Vol. 10, No. 5
JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 1979, p. 633-636 0095-1137/79/1 1-0633/04$02.00/0
Enzyme-Linked Immunosorbent Assay for Serology of Infectious Bovine Rhinotracheitis Virus Infections PIERRE PAYMENT,`* ROBERT ASSAF,2 MICHEL TRUDEL,' AND PAUL MAROIS2 Centre de Recherche en Virologie1 and Centre de Recherche en Medecine Vitrinaire,2 Institut ArmandFrappier, Ville de Laval, Quebec, Canada H7N 4Z3
Received for publication 7 August 1979
An enzyme-linked immunosorbent assay was developed to detect antibodies to infectious bovine rhinotracheitis virus in bovine sera. The test used viral antigen purified in a sucrose density gradient, and the antibody titers detected were up to 100 times higher than those seen in the serum neutralization test. When 250 sera were tested by those two tests, 146 were positive and 49 were negative in both tests. Only 1 serum was negative in the immunoassay test and positive in the neutralization test. However, 54 sera negative in the neutralization test were positive in the immunoassay test. These positive results appear to be in agreement with the increased sensitivity of the enzyme immunoassay. The measurement of antibodies to infectious bovine rhinotracheitis (IBR) virus after a natural infection or after vaccination is usually based on the serum neutralization (SN) test (2). In 1975 we reported (1) that immunofluorescence was a suitable and faster alternative. More recently the development of new immunoenzymatic assays for several animal infections (4) prompted us to investigate an enzyme-linked immunosorbent assay (ELISA) test for the measurement of antibodies in certain viral infections. We have already reported its successful use for the serology of pneumonia virus of mice (3), and we are now reporting its use for the serology of IBR virus, using semipurified virus as the antigen. MATERIALS AND METHODS Virus and cell culture. Primary and secondary bovine testis cell cultures were grown in lactalbumin hydrolysate medium containing 4% fetal bovine serum. The Colorado strain (passage 36) of IBR virus was used at 1,000 50% tissue culture infective doses to infect the confluent monolayers. Lactalbumin hydrolysate medium without fetal bovine serum was used as maintenance medium. When the cytopathic effect involved 90% of the monolayer, about 48 h later, the cultures were frozen and thawed (-70°C, 37°C) three times. Viral antigen preparation for ELISA. Frozen and thawed infected or uninfected cell culture fluids were adjusted to 1.0 M NaCl and 8% polyethylene glycol (molecular weight, 6,000). The mixture was agitated for 2 h at 4°C, and the precipitate was then recovered by centrifugation at 3,000 x g for 30 min. The pellet was resuspended in saline to 'Ao of the original volume. The concentrated viral and control preparations were sonicated for 1 min at 100 W/cm2, using the
microprobe of a model 1510 Bransonic sonicator. Five milliliters of this material was added to 15 ml of NTE buffer (0.15 M NaCl, 0.01 M tris(hydroxymethyl)aminomethane-hydrochloride, 0.001 M ethylenediaminetetraacetate, pH 7.8) and placed on a discontinuous sucrose gradient formed by layering 10 ml of 30% (wt/wt) sucrose and 10 ml of 50% sucrose in NTE buffer in a 40-ml tube. This preparation was centrifuged at 25,000 rpm in an SW-27 rotor for 2 h. Two-milliliter fractions were recovered. The 30%-50% sucrose interface contained the viral antigen. The same fractions from uninfected material were used as control antigen. ELISA procedure. The ELISA assay was set up as described by Voller et al. (5), using Cooke Engineering microtiter 96-well (round bottom) hemagglutination plates. The antigens were diluted in 0.05 M glycine buffer at pH 10. The antigen (0.2 ml per well) was left overnight at 4°C. Excess antigen was removed the next morning by three 1-min washings with distilled water containing 0.05% Tween 80. Sera were diluted in phosphate-buffered saline containing 0.05% Tween 80. Appropriate wells received 0.2 ml of the dilutions, and the plates were incubated for 2 h at 37°C in a humidified incubator. Excess serum was removed by washing with distilled water-0.05% Tween 80. Immunoglobulin G antibodies were detected by using a commercial preparation of goat immunoglobulin G anti-bovine immunoglobulin G conjugated to peroxidase (Miles Laboratories, Inc., Elkhart, Ind.). The conjugate was diluted 1:1,500 in phosphatebuffered saline-0.05% Tween 80 containing 2% egg albumin (Sigma Chemical Co., St. Louis, Mo.). Bovine serum albumin was not used because it contains 2 to 5% bovine immunoglobulins, which reacted with the conjugate and lowered its activity. The optimal dilution of the conjugate was determined as described by Voller et al. (5). The conjugate (0.2 ml per well) was allowed to react for 2 h at room temperature, and the surplus was removed by washing as described earlier. Freshly prepared 5-aminosalicylic acid at 2 mg/ml, pH
633
634
PAYMENT ET' AL.
