Quantitatively, the major circulating androgen in the blood is testosterone. Within the prostate, however, testosterone is enzymatically converted to 5a-dihydrotestosterone (DHT) (Wilson 1984). The class of enzymes responsible for the irreversible conversion of testosterone to DHT are the membrane-bound NADPH-dependent A4-3-ketosteroid 5a-oxidoreductases (i.e., 5a-reductases) (Bruchovsky and Wilson 1968). Biochemical studies have demonstrated that the irreversible conversion of testosterone to DHT by 5a-reductase (Fig. 12.1), involves a sequential series of steps (Levy et al. 1990). Initially, reduced nicotinamide-adenine din-ucleotide phosphate (NADPH) cofactor binds to the 5a-reductase enzyme to form a 5a-reductase-NADPH complex. Once formed, testosterone binds to this 5a-reductase-NADPH complex. Electrons are stereospecifically transferred from NADPH to reduce the A4 double bond of testosterone, producing a 5a-reductase-oxidized NADP+-5a-DHT complex. After 5a-DHT is produced, it must leave this complex before the bound NADP+ is able to leave, thus regenerating active 5a-reductase enzyme for another catalytic cycle (Levy etal. 1990).
There are two distinct 5a-reductase genes in man, each encoding a biochemically distinct isozyme. Both isozymes have been cloned and the complete DNA-based sequence and amino acid composition are now known (Andersson and Russell 1990; Jenkins etal. 1991; Labrie etal. 1992; Thigpen etal. 1992). The genes encoding the proteins for both 5a-reductase type 1 and 2 isozymes have a similar structure containing five exons separated by four introns. The two genes share ^46% DNA sequence homology and encode for a protein of ^29,000 molecular weight. The type 1 isozyme is encoded by a gene on human chromosome 5p15 (Jenkins etal. 1991). It has a neutral pH optimum, a requirement for high concentration of testosterone to saturate the enzyme (high Km = 3 |xM), and is rather insensitive to finasteride inhibition (Ki ~ 300 nM) (Andersson and Russell 1990; Jenkins et al. 1991). The type 1 isozyme is present at low levels in the prostate but is the predominant 5a-reductase isozyme in skin; it is also present in the liver (Jenkins etal. 1992; Normington and Russell 1992).
The type 2 isozyme is encoded by a gene on human chromosome 2p23 (Thigpen et al. 1992). It has an acidic (pH 5.0) optimum, has a lower Km (0.5 |xM) for testosterone, and is sensitive to finasteride inhibitor (Ki = 23 nM). The type 2 isozyme is the predominant 5a-reductase in androgen target tissue, including the
Type 2 and 6
Type I and II
Androgen Receptor N
Type 2 and 6
Type I and II
Androgen Receptor N
6a and 7a-triol
Fig. 12.7 Summary of the enzymatic pathway for androgen metabolism within the prostate.
prostate. Analysis of individuals with male pseudohermaphroditism caused by 5a-reductase deficiency has revealed no mutation in the type 1 isozyme gene (Jenkins etal. 1992). In contrast, molecular analysis demonstrated that mutation in the 5a-reductase type 2 gene accounts for this disorder (Andersson et al. 1991; Thigpen et al. 1992). Based on these results, it has been suggested that the type 1 isozyme functions in a catabolic manner in the metabolic removal of androgens by nontarget tissue, whereas the type 2 isozyme functions in an anabolic role to amplify the androgenicity of testosterone by effectively converting it to DHT within androgen target tissue (Normington and Russell 1992).
Once formed via the 5a-reductase type 1 or 2, DHT can reversibly bind to the androgen receptor to regulate prostatic cellular proliferation and survival. (Fig. 12.7). Alternatively, DHT can be further reductively metabolized to 5a-androstane-3a,17ß-diol (3a-diol) by the 3aHSD type 3 enzyme (i.e., also known as AKRIC2) (Rizner etal. 2003) (Fig. 12.7). Once formed,3a-diol can be re-oxidized back to DHT via an oxidative 3a-HSD enzyme not fully characterized in the normal prostate or glucuronidated at position 3 and excreted by the prostate (Rizner etal. 2003). 3a-diol can also be oxidized at its 17ß-hydroxy position by 17ß-HSD type 2ortype6enzymes to form5a-androsterone which can also be glucuronidated at position 3 and excreted (Biswas and Russell 1997; Rizner etal. 2003) (Fig. 12.7). DHT can also be either oxidatively metabolized at its 17ß-hydroxy group by the 17ß-HSD type 2 enzyme to form 5a-androstane-3,17 dione (Rizner et al. 2003) or reductively metabolized at its 3 keto group to produce 5a-androstane-3ß,17ß-diol (3ß-diol) by 17ß HSD type 7 enzyme (Torn etal. 2003) (Fig. 12.7). Interestingly, it has been documented that the endogenous estrogen in the prostate is not 17ß-estradiol but 3ß-diol (Weihua et al. 2001). Also it has been documented that within the normal prostate both the isotypes of the estrogen receptor (i.e., ERa and ERß) are expressed and both can bind 3ß-diol (Weihua etal. 2001). The ERa is expressed predominately in the prostatic stromal cells while ERß is expressed in the epithelial cells (Fixemer et al. 2003). The Gustafsson group which initially discovered ERß has postulated that 3ß-diol binding to the ERß within the prostatic epithelial cells results in antagonism of the AR signaling for proliferation (Weihua et al. 2002). The level of such an ERß dependent anti-proliferative effect is thus dependent upon the level of 3ß-diol. This 3ß-diol level is itself regulated by the activity of the CYP7B1 enzyme which hydroxylates 3ß-diol to 5a-androstane-3ß, 6a, 17ß-triol (6a-triol) and 5a-androstane-3ß, 7a, 17ß-triol (7a-triol) (Isaacs etal. 1979; Weihua etal. 2002) (Fig. 12.7).
