Detection of PCR products of the ipaH gene from Shigella and enteroinvasive Escherichia coli by enzyme linked immunosorbent assay

Abstract

PCR techniques applied to diarrheal stools reliably diagnose Shigella and enteroinvasive Escherichia coli (EIEC) infections. Identification of PCR products using agarose gel electrophoresis (AGE) and hybridization with DNA probes has several shortcomings. Automated methods of identifying PCR products that process larger numbers of specimens can facilitate epidemiologic studies and standardize results. In this study, we used ELISA following PCR to detect ipaH gene sequences of Shigella and EIEC from 89 diarrheal stools. Results of ELISA were compared with AGE with and without DNA probe, and with culture. Two specimen preparation methods were compared as well: boiling/centrifugation, and purification with silicon dioxide (SiO₂). Both PCR product-detection methods identified significantly more infections than did culture. PCR-ELISA detected significantly more infections than PCR-AGE when processed using SiO2 (P = 0.014). PCR-ELISA allows screening of larger numbers of specimens, automates test results, and avoids use of mutagenic reagents. PCR-ELISA is faster than PCR-AGE when testing large numbers of specimens, although not when testing small numbers of specimens.

Introduction

Molecular diagnostic tests applied to diarrheal stools are now well established in enteric infections research laboratories. Current application of recombinant DNA technology to diagnosis of both Shigella spp. and EIEC is based largely on identification of DNA segments of a critical 120–140 MDa virulence plasmid that is necessary for attachment and invasion of epithelial cells Sansonetti et al 1981, Sansonetti et al 1982. An invasion-associated plasmid antigen, ipaH, has been described, the genes for which are found in multiple copies on both the invasion plasmid and on the chromosome of both Shigella spp. and EIEC (Hartman et al., 1990). This gene is an appealing target for diagnostic tools, as it remains detectable despite the loss of the plasmid.

An alkaline phosphatase-labeled probe for the ipaH gene has been developed and tested on stool specimens after 6 hours enrichment on MacConkey agar, and found to have a sensitivity and specificity of 85% and 95%, respectively, when compared to idealized stool collection and culturing methods (Oberhelman et al., 1993). This promising test requires roughly 1.5 days to be conducted on a practical level in most laboratories, a considerable reduction in time from the usual requirement of 3–4 days. Under the best of circumstances, however, it is clear that probes developed to date do not improve the test characteristics of sensitivity and specificity over standard assays. Their appeal is based instead on the promise of more rapid or convenient diagnoses.

Molecular diagnosis advanced in the early 1990s with the application of PCR techniques. The plasmid invasion-associated locus (ial) Echeverria et al 1992, Frankel et al 1990 and the ipaH gene Gaudio et al 1997, Sethabutr et al 1993, Sethabutr et al 1994 have been targeted for amplification and have proven to be dramatically more sensitive than either standard methods or DNA probes when applied to stools from both symptomatic and asymptomatic subjects. In studies from Thailand, an ipaH PCR system was shown to increase detection of Shigella and EIEC infection by 36–45% among patients with acute dysentery Gaudio et al 1997, Sethabutr et al 1994. The increase in sensitivity proved even higher among patients who had received prior antibiotics or had samples cultured several days after the onset of illness. The detection method used to identify PCR products in these studies has been traditional agarose gel electrophoresis (AGE) followed by hybridization of AGE-negative specimens with DNA probes to identify faint (nonvisualized) bands. While PCR-AGE alone is simple and rapid, the result is subjective, usually involves the use of a mutagenic reagent (ethidium bromide) to illuminate the bands, requires special equipment, and limits the number of specimens processed.

The present study describes the application of ELISA methodology to detection of PCR products of ipaH gene sequences from Shigella/EIEC in diarrheal stool samples in comparison to PCR-AGE with and without ipaH DNA probe, and to culture. The PCR-ELISA system involves the initial amplification of the target sequence incorporating a nucleotide (dUTP) labeled with digoxigenin. This labeled target sequence is then hybridized to a complementary ipaH-derived oligonucleotide that itself is labeled with biotin. Biotin forms a strong association with streptavidin, which is commercially available on a ready made ELISA plate; thus the hybrid digoxigenin-target sequence-biotin complex is added to the ELISA plate and binds via the biotin-streptavidin interaction. Detection of the bound complex is accomplished via a simple alkaline phosphatase labeled antibody directed against digoxigenin, with color developed using a suitable substrate.

An additional comparison was made between two methods of target DNA preparation. The traditional method used in our laboratory involves purification of the stool suspension with a silicon dioxide binding technique (Boom et al., 1990), which is tedious and time consuming, but effective at removing inhibitors of PCR in stool. This technique was compared to a simple and rapid boiling/centrifugation method applied directly to stool rather than to isolates (Houng et al., 1997), which promises to shorten the overall time requirement for the test considerably. Both methods were conducted on each sample concomitantly and results compared.