J. CLIN. MICROBIOLJ.
6.0, with 0.005% hydrogen peroxide was used as the substrate. The plates were read visually after 60 min at room temperature, and the endpoint was determined as the highest dilution with an above-background reaction. Serum samples. Paired sera were obtained from animals suspected of having an IBR virus infection. The first sera were obtained as soon as the symptoms developed, and the second sera were taken about 3 weeks later. Individual sera were selected among those submitted to us for various analyses. All sera were kept at -20°C. SN test. The SN test was performed as described previously (1).
fractions at the 30%-50% interface. When the same sera were used to evaluate a gradient containing an uninfected-cell extract, there were no specific reactions. Antigen preparations were used at an optimal dilution of 1:200 to 1:500 as determined by checkerboard titration with known positive and negative sera. Specificity of ELISA-IBR. Paired sera from 20 animals were tested by the SN and the ELISA-IBR tests (Table 1). With 17 pairs the results were in agreement, but in three animals with high SN titers in their first sera, ELISAIBR did not detect the slight increases seen by SN. All sera negative in the SN test had ELISAIBR titers lower than or equal to 1:160. The highest ELISA-IBR titer was 1:10,240. Correlation between SN and ELISA-IBR tests. The sensitivity of the ELISA-IBR test was determined by using 250 sera with known SN titers (Table 2). For 195 (78%) sera there was complete agreement: 146 (58%) were positive and 49 (20%) were negative by both tests. Only 1 serum (0.04%) positive by SN was negative by ELISA-IBR, but the SN titer was borderline, at 1:8. In 54 (22%) sera ELISA-IBR detected antibody titers ranging from 1:10 to 1:640, with the largest number of sera having a titer of 1:40. When individual results were plotted on a graph correlating the two tests (Fig. 2), the high sensitivity of the ELISA-IBR test was revealed. Two large groups of ELISA-IBR results are
RESULTS Viral antigen. Unpurified cell extracts were not satisfactory antigens for the ELISA test. For undetermined reasons, very high nonspecific reactions were observed even with negative sera on noninfected-cell extracts. Acceptable results were obtained by using polyethylene glycol (molecular weight, 6,000)-concentrated and sucrose density gradient-purified extracts. Figure 1 illustrates the distribution of antigenic material in the density gradient. Most of the ELISA-reactive antigen was found near the 30%-50% sucrose interface (fractions 5 through 8), with a second light peak at the 0%-30% sucrose interface (fractions 9 through 12). The titers shown were obtained by using a pool of sera positive by SN and diluted 1:50. When a pool of negative sera was used, only a minimal reaction was observed with
-
SAMPLE
|
I...
QC
4040 _
20 10
6
7
9
10
11
12
13
Fractions FIG. 1. Density gradient separation of IBR virus-infected (circles) and noninfected (squares) cell extracts concentrated by polyethylene glvcol (molecular weight, 6,000) precipitation. Both extr acts were tested bv uising a pool of 10 sera positive by SN (closed symbols) and a pool of 10 negative sera (open symbols,).
VOL. 10, 1979
ELISA FOR IBR SEROLOGY
TABLE 1. ELISA-IBR and SN titers of 20 paired sera Titer Animal no.
