Design, synthesis and biological evaluation of novel androst-3,5-diene-3-carboxylic acid derivatives as inhibitors of 5α-reductase type 1 and 2

Article abstract of DOI:10.1016/j.steroids.2017.05.011

5α-Reductase is a key enzyme responsible for dihydrotestosterone biosynthesis and has been recognized as an important target for discovering new drugs against benign prostatic hyperplasia (BPH). In this study, a series of novel steroidal androst-3,5-diene-3-carboxylic acids have been designed and synthesized. Biological evaluations were performed on their 5α-reductase inhibitory activities by both in vitro enzyme inhibition assay and in vivo by prostate weighing method. Results showed that most of them displayed excellent 5α-reductase inhibitory potency. Detailed evaluation indicated that most of the compounds displayed slightly higher inhibition potency towards type 2 isozyme. Among all the compounds, 16a was found to be the most potential inhibitor with the IC₅₀ of 0.25?μM and 0.13?μM against type 1 and 2 isozymes respectively. In vivo 5a-reductase inhibitory evaluation of 16a also showed a more significant reduction effect (p?

Full text of DOI:10.1016/j.steroids.2017.05.011

Accepted Manuscript
Design, Synthesis and Biological Evaluation of Novel Androst-3,5-diene-3-
carboxylic Acid Derivatives as Inhibitors of 5α-Reductase Type 1 and 2
Kejing Lao, Jie Sun, Chong Wang, Weiting Lyu, Boshen Zhou, Ruheng Zhao,
Qian Xu, Qidong You, Hua Xiang
PII:
S0039-128X(17)30089-2
http://dx.doi.org/10.1016/j.steroids.2017.05.011
STE 8108
DOI:
Reference:
Steroids
To appear in:
Received Date:
Revised Date:
Accepted Date:
14 April 2017
12 May 2017
18 May 2017
Please cite this article as: Lao, K., Sun, J., Wang, C., Lyu, W., Zhou, B., Zhao, R., Xu, Q., You, Q., Xiang, H.,
Design, Synthesis and Biological Evaluation of Novel Androst-3,5-diene-3-carboxylic Acid Derivatives as
Inhibitors of 5α-Reductase Type 1 and 2, Steroids (2017), doi: http://dx.doi.org/10.1016/j.steroids.2017.05.011
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, Synthesis and Biological Evaluation of Novel
Androst-3,5-diene-3-carboxylic Acid Derivatives as Inhibitors
of 5α-Reductase Type 1 and 2
Kejing Laoa,b, Jie Suna,b, Chong Wanga,b, Weiting Lyua,b, Boshen Zhoua,b,
Ruheng Zhaoa,b, Qian Xua,b, Qidong Youa,b, Hua Xianga,b*
aJiangsu Key Laboratory of Drug Design and Optimization, China Pharmaceutical
University, 24 Tongjiaxiang, Nanjing 210009, PR China
bDepartment of Medicinal Chemistry, School of Pharmacy, China Pharmaceutical
University, 24 Tongjiaxiang, Nanjing 210009, PR China
* Corresponding author. Tel.: +86 025 83271096; Fax: +86 025 83271096(H. Xiang).
E-mail addresses: xianghua@cpu.edu.cn (H. Xiang)
Abstract
5α-Reductase is a key enzyme responsible for dihydrotestosterone biosynthesis and has been
recognized as an important target for discovering new drugs against benign prostatic hyperplasia
(BPH). In this study, a series of novel steroidal androst-3,5-diene-3-carboxylic acids have been
designed and synthesized. Biological evaluations were performed on their 5α-reductase inhibitory
activities by both in vitro enzyme inhibition assay and in vivo by prostate weighing method.
Results showed that most of them displayed excellent 5α-reductase inhibitory potency. Detailed
evaluation indicated that most of the compounds displayed slightly higher inhibition potency
towards type 2 isozyme. Among all the compounds, 16a was found to be the most potential
inhibitor with the IC50 of 0.25μM and 0.13μM against type 1 and 2 isozymes respectively. In vivo
5a-reductase inhibitory evaluation of 16a also showed a more significant reduction effect
(p<0.001) in rat prostate weight than epristeride. Furthermore, the results of in silico ADME study
indicated that compound 16a exhibited good pharmacokinetic properties. Thus, 16a could serve as
promising lead candidates for further study.
Keywords：benign prostatic hyperplasia (BPH); 5α-reductase inhibitors; steroidal 3-carboxylic
acids; epristeride
1. Introduction
Benign prostatic hyperplasia (BPH), the leading disorder in aging males, is characterized by a
progressive enlargement of prostatic tissue, resulting in obstruction of the proximal urethra and
causing urinary flow disturbances [1]. Nearly half of men aged over 50 years have histological
evidence of BPH and, after the age of 70, the proportion increases to 80% [2]. It has long been
established that prostatic growth is stimulated by androgens. Testosterone (T), which is
1

