Development and validation of a competitive enzyme-linked immunosorbent assay for detection of type A influenza antibodies in avian sera.

Serologic screening of avian sera for group-specific antibodies to type A influenza is currently accomplished by using the avian influenza (AI) agar gel immunodiffusion (AGID) test. A competitive enzyme-linked immunosorbent assay (CELISA) was developed using a baculovirus vector, Autographa californica nuclear polyhedrosis virus, expressing the nucleoprotein (NP) gene of A/Ann Arbor/6/60 influenza virus. The recombinant NP was obtained by inoculation of Spodoptera frugiperda (Sf9) insect cells or Trichoplusia ni insect larvae with the recombinant baculovirus. A hybridoma cell line producing monoclonal antibody against influenza virus A nucleoprotein was used to generate mouse ascitic fluid for the CELISA. The nucleoprotein and the monoclonal antibody were used without further purification in a CELISA for detection of avian-origin serum antibodies to type A influenza. The AI AGID and CELISA tests were compared for sensitivity and specificity using 1651 experimental and reference antisera. Samples discrepant in AGID and CELISA test results were further evaluated by the AI indirect fluorescent antibody (IFA), hemagglutination-inhibition (HI), and neuraminidase-inhibition (NI) tests. The results demonstrated a high degree of correlation between the AGID and CELISA test results, with the IFA, HI, and NI tests offering additional support of CELISA test specificity. The CELISA is a rapid, economical, sensitive, and specific serodiagnostic method for screening large numbers of avian sera for antibodies to avian influenza virus.

Comparative study of serological methods for the detection of antibodies to porcine reproductive and respiratory syndrome virus.

A comparison was made of serological diagnostic methods used for the detection of antibodies against porcine reproductive and respiratory syndrome (PRRS) virus. In the "phase I" PRRS test panel comparison, a panel of sera collected from 135 pigs of various ages, from North American herds with and without PRRS histories, were sent to 4 different laboratories and tested by an indirect immunofluorescent assay (IFA), an immunoperoxidase monolayer assay (IPMA) and an indirect enzyme-linked immunosorbent assay (iELISA). In the "phase II" PRRS test panel comparison, a panel of 382 sera collected from pigs of various ages, PRRS histories, and from various locations in North America and France, were divided into 2 panels (A & B) and sent to 3 Canadian laboratories and tested by the IFA and iELISA. In the phase I comparison, agreement between the IFA of laboratory 4 and the iELISA and IPMA of laboratory 3 was excellent (kappa values of 95% and 98%, respectively). This contrasted with the poor agreement between these laboratories and the IFA results of laboratories 1 and 2 in the phase I trial. In the phase II comparison, the results demonstrated good agreement between various tests both within and between laboratories. The overall performance of the iELISA was superior in the combination of sensitivity (96.1%) and specificity (100%) relative to the reference classification of the serum samples and repeatability (kappa value 98%). The iELISA is technically superior to IFA and IPMA, time efficient, cost effective and suitable for testing of a large number of samples over a short period of time. Thus, the iELISA may be a better alternative to IFA or IPMA for routine detection of PRRS viral antibodies in swine sera.

Comparison of the antigen distribution of two US porcine reproductive and respiratory syndrome virus isolates with that of the Lelystad virus.

One hundred 4-week-old cesarean-derived colostrum-deprived pigs were inoculated with one of two different US porcine reproductive and respiratory syndrome virus (PRRSV) isolates (VR2385, VR2431) or the European Lelystad virus to detect and compare the location and amount of virus antigen. Interstitial pneumonia, myocarditis, lymphadenopathy, and encephalitis were consistently seen in all three groups; however, disease and lesions were more severe in the VR2385 group. Immunohistochemical evaluation of formalin-fixed tissues revealed virus antigen in alveolar macrophages in lungs of 22/25, 14/25, 14/25, and 0/25 of the VR2385, VR2431, Lelystad, and control pigs, respectively. Follicular macrophages and dendritic cells in the lymph nodes of 14/25, 10/25, 10/25, and 0/25 pigs from the VR2385, VR2431, Lelystad, and control groups, respectively, stained positive for virus antigen. Similar cells in the tonsils from 25/25, 21/25, 23/25, and 0/25 pigs from the VR2385, VR2431, Lelystad, and control groups, respectively, stained positive for virus antigen. Other tissues and cells in which virus antigen was detected included macrophages and endothelial cells in the heart, macrophages, and interdigitating cells in the thymus, macrophages and dendritic cells in the spleen and Peyer's patches, and macrophages in hepatic sinusoids, renal medullary interstitium, and adrenal gland. PRRSV persisted in macrophages in the lung, tonsil, lymph node, and spleen for at least 28 days. Significantly more PRRSV antigen was detected in the lung (P < 0.01), lymph nodes (P < or = 0.05), and tonsils (P < 0.05) of the VR2385 pigs than was detected in the same tissues of the VR2431 and Lelystad pigs. The cell types in which PRRSV antigen was detected and the distribution of PRRSV antigen-positive cells within particular tissues and organs were generally similar for the different virus inoculation groups despite differences in virulence of the isolates.

