- Email: [email protected]
A rapid method for obtaining finasteride, a 5α-reductase inhibitor, from commercial tablets
Brain Research Protocols 9 (2002) 130–134 www.elsevier.com / locate / bres
Protocol
A rapid method for obtaining finasteride, a 5a-reductase inhibitor, from commercial tablets a, b b a Giuseppe Trapani *, Laura Dazzi , Maria Giuseppina Pisu , Antonia Reho , b b Emanuele Seu , Giovanni Biggio b
a Pharmaco-Chemistry Department, University of Bari, Via Orabona 4, 70125 Bari, Italy Department of Experimental Biology ‘ B. Loddo’ and CNR, Centre for Neuropharmacology, University of Cagliari, Cagliari, Italy
Accepted 8 April 2002
Abstract To study the effects of allopregnanolone (AP) depletion on stress-induced dopamine changes in cortical dopamine, the 5a-reductase inhibitor finasteride on a gram-scale is required. Two procedures for the extraction of finasteride from tablets are outlined (method A and B). In method A, a suspension of powdered tablets was preliminary extracted with chloroform and the extracts dried and evaporated. The resulting residue was then purified on column chromatography. Method B involves a direct chromatographic separation of the powdered tablets. In terms of isolated yields, the second procedure works well, is cheaper, and less time-consuming. The efficiency of the method was tested by measuring progesterone, AP and THDOC content in plasma and cerebral cortex of rats. The protocol enables the prompt availability of sufficient amount of finasteride in experimental grade, useful in examining the role of endogenous cerebrocortical AP in brain homeostasis. 2002 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behaviour Topic: Stress Keywords: Finasteride; Allopregnanolone; Brain; Tablets
1. Type of research 1. Procedures for extraction of finasteride from commercial tablets. 2. Investigation on the effects of depletion of allopregnanolone in plasma and cerebral cortex of rats.
2. Time required 1. Pulverization of tablets with mortar and pestle (method A and B): 10 min
*Corresponding author. Tel.: 139-080-544-2764; fax: 139-080-5442754. E-mail address: [email protected] (G. Trapani).
2. Extraction of the powder with chloroform (method A): 45 min 3. Drying of chloroform extracts and evaporation of the solvent (method A): 60–90 min 4. Column chromatography on silica gel: 480 min 5. Total protocol: 600–700 min
3. Materials
3.1. Chemicals Tablets of Prostide , each containing 5 mg of drug, were obtained from Sigma-tau, Italy. The excipients are: lactose monohydrate, pregelatinized starch, sodium carboxymethyl starch, yellow iron oxide, dioctyl sodium sulfosuccinate, microcrystalline cellulose, magnesium stearate, hypromellose, hydroxypropylcellulose, titanium diox-
1385-299X / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S1385-299X( 02 )00146-0
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
ide, talc, indigotine on aluminium hydrate. Thin layer chromatography (TLC) was performed using Merck precoated silica gel aluminium sheets. Silica gel for column chromatography was Merck silica gel 60 (63–200 mm). Chloroform and ethyl acetate was obtained from Baker.
3.2. Apparatus Melting points were determined in open capillary tubes ¨ with a Buchi apparatus and are uncorrected. Elemental analyses were carried out with a Eurovector EA 3011 instrument. 1 H NMR spectra were determined on a Varian VX Mercury operating at 300 MHz. Mass spectra were recorded on a Hewlett-Packard 6890-5973MSD GC–MS low resolution spectrometer. The column used for GC was a HP5MS column (30 m30.25 mm30.25 mm, 5% phenylmethyl silicone); initial temperature: 70 8C, final temperature: 250 8C, isothermal 3 min, rate 20 8C / min.
