Synthesis and biological evaluation of steroidal derivatives as selective inhibitors of AKR1B10

Article abstract of DOI:10.1016/j.steroids.2014.04.010

AKR1B10 is a member of human aldo-keto reductase superfamily, and a promising anti-cancer therapeutic target. In this paper, androst-5-ene-3β- ol, dehydroepiandrosterone, pregnenolone and cholesterol were used as reactants, sixteen products were obtained through Jones reaction and reduction reaction using NaBH₄. Their inhibitory activities against AKR1B10 and AKR1B1 were measured. The most active compound (3a) has the IC₅₀ of 0.50 μM for AKR1B10, and the most AKR1B10 selective compound (2a) has the IC ₅₀ of 0.81 μM with AKR1B1/AKR1B10 selectivity of 195. In addition, the binding modes of 2a and 3a in the active site of human AKR1B10 were identified by docking.

Full text of DOI:10.1016/j.steroids.2014.04.010

Steroids 86 (2014) 39–44
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Synthesis and biological evaluation of steroidal derivatives
as selective inhibitors of AKR1B10
Wei Zhang a,1, Ling Wang , Liping Zhang , Wenli Chen , Xinying Chen , Minyu Xie ,
a,1
a
b
c
a
b
a
a
a,
⇑
Guangmei Yan , Xiaopeng Hu , Jun Xu , Jingxia Zhang
a
School of Pharmaceutical Sciences, Sun Yat-sen University, 132 Waihuan East Road, Guangzhou Higher Education Mega Center, Guangzhou 510006, PR China
Department of Pharmacology, Zhongshan School of Medicine, Sun Yat-sen University, 74 Zhongshan Road II, Guangzhou 510080, PR China
Engineering School, Sun Yat-sen University, 132 Waihuan East Road, Guangzhou Higher Education Mega Center, Guangzhou 510006, PR China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
AKR1B10 is a member of human aldo–keto reductase superfamily, and a promising anti-cancer therapeu-
tic target. In this paper, androst-5-ene-3b-ol, dehydroepiandrosterone, pregnenolone and cholesterol
were used as reactants, sixteen products were obtained through Jones reaction and reduction reaction
Received 22 October 2013
Received in revised form 11 March 2014
Accepted 14 April 2014
using NaBH
compound (3a) has the IC50 of 0.50
has the IC50 of 0.81 M with AKR1B1/AKR1B10 selectivity of 195. In addition, the binding modes of 2a
and 3a in the active site of human AKR1B10 were identified by docking.
Ó 2014 Elsevier Inc. All rights reserved.
4
. Their inhibitory activities against AKR1B10 and AKR1B1 were measured. The most active
Available online 30 April 2014
lM for AKR1B10, and the most AKR1B10 selective compound (2a)
l
Keywords:
Aldose reductase
AKR1B10
Inhibitors
Steroidal derivatives
Molecular docking
1
. Introduction
Endo and his colleagues have reported that steroidal derivatives
can selectively inhibit AKR1B10 against AKR1B1 [1].
The human aldo–keto reductase AKR1B10 is
a
NADPH-
To discover higher active and selective AKR1B10 inhibitors, we
designed and synthesized a series of hydroxysteroids and ketoste-
roids, which were tested for inhibiting AKR1B10 with pyridine-3-
aldehyde as substrate. The most AKR1B10 selective inhibitor has
dependent reductase that metabolizes carbohydrates, steroids,
prostaglandins, ketones and other endogenous aldehydes [1].
AKR1B10 is highly expressed in various cancer cells including
hepatocellular carcinoma [2], lung squamous cell carcinoma [3],
lung adenocarcinoma in smokers [4], cervical cancer [5], and breast
cancer [6]. AKR1B10 plays a significant role in the metabolism of
all-trans-retinaldehyde into their corresponding retinols, which
have a close association with cell proliferation and differentiation
the IC50 of about 0.81
. Experimental procedures
The human AKR1B10 and AKR1B1 are human recombinant
lM with AKR1B1/AKR1B10 selectivity of 195.
2
[
7,8]. It is reported that the silencing of the AKR1B10 gene results
in anti-proliferation of certain types of cancer cells, while its
elevated expression promotes cell proliferation [9,10]. Therefore,
enzymes and purchased from prospecbio (Catalog NO. ENZ-416 and
ENZ-390). The reagents were purchased from Wuhan Yuancheng
Technology CO., Ltd. Other solvents used were of analytical grade pur-
chased from commercial sources. The melting points of the agents
were measured by a Tianjing technology company YRT-3 apparatus.
The NMR spectra of the synthesized compounds were measured on a
Bruker 400 MHz NMR instrument with TMS as internal standard.
Chemical shifts are given as d ppm values relative to TMS. Mass spectra
were measured on Thermo mass spectrometer by electron ionization.
AKR1B10 is
a potential tumor biomarker, and a promising
therapeutic target for cancer prevention and treatment.
AKR1B10 shares 71% of amino acid sequence identity with
AKR1B1 which plays an important role in the pathogenesis and
therapeutics of diabetic complications [11,12]. So, selective inhib-
itors of AKR1B10 over AKR1B1 are demanded for the development
of anticancer agents targeting AKR1B10. Some reports have shown
that steroidal derivatives are anti-cancer agents [13,14]. Satoshi
2.1. Chemistry
⇑
Corresponding author. Tel.: +86 20 39943029.
E-mail address: zhjingx@mail.sysu.edu.cn (J. Zhang).
Equal contribution.
