3,3-Bisaryloxindoles as mineralocorticoid receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.03.086

Syntheses and SAR studies of 3,3-bisaryloxindole analogues provided potent mineralocorticoid receptor (MR) antagonists that were selective over other steroid nuclear hormone receptors.

Full text of DOI:10.1016/j.bmcl.2005.03.086

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2553–2557
3,3-Bisaryloxindoles as mineralocorticoid receptor antagonists
David A. Neel,* Matthew L. Brown, Peter A. Lander, Timothy A. Grese, Jean M. Defauw,
Robert A. Doti, Todd Fields, Sally Ann Kelley, Stephon Smith, Karen M. Zimmerman,
Mitchell I. Steinberg and Prabhakar K. Jadhav
Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285, USA
Received 28 December 2004; revised 10 March 2005; accepted 15 March 2005
Available online 12 April 2005
Abstract—Syntheses and SAR studies of 3,3-bisaryloxindole analogues provided potent mineralocorticoid receptor (MR) antago-
nists that were selective over other steroid nuclear hormone receptors.
Ó 2005 Elsevier Ltd. All rights reserved.
Nuclear hormone receptors (NHRs) are a superfamily
of ligand-activated transcriptional regulators that are in-
volved in many physiological and pathological pro-
cesses. The steroid receptor subfamily of NHRs, use
classical steroid hormones as natural ligands to regulate
gene expression. Steroid NHRs include the glucocorti-
coid receptor (GR), progesterone receptor (PR), andro-
gen receptor (AR), estrogen receptor (ER), and
mineralocorticoid receptor (MR). MRs located in the
kidney, are used as the primary means to regulate body
sodium, potassium, and water balance by aldosterone.1
More recently, aldosterone has been shown to mediate
deleterious cardiovascular effects through MRs in brain,
heart, and vascular tissue.2
of aldosterone levels over time, a phenomenon known as
Ôaldosterone escapeÕ (RESOLVD study).4 These elevated
levels of aldosterone result in increased cardiovascular
risk for CHF patients and accentuate the potential util-
ity of MR antagonists. Moreover, a recent review sug-
gests that MR antagonists might be useful in other
vascular diseases, rather than simply those with elevated
levels of aldosterone.5
Spironolactone and more recently, eplerenone,6 are the
only therapeutic options available to antagonize the
mineralocorticoid receptor. Spironolactone lacks selec-
tivity against other steroid nuclear hormone receptors,
primarily AR and PR. Long term use is limited due to
undesirable side effects such as gynecomastia in men
and menstrual irregularity in women.7 Eplerenone is
more selective, but less potent than spironolactone in
vitro.8
Congestive heart failure (CHF) continues to be a major
contributor to morbidity and mortality. Standard thera-
pies include digitalis, ACE inhibitors, diuretics, and b-
blockers. These agents can help relieve symptoms and
prolong life, but there continues to be a need for novel
agents that target the underlying pathophysiological
basis of the disorder. The randomized aldactone
evaluation study (RALES) has demonstrated that
spironolactone, an MR antagonist, when added to stan-
dard therapy, reduces risk of death by 30% and im-
proves cardiac function of patients with severe CHF.3
The renin–angiotensin–aldosterone-system (RAAS) is
responsible for maintaining blood pressure and volume
homeostasis. Upstream inhibitors of the RAAS such
as angiotensin converting enzyme (ACE) inhibitors
and angiotensin II (AII) blockers do not prevent the rise
Our objective was to identify a non-steroidal human
MR (hMR) antagonist and thereby diminish the adverse
effects of elevated levels of aldosterone under patholog-
ical conditions. In order to avoid the deleterious side
HO
OH
O
N
2a
Ph
*
Corresponding author. E-mail: neel_david_a@lilly.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.03.086

