Effects of filtration on bacterial activity and picoplankton community structure as assessed by flow cytometry

We used flow cytometry of autofluorescent and Syto13-stained marine bacteria and the uptake of tritiated leucine to assess the effects of different filter types on picoplankton abundance, community structure and bacterial activity in the filtrate. Coastal and oceanic samples from the NW and SW Mediterranean and the Atlantic coast of Galicia were size-fractionated using polycarbonate (PC), mixed cellulose esters (CE), aluminum oxide (IM) and glass fiber (GF) filters of 0.2 to 1.2 μm nominal pore size from different brands. Flow cytometry of Syto13-stained marine bacteria and autofluorescent photosynthetic prokaryotes was used to analyze picoplankton abundance, size structure and community composition before and after filtration. We combined this capability with the detection of the changes in cell-specific heterotrophic activity in the filtrates. We found that the CE filters retained picoplankton better than the PC filters. The PC filters did not discriminate prokaryotes according to size as much as the GF and the CE filters did. In our hands the IM filters were no better than the CE filters. Bacterial activity in the filtrates increased in the PC and in the CE filtrates and this stimulation of bacterial activity was more important in the less productive environments. We conclude that care must be taken when PC filters are used for generating bacteria-free water, and that the use of CE 0.22 μm filters is the best way of creating picoplankton-free water. However, the picoplankters that will go through the filters may encounter increased nutrient levels.