Compounds 5a, 5b, 5c, and 5d were synthesized with routine
processes, and other steroids were synthesized using the protocols
from the references [15,16] (see Scheme 1).
1
http://dx.doi.org/10.1016/j.steroids.2014.04.010
039-128X/Ó 2014 Elsevier Inc. All rights reserved.
0

4
0
W. Zhang et al. / Steroids 86 (2014) 39–44
2
.1.1. Androst-4-ene-3,6-dione (2a)
(CH
2
), 23.95 (CH
), 18.63 (CH
2
), 23.78 (CH
), 17.49 (CH
2
), 22.78 (CH ), 22.53 (CH
3
3
), 20.87
Faint yellow solid; yield 89%; m.p. 142.1–143.6 °C. 1H NMR
(CH
), 11.87 (CH ); LR–EI–MS m/z 398
2
3
3
3
+
+
(
400 MHz, CDCl
3
) d 6.17 (s, 1H, 4-H), 2.44–2.55 (m, 2H, 2-H),
.05–2.19 (m, 2H, 7-H), 1.18 (s, 3H, 19-CH ), 0.77 (s, 3H, 18-CH ).
) d 202.20 (C), 199.41 (C), 160.99 (C),
25.42 (CH), 54.76 (CH), 51.25 (CH), 47.07 (CH ), 40.78 (C), 39.95
), 34.52 (CH), 33.96
), 17.52 (CH ), 17.26
). LR–EI–MS m/z 286 (M ); HR–EI–MS Calcd. for C19
(M ); HR–EI–MS Calcd for
398.3177.
C
27
H
42
O
2
:
398.3179 (M ), found:
2
3
3
1
3
C NMR (100 MHz, CDCl
3
1
2
2.1.5. Androst-4-ene-3b,6a-diol (3a)
1
(
(
(
CH
CH
CH
2
2
3
), 39.86 (C), 38.03 (CH
2
), 35.59 (CH
), 20.34 (CH
2
White solid; yield 93%; m.p. 181.6–182.6 °C; H NMR (400 MHz,
DMSO) d 5.50 (s, 1H, 4-H), 4.55(s, 1H, 6-OH), 4.48 (s, 1H, 3-OH),
3.96 (m, 2H, 3-H, 6-H), 0.96 (s, 3H, 19-CH ), 0.69 (s, 3H, 18-CH ).
), 25.25 (CH ), 20.94 (CH
2
2
+
2
3
26 2
H O :
3
3
+
13
2
86.1927 (M ), found: 286.1929.
C NMR (100 MHz, DMSO) d 146.59 (C), 121.80 (CH), 66.49 (CH),
6.18 (CH), 54.20 (CH), 53.56 (CH), 42.57 (CH ), 40.30 (C), 39.80
), 34.18 (CH), 28.85
), 19.37 (CH ), 17.18
). LR–EI–MS m/z 290 (M ); HR–EI–MS Calcd. for C19
6
2
2.1.2. Androst-4-ene-3,6,17-trione (2b)
(CH
(CH
(CH
2
2
3
), 38.12 (CH
), 25.09 (CH
2
), 37.18 (C), 36.12 (CH
), 20.57 (CH ), 20.01 (CH
2
Faint yellow solid; yield 83%; m.p. 220.1–222.4 °C. 1H NMR
2
2
+
2
3
(
400 MHz, CDCl
3
) d 6.20 (s, 1H, 4-H), 2.46–2.59 (m, 2H, 2-H),
.11–2.19 (m, 2H, 7-H), 1.21 (s, 3H, 19-CH ), 0.94 (s, 3H, 18-CH ).
) d 219.12 (C), 201.05 (C), 199.04 (C),
60.13 (C), 125.86 (CH), 51.45 (CH), 50.87 (CH), 47.53 (C), 45.34
), 39.65 (C), 35.54 (CH ), 35.45 (CH ), 33.86 (CH ), 33.66
CH), 30.90 (CH ), 21.55 (CH ), 20.18 (CH ), 17.56 (CH ), 13.64
CH ). LR–EI–MS m/z 300 (M ); HR–EI–MS Calcd. for C19
30 2
H O :
+
2
3
3
290.2240 (M ), found: 290.2242.
1
3
C NMR (100 MHz, CDCl
3
1
2.1.6. Androst-4-ene-3b,6 ,17b-triol (3b)
a
1
(
(
(
CH
2
2
2
2
White solid; yield 95%; m.p. 218.9–220.4 °C; H NMR (400 MHz,
DMSO) d 5.57 (s, 1H, 4-H), 4.62 (s, 1H, 6-OH), 4.55 (s, 1H, 3-OH),
4.42 (s, 1H, 17-OH), 3.95 (m, 2H, 3-H, 6-H), 3.42 (s, 1H, 17-H),
2
2
+
2
3
3
24 3
H O :
+
13
3
00.1720 (M ), found: 300.1721.
