J Vet Diagn Invest 3:3-9 (1991)
Detection of IgM antibodies against canine distemper virus in dog and mink sera employing enzyme-linked immunosorbent assay (ELISA) Merete Blixenkrone-Moller, Ib Rode Pedersen, Max J. Appel, Christian Griot Abstract. An enzyme-linked immunosorbent assay (ELISA) for the detection of IgM antibodies against canine distemper virus (CDV) in canine and mink serum is described. The diagnostic potential of this technique was evaluated by analyzing sera from natural or experimental infections in dog and mink and negative control sera. These results were compared with results obtained in the developed CDV IgG ELISA and in the virus neutralization test. The IgM test, which requires only a single serum specimen, is a useful method for diagnosing current or recent CDV infections in dog and mink.
A virus-specific serum antibody response after induced infection in dogs with virulent or attenuated vaccine strains of CDV may be detectable by employing an indirect immunofluorescence technique.2–4 The ELISA techniques have been described for the demonstration of the class-specific immunoglobulin response during CDV infections.6,10,14,25 However, it is not known if the techniques can distinguish between known infected patients and healthy controls. The ELISA techniques have been developed for the detection of antibodies against the closely related measles virus, and these techniques, which we essentially based on measurements of single serum samples, are currently used for serodiagnosis of present or recent measles virus infections in humans.15,18,21,22 The principle of these ELISA techniques, in contrast to other CDV ELISA techniques, 10,14,25 is based on an immunological separation of from the test sample before the reaction with viral antigen. This capture technique eliminates competing antibodies and minimizes nonspecific reactions due to rheuma factors in human sera.11,17 The objectives of the present studies were to set up an ELISA technique based on examination of single serum samples for the detection of antibodies against CDV in dog and mink sera and to evaluate the validity of the technique for diagnosis of present or recent CDV infections.
Canine distemper (CD) is a contagious systemic viral disease with often high mortality rates in dogs and other carnivores throughout the world. Canine distemper is caused by canine distemper virus a morbillivirus in the paramyxovirus family. An acute primary CDV infection may induce a specific antibody response. An antibody titer rise in paired serum samples is of decisive diagnostic importance. The detection of an increased titer of CDV antibodies employing the neutralization test,5 precipitation test,24 complement fixation test,13 cytotoxic test,20 or enzyme-linked immunosorbent assay (ELISA)14,25 is, however, not suitable for routine diagnostics because the titers are often high in the beginning of the clinical disease and may not rise further.2 Often in fatal cases of distemper, no detectable serum neutralizing 1,13 antibodies are found. In cases where a significant titer rise is demonstrated in suitable paired samples (taken with a 3-week interval), the result may have lost its importance because of the time factor. However, antibody of the class produced early in viral infection may indicate an ongoing or recent viral multiplication. For this reason, measurement of single serum samples has been investigated as an aid for viral diagnosis. From the Department of Veterinary Microbiology, Laboratory for Virology and Immunology, The Royal Veterinary and Agricultural University of Copenhagen, 13, Bülowsvej 1870 Frederiksberg C, Denmark (Blixenkrone-Moller), the Department of Medical Microbiology, University of Copenhagen, Panum 24/2, 3, Blegdamsvej 2200 Copenhagen N, Denmark (Pedersen), the James Baker Institute for Animal Health, New York State College of Veterinary Medicine, Cornell University, Ithaca, NY 14853 (Appel), and the Institute of Animal Neurology, University of Bern, Bremgartenstr. 109a, Postfach 2735, CH-3001 Bern, Switzerland (Griot). Received for publication August 4, 1990.
Materials and methods Preparation of CDV antigen. Vero cells were grown in stationary flasks (175 cm2) in Eagle’s minimum essential medium with 2% bovine fetal serum (BFS), 20 mM/liter L-glutamine, 2.2 g/liter sodium bicarbonate, and antibiotics. At 70-80% confluency, the monolayer cultures were infected at a multiplicity of infection (MOI) of 1, 0.1, or 0.01 (estimated 3
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
4
Blixenkrone-Møller et al.
by CPE in Vero cells) with the Onderstepoort strain of CDV isolated from a commercially attenuated CDV vaccine for mink. a After virus adsorption for 2 hr at 37 C, the cultures were incubated until pronounced CPE were observed (7296 hr postinoculation). Viral antigen was prepared by extraction of cell-associated virus. The culture medium was discarded, and after rinsing the cell layer in phosphate buffered saline (PBS) pH 7.2, 3 ml lysis buffer pH 8.4 (0.01 M Tris, 0.002 M ethylenediaminetetraacetic acid [EDTA], 0.2 M sucrose, 2% [v/v] Triton-X100, and 200 KIU/ml aprotinin) was added to each culture flask. Following freezingthawing once and sonication in an ice bath twice for 15 sec each, the cell lysate was clarified by 2 consecutive centrifugations (10 min, 12,100 × g), and the supernatant was centrifuged (20 min, 48,200 × g). The supernatant was used as antigen in the ELISA techniques and was stored at -40 C. Control antigen was prepared in the same way from noninfected Vero cell cultures. Production of antisera against CDV. Antisera against CDV were raised in 2 rabbits that had been immunized with purified Vero cell-grown virus. Virus was harvested by freezingthawing of infected cell cultures, and after clarification (10 min at 12,100 × g), the supernatant was centrifuged into a double layer sucrose cushion (30 and 60% w/w sucrose in TE-buffer pH 7.4 [0.01 M Tris, 0.01 M EDTA, and 0.1 M NaCl]) for 90 min at 120,000 × g at 4 C in a SW 28 rotor. b The interphase between the sucrose cushions was diluted in TE buffer, layered on top of a linear 20-60% w/w sucrose gradient in TE buffer, and centrifuged for 235 min at 155,000 × g at 4 C in a VTI 50 rotor. b Gradient fractions were collected by bottom unloading, and CDV antigen in the fractions was quantified by a modified antigen capture ELISA as previously described, 15 employing canine CDV antiserum and rabbit anti-canine IgG conjugated to horseradish peroxidase. c The rabbit antisera against CDV were used conjugated to peroxidase after purification of IgG and as “catching antibody”’ in the CDV antibody ELISA. Purification and enzyme conjugation. The IgG was purified from the rabbit antisera against CDV and conjugated to horseradish peroxidase as previously described.14 The conjugate contained 1.2 mg protein/ml. Examination of anti-IgM reagents. The following 3 rabbit antisera were tested in the IgM ELISA technique: anti-dog IgM, d anti-mink IgM,e and anti-human IgM (µ-chain specific).f ELISA techniques. The techniques illustrated in Fig. 1 were carried out in flat-bottom polystyrene microtest plates with 96 wells and 100 µ1 per well. Optimal conditions for incubation of the various reagents (time, temperature, stationary/orbital shaking) were established in preliminary experiments. Working dilutions for the reagents that made it possible to determine the minimal amounts of virus-specific antibodies detectable in serum samples were determined by checkerboard titrations. Preliminary ELISA results indicated that only IgG class antibodies were detectable using the antiIgG (H + L) peroxidase conjugate. All test samples were run in duplicate. A carbonate buffer pH 9.6 (15 mM Na2CO3, 35 mM NaHCO3, 3 mM NaN3) was used for dilution of catching antibodies. After every step of incubation with immunological reagents, the plates were rinsed 4 × for 3 min each in