2554
D. A. Neel et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2553–2557
effects of spironolactone, such an antagonist must be
selective for hMR over other steroid nuclear hormone
receptors. Initial screening identified oxindole derivative
2a as a ligand for human MR (hMR) and SAR studies
were initiated to improve potency and selectivity. To our
knowledge, these are the first non-steroidal MR
antagonists.
a phenol to provide 2a–e. Deoxygenation to 3a was
accomplished by triflation of 2a and subsequent Pd(0)
catalyzed hydrogenolysis.10
Treatment of isatin with 2,6-dimethylphenol (Scheme 2)
in a Friedel–Crafts procedure11 provided 4. Silyl protec-
tion of the phenols followed by alkylation and deprotec-
tion gave compounds with substituted benzyl groups (6).
N-Arylation of 5 with phenyl boronic acids and cupric
acetate led to compounds 7a–l.12
3,3-Bisaryloxindoles were obtained using superacid
mediated coupling methodology reported by Olah and
co-workers9 (Scheme 1). The dicationic intermediate
generated by subjecting N-benzylisatin 1 to trifluorome-
thanesulfonic acid was condensed with 2 equivalents of
The unsymmetrical bisaryloxindoles 10 were obtained as
shown in Scheme 3. Isatin was treated with 2 equiv of
HO
R2
R1
R1
OH
O
O
a
b
R2
O
N
O
N
H
Isatin
1
O
N
2a-e
Ph
2
Ph
e
c,d
R1 = R2 = CH3
MeO
O
N
OMe
3a
Ph
O
N
3b
Ph
Scheme 1. Reagents and conditions: (a) KO-t-Bu, BrBn, 84%; (b) excess phenol, TfOH, 50–70%; (c) 2,4,6-collidine, Tf2O, 67%; (d) n-Bu3N,
Ph2P(CH2)3PPh2, PdCl2(PPh3)2, HCO2H, DMF, 58%; (e) Cs2CO3, MeI, DMF, 40 °C, 76%.
HO
TBDMSO
OH
OTBDMS
O
a
b
O
N
H
O
N
H
O
N
H
4
Isatin
5
d,e
HO
c,e
OH
HO
OH
O
6
N
7a-l
O
R
N
R
Scheme 2. Reagents and conditions: (a) AcOH, AlCl3, 86%; (b) imidazole, DMF, TBDMSCl, 75%; (c) KO-t-Bu, XCH2R, 51–66%; (d) RPhB(OH)2,
˚
Cu(OAc)2, TEA, 4 A sieves, 15–52%; (e) TBAF, 86%.

D. A. Neel et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2553–2557
2555
TBDMSO
HO
OTBDMS
a
b,c
OH
O
OH
O
N
H
Br
N
8
9
R1
HO
R2
R3
R4
d
OH
O
N
10a-o
R1
˚
Scheme 3. Reagents and conditions: (a) n-BuLi/THF/À70 °C, then isatin, 80%; (b) R1PhB(OH)2, Cu(OAc)2, TEA, 4 A sieves, 64%; (c) TBAF, 86%;
(d) TFA, substituted phenol, 61–75%.
the lithio adduct of (4-bromo-2,6-dimethylphenoxy)-
tert-butyldimethylsilane to provide 8. Coupling with
an appropriately substituted phenylboronic acid fol-
lowed by deprotection of the phenol provided intermedi-
ate 9. Formation of the carbocation in trifluoroacetic
acid in the presence of an excess of a phenol gave 10a–o.
antagonist affinity (Kb) was estimated using the ratio of
the EC50 in the presence and absence of 1 lM test com-
pound using the Schild equation, Kb = [antagonist]/
(dose ratio À 1).14 Non-linear regression analysis was
used to estimate EC50 values.
We quickly recognized that simple changes to the bis-
dimethyl phenol resulted in dramatic loss of hMR affin-
ity (Table 1). The biological activity of 2a was
surprisingly sensitive to any changes to the methyl
groups as shown by 2b and e. Other changes to these
symmetrical molecules, wherein the hydroxyls were re-
moved (3a), or capped with methyls (3b), resulted in
complete loss of activity.
A radioligand binding assay was used to identify human
MR (hMR) antagonists. Full length hMR was isolated
from a human kidney cDNA library and hMR protein
was expressed in Sf9 cells using the baculovirus expres-
sion vector system (PharMingen). Affinity of com-
pounds for hMR was determined using vacuum
filtration methods. IC50 values defined as the concentra-
tion of test compound required to decrease [3H]aldoste-
rone binding by 50% were determined using non-linear
regression curve fitting. Inhibition constants (Ki) were
calculated from these IC50 values using the Cheng–Prus-
off equation.13 Where compounds exhibited weak affin-
ity and full inhibition was not achieved at 3 lM, IC50
could not be calculated, and % inhibition at 1 lM test
compound is reported. All data points represent means
of at least two determinations unless otherwise
indicated.
Extensive substitution on the N-benzyl group of 2a (giv-
ing compounds of type 6) resulted in only modest
improvements in potency or selectivity (data not
shown). However, when the phenyl ring was directly at-
tached to the oxindole core, there was a dramatic
Table 1. hMR Binding of compounds 2a–e and 3a and b
HO
R
R2
O
R1
R1
A panel of related steroid nuclear hormone receptors
was used to determine hMR selectivity. These included
binding assays for human glucocorticoid receptor
(hGR), human androgen receptor (hAR), human pro-
gesterone receptor (hPR) and human estrogen receptor
(hERa, hERb).
OH
R2
R
O
N
N
3a,b
2a-e
Ph
Ph
Cpd R1 R2 hMR Ki (nM) Cpd
R
hMR Ki (nM)
Functional activity was determined using COS-7 cells
co-transfected with an expression vector containing full
length hMR and a luciferase reporter plasmid respon-
sive to hMR activation. Ability of test compounds to in-
hibit the agonist response of aldosterone on hMR was
determined in the absence and presence of test com-
pound and measured by luciferase luminesence. Neutral
2a
2b
2c
2d
2e
Me Me 65
3a
3b
H
16%a
H
H
H
F
22%a
423
OMe 0%a
Me
F
16%a
701
Cl
Cl
a Percent inhibition at 1 lM antagonist concentration.

