Design and synthesis of 4-[3,5-dioxo-11-oxa-4,9-diazatricyclo[5.3.1.0 2,6]undec-4-yl]-2-trifluoromethyl-benzonitriles as androgen receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.06.034

A novel series of 4-[3,5-dioxo-11-oxa-4,9-diazatricyclo[5.3.1.0 ^2,6]undec-4-yl]-2-trifluoromethyl-benzonitriles has been synthesized. The ability of these compounds to act as antagonists of the androgen receptor was investigated and several wer

Full text of DOI:10.1016/j.bmcl.2010.06.034

Bioorganic & Medicinal Chemistry Letters 20 (2010) 4491–4495
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of 4-[3,5-dioxo-11-oxa-4,9-diazatricyclo[5.3.1.02,6]undec-
4-yl]-2-trifluoromethyl-benzonitriles as androgen receptor antagonists
a
a
b
b
b
Hai-Yun Xiao a, , Aaron Balog , Ricardo M. Attar , David Fairfax , Linda B. Fleming , Christian L. Holst ,
Gregory S. Martin b, Lana M. Rossiter b, Jing Chen a, Mary-Ellen Cvjic a, Janet Dell-John a, Jieping Geng a,
Marco M. Gottardis a, Wen-Ching Han a, Andrew Nation a, Mary Obermeier a, Cheryl A. Rizzo a,
Liang Schweizer a, Thomas Spires Jr. a, Weifang Shan a, Ashvinikumar Gavai a, Mark E. Salvati a, Gregory Vite a
*
a Bristol-Myers Squibb Company, Research and Development, Princeton, NJ 08543-4000, USA
b Albany Molecular Research Institute, Albany, NY 12203, USA
a r t i c l e i n f o
a b s t r a c t
A novel series of 4-[3,5-dioxo-11-oxa-4,9-diazatricyclo[5.3.1.02,6]undec-4-yl]-2-trifluoromethyl-benzo-
nitriles has been synthesized. The ability of these compounds to act as antagonists of the androgen recep-
tor was investigated and several were found to have potent activity in vitro and in vivo.
Ó 2010 Elsevier Ltd. All rights reserved.
Article history:
Received 22 April 2010
Revised 3 June 2010
Accepted 7 June 2010
Available online 10 June 2010
Keywords:
Androgen receptor
Prostate cancer
Antagonist
Diazatricycloundecane
Carcinoma of the prostate (CaP) is one of the leading causes of
cancer related death in men in the United States.1 The androgen
receptor (AR) is a ligand binding transcription factor in the nuclear
hormone receptor super family and is a key molecular target in the
etiology and progression of prostate cancer. Binding of androgens,
such as dihydro-testosterone (DHT), to the AR provides the mito-
genic signal for growth of prostate cancer cells. Androgen ablation
via surgical castration or by chemical castration with a luteinizing
hormone releasing hormone agonist, in combination with an anti-
androgen,2 is currently the treatment of choice for advanced CaP.
Although this therapy initially shows an 80–90% response rate,3
approximately 50% of patients progress to fatal androgen indepen-
dent CaP (AI-CaP) after about 18 months of treatment.4 Recent ad-
vances in the field have shown that reactivation of the AR signaling
pathway is the root cause for the development of AI-CaP.5 The
identification of the role of the AR in AI-CaP suggests that new
agents which act at the level of the AR may be effective in the treat-
ment of this disease. For this reason, we are interested in identify-
ing novel small molecule antagonists of the AR that are more
effective than the current AR antagonists at targeting the AR in
AI-CaP.
Our initial screening and optimization efforts led to a series of
hydantoin6 and cyclic imide7-based AR antagonists, exemplified
by compounds 1 and 2 (Fig. 1), that bound tightly to the AR and
functioned as antagonists to the wild-type (WT) and the mutant
(MT) isoforms of the AR in vitro.
Figure 2 shows the structures of flutamide (3), nilutamide (4),
and bicalutamide (5)—the nonsteroidal AR antagonists that have
been used in the treatment of prostate cancer.
