COMPARISON of ANALYTICAL SENSITIVITIES of THREE ENZYME IMMUNOASSAY PROCEDURES FOR the DETECTION of SALMONELLA LIPOPOLYSACCHARIDE

Abstract

The analytical sensitivities of three different enzyme linked immunoassays (ELISA), two competitive and a capture format were assessed. the assay systems employed monoclonal antibodies to Salmonella lipopolysaccharide (LPS) outer core epitopes to detect crude LPS antigens from Salmonella typhimurium. the most sensitive ELISA was the capture procedure, being capable of detection 1.3 ng/ml of LPS. This technique, however also gave the greatest between‐test variation and as a result, the lowest amount that could be detected with a 95% confidence limit was actually 12.8 ng/ml and it took the longest time to perform (3 h, 30 min). A competitive ELISA using limiting monoclonal antibody to compete between solid phase antigen and soluble antigen in the sample, ranked second in sensitivity, and can detect 2.8 and 3.8 ng/ml of LPS when tested with two different monoclonal antibodies. However, because of the slight between test variation, the actual sensitivities that could be detected with a 95% confidence limit were 3.1 and 4.6 ng/ml, respectively. This test takes approximately 1 h and 30 min to perform.The classical type of competitive assay, employing a labelled antigen, was the least sensitive being capable of detecting 5.8 ng/ml if the LPS was conjugated with horseradish peroxidase and 16.0 ng/ml if alkaline phosphatase was used as a label. to account for the between‐test variation, the sensitivities with a 95% confidence limit were 8.6 and 18.7 ng/ml for the respective assays, which take 2 h and 15 min to perform.These sensitivities compare favorably with those published for similar assays, but all of the procedures were judged insufficiently sensitive for direct use on food samples to be tested for the presence of Salmonella species. However, the assays would be quite suitable for demonstration of Salmonella sp. after an enrichment procedure.