A sensitive radioimmunoassay optimized for reproducible measurement of rat plasma insulin.

There has been a notable lack of consistency in plasma insulin values reported in the literature. In an attempt to find a more reproducible and sensitive insulin radioimmunoassay (RIA) method for the measurement of plasma insulin, the charcoal adsorption method for separation of free from antibody-bound antigen was investigated. The applicability of this method towards the measurement of plasma insulin, however, has been known to be severely limited because of interference by plasma proteins with charcoal binding. Indeed, the assay resulted in negative fasting insulin values and unacceptable interassay variability, which was not improved by adding charcoal-extracted plasma (CEP) to the standard curve. In this paper, we demonstrate a process by which relatively small volumes (25 microL) of plasma from control, diabetic, and fasted rats can be assayed reproducibly with charcoal in the final separation step. It was observed that the assay of insulin-free plasma, in the form of CEP, could represent a distinct "zero insulin" level specific for plasma samples, in place of the zero insulin standard (insulin-free standard buffer) routinely used. Normalizing plasma samples against CEP resulted in lower intra-assay and interassay variability (coefficient of variation < 10%) and enhanced sensitivity to 7 microU/mL. Recovery was tested by adding 25 microL of known insulin standards to plasma samples in which insulin levels had been determined previously. Whereas the recovery of total insulin in the presence of plasma was low (38%-89%), recovery of total insulin from two standards was > 95%. Following correction, recovery of total insulin in 25-microL plasma was improved to > 90%. Hence, this approach provides a simplified way of correcting for plasma effects and greatly enhances the feasibility of using the charcoal-separation technique for the measurement of plasma insulin.