This paper describes an experimental study of microbial communities of three moving bed biofilm reactors (MBBR) inoculated with nitrifying cultures originated from environments with different salinity; freshwater, brackish (20‰) and seawater. All reactors were run until they operated at a conversion efficiency of >96%. The microbial communities were profiled using 454-pyrosequencing of 16S rRNA gene amplicons. Statistical analysis was used to investigate the differences in microbial community structure and distribution of the nitrifying populations with different salinity environments. Nonmetric multidimensional scaling analysis (NMDS) and the PERMANOVA test based on Bray-Curtis similarities revealed significantly different community structure in the three reactors. The brackish reactor showed lower diversity index than fresh and seawater reactors. Venn diagram showed that 60 and 78% of the total operational taxonomic units (OTUs) in the ammonia-oxidizing bacteria (AOB) and nitrite-oxidizing bacteria (NOB) guild, respectively, were unique OTUs for a given reactor. Similarity Percentages (SIMPER) analysis showed that two-thirds of the total difference in community structure between the reactors was explained by 10 OTUs, indicating that only a small number of OTUs play a numerically dominant role in the nitrification process. Acute toxicity of salt stress on ammonium and nitrite oxidizing activities showed distinctly different patterns, reaching 97% inhibition of the freshwater reactor for ammonium oxidation rate. In the brackish culture, inhibition was only observed at maximal level of salinity, 32‰. In the fully adapted seawater culture, higher activities were observed at 32‰ than at any of the lower salinities.
Keywords: Amplicon pyrosequencing; Bacterial community; Nitrification; Salinity; Statistical analysis.
Copyright © 2016 Elsevier Ltd. All rights reserved.