Abstract
An immunoglobulin M (IgM) capture enzyme-linked immunosorbent assay (MC-ELISA) was developed for the detection of primary infection of vesicular stomatitis virus (VSV) in equine and swine sera. The test was based on the use of biotinylated sheep antibodies against equine or swine IgM molecules bound to a streptavidin-coated ELISA plate. The captured IgM antibodies were detected by application of antigens prepared from the New Jersey and the Indiana VSV serotypes (VSV-NJ and VSV-IN, respectively) and mouse polyclonal antibodies against VSV-NJ and VSV-IN. The MC-ELISA was compared to a competitive ELISA (C-ELISA) and the standard microtiter serum neutralization (MTSN) assay by testing serum samples from horses and pigs experimentally infected with VSV-NJ or VSV-IN. The MC-ELISA detected specific homologous IgM antibodies from equine and swine sera as early as 5 and 4 days postinfection (DPI), respectively, and as late as 35 DPI. The MTSN test also detected antibodies as early as 5 DPI and as late as 160 DPI. In a similar fashion, the C-ELISA detected antibodies from 6 to 7 DPI and as late as 160 DPI. These results demonstrated that the MC-ELISA is a useful test for serodiagnosis of primary VSV infection in horses and pigs.
Vesicular stomatitis (VS) is a contagious viral disease that primarily affects cattle, horses, swine (7, 19), and some wild ungulates (9) in enzootic and epizootic forms in the tropical and subtropical areas of the Americas. VS, an important disease in the United States and many South American countries, spreads rapidly and has serious socio-economic and public health consequences. It is identified as a List A disease by the Office International des Epizooties (13) and is important in the international trade of animals and animal products. Vesicular stomatitis virus (VSV) is a Rhabdovirus of the Vesiculovirus genus with potential for arthropod transmission (7, 8, 17). Two VSV serotypes, VSV New Jersey (VSV-NJ) and VSV Indiana (VSV-IN), are serologically distinct and are of major etiological concern because both cause infections in cattle, horses, and swine. These viruses are morphologically similar, have some common antigens, and produce overt infection with similar lesions in susceptible animals. Early diagnosis of VS is necessary and important for the differential diagnosis of other vesicular diseases. Serodiagnostic tests available for distinguishing VSV-NJ and VSV-IN are neutralization tests (5, 15, 18), indirect enzyme-linked immunosorbent assay (I-ELISA) (1), and competitive ELISA (C-ELISA) (2). These tests are reliable but have some limitations. They can detect early and long-lasting antibody responses, but because of the nature of these assays, they are not able to differentiate a primary from a secondary VSV infection.
In a primary viral infection, immunoglobulin M (IgM) class antibody is the first to appear in the blood circulation, and it disappears shortly after the IgG antibodies develop (6). Because of the ontogeny of the antibody response, detection of specific IgM antibodies provides a differential serodiagnosis of virus infection. Thus, Vernon and Webb developed an IgM capture ELISA (MC-ELISA) that detected the recent infection of horse and cattle with VSV-NJ (16). In their assay, the ELISA plates were directly coated with rabbit anti-equine and anti-bovine IgM antibodies, and IgM antibodies to VSV-NJ were detected as early as 6 days postinfection (DPI). The objective of our study was to develop an MC-ELISA using an avidin-biotin system for detection of IgM anti-VSV-NJ and anti-VSV-IN antibodies from horses and pigs. The MC-ELISA utilized the unique nature of a biotin derivative, EZ-Link Sulfo-NHS-LC-Biotin, which has an extended space arm to reduce both steric hindrance and interference with the biological activity of the coupled IgG (3). The MC-ELISA was used for detection of both anti-VSV-NJ and anti-VSV-IN antibodies from horses and pigs in comparison with the C-ELISA and serum neutralization test.
MATERIALS AND METHODS
VSV antigens.
VSV antigens were produced from Vero cells infected with the Ogden strain of VSV-NJ and San Juan strain of VSV-IN serotypes using a method described by Afshar et al. (1). These antigens were used in the MC-ELISA and the C-ELISA for detection of antibodies to VSV-NJ and VSV-IN, respectively, as described below.
Purification and titration of mouse antibodies to VSV-NJ and VSV-IN.