Of the compounds outlined in Figures 1 and 2, the 3,4-disubsti-
tuted aniline amide structure is very intriguing. It appears that this
motif of the molecules is essential for high affinity binding to the
AR. We reasoned that combining the 3,4-disubstituted aniline moi-
ety with a symmetric bicyclic system similar to our lead com-
pounds (1) and (2), as shown for the generic structure (6) in
Me
O
O
I
O
CF3
CN
O
F
N
O
N
CN
N
N
HO
O
2
O
Me
1
* Corresponding author. Tel.: +1 609 252 6469; fax: +1 609 252 6804.
E-mail address: haiyun.xiao@bms.com (H.-Y. Xiao).
Figure 1. Our lead antiandrogens.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.06.034

4492
H.-Y. Xiao et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4491–4495
O
sodium cyanoborohydride afforded the desired compounds. N-Aryl
Me
HN
O
H
N
analogs of 6a were successfully synthesized employing aryl amines
in the presence of acetic acid. When aliphatic amines were used,
the reaction did not proceed due to the protonation of the amine
by acetic acid. To synthesize aliphatic amine analogs, ozonolysis
was carried out in anhydrous dichloromethane and the ozonide
was reduced with dimethylsulfide to give the dialdehyde interme-
diate that was then subjected to reductive amination conditions to
provide the desired N-alkyl analogs. The parent amine (6c) was
synthesized from the hydroxylamine intermediate (6a, R = OH)
via reduction with titanium trichloride. A number of analogs were
synthesized from (6c) employing acylation and reductive amina-
tion protocols. The ozonide was also oxidized9 to give the diacid
that was converted into imides (6b) via the anhydride (11).
To evaluate the activity of this new series, we investigated the
ability of compounds to bind to (Ki) and functionally antagonize
(IC50) the WT AR found in the MDA-453 cell line as well as antag-
onize the MT AR (T877A) found in the LNCaP cell line.11
Modification of the tricyclic core has dramatic effect on the
activities. The data in Table 1 clearly shows that although the bind-
ing potency is maintained with the imides (12, 13), the functional
potency in the MDA-453 assay is significantly reduced compared
to bicalutamide (5). However, the tertiary amine analog (15) was
significantly more potent than bicalutamide across all the
in vitro assays examined and was the subject of further SAR studies
as outlined in Table 2.
Me
Me
CF3
N
CF3
NO2
Me
O
NO2
Flutamide
(3)
Nilutamide
(4)
Me
O
OH
O
O
H
N
CF3
CN
S
F
Bicalutamide
(5)
Figure 2. Clinically used antiandrogens.
O
CF3
CN
O
RN
N
O
6
Figure 3. Proposed diazatricyclic antiandrogens.
Figure 3, would allow us to rapidly explore the SAR in a novel series
of compounds.
The parent amine (6c) and N-hydroxyl analog (16) had similar
binding and wild-type functional activity compared to bicaluta-
mide, although both compounds are significantly less potent in
the LNCaP functional assay. However, the N-methoxy analog (17)
had significantly improved activity versus bicalutamide in all three
of the in vitro assays. The SAR trend in Table 2 appears to suggest
that lipophilic substitution in general improved potency across all
three in vitro assays (23 vs 24; 30 vs 31; 32 vs 34; 36 vs 37). Aryl
substitution improved potency in amides (22 vs 20) and ureas (28
vs 27) series. Carbamate and sulfonamide analogs were better than
The synthetic pathway8 utilized in the preparation of analogs of
compound 6 is outlined in Scheme 1. Furan-2,5-dione (7) and 4-
amino-2-trifluoroenzonitrile (8) were heated together in the pres-
ence of acetic acid to yield 4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-
yl)-2-(trifluoromethyl)benzonitrile (9). Diels–Alder reaction of 9
with 2,5-dimethylfuran gave exclusively the exo adduct (10) that
precipitated out from the reaction mixture. Ozonolysis of the
Diels–Alder adduct (10), reduction and reductive amination with
Me
O
O
O
CF3
CN
CF3
CN
CF3
CN
a
b
O
O
N
N
+
H2N
O
O
O
Me
7
8
9
10
f, g, h
c or d
Me
Me
Me
O
N
O
O
O
O
O
O
O
N
O
CF3
CN
CF3
CN
CF3
CN
i, j
R
R
N
O
N
O
N
O
O
O
Me
Me
Me
Me
6b
11
6a
e
(when R = OH)
Me
O
O
CF3
CN
CF3
CN
derivitization
R1
N
HN
N
N
O
O
O
O
Me
Me
6c
6d
Scheme 1. Synthesis of compound 6. Reagents and conditions: (a) AcOH, 120 °C; (b) 2,5-dimethylfuran, PhMe, 60 °C, 3 h, 90%; (c) O3, MeOH, CH2Cl2, À78 °C, then NaBH3CN,
ArNH2, HOAc, 20–50%; (d) O3, CH2Cl2, À78 °C, then SMe2, then RNH2, NaBH3CN, TEA, DMF, 20–40%; (e) TiCl3, MeOH, H2O, 70–90%; (f) O3, CH2Cl2, À78 °C; (g) Jones’ reagent,
acetone; (h) Ac2O; (i) RNH2, THF; (j) Ac2O, KOAc, HOAc; 20–50% yield for steps f–j.

