Development of rep-PCR and PFGE methods for bacterial source tracking

Wail Mostafa Hassan

Abstract

The goal of the current study was to develop efficient bacterial source tracking capabilities to identify sources of fecal pollution in south Mississippi waters. The use of repetitive extragenic palindromic polymerase chain reaction (rep-PCR) was compared to the use of pulsed field gel electrophoresis (PFGE) in terms of source tracking fidelity. Rep-PCR was tested using BOX- and REP-specific primers, where the technique is called BOX-PCR and REP-PCR, respectively. In addition, the fidelity of source tracking was compared using two target organisms, Escherichia coli and Enterococcus spp. The current study showed that the highest fidelity of source tracking was achieved using BOX-PCR for fingerprinting and Enterococcus spp. as a target organism for source tracking. In addition, the method by which unknown isolates were assigned to animal sources was optimized. Source assignment based on maximum similarity (in which isolates were assigned to the animal source containing the most similar fingerprint) yielded higher fidelity compared to average similarity (in which isolates were assigned to the animal source with the highest average fingerprint similarity to the unknown's fingerprint). The current study also showed that applying a similarity value (the similarity of the unknown fingerprint to the source group to which it was assigned) threshold and/or quality factor (the ratio between the average similarity of a source's fingerprints to each other, on one hand, and their average similarity to the fingerprint of an isolate assigned to this source, on the other hand) threshold was essential for the fidelity of source tracking. Finally, enterococcal isolates obtained from the major rivers near Hattiesburg, Mississippi and from Mississippi Gulf Coast were assigned to animal sources using the constructed library. The freshwater samples collected from the Hattiesburg area showed no human fecal contamination. The main source of fecal contamination appeared to be chicken feces. The Gulf Coast samples showed occasional human fecal contamination, although most enterococcal isolates were of gull or chicken origin.