Veterinary Microbiology, 22 (1990) 1-10 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
An E n z y m e - l i n k e d I m m u n o s o r b e n t Assay (ELISA) for Antibodies to B o v i n e Viral D i a r r h e a Virus P.J.K. DURHAM* and L.E. HASSARD
Department of Veterinary Microbiology, Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Sask. S7N OWO (Canada) (Accepted 12 September 1989)
ABSTRACT Durham, P.J.K. and Hassard, L.E., 1990. An enzyme-linked immunosorbent assay (ELISA) for antibodies to bovine viral diarrhea virus. Vet. Microbiol., 22: 1-10. A single dilution enzyme-linked immunosorbent assay (ELISA) was developed to detect antibodies to bovine viral diarrhea (BVD) virus in cattle sera. Viral antigen (NADL strain) was grown in a pig kidney cell line (PK15), and after removal of nuclear debris, was purified by ultracentrifugation through a potassium tartrate cushion. Antigen grown in embryonic bovine tracheal epithelial cells was also satisfactory. The test used a high salt buffer to minimize nonspecific reactivity, polyethylene glycol to enhance the reaction, and Protein G as the labelling agent. Comparative testing with the virus neutralization test (VNT) showed the ELISA results to have a high level of correlation with the VNT titers (r = 0.83). In vaccinated animals the ELISA detected antibodies earlier than the VNT. All animals sampled from a BVD-free herd were negative for BVD antibody. The single dilution test showed close agreement (r=0.84) with ELISA values obtained using a serial dilution technique, and also proved to have a high level of reproducibility. The test procedures were relatively easy to carry out, and were economic in their use of materials.
INTRODUCTION B o v i n e viral d i a r r h e a ( B V D ) virus is p r e s e n t in cattle in m a n y areas o f t h e world, a n d is i n c r i m i n a t e d as a cause o f a b o r t i o n , d i a r r h e a a n d m u c o s a l disease. A r a n g e of serological p r o c e d u r e s s u c h as virus n e u t r a l i z a t i o n , c o m p l e m e n t f i x a t i o n a n d gel-precipitin tests h a v e b e e n u s e d to d e t e c t t h e p r e s e n c e of B V D viral antibodies, b u t t h e y suffer f r o m a n u m b e r of d i s a d v a n t a g e s . C o n s e q u e n t l y several e n z y m e - l i n k e d i m m u n o s o r b e n t a s s a y s ( E L I S A ) ( H o w a r d et al., 1985; C h u et al., 1985; B o c k et al., 1986; L i a u w a n d Eugster, 1986; J u n t t i et al., 1987; J u s t e w i c z et al., 1987; C o q u i n e a u et al., 1988) have b e e n developed r e c e n t l y in *Present address: c/o B.P.P.H. Complex, P.O. Box 79, Yogyakarta, Indonesia.
© 1990 Elsevier Science Publishers B.V.
P.J.K. DURHAM AND L.E. HASSARD
an effort to produce a simpler, faster and more accurate test. The reported procedures still have some limitations, such as elaborate antigen preparation protocols, relatively low antigen yields and inadequate accuracy and evaluation. The present paper describes an ELISA procedure which was developed and found effective in this laboratory. MATERIALS AND METHODS
Virus culture The NADL strain of BVD virus (Americal Type Culture Collection, Rockville, MD ) was cultured in embryonic bovine tracheal epithelial (EBTr) cells, and caused typical cytopathic effect (CPE). The culture medium consisted of Earle's minimal essential medium (Gibco, Grand Island, NY) with 1% nonessential amino acids (Gibco, Grand Island, NY), 1% BVD antibody-free fetal calf serum (FCS) and antibiotics. High titer virus stocks were clarified by centrifugation at 600 g and stored at - 80 ° C. Direct inoculation of the NADL strain of BVD virus into a pig kidney cell line (PK15) (ATCC) failed to establish an infection detectable by cytopathic effect, immunofluorescence or backpassage into EBTr cells. Addition of BVD (NADL) infected EBTr cells into PK15 cells initiated successful infection, though growth was noncytopathic and initially detected only by backpassage into EBTr cells. The infected PK15 cells were cultured in medium as above but using 5% FCS. The cells were serially passaged 20 times, at which stage they showed a high level of virus content by fluorescent antibody test. Uninfected cultures were retained for control purposes.
Virus neutralization test This was carried out following a standard microtiter procedure (Rossi and Kiesel, 1971 ) using doubling dilutions of heat-inactivated sera, routine controls and 100 TCIDs0 of virus. The titre was expressed as the reciprocal of the serum dilution that completely or nearly completely inhibited viral CPE.