0.96 (s, 3H, 19-CH ), 0.64 (s, 3H, 18-CH ). C NMR (100 MHz,
3
3
DMSO) d 146.66 (C), 121.79 (CH), 79.87 (CH), 66.48 (CH), 66.17
(CH), 54.19 (CH), 50.08 (CH), 42.44 (C), 41.84 (CH ), 37.16 (C),
36.36 (CH ), 36.09 (CH ), 33.99 (CH), 29.77 (CH ), 28.85 (CH ),
23.06 (CH ), 20.23 (CH ), 19.38 (CH ), 11.17 (CH ). LR–EI–MS m/z
2
.1.3. Pregn-4-ene-3,6,20-trione (2c)
Faint yellow solid; yield 78%; m.p. 179.7–182.1 °C. 1H NMR
) d 6.18 (s, 1H, 4-H), 2.54–2.60 (m, 2H, 2-H),
.45–2.52 (m, 2H, 7-H), 2.14 (s, 3H, 21-CH ), 1.18 (s, 3H, 19-CH ),
). C NMR (100 MHz, CDCl ) d 208.72 (C),
01.52 (C), 199.19 (C), 160.46 (C), 125.66 (CH), 63.11 (CH), 56.50
2
2
2
2
2
(
400 MHz, CDCl
3
2
2
3
3
+
+
2
0
2
3
3
306 (M ); HR–EI–MS Calcd. for C19
306.2270.
H
30
O
3
: 306.2271 (M ), found:
1
3
.69 (s, 3H, 18-CH
3
3
(
CH), 50.73 (CH), 46.45 (CH
2
), 43.83 (C), 39.65 (C), 38.13 (CH
2
),
),
2.1.7. Pregn-4-ene-3b,6a,20b-triol (3c)
1
3
2
3
3
5.15 (CH
2
), 33.99 (CH), 33.90 (CH
), 20.84 (CH ), 17.51 (CH
2
), 31.38 (CH
3
), 24.11 (CH
2
White solid; yield 89%; m.p. 200.6–202.0 °C; H NMR (400 MHz,
DMSO) d 5.57 (s, 1H, 4-H), 4.60 (s, 1H, 6-OH), 4.54 (d, J = 2.0 Hz,1H,
3-OH), 4.05 (s, 1H, 20-OH), 3.96 (s, 2H, 3-H, 6-H), 3.48 (s, 1H, 17-H),
2.11 (t, 1H, 17-H), 1.54 (s, 3H, 21-CH ), 0.96 (s, 3H, 19-CH ), 0.68 (s,
2.83 (CH
2
2
3
3
), 13.21 (CH ). LR–EI–MS m/z
+
+
28 (M ); HR–EI–MS Calcd. for C21
28.2032.
28 3
H O : 328.2033 (M ), found:
3
3
1
3
3
H, 18-CH
68.23 (CH), 66.54 (CH), 66.20 (CH), 57.62 (CH), 55.11 (CH), 54.19
(CH), 42.39 (C), 39.18 (CH ), 37.14 (C), 36.09 (CH ), 33.78 (CH),
28.86 (CH ), 25.22 (CH ), 24.17 (CH ), 23.71 (CH ), 20.41 (CH ),
19.36 (CH ), 11.96 (CH ). LR–EI–MS m/z 334 (M ); HR–EI–MS
Calcd. for C21 : 334.2532 (M ), found: 334.2533.
3
). C NMR (100 MHz, DMSO) d 146.73 (C), 121.71 (CH),
2
.1.4. Cholest-4-ene-3,6-dione (2d)
Faint yellow solid; yield 86%; m.p. 108.0–110.2 °C. 1H NMR
2
2
(
2
400 MHz, CDCl
3
) d 6.10 (s, 1H, 4-H), 2.62 (m, 2H, 2-H), 2.43 (m,
H, 7-H), 1.09 (s, 3H, 19-CH ), 0.86 (d, J = 5.2 Hz, 3H, 21-H), 0.81
), 0.79 (s, 3H, 26 or 27-CH ), 0.66 (s, 3H, 18-
). C NMR (100 MHz, CDCl ) d 202.24 (C), 199.42 (C), 161.05
),
), 35.65
), 27.98 (CH), 27.97
2
2
2
3
2
+
3
3
3
+
(
s, 3H, 26 or 27-CH
3
3
34 3
H O
1
3
CH
(
4
(
3
3
C), 125.42 (CH), 56.55 (CH), 55.97 (CH), 50.98 (CH), 46.78 (CH
2.53 (C), 39.79 (C), 39.45 (CH ), 39.14 (CH ), 36.05 (CH
2
2.1.8. Cholest-4-ene-3b,6a-diol (3d)
White solid; yield 97%; m.p. 119.2–121.2 °C; H NMR (400 MHz,
DMSO) d 5.52 (s, 1H, 4-H), 4.57 (s, 1H, 6-OH), 4.47 (s, 1H, 3-OH),
1
2
2
2
CH), 35.53 (CH ), 34.20 (CH), 33.94 (CH
2 2
R
1
R1
R2
(
i)
(ii)
HO
O
HO
O
OH
1
2
3
(iii)
R2
R1
(iv)
HO
O
OH
O
4
5
Scheme 1. 1a–1d, 2a–2d and 5a–5d: R
1
= H, O, Ac, CH(CH
3
)CH
2
CH
2
CH(CH
3
2
) , 3a–3d and 4a–4d: R2 = H, OH, CH(OH)CH
3 3 2 2 3 2
, CH(CH )CH CH CH(CH ) , Condition: (i) Jones
reagent; (ii) NaBH , CeCl ; (iii) NaBH , NiCl
4
3
4
2
; (iv) Jones reagent.