biosynthesized in testicles and adrenal glands, can be further converted to more potent
dihydrotestosterone (DHT) by 5α-reductase in prostate. 5α-Reductase is a membrane-bound,
NADPH-dependent enzyme that irreversibly catalyzes the reduction of 4-ene-3-oxosteroids to the
corresponding 5α-3-oxosteroids (Figure 1). The 5α-reductase family is composed of three
isozymes. 5α-Reductase type 1 (5αR-1) is mainly expressed in skin and liver with the optimum pH
range of 6.0~8.5, while the 5α-reductase type 2 (5αR-2) with the optimum pH 5.5, is mainly
expressed in prostate and other genital tissues [3]. More recently, type 3 isozyme was identified in
castration-resistant prostate cancer cells as well as other tissues such as pancreas, brain, skin and
adipose tissues [4, 5]. 5α-Reductase has emerged as a target for the pharmaceutical treatment of
BPH as abnormally high activity of the enzyme in human results in excessive DHT levels. Thus
the suppression of DHT biosynthesis by 5α-reductase inhibitors is a logical treatment for BPH [6]
.
Fig. 1 The reaction catalyzed by 5α-reductase.
Finasteride (1) and dutasteride (2), the two widely used 5α-reductase inhibitors in clinical practice,
are both 4-azasteriods. However, in spite of their success in improving the quality of life in men
suffering from BPH symptoms, a substantial body of evidence exists which points to serious and
potentially ill-health effects including loss or reduced libido, erectile dysfunction, orgasmic and
ejaculatory dysfunction, development of high grade prostate tumors, potential negative
cardiovascular events, and depression, which was suspicious of the presence of two nitrogen
atoms at C-4 and C-20 positions [7-8]. Steroidal 3-carboxylic acids are a kind of novel 5α-reductase
inhibitors, among which, epristeride (3) was the first one launched in 2000 as a therapy for BPH[9].
Fig. 2 The structures of the finasteride (1), dutasteride (2), and epristeride (3).
Since the 5α-reductase isozymes have not been purified and crystalized due to their unstable
nature, the design of novel 5α-reductase inhibitor mainly relies on structure-based optimization
approach. The ring A of the 3-carboxylic acids was designed to mimic the putative enzyme-bound
enolate intermediate by incorporating sp2-hybridized centers at C-3 and C-4 and, most critically,
an anionic carboxylic acid at C-3 as a charged replacement for the enolate oxyanion. Because of
this presumably favorable electrostatic interaction between the carboxylate and the positively
charged oxidized cofactor, the acrylate preferentially binds in a ternary complex with enzyme and
NADP+, which leads to the noncompetitive kinetic mechanism [9-12]. The 3D-QSAR study of
3-carboxysteroids indicated that presence of bulky electronegative substitutions at 17-C was
2

favorable for maximal activity, and around position C-3 of steroid skeleton, electronegative
groups and small steric substituents were required for optimal inhibitory activity [13]
.
Interestingly, considerable attention has been also focused on the modifications of substitutions at
the C-17 position of steroid skeleton. Among them, aniline moiety has been widely used as
functional groups in compounds targeting androgen axis as exemplified in Figure 3 [14-17]. A series
of androstene 17β-carboxamides were reported by C.Amaral et al. as 5α-reductase inhibitors with
anti-proliferate activities. In our previous study, several androstene-17β-carboxamides has been
disclosed for their strong inhibition of 5α-reductase inhibition and modulations of androgen
receptor (Figure 3) [18]
.
Fig.3 Structures of known compounds bearing aniline substitutions
Inspired by the promising research results mentioned above, we set out to design a number of
3-carboxylic acids with various substituted 17β-N-phenylcarbamoyl groups. Their 5α-reductase
inhibitory activities in vitro toward 5α-reductase type 1 and 2 isozymes was evaluated. The in vivo
5α-reductase inhibitory activity of the most active compound 16a was further assessed by prostate
weighting method. In silico ADME analysis was also performed to predict important
pharmacokinetic parameters.
2. Experimental
2.1. Chemistry
2.1.1. General procedure
Melting points of compounds were measured on a RY-1 melting point apparatus and were
uncorrected. Nuclear magnetic resonance (1H NMR) spectra were recorded on a Bruker AV-300
(300 MHz) spectrometer as deuterochloroform (CDCl3) solutions using tetramethylsilane (TMS)
as an internal standard (δ = 0) unless noted otherwise. Electron impact mass spectral (EI-MS) data
were obtained on a SHIMADZU GCMS-QP2010 system. All chemicals were purchased from
commercial sources and were used without further purification unless otherwise noted. The
solvents (such as MeOH, EtOAc, EtOH, CH2Cl2 and others) were C.P. grade purchased from
Nanjing Chemical Co., Ltd. and used without further purification. Column chromatography (CC)
was carried out on silica gel (200–300 mesh, Qingdao Ocean Chemical Company, China).
Thin-layer chromatography (TLC) analyses were carried out on silica gel GF254 (Qingdao Ocean
Chemical Company, China) glass plates (2.5 cm × 10 cm with 250 μm layer). Concentration and
evaporation of the solvent after reaction or extraction was carried out on a rotary evaporator
operated at reduced pressure.
2.1.2. Synthesis of 3β-hydroxyl-5-pregnen-20-one(10)
3

