Preparation of 4-aryl-2-trifluoromethylbenzonitrile derivatives as androgen receptor antagonists for topical suppression of sebum production

Article abstract of DOI:10.1016/j.bmcl.2007.08.034

A series of substituted 4-aryl-2-trifluoromethylbenzonitrile analogs were evaluated in the human androgen receptor binding and cellular functional assays. Analogs with sufficient in vitro binding and cellular potency (IC₅₀ < 200 nM) were tested in the progesterone receptor binding assay for selectivity and in the Golden Syrian hamster ear model for in vivo efficacy. Within the series, compound 4e was identified to be the most active analog in vivo (wax ester inhibition = 86%).

Full text of DOI:10.1016/j.bmcl.2007.08.034

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5529–5532
Preparation of 4-aryl-2-trifluoromethylbenzonitrile derivatives
as androgen receptor antagonists for topical
suppression of sebum production
a,
a
a
b
a
a
*
Jennifer A. Van Camp, Lain-Yen Hu, Catherine Kostlan, Bruce Lefker, Jie Li,
a
a
a
a
Lorna Mitchell, Zhi Wang, Wen-Song Yue, Matthew Carroll, Danielle Dettling,
a
a
a
Daniel Du, David Pocalyko and Kimberly Wade
a
Pfizer Global Research and Development, Michigan Laboratories, Ann Arbor, MI 48105, USA
b
Pfizer Global Research and Development, Groton Laboratories, Groton, CT 06340, USA
Received 2 July 2007; revised 13 August 2007; accepted 15 August 2007
Available online 19 August 2007
Abstract—A series of substituted 4-aryl-2-trifluoromethylbenzonitrile analogs were evaluated in the human androgen receptor bind-
ing and cellular functional assays. Analogs with sufficient in vitro binding and cellular potency (IC50 < 200 nM) were tested in the
progesterone receptor binding assay for selectivity and in the Golden Syrian hamster ear model for in vivo efficacy. Within the series,
compound 4e was identified to be the most active analog in vivo (wax ester inhibition = 86%).
ꢀ
2007 Elsevier Ltd. All rights reserved.
One of the most common skin conditions treated by
physicians is acne vulgaris, which affects about 17 mil-
lion people in the USA. In addition to permanent scar-
ring, many people with this disease develop
psychological problems such as anxiety, social inhibition
the androgen receptor, but exhibited weak AR agonist
activity. This compound was reevaluated as part of a
project to identify androgen receptor modulators for
controlling sebum production and was found to be
potent as an AR antagonist (cellular IC50 = 3 nM).
However, due to the potential phototoxicity of the aryl
1
and depression. Sebum production is a pathogenic
factor that increases the onset of acne vulgaris infec-
7
_
chloride group, the chloride of compound 1 was
tions in the pilosebaceous unit. The production of se-
bum is regulated by the androgens testosterone and
replaced with a trifluoromethyl group. In this paper,
we report a series of potent AR antagonists, some of
which are active in both the binding and cellular
functional assays. In particular, we highlight the analog
4e as the most potent example, both in vitro and in vivo,
within the 4-aryl-2-trifluoromethylbenzonitrile series.
5
a-dihydrotestosterone (5a-DHT). Testosterone is con-
verted to 5a-DHT by the Type I 5a-reductase enzyme
prevalent in the sebaceous glands and epidermis. Fur-
2
thermore, DHT is thought to mediate androgen-depen-
dent skin diseases such as hirsutism, acne, and
3
androgenic alopecia. Androgen receptor (AR) antago-
nists known to suppress sebum production when applied
Cl
CF3
NC
NC
4
,5
OH
topically are steroidal cyproterone acetate
and the
6
hydantoin analog RU-58841. There is an increased
need for additional non-steroidal AR antagonists for
topical applications.
1
4e
The preparation of the biphenyl analogs described is
8
outlined in Scheme 1. Synthesis began with the diazoti-
Earlier work in the Pfizer laboratories identified the
biphenyl derivative 1 as a compound that bound to
zation of commercially available aniline 2, followed by
iodine insertion to afford the aryl iodide 3 in 80% yield
over two steps. Further elaboration to their biphenyl
targets 4a–m was achieved under Suzuki coupling condi-
tions with the appropriate aryl boronic acid.
Keywords: Androgen receptor antagonists; Sebum production inhib-
itors; Biaryl analogs.
*
Corresponding author. E-mail: jennifer_vancamp@yahoo.com
0
960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.08.034