W. Zhang et al. / Steroids 86 (2014) 39–44
41
3
2
3
6
.99 (m, 2H, 3-H, 6-H), 0.99 (s, 3H, 19-CH
1-H), 0.83 (s, 3H, 26 or 27-CH ), 0.81 (s, 3H, 26 or 27-CH
). C NMR (100 MHz, DMSO) d 146.61 (C), 121.78 (CH),
6.50 (CH), 66.18 (CH), 55.58 (CH), 55.44 (CH), 53.92 (CH), 42.21
), 42.01 (C), 39.26 (C), 38.91 (CH ), 37.07 (CH ), 36.06 (CH ),
), 33.84 (CH), 28.81 (CH ), 27.70 (CH),
), 22.33 (CH ),
). LR–EI–MS m/z
3
), 0.89 (d, J = 5.2 Hz, 3H,
21.68 (CH
288 (M ); HR–EI–MS Calcd. for C19
208.2080.
2
), 20.35 (CH
2
), 17.43 (CH
3
), 13.57 (CH
3
). LR–EI–MS m/z
+
+
3
3
), 0.65 (s,
H
28
O
2
: 288.2084 (M ), found:
1
3
H, 18-CH
3
(
CH
2
2
2
2
35.60 (CH), 35.14 (CH
27.35 (CH ), 23.80 (CH
20.57 (CH ), 19.32 (CH
4
4
2
2
2.1.14. 5a-Androst-3,6,17-trione (5b)
2
2
), 23.19 (CH
), 18.45 (CH
2
), 22.60 (CH
3
3
1
White solid; yield 85%; m.p. 177.1–178.6 °C; H NMR (400 MHz,
2
3
3
), 11.72 (CH
3
CDCl
.96 (s, 3H, 19-CH
CDCl ) d 219.31 (C), 210.58 (C), 208.03 (C), 57.46 (CH), 53.46
CH), 51.50 (CH), 48.02 (C), 45.26 (CH
3
) d 2.59 (m, 1H, 5-H), 2.45 (m, 2H, 7-H), 2.32 (m, 2H, 2-H),
+
+
02 (M ); HR–EI–MS Calcd. for C27
02.3520.
H
46
O
2
: 402.3518 (M ), found:
13
0
3 3
), 0.87 (s, 3H, 18-CH ). C NMR (100 MHz,
3
(
2
), 41.12 (C), 37.95 (CH
), 35.56 (CH ), 31.08 (CH
2
),
),
2.1.9. 5a-Androst-3b,6b-diol (4a)
3
2
3
3
7.42 (CH), 37.23 (CH
1.61 (CH ), 20.91 (CH
02 (M ); HR–EI–MS Calcd. for C19
2
), 36.86 (CH
), 13.77 (CH
2
2
2
1
White solid; yield 76%; m.p. 169.9–172.3 °C; H NMR (400 MHz,
2
2
3
3
), 12.55 (CH ); LR–EI–MS m/z
DMSO) d 4.39 (d, J = 4.8 Hz, 1H,3-OH), 4.08 (d, J = 3.6 Hz, 1H, 6-OH),
+
+
H
26
O
3
: 302.1876 (M ), found:
3
1
5
.57 (s, 1H, 3-H), 3.35 (m, 1H, 6-H), 0.93 (s, 3H, 19-CH
3
), 0.69 (s, 3H,
). C NMR (100 MHz, DMSO) d 69.97 (CH), 69.67 (CH),
4.11 (CH), 53.87 (CH), 47.24 (CH), 40.43 (C), 40.25 (CH ), 39.98
), 35.18 (C), 31.44
), 20.06 (CH ), 17.31
). LR–EI–MS m/z 292 (M ); HR–EI–MS Calcd. for
02.1872.
1
3
8-CH
3
2
(
(
(
CH
CH
CH
2
2
3
), 38.41 (CH
), 30.36 (CH), 25.19 (CH
), 15.49 (CH
: 292.2397 (M ), found: 292.2399.
2
), 38.23 (CH
2 2
), 35.83 (CH
2.1.15. 5a-Pregn-3,6,20-trione (5c)
2
), 20.64 (CH
2
2
1
+
White solid; yield 76%; m.p. 191.5–192.9 °C; H NMR (400 MHz,
3
+
CDCl
3
) d 2.61 (t, J = 10.8 Hz, 1H, 5-H), 2.56 (t, J = 4 Hz, 1H, 17-H),
.42 (m, 2H, 7-H), 2.35 (m, 2H, 2-H), 0.96 (s, 3H, 19-CH ), 0.65 (s,
) d 210.88 (C), 208.82 (C),
08.43 (C), 63.36 (CH), 57.52 (CH), 56.64 (CH), 53.37 (CH), 46.39
C
19
H
32
O
2
2
3
2
3
1
3
3 3
H, 18-CH ). C NMR (100 MHz, CDCl
2
.1.10. 5
White solid; yield 73%; m.p. 207.1–208.8 °C; H NMR (400 MHz,
DMSO) d 4.40 (d, J = 4.8 Hz, 1H, 3-OH), 4.37(d, J = 4.0 Hz, 1H, 6-OH),
.08 (d, J = 3.2 Hz, 1H, 17-OH), 3.55 (m, 1H, 17-H), 3.42 (m, 1H, 3-
a-Androst-3b,6b,17b-triol (4b)
1
(
C), 44.35 (CH
2
), 41.18 (C), 38.46 (CH
), 36.91 (CH ), 31.42 (CH ), 24.23 (CH
), 13.46 (CH ), 12.63 (CH
: 330.2237 (M ), found: 330.2239.
