INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

During the summer season, enteroviruses are responsible for the majority of viral diseases among pediatric and adult patients. Approximately 10 to 15 million symptomatic infections due to enteroviruses occur each year, resulting in a variety of disease syndromes (9). A rapid laboratory diagnosis of an enterovirus infection is important in patient care and management (e.g., decisions about antibiotic use and length of hospital stay). The significance of rapid enterovirus diagnostics is further underscored by recent progress in enterovirus drug research (8).

Enteroviruses in general grow rapidly in cell culture. Most clinical enterovirus strains are isolated within 4 to 5 days of conventional culture inoculation (S. M. Lipson, unpublished observations). The need to perform subpassages on toxic cultures and/or the performance of culture confirmation may add an additional 24 or more h to test turnaround time. In one study with stool isolates, for example, a mean time to isolation and confirmation of 11.5 ± 5 days was reported (1).

Molecular diagnostics, both "automated" and nonautomated, have been introduced to the laboratory medicine community. Molecular enterovirus testing, although more sensitive than conventional cell culture (14) is expensive and requires a designated laboratory facility. Furthermore, the enterovirus season is short-lived (viz., commonly 6 to 8 weeks), which raises questions about the cost-effectiveness of bringing enterovirus genome amplification technology into the general virology or microbiology laboratory setting.

Antigen detection-cell culture amplification (Ag-CCA) has been shown to be effective for the rapid detection of viruses in clinical specimens (3, 6). However, in the clinical laboratory setting, use of this technology has been limited primarily, if not totally, to nonenterovirus groups. Several research teams have addressed the potential application of an enterovirus Ag-CCA assay (4, 11, 12). However, those studies did not employ freshly collected specimens and utilized in-house-seeded shell vials or microtiter platescharacteristic of those found in the research setting.

The purpose of this study was to determine the efficacy of the enterovirus Ag-CCA assay by using commercially manufactured shell vials and a current generation (and heretofore untested) monoclonal antibody (MAb), as well as by incorporating freshly collected clinical (field) specimens.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Clinical specimens and reference viruses. The initial phase of this study utilized reference enterovirus isolates, including 10 coxsackievirus A and B, echovirus, and poliovirus strains. All strains were obtained from the Virology Service (North Shore University Hospital, Manhasset, N.Y.) Culture Collection and from New York State Proficiency test samples.

Cell culture. All clinical (field) specimens were inoculated into A549, rhesus monkey kidney (RhMk), and MRC-5 conventional tube cultures (TCs). Hep-2 and HNK TC, were added to the routine cell culture panel based upon the specimen source and clinical symptoms (e.g., vesicular lesions) upon clinical presentation. Culture confirmation of the enterovirus cytopathic effect (CPE) was performed by indirect immunofluorescence testing with reagents 5-D8/1 and 2E11 (see below).

Virus titration. The titers of all reference enterovirus strains were determined in primary RhMk conventional TCs. Briefly, RhMk TCs were inoculated in pentuplicate at 10-fold serial dilutions. Cultures were incubated at 36.5°C for 14 days. Viral titers were determined by the Reed-Muench end point procedure (7).

Ag-CCA. (i) Immunoreagents. Enterovirus clone 5-D8/1 (catalog no. M7064; DAKO Corporation, Carpinteria, Calif.) and the Pan-Enterovirus 2E11 MAb (catalog no. 3362; Chemicon International, Inc., Temecula, Calif.) immunoreagents were used in the establishment of the enterovirus Ag-CCA assay. Reagent 5-D8/1 reacts with the highly conserved VP1 region of the enterovirus (11). Reagent 2E11 is an enterovirus group-specific MAb to the viral capsid (12). Preliminary titration experiments revealed that each immunoreagent attained an optimal immunofluorescent signal at the concentration supplied by the manufacturer.

