Engineered BGMK Cells for Sensitive and Rapid Detection of Enteroviruses

Abstract

Decay-accelerating factor (DAF) has been reported to be a cellular receptor for several enteroviruses. Buffalo green monkey kidney (BGMK) cells expressing human DAF (BGMK-hDAF cells) showed increased susceptibility and sensitivity to several types of enteroviruses compared to wild-type BGMK cells. When 17 frozen positive clinical samples were tested, BGMK cells detected 8 and BGMK-hDAF cells detected 16. Since the CaCo-2 cell line has been documented to support the replication of most enteroviruses, CaCo-2 cells were mixed with BGMK-hDAF cells in order to increase the number of viruses detected. Thirty-four frozen clinical samples that previously had tested positive for enteroviruses were tested, and the following numbers were detected: 33 of 34 by CaCo-2/BGMK-hDAF cells, 29 of 34 by CaCo-2/BGMK cells, 28 of 34 by H292/RD (E-mix A) and A-549/BGMK (E-mix B) cells, and 26 of 34 by MRC-5 and pRhMK cells.

Detection and isolation of enteroviruses from clinical specimens require a number of different cell lines that may include primary monkey kidney (pMK) cells, human rhabdomyosarcoma (RD) cells, human epidermoid carcinoma (A-549) cells, Buffalo green monkey kidney (BGMK) cells, human embryonic lung (HEL) cells (WI-38 or MRC-5), and human embryonic kidney (HEK) cells. Despite the use of combinations of the above cells, it takes an average of 5 days before cytopathic effect (CPE) is detectable in tube culture. By use of the shell vial method (7, 8, 16), detection time can be reduced to about 3 days. In order to improve patient management, more-rapid cell culture procedures are required.

Decay-accelerating factor (DAF; also known as CD55) is a 70-kDa glycosylphosphatidylinositol-anchored glycoprotein involved in the regulation of complement activation and in cell signaling (9). The echoviruses that hemagglutinate human erythrocytes, including serotypes 3, 6, 7, 11, 12, 13, 19, 21, 24, 25, 29, 30, and 33, bind to DAF receptor in human cells, and the binding can be blocked by an anti-DAF monoclonal antibody (MAb) (1, 11). In addition, coxsackievirus A21 also binds to DAF of HEp-2 and HeLa cells, and the binding of coxsackieviruses B1, B3, and B5 to HeLa cells can be blocked by an anti-DAF MAb (2, 14, 15). An anti-DAF MAb has also been shown to inhibit enterovirus 70 binding to HeLa cells and to protect them against enterovirus 70 infection (6). In summary, DAF is involved in the entry of quite a number of enteroviruses into host cells.

BGMK cells have been shown to be quite sensitive for detection of coxsackie B viruses but show relatively poor sensitivity for detection of echoviruses (3, 5, 10). Since human DAF (hDAF) is a cellular receptor for a number of enteroviruses, especially the echoviruses, it follows that the expression of hDAF in BGMK cells may expand the host range and enhance their sensitivity in the detection of enteroviruses.

CaCo-2, a human colon adenocarcinoma cell line, has been demonstrated to support the replication of, and show visible CPE for, most enteroviruses (12). On the basis of this report and the evidence that only 13 types of echoviruses use hDAF as a receptor, a mixed monolayer of BGMK-hDAF and CaCo-2 cells might provide a broader range of enterovirus detection in a shorter time than other single- or multi-cell-line monolayers.

MATERIALS AND METHODS

Cells.

BGMK and BGMK-hDAF cells were prepared in-house and seeded in 48-well plates. Primary rhesus monkey kidney (pRhMK) cells and MRC-5 cells were obtained from ViroMed Inc. (Minneapolis, Minn.) in 48-well plates. E-Mix A (RD/H292), E-Mix B (BGMK/A-549), CaCo-2/BGMK, and CaCo-2/BGMK-hDAF mixed cells were also prepared in 48-well plates and supplied by Diagnostic Hybrids Inc., Athens, Ohio. All cells were used within a week after receiving or preparation. Each specimen or virus dilution was inoculated into duplicate wells with 0.2 ml of inoculum. Inoculated plates were centrifuged at 700 × g for 45 min before being incubated at 37°C in a humidified, 5% CO₂ atmosphere.

Transfection of hDAF in BGMK cells.

