|[Frontiers in Bioscience 1, d59-71, March 1, 1996]|
THE ANDROGEN RECEPTOR: A MEDIATOR OF DIVERSE RESPONSES
Evan T. Keller, William B. Ershler, and Chawnshang Chang
The Institute on Aging and the Department of Human Oncology, University of Wisconsin, Madison, WI 53706, USA.
Received 01/16/96; Accepted 02/22/96; On-line 03/01/96
Perhaps the most dramatic example of the difference of T and DHT actions are observed during male sexual development. It has long been recognized that T is required for the development of the internal accessory sexual organs, whereas DHT is required for the development of external sex organs. In humans, these events appear to be triggered by T expression which begins at the 8th week of gestation (68). T, secreted by the Leydig cells, stimulates the Wolffian duct to differentiate into the epididymis, vas deferens, and seminal vesicles. The observation that unilateral orchiectomy results only in ipsilateral loss of Wolffian duct differentiation suggests that high concentrations of T are required for this event. The action of 5a-reductase on T results in the production of DHT which stimulates the differentiation of the urogenital sinus into the prostate, penis, and scrotum. The importance of DHT in this role is demonstrated in individuals with congenital 5a-reductase deficiency. Even though genotypically male (46, X, Y), these individuals cannot convert T to DHT and thus develop feminized external genitalia (69). However, because T is present during development, Wollfian duct differentiation proceeds normally resulting in formation of a pseudohermaphrodite. When coincident with puberty the level of T increases some of these individuals develop male characteristics such as increased skeletal muscle bulk, axillary and pubic hair growth, and maturation of the penis (70). This observation suggests that high levels of T may replace the loss of DHT. In agreement with this hypothesis are the following observations made by Imperato-McGinley et al (71): 1) flutamide, an androgen receptor blocker, administered to male rats in utero during the period of sexual differentiation completely inhibited the development prostate; and 2) administration of finasteride, a 5-a reductase inhibitor, resulted only in partial inhibition of development of prostate. Taken together, these results suggest that T, via the AR, can compensate for lack of DHT.
If as discussed below one subscribes to the idea that both T and DHT mediate their effects by interacting with a single nuclear receptor protein, then how can we account for the requirement of both T and DHT for sexual development? The answer may be found by examining the ability of these androgens to interact with the AR. Several investigators have demonstrated that T has approximately 3 times faster association and dissociation rates than DHT on both the rat (72) and human (73, 74) AR. In agreement with these binding kinetics, Zhou et al. have demonstrated that T is less effective at stabilizing AR than DHT (74). These observations suggest that DHT, by enhancing the stabilization of AR and its action, amplifies the T signal in those tissues which contain 5a-reductase. Perhaps in these tissues, the AR expression is not sufficient for T to mediate a physiologic response, but due to the ability of DHT to enhance AR activity, a response is observed.
In contrast to the inability of T to mediate various androgen-dependent events during sexual development, the biology of hair growth offers an example of the different effects that androgens exert on the proliferation of similar populations of epithelial cells. Specifically, T can stimulate facial hair growth, but causes the regression of scalp hair in aging individuals (75). Hair follicles are intimately associated with the mesenchymally-derived dermal papilla which is believed to provide an important influence on the follicular proliferation. Several lines of evidence support the role of androgens in controlling growth of hair follicles through modulation of the dermal papilla activity: 1) AR has been identified by both ligand binding assay (76) and by immunohistochemical staining in the dermal papilla (77); 2) dermal papillae from androgen-dependent hair follicles contain a greater number of AR than those in non-balding areas (78); 3) in primary dermal papilla cell lines from either androgen-dependent (i.e., beard) or -independent (i.e., non-balding scalp) hair follicles, only the beard dermal papilla-derived cells were able to synthesize DHT when T was added to the media (75); 4) the level of 5a-reductase varied between hair follicles from frontal and occipital sites (79); and 5) dermal papilla can produce extracellular matrix components and mitogenic factors (75). Taken together, these findings suggest that androgen, via the AR, can indirectly mediate an effect on hair follicle proliferation through modulating dermal papilla activity. Though these data do not explain how T modulates cell proliferation in hair follicles, the differences in androgen metabolism and AR expression may, in part, account for the opposite proliferative responses observed in various epithelial tissues. The effect of androgens on cell proliferation in hair follicles may be regulated through regulation of expression of growth factors. Androgens can modulate expression of a variety of growth factors in the prostate stroma.
Mesenchymal-epithelial interactions are critical in the development of prostate tissue (reviewed in (80)). Prostate mesenchyme, when coincubated with Sertoli cells, synthesizes extracellular matrix in an androgen-independent manner; however, when exposed to androgen it produces a diffusible substance similar in action to P-Mod-S (81). P-Mod-S, a protein secreted by peritubular cells, can activate a variety of responses such as inhibin production and aromatase activity in Sertoli cells. Its effect on the prostate epithelia is currently unknown; however, it may also stimulate the activity similar to that observed in Sertoli cells thus accounting for one mechanism of mesenchymal-epithelial interaction within the prostate. Gleave et al. demonstrated that prostate fibroblasts secrete a diffusible substance which can stimulate the growth of LNCaP prostate carcinoma cells (82). Androgen deprivation mediated by castration resulted in expression of tenascin in the rat prostate (83).