Each 5 ml of mouse ascitic fluids produced and provided by Afshar et al. (2) against either VSV-NJ or VSV-IN serotypes was precipitated with 50% (vol/vol) saturated ammonium sulfate. Each globulin preparation was resuspended in 5 ml of phosphate-buffered saline (PBS) and was passed through a Sephacryl S-300 high-resolution gel filtration column (Pharmacia Biotech, Inc., Baire d'Urfé, Quebec, Canada). The fractions representing peak IgG were collected and concentrated using a microconcentrator with a molecular weight (MW) cutoff of 50,000 (Millipore Canada Ltd., Nepean, Ontario, Canada). Protein concentration was calculated based on the extinction coefficient of 13.5 for a 1% preparation at an optical density at 280 nm (OD280).
An I-ELISA was used to titrate the purified mouse anti-VSV antibodies. The optimal dilutions of the antigens were predetermined by a checkerboard titration (12) with the use of mouse polyclonal antibodies in ascitic fluid. ELISA plates (Gibco, Burlington, Ontario, Canada) were coated with either VSV-NJ or VSV-IN antigens at the predetermined dilutions of 1:4,000 or 1:2,500, respectively, in carbonate buffer (0.06 M, pH 9.6) and stored at −70°C for up to 3 months. Plates were thawed at room temperature and washed, without blocking agent, using the washing buffer PBS-T (0.01 M PBS [pH 7.2] containing 0.05% Tween 20 [vol/vol]). Purified mouse antibodies at various dilutions in PBS-T were added into the wells (100 μl/well) and incubated for 1 h at 37°C. After removing unbound materials by washing five times with PBS-T, wells were filled with 100 μl of a 1:2,000 dilution of horseradish peroxidase-labeled goat anti-mouse IgG (Bethyl Laboratories, Montgomery, Tex.) in PBS-T and incubated at 37°C for 1 h. The substrate used was 2,2′-azinobis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) (Sigma Chemical Company, St. Louis, Mo.) in 0.05 M citrate buffer (pH 4.5) and 0.3% H2O2. The OD410 of each well was measured on an automatic ELISA plate reader (TiterTek plus; ICN Flow; Biomedical, Inc., Costa Mesa, Calif.).
Equine and swine serum samples.
Sequential blood samples were collected from one horse (H-1) and two pigs (P-35-94 and P-36-94) experimentally infected with 107 50% tissue culture infective doses (TCID50) of VSV-NJ Ogden strain and one horse (H-2) and two pigs (P-39-94 and P-40-94) infected with 104 TCID50 of VSV-IN San Juan strain. The details of the experimental infection and serological results have been described previously (1, 4). In addition, sequential blood samples collected from three horses were obtained from the National Veterinary Services Laboratories, U.S. Department of Agriculture, Ames, Iowa. These sera were collected from two horses (H-106 and H-109) experimentally infected with 105 TCID50 of VSV-NJ and from one horse (H-107) infected with the same dose of VSV-IN, as described previously (10).
Purification of IgM antibodies from equine serum samples.
IgM molecules were purified from two serum samples collected at 6 DPI from two horses (H-1 and H-2) experimentally infected with VSV-NJ and VSV-IN serotypes, respectively. Each 5 ml of equine serum was passed through a Sephacryl S-300 high-resolution gel filtration column (Pharmacia Biotech, Inc.). Fractions containing IgM molecules were collected and concentrated to their original 5-ml volume using a microconcentrator with an MW cutoff of 100,000) (Millipore Canada Ltd.). A concentration of total IgM molecules of 3 mg/ml was determined using the methods described by Lowery et al. (11). Their specific binding to VSV antigens was detected and titrated by the MC-ELISA, as described below.
Biotinylation of sheep antibodies against equine and swine IgM molecules.