The hDAF gene cloned into the pcDNA3 vector (Invitrogen, Inc., Carlsbad, Calif.) was kindly provided by Edward Medof of the Department of Pathology, Case Western Reserve University, Cleveland, Ohio. The plasmid DNA was used to transform Escherichia coli JM109, and colonies with the insert were selected as described previously (13). DNA from hDAF-containing plasmids was prepared and used to transfect BGMK cells.

Subconfluent BGMK cells seeded in 12-well plates were washed with Opti-MEM (Gibco-BRL) without antibiotics or serum. DNA (1.5 μg) was diluted to 150 μl with Opti-MEM, and 10 μl of Superfect (Qiagen Inc., Valencia, Calif.) was separately diluted to 150 μl with Opti-MEM. Diluted DNA and Superfect were combined and added to BGMK cells according to the manufacturer's recommendations (Qiagen News, no. 2, p. 4, 1997). Transfected cells were incubated for another 2 days, and Geneticin (G418) was added to select for stable transfectants according to the reported procedure (13). In order to select and obtain cells expressing the highest level of hDAF, cells were sorted by fluorescence-activated cell sorting (4).

Viruses.

Coxsackieviruses A9, B1, B2, B4, and B5 and echovirus types 4, 7, 9, 21, 25,and 30 were from clinical samples, propagated in our laboratory and typed by the virology laboratory of the Ohio State Department of Health using a neutralization test.

Clinical specimens.

The clinical specimens used in this study, which originally tested positive by different cell culture systems, were frozen. The vast majority were rectal or stool specimens; in addition, there were four nasopharyngeal swabs, four throat swabs, and one eye swab. Some of these samples were shipped in dry ice from University Hospitals of North Carolina, Chapel Hill, N.C., and Children's Hospital of Omaha, Omaha, Nebr., and kept in a −70°C freezer prior to use for inoculation.

Immunofluorescence assay.

To monitor hDAF expression in BGMK cells, these cells were stained with an anti-DAF MAb (PharMingen International Inc., San Diego, Calif.) as the primary antibody and fluorescein-labeled goat anti-mouse immunoglobulin G as the secondary antibody (Chemicon International Inc., Temecula, Calif.).

To confirm and identify the enterovirus isolates, blends of MAbs against panenteroviruses, Coxsackie B viruses, echoviruses, enteroviruses, and polioviruses (all from Chemicon International Inc.), as well as an enterovirus MAb (Dako Co., Carpinteria, Calif.), were used according to the manufacturer's instructions to stain infected cells prepared in the multiwell slide and were followed with a fluorescein-labeled goat anti-mouse immunoglobulin G antibody (Chemicon) as the secondary antibody. Slides were examined with a UV microscope.

RESULTS

Expression of hDAF in BGMK cells.

Transfected BGMK-hDAF cells selected by cell sorting were cultured and subsequently cloned by limiting dilution. Two clones were selected, and one was used throughout the study. hDAF was uniformly expressed to high levels in these transfected BGMK cells (Fig. 1). The morphologies and growth rates of the wild-type and transfected BGMK cells were indistinguishable. However, about 5 days postconfluence, the overgrown cells on top of the monolayer were slightly more abundant in transfected than in wild-type BGMK cells. However, the monolayer of transfected BGMK cells mixed with CaCo-2 cells did not display this overgrowth.

FIG. 1.

Immunofluorescent staining of hDAF protein on BGMK cells transfected with hDAF (A) or mock transfected (B).

Comparison of BGMK cells to BGMK-hDAF cells for detection of echoviruses.

Initially, several echoviruses of types 4, 7, 9, 21, 25, and 30 were diluted from 10⁻¹ to 10⁻⁷ and inoculated in duplicate onto single monolayers of BGMK and BGMK-hDAF cells in 48-well plates. Monolayers were examined for CPE daily for 5 days. Of the echoviruses tested, types 7, 21, 25, and 30 have been reported to use hDAF as a receptor. As shown in Fig. 2A, no CPE was detected in BGMK cells with type 21, even at a dilution as low as 10⁻¹ after 5 days of incubation, while CPE was detected in BGMK-hDAF cells at dilutions as high as 10⁻⁵ at day 1. Types 7, 25, and 30 enter into and replicate in both BGMK and BGMK-hDAF cells, but CPE was detected in BGMK-hDAF cells at dilutions 1 to 2 log units higher at day 1 and at dilutions 1 1/2 to 2 1/2 log units higher at day 5. BGMK-hDAF cells detected types 4 and 9 at a dilution 1 log unit higher at day 1, but the end titer detected was the same. Monkey DAF or another cellular receptor(s) in BGMK cells may be used by types 4, 7, 9, 25, and 30; however, hDAF facilitates more-efficient entry, and only hDAF permits entry of type 21.

