Report No FR0190

P Gale and P Broberg

Mar 1991



To investigate, in terms of species specificity, sensitivity and detection time, the potential of integrating the GENE-TRAK Escherichia coli and Salmonella assay kits into most probable number (MPN) methods for enumerating confirmed E coli or salmonellae in water samples.


Gene probe technology offers the potential advantage of absolute specificity in the species of micro-organism detected and hence the possibility for confirmation techniques which are not only more rapid than the standard methods routinely used by the water industry but which are also less susceptible to registering false identifications. GENE-TRAK have developed gene probe-based methods for confirming the presence or absence of E coli and Salmonella bacteria in food samples within three working days. However, for the water industry a simple presence/absence result is not sufficient and quantification of bacterial numbers is also required with a limit of sensitivity of one organism per 100 ml of water. WRc has therefore attempted to integrate the GENE-TRAK assay kits into an MPN method to produce a technique which is not only as sensitive and specific as the current UK standard methods for the enumeration of confirmed E coli and salmonellae in water samples, but which is also more rapid.


  1. The GENE-TRAK E coli and Salmonella assay kits have been integrated into six tube format (1 x 50 ml, 5 x 10 ml) MPN methods, devised by WRc, such that the number of confirmed E coli or salmonellae in 100 ml volume water samples can be determined within two working days. In terms of detection time this is a considerable improvement over current UK standard methods.

  2. One problem encountered with the GENE-TRAK E coli assay kits was that the readings obtained for the Positive Control often failed to meet that required for validation of the test as stated by the manufacturers.

  3. The method devised by WRc and utilising GENE-TRAK for E coli appears very encouraging in that it may enumerate E coli in river water samples with as few as 1 E coli per 100 ml (as judged by UK standard MPN method) within two working days. Statistically, however, the sensitivity of this method is significantly less than that of the UK MPN standard method, although this may reflect the GENE-TRAK kit not working to design specifications (see above).

  4. In terms of specificity of the GENE-TRAK E coli assay kit, very good agreement was obtained between colonies which were isolated from river water samples, confirmed as E coli by GENE-TRAK and identified by API 20E as E coli. Fewer colonies were confirmed as E coli (production of indole from tryptone water and acid and gas from lactose at 44C) by accepted UK standard methods although all of these, with the exception of one, were identified as positive by both GENE-TRAK and API 20E.

  5. Evaluations performed so far on the method for salmonellae, utilising the GENE-TRAK Salmonella assay kit and requiring only two working days to complete, demonstrate that for deionised water samples spiked with pure Salmonella cultures, the method begins to fail at Salmonella concentrations below 30 per 100 ml, although as few as two salmonellae per 100 ml (as judged by the standard method) were successfully enumerated in one water sample. Difficulties in obtaining GENE-TRAK Salmonella assay kits have prevented evaluation of the sensitivity from being completed and tests on natural water samples with other heterotrophic bacteria present from being performed.

  6. Preliminary studies on the specificity of the GENE-TRAK Salmonella kit show that non-Salmonella species of bacteria commonly found in river water samples are not detected.


The GENE-TRAK assay kits are readily integrated into MPN methods for enumeration of bacteria in water samples and thus have potential use in the water industry in replacing conventional detection methods. Of particular significance to the water industry is that, in terms of species specificity, the GENE-TRAK E coli assay kit appears to detect E coli which are identified by API 20E but not by the standard UK method and in this respect GENE-TRAK may generate fewer false negatives.

The GENE-TRAK based method, which permits enumeration of E coli in water samples within two working days, is encouraging and further research should be directed at developing a culturing method such that the sensitivity requirements of the GENE-TRAK test kit are consistently satisfied. One possibility would be to perform the 21 hr culture step at 44C, allowing only the thermotolerant organisms to grow. Further work should be performed to complete the assessment of the potential of the Salmonella assay kit.

The protocol for testing food samples for the presence of salmonellae as recommended by GENE-TRAK requires three working days to complete. This method was not applied to water samples in this study and it certainly deserves assessment since the standard WRc method for salmonellae requires at least six working days to complete.


This report does not attempt to evaluate the GENE-TRAK assay kits in the mode of operation or the applications recommended by the manufacturers. Instead the GENE-TRAK kits were integrated into MPN methods, devised by WRc, with the purpose of assessing if in this form they provided a more rapid and specific means of analysing water samples than current standard methods. Section 1 outlines the requirements of any enumeration method used for analysing water samples and overviews the potential advantages of using gene probe based techniques for such purposes. In Section 2, the GENE-TRAK kit is described, although details of the molecular biology involved and the protocol of the assay are omitted. The strategy for integrating the GENE-TRAK assay kits into MPN methods and for conducting the experiments to compare the sensitivity with that of the standard technique is outlined in Section 3. The design of an MPN method, based on the GENE-TRAK E coli assay kit, for enumerating confirmed coli in water samples within two working days is explained in Section 4 and, on the basis of results obtained from analysis of 18 water samples by six tube format MPN, the sensitivity of the new method is compared with that of the standard method. In a similar manner, Section 5 assesses the potential of the WRc devised method, using GENE-TRAK and requiring two days for completion, to enumerate salmonellae in water samples. Of paramount importance to any gene probe method is that it is indeed absolutely specific for the micro-organism it is designed to detect. Species specificity of the GENE-TRAK E coli probe is addressed in Section 6, where 50 colonies isolated from river water samples are identified by API 20E and UK standard methods (for E coli) and tested for detection by GENE-TRAK. In Section 7, preliminary studies on the species specificity of the Salmonella GENE-TRAK kit are reported.

Copies of the Report are available from FWR, price 25.00 less 20% to FWR Members