2556
D. A. Neel et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2553–2557
Table 2. Steroid NHR binding of compounds 7a–l
The unsymmetrical phenols (Table 3) provided com-
pounds with the optimum selectivity. In particular, the
resorcinol analogues, 10d,e,g, and o were less potent at
hGR than any of the other oxindoles in this series. We
speculate that the ortho-phenol provides a rigid con-
former through hydrogen bonding to the carbonyl of
the oxindole, allowing for optimum discrimination be-
tween the binding pockets of hMR and hGR. Also nota-
ble is the lack of space available for larger substituents.
Ethyl 10i, was dramatically less potent than methyl 10h.
The resorcinol analogues, 10d,e,g, and o had the best
selectivity across the panel of steroid NHRs tested.
Compound 10o was separated by chiral chromatogra-
phy into its (+)- and (À)-enantiomers. The (À)-enantio-
mer was significantly more potent and selective.
HO
OH
7a-l
O
N
R
Cpd
R
hMR
Ki (nM)
hGR hPR hAR hERa hERb
7a
7b
7c
7d
7e
7h
7i
H
2-Cl
27
15
24
7
147
188b
115
84
44%a 41%a
36 51%a
733 >1100
179
594 >1100
354
1044
844
2-Me
3-Me
4-Me
3-OMe
4-OMe
3-OEt
3-CF3
4-CF3
218
138
306
425
402
316
277
390
183
315
242
Table 4. Transcriptional activity of select compounds
20
0.6
39
3
180b
51
297
715
890
930
Cpd
hMR Ki (nM)
Kb (nM)
158
224
275b
145
36%a
402 >1100
7d
7h
7
56
26
7j
284
194
196
163
771
926
316
0.6
1
7k
7l
12
156
10a
10d
89
821 43%a
1016
3
150
181
93
a Percent inhibition at 1 lM antagonist concentration.
10f
10g
3
b One determination.
2
10o(À)
1
41
18
Spironolactone
2
improvement in potency and selectivity (Table 2). meta-
Substitution provided the best combination of potency
and selectivity as exemplified by 7d,h, and j.
The functional activity of a select group of ligands is
shown in Table 4. The resorcinol analogue 10o(À) as
Table 3. Steroid NHR binding of compounds 10a–o
HO
R2
R3
R4
OH
O
N
10a-o
R1
Cpd
R1
R2
R3
R4
hMR Ki (nM)
hGR
hPR
hAR
hERa
hERb
10a
10b
10c
OMe
OMe
OMe
OMe
OMe
OCF3
OCF3
OCF3
OCF3
Me
H
Me
H
Me
Me
H
1
2
165
66
485
453
573
678
395
762
796
217
378
206
228
251
448
310
245
33%
49%
539
>1100
>1100
>1100
>1100
>1100
>1100
335
>1100
>1100
>1100
>1100
>1100
>1100
896
Me
Me
H
44%a
44%a
19%a
1105
29%
1203
41%
47%
40%
53%
52%
50%
242
H
3
62
10d
10e
OH
OH
Me
OH
H
H
3
793b
693b
470
519
843b
72
H
Me
Me
Me
H
1
10f
H
1
10g
10h
10i
H
2
Me
Et
H
7
>1100
>1100
>1100
>1100
>1100
346
>1100
>1100
>1100
>1100
>1100
>1100
>1100
>1100
>1100
>1100
H
H
395
8
10j
Me
H
Me
H
100
146b
79
10k
10l
Me
Me
Me
H
6
Me
H
H
9
10m
10n
10o
10o(À)
10o(+)
Me
Me
F
OMe
Me
Me
Me
Me
130
6
143
77
F
H
650
579
Me
Me
H
OH
OH
OH
2
334
582
686b
37%
16%a
20%
H
1
40
359
519
Me
H
a Percent inhibition at 1 lM antagonist concentration.
b One determination.

