Spatiotemporal dynamics and determinants of planktonic bacterial and microeukaryotic communities in a Chinese subtropical river.

The spatiotemporal distribution of microbial diversity, community composition, and their major drivers are fundamental issues in microbial ecology. In this study, the planktonic bacterial and microeukaryotic communities of the Jiulong River were investigated across both wet and dry seasons by using denaturing gradient gel electrophoresis (DGGE). We found evidence of temporal change between wet and dry seasons and distinct spatial patterns of bacterial and microeukaryotic communities. Both bacterial and microeukaryotic communities were strongly correlated with temperature, NH4-N, PO4-P, and chlorophyll a, and these environmental factors were significant but incomplete predictors of microbial community composition. Local environmental factors combined with spatial and temporal factors strongly controlled both bacterial and microeukaryotic communities in complex ways, whereas the direct influence of spatial and temporal factors appeared to be relatively small. Path analysis revealed that the microeukaryotic community played key roles in shaping bacterial community composition, perhaps through grazing effects and multiple interactions. Both Betaproteobacteria and Actinobacteria were the most dominant and diverse taxa in bacterial communities, while the microeukaryotic communities were dominated by Ciliophora (zooplankton) and Chlorophyta (phytoplankton). Our results demonstrated that both bacterial and microeukaryotic communities along the Jiulong River displayed a distinct spatiotemporal pattern; however, microeukaryotic communities exhibited a stronger distance-decay relationship than bacterial communities and their spatial patterns were mostly driven by local environmental variables rather than season or spatial processes of the river. Therefore, we have provided baseline data to support further research on river microbial food webs and integrating different microbial groups into river models.