Design and synthesis of alkoxyindolyl-3-acetic acid analogs as peroxisome proliferator-activated receptor-γ/δ agonists

Article abstract of DOI:10.1016/j.bmcl.2012.11.033

A series of carbazole or phenoxazine containing alkoxyindole-3-acetic acid analogs were prepared as PPARγ/δ agonists and their transactivation activities for PPAR receptor subtypes (α, γ and δ) were investigated. Structure-activity relationship studies disclosed the effect of the lipophilic tail, attaching position of the alkoxy group and N-benzyl substitution at indole. Compound 1b was the most potent for PPARδ and 3b for PPARγ. Molecular modeling suggested two different binding modes of our alkoxyindole-3-acetic acid analogs providing the insight into their PPAR activity.

Full text of DOI:10.1016/j.bmcl.2012.11.033

Bioorganic & Medicinal Chemistry Letters 23 (2013) 513–517
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of alkoxyindolyl-3-acetic acid analogs as peroxisome
proliferator-activated receptor-
c/d agonists
⇑
Hyo Jin Gim, Hua Li, Eun Lee, Jae-Ha Ryu, Raok Jeon
Research Center for Cell Fate Control, College of Pharmacy, Sookmyung Women’s University, 52 Hyochangwon-Gil, Yongsan-Ku, Seoul 140-742, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of carbazole or phenoxazine containing alkoxyindole-3-acetic acid analogs were prepared as
Received 10 August 2012
Revised 21 October 2012
Accepted 8 November 2012
Available online 22 November 2012
PPARc/d agonists and their transactivation activities for PPAR receptor subtypes (a, c and d) were inves-
tigated. Structure–activity relationship studies disclosed the effect of the lipophilic tail, attaching position
of the alkoxy group and N-benzyl substitution at indole. Compound 1b was the most potent for PPARd
and 3b for PPAR
c. Molecular modeling suggested two different binding modes of our alkoxyindole-3-ace-
tic acid analogs providing the insight into their PPAR activity.
This paper is dedicated to Professor Young-
Ger Suh on the occasion of his 60th birthday
Ó 2012 Elsevier Ltd. All rights reserved.
Peroxisome proliferators-activated receptors (PPARs) are
a
(PPARacd) of PPARs, have been reported in an effort to reduce side
effects and increase efficacy.10 Despite the high potential of PPAR
agonists, the clinical benefits are nevertheless limited by several
adverse effects such as weight gain, renal fluid retention and in-
creased risk of heart failure.11–13 It is necessary to improve phar-
group of nuclear receptor proteins that function as a ligand acti-
vated transcription factors regulating the expression of gene
including cellular differentiation, development and metabolism,
therefore they are targets of drugs effective in treatment of meta-
bolic diseases.1,2 PPARs are a group of three isoforms, PPAR
a,
macological profiles of PPAR agonists. PPARacd agonists are
PPAR
c
and PPARd.3 PPAR
a
which is expressed at high levels in
expected to have superior therapeutic utility for the treatment of
altered lipid homeostasis and glucose homeostasis in target organs.
Recently, we reported benzoxazole containing indolylacetic
liver, kidney, heart, adipose tissue and others, has been shown to
play a critical role in the regulation of cellular uptake, activation
and b-oxidation of fatty acid. Increased fatty acid oxidation by acti-
acids and indole carboxylic acid analogs 1 as PPARc
/d agonists.14
vated PPAR
a
lowers circulating triglyceride levels and reduces adi-
which is mainly
The representative structures of PPAR agonists and our indole com-
pound depicted in Figure 1. Typically, a number of PPAR agonists
have common structural features which include polar head, aro-
matic ring, proper linker and hydrophobic tail in order (Fig. 2).
In previous study, we have found that indolylacetic acid analogs
are more active than the corresponding indole carboxylic acids
among our indole compounds. Herein we focused on the influence
of alkoxy position at the indolyl-3-acetic acid, the replacement of
bicyclic hydrophobic tail with new tricyclic rings and introduction
of benzyl substituent at the nitrogen of indole core.
Synthetic routes for the preparation of final compounds 1a–4c
are outlined in Schemes 1–4. The targeted 5- or 6-alkoxyindo-
lyly-3-acetic acid analogs were prepared by Mitsunobu reaction
of the corresponding hydroxyindoles with the various tricyclic-
linked alcohols. First, 5-hydroxyindolyl-3-acetic acid methyl ester
5a was simply prepared by esterification of commercially available
5-hydroxyindole-3-acetic acid (Scheme 1).
For the modification of position of alkoxy-linker at indole ring,
6-hydroxyindolyl-3-acetic acid methyl ester 5b was prepared by
Fukuyama indole synthesis as a key step (Scheme 2).15 The
commercially available 4-amino-3-nitrophenol was treated with
sodium nitrite under acidic condition and then potassium iodide
to give iodophenol 6, which was protected with a benzyl group
to afford compound 7. The Heck reaction of compound 7 with
posity, which improves insulin sensitivity.1 PPAR
c
associated with adipose-related functions, regulates lipid metabo-
lism, lipid uptake into adipocytes, glucose homeostasis and insulin
sensitivity, and its agonists, specifically the thiazolidinedi-
ones(TZDs) including rosiglitazone and pioglitazone, are used as
efficient insulin sensitizers in type 2 diabetes.4 To combine the
beneficial effects of insulin sensitization and lipid regulation,5
PPARac dual agonists including muraglitazar and tesaglitazar have
been paid much attention. However, the adverse toxicity profiles
PPARac dual agonists have been reported and raised critical safety
issues, which have led to the discontinuation of clinical develop-
ment.6 Recent studies revealed that PPARd is ubiquitously ex-
pressed with highest levels in adipose tissue, skeletal muscle and
intestine. PPARd is also involved in lipid metabolism and glucose
homeostasis.7–9 A number of selective PPARd agonists have been
identified, and GW501516, a potent and selective PPARd agonist
to have entered clinical trials, has shown enhanced fatty acid b-
oxidation in skeletal muscle, increased HDL cholesterol level and
decreased elevated triglyceride and insulin levels in animal mod-
el.7 Also, PPAR pan agonists which act on all the three subtypes
⇑
Corresponding author.
E-mail address: rjeon@sookmyung.ac.kr (R. Jeon).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.11.033

