Detection of Toxin Producing Strains of E. coli
Report No DWI0674

Apr 1995


This research was undertaken under a Department of the Environment Research Contract reference EPG 1\9\22 and was managed by the Drinking Water Inspectorate. The work was carried out by Strathclyde Water Services in three parts, in accordance with the objectives set out in the contract:

  1. Techniques suitable for routine identification and enumeration of E. coli 0157:H7 in water were reviewed.

  2. A suitable method was selected and evaluated in terms of its sensitivity, reproducibility, robustness and convenience in use.

  3. The method was more fully assessed in a collaborative trial involving five laboratories in Scotland, England and Wales. In the review it was concluded that the proposed method in 'Methods for the Examination of Water and Associated Materials' (Report 71)' which recommends 24 hour culture in selective broth, subculture to selective sorbitol MacConkey agar and confirmation of suspect colonies by latex agglutination and biochemical profile, remains the most suitable procedure for use in a routine laboratory. The method also recommends Immunomagnetic Separation (IMS) as an additional selective tool.

Contrary to this, the method described by Chapman et al, from which the proposed Report 71 method was derived, recommends incubation of the selective broth for 6 hours. The effects of incubation for 6 hours and 24 hours were examined, and it was found that 6 hours incubation was adequate for isolation of E coli 0157:H7. whereas 24 hours incubation allowed overgrowth of other coliform species, resulting in a loss of sensitivity. The method tested in the remainder of the study was that requiring a 6 hour incubation period.

This method was evaluated for its ability to isolate E. coli 0157:H7 from a range of seeded source waters. The method was sensitive, and its performance was enhanced by IMS.

Further evaluation of the method in the collaborative trial using a broader range of water qualities confirmed its sensitivity and reproducibility, with isolations from 73% of seeded samples .

When the coliform counts of water samples were below 1000 per 100ml, E. coli 0157:H7 was isolated from 81 - 82% of seeded samples, however with sample coliform counts higher than this only 38% yielded the target organism.

Those working in the collaborative trial commented on the straightforwardness of the method and the minimal cost of equipment and reagents. All found the IMS procedure easy to implement.

Copies of this report may be available as an Acrobat pdf download under the 'Find Completed Research' heading on the DWI website.