Xiang Li

Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Civil and Environmental Engineering

Committee Chair

Jennifer Weidhaas

Committee Co-Chair

Gary Bissonnette

Committee Member

Antarpreet Jutla

Committee Member

Lian-Shin Lin

Committee Member

Jianbo Yao


Pathogen detection and the identification of fecal contamination sources can be challenging in environmental and engineered treatment systems. Factors including pathogen diversity and ubiquity of fecal indicator bacteria hamper risk assessment and remediation of contamination sources. Therefore, a quick method that can detect and identify waterborne pathogens in environmental systems is needed. In this work, a custom microarray targeting pathogens (viruses, bacteria, protozoa), microbial source tracking (MST) markers, mitochondria DNA (mtDNA) and antibiotic resistance genes was used to detect over 430 selected gene targets in whole genome amplification (WGA) DNA and complementary DNA (cDNA) isolated from sewage and animal (avian, cattle, poultry and swine) feces, freshwater and marine water samples, sewage spiked surface water samples, treated wastewater and sewage contaminated produce.;A combination of perfect match and mismatch probes on the microarray reduced the likelihood of false positive detections, thus increasing the specificity of the microarray for various gene targets. A linear decrease in fluorescence of positive probes over a 1:10 dilution series demonstrated a semi-quantitative relationship between gene concentrations in a sample and microarray fluorescence. Various pathogens, including norovirus, Campylobacter fetus, Helicobacter pylori, Salmonella enterica, and Giardia lamblia were detected in sewage via the microarray, as well as MST markers and resistance genes to aminoglycosides, beta-lactams, and tetracycline. Sensitivity (percentage true positives) of MST results in sewage and animal waste samples (21--33%) was lower than specificity (83--90%, percentage of true negatives). Next generation sequencing (NGS) of DNA from the fecal samples revealed two dominant bacterial families that were common to all sample types: Ruminococcaceae and Lachnospiraceae. Five dominant phyla and 15 dominant families comprised 97% and 74%, respectively, of sequences from all fecal sources.;Waterborne pathogens were also detectable via the microarray in freshwater, marine water and sewage spiked surface water samples as well as treated wastewater. Ultrafiltration was used to concentrate microorganisms (bacteria, viruses, protozoa and parasites) from several liters of environmental and treated water samples. Dead-end ultrafiltration (DEUF) was shown to have a 61.4 +/- 47.8 % recovery efficiency and 46-fold concentration increasing ability. Then WGA was utilized to increase gene copies and lower the microarray detection limit. Viruses, including adenovirus, bocavirus, Hepatitis A virus, and polyomavirus were detected in human associated water samples as well as pathogens like Legionella pneumophila, Shigella flexneri, C. fetus and genes coding for resistance to aminoglycosides, beta-lactams, tetracycline. Microbial source tracking results indicate that sewage spiked freshwater and marine samples clustered separately from other fecal sources including wild and domestic animals via non-metric dimensional scaling. A linear relationship between qPCR and microarray fluorescence was found, indicating the semi-quantitative nature of the MST microarray.;Multiple displacement amplification (MDA), which is an important type of WGA, is a widely used tool to amplify genomic nucleic acids. The strong amplification efficiency of MDA and low initial template requirement make MDA an attractive method for environmental molecular and NGS studies. However, like other nucleic acid amplification techniques, various factors may influence MDA efficiency including template concentration (e.g. rare species swamping out), GC amplification bias and genome length favoring amplification of longer genomes. It was found that MDA increased nucleic acids in mixed environmental samples approximately 4.24 +/- 1.40 (log, average +/- standard deviation) for 16S rRNA gene of Enterococcus faecalis, 1.90 +/- 1.70 for RNA polymerase gene of human norovirus, 8.83 +/- 2.88 for T antigen gene of human polyomavirus, 3.83 +/- 0.93 for uidA gene of Escherichia coli, 4.96 +/- 0.32 for invA gene of S. enterica and 8.77 +/- 2.85 for 16S rRNA gene of human Bacteroidales. The template length, concentration and GC content were found to influence MDA efficiency. The results mainly show that the MDA will be more efficient the longer the template length, the greater the initial concentration of nucleic acids and the lower the GC content of the template.;Overall, the results of this work show that 1) the microarray and sample handling technique is suitable for pathogen detection from feces and sewage; 2) when combined with ultrafiltration techniques, the microarray can also be used as a pathogen detection tool in environmental waters; 3) template length, and initial concentration increase MDA efficiency, but higher GC content template negatively effects MDA efficiency. The proposed microarray can be used for pathogen detection in feces, wastewater treatment plant sewage, treated wastewater and environmental waters. Further the proposed method is potentially applicable to pathogen/microorganism detections on vegetables, seafood, in hospital settings, industrial wastewater, and aquaculture settings.