The EZ-Link Sulfo-NHS-LC-Biotin purchased from Pierce (Rockford, Ill.) was used for biotinylating sheep IgG antibodies against equine and swine IgM according to the manufacturer's instructions. Briefly, 2 mg of sheep IgG antibodies against equine or swine IgM molecules (Cedarlane Laboratories Ltd., Canada, Hornby, Ontario, Canada) in 1 ml of 50 mM sodium bicarbonate buffer, pH 8.5, was mixed with 74 μl of EZ-Link Sulfo-NHS-LC-Biotin (1 mg/ml in water). The reaction between sheep IgG and biotin was allowed to proceed for 2 h on ice, and free biotin was removed using a microconcentrator (MW cutoff of 5,000). The biotinylated sheep antibody preparations were mixed with equal volume of glycerol and stored at −20°C. The optimal dilution of the biotinylated sheep antibody preparations was determined using a checkerboard titration method. Various dilutions of the biotinylated sheep antibody preparations were bound to the streptavidin-coated plate for 1 h at 37°C followed by three washing steps with PBS-T. Equine and swine IgM antibodies at various concentrations were added to the plate and incubated for 1 h at 37°C, and this was followed by washing steps. The conjugate was sheep antibody against equine or swine IgM coupled to horseradish peroxidase at a dilution of 1:3,000 as recommended by the manufacturer (Cedarlane Laboratories Ltd., Canada), and the substrate was ABTS, as described above. The optimal dilution of 1:2,000 for biotinylated sheep antibodies against both equine and swine IgM was selected and used in the following MC-ELISA.
MC-ELISA.
The MC-ELISA was developed by coating the solid phase of an ELISA plate (Gibco) with recombinant streptavidin (Boehringer Mannheim Canada, Laval, Quebec, Canada) at 10 μg/well in 100 μl of PBS (0.01 M, pH 7.2) without using blocking agent. After drying the wells by incubating the plate 18 h at 37°C, biotinylated sheep antibodies against equine or swine IgM molecules at the predetermined dilution of 1:2,000 in 100 μl of dilution buffer (0.13 M sodium chloride, 0.05 M Tris-hydrochloride, and 1 mM EDTA at pH 7.8) per well were added. The plate was incubated at 37°C for 1 h and then washed five times with PBS-T. The purified IgM antibodies and the equine or swine serum samples in the dilution buffer were added into the wells at 100 μl/well. The plate was subsequently incubated and washed as described above, and the VSV antigens diluted 1:500 in dilution buffer (100 μl/well) were added into the wells. Following another incubation period and washing steps, the purified mouse antibodies to either VSV-NJ or VSV-IN at optimal dilutions in dilution buffer (100 μl/well) were added into the wells. The plate was incubated at 37°C for 1 h and then washed. The binding of antigens by antibodies was detected by applying the goat antibodies against mouse IgG conjugated to horseradish peroxidase (Cedarlane Laboratories Ltd., Canada) diluted 1:3,000 in dilution buffer (100 μl/well). After the final incubation (37°C for 1 h) and washing steps, the substrate was applied and the plate was read, as described above. The cutoff value was established at an OD of 0.3.
C-ELISA.
The C-ELISA was performed based on a procedure described by Afshar et al. (2) for detection of equine and swine antibodies to VSV-NJ and VSV-IN. In brief, viral antigens were prepared from Vero cells infected with the Ogden strain of VSV-NJ or San Juan strain of VSV-IN serotypes, as described in detail previously (1), and then adsorbed to ELISA plate wells. Serum samples were diluted in PBS-T and 50 μl of sample was added to each well, followed immediately by the addition of 50 μl of the mouse polyclonal antibodies to VSV-NJ or VSV-IN. Conjugate and substrate were the same as in the I-ELISA described above. The cutoff value of 50% inhibition was used.
MTSN test.
The microtiter serum neutralization (MTSN) test was performed according to a method previously described (14) for testing neutralizing serum antibodies to VSV-NJ strain VS 20-022-522 and VSV-IN strain VS 20-001-022 obtained from the American Type Culture Collection (Manassas, Va.). The serum neutralizing antibody titer was expressed as the reciprocal of the dilution giving complete protection against cytopathic effect of 1,000 TCID50 of virus at a titer of 1:32 or higher.
RESULTS
Purification and titration of mouse antibodies.