514
H. J. Gim et al. / Bioorg. Med. Chem. Lett. 23 (2013) 513–517
O
F3C
O
CO2H
NH
S
N
S
O
S
O
N
N
Rosiglitazone
GW501516
PPARγ agonist
PPARδ agonist
CO2H
O
O
O
OH
O
N
O
N
S
O
O
O
Ragaglitazar
Indeglitazar
PPARα/γ agonist
PPAR α/γ/δ agonist
N
N
CO H
O
O
n
2
O
n
N
R
X
N
X = O, S
O
HO2C
compound 1
PPARγ/δ agonist
PPAR α/γ/δ agonist
Figure 1. Chemical structures of PPAR agonists.
of the tested compounds was compared to the reference com-
pounds GW0746, rosiglitazone and GW0742 in the PPAR and
d transactivation assays, respectively. PPAR and d transactiva-
Aromatic Polar
Linker
a, c
ring
head
a, c
CO2R1
tion activity of alkoxyindolyl-3-acetic acid analogs were described
Hydrophobic
tail
in Table 1.
O
O
To compare the effect of hydrophobic ring system, tricyclic car-
bazole or phenoxazine was conjugated into 5- or 6-position of in-
dole scaffold through C3-methylene linker. Phenoxazine-linked
indole analogs 3a and 3b showed slightly increased activity for
N
R2
Figure 2. Design of alkoxyindolyl-3-acetic acid based PPAR
c
d agonist.
PPARc or d as compare with carbazole-linked analogs 1a and 1b,
respectively.
CO2Me
CO2H
The activity was also influenced by attaching position of the alk-
oxy-linker at indole ring. In general, 5-alkoxyindole analogs
HO
HO
a
N
H
N
H
showed higher PPAR
analogs. 5-Alkyoxy analogs 1a, 1b and 3a were more active for
both PPAR and d transactivation than those of corresponding 6-
c or d activity than those of 6-alkoxyindole
5a
c
alkoxy analogs 2a, 2b and 4a with exception of phenoxazine con-
taining N-benzylated 5-alkoxyindole 3b.
Scheme 1. Synthesis of 5-hydroxyindolyl-3-acetic acid methyl ester. Reagent and
condition: (a) SOCl2, MeOH, 0 °C–rt.
As expected from the previous study, introduction of the benzyl
group at nitrogen of indole core led to improved activity for PPAR
c
and d as compared 1a, 3a and 4a with 1b, 3b and 4b, respectively.
However, carbazole-linked 6-alkoxy analogs were not active even
after introduction of N-benzyl group while phenoxazine linked 6-
methyl acrylate gave compound 8 in good yield. Chemoselective
reduction of the nitro group of compound 8 afforded amine 9
which was formylated by treating acetoformic anhydride. The
compound 10 was treated with triphosgene in the presence of tri-
ethylamine to obtain isonitrile 11. Treatment of the isonitrile 11
with tributyltin hydride in the presence of AIBN gave 3-substituted
indole 12, and followed debenzylation provided the desired 6-
hydroxyindole 5b.
alkoxy analogs brought up the higher activity for PPARc and/or d
by N-benzylation. Among the tested compounds, the most active
compounds were carbazole containing N-benzylated analog 1b
for PPARd and phenoxazine containing N-benzylated analog 3b
for PPARc.
To rationalize the different activities between 5-alkoxy and 6-
alkoxyindolyl-3-acetic acids and effect of N-benzylation, docking
Alcohols 13 and 14 were obtained from commercially available
carbazole and phenoxazine by alkylation with 3-iodopropanol in
the presence sodium hydride. Mitsunobu reaction of hydroxyin-
doles 5 with alcohols 13 and 14 in the presence of tributylphosphine
and 1,10-(azodicarbonyl)dipiperidine (ADDP) gave compounds
15–18 which were hydrolyzed to final acids 1a–4a (Scheme 3). N-
benzylation of compounds 15–18 and the following hydrolysis gave
the final acids 1b–4c as shown in Scheme 4.
experiments into the hPPARc receptor binding domain were car-
ried out using Surflex Dock interfaced with SYBYL-X version 1.3.
The crystal structure of the human PPARc in complex with rosiglit-
azone (PDB code: 2PRG)16 was employed for the docking study. In
this automated docking program, the flexibility of the ligand is
considered while the protein is considered as a rigid structure.
The molecules for docking were sketched in the SYBYL and energy
minimizations were performed using Tripos Force Field and
PPAR agonist activities of the synthesized compounds were
evaluated by in vitro transient transactivation assays. The efficacy

