RAPID DETECTION OF RESPIRATORY VIRUSES AND ENTEROVIRUSES USING MIXTURES OF CONTINUOUS CELLS

Yung T. Huang, Ph.D.

Department of Pathology, University Hospitals of Cleveland, Case Western Reserve University, Cleveland, Ohio

Identifying a continuous cell line(s) that could greatly reduce or eliminate reliance on Primary Monkey Kidney (PMK) cells is of significant interest because of the numerous problems associated with preparations of primary monkey kidney cells. Two candidate cell lines that have been studied include NCI-H292, which has been suggested as a sensitive cell line for propagating a wide range of viruses including all of the respiratory viruses except influenza viruses (1,2) and Madin-Darby Canine Kidney (MDCK), which has been used for isolation of influenza viruses (3).

This article reports the first use of Mink Lung cells (Mv1Lu), a continuous cell line, for rapid Influenza Virus detection. Mink Lung cells produced more fluorescent cells and were found to be more sensitive than PMK or MDCK cells when inoculated specimen was enhanced by centrifugation and virus detected 16-24 hrs post-infection using a commercial antibody kit.

Novel mixtures of continuous cell lines (Mixed FreshCells^TM; Diagnostic Hybrids, Inc., Athens, OH) are also delineated which effectively broaden the susceptibility profile and virus detection coverage contained within a single cell culture unit, thus reducing or eliminating the need for multiple cultures. Mv1Lu/NCI-H292 cell mixtures provided rapid detection of several respiratory viruses in a single container, and BGMK/NCI-H292 cell mixtures provided rapid detection of several enteroviruses, also in a single cell culture unit.

One hundred and thirteen (113) clinical specimens submitted to our clinical laboratory for direct smear testing were included in this study. Specimens were processed by centrifugation at 700xg for 10 min to obtain the cellular material for preparing a direct smear. Two-tenths ml (0.2ml) of specimen supernatant was inoculated into a single shell vial each of Mv1Lu, pRhMK and Mv1Lu/NCI-H292 cell mixture, centrifuged at 700xg for 45 min, incubated overnight at 36%C, and stained with anti-influenza A virus monoclonal antibody (MAb) from Chemicon (Chemicon Inc., Temecula, CA).

Table 1 shows influenza A virus detection on two different cell lines and Mixed FreshCells in comparison to the direct smear (IFA) result. Mv1Lu detected 56/59 direct smear positive specimens while pRhMK detected only 44/59; the Mv1Lu/NCI-H292 cell mixture detected 55 specimens, thus missing one low titer specimen (low number of positive cells in Mv1Lu shell vial). All influenza A isolates were subtype H3N2. Mv1Lu and the Mv1Lu/NCI-H292 cell mixture identified 7/54 specimens positive for influenza A virus that were identified by direct smear as being negative, while pRhMK found 1/54. Additionally, the number of positive fluorescing cells in Mv1Lu and Mv1Lu/NCI-H292 shell vials was much greater than in pRhMK shell vials in virtually every specimen tested.

Trypsin has been used to enhance the yield of influenza A virus from clinical specimens. To compare the effect of trypsin, 24 frozen specimens were evaluated in the following three cell preparations in shell vials with or without 2 g/ml of trypsin in the culture medium (0% FBS) during incubation: Mv1Lu/NCI-H292 Mixed FreshCells, pRhMK, and MDCK. After overnight incubation, the cells were fixed, stained with Mab (IFA) and the number of positive cells in each vial were counted.

Table 2 shows the positive cell counts for each specimen using three cell preparations with and without trypsin. Positive stained cells were identified in 24/24, and 23/24, of the Mv1Lu/NCI-H292 Mixed FreshCells, with and without trypsin, respectively. Specimen #7893 demonstrated a single positive cell in the trypsin-treated shell vial only, which is difficult to judge as significant. In all other specimens, there did not appear to be a significant detection advantage as a result of adding trypsin to the culture medium. pRhMK and MDCK cell preparations were substantially less sensitive than the Mv1Lu/NCI-H292 cell mixture, regardless of whether trypsin was added or not. pRhMK cells identified influenza A-positive cells in 12/24 and 13/24, with and without trypsin, while MDCK cells found influenza A-positive cells in 11/24 and 10/24, with and without trypsin, respectively.