Mouse polyclonal IgG antibodies against VSV-NJ and VSV-IN were purified individually and titrated against the homologous and heterologous antigens. The results shown in Fig. 1 demonstrated that the end point titers of 0.78 and 1.2 μg/ml were generated for the anti-VSV-NJ antibody (F38–51) and the anti-VSV-IN antibody (F39–52), respectively. The end point titer was determined by the last dilution of antibody preparation giving an OD410 value of 1.0 after 10 min of development time in the I-ELISA. The 0.78- and 1.2-μg/ml concentrations of antibody preparations against VSV-NJ and VSV-IN, respectively, were used in the MC-ELISA and C-ELISA.
FIG. 1.
Titration of mouse anti-VSV antibodies. IgG antibodies were separated from ascitic fluid collected from mice infected with VSV-NJ (F38-51-NJ) or VSV-IN (F39-52-IN). Their levels of binding to VSV-NJ and VSV-IN antigens were detected by an I-ELISA for VSV-NJ (A) and VSV-IN (B). A cutoff value of an OD410 of 1.0 (horizontal lines) was used to determine the end point titer of the antibodies.
Sensitivity of MC-ELISA.
The purified IgM antibodies and nonpurified sera from two horses (H-1 and H-2) were used to determine the sensitivity of the MC-ELISA. As demonstrated in Fig. 2, the end point titer for both the serum sample and purified IgM antibodies against VSV-NJ was 1:50,000. Similarly, the antibodies against VSV-IN were detected with the end point titer of 1:8,000 for the serum sample and 1:50,000 for the purified IgM antibodies. No cross-reaction between homologous and heterologous anti-VSV antibodies was observed, and no IgM antibodies in prebleed serum samples were detected.
FIG. 2.
Sensitivities of sera and IgM antibodies to VSV-NJ (A) and VSV-IN (B) measured in the VSV-NJ MC-ELISA (A) and VSV-IN MC-ELISA (B). Serum samples and purified IgM antibodies were collected from horses experimentally infected with VSV-NJ (H-1) and VSV-IN (H-2) at 0 DPI (0d) and 6 DPI (6d). An OD410 of 0.3 (horizontal lines) was selected as the cutoff.
Detection of antibodies from equine serial bleed samples.
The MC-ELISA was compared with the C-ELISA and MTSN test for detection of equine antibodies to VSV (Table 1). For VSV-NJ IgM antibody detection, sequential serum samples were collected from three horses infected with VSV-NJ. In the MC-ELISA, using a cutoff value of an OD of 0.3, IgM antibodies from animal H-1 were detected in the 5-DPI sample. Earlier samples (0, 1, and 3 DPI) were negative for IgM antibodies. Similarly, IgM anti-VSV-NJ antibodies were detected at 6 DPI in sera collected from two other animals (H-106 and H-109). The samples collected from these two animals between 0 and 4 DPI were not available in this study, but Katz et al. (10) previously reported them to be negative for VSV antibodies by the complement fixation test, IgM capture ELISA, serum neutralization test, and C-ELISA (10). The IgM antibodies from these animals were detected for up to 35 DPI. The peak IgM titer was observed between 8 and 11 DPI, with an average OD of 1.04.
TABLE 1.
