J Vet Diagn Invest 2:191-196 (1990)

Comparison of a radioimmunoprecipitation assay to immunoblotting and ELISA for detection of antibody to African swine fever virus Carlos Alcaraz, Maribel De Diego, Maria J. Pastor, José M. Escribano Abstract. A radioimmunoprecipitation assay (RIPA) has been developed for detection of antibody to African swine fever virus (ASFV) and compared with the immunoblot assay with regard to sensitivity and specificity. Two hundred seven field sera, obtained from pigs in Spain from different geographic areas between 1975 and 1986, that were positive by ASFV enzyme-linked immunosorbent assay (ELISA) were also analysed by immunoblot assay and RIPA. By serum dilution experiments, the RIPA appeared at least as sensitive as the ELISA and immunoblotting tests, although ELISA and RIPA detected antibodies to ASFV earlier in natural infection than did the immunoblot assay, as disclosed by animal inoculation studies. The most antigenic ASFV-induced proteins in natural infection detected by RIPA were the viral proteins p243, p172, p73, p25.5, p15, and p12 and the infection proteins p30 and p23.5. In the immunoblot assay, the proteins that were most reactive with the same sera were the viral protein p25.5 and the infection proteins p30, p25, and p21.5. Only 1 serum, from an animal infected with ASFV, was negative by immunoblot assay but showed a positive result by RIPA. A modification of conventional RIPA was performed using a dot transference of immunoprecipitated proteins to a nitrocellulose filter. This modification simplified the conventional RIPA procedures by eliminating the electrophoresis of immunoprecipitated proteins without affecting sensitivity and specificity. The ease of use, specificity, and the sensitivity comparable to that of the immunoblot assay make the RIPA a useful confirmatory assay for sera that yield conflicting results in other ASFV antibody assays.

African swine fever virus (ASFV) is an icosahedral

tation. Additionally, many sera with low ELISA titers

cytoplasmic deoxyvirus, tentatively classified as an iridovirus,15 that infects porcine species. The African

are negative by antibody immunofluorescence, but they react with at least 2 proteins, p25.5 and p30, in immunoblotting.9 In the present study, we compared the immunoblotting and RIPA tests for confirmation of ELISApositive sera. We analyzed the sensitivity of the techniques, and we discuss the ASFV-induced proteins and how they affect the results in each test system.

swine fever (ASF) disease is one of the most important problems of the pig industry in affected countries. Acute, subacute, and inapparent carrier forms are found in nature. In most cases, infection is diagnosed by the presence of antibody because of the incidence of strains with reduced virulence that results in low mortality.5,11 The most widely employed technique for the diagnosis of ASF subacute outbreaks and inapparent carriers is the enzyme-linked immunosorbent assay (ELISA).3 In infectious diseases where diagnoses are based on antibody determination, confirmatory tests are necessary. These techniques are based, in general, on the detection of specific proteins of the infectious agent that react with the test sera. The most common methods are immunoblotting12 and radioimmunoprecipitation (RIPA).10,14 In ASF diagnosis, positive ELISA results must be routinely confirmed because of the sanitary importance in both ASFV-affected and -free countries. In many diagnostic laboratories, antibody immunofluorescence has been replaced by the immunoblot technique, which provides increased sensitivity and objective interpre-

Materials and methods Cells, viruses, and Sera A Spanish strain of the ASFV isolated in 1970 (E70) was adapted to grow in monkey stable (MS) cell line1 and used for antigen production after 48 passages (E70 MS 48). A second Spanish strain of ASFV isolated in 1975 (E75) was attenuated by 4 passages on CV1 (ATCC, CCL 70) cell line (E75 CV14) and used in animal studies. An experimental pig inoculated with the attenuated ASFV E75 CV148 was bled daily and tested by the RIPA, ELISA, and immunoblot assay. The serum on postinoculation day (PID) 20 was used as hyperimmune ASFV serum. Two hundred seven field pig sera with values of absorbances >0.3 O.D. in ASFV-ELISA (arbitrary value considered to discriminate positive and negative sera, using the sera at the screening dilution of 1:30) were obtained from different geographic areas in Spain between 1975 and 1986. Two hundred field sera were tested by RIPA and immunoblot assay using 1:100 and 1:30 dilution, respectively. Seven sera were poorly preserved and were tested by ELISA, RIPA, and immunoblot assay. A second bleeding from these 7 animals

