Coherent dynamics and association networks among lake bacterioplankton taxa

Bacteria play important roles in freshwater food webs and in cycling of elements. Yet specific ecological features of individual phylogenetic groups and interactions among these are largely unknown. We used 454 pyrosequencing of 16S rRNA genes to study the association of different bacterioplankton groups with environmental properties and their co-occurrence patterns over an annual cycle in a dimictic lake. Clear seasonal succession of the bacterioplankton community was observed. After phylogenetic clustering of sequences into previously described and highly resolved groups (tribes), their temporal dynamics revealed extensive synchronization among them and associations with seasonal events such as ice coverage, ice-off, mixing and phytoplankton blooms. Complex relationships among closely and distantly related tribes were resolved by time-dependent rank-correlations suggesting ecological coherence and incoherence often independent of taxonomic relatedness. Local similarity analysis-based association networks centred around Actinobacteria and a-Proteobacteria suggest complex interactions within bacterioplankton communities and with changing environmental conditions, with unique ecological features inferred for each tribe. The study reveals the shifts in the abundant bacterioplankton taxa through time (seasonal succession) and provide novel insights into the natural history of abundant cultured and uncultured freshwater bacteria.