Enzyme-Linked Immunosorbent Assay-Based ... - Europe PMC

1 downloads 0 Views 347KB Size Report
Apr 4, 1996 - Greene Publishing Associates and John Wiley & Sons, New York. ... Fahey, K. J., A. J. Chapman, I. G. Macreadie, P. R. Vaughan, N. M. McKern,.
CLINICAL AND DIAGNOSTIC LABORATORY IMMUNOLOGY, July 1996, p. 456–463 1071-412X/96/$04.0010 Copyright q 1996, American Society for Microbiology

Vol. 3, No. 4

Enzyme-Linked Immunosorbent Assay-Based Detection of Antibodies to Antigenic Subtypes of Infectious Bursal Disease Viruses of Chickens† D. J. JACKWOOD,* K. S. HENDERSON,‡

AND

R. J. JACKWOOD

Food Animal Health Research Program, Department of Veterinary Preventive Medicine, Ohio Agricultural Research and Development Center, Ohio State University, Wooster, Ohio Received 16 February 1996/Returned for modification 4 April 1996/Accepted 24 April 1996

An enzyme-linked immunosorbent assay (ELISA) using a fragment of the infectious bursal disease virus (IBDV) VP2 gene expressed in baculovirus was developed. A 944-bp portion of the VP2 gene from the Del-A strain of IBDV was ligated into the pAc360 transfer vector and transfected into baculovirus. Recombinant baculoviruses were identified by dot blot hybridization. The recombinant baculovirus 9A5 expressed a 57-kDa VP2 fusion protein, which was immunoprecipitated. This baculovirus-expressed VP2 was used as an antigen in an ELISA (Ohio State University [OSU]-ELISA). Titers of sera from specific-pathogen-free chickens infected with different strains of IBDV in our laboratory were determined by the OSU-ELISA, commercial ELISAs, and a virus neutralization assay. The results indicate that all sera from the specific-pathogen-free chickens infected with IBDV strains contained high titers of neutralizing antibodies. Each of these antisera also tested positive with the commercial ELISA kits. The OSU-ELISA did not detect antibodies to all strains of IBDV tested. This ELISA detected antibodies to the antigenically similar Delaware variant strains Del-A, Del-E, and GLS but did not detect or detected very poorly antibodies to antigenically heterologous classic IBDV strains STC, D78, and BVM. Although the antiserum to the IN strain of IBDV contained a virus-neutralizing antibody titer, the OSU-ELISA did not detect antibodies in this serum, suggesting IN is antigenically heterologous to Del-A. Titers to the IN strain were detected with the commercial ELISA kits. The OSU-ELISA detected antibodies to the SAL strain, suggesting this virus is antigenically similar to Del-A, which is supported by previously reported vaccination and challenge studies. In conclusion, the OSU-ELISA could be used to detect antibodies to a subgroup of IBDV strains, while commercially available ELISA kits detected antibodies to all the antigenic subtypes of IBDV tested. progeny correlated with antibody titers in the breeders at the time of lay. Two serotypes of IBDV exist, but only serotype 1 viruses cause disease in poultry (17, 23, 24). At least six antigenic subtypes of IBDV serotype 1 viruses have been identified by the in vitro cross-virus neutralization (VN) assay (19). Viruses in one of these antigenic subtypes are commonly known as variants, and viruses in the other antigenic subtypes are known as classic viruses. Variant IBDV strains are more recent isolates from commercial flocks of chickens that have high levels of maternal antibodies to classic IBDV strains (18, 36, 37, 40). The failure of maternal immunity and some vaccination programs to protect against IBD is probably due to these antigenic subtypes of serotype 1 (16, 19, 29, 40). Thayer et al. (48) reported a good correlation between serum antibody titers obtained with commercially available ELISA kits and by the VN assay for IBDV. Since the development of commercially available ELISA kits for IBDV antibodies, little improvement of this technology has been reported. These kits are widely used by diagnostic laboratories and poultry producers. Commercially available ELISA kits for antibodies to IBDV strains detect antibodies to both serotypes 1 and 2 in addition to all the known subtypes of serotype 1 viruses (15). Identification of antibodies to different antigenic subtypes of IBDV is currently possible only by the VN assay (19, 29). Thus, an immune response to antigenic variants of IBDV cannot be distinguished from an immune response to other antigenic types of IBDV or serotype 2 viruses with ELISA kits. ¨ ppling More than one immunogenic epitope exists on VP2. O

Infectious bursal disease virus (IBDV) is the etiologic agent of IBD, an immunosuppressive disease of young chickens. The immunosuppression results from a depletion of B lymphocytes (8). Secondary infections are commonly associated with IBD (10, 13, 38, 39, 51). The IBDV virion contains two segments of double-stranded RNA (31). Segment A encodes a polyprotein, which is cleaved into the structural proteins VP2, VP3, and VP4 (14). Segment B encodes the putative viral polymerase VP1 (30). A neutralizing monoclonal antibody binding site was mapped to a region of VP2 between amino acids 206 and 350 (2). An area of sequence variability exists among strains of IBDV within the VP2 region containing amino acids 239 to 332 (5). Detection of antibodies to IBDV has been accomplished by a number of different assays, but most often the enzyme-linked immunosorbent assay (ELISA) is used because it is economical and can quickly test large numbers of samples. Since Marquardt et al. (28) first described the ELISA for antibodies to IBDV, many improvements have been made to the assay (7, 44, 45, 48). A kinetics-based ELISA was described previously (46). This assay was designed to determine the rate at which maternal antibodies decrease in serum. Kinetic-ELISA results from

* Corresponding author. Phone: (216) 263-3964. Fax: (216) 2633677. † Ohio Agricultural Research and Development Center, manuscript no. 175-95. ‡ Present address: Charles River Laboratories, Wilmington, MA 01887. 456