4. Detailed procedure (i) Method A: 180 tablets were powdered in a mortar and the powder suspended in about 200 ml of deionized water. The suspension was transferred to a flask shaker and extracted with chloroform (200 ml). The water layer was separated in two portions (each about
131
100 ml), and extracted further with 100 ml of chloroform. The combined chloroform extracts were dried over Na 2 SO 4 , filtered and evaporated. The residue was purified by silica gel column chromatography with ethyl acetate–methanol (95:5, v / v) to give 350 mg of pure (TLC) finasteride (yield 39%). (ii) Method B: 135 tablets were powdered in a mortar and the powder was purified by silica gel column chromatography with ethyl acetate as eluent to give 520 mg of pure (TLC) finasteride (yield 77%). (iii) Finasteride purity was checked (a) by the presence of a single spot on TLC (chloroform as eluent); (b) by a sharp melting point at 245 8C hlit. 252–254 [8]j; (c) by a single peak in gas chromatography (retention time 23.2 min) (Fig. 1); (d) by elemental analysis (C 23 H 36 N 2 O 2 ): Requires C, 74.15; H, 9.74; N, 7.52. Found C, 73.92; H, 9.83; N, 7.31. (iv) Identity of finasteride was checked (a) by proton magnetic resonance: 1 H NMR (CDCl 3 ) d 0.7–2.2 (m, 16 H, aliphatic), 0.70 (s, 3H, CH 3 ), 0.97 (s, 3H, CH 3 ), 1.35 (s, 9H, C(CH 3 ) 3 ), 3.3–3.5 (m, 1H, NHCH ), 5.11 (s, 1H, CONHC(CH 3 ) 3 ), 5.80 (d, 1H, CH=), 6.62 (s, 1H, CHNHCO), 6.79 (d, 1H, CH=); (b) by mass spectrometry: MS m /z 372 (M 1 , base), 357 (M 1 – CH 3 , 54) (Fig. 2). (v) The efficiency of the method was tested by measuring progesterone, allopregnanolone, and THDOC content
Fig. 1. Gas chromatogram of finasteride indicating the purity of the sample.
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
132
Fig. 2. Fragmentation pattern of finasteride.
in plasma and cerebral cortex of rats. Steroids were extracted and purified by HPLC, and quantified by radioimmunoassay as previously described [3].
5. Results To study the effects of allopregnanolone (AP) depletion on stress-induced dopamine changes in cortical dopamine [3], finasteride on a gram-scale was required. To circumvent this problem, it was decided to extract the active principle from a batch of tablets commercially available. Examination of the tablet composition revealed that most of the excipients utilized are water-soluble or water-dispersible. Therefore, it seemed appropriate to use a preliminary partition between chloroform and water of the powdered tablets and successively a purification on column chromatography (method A). In principle, in such a way the poorly water-soluble finasteride might be separated from excipients. However, following this procedure we obtained samples which resulted in mixtures of the required drug and the colorant indigotine. The presence of this last compound is easily detected by a blue spot on TLC. Thus, following method A, a further purification on column chromatography was necessary resulting, on the whole, in a lower yield of finasteride. Thus, we suspected that the considerable amount of indigotine was a consequence of the chloroform–water partition preliminary at the chromatographic purification. Therefore, according to
method B, we performed the direct separation on column chromatography of the powdered tablets, without the preliminary chloroform–water partition. In such a way, with pleasure it was observed that, the overall yield of finasteride considerably increased. To demonstrate that the procedures described here to obtain finasteride in experimental grade were efficient, we measured the brain content of progesterone and of its GABAA receptor active metabolites AP and THDOC. Rats were treated with finasteride at a dose (25 mg / kg, s.c.) that has already been shown to reduce AP levels in rat brain [2,9] and steroids were measured as previously described [3]. As expected [3,2,9], administration of finasteride (25 mg / kg, s.c., once daily for 2 days) resulted in a marked decrease (273%) in the cerebrocortical concentration of AP and THDOC, measured 5 h after the last injection (Fig. 3). In contrast, finasteride did not significantly affect the basal cortical concentration of progesterone. Similarly, finasteride induced a marked decrease (about 270%) in the plasmatic concentrations of AP and THDOC (Fig. 4).