The extensive metabolic pathway for androgen within the prostate functions as a means for autoregulation so that the prostatic level of DHT remains constant during the episodic and diurnal variations in both total and free serum testosterone levels (Plymate etal. 1989). Because growth versus regression (i.e. death) of the prostate is determined by the specific level of prostatic DHT (Kyprianou and Isaacs 1987), a constant prostatic DHT level is critical and is in turn required for the dose-dependent ability of DHT to bind to and regulate the function of the androgen receptors (Liao etal. 1972).
Androgen receptors are ligand-dependent zinc finger DNA binding proteins whose genomic binding co-ordinates formation of transcriptional complexes at the regulatory elements of targeted genes. The AR gene is located on the long arm of the X chromosome (i.e. Xq11.2), and encodes a protein with three critical domains: 1) anN-terminal domain (NTD) involved inhomotypic dimerizationand binding with other transcriptional co-activator or co-repressor proteins; 2) a DNA binding domain with two zinc finger binding motifs and hinge region, and 3) a C-terminal steroid ligand binding domain (LBD), which is also involved in homotypic dimerization and co-activation binding. This latter C-terminal LBD domain is also where 90-Kda heat shock protein (i.e., Hsp-90) dimers bind to stabilize the AR protein during folding subsequent to its synthesis (Chadli et al. 2000). Specific interaction with androgenic ligands results in the conformational activation of the androgen receptor. This allows the dissociation of the Hsp-90 dimer proteins and thus the binding and dimerization of the occupied androgen receptor (Langley et al. 1995) to androgen-response elements present in the promoter and enhancer regions in AR-regulated genes (Jain et al. 2002; Mitchell et al. 2000; Schuur et al. 1996; Watt et al. 2001; Zelivianski etal. 2002).
This initial genomic AR binding allows further binding to specific regions of the bound AR by additional nuclear proteins (i.e., transcriptional coactivator proteins like SRC-1, ARA 70, etc., and general transcription factors [GTF] like TFIIF and H) to produce transcriptional complexes which can activate or repress specific gene expression (Sampson etal. 2001). For activation, formation of an active tran-scriptional complex is required, resulting in site-directed chromatin remodeling via histone acetylation and methylation which enhances target gene expression (He et al. 2001; Kang et al. 2002; Sampson et al. 2001; Shang et al. 2002; Xu et al. 1998). SRC-1 is a member of the p160 transcriptional coactivator gene family that includes SRC-1, TIF2 (also termed GRIP-1 and SRC-2), and p/CIP (also termed RAC3, ACTR, AIBI, and SRC-3) (89). Cell-free in vitro transcription and in vivo experiments have indicated that the SRC-1 family members enhance androgen receptor-dependent transactivation of nuclear genes. The mechanism for such enhancement involves binding ofp160 proteins to the DNA-bound AR. This allows the p160 to acetylate histones via its histone acetyltransferase (HAT) activity. Additional coactivators with HAT activity such as CBP, p300, or p/CAF also bind to the p160/AR complex. This results in chromatin remodeling and additional binding of GTFs such as TBp and TIFIIB with the AR coactivation complexes (He et al. 2001; Kang et al. 2002; Sampson et al. 2001; Shang et al. 2002; Xu et al.). These AR-coordinated complexes regulate the expression of a series of genes resulting in the complex differentiation and growth of the prostate (Coffey 1992). The critical importance of DHT and its receptor in this developmental process has been demonstrated by the fact that the prostate does not develop in males who have inherited either a stop mutation that prevents AR expression (Gottlieb etal. 1999) or an inactivating mutation in the type II 5a-reductase gene, thus preventing high prostatic DHT formation (Imperato-McGinley et al. 1980), even though serum testosterone levels are normal in individuals with either type of mutation.
Was this article helpful?