Sequential transformation of 4-androstenedione into dihydrotestosterone in prostate carcinoma (DU-145) cells indicates that 4-androstenedione and not testosterone is the substrate of 5α-reductase
related.springerprotocols.com/lp/de-gruyter/sequential-transformation-of-4-androstenedione-into-mJDgAzg6cg
Samson, Melanie; Labrie, Fernand; Luu-The, Van
Hormone Molecular Biology and Clinical Investigation , Volume 1 (2)
de Gruyter – Jan 1, 2010
Introduction Androgens play a crucial role in the development, growth, and control of function of the prostate as well as the diseases associated with the prostatic gland, such as benign prostatic hyperplasia (BPH) and prostate cancer. It is well recognized that there are two androgens, namely testosterone (T) and dihydrotestosterone (DHT). T is produced in the testis by the conversion of 4-dione into T by the enzyme type 3 17bHSD (17b-HSD3) (1) and promotes the formation of the internal male reproductive structures (epididymes, seminal vesicles, and vas deferens). A defect in the HSD17B3 gene causes disorders of human intersex called male pseudohermaphroditism. The production of DHT requires a 5a-reduction step of 3-keto-4-enesteroids catalyzed by the enzyme 5a-reductase. A defect in 5a-reductase also causes male pseudohermaphroditism (2) with ambiguous external genitalia. However, in contrast to HSD17B3 gene deficiency, the Wolffian structures are normally differentiated (3). DHT is mainly produced in a long series of peripheral tissues, including the liver, skin, and prostate where 5areductases are expressed (46). T and DHT act via the same androgen receptor (AR), because a defect in AR causes the X-linked androgen-insensitive syndrome (AIS) (3, 7), in which the development of both the internal and external male reproductive structures are altered. The above-mentioned data strongly suggest that depending upon the steroid precursors and enzymes present in the cells or tissues, AR is modulated by T or DHT. This is, somehow, in contradiction with the general belief that DHT is globally produced by 5a-reduction of T (6, 8). In addition, the higher affinity of T for AR (Km;10 8, 10 9 M) than 5a-reductases (Km;10 6 M) strongly suggests that in tissues that express both AR and 5a-reductases T will preferentially bind to AR. The reduction of T to DHT step does not seem substantiated by enzymatic kinetics. This is in agreement with the finding 2010/3 Article in press - uncorrected proof 68 Samson et al.: Pathway of DHT biosynthesis that does not require testosterone as intermediate Materials and methods Cell culture The prostate carcinoma cells DU-145 obtained from ATCC (HTB81) (Manassas, VA, USA) were maintained in MEM (Invitrogen Life Technologies, Inc.) supplemented with 10% (vol/vol) FBS (Wisent Inc., St-Bruno, QC, Canada) at 378C under a 95% air/5% CO2 humidified atmosphere. 5a-reductase inhibitor Finasteride (1 mM), synthesized in our laboratory under the name EM-351, was used as 5a-reductase inhibitor. It was added directly in cell culture media 10 min before addition of 4-dione. Figure 1 Schematic representation of the two last steps of 4androstenedione (4-dione) metabolism. DHT, dihydrotestosterone; ADT, androsterone; HSD, hydroxysteroid dehydrogenase. Assay of enzymatic activity that 4-dione is a better substrate for 5a-reductases than T (4, 9). Together, these data strongly suggest that DHT is produced by a pathway that does not require T biosynthesis, as suggested by Luu-The et al. (10). According to this pathway (10), the 5a-reductase activity step precedes the 17b-HSD activity step (Figure 1). The profile of 5a-reduced metabolites described in the literature (11) showing very high levels of 5a-reduced steroids in the circulation, namely in the micromolar range, also supports the existence of a predominant pathway where the 5a-reduction step precedes the 17keto-reduction step. Indeed, the concentration of 5a-reduced steroids in men between 20 and 30 years (11) is highest for androsterone-sulfate (ADT-S) with ;1400 nM, followed by ADT-G (80 nM) )5a-androstane-3b,17b-diol-G (3b-diol) (;47 nM) )5a-androstane3a,17b-diol-G (3a-diol-G) )ADT (5 nM) )3b-diol (3 nM) )DHT (2.8 nM) )3a-diol (2.2 nM). The presence of large amounts of ADT derivatives (ADT-S and ADT-G) that contain a 17-keto group in the circulation compared with compounds having a 17b-hydroxy group (DHT, 3b-diol-G, 3a-diol-G, 3b-diol, and 3a-diol) provides support for the pathway suggested by Luu-The et al. (10) in which ADT metabolites are produced from 5a-dione obtained from 5areduction of 4-dione. It is worth noting that the amount of steroids measured in the blood circulation represent most probably the amount remained after transformation'' and not the amount available for transformation’’. Thus, a steroid amount available for