A highly specific heterologous enzyme-linked immunosorbent assay for measuring testosterone in plasma using antibody-coated immunoassay plates or polypropylene tubes.

A highly specific and sensitive enzyme-linked immunosorbent assay, using a heterologous combination of antiserum raised against testosterone-3-(O-carboxymethyl) oxime-bovine serum albumin and penicillinase-labeled testosterone-11 beta-carboxymethyl ether, was developed for measuring testosterone in human plasma. Immunoassay plates (96 wells) provided a sensitivity of 2.5 pg/well. This was achieved by maintaining the molar ratios of steroid to enzyme between 10 and 20. The assay was very specific for testosterone and did not show any cross-reaction with the related C19 steroids tested. Replacement of immunoassay plates with the locally available polypropylene tubes raised the detection limits to 25 pg/tube, but improved the range of doses of testosterone that could be measured up to 10,000 pg. The antiserum to testosterone derivative was linked to both immunoassay plates and polypropylene tubes through immunochemical bridges. Comparison of testosterone values of 52 plasma specimens obtained by both solid phase methods with those of radioimmunoassay showed good correlation.

Influence of different combinations of antibodies and penicillinase-labeled testosterone derivatives on sensitivity and specificity of immunoassays.

Three antisera raised against bovine serum albumin (BSA) conjugates of testosterone-3-(O-carboxy-methyl)-oxime (T-3-CMO), 11 beta-hydroxytestosterone-11-carboxymethyl ether (T-11 beta-O-CME) and 19-hydroxytestosterone-19-carboxymethyl-ether (T-19-O-CME) were evaluated in enzyme immunoassays (EIAs) in combinations with penicillinase-labeled T-3-CMO, T-11 beta-O-CME, T-19-O-CME, and testosterone-17 beta-hemisuccinate (T-17 beta-HS) for their influence on the sensitivity and specificity of EIAs. Of the various combinations, anti-T-3-CMO antiserum along with T-11 beta-O-CME-penicillinase showed no cross-reaction with any of the closely related steroids, although the same antibody had 21.6% binding to 5 alpha-dihydrotestosterone (5 alpha-DHT) in radioimmunoassay. All the homologous combinations appeared to be less sensitive due to their low affinity for testosterone. It was also apparent that of all the heterologous systems tested, only two combinations, (a) anti-T-19-O-CME antiserum and T-3-CMO-penicillinase and (b) anti-T-3-CMO antiserum and T-11 beta-O-CME-penicillinase, were found to be more sensitive. The former was less specific; it showed 70% cross-reaction with 5 alpha-DHT. The ability of testosterone to displace the hapten-enzyme conjugate and the specificity of the assay appear to depend on the position of the enzyme label on the steroid molecule as well as on the availability of antigenic sites in particular combinations of antibody and hapten-enzyme conjugates.