VOL. 3, 1996

ELISA-BASED DETECTION OF ANTIBODIES TO IBDV SUBTYPES

et al. (33) reported that VP2 contains at least three independent epitopes. Using monoclonal antibodies, Whetzel and Jackwood (50) showed that differences in neutralizing epitopes on VP2 exist between variant and classic viruses. These results suggest that it may be possible to express IBDV epitopes that will bind only antibodies to specific subtypes of IBDV. Several expression systems have been utilized to produce IBDV proteins. Azad et al. (2) expressed VP2 in Escherichia coli. A recombinant fowlpox virus expressing VP2 protected birds challenged with IBDV against mortality but not against damage to bursas (4). A VP2 fusion protein produced in Saccharomyces cerevisiae induced neutralizing antibodies in specific-pathogen-free (SPF) chickens, and progeny chicks were passively protected from IBDV challenge (12). IBDV proteins have also been expressed by the baculovirus system (49). It was reported that these proteins reacted with IBDV-specific monoclonal antibodies and resembled the native viral proteins. Proteins expressed in baculovirus have been produced as fusion and nonfusion proteins in amounts of 1 to 500 mg/ml (26, 27, 32, 42). There are no reports of studies using these expression products as antigens in an ELISA. Ismail and Saif (16) demonstrated that vaccination with one serotype 1 subtype did not always protect chickens from challenge with another serotype 1 subtype. When a vaccine dose containing low virus titers was used, the results correlated with cross-VN assay data (19). However, cross-neutralization between variant and classic viruses was observed at higher vaccine doses in the studies of Ismail and Saif (16). Other studies have indicated a lack of correlation between in vitro cross-VN assay data and in vivo vaccination and challenge data (18, 35). Thus, the antigenic relationships between variant and classic IBDV strains are ambiguous. The goal of our study was to develop an ELISA-based technique that can be used to determine the antigenic relatedness of Del-A to other IBDV strains. As a first step to achieving this goal, a portion of VP2 from a Delaware variant virus (Del-A) was expressed in the baculovirus system. This expression product was used as an antigen in an ELISA for the detection of antibodies to IBDV strains that were reported to be antigenically related or unrelated to Del-A (18, 19). The ELISA developed using this antigen is described and was designated the Ohio State University (OSU)-ELISA. MATERIALS AND METHODS Viruses and antisera. Wild-type baculovirus (Autographa californica nuclear polyhedrosis virus) was propagated in Spodoptera frugiperda (Sf9) cells (Invitrogen, San Diego, Calif.). Cell culture-adapted IBDV strains were propagated on baby Grivet monkey kidney (BGM70) cells (20). The serotype 1 variant-subtype IBDV strains used included IN (18), MD (40), GLS (43), Del-E, and Del-A (37). These viruses were reported to be antigenically similar by the in vitro cross-VN assay (18, 19, 40). The serotype 1 classic-subtype IBDV strains included STC (39), D78 (19), BVM (19), and SAL (19). These viruses are antigenically distinct from the variant strains and also antigenically distinct from each other (19). A serotype 2 virus designated OH was used as a control (22). Antisera against these IBDV strains were prepared in SPF chickens (Hy-Vac, Gowrie, Iowa). Birds were inoculated via the oronasal route at 3 weeks of age with 0.05 ml of virus inoculum. Virus inoculum titers in BGM70 cells ranged from 104.5 to 106 50% tissue culture infective doses (TCID50)/ml. Antisera from blood samples collected at 21 days following inoculation were tested by the VN assay, the OSUELISA, and two commercial ELISAs. In a second experiment, antisera to the Del-A, Del-E, GLS, and D78 IBDV strains were prepared in 3-week-old SPF chickens. Five groups containing six SPF birds each were used. The IBDV strains were propagated in BGM70 cells, adjusted to a titer of 104.5 TCID50/ml, and then inactivated with 0.1% betapropiolactone (16). A 1:1 (vol/vol) oil-in-water emulsion was prepared as previously described (16) for each virus inoculum, and a 1.0-ml volume of the emulsions was inoculated via a subcutaneous route into each bird. Four groups of birds were given Del-A, Del-E, GLS, or D78, and the fifth group of control birds was inoculated with the oil-in-water emulsion with phosphate-buffered saline (PBS; pH 7.2) in place of virus. A second 1.0-ml inoculation was administered subcutaneously without the oil-in-water adjuvant, 49 days following the first