From the Instituto National de Investigaciones Agrarias, Departamento de Sanidad Animal, Embajadores 68, 28012 Madrid, Spain. Received for publication December 3, 1989. 191

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

192

Alcaraz et al.

was obtained and was similarly tested. Serum from 2 animals infected with foot-and-mouth disease virus (FMDV) and transmissible gastroenteritis virus (TGEV) were used as specificity control sera at the same dilution as ASFV sera in each technique. A negative ASFV serum was used as a negative control for the ELISA, RIPA, and immunoblot assay. Isotopic labeling of infected cells The MS cells inoculated at a multiplicity of infection factor of 1 were incubated for 24 hr and labeled with 35S-methioninea (800 Ci/mmol) in Eagle minimal essential medium lacking methionine and supplemented with 1% calf serum. Infected cells were pulse labeled for 2 hr with 200 µCi/ml of isotope and harvested immediately. Preparation of cytoplasmic soluble (CS-P) antigen The CS-P antigen was prepared as described.3 After isotopic labeling, the ASFV-infected MS cells were harvested and sedimented. The pellet was washed with 0.067 M sucrose in 5 mM Tris-HCl, pH 8, for 10 min at 0 C, lysed by adding nonidet P40 detergentb at a final concentration of 1% (w/v), and the nuclei were pelleted at 1,000 x g. The cytoplasmic fraction was treated with 2 mM ethylenediaminetetraacetic acid (EDTA) and 0.05 M ß-mercaptoethanol, and after 15 min at 25 C, the mixture was centrifuged at 100,000 x g for 1 hr at 4 C over a 20% (w/w) sucrose bed in 50 mM TrisHCl, pH 8. The fraction above the sucrose layer was removed and used as cytoplasmic soluble antigen. Enzyme-linked immunosorbent assay The ELISA was performed as described.7 Cytoplasmic soluble antigen,3 at optimal concentration determined by pretitration, was used to coat microtiter plate wells. One hundred microliters of the 1:40 dilutions of test sera were added and incubated for 1 hr at 37 C. After washing the wells, protein Ac conjugated with peroxidase was added. The plates were reincubated for 1 hr at 25 C and, after being washed again, were incubated for 10 min at 25 C with Orthophenylenediamined as substrate. The reaction was stopped by addition of 100 µ1 of 3 N H 2SO4. Finally, the reactions were read spectrophotometrically at O.D.492; sera with absorbance values >0.3 O.D. were considered to be positive. Most of the seven sera were poorly preserved and gave nonrepetitive ELISA values among assays; 4 sera had an O.D. variation >20% in 3 different assays. Immunoblotting assay Immunoblotting was performed as described.9 Cytoplasmic soluble antigen proteins3 were resolved in 17% acrylamide gels and transferred to a nitrocellulose filter.” The filter was cut into strips that were then incubated with 2% nonfat dry milkf as blocking solution for 15 min and reacted with test sera diluted to 1:30 for 30 min. The immunocomplexes were detected using peroxidase-labeled protein A of Staphylococcus aureus and 4-Chloronaphtold as substrate. Radioimmunoprecipitation assay, polyacrylamide gel electrophoresis (PAGE), and dot blot Radioimmunoprecipitation assay was performed as described4 with modifications. A CS-P antigen (4 x 105 counts

