Bacterial communities in a full-scale drinking water treatment plant (DWTP) were characterized using matrix-assisted laser desorption/ionization time of flight mass-spectrometry (MALDI-TOF MS) to identify HPC isolates and the obtained results were compared to 16S rRNA (V4) metabarcoding data acquired in a previous study. Sixty-three samples were collected at nine stages of the potabilization process: river water and groundwater intake, decantation, sand filtration, ozonization, carbon filtration, reverse osmosis, the mixing chamber and post-chlorination drinking water. In total, 1807 bacterial colonies were isolated, 32 % of which were successfully identified to at least the genus level by MALDI-TOF MS using our previously developed Drinking Water Library. Trends in diversity were similar by both approaches, but differences were observed in the detection of taxa, especially at lower hierarchy levels. High bacterial diversity was observed in river and groundwater, where Proteobacteria predominated. The diversity decreased significantly after the chlorination step, where Bacillus sp. (Firmicutes) and an unknown genus of Obscuribacteraceae (Cyanobacteria) were the most prevalent genera according to MALDI-TOF MS and metabarcoding, respectively. The two approaches gave similar results for the decantation, sand filtration and mixing chamber steps, where the most abundant taxon was Flavobacterium. The combined use of these culture-based and culture-independent methods to characterize microbial populations may help to better understand the role of bacteria in water treatment and quality, which will be of value for DWTP management.
Keywords: 16S rRNA metabarcoding; Bacterial diversity; Drinking water treatment plant; MALDI-TOF MS; Water quality monitoring.
Copyright © 2022 The Authors. Published by Elsevier B.V. All rights reserved.