457

inoculation. Blood samples were collected from each bird on days 0, 7, 10, 14, 21, 28, 35, 43, 49, 56, 63, 71, 77, 85, 91, and 98. Antisera from blood samples were tested by the OSU-ELISA. VN assay. The VN assay was conducted as previously described for BGM70 cells grown in 96-well culture plates (Costar; Corning, Ithaca, N.Y.) (19). The serum samples were diluted twofold, beginning at a 1:100 dilution. A constant amount of virus (100 TCID50) was added to each test well. Insertion of the VP2 gene into the pAc360 transfer vector. The VP2 cDNA from IBDV Del-A was previously described (11, 50). It was 1,006 bp and contained bases 380 to 1386 of the VP2 gene (25). This fragment was ligated into pGem3Zf(1) (Promega Corporation, Madison, Wis.) (11). The resulting recombinant plasmid (pV-17) was introduced into E. coli and plated on Luria agar plates containing 50 mg of ampicillin (Sigma Chemical Co., St. Louis, Mo.) per ml (11). An isolated colony was selected and grown in terrific broth (41) containing 50 mg of ampicillin per ml. Plasmid DNA was isolated by an alkaline extraction procedure (41). Plasmid pV-17 was digested with BalI (Promega) and partially digested with PstI (Promega) (Fig. 1). The DNA fragments were separated by electrophoresis using 1% agarose (FMC Corporation, Rockland, Maine) in TBE buffer (89 mM Tris base [pH 8.0], 89 mM boric acid, 2 mM EDTA) (AMRESCO, Solon, Ohio). A 944-bp fragment was excised from the gel and collected into TBE buffer with an electro-eluter (Bio-Rad Laboratories, Richmond, Calif.). The 944-bp fragment was extracted with phenol and chloroform, precipitated with ethanol, and after centrifugation, suspended in double-distilled H2O. The protruding 39 PstI end was removed with T4 DNA polymerase (Stratagene, La Jolla, Calif.) (41). A BamHI 10-mer linker (Promega) was ligated onto each end with T4 DNA ligase (Bethesda Research Laboratories, Gaithersburg, Md. (41). The fragments were then digested to completion with BamHI (Promega). Removal of linker fragments and isolation of the 944-bp fragment containing BamHI sites on both ends was performed by electrophoresis with 1.25% SeaPlaque low-gelling-temperature agarose (FMC Corporation) in TAE (40 mM Tris-acetate [pH 8.0], 1 mM EDTA) (AMRESCO). The 944-bp fragment was excised from the gel and ligated into the transfer vector by an in-gel ligation procedure (1). Prior to ligation, the pAc360 vector was digested with BamHI (Promega) and treated with shrimp alkaline phosphatase (United States Biochemical Corporation, Cleveland, Ohio). The treated pAc360 DNA was extracted with phenol and chloroform, precipitated with ethanol, and resuspended in TE (10 mM Tris-HCl [pH 7.6], 1 mM EDTA). Ligation reaction products were introduced into E. coli Max Efficiency DH5a-competent cells (Bethesda Research Laboratories) according to the manufacturer’s instructions. E. coli was plated on Luria agar plates containing 50 mg of ampicillin (Sigma) per ml. Selected colonies were grown in terrific broth (41) with 50 mg of ampicillin per ml. Plasmid DNA from these colonies was isolated by a miniprep plasmid extraction procedure (41). Restriction enzymes BamHI, HindIII, PstI, and SalI were used to confirm the orientation of the VP2 fragment. Transfection and selection of recombinant virus. Transfection was conducted by the MAXBAC baculovirus expression system (Invitrogen). Recombinant plaques on Sf9 cells were identified visually and prepared for dot blot hybridization (47). A 612-bp PstI-BalI fragment within the 944-bp fragment was radiolabeled with [32P]dCTP (ICN Radiochemicals, Irvine, Calif.) by nick translation (Bethesda Research Laboratories). Dot blot hybridization using procedures previously described (21) was conducted. Positive baculoviruses which contained the VP2 gene fragment were plaque purified an additional two times. Following each plaque purification, hybridization was used to confirm the presence of the VP2 gene fragment. Recombinant baculoviruses were propagated on Sf9 cells for 72 h at 278C. Radioimmunoprecipitation. VP2 fusion protein was metabolically labeled with [35S]methionine as previously described (34). Cells infected with 9A5 were lysed with a double-detergent lysis buffer with protease inhibitors (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 0.02% sodium azide, 0.1% sodium dodecyl sulfate [SDS], 100 mg of phenylmethylsulfonyl fluoride per ml, 1 mg of pepstatin per ml, 0.5 mg of leupeptin per ml, 1% Nonidet P-40). A 5-ml volume of lysate was mixed with 5 ml of polyclonal MD antiserum, and the mixture was incubated at 48C for 3 h. Following incubation, the mixture was combined with 2 ml of polyclonal rabbit anti-chicken serum (Sigma Chemical Co.) and incubated at 48C for 1 h. This mixture was added to 100 ml of a 1:1 (vol/vol) mixture of protein ASepharose and NET buffer (50 mM Tris-HCl [pH 8.0], 0.14 M NaCl, 5 mM EDTA, 0.05% [vol/vol] Nonidet P-40) and shaken gently for 1.25 h at 48C. The mixture was washed three times with NET buffer and suspended in SDS loading buffer (34). Samples were analyzed by SDS–12.5% polyacrylamide gel electrophoresis (PAGE) (34). Gels were vacuum dried and then exposed to X-Omat AR film (Eastman Kodak Co., Rochester, N.Y.). OSU-ELISA. Antigen for coating ELISA plates was prepared by infecting High Five insect cells (Invitrogen) at a high multiplicity of infection (.5). The High Five cells were harvested 4 days following infection, and the infected cells were placed in PBS (1.9 mM NaH2PO4, 8.1 mM Na2HPO4, 154 mM NaCl [pH 7.2]) containing 0.05% (wt/vol) sodium azide. The cells were lysed by sonication, and following a 1:1,000 dilution in PBS-sodium azide, the antigen was used to coat 96-well flat-bottom plates (Falcon; Becton Dickinson, Lincoln Park, N.J.). This dilution of antigen was determined to be optimal using procedures previously described (9). A 50-ml volume per well was used to coat the plates for 24 h

458

JACKWOOD ET AL.

CLIN. DIAGN. LAB. IMMUNOL.