SN
Serum 1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 1979, p. 633-636 0095-1137/79/1 1-0633/04$02.00/0
Enzyme-Linked Immunosorbent Assay for Serology of Infectious Bovine Rhinotracheitis Virus Infections PIERRE PAYMENT,`* ROBERT ASSAF,2 MICHEL TRUDEL,' AND PAUL MAROIS2 Centre de Recherche en Virologie1 and Centre de Recherche en Medecine Vitrinaire,2 Institut ArmandFrappier, Ville de Laval, Quebec, Canada H7N 4Z3
Received for publication 7 August 1979
An enzyme-linked immunosorbent assay was developed to detect antibodies to infectious bovine rhinotracheitis virus in bovine sera. The test used viral antigen purified in a sucrose density gradient, and the antibody titers detected were up to 100 times higher than those seen in the serum neutralization test. When 250 sera were tested by those two tests, 146 were positive and 49 were negative in both tests. Only 1 serum was negative in the immunoassay test and positive in the neutralization test. However, 54 sera negative in the neutralization test were positive in the immunoassay test. These positive results appear to be in agreement with the increased sensitivity of the enzyme immunoassay. The measurement of antibodies to infectious bovine rhinotracheitis (IBR) virus after a natural infection or after vaccination is usually based on the serum neutralization (SN) test (2). In 1975 we reported (1) that immunofluorescence was a suitable and faster alternative. More recently the development of new immunoenzymatic assays for several animal infections (4) prompted us to investigate an enzyme-linked immunosorbent assay (ELISA) test for the measurement of antibodies in certain viral infections. We have already reported its successful use for the serology of pneumonia virus of mice (3), and we are now reporting its use for the serology of IBR virus, using semipurified virus as the antigen. MATERIALS AND METHODS Virus and cell culture. Primary and secondary bovine testis cell cultures were grown in lactalbumin hydrolysate medium containing 4% fetal bovine serum. The Colorado strain (passage 36) of IBR virus was used at 1,000 50% tissue culture infective doses to infect the confluent monolayers. Lactalbumin hydrolysate medium without fetal bovine serum was used as maintenance medium. When the cytopathic effect involved 90% of the monolayer, about 48 h later, the cultures were frozen and thawed (-70°C, 37°C) three times. Viral antigen preparation for ELISA. Frozen and thawed infected or uninfected cell culture fluids were adjusted to 1.0 M NaCl and 8% polyethylene glycol (molecular weight, 6,000). The mixture was agitated for 2 h at 4°C, and the precipitate was then recovered by centrifugation at 3,000 x g for 30 min. The pellet was resuspended in saline to 'Ao of the original volume. The concentrated viral and control preparations were sonicated for 1 min at 100 W/cm2, using the
microprobe of a model 1510 Bransonic sonicator. Five milliliters of this material was added to 15 ml of NTE buffer (0.15 M NaCl, 0.01 M tris(hydroxymethyl)aminomethane-hydrochloride, 0.001 M ethylenediaminetetraacetate, pH 7.8) and placed on a discontinuous sucrose gradient formed by layering 10 ml of 30% (wt/wt) sucrose and 10 ml of 50% sucrose in NTE buffer in a 40-ml tube. This preparation was centrifuged at 25,000 rpm in an SW-27 rotor for 2 h. Two-milliliter fractions were recovered. The 30%-50% sucrose interface contained the viral antigen. The same fractions from uninfected material were used as control antigen. ELISA procedure. The ELISA assay was set up as described by Voller et al. (5), using Cooke Engineering microtiter 96-well (round bottom) hemagglutination plates. The antigens were diluted in 0.05 M glycine buffer at pH 10. The antigen (0.2 ml per well) was left overnight at 4°C. Excess antigen was removed the next morning by three 1-min washings with distilled water containing 0.05% Tween 80. Sera were diluted in phosphate-buffered saline containing 0.05% Tween 80. Appropriate wells received 0.2 ml of the dilutions, and the plates were incubated for 2 h at 37°C in a humidified incubator. Excess serum was removed by washing with distilled water-0.05% Tween 80. Immunoglobulin G antibodies were detected by using a commercial preparation of goat immunoglobulin G anti-bovine immunoglobulin G conjugated to peroxidase (Miles Laboratories, Inc., Elkhart, Ind.). The conjugate was diluted 1:1,500 in phosphatebuffered saline-0.05% Tween 80 containing 2% egg albumin (Sigma Chemical Co., St. Louis, Mo.). Bovine serum albumin was not used because it contains 2 to 5% bovine immunoglobulins, which reacted with the conjugate and lowered its activity. The optimal dilution of the conjugate was determined as described by Voller et al. (5). The conjugate (0.2 ml per well) was allowed to react for 2 h at room temperature, and the surplus was removed by washing as described earlier. Freshly prepared 5-aminosalicylic acid at 2 mg/ml, pH
633
634
PAYMENT ET' AL.
