Tissue-specific pharmacology of testosterone and 5 alpha-dihydrotestosterone analogues: characterization of a novel canine liver androgen-binding protein.

The mechanism by which the hormones 5 alpha-dihydrotestosterone and testosterone differentially regulate such diverse functions as development of male internal and external genitalia and maintenance of prostatic growth via a single androgen receptor (AR) is not well understood. To search for potential AR isoforms, an extensive pharmacological survey of the binding of [3H]mibolerone (7 alpha,17 alpha-[3H]dimethyl-19-nortestosterone) in dog prostate, adrenal gland, testis, liver, kidney, brain, muscle, and spleen cytosolic extracts was carried out. The antagonist androst-4-en-3,17-dione (ATD), as well as a series of unsaturated analogues of testosterone, exhibited marked tissue specificity for binding to mibolerone-binding proteins (MBPs), with ATD having a 10-fold higher affinity for the MBPs present in liver than for those in prostate and testis. The difference in affinity was not due to tissue-specific metabolism of ATD. Competition binding profiles for ATD with mixtures of prostate and liver extracts were consistent with two distinct populations of binding sites. Both wild-type human AR-B and the recently discovered human AR-A isoform were expressed in COS cells and were found to exhibit pharmacology similar to that of the prostatic MBPs in dogs. Analogues of ATD or testosterone could prove to be useful probes for delineating the differential effects of 5 alpha-dihydrotestosterone and testosterone on the biological actions of the AR and related proteins.

Active-site-directed photoaffinity radioiodination of androgen-binding proteins.

Androgen-binding protein (ABP) in rat epididymal cytosol and sex hormone-binding globulin (SHBG) in rabbit serum and SHBG purified from human serum were active-site-directed photoaffinity radiolabeled with 17 alpha-[(E)-2-[125I]iodoethenyl]androstan-4,6-dien-17 beta-ol-3-one ([125I]1). The interaction of this compound with binding components in epididymal cytosol was dependent on exposure of the mixture to ultraviolet light and on the duration of exposure. Photolysis in the presence of [125I]1 and 5 alpha-dihydrotestosterone (5 alpha-DHT) resulted in a 40% inhibition of binding of [125I]1 to cytosolic components. These result indicate that, while [125I]1 interacted with 5 alpha-DHT binding sites, it also formed adducts with other sites. To characterize the labeled species, the photolysis mixture was subjected to electrophoresis under denaturing and reducing conditions. Autoradiography of the gel revealed that ABP and SHBG were labeled with [125I]1, but in cytosol and serum, higher and lower molecular weight components were also labeled. Purified SHBG was labeled, but no labeled contaminating protein was detected. The presence of 5 alpha-DHT completely inhibited [125I]1 photolabeling of human and rabbit SHBG and of ABP. However, in cytosol, the presence of 5 alpha-DHT also eliminated photolabeling to a component that may be albumin, but 5 alpha-DHT did not affect [125I]1 photolabeling of other contaminating proteins in cytosol. Thus, while [125I]1 is an effective photoaffinity radiolabel for ABP and SHBG, the observation that it also photolabels other proteins limits its practical use to the radiolabeling of purified ABP and SHBG preparations.