Potential pitfalls in the quantitative molecular detection of Escherichia coli O157:H7 in environmental matrices

Rebekka R.E. Artz, Lisa M. Avery, Davey L. Jones, Ken Killham

Research output: Contribution to journalArticlepeer-review

30 Citations (Scopus)


The detection sensitivity and potential interference factors of a commonly used assay based on real-time polymerase chain reaction (PCR) for Escherichia coli O157:H7 using eae gene-specific primers were assessed. Animal wastes and soil samples were spiked with known replicate quantities of a nontoxigenic strain of E. coli O157:H7 in a viable or dead state and as unprotected DNA. The detection sensitivity and accuracy of real-time PCR for E. coli O157:H7 in animal wastes and soil is low compared to enrichment culturing. Nonviable cells and unprotected DNA were shown to produce positive results in several of the environmental samples tested, leading to potential overestimates of cell numbers due to prolonged detection of nonviable cells. This demonstrates the necessity for the specific calibration of real-time PCR assays in environmental samples. The accuracy of the eae gene-based detection method was further evaluated over time in a soil system against an activity measurement, using the bioluminescent properties of an E. coli O157:H7 Tn5luxCDABE construct. The detection of significant numbers of viable but nonculturable (VBNC) as well as nonviable and possibly physically protected cells as shown over a period of 90 days further complicates the use of real-time PCR assays for quick diagnostics in environmental samples and infers that enrichment culturing is still required for the final verification of samples found positive by real-time PCR methods.

Original languageEnglish
Pages (from-to)482-488
Number of pages7
JournalCanadian Journal of Microbiology
Issue number5
Publication statusPublished - 1 May 2006
Externally publishedYes


Dive into the research topics of 'Potential pitfalls in the quantitative molecular detection of Escherichia coli O157:H7 in environmental matrices'. Together they form a unique fingerprint.

Cite this