FIG. 1. Schematic diagram depicting the construction of transfer vector p360V17.

at room temperature. The antigen-coated 96-well plates were washed three times in distilled water and then incubated at room temperature for 30 min in blocking buffer (170 mM H3BO4 [pH 8.5], 120 mM NaCl, 1 mM EDTA, 0.05% [wt/vol] sodium azide, 0.25% [wt/vol] bovine serum albumin, 0.05% [vol/vol] Tween 20). After three washes in distilled water, a 50-ml volume of serum diluted in blocking buffer was added to each well. Incubation continued at room temperature for 30 min. The ELISA plates were then washed three times in distilled water, once for 10 min in blocking buffer, and then three times in distilled water. A peroxidaselabeled goat anti-chicken immunoglobulin G (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Md.) was diluted 1:100 in blocking buffer, and 50-ml was added to each well. Plates were incubated at room temperature for 30 min and then washed in distilled water, in blocking buffer, and again in distilled water as described above. A 75-ml volume of the ABTS substrate (Kirkegaard & Perry Laboratories, Inc.) was added, and after 15 min the color development was stopped with 5% (wt/vol) SDS in water. Test wells were read on an ELISA reader (Dynatech Laboratories, Alexandria, Va.) at a wavelength of 410 nm. Data analysis. The OSU-ELISA data are presented as the sample absorbance values minus the negative control absorbance values, and each datum point represents the mean of at least four separate assays. We selected a 0.2 absorbance as the cutoff for a positive assay; this value is four standard deviations above the absorbance values for negative control antisera at a 1:400 dilution. Statistically significant differences were determined by a one-way analysis of variance, and differences between samples were determined by Fisher’s least significant difference assay.

tein was not observed in the wild-type baculovirus. Furthermore, the polyhedron protein (32 kDa) was observed in the wild-type virus but not in the recombinant viruses. Polyclonal chicken antiserum against the variant IBDV strain MD was used to verify the authenticity of the 57-kDa protein in a radioimmunoprecipitation assay. Immunoprecipitated proteins were separated by SDS–12.5% PAGE and visualized by autoradiography. A 57-kDa protein band was observed (Fig. 3). This band size corresponded to that of the predicted fusion protein. No proteins were observed in the lanes containing protein samples from the wild-type baculovirus and mock-infected control cells. OSU-ELISA. Two experiments using the OSU-ELISA were conducted. The first was designed to determine if antibodies to

RESULTS Identification of the VP2 fusion protein. The 944-bp PstIBalI fragment contains bases 434 through 1377 of the VP2 gene, which includes the variable sequence region (5). The recombinant transfer vector p360v17 (Fig. 1) was identified and transfected into baculovirus DNA. Recombinant baculovirus plaques that were not refractive when illuminated by a fiber-optic light source (Dolan-Jenner Industries, Inc., Woburn, Mass.) were selected. Dot blot hybridization confirmed the presence of the 944-bp fragment in recombinant baculoviruses following each plaque purification. To determine if the recombinant baculoviruses were expressing the VP2 fusion protein, viral proteins were metabolically labeled with [35S]methionine after inoculation of Sf9 cells and analyzed by SDS–12% PAGE (34). A polypeptide with the expected size of the VP2 fusion protein (57 kDa) was observed in recombinant baculovirus 9A5 (Fig. 2). A similar-sized pro-

FIG. 2. Metabolically [35S]methionine-labeled viral and cellular proteins separated by SDS–12% PAGE. Lane 1, mock-infected Sf9 cells; lane 2, wild-type baculovirus; lanes 3 and 4, recombinant baculovirus. Locations of the wild-type polyhedron protein (32 kDa) and recombinant proteins (57 kDa) are indicated in the right margin. Molecular mass markers (in kilodaltons) are indicated in the left margin.

VOL. 3, 1996

ELISA-BASED DETECTION OF ANTIBODIES TO IBDV SUBTYPES

FIG. 3. Metabolically [35S]methionine-labeled viral and cellular proteins were immunoprecipitated with polyclonal antiserum to IBDV MD and separated by SDS–12% PAGE. Lane 1, mock-infected Sf9 cells; lane 2, wild-type baculovirus; lanes 3 and 4, recombinant baculovirus. The position of recombinant proteins (57 kDa) are indicated in the right margin. Molecular mass markers (in kilodaltons) are indicated in the left margin.

antigenically homologous (MD, IN, GLS, Del-A, and Del-E) and heterologous (STC, D78, BVM, and SAL) IBDV strains could be detected by the OSU-ELISA. The antigenic relatedness of these viruses was previously reported by using the cross-VN assay (19). In this experiment, the OSU-ELISA data were compared with antibody titers determined by the VN assay and titers obtained with commercial ELISA kits on antisera monospecific to the antigenically related and unrelated IBDV strains. The second experiment was designed to determine if OSU-ELISA titers reflected the development of an antibody response to IBDV in sera collected over a 14-week period following inoculation with the Del-A, Del-E, GLS, or D78 strains. Experiment 1. We prepared antibodies to four antigenically related classic viruses (STC, D78, BVM, and SAL), five antigenically related variant viruses (MD, IN, GLS, Del-A, and Del-E), and the OH serotype 2 virus in SPF chickens (HyVac). The classic viruses are reported to be antigenically heterologous to the variant viruses (19, 36). The resulting antisera were tested by the VN assay to ensure that seroconversion had occurred. Serum samples containing antibodies to each virus plus a negative control serum were tested by the VN assay with each of four different IBDV strains (D78, SAL, Del-A, and MD). The titers ranged from 1,600 to 51,200 when 100 TCID50 of a homologous virus (variant with variant and classic with classic) per well were used to test these antisera (Table 1). Virus-neutralizing antibodies were not detected in the negative control or serotype 2 OH antisera. These antisera were then tested by the OSU-ELISA. The OSU-ELISA did not detect antibodies to all viruses tested (Fig. 4). Titers of antisera to STC-2, BVM, and D78 at a 1:400 dilution were negative by the OSU-ELISA. These viruses are reportedly antigenically heterologous to the Del-A strain used to prepare the VP2 antigen used in the OSUELISA (19). The SAL virus antiserum (VN assay titer 5 51,000) reacted strongly in the OSU-ELISA. This virus was previously reported to be antigenically heterologous to the Del-A antigen (19). All antisera to the viruses antigenically related to Del-A (MD, GLS, Del-A, and Del-E) were positive by the OSU-ELISA with the exception of antiserum to IN, which was negative by the OSU-ELISA although it had a VN