H. J. Gim et al. / Bioorg. Med. Chem. Lett. 23 (2013) 513–517
515
NH2
NO2
I
I
a
c
b
HO
HO
NO2
NH2
BnO
NO2
7
6
CO2Me
CO2Me
CO2Me
e
d
BnO
NHCHO
BnO
g
BnO
f
NO2
8
9
10
CO2Me
CO2Me
h
CO2Me
N
H
N
H
BnO
BnO
NC
HO
5b
11
12
Scheme 2. Synthesis of 6-hydroxyindolyl-3-acetic acid methyl ester. Reagents and conditions: (a) NaNO2, HCl, then KI, rt, 89%; (b) BnBr, K2CO3, acetone, rt, 95%; (c) methyl
acrylate, Pd(OAc)2, (Tol)3P, TEA, MeCN, reflux, 84%; (d) Zn, AcOH, rt, 72%; (e) AcOCHO, Py, CH2Cl2, rt, 65%; (f) triphosgene, TEA, CH2Cl2, rt, 73%; (g) Bu3SnH, AIBN, Toluene,
80 °C, then HCl, 84%; (h) H2, Pd(OH)2/C, AcOH, rt, 87%.
CO2Me
5
L
a
L
b
N
OH
NH
HO
6
N
H
L: carbazole
13: carbazole
5a: 5-hydroxyindole
5b: 6-hydroxyindole
phenoxazine
14: phenoxazine
CO2Me
CO2H
5
5
c
L
L
N
O
N
O
6
6
N
H
N
H
15: carbazole, 5-alkoxy
16: carbazole, 6-alkoxy
1a: carbazole, 5-alkoxy
2a: carbazole, 6-alkoxy
3a: phenoxazine, 5-alkoxy
4a: phenoxazine, 6-alkoxy
17: phenoxazine, 5-alkoxy
18: phenoxazine, 6-alkoxy
Scheme 3. Synthesis of carbazole or phenoxazine linked indole analogs 1a–4a. Reagents and conditions: (a) 3-iodopropanol, NaH, DMF, rt; (b) ADDP, P(Bu)3, THF/toluene, rt;
(c) NaOH, THF/MeOH/H2O, rt.
CO2Me
CO2R1
5
6
5
6
L
a,b
N
O
N
O
N
H
N
R2
1b: carbazole, 5-alkoxy, R1 = H, R2= Bn
2b: carbazole, 6-alkoxy, R1 = H, R2= Bn
3b: phenoxazine, 5-alkoxy, R1 = H, R2= Bn
4b: phenoxazine, 6-alkoxy, R1 = H, R2= Bn
4c: phenoxazine, 6-alkoxy, R1 = H, R2= 4-F-Bn
15: carbazole, 5-alkoxy
16: carbazole, 6-alkoxy
17: phenoxazine, 5-alkoxy
18: phenoxazine, 6-alkoxy
Scheme 4. Synthesis of carbazole or phenoxazine linked N-benzylated indole analogs 1b–4c. Reagents and conditions: (a) benzyl or p-fluorobenzylbromide, NaH, THF, rt; (b)
NaOH, THF/MeOH/H2O, rt.
Gasteiger–Huckel charge with 20,000 iterations of conjugate gradi-
ent method in convergence criterion of 0.05 kcal/mol. The docking
analyses of compound 1a, 3a and 3b were performed using the
internal default parameters for all the variables.
ligands (Fig. 3). The carboxyl group of both compounds 1a (red)
and 3a (magenta) interacted with His323(1a) or 449(3a), Tyr473
and Gln286 and NH of indole ring interacted with Ser289 residue.
The lipophilic tails were inserted into the hydrophobic pocket of
the binding site. In there, phenoxazine analog 3a showed addi-
tional H-bonding interaction between the oxygen of phenoxazine
and Ser342, whereas the carbazole analog 1a did not.
Most of PPARc agonists including rosiglitazone interact with
critical residues in helix H12 of the LBD. Thiazolidinedione ring
of rosiglitazone forms H-bonding with Try473, His449, His323
and Ser289, where Tyr473 was located in the AF-2 helix as a key
residue for the stabilization of conformation favoring the binding
of co-activators.16 As expected, 1a and 3a were bound to the active
However, N-benzylated indole analogs did not fit into the same
binding site of PPARc which was occupied by rosiglitazone and
compounds 1a and 3a (Fig. 4). Hydrophilic head of compound 3b
occupied other binding pocket which was away from the helix
H12 and thereby louse the critical H-bonding interaction. Docking
site of PPARc co-crystallized with rosiglitazone through the known
key interaction between specific residues of the protein and the