5530
J. A. Van Camp et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5529–5532
CF3
CF3
CF3
NC
a,b
NC
c
NC
NH2
I
R1
2
3
4a-m
Scheme 1. Reagents and conditions: (a) NaNO
DME/H
O, 80 ꢁC, 16 h, 62–88%.
2
, H
2
O/HCl, 0 ꢁC; (b) KI, H
2 2 3 4 2 3
O, 0 ꢁC to rt, 18 h; 80% over 2 steps; (c), Ar–B(OH) , Pd(PPh ) , K CO ,
2
All new analogs were evaluated in the human androgen
,10
The 2-OMe compound (4b) was not as potent as its
desmethyl counterpart; however, it still exhibited both
good androgen receptor binding as well as cellular activ-
ity. Thus, we hypothesized that a ligand with enough
bulk and/or length at the 2-position may interact with
the receptor at the ligand-binding domain to such an
9
receptor binding assays.
AR binding was assessed by
3
a competitive [ H]DHT ligand binding assay using
recombinant human AR. Compounds that exhibited po-
tent binding affinity (IC50 < 200 nm) were tested in a hu-
man androgen receptor cellular functional assay
utilizing an MDA-MB453-MMTV-luci cell line for their
biological activities on human AR. A human progester-
one receptor (PR) binding assay was used to investigate
the selectivity of the analogs versus AR.
6
extent as to alter the conformation of the H12 helix,
and therefore achieve improved antagonistic activity
over compound 4e. To this end, the hydroxyl group of
1
compound 4e was alkylated as indicated in Scheme 2.
2
The structure–activity relationship (SAR) of the 4-aryl-
-trifluoromethylbenzonitriles was explored for various
substitutions on the B-ring (Table 1). Analogs were cho-
sen for preparation based upon properties required for
As shown in Table 2, there was a dramatic decrease in
binding potency as the size of the alkyl groups increased
(compounds 5a–e).
2
enhanced transdermal delivery (clogP = 2–4, M
1
<
Substitution of the 2-hydroxy group of compound 4e
was found to be detrimental to androgen receptor bind-
W
1
4
00).
In general, the receptor binding potency was dictated by
the substitution pattern of the B-ring (2- > 4- ꢀ 3-).
Additionally, the most potent 2-substituted analog in
the functional assay was the hydroxyl compound 4e.
However, 2,6-di-hydroxy substitution (4m) resulted in
loss of binding to the receptor. Multiple substitution
of the B-ring also resulted in a decrease in cellular
potency (4l).
CF3
CF3
NC
NC
R2
a
OH
O
4
e
5a-e
2
Scheme 2. Reagent and condition: (a) R –Br, K CO , acetone, rt, 72–
2
3
77%.
a
Table 1. Human androgen and progesterone receptor binding activities and calculated logP (BioByte) values for biphenyl analogs
CF3
NC
R1
1
R
Compound
clogP
ARB IC50 (nM)
ARCELL IC50 (nM)
PRB IC50 (nM)
4
4
4
4
4
4
4
4
4
4
4
4
4
a
b
c
d
e
f
2-Cl
4.81
3.79
4.35
4.35
3.55
3.85
3.85
4.55
4.85
4.85
5.23
3.27
2.64
9.8
37.0
955
nt
nt
2-OMe
3-OMe
4-OMe
2-OH
3-OH
4-OH
2-Me
20.1
140
0.8
3250
9460
5520
4450
nt
56.7
39.9
229
0.02
nt
g
h
i
61.2
146
11.9
>1000
nt
>10,000
nt
3-Me
4-Me
1900
200
nt
j
843
>1.0
399
nt
nt
k
l
2-CF
2,6-(OMe)
2,6-(OH)
3
10.9
14.7
989
nt
2
nt
m
2
nt
a
Values (IC50) are given as an average of 2–12 experiments; nt, not tested; ARB, human androgen receptor binding assay; ARCELL, human
androgen receptor cellular functional assay; PRB, human progesterone receptor binding assay.