6. Discussion Allopregnanolone (AP) and allotetrahydrodeoxycorticosterone (THDOC) are endogenous neuroactive steroids that enhance the action of GABA at GABAA receptors with high potency in vitro [4,7] and in vivo [5,1]. AP and THDOC are biosynthetically formed from progesterone
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
133
Fig. 3. Effects of finasteride pretreatment on cerebrocortical concentrations of progesterone, AP and THDOC. Rats were pretreated with finasteride (25 mg / kg, s.c.; hatched columns) or vehicle (open columns) once a day for 2 days. Cerebrocortical concentrations of neurosteroids were measured 5 h after the last injection. Data are expressed as nanograms of steroid per gram of protein and are means6S.E.M. of values from at least six rats per group. a P,0.01 vs. vehicle-treated animals.
with the intervention of the enzyme 5a-reductase. Finasteride, N-(1,1-dimethylethyl)-3-oxo-(5a, 17b)-4-azaandrost-1-ene-17-carboxamide, is the prototype of a new class of antiprostatic agents being a potent inhibitor of the enzyme 5a-reductase [6]. Therefore, it constitutes an essential pharmacological tool for studying the effects of inhibition of AP biosynthesis in the brain. Our present results show that finasteride purified with the method herein described has an efficacy in decreasing allopregnanolone and THDOC content which is similar to that obtained by administration of the commercially available drug [2,9]. The prompt availability of finasteride in experimental grade and on a gram-scale is not a simple task. The
chemical synthesis of finasteride, indeed, is very difficult since it involves several delicate, time-consuming and expensive steps and overall proceeds in very low yield. The procedures outlined herein have proved to be rapid, simple and efficient methods and their main advantage is that they allow us to avoid the chemical synthesis. Furthermore, in terms of isolated yields, the second procedure (method B) works well, is cheaper and less time-consuming.
6.1. Troubleshooting Although the procedures herein described are straightforward and simple, the single relevant problem encoun-
Fig. 4. Effects of finasteride pretreatment on plasmatic concentrations of progesterone, AP and THDOC. Rats were pretreated with finasteride (25 mg / kg, s.c.; hatched columns) or vehicle (open columns) once a day for 2 days. Plasmatic concentrations of neurosteroids were measured 5 h after the last injection. Data are expressed as nanograms of steroid per ml of plasma and are means6S.E.M. of values from at least six rats per group. a P,0.01 vs. vehicle-treated animals.
134
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
tered was the possibility to obtain mixtures of finasteride and indigotine. The solution to this drawback is to strictly follow method B and collect small fractions during the chromatography on column.
7. Quick procedure (i) Pulverization of tablets with mortar and pestle (ii) Fill a column chromatography with a slurry of silica gel using ethyl acetate (iii) Pour the powdered tablets onto the silica gel (iv) Elute with ethyl acetate and collect the fractions which show a single spot on TLC (chloroform as eluent) (v) Evaporation of the solvent yields the final product
8. Essential literature references Ref. [3]: L. Dazzi, M. Serra, G. Vacca, S. Ladu, A. Latrofa, G. Trapani, G. Biggio, Depletion of cortical allopregnenolone potentiates stress-induced increase in cortical dopamine output. Brain Res. 2002 (in press).
[2]
[3]
[4]
[5] [6]
[7]
[8]
[9]
Acknowledgements The authors wish to thank Sigma-tau (Pomezia, Italy) for the gift of Prostide .