FIG. 2.

Dilutions of virus detected (CPE) on days 1 and 5 in BGMK (B) and BGMK-hDAF (B-D) cells. (A) Different types of echoviruses; (B) different types of coxsackieviruses.

Comparison of BGMK to BGMK-hDAF cells for detection of coxsackieviruses.

Coxsackieviruses A9, B1, B2, B4, and B5 were studied as described above for echoviruses. B1 and B5 have been reported to utilize hDAF as a receptor. As shown in Fig. 2B, for these coxsackieviruses the final dilutions of virus detected by BGMK and BGMK-hDAF cells were about the same. However, at day 1, BGMK-hDAF cells detected higher dilutions of viruses than BGMK cells, i.e., 2.5 log units higher forA9, 2 log units higher for B1, 1.5 log units higher for B5, and 1 log unit higher for B2, while B4 showed no difference. Thus, high-level expression of hDAF in BGMK cells also enhances the entry of coxsackieviruses, which leads to earlier detection of low levels of these viruses, as was shown for echoviruses.

Comparison of BGMK to BGMK-hDAF for detection of enteroviruses from frozen clinical specimens.

The above results prompted us to evaluate clinical specimens. A total of 17 frozen specimens were inoculated. These samples originally tested positive by different cell culture systems. Due to the small volumes of the original specimens, all samples were diluted 1 to 20 in order to generate large enough volumes for inoculation. Inoculated monolayers were examined for CPE daily for 5 days. All 17 enteroviruses from the frozen specimens were isolated by one or more cells and confirmed by a blend of MAbs against panenteroviruses (Chemicon) and an enterovirus MAb (Dako). Viruses were typed with a group-typing reagent (Chemicon). The staining characteristics of BGMK-hDAF and BGMK cells were indistinguishable. Group type staining yielded the following results: 10 echoviruses, 3 polioviruses, 1 coxsackie B virus, 1 echovirus or coxsackievirus, and 2 untyped enteroviruses. BGMK cells detected 8 enterovirus-positive samples, while BGMK-hDAF cells detected 16 enterovirus-positive samples. BGMK cells missed one poliovirus and eight echovirus samples. BGMK-hDAF cells missed one untyped enterovirus sample which was detected by BGMK cells at day 5, suggesting that the specimen had a very low titer of virus. But the inability of this untyped enterovirus to grow in BGMK-hDAF cells cannot be ruled out. These results are consistent with reports (3, 5, 10) that BGMK cells have relatively poor sensitivity for detection of echoviruses. However, with the expression of hDAF, sensitivity was markedly enhanced for these 10 echovirus-containing samples.

Comparison of BGMK-hDAF cells mixed with CaCo-2 cells to other cell culture systems for detection of enteroviruses.

CaCo-2 cells have been shown to be quite sensitive to echovirus types 2, 4, 6, 11, 14, and 27 (data not shown). In view of the fact that hDAF has been reported to be the receptor for only 13 types of echoviruses, a mixed monolayer of BGMK-hDAF and CaCo-2 cells might increase the number of enteroviruses detected. BGMK-hDAF and CaCo-2 cells were found to coexist well in this mixed-cell monolayer, with neither overgrowing or inhibiting the other, and were able to maintain the desired ratio. Virus-induced CPE was easily detected in this mixed monolayer of cells (Fig. 3). This new mixture of CaCo-2/BGMK-hDAF cells was compared to E-mix A (RD/H292), E-mix B (BGMK/A-549), CaCo-2/BGMK, MRC-5, and pRhMK cells for detection of enteroviruses.

FIG. 3.

CaCo-2/BGMK-hDAF mixed cells infected with echovirus. (A) Uninfected monolayer; (B) early CPE; (C) moderate CPE; (D) extensive CPE.