459

assay titer of 4,800. The background observed in the OSUELISA was extremely low, and no antibodies were detected in the control or OH antisera. Antisera to the MD, GLS, Del-A, Del-E, and IN viruses were also tested with two commercial ELISA kits: KPL (Kirkegaard & Perry Laboratories, Inc.) and IDEXX (IDEXX Laboratories, Westbrook, Maine). The IDEXX kit detected antibodies to all the viruses tested at the recommended 1:500 dilution (Fig. 5). The KPL ELISA did not detect antibodies to IBDV Del-A and GLS at the recommended 1:100 dilution but did detect antibodies to Del-E, IN, and MD (Fig. 5). ELISA data generated by the OSU-ELISA, IDEXX ELISA, and KPL ELISA were compared statistically. Each serum was tested at the following dilutions: OSU-ELISA, 1:400; IDEXX ELISA, 1:500 (manufacturer’s recommendation); and KPL ELISA, 1:100 (manufacturer’s recommendation). The absorbance results for a given serum in each assay were compared for statistically significant differences (Table 2). Significant differences were not observed among the three ELISAs using control serum or MD antiserum. Differences were detected with variant IBDV antiserum to the IN, GLS, Del-A, and Del-E strains. The IN and Del-E absorbance results from the OSU-ELISA were significantly lower than the absorbance values observed in the IDEXX and KPL assays. The IDEXX assay gave a significantly higher absorbance than the KPL assay for the Del-E serum. The GLS serum absorbance was significantly higher in the OSU-ELISA than those of the commercial ELISA kits. Using the GLS serum, the KPL assay gave an absorbance which was not different from that of the control serum and the IDEXX assay result was only slightly above that of the negative control serum. The OSU-ELISA result with the Del-A serum was significantly higher than those of both commercial assays. The results obtained with the commercial ELISA kits and Del-A serum were not significantly different from those obtained with the control. Experiment 2. The OSU-ELISA titers were determined for antisera collected weekly from each bird over a 14-week period. The geometric mean titer was determined for each time point within a given group (Fig. 6). Antibodies were detected in antisera collected from birds in each group except the control group. Antibody titers to the three variant viruses Del-A, Del-E, and GLS appeared to follow typical primary and anamnestic humoral responses. Titers of antibodies to the D78 virus

TABLE 1. Geometric mean virus-neutralizing antibody titers for IBDV antisera determined by the VN assaya Titer of antibodies to antigen:

Antiserum to strain:

D78

SAL

Del-A

MD

Control MD IN GLS Del-A Del-E STC-1 STC-2 SAL BVM D78 OH

,100 23,360 730 1,460 600 4,800 3,728 1,600 19,136 466 1,600 ,100

,100 25,600 1,600 1,600 1,600 3,200 9,600 1,600 51,200 2,400 1,600 ,100

,100 19,200 1,600 2,400 2,400 6,400 4,800 400 12,800 3,200 800 ,100

,100 25,600 4,800 2,400 1,600 2,400 2,400 400 19,200 1,600 1,600 ,100

a The VN assay was conducted with BGM70 cells grown in 96-well microtiter plates. A titer was determined with 100 TCID50 of viral antigen per well and reported as the last serial twofold dilution without cytopathic effects in 50% of the wells.

460

JACKWOOD ET AL.

CLIN. DIAGN. LAB. IMMUNOL.

FIG. 4. OSU-ELISA results with chicken anti-IBDV serum to variant strains (A) and classic strains (B) of the virus. Serial twofold dilutions of each serum were tested. Samples were considered positive if their absorbance (ABS) was equal to or greater than 0.2, which is at least four standard deviations above that of the control. The titers of these antisera determined by the VN assay are shown in Table 1.

were significantly lower (P , 0.05) than the variant virus titers. Titers of antibodies to the D78 virus were detected in antisera with the IDEXX kit and by the VN assay (data not shown). Titers in the D78 antisera observed by the IDEXX ELISA were similar to those observed for Del-A and Del-E by the OSU-ELISA. VN assay titers in the D78 antisera were detected on days 35, 43, 63, 71, 77, and 85.

DISCUSSION The VP2 protein produced by IBDV plays an important role in the production of a protective immune response to IBD (2, 3, 4, 12). Immunologic studies involving IBDV have suggested that VP2 contains a conformationally dependent neutralizing epitope which could be used to distinguish serotypes (6). The ¨ ppling et al. (33) and Crisman et al. (11) suggest the results of O presence of additional conformationally dependent epitopes on VP2. If a different cassette of epitopes is present on each IBDV strain, antibody profiles to different IBDV strains may

be different. Our previous studies demonstrated that in vitro expression products from the VP2 cDNA reacted with polyclonal antisera to variant viruses but not to the classic serotype 1 virus STC in the radioimmunoprecipitation assay (11). These results indicate that this protein may contain a variant specific epitope which could be exploited in diagnostic assays, such as the ELISA. Our hypothesis was that epitopes specific for only some IBDV strains might be detected with a portion of the VP2 protein as antigen in an ELISA. In this study, a portion of the Del-A VP2 protein was expressed as a fusion protein and was used to develop the OSU-ELISA. The recombinant baculovirus 9A5 expressed a 57-kDa fusion protein. The VP2 gene fragment was 944 bp in length and did not include the initial 300 bp, which results in the absence of 100 amino acids located at the amino terminus of native VP2. Vakharia et al. (49) used baculovirus to express the entire VP2 protein in addition to the VP3 and VP4 proteins of IBDV and obtained an average virus-neutralizing antibody titer of

VOL. 3, 1996

ELISA-BASED DETECTION OF ANTIBODIES TO IBDV SUBTYPES

461

FIG. 5. Antisera to variant IBDV strains were tested with IDEXX (A) and KPL (B) ELISA kits. The six antisera tested in both assays were identical. Serial twofold dilutions of each serum were tested. The manufacturer-recommended serum dilutions for these assays are 1:500 (IDEXX) and 1:100 (KPL). Samples were considered positive if their absorbance (ABS) was equal to or greater than 0.2, which is at least four standard deviations above that of the control at the manufacturer-recommended serum dilution. The VN assay titers of these antisera are shown in Table 1.

