Diagnostic dilemma: application of real-time PCR assays for the detection of Dientamoeba fragilis in medical and veterinary specimens.

Background: Real-time PCR (qPCR) diagnostics developed for use in human clinical settings have been implemented to identify new animal hosts of the gastrointestinal protozoan Dientamoeba fragilis. The gut microbiome varies between species; unrecognised cross-reactivity could occur when applying these assays to new animal hosts. The use of qPCR diagnostics was assessed for the identification of new animal hosts of the gastrointestinal protozoan Dientamoeba fragilis.

Methods: Forty-nine cattle, 84 dogs, 39 cats and 254 humans were screened for D. fragilis using two qPCR assays: EasyScreen (Genetic Signatures) and a laboratory-based assay commonly used in Europe. The reliability of the identifications made by these assays were assessed using melt curve analysis of qPCR products, conventional PCR targeting the SSU rDNA sequencing and NGS amplicon sequencing of qPCR product.

Results: PCR products from the D. fragilis identified in cattle had a 9 °C cooler melt curve than when detected in humans. This melt curve discrepancy, indicative of cross-reactivity with an unknown organism, was investigated further. DNA sequencing determined that Simplicimonas sp. was the genera responsible for this cross-reactivity in cattle specimens. Dientamoeba fragilis was not detected in either dogs or cats. There was a discrepancy in the number of positive samples detected using the two qPCR assays when applied to human samples. The EasyScreen assay detected 24 positive samples; the laboratory-based assay detected an additional 34 positive samples. Of the discrepant samples, 5 returned sequence data for D. fragilis, and 29 were unsupported (false) positive samples.

Conclusions: Analysis of the melt curve after the qPCR reaction is a valuable technique to help differentiate samples containing D. fragilis compared to cross-reactions with non-target organisms. The identification of new animal hosts requires further evidence from either microscopy or DNA sequencing to confirm the presence of D. fragilis. Additionally, to reduce the risk of false-positive results due to non-specific amplification, we recommend reducing the number of PCR cycles to less than 40. Based on these results, we consider the ramifications of this identified cross-reactivity to the known host species distribution of D. fragilis.