Genomic Characterization of Salmonella In Municipal Wastewater For Community Enteric Disease Monitoring

Abstract

Municipal wastewater (MW) contains human enteric microorganisms: this includes normal gut flora and as pathogens from infected persons. These pathogens can include bacteria and viruses, as well as microorganisms from other domains. From a civil engineering perspective, these microorganisms normally present a public health threat which must be removed or inactivated during wastewater treatment. However, they can also represent a rich source of information about community health: this collective microbiome can be monitored for disease patterns, and used to inform public health interventions and policies. Traditional clinical surveillance methods underestimate community enteric disease burden; for salmonellosis, it is often assumed that the true number of infected persons is approximately 30 times the clinically reported rate. Municipal wastewater infectious disease surveillance may provide a less-biased option for public health monitoring. The overall objective of this dissertation was to use municipal water surveillance to elucidate community salmonellosis trends in Honolulu, Hawaii. Whole genome sequencing (WGS) was conducted on Salmonella isolates (n = 272) collected in weekly 24-hour composite samples from Sand Island Wastewater Treatment Plant in a yearlong monitoring campaign from 2010-2011. First, comparative genomics approaches were used to identify a salmonellosis outbreak which was previously missed by public health clinics in Honolulu: this demonstrates the sensitivity of MW surveillance for clinically relevant infections. Second, subclinical salmonellosis trends were examined, as an abundant strain of Salmonella in MW was found in low frequencies in health clinics (S. Derby). This strain was a previously under-reported sequence type of Derby, which exhibits low virulence potential but an increased ability to establish infections than most Derby strains, allowing it to maintain persistence in the community. Third, the antibiotic resistance gene (ARG) profiles of MW Salmonella was compared with clinical nontyphoidal salmonellosis ARG patterns: MW isolates reflect a subset of the clinical ARG diversity, with ARG rates appearing lower in MW than in clinical isolates. However, MW ARG patterns correspond with the prevalent serovars, and may be indicative of the Salmonella resistome for asymptomatic or minor infections. Taken together, the results of this dissertation demonstrate that MW surveillance can discern relevant public health trends which are inaccessible using traditional clinical monitoring approaches.