per min [cpm] of radioactivity) was prepared in immunoprecipitation buffer (0.5% NP40, 2 mM EDTA, 0.5 M NaCl, 0.2% sodium dodecyl sulfate [SDS], in 50 mM Tris-HC1 pH 8) and then incubated with 10 µl of serum for 1 hr at 25 C. Staphylococcus aureus, g washed twice with the same immunoprecipitation buffer lacking NaCl and SDS, was diluted to a concentration of 20% (w/v) and used for the adsorption of immunocomplexes during 15 min at 25 C. The Staphylococcus was washed twice with washing buffer (0.5% NP40, 2 mM EDTA, 0.1% NaCl, 0.2% SDS, 5 mM ß-mercaptoethanol, in 50 mM Tris-HC1 pH 8) and once with phosphate buffered saline (PBS), pH 7.2. Finally, antigen-antibody complexes were detached from the Staphylococcus by resuspending the pellet in 5 mM Tris-HCl, pH 7, 2% SDS, and 0.05 M ß-mercaptoethanol and incubating for 5 min at 60 C before analysis by electrophoresis or dot blot. All procedures for electrophoresis and fluorography were as previously described.4 Conditions of transference to nitrocellulose filters were the same as those described above for immunoblot. Dot blot was carried out by filtering samples through a nitrocellulose filter using a biodot vacuum manifold.e The filter was washed with PBS, dried, and exposed to X-ray film.h All the sera were analyzed by both electrophoresis and dot blot. Optimal conditions for RIPA The 35S-methionine-labeled CS-P antigen was used to establish the optimal conditions for RIPA. Three different aliquots of 8 x 105 cpm of 35S-antigen were immunoprecipitated by the hyperimmune serum and analyzed by SDSPAGE. To establish the minimum time of X-ray film exposure that provides clear patterns, the gel was divided into 3 parts and either dried, transferred to a nitrocellulose filter, or fluorographed. Sensitivity of RIPA To compare the sensitivity of the RIPA with that of the immunoblot, dot, and ELISA techniques, serial dilutions of the hyperimmune serum, starting at 1:100 in RIPA, dot, and ELISA and 1:200 in immunoblot, were tested.

Results Optimal conditions for RIPA and sensitivity of the technique

After 17 hours of exposure, the fluorographed gel showed a clear pattern of immunoprecipitated ASFV proteins. The other methods showed, after 72 hours of exposure, an immunoprecipitated protein pattern equivalent to a 17-hour X-ray exposure of the fluorographed gel (Fig. 1). The results of the comparative sensitivity of immunoblot and RIPA are shown in Fig. 2. The titer of hyperimmune serum with RIPA and ELISA was 1:6,400, and with the immunoblot it was 1:3,200. The RIPA demonstrated more sensitivity than immunoblot for antibodies against p243, p172, p73, and p12 (Fig. 2), whereas immunoblot showed more sensitivity

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

Detection of antibody to ASFV by radioimmunoprecipitation assay

193

Figure 1. Comparison of 3 different procedures for revealing radioimmunoprecipitated viral proteins after electrophoresis. A 72hour X-ray exposure of the gel after drying (A) or after the proteins were transferred to a nitrocellulose filter (B). The channels C and D represent the same radioimmunoprecipitation after fluorography at 17 and 72 hours of X-ray exposure, respectively.

Figure 2. Reactivity of the proteins with serial dilutions of the hyperimmune serum starting at 1:100 in RIPA (1) and 1:200 in immunoblotting (2). The radioimmunoprecipitated proteins by the serial dilutions (1:100-l : 12,800) of the hyperimmune serum were dot blotted (lowest panel).

for antibodies against p34, p30, p25 and p23.5 at higher dilutions (Fig. 2). When 35S-antigen aliquots, immunoprecipitated with serial dilutions of the hyperimmune serum, were dotted on nitrocellulose filter and exposed for 72 hours to X-ray film, the titer of the serum was 1:6,400. In animal inoculation studies, ASFV antibodies were detected as early as the 7th day postinoculation in ELISA and RIPA and 1 day later by immunoblot. The sequence of appearance of antibodies to the different ASFV-induced proteins detected by RIPA in ASFV infection is shown in Fig. 3. The first reacting protein by RIPA, at 7th day postinfection, was the p12.