J. A. Van Camp et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5529–5532
5531
a
Table 2. Human androgen receptor binding activity and calculated logP (BioByte) values for ortho-substituted biphenyl analogs
CF3
NC
R2
O
2
R
Compound
clogP
ARB, IC50 (nM)
ARCELL, IC50 (nM)
5a
5b
5c
5d
5e
CH
CH
2
CO
CO
2
Me
H
3.41
3.08
3.68
2.91
3.54
1310
>10,000
2230
nt
nt
nt
nt
nt
2
2
(CH
(CH
(CH
2
2
2
)
2
)
2
)
2
OMe
OH
O(CH
1540
2
)
2
OMe
8140
a
Values (IC50) are given as an average of 2–12 experiments; nt, not tested; ARB, human androgen receptor binding assay; ARCELL, human
androgen receptor cellular functional assay.
ing. Thus, the most biologically potent analog (4e) was
studied for its effect in vivo. It is known that there is a
direct correlation between wax ester reduction and a
52.9 mmol) in cHCl/H
slowly a NaNO2 (4.4 g, 63.4 mmol) solution in water
15 mL). After stirring for 20 min at 0 ꢁC, a solution of KI
(17.6g, 106.0mmol) in water (25mL) was added slowly.
After stirring for 4 h, the reaction mixture was extracted
with dichloromethane (4 · 100 mL). The combined organic
extracts were washed with 5% aq NaOH (300 mL), 5%
2
O (1:1, 100 mL) at 0 ꢁC was added
(
1
3,14
reduction in total sebum production.
As it measures
the reduction in wax esters, the Golden Syrian hamster
ear model is widely used as an in vivo model for testing
1
5
drug effects on sebaceous glands. Compound 4e was
evaluated in this in vivo model and exhibited excellent
activity. Tested topically, compound 4e exhibited an
3
aq NaHCO (300 mL), and brine (300 mL). The combined
organics were dried (MgSO ), filtered, and concentrated in
4
vacuo. The crude solid was purified by flash chromatog-
raphy (8:1 hexanes/EtOAc) to afford the intermediate aryl
iodide 3 as a white crystalline solid (12.6 g, 80% yield). A
mixture of the aryl iodide 3 (5.0 g, 16.8 mmol), K CO
3
8
7
6% reduction in wax esters versus vehicle (3% dose in
0/30 ethanol/propylene glycol). To compare, the posi-
tive control (RU-58841) reduced wax esters 95% as a
% formulation.
2
(
7.0 g, 50.5 mmol), and 3-methoxybenzeneboronic acid
3.0 g, 26.8 mmol) was suspended in DME/H O (80 mL/
mL) at rt. The mixture was degassed with nitrogen and
Pd(Ph P) (1.0 g, 1.1 mmol) added. The reaction mixture
1
(
2
8
In summary, the 2-hydroxy compound 4e was shown to
be the most potent compound in the functional assay
from the 4-aryl-2-trifluoromethyl-benzonitrile series.
Although in vitro potency was maintained for the 2-hy-
droxy and 2-methoxy analogs (4e and 4b), increased ste-
ric bulk at this position led to a loss of receptor binding.
Compound 4e exhibited good efficacy in the in vivo
model (86% reduction of wax esters); however, further
testing of this compound revealed a potential for
phototoxicity.
3
4
was heated at 80 ꢁC over 18 h. After this time, the mixture
was diluted with EtOAc/H O (1:1, 100 mL) and filtered
2
through a pad of celite. The filtrate was collected and the
phases separated. The organic phase was dried (MgSO4),
filtered, and concentrated in vacuo. Purification by flash
chromatography (10:1 hexanes/EtOAc) afforded the title
compound as a colorless solid (2.9 g, 62%); TLC: R = 0.25
f
ꢁ
+
4:1 hexanes/EtOAc); MS (APCI )m/z 277.1 (M );
1
(
H
6
NMR (400 MHz, DMSO-d )d 8.18–8.24 (m, 3H), 7.36–
7
C, 64.97; H, 3.46; N, 4.92%. C H F NO requires: C,
.46 (m, 3H), 7.02–7.11 (m, 1H), 3.83 (s, 3H). CHN found:
1
5
10 3
6
4.98; H, 3.64; N, 5.05%.
References and notes
9. The androgen receptor binding assay was run essentially
as described in Liao, S.; White, D.; Schilling, K.; Chang,
C. J. Steroid Biochem. 1984, 1, 11, The human AR cDNA
cloned in baculovirus was expressed in Sf9 cells. Cell
lysates from transfected Sf9 cells were isolated and used as
the source of human AR in the radio-ligand binding assay.
Different concentrations of test compounds (10,000, 1000,
200, 40, 8, 1.6, and 0.16 nM) were incubated in the
presence of human AR extract, hydroxylapatite, and 1 nM
1
2
. Thiboutot, D. Arch. Fam. Med. 2000, 9, 179.
. Chen, W.; Thiboutot, D.; Zouboulis, C. C. J. Invest.
Dermatol. 2002, 119, 992.
. Thiboutot, D. J. Am. Acad. Dermatol. 2002, 47, 109.
. Bingham, K. D.; Low, M.; Wyatt, E. H. Lancet 1979, 314,
304.
. Gruber, D. M.; Sator, M. O.; Joura, E. A.; Kokoschka, E.
M.; Heinza, G.; Huber, J. C. Arch. Dermatol. 1998, 134,
59.
. Gao, W.; Bohl, C. E.; Dalton, J. T. Chem. Rev. 2005, 105,
352.
3
4
5
6
3
H-DHT for one hour at 4 ꢁC with gentle rocking. After
4
incubation, plates were placed on a filter apparatus and
the reaction mixture was removed under 15 psi vacuum
pressure, then washed three times with cold buffer. The
plates were then dried at room temperature overnight. The
next day, scintillation fluid was added to each well and the
plates were counted using a Microbeta Trillux. To
determine non-specific binding, 1000-fold concentration
of cold DHT was added to each well. Triplicate wells were
tested for each concentration. The experimental results
3
7
8
. Moore, D. E. Drug Safety 2002, 25, 345.
. The general experimental procedure for preparing 4-aryl-
2
-trifluoromethylbenzonitriles is described below with the
0
synthesis of analog 4c as an example. 3 -Methoxy-3-
trifluoromethyl-biphenyl-4-carbonitrile (4c): To a stirred
solution of 4-amino-2-(trifluoromethyl)benzonitrile (9.8 g,