washing buffer pH 7.2 (0.5 M NaCl, 2.7 mM KCl, 1.5 mM KH2PO 4, 6.5 mM Na2HPO 4· 2H2O, 1% [v/v] Triton-X100). Subsequent immunological reagents were diluted in dilution buffer, i.e., washing buffer supplemented with 0.01% w/v phenol red and 1% w/v bovine serum albumin. The substrate for the enzyme reaction was 8 mg phenylendiamine dihydrochloride (OPD) and 5 µ1 30% H O 2 2 per 12 ml 0.1 M citrate buffer pH 5.0. The reaction was stopped with 1 M H2SO 4. The color was read at 492 nm with a 620-nm reference wavelength on an ELISA reader.g The ELISA OD492 values (Evalues) of the serum samples were defined as the OD492 difference between the mean of 2 measures in wells with viral antigen and control antigen. To minimize false positive results, the cutoff values were established with a nonparametric method. 16,19 Virus neutralization test (VNT). The VNT was performed as described5 using flat-bottom microtest plates with 96 wells, and every serial dilution of test serum was examined in quadruplicate. Each test well was incubated for 1 hr at 37 C with 50 µ1 2-fold serum dilutions and 50 µ1 dilution of the Onderstepoort strain of CDV containing 5-56 TCID50. Fifty microliters of a Vero cell suspension (5 × 105 cells/ml) were then added to each well. After 2 hr, the supernatants in the wells were replaced with 200 µ1 Eagle’s medium supplemented with 2% BFS. The plates were sealed, and after an incubation period of 4-5 days at 37 C, the cultures were examined under the microscope for cytopathic changes and the titer was calculated ad modum Kärber.8 Patients and serum samples. One hundred thirty-one serum samples were taken from 126 dogs of various breeds and ages and 256 serum samples were taken from 120 farmed mink (6-52 wk of age). Samples were kept at -20 C. There were 10 paired samples from 5 Beagles before and after inoculation with virulent Cornell A75-17 strain of CDV and 43 samples from naturally occurring cases of canine distemper. These 43 samples were from acute cases of canine distemper (7 samples from unvaccinated dogs and 24 samples from dogs that had been vaccinated within 2-3 wk of a distemper outbreak in Northern Greenland in 19887), from 2 dogs in the convalescent phase after acute canine distemper, and from 10 postmortem cases of chronic, progressive inflammatory distemper encephalitis. One hundred twenty-four samples were from 26 mink 0-56 days after subcutaneous vaccination with an attenuated CDV vaccine,h and 71 samples were taken from 33 mink 0-56 days after inoculation of the virulent Snyder-Hill strain of CDV.i Negative control sera. Negative control sera from dog and mink were analyzed with the ELISA techniques. Ten samples were from unvaccinated sled dogs from Greenland before the canine distemper outbreak in 1988, 31 samples were from unvaccinated dogs from Iceland,j and 9 samples came from unvaccinated 14-wk-old Beaglesk raised for experimental purposes and kept isolated from other dogs. Another 28 canine samples from randomly selected Danish dogs that had not been vaccinated within the last 3 mo were also classified as CDV IgM negative control sera. Sixty-one negative mink control sera were from unvaccinated wild mink from farms with no distemper outbreaks during the last 8 yr. Furthermore, the 5 canine sera taken before inoculation of virulent CDV and 24 mink sera taken before vaccination or before
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
Figure 2. Determination of IgM ELISA OD492 values of rabbit anti-IgM used as catching antibody. A serial 2-fold dilution of antiIgM reacted with sera from CDV-infected dog and mink and with sera from uninfected dog and mink Figure 1. Principles of ELISA for CDV IgM antibody detection (A) and CDV IgG antibody detection (B).
inoculation with a virulent strain of CDV were classed among the negative control sera. Reference sera. The 2 positive-IgM reference sera were from a dog with an acute clinical CDV infection and from a vaccinated mink. A serum pool from vaccinated dogs and hyperimmune serum from mink served as positive references in the CDV IgG antibody ELISA.
Results Determination of the optimal anti-IgM concentration for the CDV IgM ELISA. Optimal preparations of CDV antigen for the ELISA techniques were obtained from cell cultures infected at MOI 0.1. Of the different anti-IgM reagents (anti-human, anti-dog, antimink) tested as capture antibodies in the IgM ELISA, anti-human IgM gave the best results employing canine as well as mink sera. Serial 2-fold dilutions of antihuman IgM used as catching antibody in IgM ELISA are shown in Fig. 2. The background was reduced 10% by including a blocking step with 1% rabbit serum in PBS after coating of the solid phase. The incubation conditions and dilutions of ELISA reagents that made it possible to detect minimal amounts of virus-specific antibodies in test samples are given. CDV IgM antibody ELISA procedure. The wells were coated with anti-human IgM diluted 1:200 for 1 hour at 37 C. Before the next incubation step the plates were blocked with 1% rabbit serum in PBS for 1.5 hours at 37 C. Dilutions (1:25) of serum samples were incubated overnight at 4 C. Then dilutions (150) of viral antigen and control antigen were added, and the plates were incubated at 37 C on an orbital shaker for 3 hours, followed by shaking with anti-CDV peroxidase conjugate (1:100 dilution) for 1.5 hours at 37 C. Finally the substrate reaction, carried out in the dark at 21 C, was stopped after 13 minutes.
CDV IgM antibody ELISA procedure. The wells were coated for 1 hour at 37 C with rabbit antisera against CDV (1:3,000 dilution). Viral antigen and control antigen (1:100 dilution) were added to the plates, incubated overnight at 4 C, then incubated with serum samples (canine sera in 1: 100 dilution, mink sera in 1:50 dilution) for 1 hour at 37 C, followed by incubation with anti-canine IgG conjugate for 1 hour at 37 C. The rest of the procedure was as described for the IgM ELISA. Reproducibility and sensitivity. The E-value of a serum sample was corrected in accordance with the E-value for a positive reference serum to minimize the day to day variation. The variation of the E-values was 18% for 8 parallel testings on 8 different days for IgM reference mink serum, with a mean of 1.069 (0.9541.187). Corresponding variation for IgG reference mink serum with a mean of 1.326 (1.400-1.071) was 12%. The sensitivity of the IgM ELISA technique, defined as the highest serum dilution with positive reaction, was examined employing serial 10-fold dilutions of 6 sera. The IgM antibodies were detected in dilutions up to 1:25,000 in dog sera and 1:2,500 in mink sera. Determination of cutoff values. To establish the cutoff values between positive and negative sera in the CDV IgM antibody ELISA, 121 canine serum samples (83 negative control samples, 36 samples from clinical infections, and 2 samples from convalescent dogs) and 103 mink serum samples (85 negative control samples and 18 samples from clinical infections) were examined (Fig. 3). The cutoff for negative E-values was established at 100% of the negative control group. Considering the limited number of sera in the negative control groups, the cutoff for positive E-values was set at 1.5 times the negative cutoff. The inconclusive values were defined between the positive and negative
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
6
Blixenkrone-Møller et al.