To the solution of 3β-hydroxyl-5,16-Pregnadien-20-one-3-acetate (6g, 16.8mmol) in 180ml
EtOAc, 0.6g Raney Ni was slowly added at 0℃. The reaction mixture was stirred at 25℃for 1h.
After filtration, KOH (0.94g, 16.8mmol) was added. The reaction mixture was refluxed for
another 2h. After cooled to room temperature, the resulting precipitate was filtered, washed with
1
water, and dried to give white solide 4.85g, yield 92.81%. H NMR (CDCl3, 300MHz) δ: 5.3 (s,
1H, 6-H), 3.5 (s, 1H, 3-H), 2.5 (1H, t, 17-H), 2.1(s, 3H, 21-H), 1.0 (s, 3H, 19-CH3), 0.6 (s, 3H,
18-CH3) ppm; ESI-MS m/z: 339[M+Na]+.
2.2. Synthesis of 3-oxo-4-pregnen-20-one(11)
To the solution of 10 (7g, 22mmol) in 112ml totuene, cyclohexanone(21ml, 0.2mol) was added.
After circumfluence to repel water for 2h, aluminium isopropoxide (1.05g, 5.1mmol) was added,
and then refluxed for another 1.5g. The resulting mixture was poured into 10ml cold 10% NaOH
aqueous solution. The solvent was evaporated by steam distillation to provide yellow solid 6.3g,
yield 90.03%. 1H NMR (CDCl3, 300MHz) δ: 5.8 (s, 1H, 4-H), 2.6 (1H, t, 17-H), 2.4 (m, 2H, 2-H),
2.1(s, 3H, 21-H), 1.2 (s, 3H, 19-CH3), 0.7 (s, 3H, 18-CH3) ppm; ESI-MS m/z: 315[M+H]+.
2.3. Synthesis of17β-carboxy-androst-4-ene-3-one(12)
To the solution of NaOH (5.9g, 143mmol) in 50ml water, Br2 (1.9ml, 37mmol) was slowly added
at 0℃ to provide NaOBr aqueous solution. Resolving 11 (3.5g, 11mmol) in 105ml dioxane and
36ml water, the NaOBr aqueous solution was added. After stirred at room temperature for 1.5h,
19ml 10% Na2SO3 was added and refluxed for 15min. After cooled to room temperature, 2N
hydrochloric acid was added to adjust pH to 2 with in ice bath. Dioxane was evaporate and The
resulting precipitate was filtered, washed with water, and dried to give crude product.
Recrystallized by acetone to provide white solid 2.6g, yield 75.48%. 1H NMR (CDCl3, 300MHz) δ:
5.7 (s, 1H, 4-H), 2.4-2.6 (5H, m, 2-H, 6-H, 17-H), 1.2 (s, 3H, 19-CH3), 0.9 (s, 3H, 18-CH3) ppm;
ESI-MS m/z: 315[M-H]-.
2.3.1. Synthesis of 3-Bromo-androsta-5,16-dien-17β-carboxylic acid (13)
To solution of 12(10g, 32mmol) in 52ml acetic acid, PBr3 (4.6ml, ) was added dropwise. The
resultant mixture was then stirred at room temperature for 1h, poured into water and filtered to get
white solid 5 (7.8g, 65% ). ESI-MS m/z: 378[M-H]-.
2.3.2. General procedures of synthesis of 14a~14j
To a solution of 3-Bromo-androsta-5,16-dien-17β-carboxylic acid (13) (1g, 3.16mmol), anhydrous
pyridine (0.32ml, 4.12mmol) in 15ml anhydrous toluene, oxalyl chloride (0.34ml, 3.56mmol) in
7ml anhydrous toluene was added drop wise at 0℃. The resulting solution was stirred at room
temperature 1.5h then corresponding amine (7.9mmol) was added. The mixture was stirred at
room temperature for another 8h. 30ml water was added and the reaction mixture was extracted
with ethyl acetate. It was dried with sodium sulfate and the solvent was removed in vacuum. The
compound was purified by column chromatography (PE /EA, 9:1).
2.3.3. 2.3.2. General procedures of synthesis of 16a~16j
To a solution of 14a~14j (1.35mmol) in 15ml DMF, CuCN (0.24g, 2.7mmol) was added. The
resultant mixture was then stirred at 180℃ for 12 h. After cooled to room temperature, the residue
was removed by filtration. The filtrate was quenched by 75ml water and filtered to get dark yellow
powder without further purification. To a solution of 15a~15j (0.84mmol) in 10ml ethanol, 0.17ml
20% NaOH aqueous solution was added. The resultant mixture was then refluxed for 6h, cooled to
room temperature and removed the ethanol under vacuum. Adjust pH to 2 with 2N hydrochloric
acid in ice bath. The resulting precipitate was filtered, washed with water, and dried to give crude
4

