B09008: Accelerated detection of salmonella and verocytotoxigenic E.coli in food

Study Duration: June 1999 to July 2002

Contractor: University of Reading

The Food Standards Agency requires reliable information on the sources of foodborne pathogens and how they are transmitted through the food chain, in order to formulate strategies for reducing the incidence of foodborne illness in the population. This information is gained by microbiological examination of food, water, animals and the environment.

Conventional culture methods for detecting foodborne pathogenic bacteria have proved their usefulness over many years, but are often slow, labour-intensive and lacking in specificity. One such technique, the polymerase chain reaction (PCR), allows a particular section of an organism's DNA, a unique signature of its presence, to be amplified and detected in a few hours. Despite the potential advantages of such methods, their application to food has been slow in developing. Simple reliable protocols are needed that are specifically tailored to food applications.

The purpose of this project was to develop and characterise accelerated methods for detecting two important groups of foodborne pathogens: Salmonella and verocytotoxin-producing strains of Escherichia coli (VTEC).

The method will be based on a preliminary incubation of the food sample in growth medium (enrichment culture) to allow resuscitation of debilitated bacteria and cell multiplication, followed by a PCR assay that is specific for the target organism.

A new Polymerase Chain Reaction (PCR) method was developed that detects all Salmonella strains and indicates the presence of the specific Salmonella strain that is responsible for most food poisoning associated with eggs and poultry (Salmonella Enteritidis). This PCR, plus one for VTEC organisms based on published information, was used to evaluate the performance of enrichment PCR detection methods.

The simplest method of extracting DNA from cells was by boiling of the cells. With this method, 1,000 Salmonella cells per ml could be detected by PCR under ideal conditions. The VTEC PCR was less sensitive and 1,000 to 100,000 cells were needed for a positive result. The sensitivity of both PCR assays was much reduced when applied to enrichment broths containing minced beef, bean sprouts, milk or raw egg, particularly when the simple boiling procedure was used. Under these conditions, tens of millions of cells were needed. A modified DNA extraction method was devised and this improved sensitivity, allowing 1,000,000 cells per ml to be detected in the presence of food. A commercially available enrichment medium used for Salmonella detection inhibited the PCR reaction severely, but this could be overcome using the modified DNA extraction procedure.

A method for following the growth of salmonellae or VTEC in enrichment broths containing a large excess of other food bacteria was developed. Growth of Salmonella starting from 2 or 20 cells was examined in enrichment broths containing various foods. Raw egg contained only low numbers of competing microbes and did not inhibit growth of salmonellae. The microbes in bean sprouts or beef mince were inhibitory, but the final numbers of salmonellae reached after 24 hours enrichment could still be detected by PCR. The microbes in milk were very inhibitory to salmonellae and, in some cases, final numbers reached only 1,000 per ml, which would make PCR detection doubtful. However, the use of Salmonella Sprint Broth (SSB) overcame the growth inhibition problem and allowed detection by PCR. Exposure to heat stress reduced final cell counts to numbers that would be difficult to detect by PCR.

The growth of VTEC organisms in enrichment broth was not suppressed by the natural microflora of salami or apple juice and cell numbers present after enrichment were easily detected by PCR. Storage of VTEC organisms in apple juice for periods of up to two weeks did not appear to impair their ability to grow in enrichment broths. With minced beef, pronounced growth inhibition was seen. The final number of VTEC organisms present after enrichment was directly related to the number of VTEC cells originally present in the food, and varied between 1,000 and 100,000,000 cells per ml. The lowest number in this range would be marginal for PCR detection.

Final report is available from the FSA Library and Information Centre.
To obtain a copy, please contact the Enquiry Desk, Dr Elsie Widdowson Library and Information Services, Food Standards Agency ( Tel: 020 7276 8181/8182 or e-mail: library&info@foodstandards.gsi.gov.uk ).

Contact: Email: science@foodstandards.gsi.gov.uk