Antigen preparation for ELISA EBTr-derived antigen EBTr cells were inoculated with high titer stocks of virus and harvested when the cytopathic effect (CPE) was well advanced. Following three cycles of rapid freezing and thawing, the mixture of cell debris and medium was sonicated at 20 kHz for 60 s in an ultrasonic processor (Sonics and Materials, Danbury, CT) using a 70% pulse mode. The suspension was clarified by centrifugation at 600 g for 20 rain, and ultracentrifuged at 100 000 g for 3 h through
AN ELISA FOR ANTIBODIES TO BVD VIRUS
a 16% potassium tartrate cushion. The pellet was resuspended in a small volume of phosphate buffered saline ( P B S ) , and stored in aliquots at - 8 0 ° C . Uninfected cell cultures were processed similarly for use as controls. The antigens were thawed and briefly sonicated in an ultrasonic bath (Bransonic Cleaning Equipment Company, Shelton, CT) before use.
PK15-derived antigen BVD infected PK15 cells (passage level > 20) were grown to confluence in 75 cm 2 flasks. The medium was then removed and held at 4 ° C. The monolayer was washed and the cells detached using a trypsin-versene mixture, after which a small volume of fetal-calf serum was added to inactivate the trypsin. The cell suspension was combined with the previously removed medium and centrifuged at 600 g. The cell pellet was then suspended in deionized water (2 ml per flask). The cells were allowed to swell for 15 min at 4 ° C and then processed in a Dounce homogenizer (Fisher Scientific, Edmonton, AB). The suspension was centrifuged for 15 min at 225 g in a conical centrifuge tube to deposit the nuclei. The supernatant was recovered, diluted 4-fold in P B S and sonicated at 20 kHz for 60 s using a 70% pulse mode. It was then ultracentrifuged over a cushion of 16% potassium tartrate for 2 h at 100 000 g. The pellet was rinsed gently with PBS, resuspended in P B S (1 ml per flask), sonicated as above, and stored at - 80 ° C. Uninfected cultures were processed similarly for use as control antigen. BVD-infected cells grew slightly slower than uninfected cells, making it necessary to allow both cultures to reach full confluence before harvesting. Optimal antigen and cell control concentration was determined by titration against a standard-positive serum and a negative-control serum.
Sera (a): 74 bovine sera that were negative to the BVD V N T and 105 VNTpositive sera were submitted for diagnostic purposes from a number of different properties in the field. The sera were stored at - 18°C. (b): Five BVD seronegative cows were vaccinated with two doses of a killed BVD vaccine (Triangle 1, Fort Dodge Laboratories, Fort Dodge, IA) 14 days apart, and with a further 5-fold dose 7 days later. A total of 50 serum samples were collected at intervals before and after vaccination. (c): 30 serum samples from a BVD-free herd were also tested. The herd had been held in isolation over a 5-year period, and was considered BVD virus-free on the basis of negative results in repeated virus isolation and V N T tests over this time. (d): Positive control serum for the test was produced in a cow by vaccination with the killed vaccine, followed by two intramuscular injections of 106 TCIDso
P.J.K. DURHAM AND L.E. HASSARD
of a local isolate of BVD virus 1 week apart. Fetal calf serum that was VNT negative for BVD antibody was used as the negative control.
