Matrix effects on an antigen immobilized format for competitive enzyme immunoassay of salivary testosterone

Matrix interferences in salivary testosterone enzyme immunoassays are important in the development of direct ELISA. An alternative format for sensitive enzyme immunoassay of testosterone was developed using immobilization of the antigen as part of a protein conjugate using oligoethylene glycol linker to project the antigen, with color development via enzyme labeled secondary antibody. This technique gave the required sensitivity for detection of testosterone from male saliva with a limit of detection (LOD) of 8.9 pg/mL (31 pmol/L). Application of the immunoassay directly in human saliva gave suppression of binding signals for the samples, indicating clear matrix interference with antibody binding. A wide variety of treatments were used in an attempt to overcome this effect, including use of both synthetic saliva and stripped human saliva for standard preparation, use of bovine serum albumin (BSA) to reduce non-specific binding of contaminants, pH adjustment of samples and/or standards and use of different antibodies. None of these techniques proved effective, causing either substantial suppression or enhancement of signal, and dilution was not possible because of the very low physiological concentrations. Complete removal of the saliva medium by chemical extraction was the only technique studied that could overcome this problem. These issues have not been explored previously for direct ELISA of salivary testosterone using alternative assay formats and have implications for the design of small molecule plate-based immunoassays in this medium.