product. The compound was purified by column chromatography (PE/Ethyl acetate, 4:1).
2.3.4. 17β-[N-[2,5-bis(trifluoromethyl)phenyl]amide]-androsta-3,5-dien-3-carboxylic acid (16a)
1
Yield 43%; m.p.: 280-284℃; H NMR (DMSO-d6, 300MHz) δ: 12 (s, 1H, COOH), 9.4 (s, 1H,
NH), 8.0~7.7 (m, 3H, Ar-H), 6.8 (s, 1H, 4-H), 5.9 (s, 1H, 6-H), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H,
18-CH3) ppm; 13C NMR (DMSO-d6, 75 MHz) δ: 175.04, 168.77, 141.20, 137.79, 131.37, 126.52,
55.97,54.99, 48.02, 43.52, 38.11, 34.64, 33.50, 32.12, 31.78, 24.40, 23.73, 22.02, 20.96, 19.13,
13.50;IR (KBr) ν:3434, 2930, 1725, 1384, 1097 cm-1; ESI-MS m/z:554[M-H]+; Anal.calcd for
C29H31F6NO3: C 62.70, H 5.62, N 2.52; Found: C 62.84, H 5.384, N 2.318.
2.3.5. 17β-[N-[3,5-bis(trifluoromethyl)phenyl]amide]-androsta-3,5-dien-3-carboxylic acid (16b)
Yield 23%; m.p.: 140-142℃; 1H NMR (DMSO, 300MHz) δ: 10.2 (s, 1H, COOH), 8.4 (s, 1H, NH),
7.73~7.68 (m, 3H, Ar-H), 7.3 (1H, s, Ar-H), 6.8 (s, 1H, 4-H), 5.7 (s, 1H, 6-H), 0.9 (s, 3H, 19-CH3),
0.8 (s, 3H, 18-CH3) ppm; 13C NMR (DMSO-d6, 75 MHz) δ: 175.1, 168.85, 141.21, 137.76,
131.46, 126.50, 55.95, 54.94, 48.00, 43.56, 38.08, 34.67, 33.48, 32.12, 31.79, 24.42, 23.69, 21.79,
20.92, 19.17, 13.57; IR (KBr) ν:3357, 2941, 1670, 1436, 1134 cm-1; ESI-MS m/z:554[M-H]+;
Anal.calcd for C29H31F6NO3: C 62.70, H 5.62, N 2.52; Found: C 62.95, H 5.298, N 2.309.
2.3.6. 17β-[N-[4-carboxy-3-methylphenyl]amide]-androsta-3,5-dien-3- carboxylic acid (16c)
1
Yield 35%; m.p.: 294-296℃; H NMR (DMSO, 300MHz) δ: 9.0 (s, 1H, NH), 7.8~7.6 (m, 3H,
Ar-H), 6.9 (s, 1H, 4-H), 5.9 (s, 1H, 6-H), 2.3 (s, 3H, Ar-CH3), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H,
18-CH3) ppm; 13C NMR (CDCl3, 75 MHz) δ: 141.41, 132.31, 132.25, 129.58, 87.84, 58.50, 56.53,
53.05, 34.52, 32.77, 29.68, 24.34, 18.97, 17.73, 13.29; IR (KBr) ν: 3439, 2967, 2200, 1681, 1292
cm-1; ESI-MS m/z:476[M-H]+; Anal.calcd for C29H35NO5: C 72.93, H 7.39, N 2.93; Found: C
73.27, H 7.71, N 2.54.
2.3.7. 17β-[N-[4-methoxypheny]amide]-androsta-3,5-dien-3- carboxylic acid (16d)
1
Yield 24%; m.p.: 194-196℃; H NMR (DMSO, 300MHz) δ: 9.4 (s, 1H, NH), 7.5 (d, 2H,
J=8.67Hz, 2’,6’-H), 6.7 (d, 2H, J=8.67Hz, 3’,5’-H), 6.7 (s, 1H, 4-H), 5.7 (s, 1H, 6-H), 3.7 (s, 3H,
-OCH3), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR (CDCl3, 75 MHz) δ: 170.90,
170.27, 156.29, 140.84, 134.88, 131.25, 130.70, 127.58, 121.78, 114.13, 57.84, 56.59, 55.51,
48.06, 44.30, 38.42, 34.75, 33.47, 32.15, 31.76, 24.42, 23, 69, 21.99, 21.03, 18.95, 13.30; IR (KBr)
ν: 3414, 2940, 1661, 1511, 1234 cm-1; ESI-MS m/z:450[M+H]+; Anal.calcd for C28H35NO4: C
74.80, H 7.85, N 3.12; Found: C 74.55, H 8.117, N 2.755.
2.3.8. 17β-[N-[4-chloro-3-trifluoromethylphenyl]amide]-androsta-3,5-dien-3-carboxylic
(16e)
acid
Yield 13%; m.p.: 158-164℃; 1H NMR (DMSO, 300MHz) δ: 10.0 (s, 1H, COOH), 8.2 (s, 1H, NH),
7.9 (d, 1H, J=8.7Hz, 2’-H), 7.7 (d, 1H, J=8.67Hz, 5’-H), 7.3 (m, 1H, 6’-H), 6.7 (s, 1H, 4-H), 5.7 (s,
1H, 6-H), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR (CDCl3, 75 MHz) δ: 171.44,
170.22, 140.81, 136.96, 134.96, 131.93, 130.71, 127.56, 123.66, 58.07, 56.64, 48.00, 44.53, 38.40,
34.74, 33.49, 32.09, 31.74, 30.88, 29.68, 23.69, 21.99, 21.00; IR (KBr) ν: 3339. 2940, 1659, 1413,
1173 cm-1; ESI-MS m/z:523[M+H]+; Anal.calcd for C28H31ClF3NO4: C 64.43, H 5.99, N 2.68;
Found: C 64.55, H 5.776, N 2.556.
2.3.9. 17β-[N-[3-trifluoromethylphenyl]amide]-androsta-3,5-dien-3-carboxylic acid (16f)
1
Yield 17%; m.p.: 140-142℃; H NMR (DMSO, 300MHz) δ: 9.9 (s, 1H, NH), 8.1 (s, 1H, 2’-H),
7.8 (d, 1H, J=8.01Hz, 6’-H), 7.5~7.3 (m, 1H, 4’,5’-H), 6.7 (s, 1H, 4-H), 5.7 (s, 1H, 6-H), 0.9 (s,
3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR (DMSO-d6, 75 MHz) δ: 171.53, 142.92,139.99,
139.47, 132.70, 129.74, 122.79, 120.10, 119.16, 115.35, 105.99, 56.27, 55.7, 47.27, 44.06, 37.35,
5

