Analysis of the microbiome of selected drinking water treatment plants, drinking water treated sludge, and water distribution systems

Abstract

Microorganisms are crucial for water quality and purification in water treatment and distribution systems. The microbial communities present in untreated water can aid in purification through the biodegradation of contaminants. However, some microbes may be human pathogens that threaten consumer health. The current study analysed microbial diversity in selected drinking water treatment plants, water treatment sludge and drinking water distribution point seasonally over 12 months. Samples were drawn from five selected water treatment plant sites in the Gauteng (A, D and E), Limpopo (B) and Mpumalanga (C) Provinces in South Africa. Selective media culture studies were used for Escherichia coli and Enterococcus species (spp.) quantification and antibiotic resistance profiling for isolates from the samples. Colilert and Enterolert Quanti-Tray 2000 IDEXX methods were used to enumerate total coliforms, E. coli and Enterococcus spp. from water samples from the different water treatment stages. The Kirby-Bauer disc-diffusion technique was used to assess the antibiotic susceptibility of the indicator bacteria isolates. Shotgun metagenomics sequencing using the MGI DNBSEQ-G400 sequencing platform was performed to investigate microbial metagenomic diversity and their functions from the raw water sources, the filtration, sedimentation and disinfection stages, final treated water, treated sludge and distribution system samples. Water physicochemical parameters were assessed, and the obtained results in terms of pH, temperature, turbidity, free chlorine, manganese, ammonia, nitrate, sulphate, iron, and total organic carbon for different stages of treatment, final treated water, drinking water treated sludge and first point in the distribution system varied across the five sampling sites. The physicochemical and microbiological parameters were measured and compared to the drinking water quality guidelines set by the World Health Organization and South African National Standard 241. The results obtained from the IDEXX-defined substrate Colilert/ Quanti-Tray 2000 System (ISO 9308-2:2012) showed the presence of the indicator microorganisms (total coliforms, E. coli and Enterococcus spp.) with variations in their abundance (based on the Most Probable Number) in the raw water sources, sedimentation stage and drinking water treated sludge across the seasons. However, only two of the five treatment plants maintained the acceptable limit of microbiological quality for the final treated water stage, and the distribution systems sampling points complied with the drinking water standards. Using the MGI DNBSEQ-G400 sequencing platform, the following dominant phyla and classes were identified: Euryarcheota and Nitrososphaerota (Archaea phyla), Halobacteria and Methanomicrobia (Archaea classes), Pseudomonadota (bacterial phylum), v Alphaproteobacteria, Gammaproteobacteria and Betaproteobacteria (bacterial classes), Ascomycota and Basidiomycota (Eukaryota phyla), and Aconoidasida and Sordariomycetes (Eukaryota classes). Despite the observed overlap of dominant phyla and classes in the treatment plant stages, the antibiotic-resistance gene composition and antibiotic class phenotypes exhibit a general trend of a downward shift, showing the efficiency of treatment plants in reducing opportunistic pathogens. Correlation analysis done between physicochemical parameters in the drinking water treatment, distribution, and the Archaea, bacterial and Eukaryota phyla and classes respectively demonstrated a link that existed between the physicochemical parameters and the proliferation of the members of microbial groups. Temperature correlated significantly positively with Gammaproteobacteria (filtration stage at Site D), Alphaproteobacteria (raw water at Site A), sulphate and Alphaproteobacteria (final treated water at Site C), Cynophyceae (raw water at Site A), sulphate and Betaproteobacteria (raw water at Site D and filtration stage at Site A), and nitrate and Bacilli (final treated water at Site E). This correlation clearly showed that changes in the physicochemical parameters could contribute to the selection of the microorganisms that survive in each stage of treatment, including drinking water treated sludge and distribution systems. Functional metagenomes revealed the abundance of antibiotic resistance functions, especially cationic antimicrobial peptide, ßeta-lactam and vancomycin resistance in the final treated water and distribution systems across the selected sampling sites. It can be concluded that selected treatment plant processes effectively reduced the number of indicator organisms. In addition, seasons and different treatment stages influence the microbial diversity of the drinking water samples. In Gauteng Province at Sites A and D, most bacteria, including potential pathogens, were effectively removed by chlorine disinfection. However, some bacteria presented excellent resistance to chlorine. The study was able to provide a profile of microbial diversity (metagenomic diversity of Proteobacteria and Firmicutes in drinking water treatment process). The presence of antibiotic-resistant microorganisms and residual microorganisms in some treated water samples means that these microorganisms are not completely removed during drinking water treatment process and optimisation of drinking water treatment systems can be recommended, giving more attention to the raw water sources, DWTS, and distribution systems in removing the indicator organism and maintaining high-quality drinking water free from pathogens and contaminants.