These results indicate that MvlLu cells are more sensitive for rapid detection of influenza A virus from clinical specimens, that mixing the Mv1Lu cells with NCI-H292 cells does not appreciably reduce the sensitivity of the Mv1Lu cells for influenza A detection, and that addition of trypsin to the culture medium post-infection does not appear to enhance the detection of influenza A virus in an overnight incubation.

Since in the 1997-1998 season the influenza A isolates were all subtype H3N2, H1N1 from an earlier isolate was titrated in Mv1Lu, Mv1Lu/NCI-H292, MDCK and pRhMK cells to determine the performance of Mv1Lu cells (alone or in the mixed cells) in comparison to MDCK and pRhMK. Similar to H3N2 virus, Mv1Lu (alone or in the mixed cells) is also more sensitive for H1N1 virus compared to MDCK and pRhMK cells (data not shown).

Rapid detection of four strains of influenza B virus obtained from ATCC was also examined using Mv1Lu, pRhMK and MDCK cells. Stock virus was serially diluted ten-fold to low titer and the sensitivity compared after overnight incubation and staining with anti-influenza B virus MAb (IFA), from Chemicon. The results indicated that Mv1Lu cells are equivalent or better than pRhMK or MDCK cells under these conditions (data not shown).

Mixed FreshCells contain at least two cell lines in a single unit, each cell line possessing its unique susceptibility profile to various viruses. Thus, the cell mixture Mv1Lu/NCI-H292 was studied to determine if two other common respiratory viruses that can be identified in NCI-H292, i.e. adenoviruses and respiratory syncytial virus (RSV), could be detected as well in Mixed FreshCells.

Nine specimens that were adenovirus-positive upon direct smear were inoculated into a shell vial each of Mv1Lu/NCI-H292, NCI-H292 and A-549 cells. After centrifugation, the cells were incubated for 48 hrs before being stained with anti-adenovirus MAb. All three cell systems detected the same seven specimens and missed the same two specimens. A-549 cells showed more positive cells than the other two cell systems, a finding consistent with a previous report that showed A-549 is a sensitive cell line for rapid detection of adenoviruses (4). However, our results demonstrate that detection of adenovirus in NCI-H292 cells is similar regardless of whether the NCI-H292 cells are mixed with Mv1Lu cells or provided alone.

Twenty-eight frozen samples that were found to be RSV positive by direct enzyme-immunoassay (EIA; Abbott TestPack), were inoculated into duplicate shell vials each of Mv1Lu/NCI-H292, NCI-H292, and HEp-2 cells. After inoculation, the shell vials were centrifuged and incubated at 36^oC for 48 hrs prior to staining with anti-RSV MAb (IFA). NCI-H292, Mv1Lu/NCI-H292 and HEp-2 cells detected 21, 19 and 14 positive specimens, respectively. These results are consistent with previous data (5) indicating that NCI-H292 is more sensitive than the HEp-2 cell line for detection of RSV. As in the case above with adenovirus, this data indicates that detection of RSV in NCI-H292 cells is similar regardless of whether the NCI-H292 cells are mixed with Mv1Lu cells or provided alone.

We are currently studying additional cell combinations, including mixing three cell types, to further optimize the use of Mixed FreshCells in diagnostic virology applications.

For isolation of enteroviruses from clinical specimens, cultures of PMK, MRC-5, BGMK and RD cells are used in the majority of laboratories. BGMK cells have been reported to be the most sensitive cells for coxsackie viruses (6, 7, 8) and NCI-H292 cells have been shown to be broadly sensitive to enteroviruses (2). A pilot study was done comparing inoculation of specimen into one shell vial each of BGMK and NCI-H292 cells (shell vial method) to one culture tube each of MRC-5 and PMK cells (tube culture method). The shell vials were centrifuged for

45 min at 700xg and incubated for 72 hrs; the tube cultures were observed daily for CPE up to 2 weeks post-inoculation. Shell vials were fixed and stained with entero–pool from Dako (Dako Corporation, Carpinteria, CA). Cells from CPE-positive tube cultures were scraped, spotted on slides and fixed and stained with entero-panel, echo-pool and Coxsackie-pool from Chemicon (Chemicon Inc., Temecula, CA).