Results of MC-ELISA, C-ELISA, and MTSN assay with samples from horses experimentally infected with VSV-NJ or VSV-IN
| Infection and animal | DPI | Result of test for indicated VSV serotype
|
|||||
|---|---|---|---|---|---|---|---|
| MC-ELISAa
|
C-ELISAb
|
MTSNc
|
|||||
| NJ | IN | NJ | IN | NJ | IN | ||
| VSV-NJ | |||||||
| H-1 | 0 | − | − | − | − | − | − |
| 1 | − | − | − | − | − | − | |
| 3 | − | − | − | − | − | − | |
| 5 | + | − | − | − | + | − | |
| 6 | + | − | + | − | + | − | |
| 14 | + | − | + | − | + | − | |
| 21 | + | − | + | − | + | − | |
| 28 | + | − | + | − | + | − | |
| H-106 | 6 | + | − | − | − | + | − |
| 7 | + | − | − | − | + | − | |
| 8 | + | − | − | − | + | − | |
| 10 | + | + | − | − | + | − | |
| 11 | + | + | + | − | + | − | |
| 14 | + | + | + | − | + | + | |
| 21 | + | − | + | − | + | + | |
| 28 | + | − | + | − | + | + | |
| H-109 | 5 | − | − | − | − | + | − |
| 6 | + | − | − | − | + | − | |
| 7 | + | − | − | − | + | − | |
| 8 | + | − | − | − | + | − | |
| 10 | + | + | + | − | + | − | |
| 11 | + | + | + | − | + | + | |
| 14 | + | − | + | − | + | − | |
| 21 | + | − | + | − | + | − | |
| 28 | + | − | + | − | + | − | |
| 35 | + | − | + | − | + | − | |
| VSV-IN | |||||||
| H-2 | 0 | − | − | − | − | − | − |
| 1 | − | − | − | − | − | − | |
| 3 | − | − | − | − | − | − | |
| 5 | − | + | − | − | − | − | |
| 6 | − | + | − | − | − | + | |
| 12 | − | + | − | + | − | + | |
| 21 | − | + | − | + | − | + | |
| 28 | − | + | − | + | − | + | |
| H-107 | 6 | − | + | − | − | − | + |
| 7 | − | + | − | − | − | + | |
| 8 | − | + | − | − | − | + | |
| 11 | + | + | − | + | − | + | |
| 14 | + | + | − | + | + | + | |
| 21 | − | + | − | + | + | + | |
| 28 | − | + | − | + | + | + | |
Cutoff value of OD410 of 0.3.
Cutoff value of 50% inhibition.
Cutoff value of 1:32.
For VSV-IN IgM antibody detection, two horses (H-2 and H-107) were infected with VSV-IN. IgM antibodies against VSV-IN were detected in these samples by the MC-ELISA at 5 and 6 DPI. Antibodies remained at high levels up to 12 DPI (OD > 1.2) and diminished after 28 DPI. No IgM antibodies were detected before 5 DPI in samples from animal H-2 (Table 1) or animal H-107 by another IgM capture ELISA (10).
In the heterologous MC-ELISA, no cross-reactions were observed in samples from animals H-1 and H-2 (Table 1), and only a low level of cross-reaction was seen in samples from three other animals (H-106, H-109, and H-107). The cross-reactions in the VSV-IN MC-ELISA were detected for animals H-106 and H-109 (infected with VSV-NJ) at 10, 11, and 14 DPI and at 10 and 11 DPI, respectively. Cross-reactive samples produced ODs of 0.46 or less. A similar cross-reaction was observed in samples from one VSV-IN-infected animal (H-107) in the VSV-NJ MC-ELISA at 11 and 14 DPI, with an OD of ≤0.38.
The C-ELISA, with a 50% inhibition cutoff value, detected equine anti-VSV-NJ antibodies as early as 6, 10, and 11 DPI from animals H-1, H-109, and H-106, respectively (Table 1). Similarly, anti-VSV-IN antibodies were also detected at 11 and 12 DPI from animals H-107 and H-2, respectively. No cross-reaction in the heterologous C-ELISAs was observed in any of five horses.
The MTSN test detected equine antibodies to VSV-NJ as early as at 5, 6, and 5 DPI for animals H-1, H-106, and H-109, respectively, and to VSV-IN at 6 DPI in horses H-2 and H-107. No cross-reactivity was detected for samples from animals H-1 and H-2. Some degree of crossreactivity was observed in samples from three other animals when tested in the heterologous assays (Table 1). One animal (H-109) gave a transient cross-reaction at 11 DPI (titer of 1:32), while the sera of two other animals (H-106 and H-107) cross-reacted in the heterologous assays at 14, 21, and 28 DPI (antibody titer of 1:32 to 1:128).
Detection of antibodies from swine serum samples.
The MC-ELISA was compared to the C-ELISA and MTSN assay by testing sequential blood samples from pigs experimentally infected with VSV-NJ or VSV-IN. As shown in Table 2, IgM antibodies were detected in the MC-ELISA as early as 4 DPI and up to 28 DPI from pigs infected with both VSV serotypes. Cross-reactions were observed in samples from these animals in the heterologous MC-ELISAs, with ODs of <0.45. The cross-reactions were from samples collected at 6, 7, and 14 DPI from two pigs infected with VSV-NJ and at 6 and 7 DPI from two other pigs infected with VSV-IN.
TABLE 2.