D. A. Neel et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2553–2557
2557
Delvani, J. A. Am. J. Physiol. Heart Circ. Physiol. 2002,
283, H1802; (c) McMahon, E. G. Curr. Opin. Pharmacol.
2001, 1, 190; (d) Rocha, R.; Stier, C. T. Trends Endocrinol.
Metab. 2001, 12, 308.
well as 7d and h had the greatest potency as antagonists.
There was no agonist or partial agonist activity
observed.
3. Pitt, B.; Zannad, F.; Remme, W. J.; Cody, R.; Castaigne,
A.; Perez, A.; Palensky, J.; Wittes, J. N. Engl. J. Med.
1999, 341, 709.
4. McKelvie, R. S.; Yusuf, S.; Pericak, D.; Avezum, A.;
Burns, R. J.; Probstfield, J.; Tsuyuki, R. T.; White, M.;
Rouleau, J.; Latini, R.; Maggioni, A.; Young, J.; Pogue, J.
Circulation 1999, 100, 1056.
5. Funder, J. W. Expert Opin. Investig. Drugs 2003, 12, 1963.
6. Brown, R.; Quirk, J.; Kirkpatrick, P. Nat. Rev. Drug
Discov. 2003, 2, 177.
7. Gasparo, M. A.; Whitebread, S. E.; Preiswerk, G.;
Jeunemaitre, X.; Corvol, P.; Menard, J. J. Steroid
Biochem. 1989, 32(1B), 223.
In summary, we have identified a series of 3,3-bisaryl-
oxindoles that are potent non-steroidal hMR antago-
nists. These ligands display excellent selectivity for
hMR over hGR, hPR, hAR, and hER. The best combi-
nation of hMR potency and selectivity was demon-
strated with analogues wherein one of the aryl groups
was a resorcinol derivative. On the basis of our results,
further studies to investigate the potential of selective
non-steroidal hMR antagonists to treat CHF are
warranted.
8. De Gasparo, M.; Joss, U.; Ramjoe, H.; Whitebread, S.;
Haenni, H.; Schenkel, L.; Kraehenbuehl, C.; Biollas, M.;
Grob, J.; Schmidlin, J., et al. J. Pharmacol. Exp. Ther.
1987, 240, 650.
9. Klumpp, D. A.; Yeung, K.; Surya Prakash, G. K.; Olah,
G. A. J. Org. Chem. 1998, 63, 4481.
10. Saa, J. M.; Dopico, M.; Martorell, G.; Garcia-Raso, A. J.
Org. Chem. 1990, 55, 991.
11. Song, H. N.; Lee, H. J.; Kim, H. R.; Ryu, E. K.; Kim, J.
N. Synth. Commun. 1999, 29, 3303.
Acknowledgments
The authors gratefully acknowledge Pete Borromeo for
provision of multigram quantities of 5 and Ryan Brady
for multigram quantities of 8. We thank Jeff Williams
for chiral separation of 10o. We also thank Keith Stay-
rook for providing hGR binding data and Ed Osborne
for hER binding data.
12. Chan, D. M. T.; Monaco, K. L.; Wang, R.; Winters, M. P.
Tetrahedron Lett. 1998, 39, 2933.
References and notes
13. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099.
14. Limbird, L. E. Cell Surface Receptors: A Short Course
on Theory and Methods; Kluwer Academic, 1996, pp 47–
51.
1. Fardella, C.; Miller, W. Annu. Rev. Nutr. 1996, 16, 443.
2. (a) Weber, K. Y. N. Engl. J. Med. 2001, 345, 1689; (b)
Rocha, R.; Rudolph, A. E.; Frierdich, G. E.; Nachowiak,
D. A.; Kekec, B. K.; Bloome, E. A. G.; McMahon, E. G.;