33.83, 32.19, 31.50, 31.26, 31.06, 24.08, 23.63, 23.34, 20.22, 18.60, 13.38; IR (KBr) ν: 3342.
2937, 1661, 1333, 698 cm-1; ESI-MS m/z:523[M+H]+; Anal.calcd for C28H32F3NO3: C 68.98, H
6.62, N 2.87; Found: C 68.56, H 7.173 N 2.952.
2.3.10. 17β-[N-phenylamide]-androsta-3,5-dien-3-carboxylic acid (16g)
1
Yield 24%; m.p.: 138-140℃; H NMR (DMSO, 300MHz) δ: 9.5 (s, 1H, NH), 7.6 (d, 2H,
J=7.89Hz, 2’,6’-H), 7.3 (t, 2H, J=7.53Hz, 3’,5’-H), 7.0 (t, 1H, J=7.26Hz, 4’-H), 6.7 (s, 1H, 4-H),
5.7 (s, 1H, 6-H), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR (CDCl3, 75 MHz) δ:
171.01, 138.02, 134.97, 130.74, 128.99, 127.46, 124.10, 119.84, 56.62, 48.05, 44.37, 38.51, 34.76,
33.47, 32.14, 31.77, 29.68, 24.41, 23.63, 21.98, 21.04, 18.95, 14.08, 13.28; IR (KBr) ν: 3474,
2936, 1661, 1439, 753 cm-1; ESI-MS m/z:420[M+H]+; Anal.calcd for C27H33NO3: C 77.29, H7.93,
N 3.34; Found: C 77.20, H 7.66 N 3.74.
2.3.11. 17β-[N-phenylamide]-androsta-3,5-dien-3-carboxylic acid (16h)
1
Yield 15%; m.p.: 175-180℃; H NMR (DMSO, 300MHz) δ: 10.0 (s, 1H, COOH), 11.9 (s, 1H,
NH), 8.1 (d, 1H, J=7.4Hz, 6’-H), 7.8 (m, 1H, 5’-H), 7.6 (d, 1H, J=7.52Hz, 3’-H), 7.4 (m, 1H,
4’-H), 6.9 (s, 1H, 4-H), 5.9 (s, 1H, 6-H), 0.85 (s, 3H, 19-CH3), 0.6 (s, 3H, 18-CH3) ppm; 13C NMR
(DMSO-d6, 75 MHz) δ: 175.08, 169.63, 168.84, 161.99, 141.24, 137.75, 134.85, 132.99, 131.39,
129.01, 126.53, 121.30, 56.59, 54.93, 48.01, 45.85, 43.53, 38.07, 37.40, 34.64, 33.61, 31.97, 29.89,
24.39, 23.69, 22.07, 20.92, 19.14, 13.66; IR (KBr) ν: 3457, 2938, 1672, 1610, 774 cm-1; ESI-MS
m/z:423[M-H]+; Anal.calcd for C28H32N2O31/4CH3OH: C 74.97, H 7.35, N 6.19; Found: C 75.06,
H 7.831 N 5.986.
2.3.12. 17β-[N-[2,5-dimethylphenyl]amide]-androsta-3,5-dien-3-carboxylic acid (16i)
Yield 28%; m.p.: 200-205℃; 1H NMR (DMSO, 300MHz) δ: 12.1 (s, 1H, COOH), 8.9 (s, 1H, NH),
7.15 (s, 1H, 6’-H), 7.08 (d, 1H, J=7.68Hz, 3’-H), 6.9 (s, 1H, 4-H), 5.85 (s, 1H, 6-H), 2.3 (s, 3H,
Ar-CH3), 2.2 (s, 3H, Ar-CH3), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR
(DMSO-d6, 75 MHz) δ: 171.17, 141.23, 137.70, 137.03, 135.26, 132.99, 131.43, 130.41, 129.58,
129.32, 126.47, 56.44, 49.04, 48.17, 44.20, 37.96, 34.70, 33.51, 31.98, 29.89, 24.65, 22.01, 20.99 ,
19.14, 18.27, 13.88; IR (KBr) ν: 2963, 2360, 1669, 1284, 804 cm-1; ESI-MS m/z:446[M-H]+;
Anal.calcd for C29H37NO3: C 77.82, H 8.33, N 3.13; Found: C 77.71, H 7.895 N 3.056.
2.3.13. 17β-[N-[2-methoxyphenyl]amide]-androsta-3,5-dien-3-carboxylic acid (16j)
Yield 26%; m.p.: 232-235℃; 1H NMR (DMSO, 300MHz) δ: 12.1 (s, 1H, COOH), 8.4 (s, 1H, NH),
7.9 (d, 1H, J=8.07Hz, 6’-H), 7.1~6.9 (m, 3H,3’,4’,5’-H), 6.9 (s, 1H, 4-H), 5.85 (s, 1H, 6-H), 3.8 (s,
3H, -OCH3), 0.9 (s, 3H, 19-CH3), 0.8 (s, 3H, 18-CH3) ppm; 13C NMR (CDCl3, 75 MHz) δ: 147.71,
141.26, 132.91, 127.90, 123.33, 121.16, 119.80, 109.88, 58.35, 6.58, 55.79, 48.08, 44.15, 38.13,
34.78, 33.39, 32.30, 31.92, 29.69, 24.41, 23.42, 22.68, 21.43, 19.03, 14.10, 13.19; IR (KBr) ν:
3425, 2946, 1673, 14833, 741 cm-1; ESI-MS m/z:448[M-H]+; Anal.calcd for C28H35NO41/2H2O:
C 74.06, H 7.88, N 3.08; Found: C 73.68, H 8.135 N 2.738.
2.4. Biological activity
2.4.1. In vitro rat 5α-reductase inhibitory activity against type 1 and type 2 5α-reductase enzyme
Preparation of tissue. Rat prostatic enzyme was prepared according to the method of Liang et
al.with slight modifications[19]. Male rats were killed and prostates were taken within 5 min and
put in ice cold 0.9% NaCl solution. All the following operations were performed at 0±4℃. The
prostates were dissected free from fat and connective tissue, cut into pieces and weighed. Per 1 g
of tissue, 3 mL of 20 mM phosphate buffer, pH 6.5, were added. The tissue was homogenized,
centrifuged for 30 min at 10,000 g. The supernatant was obtained and the centrifugation was
6

