Solid phase enzyme immunoassay of cyclic adenosine 3',5'-monophosphate. Effect of coating strategy upon assay performance in comparison with radioimmunoassay.

We have evaluated two novel enzyme-linked immunosorbent assays (ELISAs) used to quantitate cyclic AMP. In one assay ELISA plates are coated with antigen consisting of a cyclic AMP-polylysine conjugate. Cyclic AMP samples added to plates are quantified by their ability to decrease the binding and anti-cyclic AMP antibodies to the coated antigen. A second ELISA utilizes a plated anti-immunoglobin technique in which plates are coated first with anti-goat IgG and then with goat anti-cyclic AMP antiserum. Cyclic AMP samples are quantified by their ability to compete with cyclic AMP-peroxidase conjugates for binding to the plated anti-cyclic AMP antibodies. The plated anti-immunoglobin ELISA proved to be somewhat more sensitive than the plated antigen ELISA and was comparable in sensitivity to an automated RIA for measuring cyclic AMP in standards and urine samples. Our data fit with the generalization that optimal ELISA sensitivity is obtained through the use of plated anti-immunoglobins rather than plated antigens. Further they demonstrate the practicality of utilizing small ligand-enzyme conjugates for ELISAs.

Measurement of glycosylated hemoglobins using boronate affinity chromatography.

A boronate affinity column method for the measurement of glycosylated hemoglobins was evaluated. In the procedure the glycosylated hemoglobins were bound by immobilized boronic acid to separate them from nonglycosylated hemoglobins. Elution of bound glycosylated hemoglobins was carried out with sorbitol buffer, and the absorbance was read at 414 nm. The method was linear to a glycosylated hemoglobin concentration of at least 20 percent. The precision of the method ranged from 1.2 to 2.8 percent (C.V.) within-run, and 3.4 to 5.3 percent day-to-day. The reference interval was 4.8 to 6.4 percent. The method correlated with a cation exchange resin mini-column method (r = 0.94) and a colorimetric method (r = 0.93) but results from the boronate affinity method were higher in diabetic patients. The measured glycosylated hemoglobin was significantly correlated with estimated one-day-mean plasma glucose in diabetic patients (r = 0.54, n = 52, p less than 0.002). The affinity method provides an attractive alternative to earlier methods for measuring glycosylated hemoglobins.