Indanylacetic acids as PPAR-δ activator insulin sensitizers

Article abstract of DOI:10.1016/j.bmcl.2007.03.057

A series of indane acetic acid derivatives were prepared which show a spectrum of activity as insulin sensitizers and PPAR-α and PPAR-δ ligands. In vivo data are presented for insulin sensitizers with selectivity for PPAR-δ over PPAR-α.

Full text of DOI:10.1016/j.bmcl.2007.03.057

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4369–4373
Indanylacetic acids as PPAR-d activator insulin sensitizers
Philip Wickens,a,* Chengzhi Zhang,a Xin Ma,a Qian Zhao,a John Amatruda,b
William Bullock,a Michael Burns,b Louis-David Cantin,a Chih-Yuan Chuang,a
Thomas Claus,b Miao Dai,a Fernando Dela Cruz,b David Dickson,a Frederick J. Ehrgott,a
Dongping Fan,a Sarah Heald,a Martin Hentemann,a Christiana I. Iwuagwu,a
Jeffrey S. Johnson,a Ellalahewage Kumarasinghe,a David Ladner,a Rico Lavoie,a
Sidney Liang,a James N. Livingston,b Derek Lowe,a Steve Magnuson,a
Gretchen Mannelly,a Ingo Mugge,a Herbert Ogutu,a Susan Pleasic-Williams,b
Robert W. Schoenleber,a Jeff Shapiro,b Tatiana Shelekhin,a Laurel Sweet,b
Christopher Townb and Manami Tsutsumib
aDepartment of Chemistry Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
bDepartment of Metabolics Disorders Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
Received 11 January 2007; revised 17 March 2007; accepted 19 March 2007
Available online 23 March 2007
Abstract—A series of indane acetic acid derivatives were prepared which show a spectrum of activity as insulin sensitizers and
PPAR-a and PPAR-d ligands. In vivo data are presented for insulin sensitizers with selectivity for PPAR-d over PPAR-a.
ꢀ 2007 Published by Elsevier Ltd.
Peroxisome proliferator-activated receptors (PPARs)
are part of the nuclear receptor superfamily and have
captured the interest of the pharmaceutical industry as
drug targets.1,2 The PPAR receptors regulate lipid
metabolism, with fatty acid catabolism controlled by
PPAR-a and PPAR-d, and lipid storage and adipogene-
sis by PPAR-c.3,4 PPAR-c agonists such as rosiglitazone
and pioglitazone are marketed for the treatment of dia-
betes.5 Dual acting PPAR-a/-c compounds for type II
diabetes have been investigated by a number of pharma-
ceutical companies6 including ourselves,7 since the com-
bined profile of insulin resistance and dyslipidemia
could be treated through such activities. Since PPAR-d
also has great promise in treating dyslipidemia,8,9 the
industry has also moved into the area of PPAR pan
agonists (a,c,d).10,11
agonists as novel euglycemic agents with a reduced
weight gain profile, and we have observed reduced
weight gain also with our dual activator when compared
to Rosiglitazone (data not disclosed).
Many of the PPAR agonists to date (generally c or dual
a/c agonists) contain an aryl ring with a 1–3 atom spacer
to an acid or an acid isostere. A recent paper from our
group7 used a substituted indane acetic acid as a head
group for molecules which showed potent PPAR-a/-c
agonist activity. Compounds such as 1 without substitu-
tion a to the acid were more selective for PPAR-d over
PPAR-a (see examples in Table 1). Insulin receptor
induction was also measured as a readout of insulin-sen-
sitizing activity.
O
OH
O
The PPAR-c/-d combination remains relatively unex-
plored, however.11,12 A recent publication reported by
Lilly Research Laboratories13 describes PPAR-c/-d
N
O
1
Racemic and chiral syntheses of our indane acetic acid
derivatives have been published.14–16 The synthesis of
the chiral indane phenol is shown in Scheme 1. Starting
Keywords: PPAR; Insulin sensitizer; Indane acetic acids.
*
Corresponding author. Tel.: +1 416 661 2102; fax: +1 416 661
2108; e-mail: pcwickens@hotmail.com
0960-894X/$ - see front matter ꢀ 2007 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.03.057

