Search

Bacterial communities were examined in replicate lab-scale activated sludge reactors after a period of several months of enrichment with non-ionic nonylphenol ethoxylate (NPE) surfactants. Four sequential batch reactors were fed with synthetic sewage, two of which received additionally NPE. Small subunit rDNA-derived denaturing gel gradient electrophoresis (DGGE) profiles and 16S rDNA clone libraries were dominated by clones of Gammaproteobacteria class. Sequences of the other codominant rDNA phylotypes observed only in DGGE from NPE-amended reactors were, respectively, associated with the Group III of the Acidobacteria phylum. Intriguingly, 16S rRNA content from abundant Gammaproteobacteria cells was unexpectedly low. In addition to Acidobacteria, rRNA-derived DGGE profiles were dominated by members of the order Burkholderiales (of the Betaproteobacteria) and of the genus Sphingomonas (a member of the Alphaproteobacteria). Specific oligonucleotide probes for the selected ribotypes were designed and applied for quantitative real time polymerase chain reaction and fluorescence in situ hybridization, confirming their dominance in treated reactors. The parallel abundance of unique phylotypes in replicate reactors implies a distinctive selection of dominant organisms, which are better adapted to specialized niches in the highly selective environment. [ABSTRACT FROM AUTHOR]

The performance of two sets of primers targeting variable regions of the 16S rRNA gene V1–V3 and V4 was compared in their ability to describe changes of bacterial diversity and temporal turnover in full-scale activated sludge. Duplicate sets of high-throughput amplicon sequencing data of the two 16S rRNA regions shared a collection of core taxa that were observed across a series of twelve monthly samples, although the relative abundance of each taxon was substantially different between regions. A case in point was the changes in the relative abundance of filamentous bacteria Thiothrix, which caused a large effect on diversity indices, but only in the V1–V3 data set. Yet the relative abundance of Thiothrix in the amplicon sequencing data from both regions correlated with the estimation of its abundance determined using fluorescence in situ hybridization. In nonmetric multidimensional analysis samples were distributed along the first ordination axis according to the sequenced region rather than according to sample identities. The dynamics of microbial communities indicated that V1–V3 and the V4 regions of the 16S rRNA gene yielded comparable patterns of: 1) the changes occurring within the communities along fixed time intervals, 2) the slow turnover of activated sludge communities and 3) the rate of species replacement calculated from the taxa–time relationships. The temperature was the only operational variable that showed significant correlation with the composition of bacterial communities over time for the sets of data obtained with both pairs of primers. In conclusion, we show that despite the bias introduced by amplicon sequencing, the variable regions V1–V3 and V4 can be confidently used for the quantitative assessment of bacterial community dynamics, and provide a proper qualitative account of general taxa in the community, especially when the data are obtained over a convenient time window rather than at a single time point. [ABSTRACT FROM AUTHOR]