Figure 3. Determination of positive/negative IgM discrimination levels (dashed lines). A. ELISA OD 492 values for 38 canine sera There were 31 sera from acute clinical cases of canine distemper, 5 sera from experimental inoculation with a virulent strain of CDV, and 2 samples from convalescent dogs. There were 83 negative control sera B. ELISA OD492 values for 18 mink sera from acute clinical CDV infections after experimental inoculation. There were 85 negative control sera
cutoffs. Thus in the IgM ELISA E-values ≤0.27 were considered negative, 0.28-0.41 were inconclusive values, and >0.41 were considered positive (Fig. 3A). Equivalent values for mink sera were ≤ 0.30 (negative), 0.31-0.45 (inconclusive), and >0.45 (positive) (Fig. 3B). The true cutoff values seem to cover the same intervals within the accuracy of the measurement. The cutoff values for the IgG ELISA were established as described for the IgM ELISA. Examination of CDV IgM antibodies in sera from clinical CD V infections. Forty-four of 49 examined sera from acute clinical infections in dog and mink were IgM positive, and 2 samples from convalescent dogs after natural infections were IgM positive (Fig. 3). In chronic cases with central nervous system symptoms, we detected virus-specific IgM antibodies in 6 out of 10 cases (Fig. 4). Comparative study of sera in CDV IgM antibody ELISA, CDV IgG antibody ELISA, and VNT. Thir-
teen sera from acute clinical cases in dogs were found positive in both ELISA techniques, but in only 1 of these were CDV antibodies detected by VNT. In 7 serum samples from mink inoculated with a virulent strain of CDV, viral antibodies were detected in the IgM ELISA but not in the IgG ELISA. Thirty-six canine serum samples and 7 serum samples from mink that had been found negative in both ELISA techniques were also negative by VNT. Examination of the IgM response in mink inoculated with a virulent or attenuated strain of CDV. Thirtytwo serum samples from 18 mink that had been inoculated with the virulent Snyder-Hill strain of CDV were analyzed for IgM antibodies. Virus-specific IgM antibodies were detected from day 6 to day 50 postinoculation, and all samples from days 8-35 were positive (Fig. 5). IgM antibodies were detected in sera from 24 to 26 mink after a single vaccination with attenuated CDV vaccine at 6-15 weeks of age. The IgM-positive
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
Figure 4. Determination of IgM antibodies in sera from dogs suffering from chronic progressive inflammatory canine distemper encephalitis with a clinical duration of >6 weeks. ELISA OD492 values are illustrated for 10 postmortem sera. Positive/negative discrimination levels are marked with dashed lines.
samples were found between days 4 and 21 postvaccination (Fig. 6). Serum samples from 2 mink that received first vaccination at 6 weeks and 3 revaccinations at 9, 12, and 15 weeks of age were examined. In both mink, we detected IgM antibodies in serum on days 7 and 14 after first vaccination. On days 7 and 14 after the second vaccination, IgM antibodies were detected in only 1 of the mink, and no IgM antibodies were detected in serum samples examined during the first 3 weeks after subsequent revaccinations. Discussion With the IgM capture ELISA technique for detection of IgM antibodies against CDV, it is possible to diagnose an ongoing or recent CDV infection in dog and mink based on the detection of virus-specific antibodies in single serum samples from natural or experimental infections. Serodiagnosis of CDV infections is normally based on the detection of a significant (2 dilution step) titer rise in paired serum samples drawn at a sufficient (often 2-3 weeks) interval. The diagnostic potential of the IgM technique was evaluated by analyzing sera from natural or experimental infections in dog and mink. These results were compared with results obtained in the CDV IgG antibody ELISA and by VNT. A crude extract from a cell culture was employed in these ELISA techniques. Anti-human IgM antibodies that cross-reacted with canine and mink IgM were used for IgM capture. This cross-reactivity has also been demonstrated by rocketimmunoelectrophoresis. 12 The advantage of the rocketimmunoelectrophoresis assay is that it requires the presence of several antigenic determinants on the protein surface that react with the antibodies to form an immobile precipitate. This process ensures that the cross-reaction is substantial. In
7
Figure 5. Determination of the IgM antibody response in 18 mink inoculated with virulent CDV. ELISA OD492 values are illustrated for 32 sera from day 3 to day 50 after inoculation. Positive discrimination levels are marked with dashed lines.
this study, the ELISA results of the test sera were classed as positive, negative, or inconclusive. The cutoff values were determined using a nonparametric method because of the relatively small number of sera in the negative groups. 16,19 A reexamination of paired serum samples may offer further information of diagnostic importance, especially when dealing with negative or inconclusive results of the first sample tested. Reactivity of canine and human sera to cellular components has occurred in ELISA tests for the detection of morbillivirus antibodies.6,22 In the IgM ELISA technique of this study, false positive reactions were avoided by measuring the reaction of the test serum with CDV antigen and with control antigen and defining the E-value of a test serum as the OD492 difference between wells containing CDV antigen and control antigen.15,21 There was a relatively high background and a marked variation in the results obtained from negative sera.
Figure 6. Determination of the IgM antibody response in 24 mink after vaccination with an attenuated CDV vaccine. ELISA OD 492 values are illustrated for 103 sera from day 3 to day 56 after vaccination. Positive/negative discrimination levels are marked with dashed lines.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
8
Blixenkrone-Møller et al.
Similar results have been described in ELISA techniques for detecting IgM antibodies against measles 23 viruP and rubella virus. An unspecific binding of immunoglobulins of the test serum to solid phase or cellular components could contribute to the high background. Furthermore, it is possible that a high background value could arise from a direct binding of the conjugate to the immobilized antibody on the solid phase because of possible common epitopes on human and rabbit immunoglobulin.9 In the IgM ELISA technique, the background was reduced 10% by blocking the anti-IgM-coated solid phase with 1% normal rabbit serum. Interfering cross-reactions are minimized by using monoclonal antibodies against immunoglobulin and virus. The sensitivity of an ELISA for detection of IgM antibodies against rubella virus was increased by employing monoclonal instead of polyclonal antibodies.9 The reproducibility of the ELISA techniques presented here (expressed as a variation coefficient of 912% of the E-values of the reference sera) allowed an adequate differentiation of the results. We detected virus-specific IgM antibodies in most of the examined sera (44/49) from acute clinical infections in naturally infected dogs and experimentally infected mink. Six of 10 clinical samples from chronic cases of canine distemper encephalitis and 2 canine convalescent cases were IgM positive. These results illustrate the diagnostic potential of virus-specific IgM antibody detection for diagnosis of ongoing or recent CDV infections. For the IgM antibody response in mink, virus-specific IgM antibodies were detected up to 21 days after vaccination with attenuated CDV and up to 50 days after inoculation of virulent CDV. These results are similar to earlier undocumented investigations of IgM antibodies in dog sera that used an indirect immunofluorescence technique.2–4 In the present study, a primary IgM response was detected as early as 4 days after vaccination and 6 days after inoculation with a virulent strain of CDV, which is 1 and 4 days, respectively, earlier than indicated by FA studies. 4 This discrepancy might reflect a higher sensitivity of the IgM ELISA and differences in the experimental conditions (route of inoculation) or actual differences in the virus-host relations between mink and dog. Studies have indicated that revaccinations of dogs do not elicit a new IgM antibody response.4 Further investigations are needed to evaluate for how long it is possible to detect IgM antibodies in serum after vaccination with attenuated vaccine virus. The present investigation of the IgM response, however, emphasizes the potential of the IgM ELISA for diagnosing current or recent primary CDV infections. The results further indicate that virus-specific IgM antibodies in
serum may arise from natural infections as well as infections with attenuated vaccine virus. The high sensitivity and specificity of the IgM technique are indicated by the detection of IgM antibodies in dilutions of the test samples up to 1:25,000 and by comparison of results from sera tested with IgM ELISA, IgG ELISA, and VNT. In 12 sera from acute clinical infections in dogs, we detected CDV antibodies employing the ELISA techniques, whereas we were not able to detect CDV neutralizing antibodies. These results might reflect the detection by the ELISA techniques of antibodies to major internal viral antigens. Studies with sera from dogs infected with virulent CDV indicate that antibodies to the nucleoprotein and the phosphoprotein are detectable before neutralizing titers, which correspond with the appearance of antibodies to the hemagglutinin and fusion protein (B. Rima, unpublished data). The high sensitivity and specificity of the IgM ELISA are supported by results obtained in similar IgM capture techniques employed for the detection of measles virus IgM antibodies.15,18,21 A reduced sensitivity arising from competition between different classes of antibodies for the antigenic determinants is eliminated in the IgM capture ELISA, and the occurrence of false positive reactions due to rheumatoid factors is reduced. 11,15,18,22 The IgM ELISA presented here adds a new perspective to serodiagnosis of current CDV infections in dog and mink by detection of virus-specific IgM antibodies in single serum samples. The detection of IgM antibodies thus appears to be a useful supplement to the detection of intracellular viral antigen for both in vivo and postmortem diagnoses of current CDV infections. Acknowledgements We are grateful to Agnete Ingild, Dakopatts, for peroxidase conjugation of CDV antiserum. We also thank Jeanette Nielsen, Birte Nissen, Eva Lund, and Amalie Heiligstadt for technical assistance. This study was supported by grants from The Danish Fur Breeders Association Research Foundation.