repeated at 100,000 for 1h. The final pellet resuspended in a minimum volume of phosphate buffer
and stored in 300μL portions at -70℃.
Incubation Procedure. The assay was performed as described with modifications[19]. All values
were run in duplicate. The incubation was carried out for 10 min at 37°C in a total volume of
150μL. In the case of rat enzyme preparation, phosphate buffer (40 mM, pH 6.6 for type 1) or
citrate buffer (40 mM, pH 5.5 for type 2) was used. The incubation mixture contained
approximately 200μM NADPH, 0.21 μM T, and 2% DMSO with or without test compound
(10μM). The absorbance at 340nm was tested at beginning and 10min.
2.4.2. In vivo 5α-reductase inhibitory activity
Experiments were performed on SD male rats weighing from 180 to 220g. After gonadectomy, the
rats were was administered 16a orally once daily with the dose of 0.1mg/ml, 0.3mg/ml, 0.6mg/ml
respectively. The rats in positive control group were received an orally once daily of 0.3mg/ml
epristeride Two groups of gonadectomized animals were kept as control; one was injected with
1ml/100g of 2.5% Tween 80, the second with 1 ml/100g of testosterone. After 30 days’ treatment,
the animals were sacrificed. The prostate and seminal vesicles of each animal were dissected and
weighed.
3. Results and discussion
3.1. Chemistry
Taking commercial available compound 16-dehydropregnenolone acetate (9) as starting material,
the compounds 16a~16j can be easily prepared through 8 steps (Scheme 1). Compound 9 could be
selective reduced with Raney Ni in EtOH. After filtration, the hydrolysis of the acetate group at
C-3 was produced by directly adding KOH to resulting filtrate to obtain compound 10. The
resulting secondary alcohol at C-3 was oxidized to the α,β-unsaturated ketone compound 11 by
Oppenauer oxidation using aluminum isopropoxide and cyclohexanone in refluxing toluene.
Compound 11 then reacted with bromine and sodium hydroxide to form the carboxylic acid
compound 12. The resulting compound 12 was treated with PBr3 in HOAc to afford the key
intermediate compound 13. The 17-amide derivatives 14a~14k were achieved by combining
different amides with key intermediate compound 13. The cyanidation of 14a~14j in DMF at 180℃
gave 15a~15j. The compound 16a~16j was eventually prepared form their corresponding 3-CN
compounds by hydrolysis in ethanol containing 20% NaOH.
7

Scheme 1. Synthesis of 16a~16j
3.2. Biological evaluation
3.2.1. In vitro 5α-reductase inhibition assay
To evaluate the inhibitory potencies of the synthesized compounds, we first tested their activities
towards 5α-reductase 1 and 2 isozymes altogether and the result was shown in Table 1. It is
obvious that androsta-3,5-diene-3-carboxylic acid is a favorable skeleton for 5α-reductase
inhibitors. Generally speaking, most of the compounds (16a, 16b, 16g~16j) showed the inhibition
rate more than 50%, especially compound 16a with the IC50 of 0.21μM, which was better than the
positive controls. Different amide substituents were also investigated. The high potencies observed
among 16a and 16h~16j with nanomolar IC50 (0.21-0.61μM) indicated that the substitution at
2’-aniline might be beneficial. The compounds with 3’-subsititution, such as 16c, 16e and 16f,
presented a slightly weaker potency than the unsubstituted compound 16g with the inhibition rate
lower than 60% at 50μM. Meanwhile, it seemed that compounds 16d and 16e were much more
inferior to the others, which might be blamed to the electron donor substitution at 4’-aniline.
Table 1
The 5α-reductase inhibitory activities of compounds 16a~16j
Compound
16a
R
Inhibition%（50μM） IC50(μM)
2,5-CF3
80
60.7
45
0.21
6.51
-
3,5-CF3
16b
3-CH3, 4-COOH
16c
8

