Abstract
Escherichia coli (E. coli) is a universally accepted indicator of faecal pollution in water, and can be divided into several groups of which one group would be commensal E. coli (generally the indicator organism), and five diarrhoeagenic E. coli types. Pathogenic E. coli offers attractive possibilities to model the pathogenicity of polluted water that people drink. To date, techniques used for the detection of these pathogens include standard culturing techniques, the use of molecular probes directed against known pathogens and polymerase chain reaction (PCR) for the amplification of known virulence genes, usually after a pre-culturing step. There is however a need to speed up the process and to accurately quantify (for risk purposes) the type and number of E. coli (commensal and pathogenic) present in any given water sample. The use of more recently developed techniques such as real-time PCR and competitive PCR (c-PCR) offers us a way of quantifying the target organism and has been successfully applied in various parts of the world. The aim of this study was to adapt and use developed methodologies that employs both c-PCR and multiplex PCR (m-PCR) to quantify total E. coli, detect and distinguish between the diarrhoeagenic E. coli patho-types present in selected South African water samples without prior culture or enrichment.
Using E. coli as a model pathogen, a technique was developed for the concentration of bacteria from water samples, isolation of DNA from bacteria and performing PCR’s on the extracted DNA. Three m-PCR’s were developed directed towards the housekeeping (Mdh) and virulence (eaeA, Stx1, Stx2, ST, LT, Ial and Eagg) genes associated with entero-pathogenic E. coli. c-PCR was performed using the Mdh housekeeping gene, with our modified version of the PCR product used as competitor DNA. Results obtained lead to a protocol consisting of the filtration of 100 mℓ environmental water, DNA extraction directly from the membranes, followed by quantitative c-PCR to screen for PCR inhibition as well as to quantify isolated DNA, thereafter screening of the DNA for the presence of virulence genes with m-PCR.
Initial testing with known pathogens showed that this methodology was a viable option. DNA could be recovered from the filters, yielding PCR-ready templates. A total of 49 water samples were collected, these samples included household containers from rural areas, river water from three provinces in South Africa and wastewater from 4 different wastewater treatment plants around Johannesburg (Gauteng province). These water samples were subjected to the above-mentioned protocol with 100 % (49/49) of the samples testing positive for presence of the E. coli Mdh house keeping gene. Of the samples tested, 57 % (28/49) tested positive for EAEC, 0 % (0/49) tested positive for EIEC, 92 % (45/49) tested positive for ETEC, 2 % (1/49) tested positive for atypical EHEC and 0 % (0/49) tested positive for atypical EPEC. 38 percent of the water samples could be successfully quantified by c-PCR and were able to detect as low as 3 cells/mℓ. However, the remaining 62 percent DNA samples (isolated from water samples) was diluted to overcome PCR inhibition but failed to be quantified by c-PCR because the level of target DNA was too low to detect which allowed over competition of the Mdh competitor DNA.
In conclusion, this method could successfully isolate E. coli DNA from various water samples. The isolated DNA could be used as PCR template and PCR inhibition could be overcome in all the samples by either diluting the sample or adding PCR facilitators. The PCR was able to amplify low levels of isolated E. coli DNA from most of the water samples and the presence of pathogens could be detected in the water with the m-PCR. Molecular quantification could be used to quantify DNA isolated from the water samples, but one limitation is the detection of very low numbers of E. coli DNA due to the nature of c-PCR, i.e. co-amplification of the target and competitor DNA.
Prof. Paul Jagals
Dr. T.G. Barnard
Mr. K. Maclean