5532
J. A. Van Camp et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5529–5532
were expressed as the concentration of compound at which
0% of maximum binding was inhibited (IC50). Com-
pound RU-58841 was run as a standard revealing an inter-
run variability within 2-fold.
0. Chang, C.; Wang, C.; DeLuca, H. F.; Ross, T. F.; Shih, C.
filtrate concentrated in vacuo. The crude solid was purified
by flash chromatography (10:1 to 6:1 hexanes/EtOAc
gradient elution) to afford the title compound as a white
5
f
solid (230 mg, 72% yield); TLC: R = 0.43 (1:1 hexanes/
+
1
1
EtOAc); MS (ESI ) = 336.1; H NMR (400 MHz,
DMSO-d )d 8.19–8.24 (m, 2H), 8.06 (dd, J = 8.4, 1.6 Hz,
1H), 7.46–7.50 (m, 1H), 7.42 (ddd, J = 8.4, 7.4,
C.-Y. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 5946.
1. Naik, A.; Kalia, Y. N.; Guy, R. H. PSTT 2000, 3, 318.
6
1
1
0
2. The general experimental procedure for preparing 2 -O-
alkylated-4-aryl-2-trifluoromethylbenzo-nitriles is described
1.6 Hz, 1H), 4.89 (s,2H), 3.67 (s, 3H). CHN found: C,
12 3 3
61.17; H, 3.54; N, 4.12%. C17H F NO requires: C, 60.90;
below with the synthesis of analog 5a as an example.
0
H, 3.61; N, 4.18%.
13. Thiboutot, D. JID 2004, 123, 1.
14. Miller, C. G.; Wojnarowska, F. T.; Dowd, P. M.; Ashton,
R. E.; O’brien, T. J.; Giffiths, W. A. D.; Jacobs, H. S. Brit.
J. Dermatol. 1986, 114, 705.
0
Methyl{[4 -cyano-3 -(trifluoromethyl)biphenyl-2-yl]oxy}ace-
tate (5a): Methyl bromoacetate (0.09 mL, 145 mg,
0.95 mmol) was added dropwise to a stirred solution of
compound 4e (250 mg, 0.95 mmol) and K CO (132 mg,
2
3
0.96 mmol) in acetone (5 mL) at rt. After stirring for 5 h,
the mixture was filtered through a pad of celite and the
15. Luderschmidt, C.; Plewig, G. Arch. Derm. Res. 1977, 258,
185.