2-Benzenesulfonyl-8a-benzyl-hexahydro-2H-isoquinolin-6-ones as selective glucocorticoid receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2007.07.055

The 2-azadecalin ring system was evaluated as a scaffold for the preparation of glucocorticoid receptor (GR) antagonists. High affinity, selective GR antagonists were discovered based on a hypothetical binding mode related to the steroidal GR antagonist R

Full text of DOI:10.1016/j.bmcl.2007.07.055

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5704–5708
2-Benzenesulfonyl-8a-benzyl-hexahydro-2H-isoquinolin-6-ones
as selective glucocorticoid receptor antagonists
Robin D. Clark,a,* Nicholas C. Ray,b Paul Blaney,b Peter H. Crackett,b
Christopher Hurley,b Karen Williams,b Hazel J. Dyke,b David E. Clark,b
Peter M. Lockey,b Rene Devos,b Melanie Wong,b
Anne Whiteb and Joseph K. Belanoffa
aCorcept Therapeutics, 149 Commonwealth Avenue, Menlo Park, CA 94025, USA
bArgenta Discovery, 8/9 Spire Green Centre, Flex Meadow, Harlow, Essex CM19 5TR, United Kingdom
Received 10 April 2007; revised 12 July 2007; accepted 16 July 2007
Available online 19 August 2007
Abstract—The 2-azadecalin ring system was evaluated as a scaffold for the preparation of glucocorticoid receptor (GR) antagonists.
High affinity, selective GR antagonists were discovered based on a hypothetical binding mode related to the steroidal GR antagonist
RU-43044. 2-Benzenesulfonyl substituted 8a-benzyl-hexahydro-2H-isoquinolin-6-ones exemplified by (R)-37 had low nanomolar
affinity for GR with moderate functional activity (200 nM) in a reporter gene assay. These compounds were devoid of affinity
for other steroidal receptors (ER, AR, MR, and PR). Analogues based on an alternative putative binding mode (CP-like) were
found to be inactive.
Ó 2007 Elsevier Ltd. All rights reserved.
Mifepristone (RU-486)1–3 is an antagonist at progester-
one (PR) and glucocorticoid (GR) receptors and has re-
cently been shown to be an effective treatment for
psychotic major depression (PMD).4–6 The etiology of
this condition has been associated with abnormally high
levels of cortisol 7,8; hence, the efficacy of mifepristone in
the treatment of PMD is ascribed to GR antagonism.6
Our goal was to discover a ‘second generation’ drug
which would be a potent and selective GR antagonist.
There has also been considerable interest in the develop-
ment of selective GR antagonists for other indications,9
including treatment of diabetes,10,11 obesity,12,13 endog-
enous depression,14 Alzheimer’s disease,15 neuropathic
pain,16 and Cushing’s disease.17
line) ring system as this presented an attractive scaffold
to probe GR receptor binding. With appropriate relative
Me
N
Me
Me
OH
OH
O
O
RU-43044
Mifepristone
OH
CP-409069 X = OH
CP-472555 X =
N
H
N
H
}
In addition to mifepristone, the related steroid
RU-43044,2,3 which is selective for GR over PR, and
the recently disclosed potent and selective antagonists
CP-40906918 and CP-47255519 provided structural leads
for further design. We decided to explore the benzyl-
substituted 2-azadecalin (octa- and decahydro-isoquino-
X
O
OH
X
N
H
N
O
X
Keywords: Glucocorticoid receptor (GR) antagonists; Azadecalin.
*
"CP like"
"RU like"
Corresponding author. Tel.: +1 808 332 5449; fax: +1 808 332 0389;
e-mail addresses: rclark@corcept.com; robin.clark5@hawaiiantel.net
Figure 1. Potential GR-binding modes of substituted azadecalins.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.055

R. D. Clark et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5704–5708
5705
Table 2. GR-binding affinity for compounds 14–23a
and absolute stereochemistry, the benzyl-2-azadecalin
system can be represented in either a ‘CP-like’ binding
mode or an ‘RU-like’ binding mode with nitrogen sub-
stituents X providing auxiliary binding groups (Fig. 1).
