EMA-amplicon-based sequencing informs risk assessment analysis of water treatment systems.

Illumina amplicon-based sequencing was coupled with ethidium monoazide bromide (EMA) pre-treatment to monitor the total viable bacterial community and subsequently identify and prioritise the target organisms for the health risk assessment of the untreated rainwater and rainwater treated using large-volume batch solar reactor prototypes installed in an informal settlement and rural farming community. Taxonomic assignments indicated that Legionella and Pseudomonas were the most frequently detected genera containing opportunistic bacterial pathogens in the untreated and treated rainwater at both sites. Additionally, Mycobacterium, Clostridium sensu stricto and Escherichia/Shigella displayed high (≥80%) detection frequencies in the untreated and/or treated rainwater samples at one or both sites. Numerous exposure scenarios (e.g. drinking, cleaning) were subsequently investigated and the health risk of using untreated and solar reactor treated rainwater in developing countries was quantified based on the presence of L. pneumophila, P. aeruginosa and E. coli. The solar reactor prototypes were able to reduce the health risk associated with E. coli and P. aeruginosa to below the 1 × 10-4 annual benchmark limit for all the non-potable uses of rainwater within the target communities (exception of showering for E. coli). However, the risk associated with intentional drinking of untreated or treated rainwater exceeded the benchmark limit (E. coli and P. aeruginosa). Additionally, while the solar reactor treatment reduced the risk associated with garden hosing and showering based on the presence of L. pneumophila, the risk estimates for both activities still exceeded the annual benchmark limit. The large-volume batch solar reactor prototypes were thus able to reduce the risk posed by the target bacteria for non-potable activities rainwater is commonly used for in water scarce regions of sub-Saharan Africa. This study highlights the need to assess water treatment systems in field trials using QMRA.

Validation of large-volume batch solar reactors for the treatment of rainwater in field trials in sub-Saharan Africa.

The efficiency of two large-volume batch solar reactors [Prototype I (140 L) and II (88 L)] in treating rainwater on-site in a local informal settlement and farming community was assessed. Untreated [Tank 1 and Tank 2-(First-flush)] and treated (Prototype I and II) tank water samples were routinely collected from each site and all the measured physico-chemical parameters (e.g. pH and turbidity, amongst others), anions (e.g. sulphate and chloride, amongst others) and cations (e.g. iron and lead, amongst others) were within national and international drinking water guidelines limits. Culture-based analysis indicated that Escherichia coli, total and faecal coliforms, enterococci and heterotrophic bacteria counts exceeded drinking water guideline limits in 61%, 100%, 45%, 24% and 100% of the untreated tank water samples collected from both sites. However, an 8 hour solar exposure treatment for both solar reactors was sufficient to reduce these indicator organisms to within national and international drinking water standards, with the exception of the heterotrophic bacteria which exceeded the drinking water standard limit in 43% of the samples treated with the Prototype I reactor (1 log reduction). Molecular viability analysis subsequently indicated that mean overall reductions of 75% and 74% were obtained for the analysed indicator organisms (E. coli and enterococci spp.) and opportunistic pathogens (Klebsiella spp., Legionella spp., Pseudomonas spp., Salmonella spp. and Cryptosporidium spp. oocysts) in the Prototype I and II solar reactors, respectively. The large-volume batch solar reactor prototypes could thus effectively provide four (88 L Prototype II) to seven (144 L Prototype I) people on a daily basis with the basic water requirement for human activities (20 L). Additionally, a generic Water Safety Plan was developed to aid practitioners in identifying risks and implement remedial actions in this type of installation in order to ensure the safety of the treated water.

Comparison of EMA-, PMA- and DNase qPCR for the determination of microbial cell viability.

Ethidium monoazide (EMA) quantitative polymerase chain reaction (qPCR), propidium monoazide (PMA)-qPCR and DNase treatment in combination with qPCR were compared for the determination of microbial cell viability. Additionally, varying EMA and PMA concentrations were analysed to determine which dye and concentration allowed for the optimal identification of viable cells. Viable, heat treated (70 °C for 15 min) and autoclaved cultures of Legionella pneumophila, Pseudomonas aeruginosa, Salmonella typhimurium, Staphylococcus aureus and Enterococcus faecalis were utilised in the respective viability assays. Analysis of the viable and heat-treated samples indicated that variable log reductions were recorded for both EMA [log reductions ranging from 0.01 to 2.71 (viable) and 0.27 to 2.85 (heat treated)], PMA [log reductions ranging from 0.06 to 1.02 (viable) and 0.62 to 2.46 (heat treated)] and DNase treatment [log reductions ranging from 0.06 to 0.82 (viable) and 0.70 to 2.91 (heat treated)], in comparison to the no viability treatment controls. Based on the results obtained, 6 µM EMA and 50 µM PMA were identified as the optimal dye concentrations as low log reductions were recorded (viable and heat-treated samples) in comparison to the no viability treatment control. In addition, the results recorded for the 6 µM EMA concentration were comparable to the results obtained for both the 50 µM PMA and the DNase treatment. The use of EMA-qPCR (6 µM) may therefore allow for the rapid identification and quantification of multiple intact opportunistic pathogens in water sources, which would benefit routine water quality monitoring following disinfection treatment.

EMA-qPCR to monitor the efficiency of a closed-coupled solar pasteurization system in reducing Legionella contamination of roof-harvested rainwater.

Solar pasteurization is effective in reducing the level of indicator organisms in stored rainwater to within drinking water standards. However, Legionella spp. were detected at temperatures exceeding the recommended pasteurization temperatures using polymerase chain reaction assays. The aim of the current study was thus to apply EMA quantitative polymerase chain reaction (EMA-qPCR) to determine whether the Legionella spp. detected were intact cells and therefore possibly viable at pasteurization temperatures >70°C. The BacTiter-Glo™ Microbial Cell Viability Assay was also used to detect the presence of ATP in the tested samples, as ATP indicates the presence of metabolically active cells. Chemical analysis also indicated that all anions and cations were within the respective drinking water guidelines, with the exception of iron (mean: 186.76 µg/L) and aluminium (mean: 188.13 µg/L), which were detected in the pasteurized tank water samples at levels exceeding recommended guidelines. The BacTiter-Glo™ Microbial Cell Viability Assay indicated the presence of viable cells for all pasteurized temperatures tested, with the percentage of ATP (in the form of relative light units) decreasing with increasing temperature [70-79°C (96.7%); 80- 89°C (99.2%); 90-95°C (99.7%)]. EMA-qPCR then indicated that while solar pasteurization significantly reduced (p<0.05) the genomic copy numbers of intact Legionella cells in the pasteurized tank water (~99%), no significant difference (p>0.05) in the mean copy numbers was detected with an increase in the pasteurization temperature, with 6 × 10(3) genomic copies/mL DNA sample obtained at 95°C. As intact Legionella cells were detected in the pasteurized tank water samples, quantitative microbial risk assessment studies need to be conducted to determine the potential health risk associated with using the water for domestic purposes.