(ii) Specimen inoculation and antigen detection. RhMk shell vials were purchased from BioWhittaker, Inc., Walkersville, Md. Briefly, RhMk shell vials were inoculated in duplicate with 0.2 ml of enterovirus reference strains (at the viral end point) or field (clinical) specimens. Shell vials were centrifuged at 600 × g for 60 min at 36 ± 1°C, followed by the addition of serum-free maintenance medium. After 48 h for reference strains and 24, 48, 72, 96, and 120 h for field strains, the vials were washed with 1× phosphate-buffered saline (PBS), harvested, fixed in acetone at 4°C, and then immunostained with MAb 5-D8/1 and/or 2E11. Clone 2E11 was used in initial and field strain testing. Clone 5-D8/1 was used in initial assay development only. The secondary immunostaining reagent consisted of a fluorescent-conjugated goat anti-mouse whole-molecule MAb supplemented with Evans blue counterstain (Organon Teknika Corp., Cappel Research Reagents, Durham, N.C.). The negative control consisted of PBS in place of the primary immunoreagents. Conventional RhMk, A549, MRC-5, Hep-2, and HNK TCs (Hep-2 and HNK TCs were used to supplement RhMk, A549, and MRC-5 in the testing of ca. 60% of field specimens) were inoculated in parallel with all RhMk shell vials. Positive signals were identified by the appearance of a cytoplasmic apple-green stippling or a broad cytoplasmic apple-green signal among the reagents 2E11 and 5-8D/1, respectively. Readings were reported as mean focus-forming units per two shell vials. Shell vials were read with a Nikon UFX epifluorescent microscope equipped with a 100-W mercury bulb. The RhMk monolayers were read under a ×40 objective; confirmatory readings, if necessary, were performed with a ×50 oil immersion objective.

Specimen interpretation. AG-CCA, TC-CPE-negative specimens were considered true negatives. AG-CCA-positive, TC-CPE-positive results were interpreted as true-positive specimens. Among the 273 clinical specimens tested in this study, a single AG-CCA-positive, TC-CPE-negative specimen (no. 2127) was deemed uninterpretable; resolution testing of this sole discordant specimen would not have a significant impact on the results of this study.

Statistics. The McNemar test was used to compare qualitative differences in virus detection between cell cultures. Sensitivity, specificity, and positive and negative predictive values (PPV and NPV, respectively) of the AG-CCA and TC-CPE assays were calculated according to standard procedures (10).

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Comparison of cell types for the isolation of NPEVs. Prior to laboratory testing of the Ag-CCA assay, there was a need to confirm the cell type within our (cell culture) panel that permitted maximum isolation of non-polio enterovirus (NPEV) strains seen in our clinical setting. Among 32 clinical specimens consecutively tested with A549, primary RhMk, MRC-5, Hep-2, and HNK cells, RhMk and MRC-5 cells demonstrated the greatest NPEV yields (Table 1). These findings were similar to those reported by Chonmaitree et al. (2), but indicated that MRC-5 was slightly more sensitive than monkey kidney. Although no statistical difference was recognized in our comparison of RhMk and MRC-5 cells for the isolation of NPEVs from clinical specimens (P = 0.27), the slightly higher recovery rate observed with RhMk directed our use of RhMk as the sole cell type for the performance of the Ag-CCA assay.

Ag-CCA assay standardization under laboratory conditions. The efficacies of MAb clones 2E11 and 5-D8/1 were evaluated in parallel with 10 coxsackievirus, echovirus, and poliovirus strains. Antibody 2E11 and 5-D8/1 immunostaining patterns and intensities of signal were compared to isolation in conventional RhMk TCs utilizing stock enterovirus cultures at viral end points (7). Clone 5-D8/1, in concert with Ag-CCA, was more sensitive than conventional TC isolation (TC-CPE) among 4 of 10 enterovirus strains. Clone 2E11 was more sensitive than TC-CPE among 8 of 10 enterovirus strains. Ag-CCA utilizing clone 5-D8/1 was less sensitive than TC-CPE among 2 of 10 laboratory-adapted strains. The Ag-CCA assay incorporating clone 2E11 was equal to or more sensitive than TC-CPE among all 10 laboratory-adapted enterovirus strains (Table 2). These findings may be ascribed to unique differences in immunofluorescent staining patterns and signal intensities produced by the respective MAb clones. As shown in Fig. 1, clone 5-8D/1 produced a broad cytoplasmic immunofluorescent pattern that was difficult to discern from background fluorescence near the viral end point. In contrast, MAb clone 2E11 demonstrated a more recognizably definitive cytoplasmic apple-green immunofluorescent stippling pattern, permitting a readily interpretative signal at low viral titers. It is proposed that the use of enterovirus clone 2E11, rather than clone 5-D8/1, would more appropriately address the issue of low-level viral antigen detection within the centrifugation assay system herein described.

View larger version (86K): [in this window] [in a new window]   FIG. 1.   Immunofluorescent signal patterns obtained by using reagents 2E11 and 5-8D/1. (A) Reagent 2E11, positive signal. (B) Reagent 5-8D/1, positive signal. (C) Negative control. Original magnification, ×400.

Ag-CCA testing under clinical (field) conditions. Based on preliminary data ascertained from the testing of laboratory-adapted strains, testing was extended with clone 2E11 to clinical specimens collected during the peak period of the 1999 summer enterovirus season.