transformation is equal to the amount found in the blood plus the sum of all the metabolites produced downward in the pathway having this steroid as precursor. Controversy between the general belief and the exact steroidogenic pathway could be due to misinterpretation of this concept. In the present report, to verify the importance of the pathway of DHT biosynthesis that does not require the T production step, we analyzed metabolites produced during incubation of w14Cx4-dione and a 5a-reductase inhibitor and quantified the expression levels of steroidogenic enzymes present in prostatic carcinoma (DU-145) cells. Our data clearly indicate that DHT biosynthesis from 4-dione in DU145 does not require T biosynthesis. Enzymatic activity was determined using intact cells in culture as previously described (12). Briefly, 0.1 mM w14Cx-labeled androstenedione (specific activity 56 mCi/mmol) (American Radiolabeled Chemicals Inc., St Louis, MO, USA) was added to 6-well culture plates containing 2 mL culture medium per well. After 24 h incubation the medium was removed, steroids were extracted twice with 1 mL ethyl-ether and metabolites were analyzed by thin layer chromatography (TLC). Organic phases were pooled and evaporated to dryness. Steroids were solubilized in 100 mL methylene chloride and separated on Silica gel 60 TLC plates (Merck, Darmstadt, Germany) using the chloroform/ethyl acetate (4:1) solvent system. Substrates and metabolites were identified by comparison with reference steroids and quantified by a PhosphoImager Storm 860 system (Molecular Dynamics, Inc., Sunnyvale, CA, USA). In our system, DHEA, 4-dione, T, E1, and estradiol (E2) are very well separated, whereas DHT and ADT are poorly resolved. They were analyzed further by high performance liquid chromatography (HPLC) as previously described (13). Briefly, using intact cells in culture, 0.2 mM w14Cx-labeled 4-dione (American Radiolabeled Chemicals Inc.) was added to 6-well culture plates containing 2 mL culture medium per well. After 24 h incubation, metabolites were extracted and analyzed by HPLC/MS and radiolabeled HPLC profiles were recorded by the online liquid scintillation counter (Flo One Radiomatic 500TR Series; Canberra Packard). Steroids were identified using commercial standard. RNA quantification by real-time PCR Total RNA was extracted from 1=106 cells using an RNeasy mini kit (Qiagen, Mississauga, Canada) according to the manufacturer’s protocol. Total RNA (5 mg) was converted to cDNA by incubation at 428C for 2 h with 200 U SuperScript II reverse transcriptase (Invitrogen Life Technologies, Inc.), using oligo-d(T)24 as primer in a reaction buffer containing 50 mM Tris-HCl pH 8.3, 75 mM KCl, 3 mM MgCl2, 10 mM dithiothreitol and 0.5 mM dNTPs. cDNA was purified with a QIAquick PCR purification kit (Qiagen). Quantification of mRNA levels was performed by a quantitative real-time PCR method on a LightCycler Realtime PCR apparatus (Hoffman-La Roche Inc., Nutley, NJ, USA) using SYBR Green and second derivative detection of crossing point and double correction as previously described (14). In brief, 30 ng total RNA was used to perform fluorescent-based real-time PCR quantification. Reagents obtained from the same supplier were used as described by the manufacturer. The conditions for the PCR reactions were as follows: denaturation at 948C for 15 s, annealing at 558C for 10 s, and elon- Article in press - uncorrected proof Samson et al.: Pathway of DHT biosynthesis that does not require testosterone as intermediate 69 gation at 728C for 35 s. Data were normalized using mRNA expression levels of a housekeeping gene, namely ATP5o (subunit O of ATPase) as internal standard. mRNA expression levels are expressed as numbers of copies/mg total RNA using a standard curve of Cp versus logarithm of the quantity. The standard curve is established using known cDNA amounts of 0, 102, 103, 104, 105, and 106 copies of ATP5o and a LightCycler 3.5 program provided by the manufacturer (Roche Inc., Nutley, NJ, USA). Results mRNA expression levels of steroidogenic enzymes in DU-145 cells To determine the enzymes possibly involved in the metabolism of 4-dione and biosynthesis of DHT in DU-145 cells, we quantified mRNA expression levels of steroidogenic enzymes in these cells using real-time PCR. As shown in Figure 2, P450 dependent enzymes, such as P450scc, P450 C17, and P450 aromatase, as well as 3b-HSDs are absent, whereas types 1 and 3 5a-reductases and types 5, 7, and 10 17b-HSDs are present. Metabolism of w Cx4-dione in DU-145 cells Figure 3 Quantification of w14Cxmetabolites of 4-dione in DU-145 cells in culture. Curves showing the metabolic profile of 0.2 mM w14Cx4-dione incubated for 12 and 24 h with DU-145 cells cultured in 12-well plates. Experimental procedures are as