different sera in RIPA could not be correlated with the ELISA O.D. values obtained with these same sera. At least for the major antigenic proteins, the sera with ELISA values >1.0 O.D. showed a similar protein pattern in RIPA to sera with ELISA values < 1.0 O.D. Figure 4 also shows the results obtained by dot blot after radioimmunoprecipitation of the different sera at the time of autoradiography. Consistent background was not detected in the 3 negative control sera tested by dot blot. By immunoblot assay, the 200 sera were confirmed as positive. Four sera, representative of ELISA values higher and lower than 1.0 O.D., are shown in Fig. 5. The most reactive proteins in natural infection by this technique were the viral protein p25.5 and the infection proteins p30, p25, and p21.5. Three of the 7 poorly preserved sera, all ELISA positive, were positive by RIPA, whereas only 2 of them reacted with ASFV proteins in immunoblot (Fig. 6). The second serum samples from these 3 pigs were positive by the ELISA, RIPA, and immunoblot techniques. Discussion In this study, the sensitivity and specificity of RIPA for the detection of antibodies to ASFV-induced proteins were compared with those of the conventional

Confirmation of ASFV-ELISA-positive field pig sera by RIPA and immunoblotting

The 200 well-preserved ELISA-positive field sera were positive by RIPA. The reactivities in RIPA of 15 of these sera, representating different ELISA O.D. values, are shown in Fig. 4. None of the 3 negative control sera showed either nonspecific or cross-reactions in the RIPA technique (Fig. 4). The most antigenic ASFVinduced proteins in natural infection by this technique were the viral proteins p243, p172, p73, p25.5, p15, and p12 and the infection proteins p30 and p23.5. Differences in the protein pattern obtained with the

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

194

Alcaraz et al.

Figure 3. Detection of ASFV antibodies by RIPA at different days after inoculation in a pig experimentally infected with the attenuated virus strain E75-CV14.

immunoblot and ELISA techniques.7,9 The sensitivity of RIPA was at least equal to that of immunoblot and ELISA for ASFV antibody detection. An RIPA differs from immunoblot analysis in at least 3 major criteria. First, the viral antigens are not subjected to denaturing conditions, allowing preservation of conformational epitopes. Second, the antigen-antibody interaction takes place in a fluid phase, allowing optimal exposure of epitopes and better opportunity for multivalent binding. Third, the high-molecular-weight proteins transfer poorly to nitrocellulose filters in immunoblot, preventing recognition by the antibodies. These features may explain some differences between the protein patterns obtained in RIPA and immunoblot, using either field sera or serial dilutions of ASFV hyperimmune serum. The RIPA detected viral structural proteins better than immunoblot, which only significantly detected the structural protein P25.5. Discrepancy between protein recognition by RIPA and immunoblot in comparative studies for other viruses has been described.14 However, differences inherent to the tech-

niques may also explain the detection of an early antibody response to proteins such as P12 and P73 by RIPA, prior to detection by immunoblotting and possibly ELISA.’ These results confirmed RIPA to be at least as sensitive as ELISA in the detection of ASFV infections in the early stages. The RIPA also detected antibodies against minor antigenic proteins earlier than did immunoblot; RIPA is probably superior in the detection of low-affinity immunoglobulins (resulting from the formation of more stable immune complexes) when the antigen is in so1ution.6,14 Correlations between ASFV proteins immunoprecipitated by a serum and the stage of infection, and the survival prognosis of the animal are difficult to assess. In the experimentally infected pig, the proteins p243, p172, p73, p34, p30, and pl2 induced antibodies soon after infection, whereas proteins such as p23.5 and p15 induced antibodies up to 12 days after infection (Fig. 3). In general terms, field pig sera reacted with most of the early and late antibody-inducing proteins, with the exception of p34, which was immunoprecipitated by only a small number of sera (Fig. 4). This could indicate that the sera correspond to animals in late stages of infection. The protein p12, the major ASFV-induced protein in infected cells,13 seems to be the first protein to induce antibodies in infected pigs. Experience with immunoblotting is extensive because it is used to confirm reactivity in the ELISA in Spain and in other affected countries.2 It is apparent that pigs with subacute infection, or inapparent carriers, lack sufficient levels of antibodies against ASFVinduced proteins to be consistently reactive in the currently used ELISA and immunoblotting tests. However, antibody immunofluorescence did not show sufficient sensitivity to confirm some ELISA-positive sera that could be confirmed by immunoblotting.9 Only in cases of early infection was immunoblotting less sensitive than RIPA and ELISA in ASFV antibody detection as disclosed by the experiments on the kinetics of appearance of antibodies in ASFV infection.’ Nonspecific reactions have not been detected by immunoblotting, nor have antibodies cross-reacted with epitopes of ASFV-induced proteins2 A false-negative result was obtained by immunoblotting of a poorly preserved serum sample. The poor state of the sample possibly altered its epitope recognition, and only reacted with virus-induced proteins, under favorable conditions provided by ELISA and RIPA. This result makes the use of RIPA advantageous, instead of immunoblotting, in the testing of poorly preserved sera. Dotting immunoprecipitated proteins onto nitrocellulose filters is a simple procedure that avoids the electrophoresis step in the RIPA technique. We suggest the use of this method for rapid confirmation screening or in laboratories where electrophoresis equipment is not available.