2
), 38.13 (CH), 37.84 (CH
2
),
),
3
2
7.36 (CH
1.65 (CH
2
2
2
2
), 22.87 (CH
3
4
+
1
3
2
3
3
). LR–EI–MS m/z 330 (M );
H). C NMR (100 MHz, DMSO) d 80.10 (CH), 69.97 (CH), 69.59
+
30 3
HR–EI–MS Calcd. for C21H O
(
CH), 54.12 (CH), 50.40 (CH), 47.34 (CH), 42.55 (C), 39.50 (CH
2
2
3
),
),
),
3
3
1
C
8.21 (CH
0.13 (CH), 29.85 (CH
1.29 (CH
: 308.2346 (M ), found: 308.2350.
2
), 36.64 (C), 35.82 (CH
2
), 35.15 (CH ), 31.42 (CH
2
2
), 23.15 (CH
2
), 20.29 (CH ), 15.49 (CH
2
+
3
). LR–EI–MS m/z 308 (M ); HR–EI–MS Calcd. for
2
.1.16. 5
White solid; yield 87%; m.p. 161.7–163.1 °C; H NMR (400 MHz,
CDCl ) d 2.57 (t, J = 8.8 Hz, 1H, 5-H), 2.34 (m, 2H, 4-H), 2.05 (m, 2H,
-H), 1.84 (m, 2H, 7-H), 0.94 (s, 3H, 19-CH ), 0.90 (d, J = 6 Hz, 3H,
1-CH ), 0.86 (d, J = 3.2 Hz, 3H, 26-CH or 27-CH ), 0.84 (d,
or 27-CH ), 0.67 (s, 3H, 18-CH
) d 211.15 (C), 209.03 (C), 57.50 (CH), 56.61
CH), 56.12 (CH), 53.49 (CH), 46.60 (CH ), 43.00 (C), 41.22 (C),
a-Cholest-3,6-dione (5d)
+
H
32
O
3
1
19
3
2
.1.11. 5
White solid; yield 82%; m.p. 174.0–175.3 °C; H NMR (400 MHz,
DMSO) d 4.39 (d, J = 4.4, 1H, 20-OH), 4.06 (d, J = 3.2 Hz, 1H, 3-OH),
a-Pregn-3b,6b,20b-triol (4c)
2
2
3
1
3
3
3
13
J = 3.6 Hz, 3H, 26-CH
3
3
3
). C NMR
4
0
.02 (d, J = 5.2 Hz, 1H, 6-OH), 3.56 (m, 1H, 20-H), 3.48 (m, 1H, 3-H),
(
(
100 MHz, CDCl
3
1
3
.99 (s, 3H, 21-CH
3
), 0.92 (s, 3H, 19-CH
3
), 0.68 (s, 3H, 18-CH
3
).
C
2
NMR (100 MHz, DMSO) d 70.00 (CH), 69.70 (CH), 68.30 (CH), 57.80
3
3
2
1
9.45 (CH
6.97 (CH
3.99 (CH
8.63 (CH
2
), 39.39 (CH
), 36.06 (CH
2
), 38.10 (CH
), 35.67 (CH), 28.01 (CH
), 22.78 (CH ), 22.53 (CH
), 12.00 (CH
: 400.3336 (M ), found: 400.3339.
2
), 38.02 (CH), 37.36 (CH
2
),
(
CH), 55.46 (CH), 54.09 (CH), 47.32 (CH), 42.17 (C), 40.02 (CH
2
2
2
),
),
),
2
2
2
), 27.98 (CH),
3
2
1
9.44 (CH
9.93 (CH), 25.30 (CH
5.49 (CH ), 12.08 (CH
: 336.27235 (M ), found: 336.2733.
2
), 38.21 (CH
2
), 35.83 (CH
2
), 35.10 (C), 31.43 (CH
2
), 23.79 (CH
), 12.53 (CH
2
3
3 2
), 21.67 (CH
),
), 24.26 (CH
2
), 23.73 (CH
3
), 20.48 (CH
+
2
3
); LR–EI–MS m/z 400 (M );
3
3
+
); LR–EI–MS m/z 336 (M ); HR–EI–MS
+
3
3
HR–EI–MS Calcd. for C19
H
O
3
30
+
36 3
Calcd. for C21H O
2
.1.12. 5
White solid; yield 79%; m.p. 189.2–190.3 °C; H NMR (400 MHz,
DMSO) d 4.38 (d, J = 4.4 Hz, 1H, 3-OH), 4.06 (d, J = 4.0 Hz, 1H, 6-
OH), 3.55 (m, 1H, 3-H), 3.39 (m, 1H, 6-H), 0.97 (s, 3H, 19-CH ),
), 0.82 (s,
). C NMR (100 MHz, DMSO)
a
-Cholest-3b,6b-diol (4d)
Table 1
1
The inhibitory potency of AKR1B10 and AKR1B1.
Compound
Inhibitiona
3
AKR1B10 (%)
AKR1B1 (%)
0
3
.91 (d, J = 5.2 Hz, 3H, 21-H), 0.85 (s, 3H, 26 or 27-CH
3
H, 26 or 27-CH ), 0.66 (s, 3H, 18-CH
3 3
Androst-4-ene-3,6-dione (2a)
Androst-4-ene-3,6,17-trione (2b)
Pregn-4-ene-3,6,20-trione (2c)
Cholest-4-ene-3,6-dione (2d)
Androst-4-ene-3b,6
Androst-4-ene-3b,6
87
44
33
6
86
7
17
4
63
13
8
3
7
8
5
9
22
11
0.9
15
7
3
5
13
18
10
8
13
d 70.01 (CH), 69.64 (CH), 55.95 (CH), 55.86 (CH), 53.97 (CH), 47.35
CH), 42.28 (C), 42.14 (CH ), 35.75 (CH ), 35.70 (CH ), 35.40 (CH),
), 35.00 (C), 31.38 (CH ), 30.00 (CH), 27.83 (CH
7.39 (CH), 23.85 (CH ), 23.46 (CH ), 22.49 (CH ), 22.44 (CH
2.38 (CH ), 22.24 (CH ), 22.17 (CH ), 18.40 (CH ), 15.39 (CH
1.78 (CH
: 404.3748 (M ), found: 404.3747.