H.-Y. Xiao et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4491–4495
4493
Table 1
SAR around the tricyclic core
Me
Me
Me
Me
O
N
O
O
O
CF3
CN
O
O
O
CF3
CN
CF3
CN
CF3
CN
-O
N+
Me
Me
Me
Me
O
N
Me
N
O
N
N
N
O
N
O
O
O
O
O
O
Me
13
Me
12
Me
14
Me
15
b
Compd No.
Kia, nM
MDA-453 IC50b, nM (%Inh)c
LNCaP IC50 nM, (%Inh)c
5 Bicalutamide
64
20
50
3600
14
173
2144
873
3900 (58)
32 (99)
400
>5000
3329
>5000
125 (99)
12
13
14
15
a
Binding determined through direct displacement of ligand with [3H]-DHT in the MDA-453 cell line (Ki).10
Functional antagonist activity determined through a transiently transfected reporter system utilizing a secreted alkaline phosphatase (SEAP) reporter construct and a PSA
b
AR promoter domain.10
c
Percent inhibition of DHT(1 nM)-induced increase of cell growth at 5 lM.
urea and amides in terms of binding activity. However, in terms of
the LNCaP functional assays, the differences were not significant.
(23, 31 vs 19, 26). The tertiary aliphatic amines ( 15, 18) were
among the most active compounds in the series and superior to
bicalutamide (5). They were potent not only in binding and WT
functional assays but also in MT functional assay. Unfortunately,
their in vitro microsomal instabilities across species (data not
shown) prevented the compounds from further development. The
substituted aromatic amines (37, 38), however, showed excellent
activities in the MDA-453 and LNCaP functional assays and good
in vitro microsomal stability.12
Compounds that had good in vitro functional activity and micro-
somal stability proceeded into a PK/PD study. In the model, imma-
ture rats were treated with compound and testosterone
Table 2
SAR of N-substituted analogs
Me
O
CF3
CN
R
N
O
N
O
Me
Compd No.
R
Kia, nM
MDA-453 IC50b, nM (%Inh)c
LNCaP IC50b, nM (%Inh)c
In vivo IRPWd
41
5
Bicalutamide
H
OH
OMe
n-Propyl
O
64
21
37
3.7
14
173
400
6c
16
17
18
128 (98)
182 (97)
24 (99)
14 (99)
>5000 (20)
1280 (79)
179 (72)
25 (98)
19
84
129 (95)
549 (91)
Me
O
20
39
215 (98)
1610 (72)
Me
Me
O
21
36
69 (99)
404 (95)
O
22
23
24
23
54 (99)
65 (99)
9.3 (99)
177 (98)
590 (93)
54 (99)
94
Ph
O
Me
O
O
5.7
6.6
Me
O
O
Me
Me
25
26
11
12 (99)
69 (99)
50
O
O
Me
176
313 (97)
1456 (76)
HN
(continued on next page)

4494
H.-Y. Xiao et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4491–4495
Table 2 (continued)
Compd No.
