Identifying human and animal faecal contamination in surface and drinking water : new concepts and approaches

Waterborne disease outbreaks caused by various pathogens have been reported from all over the world. They also play an important role in industrialized countries, despite good sanitation and high standards of hygiene. In Switzerland, waterborne disease outbreaks are uncommon but faecal contamination of springs occurs quite frequently in some areas. As a consequence, microbiological quality of drinking water is not constantly sufficient and there is potential for improvement. Based on the hygiene ordinance, cantonal authorities contribute with control actions to the supply of safe water to consumers. Microbiological quality assessment based on current drinking water guidelines do not discriminate between human and animal sources of contamination. However, methods for discriminating between human and animal faecal contamination are needed to identify sources of contamination and to provide specific scientific data as a basis for risk management measures and for the implementation of remedial interventions.
The main goal of this thesis was to establish approaches for the identification of human and animal faecal contamination in surface and drinking water and to develop concepts for future investigations. A number of methods can be used to help identify sources of faecal contamination in water. The general concept is referred to as microbial source tracking (MST). Different methodologies for MST were previously used mainly for analysis of faecal samples and recreational waters such as streams, lakes and beaches. Prior to this study, MST was not applied in Switzerland.
In the present thesis, source tracking approaches never before used in MST as well as methods previously described in scientific literature were tested for their potential to indicate human or animal faecal contamination in Switzerland.
Two bacterial strains (Bacteroides thetaiotaomicron ARABA 84 and B. fragilis ARABA 19) specific for bacteriophages present in human faecal contamination and three strains (B. caccae RBA 63 and RBA 64 as well as B. fragilis KBA 60) specific for bacteriophages indicating animal faecal contamination were isolated from human wastewater and animal faecal specimens. Thereafter they were used to determine the source of surface and spring water faecal contamination.
In addition, the potential of Streptococcus agalactiae was tested as a new human MST indicator. Different methods of detecting these bacteria in domestic and slaughterhouse wastewater were compared. Three DNA extraction methods and five polymerase chain reaction (PCR) assays were tested to identify the most suitable combination. The most sensitive detection method, a LightCycler real-time PCR assay, detected S. agalactiae in human wastewater but not in animal samples and showed that this bacteria is potentially useful as human MST indicator.
Rhodococcus coprophilus was previously used in MST to indicate animal faecal contamination in water. For the detection of this organism, a culture-dependent method and a conventional as well as a TaqMan real-time PCR assay were previously published. The evaluation of these existing approaches, however, did not reveal any satisfactory result. A novel LightCycler real-time PCR assay was therefore designed and validated in the course of the present thesis. Compared with previously used assays, this new molecular approach showed advantages such as improved sensitivity and specificity and was much faster than the culture-based method.
Sorbitol-fermenting bifidobacteria and phages infecting the Bacteroides host strains GA-17 and GB-124 were previously described to indicate human faecal contamination. Together with the newly developed and validated approaches, these established MST indicators were included in a comparative study and their potential to indicate human or animal faecal contamination in Swiss surface and spring water was tested. Based upon this assessment, sorbitol-fermenting bifidobacteria and phages of the human host strains B. thetaiotaomicron ARABA 84 and B. ovatus GB-124 can be recommended for detecting human faecal contamination in Swiss surface and drinking water.
The above-mentioned approaches give evidence of the faecal input sites of contamination. To localize possible faecal input sites, they can be followed by an additional approach described in this thesis. This approach is based on the screening of multiresistant Escherichia coli and on the characterisation of selected isolates by antibiotic resistance profiles and pulsed field gel electrophoresis in order to identify identical strains. Thereby, a well equipped toolbox could be provided not only to discriminate sources of faecal contamination but also to localize possible faecal input sites.