(
2
2
2
a
a
-diol (3a)
,17b-triol (3b)
3
2
2
1
C
5.31 (CH
2
2
2
2
3
),
),
),
2
2
2
Pregn-4-ene-3b,6
a
,20b-triol (3c)
-diol (3d)
-Androst-3b,6b-diol (4a)
-Androst-3b,6b,17b-triol (4b)
-Pregn-3b,6b,20b-triol (4c)
-Cholest-3b,6b-diol (4d)
-Androst-3,6-dione (5a)
-Androst-3,6,17-trione (5b)
-Pregn-3,6,20-trione (5c)
-Cholest-3,6-dione (5d)
3
3
2
3
Cholest-4-ene-3b,6
a
+
3
). LR–EI–MS m/z 404 (M ); HR–EI–MS Calcd. for
5a
5a
5a
5a
5a
5a
5a
5a
5
+
27
48 2
H O
2
2
.1.13. 5
White solid; yield 82%; m.p. 142.1–143.6 °C; H NMR (400 MHz,
CDCl ) d 2.54 (m, 2H, 4-H), 2.34 (m, 2H, 7-H), 2.25 (m, 2H, 2-H),
.90 (s, 3H, 19-CH
CDCl ) d 211.31 (C), 209.12 (C), 57.42 (CH), 54.75 (CH), 53.66
CH), 46.89 (CH ), 41.28 (C), 41.22 (C), 40.06 (CH ), 38.35 (CH),
), 38.13 (CH ), 37.38 (CH ), 36.98 (CH ), 25.25 (CH ),
a-Androst-3,6-dione (5a)
67
39
26
4
1
3
13
0
3 3
), 0.67 (s, 3H, 18-CH ); C NMR (100 MHz,
b-Cholanic acid
89
12
3
a
(
3
2
2
The inhibition percentages for AKR1B10 and AKR1B1 were tested with the
compound concentration of 10 M.
l
8.21 (CH
2
2
2
2
2

4
2
W. Zhang et al. / Steroids 86 (2014) 39–44
2.2. Biological assays
into the host crystal structure to confirm the performance of FlexX
for the human AKR1B10 system. The RMSD between the docking
pose of native ligand and the experimental pose in the structure
complex was less than 1 Å. The active sites were defined as all resi-
dues within 6.5 Å radius of the reference molecule in the optimized
structure and cofactors were retained during the docking process.
Other FlexX parameters were set to default values. 30 poses were
retained during the docking process. After running FlexX, all the
poses were visually inspected, and the most suitable docking pose
was selected on the basis of score and interactions with key residues
of the active site (e.g., Tyr49 and His111, Phe1123).
In order to probe the selection of binding modes of 2a and 3a to
AKR1B1, 2a and 3a were docked into the crystal structure of
AKR1B1 (PDB ID: 1US0 [21]) using FlexX program. The cofactor
was retained during the docking process. Other FlexX parameters
were set to default values.
The reductase activity of AKRs was determined by measuring
the rate of change in NADPH absorbance at 340 nm [17,18]. The
standard reaction mixture consisted of 0.1 M potassium phos-
phate, pH 7.4 (pH 6.4 with 0.4 M Li SO for AKR1B1), 0.1 mM
2 4
NADPH, 0.2 mM pyridine-3-aldehyde and enzyme in a total vol-
ume of 1.0 mL. The IC50 values for inhibitors were determined with
0
.2 mM pyridine-3-aldehyde as the substrate. The inhibitors,
which are hardly soluble in water, were dissolved in DMSO, and
added into the reaction mixture, in which the final concentration
of DMSO was less than 1%. The inhibition pattern was analyzed
by fitting the initial velocities in the lineweaver–burk plot and
the IC50 values are expressed as the means ± standard errors of at
least three determinations.
2.3. Molecular docking study
3
. Results and discussion
The X-ray crystal structure of human AKR1B10 (PDB ID: 1ZUA
[
13]) was retrieved from the Protein Data Bank. The complex was
3.1. Chemistry
then optimized with the steepest descent method for 200 steps
using Discovery Studio 2.1 (Accelrys Inc., San Diego, CA.). The dock-
ing program FlexX encoded in SYBYL 7.3 [19,20] was applied to iden-
tify the potential binding of steroidal derivatives (2a and 3a) to the
AKR1B10. As a consistency check native ligands were docked back
A series of keto-steroids and hydroxyl-steroids were obtained
with Jones reaction at the positions 3 and 6 of the steroid ring.
The reaction yield ranges from 78% to 89%. The keto-steroids
(2a–2d) were reduced by sodium borohydride with different
Table 2
Steroidal derivative inhibitions on the reductase activities of AKR1B10 and AKR1B1.
Compound
IC50 values (
AKR1B10
lM)
Ratio
AKR1B1
158 ± 13
AKR1B1/AKR1B10
195
0.81 ± 0.08
O
O
2a
0
.50 ± 0.05
.56 ± 0.68
68 ± 4.9
53 ± 4.1
136
HO
HO
OH
OH
3a
6
8
4
a
2
.03 ± 0.17
59 ± 5.3
29
O
O
5a
(0.51 ± 0.06)a
(82 ± 5.9)b
161
O
OH
OH
OH
β-Cholic acid
HO
H
5
The IC50 values were tested by six concentrations three times.
a
Reference value is 2.2
Reference value is 78
l
M.
b
lM.

