Comparison of diagnostic tools and molecular based techniques for the rapid identification of Escherichia coli and coliforms in contaminated river water

Abstract

Water is an important daily requirement and in a clean, pure form, it promotes health and well-being. In addition to South Africa being one of the driest countries in the world, water availability is also being compromised by massive pollution of remaining water sources. The Berg- and Plankenburg Rivers are two of the surface water sources in the Western Cape, South Africa, which are highly polluted by sewage, industrial and agricultural run-off. The current investigation was aimed at comparing diagnostic tools, which are employed by municipalities and food industries, and molecular based techniques to routinely monitor water for indicator organisms in time- and cost-effective manner. These rivers were sampled twice a month (July 2010 to January 2011) at the sites closest to the informal settlements of Kayamandi in Stellenbosch (Plankenburg River) and Mbekweni in Paarl (Berg River). The contamination levels of the two river systems were evaluated by the enumeration of Escherichia coli and coliforms using the Colilert 18® system, Membrane Filtration (MF) and Multiple Tube Fermentation (MTF) techniques. The highest faecal coliform count of 9.2 × 106 microorganisms/100 ml was obtained in weeks 21 and 28 from the Plankenburg River system by the MTF technique, while the lowest count of 1.1 × 103 microorganisms/100 ml was obtained in week one for both river systems by the MTF technique. The highest E. coli count of 1.7 × 106 microorganisms/100 ml was obtained from the Berg River system (week 9) using the MTF technique, while the lowest count of 3.6 × 102 microorganisms/100 ml was obtained by the MF technique from the Plankenburg River system. The coliform and E. coli counts obtained by the enumeration techniques thus significantly (p > 0.05) exceeded the guidelines of 2000 microorganisms/100 ml stipulated by the Department of Water Affairs and Forestry (DWAF, 1996) for water used in recreational purposes. Overall the results obtained in this study showed that the water in the Berg- and Plankenburg River systems is highly polluted, especially where these water sources are used for irrigational and recreational purposes. For the coliform and E. coli counts obtained using the three enumeration techniques, it was noted that the MTF technique was more sensitive and obtained higher counts for most of the sampling weeks. However, the media (Membrane lactose glucuronide agar) used in the MF technique also effectively recovered environmentally stressed microbial cells and it was also better for the routine selection and growth of coliforms and E. coli. While E. coli and total coliforms were detected utilising the Colilert 18® system, accurate enumeration values for these two indicator groups was not obtained for the entire sampling period for both river systems. It has previously been shown that dilutions (up to 10-3) of highly polluted waters increase the accuracy of the Colilert 18® system to enumerate colifoms and E. coli in marine waters. As the results obtained utilising the Colilert 18® system were also not comparable to the MF and MTF techniques it is recommended that highly polluted water samples be diluted to increase the accuracy of this system as a routine enumeration technique. Water samples were directly inoculated onto MacConkey, Vile Red Bile (VRB) agar and the Chromocult Coliform agar (CCA) and single colonies were inoculated onto nutrient agar. Chromocult coliform agar proved to be more sensitive than MacConkey and VRB agar for the culturing of E. coli and coliforms. Preliminary identification of these colonies was done using the RapID ONE and API 20 E systems. The most isolated Enterobacteriaceae species by both systems, included Klebsiella pneumoniae, Klebsiella oxytoca, Escherichia coli and Enterobacter cloacae in both river systems. The API 20 E system was more sensitive in the preliminary identification of the various isolates, as greater species diversity was obtained in comparison to the RapID ONE system. The Polymerase Chain Reaction (PCR) was firstly optimised using positive Enterobacteriaceae species. The optimised method was then applied to the analysis of river water samples, which were centrifuged to harvest the bacterial cells, with DNA extracted using the boiling method. The extracted DNA was amplified using conventional PCR with the aid of species specific primers. The Enterobacteriaceae species that were detected throughout the study period in both river systems include Serratia marcescens, Escherichia coli, Klebsiella pneumoniae and Bacillus cereus. Conventional PCR was the most reliable and sensitive technique to detect Enterobacteriaceae to species level in a short period of time when compared to RapID ONE and the API 20 E systems. Multiplex PCR was optimised using the positive pathogenic E. coli strains namely, Enteropathogenic E. coli (EPEC), Enteroinvasive E. coli (EIEC), Enterohaemorrhagic E. coli (EHEC) and Enteroaggregative E. coli (EAEC). It was then employed in river water sample analysis and enabled the detection of EAEC, EHEC, and EIEC strains in Berg River system, with only the EAEC detected in the Plankenburg River system. Real-time PCR was used to optimise the multiplex PCR in the amplification of E. coli strains and successfully reduced the time to obtain final results when using control organisms. Real-time PCR was found to be more sensitive and time-effective in the identification of E. coli strains, and also more pronounced DNA bands were observed in real-time PCR products compared to conventional-multiplex PCR amplicons. To sustain the services provided by the Berg- and Plankenburg Rivers in the Western Cape (South Africa), these water sources should frequently be monitored, results assessed and reported according to the practices acknowledged by responsible bodies. It is therefore recommended that the enumeration techniques be used in conjunction with the very sensitive PCR technique for the accurate detection of coliforms and E. coli in river water samples.