X
N
Racemic hexahydro-2,8a-dibenzylisoquinolinone 2 was
prepared by Robinson annelation of piperidone 120 as de-
scribed for the synthesis of the corresponding 8a-methyl
analogue21 (Scheme 1). Lithium/ammonia reduction
afforded the trans-octahydroisoquinolinone 3 which was
converted to various analogues 4 by deprotection and
standard N-substitution reactions. Treatment of com-
pounds 4 with butynyl magnesium bromide furnished
racemic target compounds 5–13 (Table 1) which were ob-
tained as a major diastereomer with the indicated relative
stereochemistry. In some cases the minor diastereomer
was also isolated by column chromatography.22
O
Compound
X
GR-binding Ki (nM)b
14
15
16
17
18
19
20
21
22
23
C(O)-phenyl
C(O)-n-butyl
C(O)NH-phenyl
S(O2)-phenyl
S(O2)-n-butyl
S(O2)-NH-phenyl
4-MeO-phenyl
Benzyl
>10,000
>10,000
>10,000
225
>10,000
1200
>5000
3700
2250
4-MeO-benzyl
CH2-(4)-pyridyl
>10,000
a All compounds are racemic.
b Values are means of two experiments.
Compounds 14–50 (Tables 2–4), including those substi-
tuted on the 8a-benzyl group (Y = OMe, F, Br, NO2),
were readily available by N-debenzylation of enone 2
Table 3. GR-binding affinity for substituted sulfonamides 24–45a
O
O
S
O
N
Bn
Bn
Y
a
b
N
N
O
H
GR-binding Ki (nM)b
O
O
N
Compound
Y
Bn
1
2
3
17
24
H
2-CN
225
1000
4450
3000
1250
800
OH
25
26
2-CF3
3-NO2
4-NO2
4-F
O
c,d
e
27
28
H
H
N
N
X
X
29
30
3-OMe
4-OMe
3,4-(OMe)2
2,5-(OMe)2
4-OPh
3-Cl
196
216
(+)-5-13
4
31
32
1300
>10,000
163
Scheme 1. Reagents and conditions: (a) methylvinyl ketone, NaOMe,
MeOH, rt; (b) Li, NH3; (c) H2, Pd(OH)2/C, AcOH (d) 5: benzyl
bromide, NaH, THF; 8–13: acyl or sulfonyl chloride, NEt3, CH2Cl2;
(e) butynyl magnesium bromide, THF.
33
34
432
193
35
36
4-Cl
4-Me
4-tert-Butyl
99
14
37
Table 1. Compounds 5–13a
(R)-37
(S)-37
38
4
>10,000
55
4-Phenyl
OH
39
40
2-NHSO2Me
3-NHSO2Me
4-NHSO2Me
4-SO2Me
1950
2100
950
41
42
H
N
X
137
750
43
44
4-NHCOMe
4-NMe2
4-(Morpholin-4-yl)
AcNH
113
64
45
}
}
}
S
a All compounds are racemic unless otherwise noted.
b Values are means of two experiments.
O
O
8
5
O
11
}
HO
}
AcNH
}
O
MeSO2NH
MeSO2NH
S
6
9
O
and subsequent derivatization (Scheme 2). Conversion
of the bromo (49) and nitro (50) compounds to ana-
logues 51–59 (Table 4) was accomplished using standard
transformations.23
O
O
12
Me
N
}
}
N
H
Me
S
}
O
10
O
S
O
7
O
O
13
The enantiomers of 37 (Table 3) and 47 (Table 4) were
separated by chiral column chromatography. An asym-
a All compounds are racemic.

