engleski

Changes in Bacterioplankton Communities Resulting From Direct and Indirect Interactions With Trace Metal Gradients in an Urbanized Marine Coastal Area

Unraveling the relative importance of both environmental conditions and ecological processes regulating bacterioplankton communities is a central goal in microbial ecology. Marine coastal environments are among the most urbanized areas and as a consequence experience environmental pressures. The highly anthropized Toulon Bay (France) was considered as a model system to investigate shifts in bacterioplankton communities along natural and anthropogenic physicochemical gradients during a 1-month survey. In depth geochemical characterization mainly revealed strong and progressive Cd, Zn, Cu, and Pb contamination gradients between the entrance of the Bay and the north-western anthropized area. On the other hand, low-amplitude natural gradients were observed for other environmental variables. Using 16S rRNA gene sequencing, we observed strong spatial patterns in bacterioplankton taxonomic and predicted function structure along the chemical contamination gradient. Variation partitioning analysis demonstrated that multiple metallic contamination explained the largest part of the spatial biological variations observed, but DOC and salinity were also significant contributors. Network analysis revealed that biotic interactions were far more numerous than direct interactions between microbial groups and environmental variables. This suggests indirect effects of the environment, and especially trace metals, on the community through a few taxonomic groups. These spatial patterns were also partially found for predicted bacterioplankton functions, thus indicating a limited functional redundancy. All these results highlight both potential direct influences of trace metals contamination on coastal bacterioplankton and indirect forcing through biotic interactions and cascading.

coastal ecosystem ; metal contamination gradients ; bacterioplankton community structure ; functional prediction ; co-occurrence network

nije evidentirano