Elsevier

Molecular and Cellular Probes

Volume 21, Issues 5–6, October–December 2007, Pages 400-404
Molecular and Cellular Probes

Real-time PCR for the detection of Dientamoeba fragilis in fecal samples

https://doi.org/10.1016/j.mcp.2007.05.006Get rights and content

Abstract

A real-time polymerase chain reaction (PCR) method targeting the 5.8S ribosomal RNA gene was developed for the detection of Dientamoeba fragilis in stool samples. The PCR also included an internal control to detect inhibition of the amplification by fecal constituents in the sample. The assay was performed on species-specific DNA controls (n=29) and a range of stool samples (n=85), and achieved high specificity and sensitivity. D. fragilis DNA could be detected in unpreserved fecal samples up to 8 weeks after storage at 4 °C. The use of this assay in a diagnostic laboratory offers the possibility of introducing DNA detection as a feasible technique for the routine diagnosis of intestinal D. fragilis infections.

Introduction

Dientamoeba fragilis was first described in 1918 as an amoeba living as a commensal in the gastro-intestinal tract of humans [1]. Antigenic comparisons, ultrastructural studies and analysis of the 16S-like ribosomal RNA have reclassified this organism as a flagellate although a flagellum is lacking [2], [3]. Since its discovery, the pathogenicity of this organism has remained controversial. Although in recent years several authors have published on the clinical importance of D. fragilis as a cause of gastro-intestinal symptoms (reviewed by Johnson et al. [4] and Stark et al. [5]), there is still no consensus on its pathogenicity [6], [7], [8]. This controversy has resulted in differences between laboratories in their diagnostic approach of intestinal protozoa. In most countries outside the United States, microscopy for the detection of protozoa and ova is performed, traditionally, on a direct smear or after formalin-ether concentration using unpreserved stool. However, because D. fragilis lacks a cyst stage microscopic examination has to be performed on freshly passed stool or by the use of fixatives and permanent stains to detect the fragile trophozoites. Moreover, day-to-day shedding of D. fragilis trophozoites appears to be even more irregular than what is observed in other intestinal protozoan infections like Giardia lamblia and Entamoeba histolytica [4], [9].

Although polymerase chain reaction (PCR)-based methods have been used successfully for the detection of intestinal parasitic infections, their application in routine diagnosis is still limited. Introduction of PCR-based methods has been hampered by difficulties in the DNA extraction from fecal samples. Moreover, the amplification and detection of DNA was prone to contamination as well as being time-consuming and expensive. In recent years, however, the isolation of parasitic DNA from fecal samples has been improved and simplified [10], [11]. Moreover, automated DNA isolation procedures are developed and are implemented quite rapidly in diagnostic laboratories. The introduction of real-time PCR using fluorescent detection probes [12] has reduced the risk of contamination, labor time and reagent costs through the possibility of combining assays for the detection of different targets in one assay.

In recent years one conventional PCR method for the detection of D. fragilis and two for the detection and RFLP typing of D. fragilis have been published [13], [14], [15]. These three conventional methods targeted the whole or a relatively large part of the small subunit ribosomal RNA gene (SSU rRNA) resulting in PCR products of 1700, 887, and 662 bp, respectively. Amplification of such large fragments, especially using DNA extracted from feces will limit the sensitivity of such assay and therefore limit its use as a diagnostic tool. Recently, also a real-time PCR for the detection of D. fragilis has been published targeting only a small fragment of 77 bp on the SSU rRNA gene [16]. However, BLAST search showed that the forward primer and the Taqman detection probe based on the SSU rRNA gene to be non-specific, which can influence the efficiency of the PCR by annealing of the forward primer and probe to non-target sequences. Moreover, as this real-time PCR was a separate assay without an internal control the advantages of using real-time PCR were not fully exploited.

Therefore, a new real-time PCR based on the 5.8S ribosomal RNA gene was developed for the detection of D. fragilis in fecal samples. Additionally, an internal control for the detection of possible inhibition of the amplification by fecal contaminants was included in the assay. The performance of the assay was evaluated using a range of controls and fecal samples.

Section snippets

Controls and samples

D. fragilis control DNA was obtained from cultured D. fragilis (ATCC 30948).

Fecal samples were collected from different patients on 3 consecutive days using the Triple Feces Test (TFT) [9]. On day 1 and day 3, stools are directly mixed with SAF preservative (TFT1 and TFT3) and on the second day an unpreserved stool specimen (TFT2) is collected. The complete set was sent by mail to the laboratory. An aliquot of the unpreserved sample was mixed with ethanol (≈1 g/ml ethanol) for DNA isolation. For

Results

The real-time PCR was optimized first as monoplex assay with a 10-fold dilution series of D. fragilis DNA. The monoplex real-time PCR was thereafter compared to multiplex PCR with the PhHV internal control. The cycle threshold (Ct) values obtained from testing the dilution series of D. fragilis in the individual assay and in the multiplex assay were similar, and the same analytical sensitivity was achieved. The individual performance of the assay was not influenced by the presence of DNA from

Discussion

Using well-defined control DNA and 30 fecal samples from patients in which two preserved fecal samples did not show any parasites the real-time PCR for the detection of D. fragilis achieved 100% specificity. In a selected group of 55 fecal samples from patients in which one or two preserved samples showed D. fragilis trophozoites, the real-time PCR performed on one fecal sample showed a sensitivity of 89 percent considering microscopy of two corresponding SAF-preserved fecal samples as gold

Acknowledgments

We thank Dr. John Ackers for critically reading the manuscript.

References (21)

There are more references available in the full text version of this article.

Cited by (97)

View all citing articles on Scopus
View full text