4-OCH3
4-Cl, 3-CF3
3-CF3
H
2.7
1.7
-
16d
-
16e
40.7
98.4
100
100
100
100
100
-
16f
16g
27.29
0.65
0.89
0.61
0.59
0.27
2-CN
2,5-CH3
2-OCH3
-
16h
16i
16j
Finasteride(1)
Epristeride(3)
-
To further investigate the inhibitory potency, the inhibition assay was performed on compounds
16a, 16h, 16i and 16j towards the 1 and 2 isozymes respectively. As shown in Table 2, all of the
tested compounds displayed good inhibitory activities to both type 1 and type 2 isozymes, in
which compound 16a was the most potential one with the IC50 of 0.25μM and 0.13μM
respectively. For most compounds, the inhibitory activities towards the type 2 isozyme were
slightly higher than type 1 only except 16i.
Table 2
The inhibitory activities towards the two 5α-reductase isozymes.
Type 1 (pH 6.6)
Type 2 (pH 5.5)
Compound
Inhibition% (1μM) IC50(μM) Inhibition% (1μM) IC50(μM)
16a
16h
70.96
79.28
70.77
71.64
73.69
0.25
0.40
0.58
0.48
1.042
71.32
72.03
79.41
66.22
70.04
0.13
0.29
0.98
0.41
0.04
16i
16j
Finasteride(1)
3.2.2. In vivo inhibitory activity (changes in rat prostate and seminal vesicles weight)
As the most active 5α-reductase inhibitor in vitro, 16a was assessed for its in vivo activity using
prostate weighing method. For each inhibitor to be evaluated, mature male rats were domesticated
and were given once daily with different dose of compounds under test for 30 days and the
inhibitory effects on the growth of prostate and seminal vesicles were examined on 31th day (24h
after last dose). After 30 days’ treatment, the prostatic/body weight ratio of the control group was
129 mg/100g. The prostatic/body weight ratio in group that underwent induced prostatic
hyperplasia by testosterone injection was 491 mg/100g, which was significantly heavier than that
of the control group. As shown in Figure 3, compound 16a produced weight reduction effects in a
dose-response manner on both prostate and seminal vesicles weight, while epristeride, the positive
control, only on prostate. At 6mg/kg, compound 16a displayed a significant weight reduction
effect with the prostatic/body weight ratio of 352mg/100g, better than that of epristeride
(381mg/100g).
9

Fig. 3 Prostatic/body weight ratio of prostate and seminal vesicles glands of castrated rats
receiving treatments for 30 days. The control group was treated with vehicle only. The
pharmacological experiment was carried out in duplicate. Values are mean ± SD (n=10). *P < 0.05,
**P < 0.01, ***P < 0.001 vs. prostate of testosterone group. #P < 0.05, ##P < 0.01, ###P < 0.001 vs.
seminal vesicles of testosterone group.
3.3. In silico ADME property study
Employing Schrödinger software, compounds 16a were subjected to the analyses of several
physiochemical properties (ADME) related to pharmacokinetics[20]. 16a follow Lipinski’s rule of
five for good bioavailability. With the calculated LogP values less than 6.5, it presented good
hydrophilicity with moderate lipophilicity and hence should be able to gain access to membrane
surfaces. The aqueous solubility (LogS) of a compound is a major driving force that leads to good
absorption and distribution characteristics. The calculated LogS value of 16a was within the
acceptable range. Other calculations related to solubility, serum protein binding, blood–brain
barrier (LogBB and apparent MDCK cell permeability), gut-blood barrier (Caco-2 cell
permeability), number of Primary Metabolites, skin permeability (Kp), and human oral absorption
in the gastrointestinal tract showed these values of 16a within the standard ranges for good
bioavailable drugs (Table 3).
Table 3
Various physicochemical (ADME) parameters calculated for 33d and 33e
Parameter
calculated values
Stand. range*
(-6.5/0.5)
(-2.0/6.5)
(-1.5/1.5)
(-3.0/1.2)
(1.0 / 8.0)
log S for aqueous solubility
log P for octanol/water
log K hsa Serum Protein Binding
log BB for brain/blood
No. of Primary Metabolites
-5.570
5.969
1.440
-0.116
4
10

Apparent Caco-2 Permeability (nm/s)
Apparent MDCK Permeability (nm/sec)
log Kp for skin permeability
221
1212
-2.964
84
(<25 poor, >500 great)
(<25 poor, >500 great)
(-8.0 to -1.0, Kp in cm/h)
<25% is poor
% Human Oral Absorption in GI
* Note: for 95% of known drugs based on Schrödinger, QikProp v4.4 (2015) software results.
4. Conclusion
In this study, a series of novel steroidal androst-3, 5-diene-3-carboxylic acid derivatives have
been designed and synthesized in a simple and convenient route. Biological evaluations were
performed on their 5α-reductase inhibitory activity in vitro against 5α-reductase type 1 and 2 and
in vivo by prostate weighing method. Results showed that most of them displayed good
5α-reductase inhibitory activities. It is obvious that the compounds bearing 2’- substituted aniline
(compound 16a, 16h~16j) presented more promising potencies with nanomolar IC50
(0.21-0.61μM). In the detailed evaluation of their inhibition activities against type 1 and 2
respectively, most compounds displayed slightly higher inhibition potency towards the type 2
isozyme than type 1 only except 16i. Among all the compounds, 16a was found to be the most
potential inhibitor with the IC50 of 0.25μM and 0.13μM respectively. In vivo study, compound 16a
displayed a significant weight reduction effects in a dose-response manner on both prostate and
seminal vesicles weight. In silico studies showed that compound 16a possessed good
physiochemical properties (ADME) related to pharmacokinetics. Overall, based on biological
activities data, 16a has been identified as a potential 5α-reductase inhibitor lead molecule which
might be of therapeutic importance for the treatment of BPH.
11