Sources and manufacturers a. Distemink Vet Aerosol, United Vaccines, Division of Harlan Sprague, Inc., Madison, WI. b. Beckman Instruments, Inc., Palo Alto, CA. c. Nordic Immunological Laboratories, Tilburg, The Netherlands. d. Pelfreez Biologicals, Rogers, AR. e. Provided by B. Aasted, The Royal Veterinary and Agricultural University, Copenhagen, Denmark. f. Dakopatts, Glostrup, Denmark. g. SLT-Labinstruments, Salzburg, Austria. h. United Vaccines, Madison, WI. i. Provided by D. Danner and M. Johannsen, Behringwerke, Marburg, Germany. j. Provided by E. Gunnarsson, Keldur, Iceland. k. Provided by F. Hojelse, Lundbeck, Copenhagen, Denmark.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
References 1. Appel MJG: 1969, Pathogenesis of canine distemper. Am J Vet Res 30:1167-1181. 2. Appel M JG: 1987, Canine distemper virus. In: Virus infections of vertebrates, vol. I. Virus infections of carnivores, ed. Appel MJG, pp. 133-199. Elsevier, New York, NY. 3. Appel MJG: 1988, Canine distemper. In: Manual of small animal infectious diseases, ed. Barlow JE, pp. 49-61. Churchill Livingstone, New York, NY. 4. Appel MJG, Gibbs EPJ, Martin SJ, et al.: 1981, Morbillivirus diseases of animals and man. In: Comparative diagnosis of viral diseases, ed. Kurstak E, Kurstak C, vol. 4, pp. 235-297. Academic Press, New York, NY. 5. Appel MJG, Robson DS: 1973, A microneutralisation test for canine distemper virus. Am J Vet Res 34:1459-1463. 6. Bernard SL, Shen DT, Gorham JR: 1982, Antigen requirements and specificity of enzyme-linked immunosorbent assay for detection of canine IgG against canine distemper viral antigens. Am J Vet Res 43:2266-2269. 7. Bohm J, Blixenkrone-Møller M, Lund E: 1989, A serious outbreak of canine distemper among sled-dogs in Northern Greenland. Arctic Med Res 48: 195-203. 8. Busby DWG, House W, Mac Donald JR: 1964, Virological technique, pp. 144-146. J. A. Churchill, London. 9. Chantler S, Evans CJ: 1986, Selection and performance of monoclonal and polyclonal antibodies in an IgM antibody capture enzyme immunoassay for rubella. J Immunol Methods 87: 109-117. 10. De Vries P, Uytdehaag FGCM, Osterhaus ADME: 1988, Canine distemper virus (CDV) immune-stimulating complexes (iscoms), but not measles virus iscoms, protect dogs against CDV infection. J Gen Virol 69:2071-2083. 11. Duermeyer W, Wieland F, Van der Veen J: 1979, A new principle for the detection of specific IgM antibodies as applied in a ELISA for hepatitis. Am J Med Virol 4:25-32. 12. Hau J, Nilsson M, Skovgaard-Jensen H-J, et al.: 1990, Analysis of animal serum proteins using antisera against human analogous proteins. Stand J Lab Anim Sci 17(1):3-7. 13. Krakowka S, Olsen R, Confer A, et al.: 1975, Serological re-
9
sponse to canine distemper virus antigen in gnotobiotic dogs infected with canine distemper virus. J Infect Dis 132:384-392. 14. Noon RF, Rogul M, Binn LN, et al.: 1980, Enzyme-linked immunoabsorbent assay for evaluation of antibody to canine distemper virus. Am J Vet Res 41:605-609. 15. Pedersen IR, Antonsdottir A, Evald T, Mordhorst CH: 1982, Detection of measles IgM antibody by enzyme-linked immunosorbent assay (ELISA). Acta Pathol Microbiol Immunol Stand B 90: 153-160. 16. Reed AH, Henry JH, Mason WB: 1971, Influence of statistical method used on the resulting estimate of normal range. Clin Chem 17(4):275-284. 17. Salonen E-M, Vaheri A, Suni J, Wager O: 1980, Rheumatoid factor in acute viral infections: interference with determination of IgM, IgG & IgA antibodies in an enzyme immunoassay. J Infect Dis 142:250-255. 18. Salonen J, Vainionpaa R, Halonen P: 1986, Assay of measles virus IgM & IgG class antibody by use of peroxidase-labeled viral antigen. Arch Virol 91:93-106. 19. Savigny D, Voller A: 1980, The communication of ELISA data from laboratory to clinician. J Immunoassay 1(1): 105-128. 20. Shek WP, Schultz RD, Appel MJG: 1980, Natural and immune cytolysis of canine distemper virus infected target cells. Infect Immun 28:724-734. 21. Tuokko H: 1984, Comparison of nonspecific reactivity in indirect and reverse immunoassays for measles and mumps immunoglobulin M antibodies. J Clin Microbiol 20:972-976. 22. Tuokko H, Salmi A: 1983, Detection of IgM antibodies to measles virus by enzyme immunoassay. Med Microbiol Immunol 171:187-198. 23. Vejtorp M: 1981, Solid phase anti-IgM ELISA for detection of rubella specific IgM antibodies. Acta Pathol Stand B 89: 123128. 24. White G, Cowan KM: 1962, Separation of soluble antigen and infectious particles of rinderpest and canine distemper virus. Virology 16:209-211. 25. Winters KA, Mathes LE, Krakowka S, Olsen RG: 1983, Immunoglobulin class response to canine distemper virus in gnotobiotic dogs. Vet Immunol Immunopathol 5:209-215.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies against canine distemper virus in dog and mink sera employing enzyme-linked immunosorbent assay (ELISA) Merete Blixenkrone-Moller, Ib Rode Pedersen, Max J. Appel, Christian Griot Abstract. An enzyme-linked immunosorbent assay (ELISA) for the detection of IgM antibodies against canine distemper virus (CDV) in canine and mink serum is described. The diagnostic potential of this technique was evaluated by analyzing sera from natural or experimental infections in dog and mink and negative control sera. These results were compared with results obtained in the developed CDV IgG ELISA and in the virus neutralization test. The IgM test, which requires only a single serum specimen, is a useful method for diagnosing current or recent CDV infections in dog and mink.