described in the Materials and methods section. DU-145 cells were incubated with 0.2 mM w14Cx4-dione and the radiolabeled steroids were analyzed at 12 and 24 h incubation using TLC and HPLC. Use of TLC permits quantification using a PhosphoImager system, whereas use of HPLC presents better separation and identification of steroids. As illustrated in Figure 3, at 12 h incubation, 4-dione concentration decreased with a concomitant increase in 5adione, and to a lesser extent ADT and DHT (these two compounds are not distinguishable in our TLC system) levels increased. Analysis at 24 h incubation indicates that 5adione decreases with a concomitant increase in DHT and ADT. HPLC analysis confirms formation of both DHT and ADT with a formation of a higher amount of ADT with no detectable quantity of T (Figure 4). These data strongly suggest that ADT and DHT are produced from 5a-dione. Figure 2 mRNA expression levels of steroidogenic enzymes in DU-145 cells. Graph showing mRNA expression levels of steroidogenic enzymes in DU-145 cells quantified by a real-time PCR method described in the Materials and methods section. a and b following the numbers 3, 5, and 17 represent Greek symbols a and b, respectively. Data are expressed as means"SD of triplicate measurements. Article in press - uncorrected proof 70 Samson et al.: Pathway of DHT biosynthesis that does not require testosterone as intermediate Figure 5 Effect of 5a-reductase inhibitor on 4-androstenedione metabolism. Experimental procedures are as described in Figure 2, except that 1 mM of the 5a-reductase inhibitor Finasteride was added at the same time as the substrate. Data are expressed as means"SD of triplicate measurements. Figure 4 Identification of w14Cxmetabolites of 4-dione in DU-145 cells in culture by high performance liquid chromatography (HPLC). (A) HPLC/MS metabolic profile of 0.2 mM w14Cx4-dione incubated for 24 h with DU-145 cells cultured in 12-well plates. (B) Graphs showing principal metabolites obtained in (A). Data are expressed as means"SEM of triplicate measurements. Experimental procedures are as described in the Materials and methods section. Metabolism of w14Cx4-dione in the presence of Finasteride, a 5a-reductase inhibitor As illustrated in Figure 5, in the presence of the 5a-reductase inhibitor Finasteride, the production of 5a-dione is blocked after 12 and 24 h incubations with a minimal production of T at 24 h. These data further confirm that, in DU-145 cells, the enzyme 5a-reductase is involved in the transformation of 4-dione into 5a-dione. Discussion The present data show that in the prostatic carcinoma cells DU-145, the pathway of the transformation of 4-dione first involves the enzyme 5a-reductases which transform 4-dione into 5a-dione, whereas subsequently, 5a-dione is transformed into DHT and ADT by 17b-HSD and 3a-HSD, respectively (Figure 1). This pathway provides particularly strong support for previous observations showing that 4dione is a better substrate for 5a-reductases than T (4, 9), and 5a-reductases have higher affinity for the substrates (Km-1 mM) (4, 6) than 17b-HSDs (Km)1 mM) (1, 12, 15). These observations, however, are in contradiction with the pathway usually described in most publications which indicate that DHT is essentially produced from 5a-reduction of T. Our data are confirmed further by experiments using transformed sebocytes, namely SZ-95, cell line (Samson et al. unpublished data), and by incubation in the presence of Finasteride, a 5a-reductase inhibitor. In fact, after 24 h incubation, there is a very low decrease of 5a-dione and a minimum increase of T formation. The low level of T formation is in agreement with the low efficiency of the 17b-HSD enzymes compared with 5a-reductases that possess much higher efficacy for the transformation of substrates (6). mRNA quantification of steroidogenic enzymes (Figure 2) using improved second derivative and the double correction real-time PCR method (16) indicates that 5a-reductase type 3, a newly identified 5a-reductase (17), which is expressed at 10-fold higher than type 1 5a-reductase, is most probably the enzyme that catalyzes the conversion of 4-dione into 5adione. The type 2 5a-reductase, by contrast, is not significantly expressed. To estimate the number of copies/cell, it is worth noting that a single liver cell contains approximately 50 pg total RNA (18); thus, an expression level of 20,000 copies/mg total RNA could correspond to approximately 1 copy/cell in a homogeneous liver cell population. Expression levels lower than 20,000 copies/mg total RNA could indicate a non-significant expression level, a higher level of total RNA in a single cell or heterogeneous cell types in a tissue. These data also suggest that type 5 17b-HSD (12) is probably the enzyme catalyzing the