Results of MC-ELISA, C-ELISA, and MTSN assay with samples from pigs experimentally infected with VSV-NJ and VSV-INa
| Infection and animal | DPI | Result of test for indicated VSV serotype
|
|||||
|---|---|---|---|---|---|---|---|
| MC-ELISA
|
C-ELISA
|
MTSN
|
|||||
| NJ | IN | NJ | IN | NJ | IN | ||
| VSV-NJ | |||||||
| P-35-94 | 0 | − | − | − | − | − | − |
| P-36-94 | 0 | − | − | − | − | − | − |
| P-35-94 | 4 | + | − | − | − | − | − |
| P-36-94 | 4 | + | − | − | − | + | − |
| P-35-94 | 5 | + | − | − | − | + | − |
| P-36-94 | 5 | + | − | − | − | + | − |
| P-35-94 | 6 | + | + | − | − | + | − |
| P-36-94 | 6 | + | − | + | − | + | − |
| P-35-94 | 7 | + | + | − | − | + | − |
| P-36-94 | 7 | + | + | + | − | + | − |
| P-35-94 | 14 | + | − | − | − | + | − |
| P-36-94 | 14 | + | + | + | − | + | − |
| P-35-94 | 21 | + | − | − | − | + | − |
| P-36-94 | 21 | + | − | + | − | + | − |
| P-35-94 | 28 | + | − | + | − | + | − |
| P-36-94 | 28 | + | − | + | − | + | − |
| P-35-94 | 35 | − | − | + | − | + | − |
| P-36-94 | 35 | − | − | + | − | + | − |
| P-35-94 | 42 | − | − | + | − | + | − |
| P-36-94 | 42 | − | − | + | − | + | − |
| VSV-IN | |||||||
| P-39-94 | 0 | − | − | − | − | − | − |
| P-40-94 | 0 | − | − | − | − | − | − |
| P-39-94 | 4 | − | + | − | − | − | − |
| P-40-94 | 4 | − | + | − | − | − | + |
| P-39-94 | 5 | − | + | − | − | + | + |
| P-40-94 | 5 | − | + | − | − | − | + |
| P-39-94 | 6 | + | + | − | − | + | + |
| P-40-94 | 6 | + | + | − | − | + | + |
| P-39-94 | 7 | + | + | − | − | − | + |
| P-40-94 | 7 | + | + | − | − | − | + |
| P-39-94 | 14 | − | + | − | + | − | + |
| P-40-94 | 14 | − | + | − | + | − | + |
| P-39-94 | 21 | − | + | − | + | − | + |
| P-40-94 | 21 | − | + | − | + | − | + |
| P-39-94 | 28 | − | − | − | + | − | + |
| P-40-94 | 28 | − | − | − | + | − | + |
| P-39-94 | 35 | − | − | − | + | − | + |
| P-40-94 | 35 | − | − | − | + | − | + |
| P-39-94 | 42 | − | − | − | + | − | + |
| P-40-94 | 42 | − | − | − | + | − | + |
See footnotes to Table 1.
The C-ELISA detected antibodies against VSV-NJ from one animal (P-36–94) at 6 DPI and at 28 DPI from the second animal (P-35–94) (Table 2). For VSV-IN, antibodies were detected from both animals (P-39–94 and P-40–94) at 14 DPI. Antibodies against both VSV serotypes remained positive at 42 DPI and over 160 DPI (data not shown). No cross-reactions were identified for these animals in the heterologous C-ELISAs (Table 2).
The MTSN test detected neutralizing antibodies to both VSV serotypes as early as 4 or 5 DPI (Table 2) that persisted more than 160 DPI (data not shown). No cross-reactions were observed for two pigs infected with the VSV-NJ and tested in the VSV-IN MTSN. However, some cross-reaction occurred with samples from two other animals infected with VSV-IN and tested in the VSV-NJ MTSN (Table 2). In this case, cross-reactions were observed in samples collected at 5 and 6 DPI with antibody titers of ≤1:64.