A virus-specific serum antibody response after induced infection in dogs with virulent or attenuated vaccine strains of CDV may be detectable by employing an indirect immunofluorescence technique.2–4 The ELISA techniques have been described for the demonstration of the class-specific immunoglobulin response during CDV infections.6,10,14,25 However, it is not known if the techniques can distinguish between known infected patients and healthy controls. The ELISA techniques have been developed for the detection of antibodies against the closely related measles virus, and these techniques, which we essentially based on measurements of single serum samples, are currently used for serodiagnosis of present or recent measles virus infections in humans.15,18,21,22 The principle of these ELISA techniques, in contrast to other CDV ELISA techniques, 10,14,25 is based on an immunological separation of from the test sample before the reaction with viral antigen. This capture technique eliminates competing antibodies and minimizes nonspecific reactions due to rheuma factors in human sera.11,17 The objectives of the present studies were to set up an ELISA technique based on examination of single serum samples for the detection of antibodies against CDV in dog and mink sera and to evaluate the validity of the technique for diagnosis of present or recent CDV infections.
Canine distemper (CD) is a contagious systemic viral disease with often high mortality rates in dogs and other carnivores throughout the world. Canine distemper is caused by canine distemper virus a morbillivirus in the paramyxovirus family. An acute primary CDV infection may induce a specific antibody response. An antibody titer rise in paired serum samples is of decisive diagnostic importance. The detection of an increased titer of CDV antibodies employing the neutralization test,5 precipitation test,24 complement fixation test,13 cytotoxic test,20 or enzyme-linked immunosorbent assay (ELISA)14,25 is, however, not suitable for routine diagnostics because the titers are often high in the beginning of the clinical disease and may not rise further.2 Often in fatal cases of distemper, no detectable serum neutralizing 1,13 antibodies are found. In cases where a significant titer rise is demonstrated in suitable paired samples (taken with a 3-week interval), the result may have lost its importance because of the time factor. However, antibody of the class produced early in viral infection may indicate an ongoing or recent viral multiplication. For this reason, measurement of single serum samples has been investigated as an aid for viral diagnosis. From the Department of Veterinary Microbiology, Laboratory for Virology and Immunology, The Royal Veterinary and Agricultural University of Copenhagen, 13, Bülowsvej 1870 Frederiksberg C, Denmark (Blixenkrone-Moller), the Department of Medical Microbiology, University of Copenhagen, Panum 24/2, 3, Blegdamsvej 2200 Copenhagen N, Denmark (Pedersen), the James Baker Institute for Animal Health, New York State College of Veterinary Medicine, Cornell University, Ithaca, NY 14853 (Appel), and the Institute of Animal Neurology, University of Bern, Bremgartenstr. 109a, Postfach 2735, CH-3001 Bern, Switzerland (Griot). Received for publication August 4, 1990.
Materials and methods Preparation of CDV antigen. Vero cells were grown in stationary flasks (175 cm2) in Eagle’s minimum essential medium with 2% bovine fetal serum (BFS), 20 mM/liter L-glutamine, 2.2 g/liter sodium bicarbonate, and antibiotics. At 70-80% confluency, the monolayer cultures were infected at a multiplicity of infection (MOI) of 1, 0.1, or 0.01 (estimated 3
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
4
Blixenkrone-Møller et al.
by CPE in Vero cells) with the Onderstepoort strain of CDV isolated from a commercially attenuated CDV vaccine for mink. a After virus adsorption for 2 hr at 37 C, the cultures were incubated until pronounced CPE were observed (7296 hr postinoculation). Viral antigen was prepared by extraction of cell-associated virus. The culture medium was discarded, and after rinsing the cell layer in phosphate buffered saline (PBS) pH 7.2, 3 ml lysis buffer pH 8.4 (0.01 M Tris, 0.002 M ethylenediaminetetraacetic acid [EDTA], 0.2 M sucrose, 2% [v/v] Triton-X100, and 200 KIU/ml aprotinin) was added to each culture flask. Following freezingthawing once and sonication in an ice bath twice for 15 sec each, the cell lysate was clarified by 2 consecutive centrifugations (10 min, 12,100 × g), and the supernatant was centrifuged (20 min, 48,200 × g). The supernatant was used as antigen in the ELISA techniques and was stored at -40 C. Control antigen was prepared in the same way from noninfected Vero cell cultures. Production of antisera against CDV. Antisera against CDV were raised in 2 rabbits that had been immunized with purified Vero cell-grown virus. Virus was harvested by freezingthawing of infected cell cultures, and after clarification (10 min at 12,100 × g), the supernatant was centrifuged into a double layer sucrose cushion (30 and 60% w/w sucrose in TE-buffer pH 7.4 [0.01 M Tris, 0.01 M EDTA, and 0.1 M NaCl]) for 90 min at 120,000 × g at 4 C in a SW 28 rotor. b The interphase between the sucrose cushions was diluted in TE buffer, layered on top of a linear 20-60% w/w sucrose gradient in TE buffer, and centrifuged for 235 min at 155,000 × g at 4 C in a VTI 50 rotor. b Gradient fractions were collected by bottom unloading, and CDV antigen in the fractions was quantified by a modified antigen capture ELISA as previously described, 15 employing canine CDV antiserum and rabbit anti-canine IgG conjugated to horseradish peroxidase. c The rabbit antisera against CDV were used conjugated to peroxidase after purification of IgG and as “catching antibody”’ in the CDV antibody ELISA. Purification and enzyme conjugation. The IgG was purified from the rabbit antisera against CDV and conjugated to horseradish peroxidase as previously described.14 The conjugate contained 1.2 mg protein/ml. Examination of anti-IgM reagents. The following 3 rabbit antisera were tested in the IgM ELISA technique: anti-dog IgM, d anti-mink IgM,e and anti-human IgM (µ-chain specific).f ELISA techniques. The techniques illustrated in Fig. 1 were carried out in flat-bottom polystyrene microtest plates with 96 wells and 100 µ1 per well. Optimal conditions for incubation of the various reagents (time, temperature, stationary/orbital shaking) were established in preliminary experiments. Working dilutions for the reagents that made it possible to determine the minimal amounts of virus-specific antibodies detectable in serum samples were determined by checkerboard titrations. Preliminary ELISA results indicated that only IgG class antibodies were detectable using the antiIgG (H + L) peroxidase conjugate. All test samples were run in duplicate. A carbonate buffer pH 9.6 (15 mM Na2CO3, 35 mM NaHCO3, 3 mM NaN3) was used for dilution of catching antibodies. After every step of incubation with immunological reagents, the plates were rinsed 4 × for 3 min each in