4370
P. Wickens et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4369–4373
Table 1. SAR of a substituted indane acetic acids
O
O
Rx
R2
OEt
X
Y
R1
+
OH
CH3
O
OH
HO
N
O
v
iv
a, b
Compound
RX
Fret d EC50
(nM)
Fret a EC50
(nM)
a/d
O
1
2
3
4
H
CH3
27
120
890
220
130
670
32
1.8
OH
R2
X
Y
R1
CH2CH3
OCH3
590
4400
0.22
0.15
O
vi
Scheme 2. Reagents and conditions: (a) ADDP/Ph3P, THF, ꢀ80%; (b)
LiOH, aq MeOH, ꢀ80%.
O
O
c-e
OEt
a, b
The two PPAR activities showed different SARs. Table
1 shows the effects of substitution a to the acid, in which
the initial evaluations of diastereomeric mixtures
showed a trend. The trend suggests the smaller the a
substitution the more active the mixture at PPAR-d
and the greater the selectivity over PPAR-a. To reduce
the number of chiral centers, we chose to concentrate
our efforts on the a-unsubstituted series which also
had the best selectivity for PPAR-d over -a.
or
f
CH3O
CH3O
i
ii
O
O
OH
OEt
g,h
CH3O
HO
iii
iv
Chromatographic resolution19 of the remaining chiral
center showed that the S-enantiomer of 1 had an EC50
of 11 nM at PPAR-d, 2400 nM at PPAR-a, and 246 nM
at m-IS (R-enantiomer PPAR-d EC50 = 1.3lM). Chiral
assignments were confirmed by X-ray. Subsequently, all
analogs were made with the (S) intermediate (Scheme 1).
Scheme 1. Reagents and conditions: (a) Zn powder, CuCl, ethyl
bromoacetate; (b) H2, Pd/C; (c) NaOH, EtOH; (d) (S)-(À)-1-pheneth-
ylamine; (e) HCl (99% ee, 27% over 5 steps); (f) amano lipase PS
(98% ee, 45% over 3 steps); (g) TMSCl, EtOH, 98%; (h) EtSH, AlCl3,
DCM, 99%.
The acetic acid substitution of 1 appears to be optimal
(Table 1). Substitutions around the indane in the 4, 6,
and 7 positions (counting around the indane starting from
the acetic acid substituted position) all resulted in inactive
compounds (not shown). Replacement of the indane moi-
ety by a six-membered ring led to loss of PPAR-d activity
(5), as did replacement by a tetrahydrofuran moiety (6).
Substitution of the 2 and 3 positions of the indane (7
and 8) resulted in weakly active compounds (Table 2).
from the commercially available 5-methoxyindanone (i),
Reformatsky reaction with ethyl bromoacetate gives the
coupled olefin. Subsequent hydrogenation gives the
ethyl ester of the indane acetic acid (ii). The separation
of the two enantiomers was achieved using one of the
two methods shown in Scheme 1. Resolution of the free
acid through its (S)-(À)-1-phenethylamine salt followed
by aqueous HCl gave the chiral acid (99% ee). Alterna-
tively, enzymatic hydrolysis (Amano Lipase PS) gave the
acid in 98% ee. In either case, the acid was reesterified
using TMSCl and EtOH, followed by the methyl ether
cleavage using AlCl3 and ethanethiol.
Extensive SAR work was then devoted to the heterocy-
clic region and its aryl substituent (Table 3). Starting
with the 5-methyloxazole moiety, the 4-substitution on
the phenyl ring seemed more desirable for PPAR-d
activity and m-IS, than 3-substitution. Very few com-
pounds were explored with substitution in the 2 posi-
tion. 2-Substituted compounds were made in previous
programs with substitution a to the acid7 which resulted
in compounds with weak PPAR-c activity when com-
pared to compounds with 3- and 4-substitution. (In
the thiazole series the 2-F compound 32, Table 4, also
had weak insulin sensitivity.)
The synthesis can then be completed by coupling a het-
erocyclic alkylalcohol (v) (prepared by various methods
previously published by our group15) with the indane
phenol iv (Scheme 2), followed by hydrolysis of the
ester.
Our indane acetic acid derivatives (vi) were evaluated
by FRET assays using the PPAR-d and PPAR-a ligand
binding domains and the biotinylated TRAP 220 coac-
tivator protein, and streptavidin-labeled APC.17 Active
compounds were then tested in a cellular transactiva-
tion assay (results not shown), which generally paral-
leled the FRET data, and in a cell based insulin
receptor assay (Insulin sensitivity-IS)18 in mouse
3T3L1 preadipocytes.
Replacements of the oxazole by thiazoles and imidazoles
are shown in Table 4. (Pyrazoles were also prepared, but
no compounds made had activity in the insulin sensitiv-
ity assay.) The thiazole compounds (24–38) have compa-
rable PPAR-d activity and insulin-sensitizing activity
to the oxazoles (9–23) but seem to be generally more