Comparison of the pathogenicity of two US porcine reproductive and respiratory syndrome virus isolates with that of the Lelystad virus.

The Lelystad virus or one of two US isolates (VR2385, VR2431) of porcine reproductive and respiratory syndrome virus were given intranasally to 25 4-week-old cesarian-derived colostrum-deprived pigs. Pigs from these groups were necropsied at 1, 2, 3, 5, 7, 10, 15, 21, or 28 days postinoculation. The Lelystad virus and VR2431 induced mild transient pyrexia, dyspnea, and tachypnea. VR2385 induced labored and rapid abdominal respiration, pyrexia, lethargy, anorexia, and patchy dermal cyanosis. All three isolates induced multifocal tan-mottled consolidation involving 6.8% (n = 9; SEM = 3.4) of the lung for Lelystad, 9.7% (n = 9, SEM = 2.7) of the lung for VR2431, and 54.2% (n = 9, SEM = 4.4) of the lung for VR2385 at 10 days postinoculation. Characteristic microscopic lung lesions consisted of type 2 pneumocyte hypertrophy and hyperplasia, necrotic debris and increased mixed inflammatory cells in alveolar spaces, and alveolar septal infiltration with mononuclear cells. Lymphadenopathy with follicular hypertrophy, hyperplasia, and necrosis was consistently seen. Similar follicular lesions were also seen in Peyer's patches and tonsils. Lymphohistiocytic myocarditis and encephalitis were reproduced with all three isolates. Clinical respiratory disease and gross and microscopic lung lesion scores were considerably and significantly more severe in the VR2385-inoculated pigs. All three viruses were readily isolated from sera, lungs, and tonsils throughout the 28 days of the study. The lymphoid and respiratory systems have the most remarkable lesions and appear to be the major site of replication of these viruses. This work demonstrated a marked difference in pathogenicity of porcine reproductive and respiratory syndrome isolates.

Construction and insect larval expression of recombinant vesicular stomatitis nucleocapsid protein and its use in competitive ELISA.

The gene encoding the nucleocapsid (N) protein of Indiana 1 serotype vesicular stomatitis virus (VSV-IN1) was transferred into the genome of Autographa californica nuclear polyhedrosis virus (baculovirus) as a full-length non-fusion construct under the control of the polyhedrin gene promoter. Recombinant N protein was obtained from Trichoplusia ni insect larvae inoculated 72-96 h previously with the recombinant baculovirus. Polyclonal antibody (PAB) against VSV-IN1 was produced in mice using VSV-IN1 whole virus antigen concentrated from virus-infected cell culture fluids. The N protein and the PAB were used without further purification in a competitive enzyme-linked immunosorbent assay (C-ELISA) for detection of bovine, porcine, and equine origin serum antibodies against VSV-IN1. A limited number of field origin, experimental, and reference VSV antisera were evaluated using the C-ELISA and with a standard serum neutralization (SN) procedure. Sensitivity of the C-ELISA was comparable to the serotypically homologous SN procedure. Subject to further validation, similar C-ELISA tests for the other VSV serotypes, used in conjunction with the test described here, may offer the best combination of rapidity, sensitivity, simplicity, economy, and laboratory biosafety of any of the methods yet developed for VSV serodiagnosis.

Characterization of the humoral immune response to porcine reproductive and respiratory syndrome (PRRS) virus infection.

The development of the humoral immune response against porcine reproductive and respiratory syndrome (PRRS) virus was monitored by an indirect fluorescent antibody (IFA) test, immunoperoxidase monolayer assay (IPMA), enzyme-linked immunosorbent assay (ELISA), and serum virus neutralization (SVN) test over a 105-day period in 8 pigs experimentally infected with ATCC strain VR-2402. Specific antibodies against PRRS virus were first detected by the IFA test, IPMA, ELISA, and the SVN test 9-11, 5-9, 9-13, and 9-28 days postinoculation (PI), respectively, and reached their maximum values by 4-5, 5-6, 4-6, and 10-11 weeks PI, respectively, thereafter. After reaching maximum value, all assays showed a decline in antibody levels. Assuming a constant rate of antibody decay, it was estimated by regression analysis that the ELISA, IFA, IPMA, and SVN antibody titers would approach the lower limits of detection by approximately days 137, 158, 324, and 356 PI, respectively. In this study, the immunoperoxidase monolayer assay appeared to offer slightly better performance relative to the IFA test, ELISA, and SVN test in terms of earlier detection and slower rate of decline in antibody titers. Western immunoblot analysis revealed that antibody specific for the 15-kD viral protein was present in all pigs by 7 days PI and persisted throughout the 105-day observation period. Initial detection of antibodies to the 19-, 23-, and 26-kD proteins varied among pigs, ranging from 9 to 35 days PI. Thereafter, the antibody responses to these 3 viral proteins of PRRS virus continued to be detected throughout the 105-day study period.(ABSTRACT TRUNCATED AT 250 WORDS)

Antigenic differences between European and American isolates of porcine reproductive and respiratory syndrome virus (PRRSV) are encoded by the carboxyterminal portion of viral open reading frame 3.