Indirect ELISA protocol The coating buffer was 0.01 M phosphate buffer (pH 8). The working buffer consisted of 0.77 M NaC1 in 0.01 M phosphate buffer (pH 7.2) with 0.05% Tween 20 (BDH Chemicals, Toronto, ON) and 0.05% gelatin. The wash fluid was distilled water containing 0.05% Tween 20. Affinity purified, peroxidaselabelled protein G (Zymed Laboratories, San Francisco, CA) was used to detect bound antibody. The substrate mixture consisted of 0.012% H202 and 0.004 M ortho-phenylenediamine (OPD) (Sigma, St Louis, MO) in citrate-phosphate buffer (pH 5). Virus and cell control antigen were thawed and briefly sonicated in the ultrasonic bath, diluted in coating buffer (usually to 1/60), and distributed in 100-pl volumes to alternate rows on flat-bottom polystyrene plates (Linbro/ Titertek EIA plates, Flow Laboratories, McLean, VA). The plates were incubated overnight at 4 ° C, washed three times with a mechanical plates washer (Biotek Instruments, Winooski, VT), dried at 37°C for 15 min, and stored in plastic bags at 4°C or - 1 8 ° C . For the test, 100 pl of sera was added at 1/50 dilution in working buffer to duplicate sets of virus and control wells. Reference-positive serum was similarly diluted and added to four sets of wells, and negative serum to two sets on each plate, leaving two sets of wells as plate controls. The plates were incubated at 37 ° C for 2 h, then washed three times and tapped dry. Conjugate was added to each well at an optimal concentration (usually 1/5000) in working buffer containing 4% polyethylene glycol (PEG 8000) (BDH Chemicals, Toronto, ON ) and incubation at 37 ° C for 2 h. Substrate ( 150 #1) was then added to each well, and the plates were incubated in the dark for about 7 min at room temperature. The reaction was stopped by adding 50 ~1 of 5 M HC1 to each well. The optical density (OD) of the plates was read in a plate reader (Titertek Multiskan, Flow Laboratories, McLean, VA) at 492 nm, and the results assessed via a coupled microcomputer. The reactivity of the sera was determined as the OD of wells with BVD viral antigen minus the OD of wells with cell control antigen. The final results were expressed in units as follows: mean net OD of test s e r u m - mean net OD of FCS mean net OD of positive standard s e r u m - m e a n net OD of FCS × 100 End-point titration assays were carried out similarly, except that sera were serially diluted and added to the BVD viral antigen and cell control antigen wells. The results were expressed as the reciprocal of the serum dilution that produced a virus antigen/control antigen ratio of > 2. A number of parameters were examined during the development of the test.
AN ELISA FOR ANTIBODIES TO BVD VIRUS
These included various concentrations of the blocking agents gelatin, BSA and casein, different concentrations of Tween 20 and NaC1 in the working buffer, and the use of PEG in the conjugate buffer. Peroxidase-labelled Protein G was also compared with affinity purified, peroxidase-labelled goat antibovine globulin (Kirkegaard and Perry Laboratories, Gaithersburg, MD). RESULTS
Antigen production The NADL strain of BVD virus required repeated passaging in PK15 cells before a high proportion of cells became infected (as judged by immunofluorescence). High yields of viral antigen were obtained in PK15 cells after 20 passages. The virus infection in PK15 cells appeared to be cell associated, as the culture fluids showed no infectivity for EBTr or PK15 cells. ELISA tests using nucleated and non-nucleated PK15 antigens gave a high level of agreement (r=0.98) with the majority (89%) of 235 tested sera. A proportion of sera (11% ) gave a high level of background reactivity when tested with the nucleated PK15 antigen. This sometimes induced false-positive results that were not confirmed when these sera were tested using the non-nucleated PK15 antigen, EBTr antigen or the VNT. Non-nucleated PK15 and EBTr antigens showed a high level of agreement (r= 0.97) in tests on 20 sera. Although EBTr antigen was satisfactory, non-nucleated PK15 antigen was considered to be more effective as it showed less background reactivity in the cell control wells, and gave higher values with the more positive sera. Nonnucleated PK15 antigen was therefore chosen for routine use. Removal of the cellular debris by use of the potassium tartrate cushion considerably reduced background reactivity. One 75 cm 2 flask of infected pK15 cells could produce sufficient antigen to coat 650 wells.
ELISA reagents The reactivity of BVD antigens was considerably improved by incorporation of PEG into the ELISA working buffer. The high salt concentration in the working buffer reduced background reactivity. The concentrations of Tween 20 and gelatin chosen were found to be optimal for the present test. Protein G was superior to the antispecies globulin in this ELISA system, giving greater discrimination between viral antigen and cell control antigen, and producing less background reactivity.
P.J.K. DURHAM AND L.E. HASSARD
Comparison of single dilution and end-point dilution ELISA Comparison of the single dilution assay with the end-point assay on 19 sera showed a high level of correlation (r = 0.84) between results.
Comparison of ELISA with V N T The ELISA with non-nucleated PK15 antigen gave good separation of 129 VNT-positive and 106 VNT-negative sera. A cut-off value of 10 was chosen for TABLE 1 Comparison of single dilution E L I S A with V N T E L I S A result
Virus neutralization test result
Positive s Negative
125 (a) 3 (c)
3 (b) 102 (d)
1Titer equal to or greater than 1/2. 2Activity greater than 10 units. Relative sensitivity (a/b + d) = 97.6% Relative specificity (d/b + d) = 97.1% Predictive value (a/a + b ) = 97.6% 120
SNT T I T R E
Fig. 1. Comparison of the single dilution E L I S A with the virus neutralization test, using 129 V N T positive sera and 106 V N T negative sera.
AN ELISA FOR ANTIBODIES TO BVD VIRUS
64LU IZ I-I--