R
Kia, nM
MDA-453 IC50b, nM (%Inh)c
876 (88)
LNCaP IC50b, nM (%Inh)c
4600 (52)
In vivo IRPWd
O
Me
Me
27
98
HN
O
28
29
40
29
16 (99)
39 (99)
123 (98)
104 (98)
64
52
HN Ph
O
F
HN
O
O
Me
S
30
31
32
56
16
12
205 (99)
85 (99)
99 (99)
3511 (61)
600 (90)
979 (91)
O
O
S
Me
O
O
S
Ph
O
O
S
33
34
32
93
54 (100)
170 (98)
685 (95)
61
F
O
O
S
1241 (80)
N
CONH2
35
36
37
9.0
22
7.4 (98)
37 (99)
7.4 (96)
127 (99)
337 (94)
150 (98)
S(O)2NH2
S(O)2Me
9.4
36
37
Cl
N
38
39
3.5
6.1
11 (98)
31 (99)
104 (100)
377 (97)
N
N
N
a,b,c
See Table 1.
d
Immature rat prostate weight assay: QD Â 3 days PO dosing at 10 mpk of drug with testosterone propionate (1 mpk). Weight of seminal vesicles (SV) and full body weight
(FB) were measured. The number is reported as percentage of SV/FB relative to those of the control group where animals were treated only with testosterone propionate
(1 mpk).
propionate. The stronger antagonism of the compound will give a
smaller ratio of seminal vesicles weight (SV) to full body weight
(FB) (i.e., SV/FB) than those of the control group where animals were
treated with testosterone propionate only. It is clear, from Table 2,
that the in vivo activity roughly followed the functional activity in
the MDA-453 cell line. This is reasonable since the growth of the
prostate is dominated by WT androgen receptor activation. The
IRPW data showed that some of the compounds (37, 38) were very
efficacious and the effect was dose-dependent (e.g., compound 37
had SV/FB of 36% at 10 mpk and 28% at 30 mpk; compound 38 had
SV/FB of 59% at 1 mpk and 37% at 10 mpk).
In summary, we have developed an efficient synthesis of a novel
series of androgen receptor antagonists. These diazatricyclic com-
pounds were proposed based on the common structure of the
existing AR antagonists. Most of the new compounds demon-
strated potent binding to the WT AR and the full functional antag-
onism in the cell lines expressing the WT and especially the MT AR.
Modification of the cyclic core and optimizing the substitution on
the nitrogen atom led to the identification of compounds (37, 38)
that were very efficacious in our PK/PD model.
Acknowledgment
The authors thank Dr. Murali Dhar for his helpful suggestions.
References and notes
1. Jemal, A.; Thomas, H.; Murray, T.; Thun, M. CA Cancer J. Clin. 2002, 52,
23.
2. (a) Gao, K. L. Drugs 1991, 43, 254; (b) Motta, M.; Serio, M. Hormonal Therapy of
Prostatic Diseases: Basic and Clinical Aspects; Medicom Europe: Amsterdam,
1998; (c) Crawford, E. D.; DeAntonio, E. P.; Labrie, F.; Schroder, F. H.; Geller, J.
J. Clin. Endocrinol. Metab. 1995, 80, 1062.
3. (a) Blackledge, G. Cancer 1993, 3830; (b) Maucher, A.; von Angerer, E. J. Cancer
Res. Clin. Oncol. 1993, 119, 669; (c) Sufrin, G.; Coffey, D. S. Invest. Urol. 1976, 13,
429; (d) Narayana, A. S.; Loening, S. A.; Culp, D. A. Br. J. Urol. 1981, 53, 152.
4. (a) Huggins, C.; Stevens, R.; Hodges, C. V. Arch. Surg. 1947, 43, 209; (b) Westin,
P.; Stattin, P.; Damber, J.-E.; Bergh, A. Am. J. Pathol. 1995, 146, 1368.