References [1] A. Concas, M.C. Mostallino, C. Perra, R. Lener, G. Roscetti, M.L. Barbaccia, R.H. Purdy, G. Biggio, Functional correlation between
allopregnanolone and [ 35 S]TBPS binding in the brain of rats exposed to isoniazid, pentylenetetrazol, or stress, Br. J. Pharmacol. 118 (1996) 839–846. A. Concas, M.C. Mostallino, P. Porcu, P. Follesa, M.L. Barbaccia, M. Trabucchi, R.H. Purdy, P. Grisenti, G. Biggio, Role of brain allopregnanolone in the plasticity of gamma-aminobutyric acid type A receptor in rat brain during pregnancy and after delivery, Proc. Natl. Acad. Sci. USA 95 (1998) 13284–13289. L. Dazzi, M. Serra, G. Vacca, S. Ladu, A. Latrofam, G. Trapani, G. Biggio, Depletion of cortical allopregnanolone potentiates stressinduced increase in cortical dopamine output, Brain Res. 2002 (in press). M.D. Majewska, N.L. Harrison, R.D. Schwartz, J.L. Barker, S.M. Paul, Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor, Science 232 (1986) 1004–1007. S.M. Paul, R.H. Purdy, Neuroactive steroids, FASEB J. 6 (1992) 2311–2322. G. Pinna, V. Uzunova, K. Matsumoto, G. Puia, J.M. Mienville, E. Costa, A. Guidotti, Brain allopregnanolone regulates the potency of the GABAA receptors, Neuron 4 (2000) 759–765. G. Puia, M.R. Santi, S. Vicini, D.B. Pritchett, R.H. Purdy, S.M. Paul, P.H. Seeburg, E. Costa, Neurosteroids act on recombinant human GABAA receptors, Neuron 4 (1990) 759–765. G.H. Rasmusson, G.F. Reynolds, N.G. Steinberg, E. Walton, G.F. Patel, T. Liang, M.A. Cascieri, A.H. Cheung, J.R. Brooks, C. Berman, Azasteroids; structure–activity relationships for inhibition of 5a-reductase androgen receptor binding, J. Med. Chem. 29 (1986) 2298–2315. G.H. Rasmusson, G.F. Reynolds, N.G. Steinberg, E. Walton, G.F. Patel, T. Liang, M.A. Cascieri, A.H. Cheung, J.R. Brooks, C. Berman, Azasteroids: structure–activity relationships for inhibition of 5a-reductase androgen receptor binding, J. Med. Chem. 29 (1986) 2298–2315.
Protocol
A rapid method for obtaining finasteride, a 5a-reductase inhibitor, from commercial tablets a, b b a Giuseppe Trapani *, Laura Dazzi , Maria Giuseppina Pisu , Antonia Reho , b b Emanuele Seu , Giovanni Biggio b
a Pharmaco-Chemistry Department, University of Bari, Via Orabona 4, 70125 Bari, Italy Department of Experimental Biology ‘ B. Loddo’ and CNR, Centre for Neuropharmacology, University of Cagliari, Cagliari, Italy
Accepted 8 April 2002
Abstract To study the effects of allopregnanolone (AP) depletion on stress-induced dopamine changes in cortical dopamine, the 5a-reductase inhibitor finasteride on a gram-scale is required. Two procedures for the extraction of finasteride from tablets are outlined (method A and B). In method A, a suspension of powdered tablets was preliminary extracted with chloroform and the extracts dried and evaporated. The resulting residue was then purified on column chromatography. Method B involves a direct chromatographic separation of the powdered tablets. In terms of isolated yields, the second procedure works well, is cheaper, and less time-consuming. The efficiency of the method was tested by measuring progesterone, AP and THDOC content in plasma and cerebral cortex of rats. The protocol enables the prompt availability of sufficient amount of finasteride in experimental grade, useful in examining the role of endogenous cerebrocortical AP in brain homeostasis. 2002 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behaviour Topic: Stress Keywords: Finasteride; Allopregnanolone; Brain; Tablets
1. Type of research 1. Procedures for extraction of finasteride from commercial tablets. 2. Investigation on the effects of depletion of allopregnanolone in plasma and cerebral cortex of rats.
2. Time required 1. Pulverization of tablets with mortar and pestle (method A and B): 10 min
*Corresponding author. Tel.: 139-080-544-2764; fax: 139-080-5442754. E-mail address: [email protected] (G. Trapani).