P. Wickens et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4369–4373
Table 2. SAR of substituted indane acetic acids and other modification to the indane
4371
O
N
RY
O
Compound
RY
Fret d EC50 (nM)
m-IS EC50 (nM)
Fret a EC50 (nM)
O
OH
5
3140 (50%)
2190 (58%)
10,000 (15%)
7470 (11%)
10,000 (7.5%)
O
OH
6
7
8
10,000
883
O
O
O
OH
OH
5440 (58%)
253 (66%)
6480 (38%)
Table 3. SAR of substitution on the phenyl of 2-phenyl oxazoles
the selectivity for PPAR-d over PPAR-a while maintain-
ing the profile of the oxazole series. However, groups
larger than ethyl (50, 51) showed decreased activity in
the insulin sensitization assay.
O
OH
CH3
R1
O
N
O
The most active and selective compounds were evalu-
ated through in vivo animal models of type II diabetes
and dyslipidemia. Preliminary PK studies showed that
the oxazoles and thiazoles had good oral exposure
when dosed as a suspension at 3 mg/kg, with the thi-
azole series showing generally better blood levels. In
vitro, the thiazoles had similar insulin sensitivity and
PPAR-d activity to the oxazoles but were more selec-
tive for PPAR-d over PPAR-a. Interestingly, when
compared in vivo, the oxazoles outperformed the
thiazoles in both the type II diabetes model and the
dyslipidemia model (perhaps due to increased protein
binding of the thiazoles). Compound 17 (3 mg/kg
po) lowered blood glucose by the same amount as ros-
iglitazone (10 mg/kg po) in the db/db mouse model. In
the hApoA1 mouse model, compound 17 raised serum
HDL by 29% and lowered serum triglycerides by 38%
when dosed orally at 30 mg/kg. By comparison, the
potent PPAR-d selective compound GW 501516 raised
serum HDL by 30% and lowered serum triglycerides
by 33% at 10 mg/kg po. (The higher dose of 17 was
required, we believe, because GW 501516 has similar
PPAR-d activity in mouse and human, while 17 is
much more active at PPAR-d in the human than the
mouse.)
Compound R1
Fret d EC50 m-IS EC50 Fret a EC50
(nM)
(nM)
(nM)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
H
4-MeO
14
3
231
29
2464
798
3-MeO 18
82
27
1890
78
4-Et
3
21
1
4-t-Bu
4-i-Pr
3-F
nd
546
300
33
13
7
371
354
45
1690
504
4-F
4-Ph
4-Me
3-Me
4-CN
3-CN
3-Cl
6
2718
444
4
24
64
35
73
13
3
101
301
164
200
212
1152
9975
10,001
1573
1188
4-Cl
selective versus PPAR-a. The imidazole analogs prefer
N-substitution on the Y position, but substituents larger
than a methyl on the imidazole nitrogen were inactive in
the m-IS assay (not shown). The methyl analogs made
again show a preference for 4-substituted aryls. The re-
versed oxazoles were also made but lacked insulin-sensi-
tizing activity (not shown). The phenol ether chain was
also modified, but 1-carbon and 3-carbon lengths re-
sulted in compounds with modest insulin-sensitizing
activity.
Thus, the in vitro PPAR-d activity and increases in insu-
lin sensitivity seen with compound 17 seemed to trans-
late well in vivo, and this complementary activity was
felt worthy of further investigation. When this series of
compounds was explored in toxicology studies no pro-
hibitive findings were observed.
Finally, substituents in the 5 position of the oxazoles
(R2, Table 5) were explored. Ethyl substitution increased

4372
P. Wickens et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4369–4373
Table 4. SAR of substitution on the phenyl of 2-phenylthiazoles and SAR on the phenyl of 2-phenyl-1-me-imidazoles
Table 5. SAR of substitution on the phenyl of 2-phenyloxazoles with substitution on the 5 position of the oxazole
O
R2
OH
R1
O
N
O
Compound
R1
R2
Fret d EC50 (nM)
m-IS EC50 (nM) (% agon)
Fret a EC50 (nM)
47
48
49
50
51
H
Et
Et
7
186
30
10,000
1020
2530
4-Me
4-Et
H
5
Et
Pr
11
10
607
35
3000 (32)
>3000 (9%)
10,000
10,000
H
PhEt–
5. Ram, V. J. Drugs Today 2003, 39, 609.
Acknowledgments
6. Henke, B. R. J. Med. Chem. 2004, 47, 4118.
7. Lowe, D. B.; Bifulco, N.; Bullock, W. H.; Claus, T.; Coish,
P.; Dai, M.; Dela Cruz, F. E.; Dickson, D.; Fan, D.;
Hoover-Litty, H.; Li, T.; Ma, X.; Mannelly, G.; Monahan,
M.; Muegge, I.; O’Connor, S.; Rodriguez, M.; Shelekin, T.;
Stolle, A.; Sweet, L.; Wang, M.; Wang, Y.; Zhang, C.;
Zhang, H.; Zhang, M.; Zhao, K.; Zhao, Q.; Zhu, J.; Zhu,
L.; Tsutsumi, M. Biorg. Med. Chem. Lett. 2006, 16, 297.
8. Barish, G. D.; Nakar, V. A.; Eveans, R. M. J. Clin. Invest.
2006, 116, 590.
The authors thank Jon Brice for his chiral HPLC meth-
od development, Jordi Benet-Buchholz for X-ray crys-
tallography, and Anthony Paiva for mass spectral
analysis.
References and notes
9. Bedu, E.; Wahli, W.; Desvergne, B. Expert Opin. Ther.
Targets 2005, 9, 861.
10. Evans, J. L.; Lin, J. J.; Goldfine, I. D. Curr. Diab. Rev.
2005, 1, 299.
1. Savkur, R. S.; Miller, A. R. Expert Opin. Investig. Drugs
2006, 15, 763.
2. Ramachandran, U.; Kumar, R.; Mittal, A. Mini Rev.
Med. Chem. 2006, 6, 563.
11. Cantin, L.-D.; Liang, S.; Ogutu, H.; Iwuagwu, C. I.;
Boakye, K.; Bullock, W.; Burns, M.; Clark, R.; Claus,
T.; delaCruz, F. E.; Daly, M.; Ehrgott, F. J.; Johnson,
J. S.; Keiper, C.; Livingston, J. N.; Schoenleber, R. W.;
3. Hummasti, S.; Tontonoz, P. Mol. Endocrinol. 2006, 20,
1261.
4. Semple, R. K.; Krishna, K.; Chatterjee, K.; O’Rahilly, S.
J. Clin. Invest. 2006, 116, 581.

