Two indirect ELISA containing outer membrane protein (OMP) and lipopolysaccharide (LPS) antigens from a field isolate of Salmonella choleraesuis var kunzendorf were developed and evaluated in experimentally infected and uninfected control pigs. Experimentally induced infection with S choleraesuis was successfully established in 10 pigs by oral inoculation with 10(8) organisms, and 3 pigs died of clinical salmonellosis at postinoculation (PI) weeks 1, 2, and 4. Swab specimens from tonsils, nostrils, and rectum of pigs were obtained for culture, and sera were evaluated at weekly intervals for 9 weeks after inoculation. The ELISA containing OMP and LPS antigens with either anti-swine IgG or protein albumin-to-globulin ratio (antiglobulin) conjugates were standardized for serologic evaluation. All 4 ELISA (2 OMP and 2 LPS) detected seroconversion by PI week 3 and had sensitivities and specificities of 97.8 and 88.8, 100 and 100, 95.6 and 88.8, and 93.3 and 72.5%, at their ideal cutoff points (negative mean optical density + 2 SD). There was excellent agreement between all 4 ELISA systems as determined by kappa values. Cultures of fecal, tonsil, and nasal swab specimens were positive for S choleraesuis until the fourth week of infection. Fecal swab specimens from 1 pig were positive for S choleraesuis until PI week 7. Persistent infection after antemortem culture results were negative was detected by all 4 ELISA, which indicated consistently high titers until the end of PI week 9. Conventional bacteriologic examination of intestines, mesenteric lymph nodes, bone marrow, lung, liver, spleen, and bile yielded positive results for S choleraesuis in the 3 pigs that died of clinical infection, whereas results were negative in the other 7 pigs infected by the end of PI week 9. Histologic examination of lung, liver, spleen, intestines, and mesenteric lymph nodes from the 3 pigs that died of S choleraesuis infection revealed severe ulceration and inflammatory cell infiltration.

An ELISA was developed to detect antibodies to the 41-kd flagellin (P41) of Borrelia burgdorferi in serum obtained from cattle. Absorption studies, immunoblot analysis, immunoelectron microscopy, and correlation of results of the P41-ELISA and the P39-ELISA as well as measurement of the antibody to P41 in calves challenge-exposed with Borrelia theileri were used to assess the specificity of the P41-ELISA. Antigens derived from Escherichia coli, Leptospira interrogans serovar hardjo, and B burgdorferi were used for absorption studies and immunoblot analysis. Antibodies to P41 of B burgdorferi cross-reacted with antigens of E coli, but were not cross-reactive with L hardjo. A value 3 SD higher than the mean of the negative-control population of cattle was defined as the minimum value cutoff value) for a positive result by the P41-ELISA. Use of this value for classification of test results reduced the predicted rate of false-positive results attributable to E coli cross-reactivity to 1%. Immunoblot analysis revealed that test-positive serum from cattle reacted mainly with 41-, 39-, 34-, and 31-kd proteins of B burgdorferi, as well as several smaller proteins. Immunoelectron microscopy revealed that serum from cattle that was test-positive by the P41-ELISA bound to the flagellin and outer membrane of B burgdorferi. Results of absorption studies, immunoblot analysis, and immunoelectron microscopy were correlated and indicated that serum from cattle that was test-positive by P41-ELISA had stronger reactivity to B burgdorferi antigens than to antigens of E coli or L hardjo. The concentrations of antibodies measured by P41-ELISA and P39-ELISA testing were highly correlated [R(2)=0.78]. Calves challenge-exposed with B theileri also had test-positive results by the P-41-ELISA as early as 2 weeks after exposure, but serum antibody concentrations decreased to prechallenge-exposure concentrations by 9 weeks after exposure.

Immunoassay of serum ferritin is currently used to evaluate the clinical iron status of human beings, horses, cattle, and swine. Because ferritins are immunologically species specific, a separate assay must be developed for each species. We have developed an ELISA for serum ferritin in dogs, using a monoclonal anti-canine ferritin antibody. Ferritin standards were linear (r = 0.997) from 0 to 80 ng/ml. Recovery of ferritin from canine serum was 94%. Dilutions of pooled canine serum were linear from 0 to 50% (r = 0.994). Within-assay coefficient of variability was 5.5%, whereas assay-to-assay coefficient of variability ranged from 12.5 to 21%. This assay should provide a nonsurgical means of accurately estimating dogs' iron stores.

To measure the concentration of serum amyloid A (SAA) protein in horses a sensitive and highly reproducible sandwich (ELISA) was established, using affinity purified SAA antibody. Results of the ELISA were found to have a high correlation (r = 0.95) with those of the single radial immunodiffusion test. Equine SAA concentration was measured by use of this ELISA. In clinically normal horses, the concentration of SAA was high immediately after birth to 2 weeks of age. After that, SAA concentration had periodic fluctuations in the range of approximately 10 to 30 microgram/ml. Mean (+/- SD) concentrations of SAA in foals (less than or equal to 12 months old) and adult horses (greater than or equal to 18 months old) were 21.23 +/- 12.20 and 14.93 +/- 9.07 microgram/ml, respectively. In mares during the perinatal period, SAA concentration remained stable within the reference range before parturition. It increased quickly after delivery, and reached a peak value of 101.29 +/- 98.82 microgram/ml on postpartum day 3, then began to decrease, at postpartum week 2, to the reference range by the end of postpartum month 1. In horses with experimentally induced inflammation, SAA concentration increased quickly and reached approximately four- to 40-fold increase over the pretreatment value on day 1 and remained high on days 2 to h after treatment. It then returned to the baseline value by 2 to 4 weeks in association with disappearance of local signs of inflammation. The SAA concentration was high in most horses with clinical signs of inflammation. It was concluded from these data that this ELISA is sensitive and reliable for measuring SAA in horses.

