[A double adsorption technique for the determination of corticosteroid-binding globulin in human serum (author's...

Despite the widespread use of many different estrogenic compounds in hormonal contraception and estrogen replacement, in the human a simple test for comparing the potencies of the compounds is lacking. Many of the estrogenic compounds are difficult to measure, and animal studies may not relate directly to those done on humans in terms of potency. Corticosteroid-binding globulin (CBG) in human serum has been known since 1958 to increase in response to estrogen administration, and the degree of its elevation has been shown to be closely dose-related and reasonably sensitive. Therefore, CBG represents a biological parameter that provides for the specific quantal analysis of estrogenicity in the human. In this paper a simple and reliable method for the determination of the cortisol-binding capacity of CBG (CBG-BC) in human serum is described. The procedure is based on the amount of cortisol required to saturate the CBG binding sites. All endogenous steroids are initially removed from the serum by adsorption on dextran-coated charcoal (DCC). A mixture of a known amount of unlabeled and labeled cortisol is added to the sample in order to saturate the "stripped" binding sites of CBG. The unbound cortisol is removed by DCC at 4 degrees C, and the supernatant represents the total cortisol-binding capacity of the serum sample. To distinguish between cortisol-binding to CBG and to albumin, "stripped" serum is heated for 30 min at 60 degrees C, the temperature which does not influence cortisol binding by albumin but which does inactivate CBG. The samples are subsequently equilibrated with a mixture of labeled and unlabeled cortisol. The free cortisol is removed by DCC and the bound cortisol remaining in the supernatant is designated as the albumin-bound cortisol. Thus, CBG-BC is assayed by a double adsorption technique as differential binding to native and heat-inactivated serum. The cortisol determined by this method corresponds to CBG-bound cortisol, and CBG-BC is expressed as micrograms of cortisol bound per 100 ml serum. The precision and accuracy of the method were found to be good. The values of CBG-BC obtained by this method were 14.9 +/- 1.9 (SD) microgram/100ml in adult men and 16.0 +/- 1.8 microgram/100ml in adult women. Daily serum samples in five women revealed no significant variations in CBG-BC throughout the normal menstrual cycle despite fluctuating levels of serum estradiol. CBG-BC was also found not to vary significantly with age. In pregnancy, however CBG-BC increased significantly.