1,024 in chickens. These proteins were not used in a diagnostic assay. The OSU-ELISA detected antibodies to the MD, GLS, Del-E, Del-A, and SAL IBDV strains but did not detect or

TABLE 2. Absorbance values for antisera to variant IBDV strains determined by the OSU-ELISA, KPL ELISA, and IDEXX ELISA Absorbance of antiserum by ELISAa:

Antiserum to strain:

OSU (1:400)

KPL (1:100)b

IDEXX (1:500)b

Control MD IN GLS Del-A Del-E

0.039 6 0.024 (a) 0.785 6 0.109 (a) 0.052 6 0.007 (a) 0.622 6 0.050 (a) 0.435 6 0.068 (a) 0.355 6 0.031 (a)

0.019 6 0.009 (a) 0.832 6 0.146 (a) 0.498 6 0.028 (b) 0.028 6 0.002 (b) 0.008 6 0.001 (b) 0.510 6 0.009 (b)

0.001 6 0.009 (a) 0.764 6 0.091 (a) 0.382 6 0.057 (b) 0.169 6 0.016 (c) 0.051 6 0.027 (b) 0.675 6 0.007 (c)

a Statistically significant differences were compared between assays for each serum sample. Each value is the positive absorbance minus the absorbance of the no-serum control 6 the standard deviation. Letters in parentheses in a given row if different from the preceding letter indicate values which are significantly different. b Sera were tested at the manufacturer’s recommended dilution.

detected poorly antibodies to the IN, STC, BVM, and D78 strains (Fig. 4). The results suggest that the OSU-ELISA detects antibodies to specific epitopes that were not present on some of the IBDV strains tested. It is possible that a different cassette of epitopes may exist on each of these viruses or that different epitopes are dominant in the IBDV strains tested. This result was also observed with the IDEXX and KPL ELISA kits but to a lesser degree. The results of both commercial ELISA kits and the OSUELISA demonstrate a statistically significant difference in the absorbance values for antisera to IN, Del-E, Del-A, and GLS (Table 2). The OSU-ELISA appeared to be more sensitive than the commercial ELISA kits in detecting antibodies to the Del-A and GLS viruses and less sensitive in detecting antibodies in the Del-E and IN antisera. These results did not correlate with the VN assay titers of these antisera. The lack of reaction in the OSU-ELISA by the IN antisera (VN assay titer 5 4,800) suggests that epitopes on this virus are unrelated to those of the Del-A variant virus. Previous cross-VN assay and vaccination and challenge data suggest that IN may be antigenically related to MD, Del-E, and SAL, but

462

JACKWOOD ET AL.

CLIN. DIAGN. LAB. IMMUNOL.

FIG. 6. OSU-ELISA titers for antibodies elicited by the IBDV strains Del-A, Del-E, D78, and GLS. Geometric mean titers (GMT) were calculated for six samples on each observation day. Antisera from control birds were negative throughout the experiment. A second vaccination of viral inoculum was administered at day 49 following the initial vaccination.

Del-A was not used in these studies (18). The positive titers obtained in the commercial ELISA kits for IN antisera support the hypothesis than IN is antigenically related to classic viruses, because these kits detect antisera to the STC, SAL, BVM, and D78 classic viruses better than the OSU-ELISA (data not shown). The OSU-ELISA titer observed in antisera to the SAL strain suggests this virus is antigenically homologous to the Del-A virus. Although this contradicts cross-VN assay data (19), vaccination and challenge studies indicated that the SAL strain may be antigenically related to the Delaware variant viruses (16). It is also possible that the high antibody titer of this serum contributed to the cross-reaction observed. Further studies are needed to determine if there is a link between the antigenic relationships observed in the OSU-ELISA and VN assay titers and protection from disease. The antisera generated in experiment 2 were tested by the OSU-ELISA to determine if the assay could detect a primary and anamnestic antibody response. Both responses were observed for antisera generated to the variant Del-A, Del-E, and GLS viruses. Minimal OSU-ELISA titers were observed in antisera from birds inoculated with the D78 strain. Results from the IDEXX ELISA and VN assay demonstrated that a primary and anamnestic antibody response occurred in antisera to the D78 virus. These results suggest the OSU-ELISA can detect antibodies to specific viruses and support the data observed in experiment 1. Additional studies are needed to determine if OSU-ELISA titers correlate with protection to specific IBDV strains. It has been suggested that titers to IBDV variant viruses detected with commercial ELISA kits do not adequately reflect the immune status of a chicken flock. Results of the current study suggest that fragments of the VP2 protein may be useful in identification of antibodies to antigenically related groups of IBDV. ACKNOWLEDGMENTS Salaries and research support were provided by state funds appropriated to the Ohio Agricultural Research and Development Center, Ohio State University. This investigation was supported in part by