2. Extraction of the powder with chloroform (method A): 45 min 3. Drying of chloroform extracts and evaporation of the solvent (method A): 60–90 min 4. Column chromatography on silica gel: 480 min 5. Total protocol: 600–700 min
3. Materials
3.1. Chemicals Tablets of Prostide , each containing 5 mg of drug, were obtained from Sigma-tau, Italy. The excipients are: lactose monohydrate, pregelatinized starch, sodium carboxymethyl starch, yellow iron oxide, dioctyl sodium sulfosuccinate, microcrystalline cellulose, magnesium stearate, hypromellose, hydroxypropylcellulose, titanium diox-
1385-299X / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S1385-299X( 02 )00146-0
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
ide, talc, indigotine on aluminium hydrate. Thin layer chromatography (TLC) was performed using Merck precoated silica gel aluminium sheets. Silica gel for column chromatography was Merck silica gel 60 (63–200 mm). Chloroform and ethyl acetate was obtained from Baker.
3.2. Apparatus Melting points were determined in open capillary tubes ¨ with a Buchi apparatus and are uncorrected. Elemental analyses were carried out with a Eurovector EA 3011 instrument. 1 H NMR spectra were determined on a Varian VX Mercury operating at 300 MHz. Mass spectra were recorded on a Hewlett-Packard 6890-5973MSD GC–MS low resolution spectrometer. The column used for GC was a HP5MS column (30 m30.25 mm30.25 mm, 5% phenylmethyl silicone); initial temperature: 70 8C, final temperature: 250 8C, isothermal 3 min, rate 20 8C / min.
4. Detailed procedure (i) Method A: 180 tablets were powdered in a mortar and the powder suspended in about 200 ml of deionized water. The suspension was transferred to a flask shaker and extracted with chloroform (200 ml). The water layer was separated in two portions (each about
131
100 ml), and extracted further with 100 ml of chloroform. The combined chloroform extracts were dried over Na 2 SO 4 , filtered and evaporated. The residue was purified by silica gel column chromatography with ethyl acetate–methanol (95:5, v / v) to give 350 mg of pure (TLC) finasteride (yield 39%). (ii) Method B: 135 tablets were powdered in a mortar and the powder was purified by silica gel column chromatography with ethyl acetate as eluent to give 520 mg of pure (TLC) finasteride (yield 77%). (iii) Finasteride purity was checked (a) by the presence of a single spot on TLC (chloroform as eluent); (b) by a sharp melting point at 245 8C hlit. 252–254 [8]j; (c) by a single peak in gas chromatography (retention time 23.2 min) (Fig. 1); (d) by elemental analysis (C 23 H 36 N 2 O 2 ): Requires C, 74.15; H, 9.74; N, 7.52. Found C, 73.92; H, 9.83; N, 7.31. (iv) Identity of finasteride was checked (a) by proton magnetic resonance: 1 H NMR (CDCl 3 ) d 0.7–2.2 (m, 16 H, aliphatic), 0.70 (s, 3H, CH 3 ), 0.97 (s, 3H, CH 3 ), 1.35 (s, 9H, C(CH 3 ) 3 ), 3.3–3.5 (m, 1H, NHCH ), 5.11 (s, 1H, CONHC(CH 3 ) 3 ), 5.80 (d, 1H, CH=), 6.62 (s, 1H, CHNHCO), 6.79 (d, 1H, CH=); (b) by mass spectrometry: MS m /z 372 (M 1 , base), 357 (M 1 – CH 3 , 54) (Fig. 2). (v) The efficiency of the method was tested by measuring progesterone, allopregnanolone, and THDOC content
Fig. 1. Gas chromatogram of finasteride indicating the purity of the sample.
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
132
Fig. 2. Fragmentation pattern of finasteride.
in plasma and cerebral cortex of rats. Steroids were extracted and purified by HPLC, and quantified by radioimmunoassay as previously described [3].