Twenty-eight (28) frozen positive specimens were identified that included 5 coxsackieviruses, 11 echoviruses and 12 untyped enteroviruses. Twenty-one specimens (21/28) were positive by both shell vial and tube methods, 4 (4/28) were positive by shell vial only, and 2 (2/28) were positive by tube culture only. The specimens positive by tube culture method only were positive at days 10 and 14 post-inoculation. These results suggested that combined NCI-H292 and BGMK cells, using the mixed cells in a shell vial format, could replace MRC-5 and PMK for enteroviruses detection (9).

NCI-H292/BGMK Mixed FreshCells were evaluated using 26 frozen, positive specimens that were provided for study by the Diagnostic Virology Laboratory, MetroHealth Medical Center, (Cleveland, Ohio). Because of limited specimen volume, specimens were variously diluted to make sufficient volume for equivalent inoculation to each cell system. Detection of enterovirus in one shell vial of the cell mixture NCI-H292 /BGMK was compared to detection in single vials of NCI-H292, BGMK, MRC-5 and PMK cells. All shell vials received 0.1ml of specimen, were centrifuged and incubated for 3 days at 36^oC prior to staining with pooled enterovirus antibody (Dako). Table 3 shows that the NCI-H292/BGMK cell mixture identified the most positive specimens, i.e. 19/26. These specimens were all echoviruses, thus explaining why BGMK cells detected only 5 positive specimens. These results strongly suggest that two cell lines mixed in a single shell vial can replace two cell lines in individual shell vials. Additionally, this data suggests that NCI-H292/BGMK Mixed FreshCells can effectively replace PMK cells for rapid enterovirus detection.

Combining cell lines has the advantage of broader virus susceptibility in a single unit. Choosing specific cell mixtures enables the user to target detection of viruses associated with certain specimen sources, e.g. respiratory viruses or enteroviruses. Mixed FreshCells makes rapid detection in shell vials and multi-well plates more cost-effective and significantly reduces the reliance on PMK cells in the diagnostic virology laboratory.

1. Castells E, George VG, and Hierholzer JC. NCI-H292 as an alternative cell line for the isolation and propagation of the human paramyxoviruses. Arch Virol 115:277-288, 1990.

2. Hierholzer JC, Castells E, Banks GG, Bryan JA and McEwen CT. Sensitivity of NCI-H292 human lung mucoepidermoid cells for respiratory and other human viruses. J Clin Microbiol 31:1504-1510, 1993.

3. Brumback BG and Wade CD. Simultaneous rapid culture for four respiratory viruses in the same cell monolayer using a differential multicolored fluorescent confirmatory stain. J Clin Microbiol 34:798-801, 1996.

4. Mahafzah AM and Landry ML. Evaluation of immunofluorescent reagents, centrifugation, and conventional cultures for the diagnosis of adenovirus infection. Diagn Microbiol Infect Dis 12:407-411, 1989.

5. Verano L, Patino C, Dalal S, Desai M and Michalski FJ. "Comparison of MRC-5, PMK, and H292 cell culture for shell vial detection of respiratory syncytial virus (RSV)." Poster presented at the 12th Annual Clinical Virology Symposium and Annual Meeting of the Pan American Society for Clinical Virology, Clearwater, Florida, April 28- May 1, 1996.

6. Menegus MA and Hollick GE. Increased efficiency of group B coxsackievirus isolation from clinical specimens by use of BGM cells. J Clin Microbiol 15:945-948, 1982.

7. Dagan R and Menegus MA. A combination of four cell types for rapid detection of enteroviruses in clinical specimens. J Med Virol 19:219-228, 1986.

8. Chonmaitree T, Ford C, Sanders C and Lucia HL. Comparison of cell cultures for rapid isolation of enteroviruses. J Clin Microbiol 26:2576-2580, 1988.

9. Klespies SL, Cebula DE, Kelley CL, Galehouse D and Maurer CC. Detection of enteroviruses from clinical specimens by spin amplification shell vial culture and monoclonal antibody assay. J Clin Microbiol 34:1465- 1467, 1996.