Reference:
[1] Tiwari, A., Krishna, N. S., Nanda, K., & Chugh, A. Benign prostatic hyperplasia: an insight into
current investigational medical therapies. Expert Opinion on Investigational Drugs 2005;14(11): 1359.
[2] Benign prostatic hyperplasia. Nature Reviews Disease Primers 2016;2: 16032.
[3] Aggarwal, S., Thareja, S., Verma, A., Bhardwaj, T. R., & Kumar, M. An overview on 5α-reductase
inhibitors. Steroids 2010; 75(2): 109-153.
[4] Li, J., Ding, Z., Wang, Z., Lu, J. F., Maity, S. N., & Navone, N. M., et al"Androgen Regulation of
5α-Reductase Isoenzymes in Prostate Cancer: Implications for Prostate Cancer Prevention." Plos
One6.12(2011):e28840.
[5] Uemura, M., Tamura, K., Chung, S., Honma, S., Okuyama, A., & Nakamura, Y., et al. "Novel
5α-steroid reductase (SRD5A3, type-3) is overexpressed in hormone-refractory prostate
cancer." Cancer Science99.1(2008):81-6.
[6] Fitzpatrick J M, Artibani W. Therapeutic Strategies for Managing BPH Progression[J]. European
Urology Supplements, 2006, 5(20):997-1003.
[7] Traish A M, Mulgaonkar A, Giordano N. The Dark Side of 5α-Reductase Inhibitors' Therapy:
Sexual Dysfunction, High Gleason Grade Prostate Cancer and Depression. Korean Journal of Urology,
2014, 55(6):367-379.
[8] Cabeza M., Sánchezmárquez, A., Garrido, M., Silva, A., Bratoeff E. Recent Advances in Drug
Design and Drug Discovery for Androgen- Dependent Diseases. Current Medicinal Chemistry, 2016,
23(8):792.
[9] Wu, J. H.; Zhu, Y.; Sun, Z. Y. Yao. Xue. Jin. Zhan. 2002; 26(1):55-57.
[10] Aggarwal S., Thareja S., Verma A., Bhardwaj T. R., Kumar M. "An overview on 5alpha-reductase
inhibitors. "Steroids 2010; 75 (2):109-53.
[11] Holt D. A., Levy M. A., Oh H. J., Erb J. M., Heaslip J. I., Brandt M., et al. "Inhibition of steroid 5
alpha-reductase by unsaturated 3-carboxysteroids. " Journal of Medicinal Chemistry 1990,33
(3):943-50.
[12] Aggarwal S., Thareja S., Bhardwaj T. R., Haupenthal J., Hartmann R. W., Kumar M. "Synthesis
and biological evaluation of novel unsaturated carboxysteroids ashuman 5α-reductase inhibitors: A
legitimate approach." European Journal of Medicinal Chemistry 2012; 54(54):728-39.
[13] Thareja S. Steroidal 5α-Reductase Inhibitors: A Comparative 3D-QSAR Study Review. Chemical
Reviews, 2015, 115(8).
[14] Frye S. V., Haffner C. D., Maloney P. R., Hiner R. N., Dorsey G. F., Noe R. A., et
al.Structure-activity relationships for inhibition of type 1 and 2 human 5 alpha-reductase and human
adrenal
3
beta-hydroxy-delta 5-steroid dehydrogenase/3-keto-delta 5-steroid isomerase by
6-azaandrost-4-en-3-ones: optimization of the C17 substituent.[J]. Journal of Medicinal Chemistry,
1995, 38(14):2621-2627.
[15] Scarpi D., Occhiato E. G., Danza G., Serio M., Guarna A. Synthesis of 17β-N-substituted
19-nor-10-aza- steroids as inhibitors of human 5α-reductases I and II. Bioorgan. Med. Chem., 2002,
10(11): 3455-3461.
[16] Cortés-Benítez F, Cabeza M, Ramírez-Apan MT, Alvarez-Manrique B, Bratoeff E. Synthesis of
17β-N-arylcarbamoylandrost-4-en-3-one derivatives and their anti-proliferative effect on human
androgen-sensitive LNCaP cell line. European Journal of Medicinal Chemistry, 2016, 121:737-746.
[17] Amaral C., Varela C., Correia-Da-Silva G., Silva E. T. D., Rui A. C., Costa S. C. P., et al. New
steroidal 17β-carboxy derivatives present anti-5α-reductase activity and anti-proliferative effects in a
12

human androgen-responsive prostate cancer cell line.[J]. Biochimie, 2013, 95(11):2097-2106.
[18] Xiang Hua, et al., Faming Zhuanli Shenqing, Steroid 5alpha-reductase Inhibitors, Preparation
Methods and Medical Applications, 2012. CN 102532236 A 20120704.
[19] Franck Picard, Stephan Barassin, Armand Mokhtarian A., R.W.H. "Synthesis and evaluation of
2'-substituted 4-(4'-carboxy- or 4'-carboxymethylbenzylidene)-N-acylpiperidines: highly potent and in
vivo active steroid 5alpha-reductase type 2 inhibitors." Journal of Medicinal Chemistry 2002;45
(16):3406-17.
[20] Kumar B S, Singh A, Kumar A, et al. Synthesis of neolignans as microtubule stabilisers[J].
Bioorganic & Medicinal Chemistry, 2014, 22(4):1342.
13

Graphical Abstract
Highlights：

A series of novel steroidal androst-3,5-diene-3-carboxylic acids have been designed and
synthesized.



Most of the synthesized compounds displayed excellent 5α-reductase inhibitory potency.
16a was found to be the most potential inhibitor against type 1 and 2 isozymes.
16a exhibited a significant reduction effect in rat prostate weight in vivo as well as good
pharmacokinetic properties in silico.
14