W. Zhang et al. / Steroids 86 (2014) 39–44
43
Fig. 1. Binding modes of compounds 2a (a), 3a (b), the right represents the 3-D surface ligand-interaction diagram and the left represents the 3-D ligand-interaction diagram,
and (c) superimposition of the binding modes of compounds 2a (green) and 3a (light blue). Hydrogen bonds are depicted by dark dotted lines. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
catalysts to produce different products. In order to boost the speed
of the reaction, cerium trichloride or nickel dichloride was added
into the reaction system as catalysts. Double bond was not affected
by Ce , while the double bond was reduced when there was Ni
in the system. The polyhydroxy-steroids (4a–4d) can be oxidized
by Jones reagent in the same condition with 2a–2d.
data, we found that the double bond at the position 4 of the steroid
ring also has a significant influence on the inhibitory activity
against AKR1B10. In addition, there is no distinct evidence showing
that hydroxyl and carbonyl at the positions 3 and 6 of the
steroid ring will show a difference in the inhibitory activity against
AKR1B10.
3
+
2+
3.2. Biological evaluation
3.3. Binding mode analysis
The inhibitory data of all the compounds toward AKR1B10 and
The binding modes of 2a and 3a in the active site of human
AKR1B10 were identified by docking. The detailed results are
shown in Fig. 1. As shown in Fig. 1a, carbonyl oxygen of 2a forms
two hydrogen bonds with the side chain of Tyr49 and His111.
The hydrophobic part of 2a also forms hydrophobic interactions
with Phe123, Trp80, Trp220, Cys299 and Val301. The hydroxyl
of 3a forms one hydrogen bond with Trp112. 3a shows a similar
hydrophobic interaction with that of 2a (Fig. 1b and c). The dock-
ing scores of 2a and 3a are À10.11 and À10.78 kcal/mol, respec-
tively, which was consistent with our bioassay results (Table 2).
AKR1B1 are listed in Table 1 and Table 2. Some of the newly syn-
thesized compounds showed potent inhibition for AKR1B10. All
the tested compounds showed low inhibitory activity against
AKR1B1. The inhibitory potency was highly affected by the R at
the position 17 in the steroid ring. When the R was H, it showed
the most potent inhibitory activity. However, cholest owns the
lowest inhibition activity against AKR1B10. It seems that R substi-
tute at the position 17 in the steroid ring by carbonyl has more
potent inhibitory activity than acetyl. As a contrast to all the test
Fig. 2. Binding modes of compounds 2a (a) and 3a (b) with human AKR1B1.

4
4
W. Zhang et al. / Steroids 86 (2014) 39–44
Tyr49, Trp112 and His111 play a key role in enzyme catalytic and
inhibitor activity [7], if compound can form three hydrogen
bonds with these three residues, and will exhibit excellent inhi-
bition activity against this enzyme. 2a and 3a cannot make full
use of these three residues to form hydrogen bond, which may
result its inhibition level at micromolar. In further design of
new steroidal derivatives, an additional carboxyl group in the
steroid ring may be helpful in improving the activity against
AKR1B10.
Compared with AKR1B10, the binding modes of 2a and 3a with
AKR1B1 are shown in Fig. 2. According to our previous study [19],
compound with AKR1B1 inhibition activity must form key hydro-
gen bonds with the catalytic domain that consists of Tyr48,
His110 and Trp111 (Fig. 2). Moreover, hydrophobic interactions
from Phe112, leu300, Phe115, and Pro310, are important in sup-
porting AKR1B1 inhibition activity. As shown in Fig. 2, 2a and 3a
can form hydrogen bonds with Tyr48, His110 or Trp111, which
may result in inactivity against AKR1B1. While the hydrogen bonds
between corresponding residues in AKR1B10 (Tyr49, His111 and
Trp112, Fig. 1) are formed. These results suggest that 2a and 3a will
show better activity against AKR1B10 than that of AKR1B1. Car-
bonyl oxygen of 2a is unfavorable to form hydrophobic interac-
tions with leu300, Phe115 and Pro310 (Fig. 2a), while these
unfavorable interactions from compound 3a are weak (only one
polar oxygen located in this hydrophobic binding pocket, Fig. 2b),
indicating that 3a will show better inhibition activity against
AKR1B1 than that of 2a. Our binding mode analysis results are con-
sistent with bioassay results (Table 2).
References
[
1] Endo S, Matsunaga T, Mamiya H, Ohta C, et al. Kinetic studies of AKR1B10,
human aldose reductase-like protein: endogenous substrates and inhibitions
by steroids. Arch. Biochem. Biophys. 2009;487:1–9.
[2] Scuric Z, Stain SC, Anderson WF, Hwang JJ. New member of aldose reductase
family proteins overexpressed in human hepatocellular carcinoma.
Hepatology 1998;27:943–50.
[
3] Heringlake S, Hofdmann M, Fiebeler A, Manns MP, et al. Identification and
expression analysis of the aldo-ketoreductase1-B10 gene in primary malignant
liver tumours. J. Hepatol. 2010;52:220–7.