5706
R. D. Clark et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5704–5708
Table 4. GR-binding affinity and functional activity for 37 and aryl
derivatives 46–59a
BOC
BOC
BOC
Y
N
a,b,c
N
N
+
O
O
O
O
O
O
S
60
61
62
N
OH
d
O
Compound
Y
H
GR-binding GR Functional
Ki (nM)b
Ki (nM)b
O
O
37
(R)-37
14
4
>1000
200
ntc
230
BOC
S
e,f
N
N
O
O
46
47
3-OMe
4-OMe
76
6
(
R)-37
(R)-63
(R)-47
48
4
200
>1000
200
Scheme 3. Reagents and conditions: (a) (R)-a-methylbenzylamine,
toluene, reflux; (b) methylvinyl ketone, toluene, 6 days; (c) AcOH,
H2O; (d) NaOMe, MeOH, 75 °C; (e) TFA; (f) 4-(t)-butylbenzenesulfo-
nyl chloride, NEt3, CH2Cl2.
4-F
4-Br
5
49
50
9
4-NO2
4-CN
45
65
99
15
13
56
13
2
nt
51
52
nt
4-NH2
4-NMe2
nt
53
54
>1000
>1000
>1000
323
Affinity for GR was determined by ligand binding mea-
suring displacement of [3H]dexamethasone from recom-
binant baculovirus derived human GR.18 Functional
activity at human GR was determined in SW1353/
MMTV-5 cells transfected with a plasmid encoding fire-
fly luciferase located behind a glucocorticoid response
element (GRE).18 GR antagonist activity was measured
as inhibition of dexamethasone induced luciferase
expression. Selected compounds were tested for GR
agonist activity by performing the assay in the absence
of dexamethasone. Selectivity over the estrogen (ERa),
androgen (AR), mineralocorticoid (MR), and progester-
one (PR) receptors was determined by ligand binding
assays.29
4-NHSO2Me
4-NHCOMe
4-NHSO2NMe2
4-(Pyrid-4-yl)
4-(Piperidin-1-yl)
55
56
57
58
180
9
>1000
400
1.2
59
Mifepristone
CP-409069
4-(Morpholin-4-yl) 10
0.4
0.6
4
a All compounds are racemic unless otherwise noted.
b Values are means of two experiments.
c Not tested.
Y
Y
Y
Compounds 5–13 (Table 1), which were based on the
hypothetical ‘CP-like’ GR-binding mode, were remark-
able in their lack of affinity for the receptor. None of
these compounds, as well as a number of related ana-
logues (not shown), inhibited 50% binding of radiola-
Bn
Bn
X
b,c
N
a
N
N
O
O
O
1
2
(+)-14-50
Scheme 2. Reagents and conditions: (a) methylvinyl ketone, NaOME,
MeOH, rt; (b) ClCO2CHClCH3, dichloroethane, reflux; MeOH, reflux;
(c) 14–19: isocyanate, or acyl or sulfonyl chloride, NEt3, CH2Cl2; 20:
4-MeO-phenylboronic acid, Cu(OAc)2, Et3N, CH2Cl2; 22–23: 4-MeO-
benzyl bromide or 4-picolyl bromide, NaH, THF.
beled dexamethasone at
a
concentration of 10
micromolar. The diastereomers of 5–13, obtained as
minor products and which are epimeric at the tertiary
alcohol center, were similarly inactive. The high affinity
of CP-472555 and related compounds indicates that the
GR tolerates fairly large groups in the ‘A-ring’ portion
of these molecules.19 The lack of affinity of the substi-
tuted 2-azadecalins 5–13 demonstrates that the N-sub-
stituents apparently do not allow binding to GR in a
similar manner.