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

Detection of antibody to ASFV by radioimmunoprecipitation assay

195

Figure 4. Detection of ASFV antibodies by RIPA and dot blot in ELISA-positive field sera. The patterns of proteins shown in this figure were representative of 2 groups of sera showing ASFV-ELISA O.D. values < 1.0 O.D. (1-5) and > 1.0 O.D. (6-15). A negative control ASFV serum (16) and sera from 2 pigs infected with transmissible gastroenteritis virus of pigs (anti-TGEV) and with foot-and-mouth disease virus (anti-FMDV) were used as ASFV-negative control sera.

The RIPA technique presented here provides a valuable alternative to the confirmation of sera that give conflicting results by other ASFV antibody assays. Furthermore, this test may allow a more rapid detection of ASF-specific antibodies than immunoblotting after exposure to the virus, and it is also useful for the validation of new generations of screening tests. Acknowledgements We are grateful to M. Moyano and M. Sevilla for technical assistance. This work was supported by a grant from Instituto National de Investigaciones Agrarias, Madrid, Spain.

Sources and manufacturers a. Amersham International, Amersham, England. b. BDH Chemicals, Ltd., Poole, England. c. Materiales y reactivos, Laboratorios Llorente, Madrid, Spain. d. Sigma Chemical Co., St. Louis, MO. e. Bio-Rad Laboratories, Richmond, CA. f. Molico, Nestle, Esplugás de Llobregat, Barcelona, Spain. g. Calbiochem Corp., La Jolla, CA. h. Agfa-gevaert SA, Spain.

References Figure 5. Detection of ASFV antibodies by immunoblotting in ELISA-positive field sera. Serum numbers 3 and 5 correspond to sera with ELISA values < 1.0 O.D., and numbers 10 and 14 correspond to sera with ELISA values > 1.0 O.D.

1. Alcaraz C, Pasamontes B, Ruiz Gonzalvo F, Escribano JM: 1989, African swine fever virus-induced proteins on the plasma membranes of infected cells. Virology 168:406-408. 2. Escribano JM, Pastor MJ, Arias M, Sanchez-Vizcaino JM: 1990, Confirmation de sueros positivos a ELISA-Peste porcina africana mediante la tecnica de “Immunoblotting”. Utilization de

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

196

Alcaraz et al.

Figure 6. Seven poorly preserved sera, 4 of them did not give repeatable ELISA values between assays. Sera numbered 19 and 20 were confirmed as positive by RIPA after electrophoresis and dot blot (A) and by immunoblot (B), whereas serum 21 only reacted by RIPA.

3.

4.

5.

6.

7.

8.

las proteinas inducidas por el virus, con pesos moleculares comprendidos entre 23 y 35 kilodaltons, en el desarrollo de un kit de diagnostico. Med Vet 7: 135-141. Escribano JM, Pastor MJ, Stinchcz-Vizcaino JM: 1989, Antibodies to bovine serum albumin in swine sera: implications for false-positive reactions in the serodiagnosis of African swine fever. Am J Vet Res 50:1118-1122. Escribano JM, Tabards E: 1987, Proteins specified by African swine fever virus: V. Identification of immediate early, early and late proteins. Arch Virol 92:221-232. Hess WR: 1971, African swine fever virus. In: Virology monographs, ed. Gard S, Hallaner G, Meyer KF, vol. 9, pp. 1-3. Springer, Berlin. Matthews TJ, Langlois AJ, Robey W, et al.: 1986, Restricted neutralization of divergent human T-lymphotropic virus type III isolates by antibodies to the major envelope glycoprotein. Proc Nat1 Acad Sci USA 83:9709-9713. Pastor MJ, Arias M, Escribano JM: 1990, Comparative study between two different antigens used in African swine fever antibody detection by ELISA. Am J Vet Res (in press). Pastor MJ, Escribano JM: 1990, Evaluation of sensitivity of different antigenic and DNA-hybridization methods in African swine fever virus detection. J Virol Methods (in press).

