Drinking water safety is of great concern all around the world. Two strategies can be pursued to maintain biological stable drinking water during distribution in the network: (i) without a residual disinfectant, in which bacterial growth is controlled by very low biodegradable nutrients in the water by extensive pre-treatment and a well-maintained network and (ii) with a residual disinfectant in the water during distribution. Independent of the strategy, maintaining water quality during the transportation processes has always been a challenge, and there is a need for a better understanding of the impact of biological stability strategies using systematic studies. In this study, the impact of no residual disinfectant, residual chlorine and residual monochloramine on chemical and microbial water quality were investigated. It is anticipated that studies with pilot distribution networks will lead to better control of drinking water safety, in terms of biological stability and DBPs formation. Meanwhile, computer models applied in this study fitted disinfectant decay kinetics, and appeared to correspond with the experimental data for THMs and biofilm formation, and hence provided more insights in terms of predicting the behaviour of the system in the future, while their accuracy can also be calibrated and verified by the future experimental data.

Drinking water safety is critical to public health and the stability of drinking water quality is highly related with drinking water distribution systems (DWDSs). Antibiotic resistance genes (ARGs) and their hosts are the two major concerns in drinking water, however, the knowledge of fate of ARGs and their hosts were limited due to the current short-read length based sequencing technologies. Nanopore sequencing developed by Oxford Nanopore Technologies (ONT) is expected to overcome the challenges on ARGs and ARB characterization owing to its ultra-long reads generated. In this study, therefore, direct gDNA Nanopore sequencing was performed on treated water, distributed water and biofilm samples in two different DWDSs located at Kamerik and Lekkerkerk to explore its potential on ARGs and ARB characterization in drinking water systems. Additionally, the performances of taxonomy classification and ARGs profiling with Oxford Nanopore gDNA sequencing with different thresholds were assessed using an artificial microbial community. However, the conflict between the high requirements of input DNA quality and quantity with Nanopore sequencing and extremely low microbial biomass content in drinking water samples led to a challenge for our research. Therefore, evaluation on different sample preparation methods was conducted to meet the requirements of the input DNA.

Biological safety of drinking water is vital for safeguarding public health. Many efforts have been made to explore the microbial universe in drinking water. Nanopore sequencing developed by Oxford Nanopore Technologies is expected to enable PCR-free and rapid identification of species with high accuracy, thus overcoming the impediments of next-generation sequencing. However, the capability of Nanopore sequencing for characterizing the microbiome in drinking water with extremely low biomass content has not been explicitly evaluated. Therefore, this research was carried out to explore the potential of Nanopore sequencing for microbial community characterization and species identification in drinking water. In this study, NanoAmpli-Seq full-length 16S rRNA sequencing and 1D² genomic DNA (gDNA) sequencing were performed on an Oxford Nanopore MinION^TM sequencer. DNA samples of artificial microbial communities were sequenced in order to assess the performance of both sequencing strategies. Subsequently, DNA extracted from tap water was subjected to Nanopore sequencing with the two methods. Results showed that NanoAmpli-Seq 16S rRNA sequencing precisely identified abundant species in artificial microbial communities with high level of reproducibility but biased community profiles due to variation in PCR efficiencies of different species, whereas only 10 species were identified in tap water samples. In addition, raw results from 1D² gDNA sequencing provided an unbiased microbial community profile of an artificial community DNA, while polished data improved the species identification accuracy at the expense of the ability to profile the community structure. Furthermore, 45 hours’ sequencing generated more reliable results than 5 hours’ sequencing with higher profiling accuracy of community structure. Nevertheless, 1D² gDNA sequencing still did not exhibit desirable species identification performance on tap water DNA samples. Notably, despite two enteropathogenic species (Enterobacter cloacae and Laribacter hongkongensis) were identified, the detection of Homo sapiens in the same sample indicated the potential existence of post sample contamination. To conclude, Nanopore sequencing possesses great potential to serve as an efficient tool for study of drinking water microbiology. Specifically, notwithstanding the dissatisfactory performance of NanoAmpli-Seq, its high reproducibility across sequencing runs, adaptability to low DNA quality and quantity, and short turnaround time indicated its potential usefulness to promptly monitor microbial community changes subjected to environmental changes in extremely low-biomass samples (i.e. drinking water). Despite that 1D² gDNA sequencing exhibited superior performance on species identification and microbial community profiling to NanoAmpli-Seq, more endeavors should be made to overcome the hurdles (e.g. demand for high molecular weight gDNA, standard methods for analyzing sequencing data), thereby improving the species identification coverage and microbial community profiling accuracy in drinking water. Understanding the presence and dynamics of the microbial community in DWDS is important for water utilities to gain a better understanding of various microbial processes in drinking water from source to customers’ taps, based on which water treatment strategies could be improved and better management of drinking water quality could be performed.