metric synthesis of (R)-37 was subsequently developed
based on the chiral imine Michael addition protocol
used for the enantioselective synthesis of quaternary
centers,24 including enantiomers of the 8a-methyl ana-
logue corresponding to enone 221 (Scheme 3). Conver-
sion of BOC-protected 2-benzyl-4-piperidone 6025 to
the imine with (R)-(a)-methylbenzylamine followed by
reaction with methylvinyl ketone in toluene afforded a
separable mixture of Michael adduct 61 and Michael-al-
dol adduct 62 in a ratio of ca. 2:1. Base treatment of 62
gave enone (R)-63 with 96–98% ee, albeit in low (10–
20%) overall yield from 60.26 The stereochemical assign-
ment for (R)-63 is based on the absolute stereochemistry
observed in the directly analogous octalone24 and hexa-
hydroisoquinolinone21 examples. Removal of the BOC-
group with TFA and subsequent sulfonylation afforded
(R)-37 (96–98% ee by chiral HPLC).27,28
Results for initial compounds based on the ‘RU-like’
binding mode were similarly disappointing with the
exception of the benzenesulfonamide derivative 17
which had weak GR binding with a Ki of 225 nM (Table
2). The related alkylsulfonamide 18 was essentially de-
void of binding affinity. The binding observed for 17
prompted the synthesis of a series of substituted ben-
zenesulfonamides (Table 3). A trend toward increased
binding for large, and generally lipophilic, para-substit-
uents was observed with the racemic 4-(tert-butyl)-
benzenesulfonamide 37 having the highest binding
affinity (14 nM). Binding affinity was enantiospecific,

R. D. Clark et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5704–5708
5707
with (R)-37 having a Ki of 4 nM. The active enantiomer
is therefore in the same absolute stereochemical series as
the A–B ring system of RU-43044.
References and notes
1. Teutsch, G.; Ojasoo, T.; Raynaud, J. P. J. Steroid
Biochem. 1988, 31, 549.
2. Teutsch, G.; Gaillard-Moguilewsky, M.; Lemoine, G.;
Nique, F.; Philibert, D. Biochem. Soc. Trans. 1991, 19,
901.
3. Philibert, D.; Costerousse, G.; Gaillard-Moguilewsky, M.;
Nedelec, F.; Tournemine, C.; Teutsch, G. Front. Horm.
Res. 1991, 19, 1.
4. Belanoff, J. K.; Flores, B. H.; Kalezhan, M.; Sund, B.;
Schatzberg, A. F. J. Clin. Psychopharmacol. 2001, 21, 516.
5. Belanoff, J. K.; Rothschild, A. J.; Cassidy, F.; DeBattista,
C.; Baulieu, E. E.; Schold, C.; Schatzberg, A. F. Biol.
Psychiatry 2002, 52, 386.
6. Flores, B. H.; Kenna, H.; Keller, J.; Solvason, H. B.;
Schatzberg, A. F. Neurosychopharmacology 2006, 31,
628.
Substituted benzyl derivatives of lead compound 37
were then evaluated (Table 4). We concentrated on
para-substitution, partly on the basis of the higher
affinity of 4-methoxy analogue 47 (6 nM) compared to
3-methoxy compound 46 (76 nM), and because the
para-position is highly tolerant of substitution in mife-
pristone.30 GR-binding affinity was retained with a vari-
ety of para-substituents including the 4-pyridyl
derivative 57 with a Ki of 2 nM. Perhaps not surprisingly
this compound showed significant Cyp inhibition (2C9:
76%: 2C19: 61%: 2D6: 44% at 1 lM), an activity not
shown by other analogues including 37 and 47.
7. Belanoff, J. K.; Kalehzan, M.; Sund, B.; Fleming Ficek, S. K.;
Schatzberg, A. F. Am. J. Psychiatry 2001, 158, 1612.
8. Gold, P. W.; Drevets, W. C.; Charney, D. S. Biol.
Psychiatry 2002, 52, 381.
9. Mohler, M. L.; He, Y.; Wu, Z.; Hong, S.-S.; Miller, D. D.
Expert Opin. Ther. Patents 2007, 17, 59.
Representative high affinity ligands were evaluated for
GR functional antagonist activity in the SW1353/
MMTV-5 reporter gene assay. None of the compounds
demonstrated agonist activity and several of them were
found to be GR antagonists with moderate activity in
the 200 nM Ki range (Table 4). This level of GR antago-
nist activity was considerably lower than that observed
for the standards mifepristone and CP-409069 for which
GR binding was much closer to the GR antagonist
activity. A similar dissociation between GR binding and
functional activity has been reported for other series of
non-steroidal GR antagonists.31 A possible explanation
for this discrepancy in our series is lack of cellular penetra-
tion. To test this hypothesis, a whole cell GR-binding
assay was carried out in SW1353 cells with (R)-47.