Antigenic differences between European and American isolates of porcine reproductive and respiratory syndrome virus (PRRSV) were revealed by serologic analysis of a recombinant protein derived from PRRSV open reading frame 3 (ORF 3). The hydrophilic carboxyterminal 199 amino acids encoded by the ORF 3 of a European (Lelystad) isolate of PRRSV were expressed as a recombinant fusion protein (BP03-P) in a baculovirus gene expression system. Sera from gnotobiotic swine exposed to prototypic reference European and American isolates of PRRSV and sera from conventionally reared European and American swine convalescing from naturally acquired PRRSV infections were used to characterize the BP03-P protein. Sera from gnotobiotic and conventionally reared swine exposed to European isolates of PRRSV were significantly more reactive (P < 0.01) with BP03-P than were the corresponding American PRRSV antisera using the indirect immunoperoxidase monolayer assay (IPMA). Prototypic European, but not American, PRRSV antisera also recognized BP03-P using western immunoblotting and radioimmunoprecipitation assay (RIPA) procedures. However, gnotobiotically derived antiserum to an atypical American-origin PRRSV was reactive with BP03-P by both IPMA and western immunoblot. Despite a predicted potential for N-linked glycosylation, studies with tunicamycin and peptide-N-glycosidase F (PNGase F) indicated that BP03-P was not N-glycosylated in either insect cell cultures or Trichoplusia ni larvae infected with the recombinant baculovirus. Sera from rabbits inoculated with BP03-P failed to neutralize both the European (Lelystad) and American (ATCC VR-2332) reference isolates of PRRSV and did not react by IPMA with PRRSV-infected cell cultures. Taken together, the data suggest that the carboxyterminal portion of PRRSV ORF 3 encodes a non-neutralizing viral peptide that is partially responsible for the serologic differences noted between European and most American isolates of PRRSV.

Feral swine as a potential amplifying host for vesicular stomatitis virus New Jersey serotype on Ossabaw Island, Georgia.

Sentinel feral swine (Sus scrofa) on Ossabaw Island, Georgia (USA), were serologically monitored for antibodies to vesicular stomatitis New Jersey serotype (VSNJ) virus from 17 April to 27 August 1990. Seroconversions to VSNJ virus were detected in 24% of swine island-wide. Differences in the incidence of seroconversion were detected between swine sampled in the Pleistocene and Holocene formations of the island suggesting that the presence of virus is forest type dependent. Based on the consistency in onset and spatial distribution of seroconversions with data from 1981 to 1985, this is a very stable host-parasite system. Sequential virus isolation attempts from nasal swabs, tonsil swabs, and blood were made on a subsample of 54 sentinel swine from 9 May to 4 July 1990. The VSNJ virus was isolated from five swine from 16 May to 20 June. Vesicular lesions were detected on only two of these animals. Although infections in these feral swine were short-lived (< 7 days) and were followed by a strong neutralizing antibody response, VSNJ virus was detected in a single group of swine for a period exceeding 1 month. From these data, it appears that feral swine could provide a source of virus to feeding arthropods for extended periods of time. The failure to detect a viremia in these animals, however, indicates that a source other than blood may be required for transmission to occur.

Antibodies to vesicular stomatitis New Jersey type virus in white-tailed deer on Ossabaw Island, Georgia, 1985 to 1989.

From 1985 to 1989, 491 serum samples were collected from white-tailed deer (Odocoileus virginianus) on Ossabaw Island, Georgia (USA) and were tested for neutralizing antibodies to New Jersey and Indiana type vesicular stomatitis viruses. Prevalence of antibodies to vesicular stomatitis New Jersey (VSNJ) virus in deer for the 5-yr period was 43%. Prevalence of antibodies differed by year (P less than 0.0001), and was dependent on age class (P less than 0.0001) and location on the island (P less than 0.0001). Of 173 deer sampled from other locations in the southeastern United States, only two had VSNJ antibody titers normally considered positive (greater than or equal to 1: 32). The positive deer were from Union County, Arkansas (USA) and Wakulla County, Florida (USA). No evidence of exposure to vesicular stomatitis Indiana Virus was observed.

Parvovirus infection in pigs with necrotic and vesicle-like lesions.

Porcine parvovirus was isolated from many visceral organs and also from the brain, serum and skin specimens of swine with vesicular-like conditions. Severe lesions were reported to have occurred in the mouth, on the tongue and snout, on the coronary band and in the interdigital spaces. Also, parvoviral antigens were demonstrated, by immunofluorescence, in the outer layers of hair follicles in skin adjacent to coronary band lesions.