grants from the Southeastern Poultry and Egg Association, project 175, Eli Lilly and Company, Lilly Research Laboratories, Greenfield, Ind., and IDEXX Laboratories, Inc., One IDEXX Dr., Westbrook, Maine. REFERENCES 1. Ausubel, F. M., B. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1992. Short protocols in molecular biology, 2nd ed. Greene Publishing Associates and John Wiley & Sons, New York. 2. Azad, A. A., M. N. Jagadish, and K. J. Fahey. 1987. Deletion mapping and expression in Escherichia coli of the large genomic segment of a birnavirus. Virology 161:145–152. 3. Azad, A. A., I. Macreadie, P. Vaughan, M. Jagadish, N. McKern, H. G. Heine, P. Failla, C. Ward, A. Chapman, and K. Fahey. 1990. Full protection against an immunodepressive viral disease by a recombinant antigen produced in yeast. Vaccines 90:59–62. 4. Bayliss, C. D., R. W. Peters, J. K. A. Cook, R. L. Reece, K. Howes, M. M. Binns, and M. E. G. Boursnell. 1991. A recombinant fowlpox virus that expresses the vp2 antigen of infectious bursal disease virus induces protection against mortality caused by the virus. Arch. Virol. 120:193–205. 5. Bayliss, C. D., K. Spies, R. W. Peters, A. Papageorgiou, H. Mu ¨ller, and M. E. G. Boursnell. 1990. A comparison of the sequences of segment A of four infectious bursal disease virus strains and identification of a variable region in VP2. J. Gen. Virol. 71:1303–1312. 6. Becht, H., and H. K. Mu ¨ller. 1988. Comparative studies on structural and antigenic properties of two serotypes of infectious bursal disease virus. J. Gen. Virol. 69:631–640. 7. Briggs, D. J., C. E. Whitfill, J. K. Skeeles, J. D. Story, and K. D. Reed. 1986. Application of the positive/negative ratio method of analysis to quantitate antibody responses to infectious bursal disease virus using a commercially available ELISA. Avian Dis. 30:216–218. 8. Burkhardt, E., and H. Mu ¨ller. 1978. Susceptibility of chicken blood lymphocytes and monocytes to infectious bursal disease virus (IBDV). Arch. Virol. 94:297–303. 9. Case, J. T., A. A. Ardans, D. C. Bolton, and B. J. Reynolds. 1983. Optimization of parameters for detecting antibodies against infectious bronchitis virus using an enzyme-linked immunosorbent assay: temporal response to vaccination and challenge with live virus. Avian Dis. 27:196–210. 10. Cervantes, H. M., L. L. Munger, D. H. Ley, and M. D. Ficken. 1988. Staphylococcus-induced gangrenous dermatitis in broilers. Avian Dis. 32:140–142. 11. Crisman, J. M., R. J. Jackwood, D. P. Lana, and D. J. Jackwood. 1993. Evaluation of VP2 epitopes of infectious bursal disease virus using in vitro expression and radioimmunoprecipitation. Arch. Virol. 128:333–344. 12. Fahey, K. J., A. J. Chapman, I. G. Macreadie, P. R. Vaughan, N. M. McKern, J. I. Skicko, C. W. Ward, and A. A. Azad. 1991. A recombinant subunit vaccine that protects progeny chickens from infectious bursal disease. Avian Pathol. 20:447–460. 13. Hofacre, C. L., J. D. French, and O. J. Fletcher. 1986. Subcutaneous clos-

VOL. 3, 1996

ELISA-BASED DETECTION OF ANTIBODIES TO IBDV SUBTYPES

tridial infection in broilers. Avian Dis. 30:620–622. 14. Hudson, P. J., N. M. McKern, B. E. Power, and A. A. Azad. 1986. Genomic structure of the large RNA segment of infectious bursal disease virus. Nucleic Acids Res. 14:5001–5012. 15. Ismail, N. M., and Y. M. Saif. 1990. Differentiation between antibodies to serotypes 1 and 2 infectious bursal disease viruses in chicken sera. Avian Dis. 34:1002–1004. 16. Ismail, N. M., and Y. M. Saif. 1991. Immunogenicity of infectious bursal disease viruses in chickens. Avian Dis. 35:460–469. 17. Ismail, N. M., Y. M. Saif, and P. D. Moorhead. 1988. Lack of pathogenicity of five serotype 2 infectious bursal disease viruses in chickens. Avian Dis. 32:757–759. 18. Ismail, N. M., Y. M. Saif, W. L. Wigle, G. B. Havenstein, and C. Jackson. 1990. Infectious bursal disease virus variant from commercial leghorn pullets. Avian Dis. 34:141–145. 19. Jackwood, D. H., and Y. M. Saif. 1987. Antigenic diversity of infectious bursal disease virus. Avian Dis. 31:766–770. 20. Jackwood, D. H., Y. M. Saif, and J. H. Hughes. 1986. Replication of infectious bursal disease virus in continuous cell lines. Avian Dis. 31:370–375. 21. Jackwood, D. J., F. S. B. Kibenge, and C. C. Mercado. 1989. Detection of infectious bursal disease virus by using cloned cDNA probes. J. Clin. Microbiol. 27:2437–2443. 22. Jackwood, D. J., Y. M. Saif, and J. H. Hughes. 1982. Characteristics and serologic studies of two serotypes of infectious bursal disease virus in turkeys. Avian Dis. 26:871–882. 23. Jackwood, D. J., Y. M. Saif, and P. D. Moorhead. 1985. Immunogenicity and antigenicity of infectious bursal disease virus serotypes I and II in chickens. Avian Dis. 29:1184–1194. 24. Jackwood, D. J., Y. M. Saif, P. D. Moorhead, and R. N. Dearth. 1982. Infectious bursal disease virus and Alcaligenes faecalis infections in turkeys. Avian Dis. 26:365–374. 25. Lana, D. P., C. E. Beisel, and R. F. Silva. 1992. Genetic mechanisms of antigenic variation in infectious bursal disease virus: analysis of a naturally occurring variant virus. Virus Genes 6:247–259. 26. Luckow, V. A., and M. D. Summers. 1988. Signals important for high-level expression of foreign genes in Autographa californica nuclear polyhedrosis virus expression vectors. Virology 167:56–71. 27. Luckow, V. A., and M. D. Summers. 1988. Trends in the development of baculovirus expression vectors. Biotec 6:47–56. 28. Marquardt, W. W., R. B. Johnson, W. F. Odenwald, and B. A. Schlotthober. 1980. An indirect enzyme-linked immunosorbent assay (ELISA) for measuring antibodies in chickens infected with infectious bursal disease virus. Avian Dis. 24:375–385. 29. McFerran, J. B., M. S. McNulty, E. McKillop, T. J. Conner, R. M. McCracken, D. S. Collins, and G. M. Allan. 1980. Isolation and serological studies with infectious bursal disease viruses from fowl, turkeys, and ducks. Demonstration of a second serotype. Avian Pathol. 9:395–404. 30. Morgan, M. M., I. G. Macreadie, V. R. Harley, P. J. Hudson, and A. A. Azad. 1988. Sequence of the small double-stranded RNA genomic segment of infectious bursal disease virus and its deduced 90-kDa product. Virology 163:240–242. 31. Mu ¨ller, H., C. Scholtissek, and H. Becht. 1979. The genome of infectious bursal disease virus consists of two segments of double-stranded RNA. J. Virol. 31:584–589. 32. Niikura, M., Y. Matsuura, M. Hattori, O. Misao, and T. Mikami. 1991. Expression of the A antigen (gp57-65) of Merek’s disease virus by a recombinant baculovirus. J. Gen. Virol. 72:1099–1104. ¨ ppling, V., H. Mu 33. O ¨ller, and H. Becht. 1991. Heterogeneity of the antigenic site responsible for the induction of neutralizing antibodies in infectious bursal disease virus. Arch. Virol. 119:211–223. 34. O’Reilly, D. R., L. K. Miller, and V. A. Luckow. 1992. Baculovirus expression vectors: a laboratory manual. W. H. Freeman & Company, New York.