This compound demonstrated almost 100-fold lower
binding in the whole cell assay (352 nM) compared to
the original isolated GR-binding assay (4 nM). Whole cell
binding of mifepristone (0.82 nM) and CP-409069
(5.5 nM) was consistent with their more potent GR func-
tional activity.
10. Gettys, T. W.; Watson, P. M.; Taylor, I. L.; Collins, S. Int.
J. Obes. 1997, 21, 865.
11. Link, J. T.; Sorensen, B.; Patel, J.; Grynfarb, M.; Goos-
Nilsson, A.; Wang, J.; Fung, S.; Wilcox, D.; Zinker, B.;
Nguyen, P.; Hickman, B.; Schmidt, J. M.; Swanson, S.;
Tian, Z.; Reisch, T. J.; Rotert, G.; Du, J.; Lane, B.; von
Geldern, T. W.; Jacobson, P. B. J. Med. Chem. 2005, 48,
5295.
12. Langley, S. C.; York, D. A. J. Physiol. 1990, 259, R539.
13. Beebe, K. L.; Block, T.; DeBattista, C.; Blasey, C.;
Belanoff, J. K. Behav. Brain Res. 2006, 171, 225.
14. Reus, V. I.; Wolkowitz, O. M. Expert Opin. Investig.
Drugs 2001, 10, 1789.
15. Dhikav, V.; Anand, K. S. Med. Hypotheses 2007, 68, 1088.
16. Takasaki, I.; Kurihara, T.; Saegusa, H.; Zong, S.; Tanabe,
T. Eur. J. Pharmacol. 2005, 524, 80.
17. Chu, J. W.; Matthias, D. F.; Belanoff, J.; Schatzberg, A.;
Hoffman, A. R.; Feldman, D. J. Clin. Endocrinol. Metab.
2001, 86, 3568.
Selectivity profiling indicated that none of the sulfon-
amide compounds in Table 4 displaced 50% binding at
ERa, AR, MR or PR at 10 lM. Consistent with its re-
ported lack of selectivity for GR over PR,1–3 mifepri-
stone had a Ki of 1.3 nM at PR. Thus, although
compounds from the current series are relatively weak
GR functional antagonists, they are nonetheless highly
selective over other steroid receptors.
18. Morgan, B. P.; Swick, A. G.; Hargrove, D. M.; LaFlam-
me, J. A.; Moynihan, M. S.; Carroll, R. S.; Martin, K. A.;
Lee, E.; Decosta, D.; Bordner, J. J. Med. Chem. 2002, 45,
2417.
19. Morgan, B. P.; Liu, K. K.-C.; Dalvie, D. K.; Swick, A. G.;
Hargrove, D. M.; Wilson, T. C.; LaFlamme, J. A.;
Moynihan, M. S.; Rushing, M. A.; Woodworth, G. F.;
Li, J.; Trilles, R. V.; Yang, X.; Harper, K. W.; Carroll, R.
S.; Martin, K. A.; Nardone, N. A.; O’Donnell, J. P.;
Faletto, M. B.; Vage, C.; Soliman, V. Lett. Drug Design
Discov. 2004, 1, 1.
20. Prepared by benzylation of 1-benzyl-4-oxo-piperidine-3-
carboxylic acid methyl ester (NaH, DMF, benzyl bro-
mide) followed by hydrolysis and decarboxylation (HCl/
MeOH): Shue, H.-J.; Shih, N.-Y.; Blythin, D. J.; Chen, X.;
Piwinski, J. J.; McCormick, K. D. US Patent 5,869,488,
1999.
21. Barth, M. M.; Binet, J. L.; Thomas, D. M.; de Fornel, D.
C.; Samreth, S.; Schuber, F. J.; Renaut, P. P. J. Med.
