Assessment of the Impact of Treatment and Distribution on the Occurrence of Nontuberculous Mycobacteria and Free-Living Amoebae in Full-Scale Drinking Water Systems

One of the central goals of drinking water treatment is the removal and inactivation of human pathogens from source waters. While the focus of pathogen removal has traditionally been on enteric pathogens (e.g. Escherichia coli, Giardia spp., Cryptosporidium spp.), it has become increasingly clear that these are not the only microorganisms in drinking water that can cause human infections. Among the emerging drinking water microorganisms of concern are opportunistic human pathogens (OPs), which are species of amoebae, fungi, and bacteria that primarily infect people with compromised immune systems or other health conditions. Although the frequency with which OPs occur in drinking water systems is not known, they have been detected in drinking water distribution and building plumbing systems. In fact, drinking water is a major source of OP infections; numerous outbreaks of OP infections traced to drinking water have been reported. OPs in drinking water are not regulated in most countries and there are few standardized methods for their detection and quantification. This lack of standardized monitoring has made it difficult to compare studies and to estimate risk. There are also many unknowns relating to how treatment processes, physicochemical characteristics of drinking water, distribution systems and building plumbing, and microbial communities may influence OPs in full-scale drinking water systems, which further complicates assessment of OP risk. This work sought to investigate the occurrence of OPs in full-scale drinking water systems while also developing and evaluating methods for OP detection and quantification, with a focus on nontuberculous mycobacteria (NTM) and certain species of free-living amoebae (FLA). I investigated the potential use of near real-time flow cytometry (NRT-FCM), a novel, automated, cultivation-independent method for quantification of total and viable bacteria, for monitoring bacterial inactivation through a full-scale drinking water ozone system. My results showed that NRT-FCM can provide rapid, high-resolution feedback on disinfection system performance, and could serve as an early warning system for performance excursions. Next, I conducted a year-long, source-to-tap investigation of a full-scale drinking water system to further elucidate the impacts of treatment, distribution, and building plumbing on drinking water microbial and physicochemical parameters. Despite a low occurrence of NTM during most months in the finished water leaving the treatment plant, high concentrations of NTM were found across all sampling events in the first draw samples from buildings. Low water use in buildings resulting from COVID-19 pandemic-related building closures was associated with increased concentrations of NTM and decreased monochloramine residuals. Results of this work emphasize the importance of quantifying OPs not just in drinking water distribution systems, but also in building plumbing, where the highest concentrations of NTM were measured (up to 107 gene copies per liter). Further, I studied the presence of FLA in the full-scale drinking water system and recovered more than 20 isolates representing several genera. Among the most abundant genera was Acanthamoeba, which is an FLA genus associated with human infection. Acanthamoeba spp. were recovered from samples within the treatment plant but also in the finished water and distribution system. This research provides valuable occurrence data for NTM and FLA and provides an assessment of the performance of drinking water systems for the removal of OPs. It is intended to inform researchers, utility personnel, practitioners, and regulators, as future regulations of OPs and mitigation efforts are being evaluated by drinking water professionals.PHDEnvironmental EngineeringUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/175714/1/kathdowd_1.pd