P. Wickens et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4369–4373
4373
Shapiro, J.; Town, C.; Yang, L.; Tsutsumi, M.; Ma, X.
Biorg. Med. Chem. Lett. 2007, in Press.
12. Liu, K. G.; Lambert, M. H.; Leesnitzer, L. M.; Oliver, W.
J.; Ott, R. J.; Plunket, K. D.; Stuart, L. W.; Brown, P. J.;
Wilson, T. M.; Sternbach, D. D. Biorg. Med. Chem. Lett.
2001, 11, 2959.
13. Xu, Y.; Etgen, G. J.; Broderick, C. L.; Canada, E.;
Gonzalez, I.; Lamar, J.; Montrose-Rafizadeh, C.; Oldham,
B. A.; Osborne, J. J.; Xie, C.; Shi, Q.; Winneroski, L. L.;
York, J.; Yumibe, N.; Zink, R.; Mantlo, N. J. Med. Chem.
2006, 49, 5649.
14. Lowe, D. B.; Wickens, P. L.; Ma, X.; Zhang, M.; Bullock,
W. H.; Coish, P. D. G.; Mugge, I. A.; Stolle, A.; Wang,
M.; Wang, Y.; Zhang, C.; Zhang, H.; Zhu, L.; Tustsumi,
M.; Livingston, J. WO 03/011842.
15. Wickens, P. L.; Cantin, L.; Chuang, C.; Dai, M.;
Hentemann, M. F.; Kumarasinghe, E.; Liang, S. X.;
Lowe, D. B.; Shelekhin, T. E.; Wang, Y.; Zhang, C.;
Zhang, H.; Zhao, Q. WO 2004/011446.
16. Wickens, P. L.; Cantin, L.; Kumarasinghe, E.; Chuang,
C.; Liang, S. X. WO 2003/089418.
17. BAY compounds were incubated with europium-labeled
anti-GST antibody, GST-tagged human PPAR-d ligand
binding domain, biotinylated TRAP220 coactivator
protein, and streptavidin-labeled APC. The samples
were placed in a fluorimeter with an excitation wave-
length of 340 nm and emission wavelengths of 615 and
665 nm.
18. Insulin receptor binding in 3T3-L1 cells treated with
compounds: 3T3-L1 cells were seeded at 9300 cells per
well in Costar flat-bottomed TC and incubated for 1 week
until they were 2 days post-confluent. The cells were then
treated for 2 days with differentiation media (Dulbecco’s
modified Eagle’s medium (DMEM), 100 lg/mL Penicillin/
Streptomycin, 2 mM L-Glutamine, 10% Fetal Bovine
Serum) containing 0.5 lM human insulin-like growth
factor (IGF-1) and test compounds. After treatment, the
media were replaced with differentiation media, and the
cells were incubated for 4 days. The cells were then
assayed for insulin receptor activity. After washing the
cells with buffer, they were incubated with 0.1 nM [125I]-
insulin and (+/À) 100 nM unlabeled insulin, and incubated
at room temperature for 1 h. The cells were then washed
3· with buffer, dissolved with 1 N NaOH, and counted on
a c counter. An EC50 value was determined if a plateau
was attained and percent maximum stimulation was
assessed.
19. Chiral HPLC method: chiralpak AD-H 4.6 · 250 mm.
A = hexane (0.4% TFA) B = IPA. 10% B at 1 mL/min.
UV = 280 nM.