DISCUSSION
Currently, the tests most commonly used for the detection of antibodies against VSV are the serum neutralization test and the C-ELISA described in the Manual of Standards for Diagnostic Tests and Vaccines (13). These assays are not able to distinguish between IgM and IgG antibodies and therefore are not able to differentiate a primary from a secondary infection. In this study, we report the development of an MC-ELISA for the detection of primary infection of VSV-NJ and VSV-IN serotypes in horses and pigs based on a streptavidin-biotin system.
The MC-ELISA had an analytical sensitivity of 60 ng of total purified IgM antibodies per ml (or 6 ng/well) against both VSV serotypes (Fig. 2), i.e., a concentration of 3 mg of IgM preparation diluted 50,000 times. The purified IgM antibodies had higher binding values than the whole serum samples (end point titers of 1:50,000 for anti-VSV-NJ and of 1:80,000 for anti-VSV-IN), which may be due to the homology of the purified IgM antibody preparation that has no other molecules to cause potential nonspecific blocking of the antigen-antibody interaction. This level of analytical sensitivity is similar to that of the IgM capture ELISA developed by Vernon and Webb (16) in which ELISA plates were coated with rabbit antibodies against bovine and equine IgM molecules and a high-salt buffer was used as a dilution buffer. In our laboratory, the same format using rabbit and sheep antibodies against equine and swine IgM diluted in the same high-salt buffer resulted in high background levels of binding (data not shown). This led us to the development of an avidin-biotin system to capture sheep antibodies to equine and swine IgM. To test the performance of the MC-ELISA for anti-VSV IgM antibody detection in VSV-infected swine and horses, the C-ELISA and MTSN test were included in this study and compared with the MC-ELISA.
As demonstrated in Table 1, the MC-ELISA detected the specific IgM anti-VSV antibodies between 6 and 35 DPI. The IgM antibodies did diminish after 49 DPI, based on the study by Katz et al. (10). In comparison, the C-ELISA detected the equine anti-VSV antibodies at later DPI, remained positive at 35 DPI and more than 49 DPI. The antibodies detected by the MTSN test were developed as early as 6 DPI but lasted more than 60 DPI as reported by Katz et al. (10). However, because it is not an immunoglobulin isotype-specific assay, the MTSN test cannot provide information of whether the antibodies are produced by a primary or a secondary stimulation of the virus infection.
It was not surprising that some degree of cross-reaction was observed between the homologous and heterologous MC-ELISAs due to the polyclonal nature of the captured IgM and the heterologous nature of the antigens used in this assay. However, this cross-reaction was not observed in every animal serum sample. As shown in this study, the value of cross-reaction was much lower than that obtained from the homologous assay and only lasted a few days until the titer of homologous antibody reached its peak. Similar results were generated with swine antisera against VSV-NJ and VSV-IN.
From this study, we can conclude that the MC-ELISA provides relatively higher sensitivity for the detection of equine and swine IgM antibodies to VSV-NJ and VSV-IN. The specificity of the MC-ELISA needs to be further evaluated by testing large numbers of negative equine and swine serum samples. The MC-ELISA, like the MTSN test, was able to detect the IgM antibodies at early days of infection. However, due to the nature of these assays, only the MC-ELISA can differentiate a primary from a secondary infection of VSV.
Therefore, the C-ELISA can be used as a screening test for the serodiagnosis of VSV infection. Samples identified as positive by the C-ELISA should be confirmed by the MTSN test and subsequently tested by the MC-ELISA to provide further information on antibody isotypes. Results of the MC-ELISA would indicate if the infection is primary or secondary. This information would be valuable for routine disease surveillance and of particular use as an epidemiological tool during VS outbreaks.
ACKNOWLEDGMENTS
We thank Lisa Fernando and Deidre Ridd for their excellent technical assistance.
