Distribution, decay, and quantification of fecal source-tracking markers

Graduation date: 2012Waterborne fecal contamination poses serious risks to human health and can disrupt aquatic ecosystems. Molecular marker methods are widely used to identify and, in some cases, quantify the sources of contamination and guide management decisions regarding water resource protection. However, methods to detect some likely sources of fecal contamination such as birds are not available. Furthermore, the effects of environmental variables, such as water type and light exposure, on marker decay could change quantitative interpretations of source tracking results. Additionally, sample interference can be a significant source of variability and error during sample analysis could prevent accurate quantification of fecal sources. Methods that are robust to sample-to-sample variability and consider variable marker decay are needed to accurately quantify molecular markers in environmental samples.\ud We characterized the geographic and species distribution of three novel molecular markers based on rRNA gene sequences from bird fecal bacteria for the detection of bird feces in the environment. The distribution of the markers differed across geographic location and host species. Two markers were found mainly in gulls,\ud but one occurred in many bird species. The wide geographic distribution of the markers suggested that they will be useful in many areas where birds are suspected contributors to aquatic fecal contamination. We developed two quantitative PCR (qPCR) assays using two of the markers and determined their limits of detection of in natural water matrices. Both qPCR assays detected down to 100 ng feces/100 ml. Although one assay was designed to detect gull contamination, the marker occurred in sheep feces at low levels (2.9x10¹ ± 9.6 x10¹ copies/ng DNA), but high concentrations of sheep feces (>0.2 g/100 ml) would be required for detection. The high specificity and sensitivity of the assays make them excellent tools for the quantification of aquatic avian fecal contamination.\ud We compared the decay of human-targeted Bacteroides markers in marine and fresh water under light and dark conditions using microcosms. Markers persisted about 2-3 days longer in marine water than in fresh water, suggesting that differential persistence of molecular markers might justify different standards between marine and fresh water bodies. Sunlight limited the persistence of DNA and RNA markers. Significant correlation between the decay of Bacteroides DNA and RNA markers suggested that most of the markers detected were bound within cells. This finding is important because the persistence of extracellular DNA could complicate estimation of fecal contaminants with molecular methods. The decay rates of Bacteroides markers differed despite their close phylogenetic relationship. Differences in physiological responses between Bacteroides clades to stresses may translate into different decay rates or persistence times in the environment.\ud We developed a novel method based on the spike and recovery of a genetically modified Escherichia coli strain to describe and limit the variability in marker quantification caused by sample interference. The spike-and-recovery approach accurately reflected low recovery of Bacteroides genomic DNA in low salt extractions and Bacteroides qPCR assay inhibition by a common carry-over reagent, ethanol. We used multivariate Z-scores to identify amplification deviants and showed that this new statistical method was more sensitive to ethanol and humic acid inhibition of qPCR than other widely used analysis methods using Ct values. These methods are useful for\ud detection of sample interference, not only in fecal source identification, but also in most environmental applications of qPCR