washing buffer pH 7.2 (0.5 M NaCl, 2.7 mM KCl, 1.5 mM KH2PO 4, 6.5 mM Na2HPO 4· 2H2O, 1% [v/v] Triton-X100). Subsequent immunological reagents were diluted in dilution buffer, i.e., washing buffer supplemented with 0.01% w/v phenol red and 1% w/v bovine serum albumin. The substrate for the enzyme reaction was 8 mg phenylendiamine dihydrochloride (OPD) and 5 µ1 30% H O 2 2 per 12 ml 0.1 M citrate buffer pH 5.0. The reaction was stopped with 1 M H2SO 4. The color was read at 492 nm with a 620-nm reference wavelength on an ELISA reader.g The ELISA OD492 values (Evalues) of the serum samples were defined as the OD492 difference between the mean of 2 measures in wells with viral antigen and control antigen. To minimize false positive results, the cutoff values were established with a nonparametric method. 16,19 Virus neutralization test (VNT). The VNT was performed as described5 using flat-bottom microtest plates with 96 wells, and every serial dilution of test serum was examined in quadruplicate. Each test well was incubated for 1 hr at 37 C with 50 µ1 2-fold serum dilutions and 50 µ1 dilution of the Onderstepoort strain of CDV containing 5-56 TCID50. Fifty microliters of a Vero cell suspension (5 × 105 cells/ml) were then added to each well. After 2 hr, the supernatants in the wells were replaced with 200 µ1 Eagle’s medium supplemented with 2% BFS. The plates were sealed, and after an incubation period of 4-5 days at 37 C, the cultures were examined under the microscope for cytopathic changes and the titer was calculated ad modum Kärber.8 Patients and serum samples. One hundred thirty-one serum samples were taken from 126 dogs of various breeds and ages and 256 serum samples were taken from 120 farmed mink (6-52 wk of age). Samples were kept at -20 C. There were 10 paired samples from 5 Beagles before and after inoculation with virulent Cornell A75-17 strain of CDV and 43 samples from naturally occurring cases of canine distemper. These 43 samples were from acute cases of canine distemper (7 samples from unvaccinated dogs and 24 samples from dogs that had been vaccinated within 2-3 wk of a distemper outbreak in Northern Greenland in 19887), from 2 dogs in the convalescent phase after acute canine distemper, and from 10 postmortem cases of chronic, progressive inflammatory distemper encephalitis. One hundred twenty-four samples were from 26 mink 0-56 days after subcutaneous vaccination with an attenuated CDV vaccine,h and 71 samples were taken from 33 mink 0-56 days after inoculation of the virulent Snyder-Hill strain of CDV.i Negative control sera. Negative control sera from dog and mink were analyzed with the ELISA techniques. Ten samples were from unvaccinated sled dogs from Greenland before the canine distemper outbreak in 1988, 31 samples were from unvaccinated dogs from Iceland,j and 9 samples came from unvaccinated 14-wk-old Beaglesk raised for experimental purposes and kept isolated from other dogs. Another 28 canine samples from randomly selected Danish dogs that had not been vaccinated within the last 3 mo were also classified as CDV IgM negative control sera. Sixty-one negative mink control sera were from unvaccinated wild mink from farms with no distemper outbreaks during the last 8 yr. Furthermore, the 5 canine sera taken before inoculation of virulent CDV and 24 mink sera taken before vaccination or before
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
Figure 2. Determination of IgM ELISA OD492 values of rabbit anti-IgM used as catching antibody. A serial 2-fold dilution of antiIgM reacted with sera from CDV-infected dog and mink and with sera from uninfected dog and mink Figure 1. Principles of ELISA for CDV IgM antibody detection (A) and CDV IgG antibody detection (B).
inoculation with a virulent strain of CDV were classed among the negative control sera. Reference sera. The 2 positive-IgM reference sera were from a dog with an acute clinical CDV infection and from a vaccinated mink. A serum pool from vaccinated dogs and hyperimmune serum from mink served as positive references in the CDV IgG antibody ELISA.
Results Determination of the optimal anti-IgM concentration for the CDV IgM ELISA. Optimal preparations of CDV antigen for the ELISA techniques were obtained from cell cultures infected at MOI 0.1. Of the different anti-IgM reagents (anti-human, anti-dog, antimink) tested as capture antibodies in the IgM ELISA, anti-human IgM gave the best results employing canine as well as mink sera. Serial 2-fold dilutions of antihuman IgM used as catching antibody in IgM ELISA are shown in Fig. 2. The background was reduced 10% by including a blocking step with 1% rabbit serum in PBS after coating of the solid phase. The incubation conditions and dilutions of ELISA reagents that made it possible to detect minimal amounts of virus-specific antibodies in test samples are given. CDV IgM antibody ELISA procedure. The wells were coated with anti-human IgM diluted 1:200 for 1 hour at 37 C. Before the next incubation step the plates were blocked with 1% rabbit serum in PBS for 1.5 hours at 37 C. Dilutions (1:25) of serum samples were incubated overnight at 4 C. Then dilutions (150) of viral antigen and control antigen were added, and the plates were incubated at 37 C on an orbital shaker for 3 hours, followed by shaking with anti-CDV peroxidase conjugate (1:100 dilution) for 1.5 hours at 37 C. Finally the substrate reaction, carried out in the dark at 21 C, was stopped after 13 minutes.
CDV IgM antibody ELISA procedure. The wells were coated for 1 hour at 37 C with rabbit antisera against CDV (1:3,000 dilution). Viral antigen and control antigen (1:100 dilution) were added to the plates, incubated overnight at 4 C, then incubated with serum samples (canine sera in 1: 100 dilution, mink sera in 1:50 dilution) for 1 hour at 37 C, followed by incubation with anti-canine IgG conjugate for 1 hour at 37 C. The rest of the procedure was as described for the IgM ELISA. Reproducibility and sensitivity. The E-value of a serum sample was corrected in accordance with the E-value for a positive reference serum to minimize the day to day variation. The variation of the E-values was 18% for 8 parallel testings on 8 different days for IgM reference mink serum, with a mean of 1.069 (0.9541.187). Corresponding variation for IgG reference mink serum with a mean of 1.326 (1.400-1.071) was 12%. The sensitivity of the IgM ELISA technique, defined as the highest serum dilution with positive reaction, was examined employing serial 10-fold dilutions of 6 sera. The IgM antibodies were detected in dilutions up to 1:25,000 in dog sera and 1:2,500 in mink sera. Determination of cutoff values. To establish the cutoff values between positive and negative sera in the CDV IgM antibody ELISA, 121 canine serum samples (83 negative control samples, 36 samples from clinical infections, and 2 samples from convalescent dogs) and 103 mink serum samples (85 negative control samples and 18 samples from clinical infections) were examined (Fig. 3). The cutoff for negative E-values was established at 100% of the negative control group. Considering the limited number of sera in the negative control groups, the cutoff for positive E-values was set at 1.5 times the negative cutoff. The inconclusive values were defined between the positive and negative
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
6
Blixenkrone-Møller et al.