Chem. 1996, 39, 2302.
This initial exploration of the 2-azadecalin system
produced GR antagonists that appear to bind in the
‘RU-like’ binding mode, as opposed to the alternative
‘CP-like’ orientation. Selective, high affinity GR ligands
were produced which had moderate GR functional
antagonist activity relative to mifepristone or CP-
409069. The 2-azadecalin system therefore provides a
useful scaffold for further investigation of high affinity
antagonists with increased GR functional activity.
22. The stereochemistry of the major diastereomer (diastereo-
meric ratio ca. 5:1) was assigned on the basis of propensity
of this system to add nucleophiles from the a-face. For
example, NaBH4 reduction affords predominantly (>80%)
the b-alcohol, the stereochemistry of which was rigorously
established by NMR.
Acknowledgment
We thank Dr. Alec Oxford for his expert advice and
encouragement.

5708
R. D. Clark et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5704–5708
23. 51: treatment of 49 with Zn(CN)2 and Pd(PPh3)4 in DMF/
microwave; 52: treatment of 50 with Fe and NH4Cl/aq
IPA; 53: treatment of 52 with CH2O, HCO2H, EtOH/
microwave; 54–56: sulfonylation or acylation of 52; 57:
Suzuki coupling of 49 with 4-pyridyl boronic acid; 58 and
59: amination of the tetramethyl[1,3,2]dioxaboran-2-yl
derivative of bromo compound 49.
one can postulate a reverse Mannich-type process fol-
lowed by ring closure.
29. Estrogen receptor: [3H]estradiol, Pan Vera 26467A ERa;
androgen receptor: [3H]dihydrotesterone, Pan Vera 24938
AR; mineralocorticoid receptor: [3H]aldosterone, Sf9 cells/
recombinant MR; progesterone receptor: [3H]progester-
one, Pan Vera 24900 PR.
24. Pfau, M.; Revial, G.; Guingand, A.; d’Angelo, J. J. Am.
Chem. Soc. 1985, 107, 273.
25. Available from 1 by hydrogenation (Pd/C) in EtOH
containing di-tert-butyl dicarbonate.
26. Somewhat unexpectedly, base treatment of the Michael
product (61 and epimer) afforded (R)-63 with a lower ee
(ca. 85%).
27. Enantiomeric excess was determined by chiral HPLC
analysis using a Chiralpack 1A column.
28. BOC protection was employed in this sequence as the
benzyl derivative corresponding to (R)-63 underwent
partial (ca. 20%) racemization upon deprotection with
ACE-Cl. The mechanism through which partial racemi-
zation of the quaternary center occurs is unclear, although
30. (a) Richards, S. J.; von Geldern, T. W.; Jacobson, P.;
Wilcox, D.; Nguyen, P.; Ohman, L.; Osterlund, M.;
Gelius, B.; Grynfarb, M.; Goos-Nilsson, A.; Wang, J.;
Fung, S.; Kalmanovich, M. Bioorg. Med. Chem. Lett.
2006, 16, 6086; (b) Backes, B. J.; Hamilton, G. L.;
Nguyen, P.; Wilcox, D.; Fung, S.; Wang, J.; Grynfarb, M.;
Goos-Nilsson, A.; Jacobson, P. B.; von Geldern, T. W.
Bioorg. Med. Chem. Lett. 2007, 17, 40.
31. (a) Link, J. T.; Sorensen, B. K.; Patel, J.; Emery, M.;
Grynfarb, M.; Goos-Nilsson, A. Bioorg. Med. Chem. Lett.
2004, 14, 2209; (b) Akritopoulou-Zanze, I.; Patel, J. R.;
Hartandi, K.; Brenneman, J.; Winn, M.; Pratt, J. K.;
Grynfarb, M.; Goos-Nisson, A.; von Geldern, T. W.;
Kym, P. R. Bioorg. Med. Chem. Lett. 2004, 14, 2079.