conversion of 5a- Article in press - uncorrected proof Samson et al.: Pathway of DHT biosynthesis that does not require testosterone as intermediate 71 dione into DHT. Because the expression levels of types 1 and 3 3a-HSD are very low, the conversion of 5a-dione into ADT is most probably catalyzed by an unidentified 3a-HSD. The high level of ADT could also be due to oxidative 17bHSD activity exerted by 17b-HSD10 and 11, the enzymes that facilitate NADq co-factor (19, 20) to transform 5aandrostane-3a,17b-diol into ADT. For many decades, it has generally been believed that sex steroid hormones are produced exclusively from the gonads (21). However, recent data obtained from studies in aromatase knockout mice (2224) and in men with aromatase deficiency (2527) reveal the importance of local and tissuespecific biosynthesis of E2. The role of local E2 synthesis is increasingly recognized and prevails over the general belief that E2 is exclusively synthesized in the gonads and delivered to peripheral tissues through the circulation. Similarly, studies in patients having a defect in androgen biosynthesis have identified that T and DHT are two androgens with different functions. T measured in the blood is produced mainly in the testis and acts directly to promote the formation of the internal male reproductive structures (epididymes, seminal vesicles, and vas deferens). A defect in 17b-HSD3 causes a disorder of human intersex termed male pseudohermaphroditism (1). The typical features of 17b-HSD deficiency show a 46,XY individual having ambiguous female external genitalia and marked virilization at puberty (21). At surgery, testes and epididymis are found in the inguinal canals, whereas lower Wolffian duct structures are male in character including seminal vesicles and ejaculatory ducts. By contrast, a defect in 5a-reductase type 2 (SRD5a2) also causes male pseudohermaphroditism (2) with ambiguous external genitalia. However, as mentioned above, in contrast with type 3 17b-HSD deficiency, the Wolffian structures are normally differentiated (3). DHT and 5a-reductases thus play an important role in local androgen biosynthesis. The importance of local biosynthesis of sex steroids is in agreement with the intracrinology concept suggested by Labrie et al. (2830). Some discrepancy exists concerning identification of the substrate precursor for local biosynthesis of sex steroids. The general belief based on traditional literature suggests that the substrate of 5a-reductase and aromatase is T, although there is increasing evidence that suggests that the substrate of 5areductase and aromatase is 4-dione (10), especially in the human which produces high amounts of steroid adrenal precursors in the circulation (28, 31). The discrepancy is somewhat difficult to overcome because the mouse and the rat do not express 17a-hydroxylase/17-20lyase (CYP17) in the adrenals (32, 33), thus removing high levels of adrenal precursor in the circulation as found in the human (10, 34, 35). However, as mentioned above, the discrepancy is most probably due to the misinterpretation of the concentration of steroid measured in the blood circulation. The understanding is completely different if we consider that this concentration is the amount to be transformed or the concentration remaining after transformation. Traditional literature generally consid- ered that the blood concentration of steroid is the concentration available to be used by the cells or tissues. However, it is more logical to consider that blood concentration of steroid is the amount remaining after transformation, and the amount available to be transformed is the sum of the blood concentration plus the concentrations of all metabolites found downward in the pathway. In this regard, T is almost not transformed. It reflects the physiological action of androgenic activity distributed through the endocrine system under the control of the hypothalamo-gonadal axis. One viewpoint is that blood concentration of 4-dione represents only a part of the available amounts. Indeed, it has been transformed into many 5a-reduced metabolites, including DHT. The androgenic activity DHT is thus controlled by the presence of enzymes involved in its biosynthesis and degradation, such as 5a-reductases, 17b-HSDs, 3a-HSDs, UDP-glucuronosyltransferases and sulfotransferases, and the concentration of precursor in the tissue. Knowledge of the exact biosynthetic pathway of DHT in peripheral tissue is important for a more specific design of inhibitor of androgen biosynthesis for the treatment of androgen-sensitive diseases, such as prostate cancer, BPH, alopecia, and acne. Data obtained in the present study in the DU-145 prostate cancer cell line strongly support the pathway of DHT biosynthesis that does not require T as intermediate