Figure 3. Determination of positive/negative IgM discrimination levels (dashed lines). A. ELISA OD 492 values for 38 canine sera There were 31 sera from acute clinical cases of canine distemper, 5 sera from experimental inoculation with a virulent strain of CDV, and 2 samples from convalescent dogs. There were 83 negative control sera B. ELISA OD492 values for 18 mink sera from acute clinical CDV infections after experimental inoculation. There were 85 negative control sera
cutoffs. Thus in the IgM ELISA E-values ≤0.27 were considered negative, 0.28-0.41 were inconclusive values, and >0.41 were considered positive (Fig. 3A). Equivalent values for mink sera were ≤ 0.30 (negative), 0.31-0.45 (inconclusive), and >0.45 (positive) (Fig. 3B). The true cutoff values seem to cover the same intervals within the accuracy of the measurement. The cutoff values for the IgG ELISA were established as described for the IgM ELISA. Examination of CDV IgM antibodies in sera from clinical CD V infections. Forty-four of 49 examined sera from acute clinical infections in dog and mink were IgM positive, and 2 samples from convalescent dogs after natural infections were IgM positive (Fig. 3). In chronic cases with central nervous system symptoms, we detected virus-specific IgM antibodies in 6 out of 10 cases (Fig. 4). Comparative study of sera in CDV IgM antibody ELISA, CDV IgG antibody ELISA, and VNT. Thir-
teen sera from acute clinical cases in dogs were found positive in both ELISA techniques, but in only 1 of these were CDV antibodies detected by VNT. In 7 serum samples from mink inoculated with a virulent strain of CDV, viral antibodies were detected in the IgM ELISA but not in the IgG ELISA. Thirty-six canine serum samples and 7 serum samples from mink that had been found negative in both ELISA techniques were also negative by VNT. Examination of the IgM response in mink inoculated with a virulent or attenuated strain of CDV. Thirtytwo serum samples from 18 mink that had been inoculated with the virulent Snyder-Hill strain of CDV were analyzed for IgM antibodies. Virus-specific IgM antibodies were detected from day 6 to day 50 postinoculation, and all samples from days 8-35 were positive (Fig. 5). IgM antibodies were detected in sera from 24 to 26 mink after a single vaccination with attenuated CDV vaccine at 6-15 weeks of age. The IgM-positive
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
Figure 4. Determination of IgM antibodies in sera from dogs suffering from chronic progressive inflammatory canine distemper encephalitis with a clinical duration of >6 weeks. ELISA OD492 values are illustrated for 10 postmortem sera. Positive/negative discrimination levels are marked with dashed lines.
samples were found between days 4 and 21 postvaccination (Fig. 6). Serum samples from 2 mink that received first vaccination at 6 weeks and 3 revaccinations at 9, 12, and 15 weeks of age were examined. In both mink, we detected IgM antibodies in serum on days 7 and 14 after first vaccination. On days 7 and 14 after the second vaccination, IgM antibodies were detected in only 1 of the mink, and no IgM antibodies were detected in serum samples examined during the first 3 weeks after subsequent revaccinations. Discussion With the IgM capture ELISA technique for detection of IgM antibodies against CDV, it is possible to diagnose an ongoing or recent CDV infection in dog and mink based on the detection of virus-specific antibodies in single serum samples from natural or experimental infections. Serodiagnosis of CDV infections is normally based on the detection of a significant (2 dilution step) titer rise in paired serum samples drawn at a sufficient (often 2-3 weeks) interval. The diagnostic potential of the IgM technique was evaluated by analyzing sera from natural or experimental infections in dog and mink. These results were compared with results obtained in the CDV IgG antibody ELISA and by VNT. A crude extract from a cell culture was employed in these ELISA techniques. Anti-human IgM antibodies that cross-reacted with canine and mink IgM were used for IgM capture. This cross-reactivity has also been demonstrated by rocketimmunoelectrophoresis. 12 The advantage of the rocketimmunoelectrophoresis assay is that it requires the presence of several antigenic determinants on the protein surface that react with the antibodies to form an immobile precipitate. This process ensures that the cross-reaction is substantial. In
7
Figure 5. Determination of the IgM antibody response in 18 mink inoculated with virulent CDV. ELISA OD492 values are illustrated for 32 sera from day 3 to day 50 after inoculation. Positive discrimination levels are marked with dashed lines.
this study, the ELISA results of the test sera were classed as positive, negative, or inconclusive. The cutoff values were determined using a nonparametric method because of the relatively small number of sera in the negative groups. 16,19 A reexamination of paired serum samples may offer further information of diagnostic importance, especially when dealing with negative or inconclusive results of the first sample tested. Reactivity of canine and human sera to cellular components has occurred in ELISA tests for the detection of morbillivirus antibodies.6,22 In the IgM ELISA technique of this study, false positive reactions were avoided by measuring the reaction of the test serum with CDV antigen and with control antigen and defining the E-value of a test serum as the OD492 difference between wells containing CDV antigen and control antigen.15,21 There was a relatively high background and a marked variation in the results obtained from negative sera.
Figure 6. Determination of the IgM antibody response in 24 mink after vaccination with an attenuated CDV vaccine. ELISA OD 492 values are illustrated for 103 sera from day 3 to day 56 after vaccination. Positive/negative discrimination levels are marked with dashed lines.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
8
Blixenkrone-Møller et al.
Similar results have been described in ELISA techniques for detecting IgM antibodies against measles 23 viruP and rubella virus. An unspecific binding of immunoglobulins of the test serum to solid phase or cellular components could contribute to the high background. Furthermore, it is possible that a high background value could arise from a direct binding of the conjugate to the immobilized antibody on the solid phase because of possible common epitopes on human and rabbit immunoglobulin.9 In the IgM ELISA technique, the background was reduced 10% by blocking the anti-IgM-coated solid phase with 1% normal rabbit serum. Interfering cross-reactions are minimized by using monoclonal antibodies against immunoglobulin and virus. The sensitivity of an ELISA for detection of IgM antibodies against rubella virus was increased by employing monoclonal instead of polyclonal antibodies.9 The reproducibility of the ELISA techniques presented here (expressed as a variation coefficient of 912% of the E-values of the reference sera) allowed an adequate differentiation of the results. We detected virus-specific IgM antibodies in most of the examined sera (44/49) from acute clinical infections in naturally infected dogs and experimentally infected mink. Six of 10 clinical samples from chronic cases of canine distemper encephalitis and 2 canine convalescent cases were IgM positive. These results illustrate the diagnostic potential of virus-specific IgM antibody detection for diagnosis of ongoing or recent CDV infections. For the IgM antibody response in mink, virus-specific IgM antibodies were detected up to 21 days after vaccination with attenuated CDV and up to 50 days after inoculation of virulent CDV. These results are similar to earlier undocumented investigations of IgM antibodies in dog sera that used an indirect immunofluorescence technique.2–4 In the present study, a primary IgM response was detected as early as 4 days after vaccination and 6 days after inoculation with a virulent strain of CDV, which is 1 and 4 days, respectively, earlier than indicated by FA studies. 4 This discrepancy might reflect a higher sensitivity of the IgM ELISA and differences in the experimental conditions (route of inoculation) or actual differences in the virus-host relations between mink and dog. Studies have indicated that revaccinations of dogs do not elicit a new IgM antibody response.4 Further investigations are needed to evaluate for how long it is possible to detect IgM antibodies in serum after vaccination with attenuated vaccine virus. The present investigation of the IgM response, however, emphasizes the potential of the IgM ELISA for diagnosing current or recent primary CDV infections. The results further indicate that virus-specific IgM antibodies in
serum may arise from natural infections as well as infections with attenuated vaccine virus. The high sensitivity and specificity of the IgM technique are indicated by the detection of IgM antibodies in dilutions of the test samples up to 1:25,000 and by comparison of results from sera tested with IgM ELISA, IgG ELISA, and VNT. In 12 sera from acute clinical infections in dogs, we detected CDV antibodies employing the ELISA techniques, whereas we were not able to detect CDV neutralizing antibodies. These results might reflect the detection by the ELISA techniques of antibodies to major internal viral antigens. Studies with sera from dogs infected with virulent CDV indicate that antibodies to the nucleoprotein and the phosphoprotein are detectable before neutralizing titers, which correspond with the appearance of antibodies to the hemagglutinin and fusion protein (B. Rima, unpublished data). The high sensitivity and specificity of the IgM ELISA are supported by results obtained in similar IgM capture techniques employed for the detection of measles virus IgM antibodies.15,18,21 A reduced sensitivity arising from competition between different classes of antibodies for the antigenic determinants is eliminated in the IgM capture ELISA, and the occurrence of false positive reactions due to rheumatoid factors is reduced. 11,15,18,22 The IgM ELISA presented here adds a new perspective to serodiagnosis of current CDV infections in dog and mink by detection of virus-specific IgM antibodies in single serum samples. The detection of IgM antibodies thus appears to be a useful supplement to the detection of intracellular viral antigen for both in vivo and postmortem diagnoses of current CDV infections. Acknowledgements We are grateful to Agnete Ingild, Dakopatts, for peroxidase conjugation of CDV antiserum. We also thank Jeanette Nielsen, Birte Nissen, Eva Lund, and Amalie Heiligstadt for technical assistance. This study was supported by grants from The Danish Fur Breeders Association Research Foundation.
