Microbial ecology of surface water from the northeast U.S. and its association with environmental factors and foodborne pathogens
Open Access
- Author:
- Chung, Taejung
- Graduate Program:
- Food Science
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 07, 2019
- Committee Members:
- Jasna Kovac, Thesis Advisor/Co-Advisor
Edward G Dudley, Committee Member
Luke F Laborde, Committee Member
Darrell William Cockburn, Committee Member - Keywords:
- agricultural water
microbial community
fresh produce
surface water - Abstract:
- Fresh produce is a common food vehicle associated with foodborne outbreaks in the U.S. Moreover, recent foodborne outbreaks demonstrated that surface water used for irrigation can act as an important source of pre-harvest produce contamination with foodborne pathogens. The US Food and Drug Administration (FDA) has outlined standard protocols for direct pathogen detection methods in variety of food and environmental sources. However, direct detection of pathogens in routine analyses of irrigation water is recognized as inefficient because it requires multiple complex test for detection of each individual target pathogen. Thus, indirect methods based on indicator microorganisms, such as generic Escherichia coli are used for monitoring of the microbiological quality of agricultural water. The FDA Produce Safety Rule (PSR) of the Food Safety Modernization Act (FSMA) proposed quantitative detection of generic E. coli for indirect detection of potential microbiological food safety hazards. However, several studies suggested that indicator microorganism that are currently used for evaluation of microbiological quality and safety of irrigation waters do not correlate well with occurrence of relevant pathogens under all relevant environmental conditions. In this thesis we sought to better understand the microbial ecology of surface waters collected in streams located in the upstate New York. The goal of this thesis was to characterize microbial communities of surface water and investigate potential associations between microbial communities’ profiles, presence of foodborne pathogens and environmental factors. We characterized the composition of the bacterial and fungal communities in 68 water samples collected between May and August 2017 from six streams located in the upstate New York. Microbial communities were determined by Illumina sequencing of PCR-amplified 16S rRNA gene V4 region and ITS2 sequences. Alpha and beta diversity indices were used to analyze and compare the microbial communities’ characteristics of the samples. Moreover, Random forest (RF) machine learning was implemented to predict the presence of pathogens based on the microbial community composition. According to the principal coordinate (PCoA) analysis and permutational multivariate analysis of variance (PERMANOVA), microbial communities differed significantly (p<0.01) between suspended sediment and water fractions. Moreover, specific physicochemical properties of water (i.e., average flow rate, pH, turbidity, and conductivity) were significantly associated with microbial composition. Additionally, the microbial communities of sediment fractions differed significantly among the sampling sites based on both alpha and beta diversities measurement. The observed differences may be due to the upstream land use of the sites; however these are solely descriptive observations. Furthermore, certain microbial families (e.g., Kallotenuaceae, Flavobacteriaceae, and Sericytochromatia) were found to be predictive of Salmonella presence using RF; however, the predictions had a very low accuracy (AUC of 0.55) and therefore cannot be used. Overall, this study provided a new insight into microbial ecology of surface waters in a limited geographic area in the northeast U.S., and its relationships with pathogen occurrence and environmental factors including physicochemical properties.