463

35. Rosales, A. G., P. Villegas, P. D. Lukert, O. J. Fletcher, M. A. Mohamed, and J. Brown. 1989. Pathogenicity of recent isolates of infectious bursal disease virus in specific-pathogen-free chickens: protection conferred by an intermediate vaccine strain. Avian Dis. 33:729–734. 36. Rosenberger, J. K., and S. S. Cloud. 1986. Isolation and characterization of variant infectious bursal disease viruses abstr. 181. In Proceedings of the Annual Meeting of the American Veterinary Medical Association. American Veterinary Medical Association, Schaumburg, Ill. 37. Rosenberger, J. K., S. S. Cloud, and A. Metz. 1987. Use of infectious bursal disease virus variant vaccines in broiler and broiler breeders p. 105–106. In Proceedings of the 36th Western Poultry Disease Conference. Western Poultry Disease Conference, Veterinary Extension, University of California, Davis. 38. Rosenberger, J. K., and J. Gelb, Jr. 1977. Response to several avian respiratory viruses as affected by infectious bursal disease virus. Avian Dis. 22: 95–105. 39. Rosenberger, J. K., S. Klopp, R. J. Eckroade, and W. C. Krauss. 1975. The role of the infectious bursal agent and several avian adenoviruses in the hemorrhagic-aplastic-anemia syndrome and gangrenous dermatitis. Avian Dis. 19:717–729. 40. Saif, Y. M., D. H. Jackwood, M. W. Jackwood, and D. J. Jackwood. 1987. Relatedness of IBD vaccine strains and field strains, p. 110–111. In Proceedings of the 36th Western Poultry Disease Conference. Western Poultry Disease Conference, Veterinary Extension, University of California, Davis. 41. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 42. Smith, G. E., M. D. Summers, and M. J. Fraser. 1983. Production of human beta interferon in insect cells infected with a baculovirus expression vector. Mol. Cell. Biol. 3:2156–2165. 43. Snyder, D. B. 1990. Changes in the field status of infectious bursal disease virus. Avian Pathol. 19:419–423. 44. Snyder, D. B., W. W. Marquardt, E. T. Mallinson, E. Russek-Cohen, P. K. Savage, and D. C. Allen. 1986. Rapid serological profiling by enzyme-linked immunosorbent assay. IV. Association of infectious bursal disease serology with broiler flock performance. Avian Dis. 30:139–148. 45. Snyder, D. B., W. W. Marquardt, E. T. Mallinson, P. K. Savage, and D. C. Allen. 1984. Rapid serological profiling by enzyme-linked immunosorbent assay. III. Simultaneous measurements of antibody titers to infectious bronchitis, infectious bursal disease, and Newcastle disease viruses in a single serum dilution. Avian Dis. 28:12–24. 46. Solano, W., J. J. Giambrone, and V. S. Panangala. 1986. Comparison of a kinetic-based enzyme-linked immunosorbent assay (KELISA) and virus neutralization test for infectious bursal disease virus. II. Decay of maternal antibody in progeny from white leghorns receiving various vaccination regimens. Avian Dis. 30:126–131. 47. Summers, M. D., and G. E. Smith. 1988. A manual of methods for baculovirus vectors and insect cell culture procedures. Texas Agricultural Experimental Station bulletin no. 1555. Texas Agricultural Experimental Station, College Station, Tex. 48. Thayer, S. G., P. Villegas, and O. J. Fletcher. 1987. Comparison of two commercial enzyme-linked immunosorbent assays and conventional methods for avian serology. Avian Dis. 31:120–124. 49. Vakharia, V. N., D. B. Snyder, J. He, G. H. Edwards, P. K. Savage, and S. A. Mengel-Whereat. 1993. Infectious bursal disease virus structural proteins expressed in a baculovirus recombinant confer protection in chickens. J. Gen. Virol. 74:1201–1206. 50. Whetzel, P. L., and D. J. Jackwood. 1995. Comparison of neutralizing epitopes among infectious bursal disease viruses using radioimmunoprecipitation. Avian Dis. 39:499–506. 51. Wyeth, P. J. 1975. Effect of infectious bursal disease on the response of chickens to S. typhimurium and E. coli infections. Vet. Rec. 96:238–243.