[
4] Fukumoto S, Yamauchi N, Moriguchi H, Hippo Y, et al. Overexpression of the
aldo–keto reductase family protein AKR1B10 is highly correlated with
smokers’ non-small cell lung carcinomas. Clin. Cancer Res. 2005;11:1776–85.
[5] Yoshitake H, Takahashi M, Ishikawa H, Nojima M, et al. Aldo-keto reductase
family 1, member B10 in uterine carcinomas: potential risk factor of
recurrence after surgical therapy in cervical cancer. Int. J. Gynecol. Cancer
007;17:1300–6.
a
2
[6] Ma J, Luo DX, Huang C, Shen Y, Bu Y, Markwell S, et al. AKR1B10 overexpression
in breast cancer: association with tumor size, lymph node metastasis and
patient survival and its potential as a novel serum marker. Int. J. Cancer
2012;131:E862–71.
[7] Gallego O, Ruiz FX, Ardèvol A, Domínguez M, Alvarez R, de Lera AR, et al.
Structural basis for the high all-trans-retinaldehyde reductase activity of the
tumor marker AKR1B10. Proc. Natl. Acad. Sci. U.S.A. 2007;104:20764–9.
[
8] Penning TM. AKR1B10: a new diagnostic marker of non-small cell lung
carcinoma in smokers. Clin. Cancer Res. 2005;11:1687–90.
[
9] Yan R, Zu X, Ma J, Liu Z, Adeyanju M, Cao D. Aldo-keto reductase family 1 B10
gene silencing results in growth inhibition of colorectal cancer cells:
implication for cancer intervention. Int. J. Cancer 2007;121:2301–6.
[10] Wang C, Yan R, Luo D, Watabe K, et al. Aldo-keto reductase family 1 member
B10 promotes cell survival by regulating lipid synthesis and eliminating
carbonyls. J. Biol. Chem. 2009;284:26742–8.
[
11] Cao D, Fan ST, Chung SSM. Identification and characterization of a novel
human aldose reductase-like gene. J. Biol. Chem. 1998;273:11429–35.
[12] Hyndman DJ, Flynn TG. Sequence and expression levels in human tissues of a
new member of the aldo-keto reductase family. Biochim. Biophys. Acta
1
998;1399:198–202.
13] Amaral C, Varela C, Correia-da-Silva G, et al. New steroidal 17b-carboxy
derivatives present anti-5 -reductase activity and anti-proliferative effects in
human androgen-responsive prostate cancer cell line. Biochimie
013;95:2097–106.
14] Yu B, Shi XJ, Ren JL, Sun XN, Qi PP, et al. Efficient construction of novel D-ring
modified steroidal dienamides and their cytotoxic activities. Eur. J. Med. Chem.
2013;66:171–9.
15] Hunter AC, Priest SM. An efficient one-pot synthesis generating 4-ene-3,6-
dione functionalised steroids from steroidal 5-en-3b-ols using a modified
Jones oxidation methodology. Steroids 2006;71:30–3.
[16] Jian-guo CUI, Long-mei ZENG, Jing-yu SU, Cui-wu LIN. Regio-and stereo-
selective reductions of steroidal 4-en-3,6-dione. Chem. Res. Chin. Univ.
[
4
. Conclusion
a
a
2
A series of steroid compounds were designed and synthesized
[
[
and their inhibitory activities against AKR1B10 and AKR1B1 were
measured. Among them, 3a exhibited the highest inhibitory activ-
ity against AKR1B10 (IC50 = 0.5
potent inhibitory activities compared with the other 12 steroids,
and 2a has the IC50 of about 0.81 M with AKR1B1/AKR1B10 selec-
tivity of 195. The cholest showed the lowest inhibitory activity
against AKR1B10. The binding modes of 2a and 3a in the active site
of human AKR1B10 were identified by docking. The 2a forms two
hydrogen bonds with the side chain of Tyr49 and His111 by its car-
bonyl oxygen, 3a forms one hydrogen bond with Trp112 by its
hydroxyl.
lM); 2a, 4a, and 5a also showed
l
2002;18:400–4.
[
[
[
[
[
17] Zhang L, Zhang H, Zhao Y, Li Z, Chen S, et al. Inhibitor selectivity between aldo-
keto reductase superfamily members AKR1B10 and AKR1B1: Role of Trp112
(
Trp111). FEBS Lett. 2013;587:3681–6.
18] Shen Y, Zhong L, Johnson S, Cao D. Human aldo-keto reductases 1B1 and 1B10:
A comparative study on their enzyme activity toward electrophilic carbonyl
compounds. Chem. Biol. Interact. 2011;191:192–8.
19] Wang L, Gu Q, Zheng X, Ye J, Liu Z, Li J, et al. Discovery of new selective human
aldose reductase inhibitors through virtual screening multiple binding pocket
conformations. J. Chem. Inf. Model. 2013;53:2409–22.
20] Kramer B, Rarey M, Lengauer T. Evaluation of the FLEXX incremental
construction algorithm for protein-ligand docking. Proteins: Struct. Funct.
Bioinf. 1999;37:228–41.
21] Howard EI, Sanishvili R, Cachau RE, Mitschler A, Chevrier B, Barth P, et al.
Ultrahigh resolution drug design I: details of interactions in human aldose
reductase-inhibitor complex at 0.66 Å. Proteins Struct. Funct. Bioinf. 2004;55:
792–804.
Acknowledgments
This work was supported by The National ‘‘Eleventh five’’ Muni-
cipal Innovative Dugs Program (2000 ZX109102-013); The Munici-
pal Science and Technology Project of Guangdong Province
(
2012A080201008).