We investigated the ability of an antibody-specific, O antigen-based ELISA to document Salmonella typhimurium herd infections by screening of milk samples. Three cattle populations, 20 herds with no history of salmonellosis, 8 herds with history of S. typhimurium episodes within the previous 7 months, and 220 herds of unknown disease status, were tested. A herd was considered ELISA positive if at least 5% of the cows had OD values > 0.3. Among the 20 herds without history of salmonellosis, only 2 herds were ELISA positive, whereas all 8 herds with a known history of salmonellosis were ELISA positive (herd specificity, 0.9 and herd sensitivity, 1.0). A significant correlation (P < 0.001) was found between the OD values of serum and milk samples from cows in the herds with a history of salmonellosis. It was concluded that ELISA testing of individual milk samples can be used for surveillance of herds for S. typhimurium infections, but further modifications are needed to test bulk tank milk samples.

A commercial rapid-absorbed ELISA developed to detect antibodies to Mycobacterium paratuberculosis in bovine serum was modified for use with goat serum. Diagnostic sensitivity was evaluated, using a group of 163 goats from a herd with endemic paratuberculosis. Blood and fecal samples were obtained simultaneously, and prevalence of shedding of M paratuberculosis in the feces was estimated by detection of DNA of the mycobacterial insertion sequence, IS900, using a commercial test kit. Diagnostic specificity was evaluated, using blood samples from a total of 123 goats in 10 herds that were considered clinically free of paratuberculosis. The IS900 DNA was detected in 35 of the 163 goats (21%) from the infected herd. Serum antibody to M paratuberculosis was detected in 19 of the 35 IS900 DNA-positive goats, for apparent sensitivity of 54%. Serum antibody was detected in 18 of the 128 IS900 DNA-negative goats from the infected herd. Negative results for serum antibody to M paratuberculosis were obtained for all 123 goats from the herds that were considered clinically free of paratuberculosis.

An ELISA was developed and tested to detect antibodies to Eperythrozoon suis in swine. Results were compare with those of the indirect hemagglutination (IHA) test. Antigen isolated from swine heavily infected with E suis was used for both tests. Comparison of the ELISA with the IHA test revealed a significant (P < 0.001) correlation between results. Of 114 samples obtained from 9 swine infected with E suis, 8 7.7% were seropositive (titer greater than or equal to 200) via the ELISA, and 80.7% were seropositive (titer greater than or equal to 20) via the IHA test. The sensitivity of the ELISA was greater than that of the IHA test. All blood samples obtained from specific-pathogen-free swine tested negative for E suis antibody. Cross-reactions were not observed between E suis antigen and antisera against various swine and cattle disease agents using ELISA. We concluded that the ELISA may be used for rapid and effective diagnosis of infection with E suis in swine.

Serum samples obtained from 40,927 swine at various locations in North Carolina between Aug 1, 1987 and July 31, 1988, were tested for antibodies to Trichinella spiralis, using an ELISA based on a larval T spiralis excretory-secretory antigen. In the ELISA, samples were considered to have positive results if the optical density (OD) reading was equal to or 5 times greater than the mean OD value of 4 negative-control sera from trichina-free swine. Of the 40,927 serum samples tested, 154 (0.38%) were positive by ELISA; the rate for breeding swine was 0.35% (105/30,162), and the rate for cull swine was 0.45% (49/10,765). Of the 49 seropositive samples from cull swine, 11 were from out of the state, 22 had no identification, and 16 were known to originate from North Carolina. Seropositivity had a bimodally seasonal distribution, with peaks in March and September. There was no difference between the mean age of seropositive and seronegative swine, but males were at greater risk for seropositivity than were females. Pigs from lots with < 100 sera tested were at increased risk for seropositivity, as were pigs from the central coastal region of North Carolina.

An ELISA was used to detect IgG, IgM, and complement (C3) on the surface of canine erythrocytes. Erythrocytes were placed in wells of a microtitration plate and incubated with affinity purified, alkaline phosphatase-conjugated anti-canine IgG, IgM, or C3. Results of the ELISA were compared with the direct antiglobulin test (DAT) by preparing standard reference curves from canine blood type A erythrocytes that had been incubated with serial dilutions (1:2 to 1:8,192) of canine anti-A serum. The ELISA detected increased erythrocyte-bound immunoglobulin and complement at two- to fourfold dilutions greater than thoe required for positive results with the DAT. The ELISA required small sample and reagent volumes and detected lower concentrations of immune components than did the DAT.