J. CLIN. MICROBIOLJ.
6.0, with 0.005% hydrogen peroxide was used as the substrate. The plates were read visually after 60 min at room temperature, and the endpoint was determined as the highest dilution with an above-background reaction. Serum samples. Paired sera were obtained from animals suspected of having an IBR virus infection. The first sera were obtained as soon as the symptoms developed, and the second sera were taken about 3 weeks later. Individual sera were selected among those submitted to us for various analyses. All sera were kept at -20°C. SN test. The SN test was performed as described previously (1).
fractions at the 30%-50% interface. When the same sera were used to evaluate a gradient containing an uninfected-cell extract, there were no specific reactions. Antigen preparations were used at an optimal dilution of 1:200 to 1:500 as determined by checkerboard titration with known positive and negative sera. Specificity of ELISA-IBR. Paired sera from 20 animals were tested by the SN and the ELISA-IBR tests (Table 1). With 17 pairs the results were in agreement, but in three animals with high SN titers in their first sera, ELISAIBR did not detect the slight increases seen by SN. All sera negative in the SN test had ELISAIBR titers lower than or equal to 1:160. The highest ELISA-IBR titer was 1:10,240. Correlation between SN and ELISA-IBR tests. The sensitivity of the ELISA-IBR test was determined by using 250 sera with known SN titers (Table 2). For 195 (78%) sera there was complete agreement: 146 (58%) were positive and 49 (20%) were negative by both tests. Only 1 serum (0.04%) positive by SN was negative by ELISA-IBR, but the SN titer was borderline, at 1:8. In 54 (22%) sera ELISA-IBR detected antibody titers ranging from 1:10 to 1:640, with the largest number of sera having a titer of 1:40. When individual results were plotted on a graph correlating the two tests (Fig. 2), the high sensitivity of the ELISA-IBR test was revealed. Two large groups of ELISA-IBR results are
RESULTS Viral antigen. Unpurified cell extracts were not satisfactory antigens for the ELISA test. For undetermined reasons, very high nonspecific reactions were observed even with negative sera on noninfected-cell extracts. Acceptable results were obtained by using polyethylene glycol (molecular weight, 6,000)-concentrated and sucrose density gradient-purified extracts. Figure 1 illustrates the distribution of antigenic material in the density gradient. Most of the ELISA-reactive antigen was found near the 30%-50% sucrose interface (fractions 5 through 8), with a second light peak at the 0%-30% sucrose interface (fractions 9 through 12). The titers shown were obtained by using a pool of sera positive by SN and diluted 1:50. When a pool of negative sera was used, only a minimal reaction was observed with
-
SAMPLE
|
I...
QC
4040 _
20 10
6
7
9
10
11
12
13
Fractions FIG. 1. Density gradient separation of IBR virus-infected (circles) and noninfected (squares) cell extracts concentrated by polyethylene glvcol (molecular weight, 6,000) precipitation. Both extr acts were tested bv uising a pool of 10 sera positive by SN (closed symbols) and a pool of 10 negative sera (open symbols,).
VOL. 10, 1979
ELISA FOR IBR SEROLOGY
TABLE 1. ELISA-IBR and SN titers of 20 paired sera Titer Animal no.
SN
Serum 1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