5. Results To study the effects of allopregnanolone (AP) depletion on stress-induced dopamine changes in cortical dopamine [3], finasteride on a gram-scale was required. To circumvent this problem, it was decided to extract the active principle from a batch of tablets commercially available. Examination of the tablet composition revealed that most of the excipients utilized are water-soluble or water-dispersible. Therefore, it seemed appropriate to use a preliminary partition between chloroform and water of the powdered tablets and successively a purification on column chromatography (method A). In principle, in such a way the poorly water-soluble finasteride might be separated from excipients. However, following this procedure we obtained samples which resulted in mixtures of the required drug and the colorant indigotine. The presence of this last compound is easily detected by a blue spot on TLC. Thus, following method A, a further purification on column chromatography was necessary resulting, on the whole, in a lower yield of finasteride. Thus, we suspected that the considerable amount of indigotine was a consequence of the chloroform–water partition preliminary at the chromatographic purification. Therefore, according to
method B, we performed the direct separation on column chromatography of the powdered tablets, without the preliminary chloroform–water partition. In such a way, with pleasure it was observed that, the overall yield of finasteride considerably increased. To demonstrate that the procedures described here to obtain finasteride in experimental grade were efficient, we measured the brain content of progesterone and of its GABAA receptor active metabolites AP and THDOC. Rats were treated with finasteride at a dose (25 mg / kg, s.c.) that has already been shown to reduce AP levels in rat brain [2,9] and steroids were measured as previously described [3]. As expected [3,2,9], administration of finasteride (25 mg / kg, s.c., once daily for 2 days) resulted in a marked decrease (273%) in the cerebrocortical concentration of AP and THDOC, measured 5 h after the last injection (Fig. 3). In contrast, finasteride did not significantly affect the basal cortical concentration of progesterone. Similarly, finasteride induced a marked decrease (about 270%) in the plasmatic concentrations of AP and THDOC (Fig. 4).
6. Discussion Allopregnanolone (AP) and allotetrahydrodeoxycorticosterone (THDOC) are endogenous neuroactive steroids that enhance the action of GABA at GABAA receptors with high potency in vitro [4,7] and in vivo [5,1]. AP and THDOC are biosynthetically formed from progesterone
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
133
Fig. 3. Effects of finasteride pretreatment on cerebrocortical concentrations of progesterone, AP and THDOC. Rats were pretreated with finasteride (25 mg / kg, s.c.; hatched columns) or vehicle (open columns) once a day for 2 days. Cerebrocortical concentrations of neurosteroids were measured 5 h after the last injection. Data are expressed as nanograms of steroid per gram of protein and are means6S.E.M. of values from at least six rats per group. a P,0.01 vs. vehicle-treated animals.
with the intervention of the enzyme 5a-reductase. Finasteride, N-(1,1-dimethylethyl)-3-oxo-(5a, 17b)-4-azaandrost-1-ene-17-carboxamide, is the prototype of a new class of antiprostatic agents being a potent inhibitor of the enzyme 5a-reductase [6]. Therefore, it constitutes an essential pharmacological tool for studying the effects of inhibition of AP biosynthesis in the brain. Our present results show that finasteride purified with the method herein described has an efficacy in decreasing allopregnanolone and THDOC content which is similar to that obtained by administration of the commercially available drug [2,9]. The prompt availability of finasteride in experimental grade and on a gram-scale is not a simple task. The
chemical synthesis of finasteride, indeed, is very difficult since it involves several delicate, time-consuming and expensive steps and overall proceeds in very low yield. The procedures outlined herein have proved to be rapid, simple and efficient methods and their main advantage is that they allow us to avoid the chemical synthesis. Furthermore, in terms of isolated yields, the second procedure (method B) works well, is cheaper and less time-consuming.