REFERENCES
- 1.Afshar A, Dulac G C, Wright P F, Martin D. Application of indirect ELISA for detection of bovine antibodies against vesicular stomatitis viruses. J Vet Diagn Investig. 1993;5:26–32. doi: 10.1177/104063879300500107. [DOI] [PubMed] [Google Scholar]
- 2.Afshar A, Shakarchi H, Dulac G C. Development of a competitive enzyme-linked immunosorbent assay for detection of bovine, ovine, porcine, and equine antibodies to vesicular stomatitis virus. J Clin Microbiol. 1993;31:1860–1865. doi: 10.1128/jcm.31.7.1860-1865.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Altin J G, Pagler E B. A one-step procedure for biotinylation and chemical cross linking of lymphocyte surface and intracellular membrane-associated molecules. Anal Biochem. 1995;224:382–389. doi: 10.1006/abio.1995.1054. [DOI] [PubMed] [Google Scholar]
- 4.Boulanger P. Complement-fixation tests of swine serum. I. In the diagnosis of vesicular stomatitis. Can J Comp Med. 1955;19:37–47. [PMC free article] [PubMed] [Google Scholar]
- 5.Brown F, Bishop D H L, Crick J. Rhabdoviridae. Intervirol. 1979;12:1–17. doi: 10.1159/000149062. [DOI] [PubMed] [Google Scholar]
- 6.Carayannopoulos L, Capra I D. Immunoglobulins: structure and function. In: Paul W E, editor. Fundamental immunology. 3rd ed. New York, N.Y: Raven Press; 1993. pp. 283–314. [Google Scholar]
- 7.Hanson R P. The natural history of vesicular stomatitis. Bacteriol Rev. 1952;16:179–204. doi: 10.1128/br.16.3.179-204.1952. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hanson R P. Vesicular stomatitis. In: Gibbs E P J, editor. Virus diseases of food animals. Vol. 2. New York, N.Y: Academic Press; 1981. pp. 517–539. [Google Scholar]
- 9.Karstad L H, Adams E V, Hanson R P, Ferris D H. Evidence for the role of wildlife in epizootics of vesicular stomatitis. J Am Vet Med Assoc. 1956;129:95–96. [PubMed] [Google Scholar]
- 10.Katz J B, Eernisse K A, Landgraf J G, Schmitt B J. Comparative performance of four serodiagnostic procedures for detecting bovine and equine vesicular stomatitis virus antibodies. J Vet Diagn Investig. 1997;9:329–331. doi: 10.1177/104063879700900321. [DOI] [PubMed] [Google Scholar]
- 11.Lowery O H, Rosebrough N J, Farr A L, Randall R J. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265–275. [PubMed] [Google Scholar]
- 12.Nielsen K H, Wright P F, Nielsen E B. Enzyme immunoassay: application to the detection of bovine antibody to Brucella abortus. Mo: Agriculture Canada; 1984. nograph. [Google Scholar]
- 13.Office International des Epizooties. Manual of standards for diagnostic tests and vaccines. Paris, France: Office International Des Epizooties, World Organisation for Animal Health; 1996. p. 57. [Google Scholar]
- 14.Snyder M L, Jenney E W, Erickson G A, Carbrey E A. The 1982 resurgence of vesicular stomatitis in the United States; a summary of laboratory diagnostic findings. Proc Annu Meet Am Assoc Vet Lab Diagn. 1982;25:221–228. [Google Scholar]
- 15.Tesh R B, Da Rosa A P A, Travassos D R, Travassos J S. Antigenic relationship among rhabdoviruses infecting terrestrial vertebrates. J Gen Virol. 1983;64:169–176. doi: 10.1099/0022-1317-64-1-169. [DOI] [PubMed] [Google Scholar]
- 16.Vernon S D, Webb P A. Recent vesicular stomatitis virus infection detected by immunoglobulin M antibody capture enzyme-linked immunosorbent assay. J Clin Microbiol. 1985;22:582–586. doi: 10.1128/jcm.22.4.582-586.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Webb P A, Holbrook F R. Vesicular stomatitis. In: Monath T P, editor. The arboviruses: epidemiology and ecology. Vol. 5. Boca Raton, Fla: CRC Press, Inc.; 1989. pp. 1–29. [Google Scholar]
- 18.Wilks C R, House J A. The glycoproteins of seven vesiculoviruses are antigenically distinct. Arch Virol. 1985;86:335–340. doi: 10.1007/BF01309837. [DOI] [PubMed] [Google Scholar]
- 19.Yuill T M. Vesicular stomatitis. In: Beran G W, editor. CRC handbook series in zoonoses, section B. Viral zoonoses. Vol. 1. Boca Raton, Fla: CRC Press, Inc.; 1981. pp. 125–142. [Google Scholar]