Comparison of quantitative PCR assays for Escherichia coli targeting ribosomal RNA and single copy genes.

AIMS: Compare specificity and sensitivity of quantitative PCR (qPCR) assays targeting single and multi-copy gene regions of Escherichia coli. METHODS AND RESULTS: A previously reported assay targeting the uidA gene (uidA405) was used as the basis for comparing the taxonomic specificity and sensitivity of qPCR assays targeting the rodA gene (rodA984) and two regions of the multi-copy 23S ribosomal RNA gene (EC23S and EC23S857). Experimental analyses of 28 culture collection strains representing E. coli and 21 related non-target species indicated that the uidA405 and rodA984 assays were both 100% specific for E. coli while the EC23S assay was only 29% specific. The EC23S857 assay was only 95% specific due to detection of E. fergusonii. The uidA405, rodA984, EC23S and EC23S857 assays were 85%, 85%, 100% and 86% sensitive, respectively, in detecting 175 presumptive E. coli culture isolates from fresh, marine and waste water samples. In analyses of DNA extracts from 32 fresh, marine and waste water samples, the rodA984, EC23S and EC23S857 assays detected mean densities of target sequences at ratios of approximately 1:1, 243:1 and 6:1 compared with the mean densities detected by the uidA405 assay. CONCLUSIONS: The EC23S assay was less specific for E. coli, whereas the rodA984 and EC23S857 assay taxonomic specificities and sensitivities were similar to those of the uidA405 gene assay. SIGNIFICANCE AND IMPACT: The EC23S857 assay has a lower limit of detection for E. coli cells than the uidA405 and rodA984 assays due to its multi-copy gene target and therefore provides greater analytical sensitivity in monitoring for these faecal pollution indicators in environmental waters by qPCR methods.

Improved real-time PCR assays for the detection of fecal indicator bacteria in surface waters with different instrument and reagent systems.

Previously reported and redesigned primer and probe assays were evaluated for the quantitative analysis of the fecal indicator bacterial groups, Enterococcus and Bacteroidetes with three real-time PCR instrument and reagent systems. The efficiency and sensitivity of the original assays varied between systems in analyses of DNA extracts from pure cultures of Enterococcus faecalis and Bacteroides fragilis, whereas the modified assays gave more consistent results. Distinctions between original and modified assays also occurred in analyses of known spike levels of E. faecalis and B. fragilis cells on filters with diverse surface water retentates. Percentages of samples causing PCR failures due to inhibition were lower using the modified assays. The accuracy and precision of spiked bacteria measurements were also generally higher, although mean measurements of both target organisms were still significantly different between systems (p < 0.05). The accuracy and precision of spiked bacteria measurements by both modified assays were further improved using a new sample matrix control spike consisting of cultured Lactococcus lactis cells and a reference assay for this organism. Corrections provided by the L. lactis assay eliminated significant differences in E. faecalis measurements between all three systems and between two of the three systems in B. fragilis measurements.