Performance of polycarbonate, cellulose nitrate and polyethersulfone filtering membranes for culture-independent microbiota analysis of clean waters

Demineralized and disinfected waters may have very low microbial loads, requiring that large volumes of water are filtered to recover enough biomass for further analysis. Extended filtration periods, often interrupted by clogging, are a major limiting factor to concentrate samples' microbiota for further examination, besides hindering the work pace. In this study, we investigated the performance of three types of filtering membranes - polycarbonate (PC), cellulose nitrate (CN), and polyethersulfone (PES) with 0.22 μm pore size for culture-independent microbiological analysis (quantitative PCR of seven housekeeping and integrase genes) of tap water, recirculating tap water in a bottle washing loop, and of demineralized water. Compared to PC membranes, CN or PES required lower filtration periods, although had slightly lower DNA extraction yields. However, genes abundance per volume of water was, in general, not significantly different. The exception was observed for bottle washing water in which PC membranes supported significantly higher quantification values than PES membranes. These differences were lower than ∼0.5 log-units and did not hamper the distinction of the types of water based on genes profile. Also, the type of membrane did not significantly affect the profile of the bacterial community determined for tap and demineralized water. A major conclusion is that CN membranes, cheaper, allowing shorter filtration periods, and producing results that are not significantly different from those obtained with PC or PES, can be a good alternative to analyze waters with low biomass loads.