether physical drivers such as salinity and redox supply, or plant-mediated drivers such as oxygen translocation and root exudate production, govern the assembly of microbial communities and their associated ecosystem services. Here, we compile microbial 16S rRNA amplicons with sediment biochemical from two separate studies that examine vegetative controls on microbial community assembly to test the hypothesis that plant-mediated drivers outweigh physical forces in driving community assembly in coastal wetland sediments. To test this hypothesis, we calculated the standardized effect size of the mean phylogenetic distance in communities under two scenarios: the first contrasted microbial community distribution across a millimeter-scale gradient in the top three centimeters of sediment from Spartina alterniflora vegetated sediments and from nearby unvegetated panne sediments. The second contrasted community assembly processes above and below the rooting zone in sediment cores taken from Spartina patens.

Results/Conclusions: We show that in both experiments, microbial community assembly was more deterministic in locations where plant roots were dominant. In the S. alterniflora experiment, the vegetated sediments had an average mean pairwise distances lower than -2, suggesting that the community structures were phylogenetically constrained whereas the unvegetated panne sediments consistently had a mean pairwise distance between -2 and 2, suggesting neither deterministic nor stochastic processes could explain community structure. Similarly, microbial communities in sediments within and below the vegetation zone show divergent responses. Microbial communities in the rooting zone of S. patens were strongly phylogenetically constrained, whereas below the active rooting zone, there was no evidence for deterministic community assembly. Taken together these studies indicate that biochemical alterations to the system driven by plant-mediated drivers likely overwhelm physical forcing to drive microbial community structure.