9. Pastor MJ, Laviada MD, Sanchez-Vizcaino JM, Escribano JM: 1989, Detection of African swine fever virus antibodies by immunoblotting assay. Can J Vet Res 53:105-107. 10. Saah AJ, Farzadegan H, Fox R, et al.: 1987, Detection of early antibodies in human immunodeficiency virus infection by ELISA, western blot, and radioimmunoprecipitation. J Clin Microbiol 25:1605-1610. 11. Sanchez-Botija AC, Ordas A. Ruiz Gonzalvo F, et al.: 1977, Procedures in use for diagnosis of ASF. Agricultural research seminar on hog cholera/classical swine fever and African swine fever, Hannover. Comm Eur Corn, Eur 5904 EN, pp. 642-652. 12. Steckelberg JM, Cockerill FR: 1988, Serologic testing for human immunodeficiency virus antibodies. Mayo Clin Proc 63: 373-380. 13. Tabarés E, Martinez J, Martin E, Escribano JM: 1983, Proteins specified by African swine fever virus: IV. Glycoproteins and phosphoproteins. Arch Virol 77: 167-180. 14. Tersmette M, Lelie PN, van der Poel CL, et al.: 1988, Confirmation of HIV seropositivity: comparison of a novel radioimmunoprecipitation assay to immunoblotting and virus culture. J Med Virol 24:109-116. 15. Viñuela E: 1985, African swine fever virus. Curr Top Microbiol Immunol 116:151-170.

Downloaded from vdi.sagepub.com at GEORGIAN COURT UNIV on April 1, 2015

450578

JVDXXX10.1177/1040638712450578

Erratum Journal of Veterinary Diagnostic Investigation 24(4) 813 © 2012 The Author(s) Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1040638712450578 http://jvdi.sagepub.com

Corrigendum

Stegelmeier, BL, et al.: 2010, Experimental rayless goldenrod (Isocoma pluriflora) toxicosis in goats. J Vet Diagn Invest. 22: 570–577

In the article “Experimental rayless goldenrod (Isocoma pluriflora) toxicosis in goats” by Bryan L. Stegelmeier et al., the published mean body weight and the means and statistics of serum biochemistries were carried out on groups of 4 animals, not 3, as described in the Material and Methods section. The additional animal in each group was part of an auxiliary physiologic study and though the animals were dosed and treated the same, they were not necropsied and were not included in the histologic study. To correct this oversight, the corrected weight and chemistry table (shaded cells indicate corrected numbers) are listed below. The differences are minimal and do not alter the conclusions. In addition, reference 7 has been deleted. Material and Methods: “Fifteen, yearling, female Spanish goats weighing 29.4 ± 3.4 kg (mean ± standard deviation) were randomly divided into 5 groups with 3 animals per group.”

References: Reference 7 should be deleted Corrected Table 1. Selected mean serum biochemical data from groups of 3 goats dosed with rayless goldenrod (Isocoma pluriflora) to obtain benzofuran ketone doses of 0, 10, 20, 40, and 60 mg/kg body weight for 7 days.* Serum result (mean ± standard deviation) Serum test (reference range†) Creatinine kinase (< 350 U/l)         Cardiac troponin-I (

Comparison of a radioimmunoprecipitation assay to immunoblotting and ELISA for detection of antibody to African swine fever virus.

A radioimmunoprecipitation assay (RIPA) has been developed for detection of antibody to African swine fever virus (ASFV) and compared with the immunob...
161KB Sizes 0 Downloads 0 Views