Microbial Quality of Rainwater Harvested from Rooftops, for Domestic use and Homestead Food Gardens

Abstract

A significant number of rural households in South Africa do not have access to piped water and depend on traditional water sources including river, stream, well or pond which have potential health risks. Rainwater harvesting, a technique used to collect and store rainwater that runs off from roofs can help mitigate potable water challenges. Roof harvested rainwater (RHRW) is generally considered to be safe and is mostly used without prior treatment, which exposes people to several health risks associated with faecal pathogens. Health risks evaluation based on faecal indicator bacteria including Escherichia coli, faecal coliforms and enterococci revealed significant levels of contamination in RHRW. Enterococci were more prevalent in RHRW than any other faecal indicator bacteria suggesting their suitability as faecal indicator bacteria in this water system. Evaluation of Enterococcus spp. prevalence in pigeon faecal samples (364 isolates) showed the four species; E. faecalis (20.5%), E. mundtii (20.51%), E. faecium(23.1%) and E. caseliflavus (17.3%) to be the most prevalent, whereas E. caseliflavus (34.6%) and E. mundtii (33.2%) were more prevalent in RHRW. The use of matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF-MS) for the characterisation of E. coli environmental isolates was evaluated as a potential tool for bacterial source tracking. Support vector machine (SVM) generated classification algorithms were able to distinguish between thirty-one different strain groups at overall accuracies between 94% and 98%. Antibiotic resistance testing among E. coli isolated from RHRW (109 isolates) and pigeon faecal samples (130 isolates) showed ampicillin (22.7.9%), gentamicin (23.6%), amikacin (24%), tetracycline (17.4) and amoxicillin (16.9%) to be the most frequently encountered forms of resistance. However, a relatively higher proportion of isolates from pigeon faeces were antibiotic resistant than those from RHRW. Pyrosequencing analysis of 16S rDNA from RHRW showed the phylum Proteobacteria (80.5%) and the classes Betaproteobacteria (50.35%), Alphaproteobacteria (16.19%), Verrucomicrobiae (6.59%), Planctomycetacia (5.69%) and Sphingobacteria (3%) to be dominant. Pathogenic signatures were detected for Acinetobacter, Pseudomonas, Clostridia, Chromobacterium, Yersinia and Serratia, with Legionella signatures being the most prevalent in the samples. The data gathered here will allow for the implementation of a more detailed and accurate health risk assessment for the safe use of RHRW in South African rural communities.