H.-Y. Xiao et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4491–4495
4495
5. (a) Scher, H.; Sawyers, C. J. Clin. Oncol. 2005, 23, 8253; (b) Chen, C. D.; Welsbie,
D. S.; Tran, C.; Baek, S.-H.; Chen, R.; Vessella, R.; Rosenfeld, M. G.; Sawyers, C. L.
Nat. Med. 2004, 10, 33; (c) Richter, E.; Srivastava, S.; Dobi, A. Prostate Cancer
Prostatic Dis. 2007, 10, 114; (d) Burd, C. J.; Morey, L. M.; Knudsen, K. E. Endocr.
Relat. Cancer 2006, 13, 979; (e) Hendriksen, P. J. M.; Dits, N. F. J.; Kokame, K.;
Veldhoven, A.; van Weerden, W. M.; Bangma, C. H.; Trapman, J.; Jenster, G.
Cancer Res. 2006, 66, 5012.
6. (a) Balog, A.; Salvati, M.; Shan, W.; Mathur, A.; Leith, L.; Wei, D.; Attar, R.; Geng,
J.; Rizzo, C.; Wang, C.; Krystek, S.; Tokarski, J.; Hunt, J.; Weinmann, R. Bioorg.
Med. Chem. Lett. 2004, 14, 6107; (b) Salvati, M.; Balog, A.; Shan, W.; Wei, D.;
Pickering, D.; Attar, R.; Geng, J.; Rizzo, C.; Gottardis, M.; Weinmann, R.; Krystek,
S.; Sack, J.; An, Y.; Kish, K. Design, Structural Analysis and Biological Profile of a
Novel Series of AR Antagonists, Poster Presentation, American Association of
Cancer Research, 95th Annual Meeting, 2004; (c) Salvati, M.; Balog, A.; Shan,
W.; Geise, S. U.S. Patent 885,798, 2003.
Chandrasena, G. Bioorg. Med. Chem. Lett. 2008, 18, 1910; (c) Salvati, M. E.; Balog,
A.; Pickering, D. A.; Giese, S.; Fura, A.; Li, W.; Patel, R.; Hanson, R. L.; Mitt, T.;
Roberge, J.; Corte, J. R.; Spergel, S. H.; Rampulla, R. A.; Misra, R.; Xiao, H.-Y.
World Patent Application WO 200306224.
8. Balog, A.; Fairfax, D.; Martin, G.; Salvati, M.; Xiao, H.-Y. U.S. 2007088029.
9. Narula, A. S.; Dev, S. Tetrahedron Lett. 1969, 22, 1733.
10. New analogs were assayed for effect on AR function in MDA-453, a cell line
expressing WT AR, and LNCaP, a cell line expressing MT AR. Transactivation
assays were done in whole cells and IC50’s were calculated. Both cell lines were
obtained from American Type Culture Collection (Rockville, MD). Binding
constants (Ki) were determined in whole cells measuring displacement by
tritiated DHT in the MDA-453 cell line. See Ref. 11 below for experimental
details on assays.
11. Salvati, M.; Gottardis, M.; Krystek, Jr., S.; Attar, R.; Sack, J. U.S. 6,960,474; WO
2002000617.
7. (a) Salvati, M.; Balog, A.; Wei, D.; Pickering, D.; Attar, R.; Geng, J.; Rizzo, C.;
Hunt, J.; Weinmann, R.; Martinez, R. Bioorg. Med. Chem. Lett. 2005, 15, 289; (b)
Salvati, M.; Balog, A.; Shan, W.; Rampulla, R.; Giese, S.; Mitt, T.; Furch, J. A.; Vite,
G. D.; Attar, R. M.; Jure-Kunkel, M.; Geng, J.; Rizzo, C. A.; Gottardis, M. M.;
Krystek, S. R.; Gougoutas, J.; Galella, M. A.; Obermeier, M.; Aberra Fura, A.;
12. For example, the in vitro metabolic stability of compound 37 was evaluated at
3 lM in the presence of human, rat, and mouse liver microsomes. After a
10 min incubation, the % compound remaining for compound 37 in humans,
rats, and mice was 76%, 100%, and 92%, respectively.