Thirty-four frozen positive clinical samples were all brought to a final volume of 2.5 ml for inoculation. All 34 enteroviruses from the frozen specimens were isolated by one or more of the cell systems and confirmed and typed as described above. Virus types were as follows: 14 echoviruses, 13 coxsackie B viruses, 3 polioviruses, 1 enterovirus, and 3 untyped enteroviruses. The number of specimens identified by CPE in different cells from day 1 to 5 is presented in Table 1. No more positives were detected after day 5. One poliovirus specimen appeared only in pRhMK cells and not in other cells. The supernatant from the infected pRhMK cells was inoculated onto all the other cell monolayers, and all propagated the virus very well, suggesting a low-titer specimen rather than a susceptibility problem with the other cells. In summary, CaCo-2/BGMK-hDAF cells missed one poliovirus. CaCo-2/BGMK cells missed one coxsackie B virus, three echoviruses, and one poliovirus. E-mix A and E-mix B together missed one echovirus, three coxsackie B viruses, and one poliovirus. MRC-5 and pRhMK together missed three echoviruses, four coxsackie B viruses, and one poliovirus. As shown in Fig. 4, the times of appearance of CPE are different in these different cell systems. CaCo-2/BGMK-hDAF cells not only detected more positives, but also detected them earlier, than all the other cells. The early identification of CPE in these diluted specimens suggests the high sensitivity of this new mixture of cells.

TABLE 1.

Comparison of different cells for detection of enterovirus from clinical specimens

Specimen	Virusa identified	Detectionb by the following cellsc:
		E-Mix A + B	Ca/BG-D	Ca/BG	MRC-5 + pRhMK
1	Echo	+	+	+	+
2	Cox B	+	+	−	−
3	Echo	+	+	+	−
4	Cox B	+	+	+	+
5	Cox B	−	+	+	−
6	Polio	−	−	−	+
7	Cox B	+	+	+	+
8	Echo	+	+	+	+
9	Echo	−	+	−	−
10	Polio	+	1	+	−
11	Echo	+	+	+	+
12	Echo	+	+	+	+
13	Echo	+	+	+	+
14	Echo	+	+	+	+
15	Cox B	+	+	+	+
16	Polio	+	+	+	+
17	Echo	+	+	−	+
18	Echo	−	+	+	+
19	Untyped	+	+	+	+
20	Cox B	+	+	+	+
21	Entero	+	+	+	+
22	Untyped	+	+	+	+
23	Cox B	+	+	+	+
24	Untyped	+	+	+	+
25	Cox B	+	+	+	+
26	Cox B	+	+	+	−
27	Cox B	+	+	+	+
28	Cox B	+	+	+	+
29	Echo	+	+	+	+
30	Cox B	−	+	+	−
31	Echo	+	+	+	+
32	Echo	+	+	−	−
33	Cox B	−	+	+	+
34	Echo	+	+	+	+
Total		28	33	29	26

^aEcho, echovirus; Cox, coxsacKievirus; Polio, poliovirus; Entero, enterovirus.

^b+, detected; −, not detected.

^cE-MixA, RD/H292; E-MixB, BGMK/A549; Ca/BG-D, CaCo-2/BGMK-hDAF; Ca/BG, CaCo-2/BGMK.

FIG. 4.

Cumulative numbers and percentages of positive specimens detected (CPE) by different cells on days 1, 2, 3, and 6. A total of 34 frozen clinical specimens were tested.

DISCUSSION

This report discussed the introduction of a cellular receptor into a cell line to enhance its sensitivity for clinical virus isolation. BGMK is the most sensitive cell line for coxsackie B virus isolation (3, 5, 10) but not for echovirus isolation. With the expression of hDAF, more echoviruses can efficiently use the receptor to gain entry into BGMK cells.

Interestingly, expression of hDAF in BGMK cells appears to mediate rapid viral entry, which leads to early detection of low titers of virus, including viruses that are not known to use hDAF as a receptor (Fig. 2 and 4).

The initial study of 17 frozen clinical samples included 1 coxsackievirus, 2 untyped enteroviruses, 3 polioviruses, 1 echovirus or coxsackievirus, and 10 echoviruses. As expected, expression of hDAF greatly enhanced the detection of echoviruses. However, the number of different types of echovirus in these 10 samples is unknown. It is possible that these echoviruses all use hDAF as a receptor.

The data presented in Table 1 were obtained with 34 frozen clinical samples, including 14 echoviruses. However, the number of different types of echovirus in these 14 samples is also unknown. It is unclear how many of the various enteroviruses can be detected in CaCo-2/BGMK-hDAF mixed cells with reasonable sensitivity.

Hopefully, these promising preliminary results will lead to large-scale prospective studies to evaluate the usefulness of this new mixture of BGMK-hDAF and CaCo-2 cells for the detection of enteroviruses.