6.1. Troubleshooting Although the procedures herein described are straightforward and simple, the single relevant problem encoun-
Fig. 4. Effects of finasteride pretreatment on plasmatic concentrations of progesterone, AP and THDOC. Rats were pretreated with finasteride (25 mg / kg, s.c.; hatched columns) or vehicle (open columns) once a day for 2 days. Plasmatic concentrations of neurosteroids were measured 5 h after the last injection. Data are expressed as nanograms of steroid per ml of plasma and are means6S.E.M. of values from at least six rats per group. a P,0.01 vs. vehicle-treated animals.
134
G. Trapani et al. / Brain Research Protocols 9 (2002) 130 – 134
tered was the possibility to obtain mixtures of finasteride and indigotine. The solution to this drawback is to strictly follow method B and collect small fractions during the chromatography on column.
7. Quick procedure (i) Pulverization of tablets with mortar and pestle (ii) Fill a column chromatography with a slurry of silica gel using ethyl acetate (iii) Pour the powdered tablets onto the silica gel (iv) Elute with ethyl acetate and collect the fractions which show a single spot on TLC (chloroform as eluent) (v) Evaporation of the solvent yields the final product
8. Essential literature references Ref. [3]: L. Dazzi, M. Serra, G. Vacca, S. Ladu, A. Latrofa, G. Trapani, G. Biggio, Depletion of cortical allopregnenolone potentiates stress-induced increase in cortical dopamine output. Brain Res. 2002 (in press).
[2]
[3]
[4]
[5] [6]
[7]
[8]
[9]
Acknowledgements The authors wish to thank Sigma-tau (Pomezia, Italy) for the gift of Prostide .
References [1] A. Concas, M.C. Mostallino, C. Perra, R. Lener, G. Roscetti, M.L. Barbaccia, R.H. Purdy, G. Biggio, Functional correlation between
allopregnanolone and [ 35 S]TBPS binding in the brain of rats exposed to isoniazid, pentylenetetrazol, or stress, Br. J. Pharmacol. 118 (1996) 839–846. A. Concas, M.C. Mostallino, P. Porcu, P. Follesa, M.L. Barbaccia, M. Trabucchi, R.H. Purdy, P. Grisenti, G. Biggio, Role of brain allopregnanolone in the plasticity of gamma-aminobutyric acid type A receptor in rat brain during pregnancy and after delivery, Proc. Natl. Acad. Sci. USA 95 (1998) 13284–13289. L. Dazzi, M. Serra, G. Vacca, S. Ladu, A. Latrofam, G. Trapani, G. Biggio, Depletion of cortical allopregnanolone potentiates stressinduced increase in cortical dopamine output, Brain Res. 2002 (in press). M.D. Majewska, N.L. Harrison, R.D. Schwartz, J.L. Barker, S.M. Paul, Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor, Science 232 (1986) 1004–1007. S.M. Paul, R.H. Purdy, Neuroactive steroids, FASEB J. 6 (1992) 2311–2322. G. Pinna, V. Uzunova, K. Matsumoto, G. Puia, J.M. Mienville, E. Costa, A. Guidotti, Brain allopregnanolone regulates the potency of the GABAA receptors, Neuron 4 (2000) 759–765. G. Puia, M.R. Santi, S. Vicini, D.B. Pritchett, R.H. Purdy, S.M. Paul, P.H. Seeburg, E. Costa, Neurosteroids act on recombinant human GABAA receptors, Neuron 4 (1990) 759–765. G.H. Rasmusson, G.F. Reynolds, N.G. Steinberg, E. Walton, G.F. Patel, T. Liang, M.A. Cascieri, A.H. Cheung, J.R. Brooks, C. Berman, Azasteroids; structure–activity relationships for inhibition of 5a-reductase androgen receptor binding, J. Med. Chem. 29 (1986) 2298–2315. G.H. Rasmusson, G.F. Reynolds, N.G. Steinberg, E. Walton, G.F. Patel, T. Liang, M.A. Cascieri, A.H. Cheung, J.R. Brooks, C. Berman, Azasteroids: structure–activity relationships for inhibition of 5a-reductase androgen receptor binding, J. Med. Chem. 29 (1986) 2298–2315.