Sources and manufacturers a. Distemink Vet Aerosol, United Vaccines, Division of Harlan Sprague, Inc., Madison, WI. b. Beckman Instruments, Inc., Palo Alto, CA. c. Nordic Immunological Laboratories, Tilburg, The Netherlands. d. Pelfreez Biologicals, Rogers, AR. e. Provided by B. Aasted, The Royal Veterinary and Agricultural University, Copenhagen, Denmark. f. Dakopatts, Glostrup, Denmark. g. SLT-Labinstruments, Salzburg, Austria. h. United Vaccines, Madison, WI. i. Provided by D. Danner and M. Johannsen, Behringwerke, Marburg, Germany. j. Provided by E. Gunnarsson, Keldur, Iceland. k. Provided by F. Hojelse, Lundbeck, Copenhagen, Denmark.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
Detection of IgM antibodies to CDV by ELISA
References 1. Appel MJG: 1969, Pathogenesis of canine distemper. Am J Vet Res 30:1167-1181. 2. Appel M JG: 1987, Canine distemper virus. In: Virus infections of vertebrates, vol. I. Virus infections of carnivores, ed. Appel MJG, pp. 133-199. Elsevier, New York, NY. 3. Appel MJG: 1988, Canine distemper. In: Manual of small animal infectious diseases, ed. Barlow JE, pp. 49-61. Churchill Livingstone, New York, NY. 4. Appel MJG, Gibbs EPJ, Martin SJ, et al.: 1981, Morbillivirus diseases of animals and man. In: Comparative diagnosis of viral diseases, ed. Kurstak E, Kurstak C, vol. 4, pp. 235-297. Academic Press, New York, NY. 5. Appel MJG, Robson DS: 1973, A microneutralisation test for canine distemper virus. Am J Vet Res 34:1459-1463. 6. Bernard SL, Shen DT, Gorham JR: 1982, Antigen requirements and specificity of enzyme-linked immunosorbent assay for detection of canine IgG against canine distemper viral antigens. Am J Vet Res 43:2266-2269. 7. Bohm J, Blixenkrone-Møller M, Lund E: 1989, A serious outbreak of canine distemper among sled-dogs in Northern Greenland. Arctic Med Res 48: 195-203. 8. Busby DWG, House W, Mac Donald JR: 1964, Virological technique, pp. 144-146. J. A. Churchill, London. 9. Chantler S, Evans CJ: 1986, Selection and performance of monoclonal and polyclonal antibodies in an IgM antibody capture enzyme immunoassay for rubella. J Immunol Methods 87: 109-117. 10. De Vries P, Uytdehaag FGCM, Osterhaus ADME: 1988, Canine distemper virus (CDV) immune-stimulating complexes (iscoms), but not measles virus iscoms, protect dogs against CDV infection. J Gen Virol 69:2071-2083. 11. Duermeyer W, Wieland F, Van der Veen J: 1979, A new principle for the detection of specific IgM antibodies as applied in a ELISA for hepatitis. Am J Med Virol 4:25-32. 12. Hau J, Nilsson M, Skovgaard-Jensen H-J, et al.: 1990, Analysis of animal serum proteins using antisera against human analogous proteins. Stand J Lab Anim Sci 17(1):3-7. 13. Krakowka S, Olsen R, Confer A, et al.: 1975, Serological re-
9
sponse to canine distemper virus antigen in gnotobiotic dogs infected with canine distemper virus. J Infect Dis 132:384-392. 14. Noon RF, Rogul M, Binn LN, et al.: 1980, Enzyme-linked immunoabsorbent assay for evaluation of antibody to canine distemper virus. Am J Vet Res 41:605-609. 15. Pedersen IR, Antonsdottir A, Evald T, Mordhorst CH: 1982, Detection of measles IgM antibody by enzyme-linked immunosorbent assay (ELISA). Acta Pathol Microbiol Immunol Stand B 90: 153-160. 16. Reed AH, Henry JH, Mason WB: 1971, Influence of statistical method used on the resulting estimate of normal range. Clin Chem 17(4):275-284. 17. Salonen E-M, Vaheri A, Suni J, Wager O: 1980, Rheumatoid factor in acute viral infections: interference with determination of IgM, IgG & IgA antibodies in an enzyme immunoassay. J Infect Dis 142:250-255. 18. Salonen J, Vainionpaa R, Halonen P: 1986, Assay of measles virus IgM & IgG class antibody by use of peroxidase-labeled viral antigen. Arch Virol 91:93-106. 19. Savigny D, Voller A: 1980, The communication of ELISA data from laboratory to clinician. J Immunoassay 1(1): 105-128. 20. Shek WP, Schultz RD, Appel MJG: 1980, Natural and immune cytolysis of canine distemper virus infected target cells. Infect Immun 28:724-734. 21. Tuokko H: 1984, Comparison of nonspecific reactivity in indirect and reverse immunoassays for measles and mumps immunoglobulin M antibodies. J Clin Microbiol 20:972-976. 22. Tuokko H, Salmi A: 1983, Detection of IgM antibodies to measles virus by enzyme immunoassay. Med Microbiol Immunol 171:187-198. 23. Vejtorp M: 1981, Solid phase anti-IgM ELISA for detection of rubella specific IgM antibodies. Acta Pathol Stand B 89: 123128. 24. White G, Cowan KM: 1962, Separation of soluble antigen and infectious particles of rinderpest and canine distemper virus. Virology 16:209-211. 25. Winters KA, Mathes LE, Krakowka S, Olsen RG: 1983, Immunoglobulin class response to canine distemper virus in gnotobiotic dogs. Vet Immunol Immunopathol 5:209-215.
Downloaded from vdi.sagepub.com at UNIV CALIFORNIA SAN DIEGO on January 13, 2015