516
H. J. Gim et al. / Bioorg. Med. Chem. Lett. 23 (2013) 513–517
Table 1
In vitro functional transactivation activity of 5- or 6-alkoxyindolylacetic acids on
murine PPARa
CO2H
5
L
N
O
6
N
R
Compd
no.
L
Alkoxy
position
R
%Maxb (10
lM)
mPPAR
a
mPPAR
c
mPPARd
1a
1b
5
5
H
Bn
10.1
NA
NA
28.7
10.5
41.3
2a
2b
6
6
H
Bn
NA
NA
NA
NA
NA
NA
N
3a
3b
5
5
H
Bn
NA
12.8
29.4
43.1
16.0
21.7
O
4a
4b
4c
6
6
6
H
NA
NA
NA
NA
NA
23.8
NA
27.3
36.6
N
Bn
p-
F-
Bn
NA means not active, which is for compounds producing transactivation activity
lower than 10% at 10 M.
Figure 4. Comparison of the docked conformation of compound 3b (blue) on the
PPAR -bound conformation of rosiglitazone (green) (PDB code: 2PRG). The connolly
surface was generated around the ligand.
l
c
a
Compounds were tested in at least three separate experiments.
b
Fold activation relative to maximum activation obtained with GW0746 (1 lM) for
PPARa, with rosiglitazone (1 lM) for PPARc, and with GW0742 (0.1 lM) for PPARd.
Figure 3. Putative binding modes of 1a (red), 3a (magenta) in the binding pocket of
PPAR co-crystallized with rosiglitazone (green) (PDB code: 2PRG). Only critical
c
amino acids interacted with the docked ligand are displayed and labeled. Several
hydrogen bonds discussed in the text are depicted as dashed red lines.
Figure 5. Putative binding modes of 3a (magenta) and 3b (blue) in the binding
pocket of PPAR
c co-crystallized with indomethacin (green) and fatty acid ligand
(cyan) (PDB code: 3ADX). Only critical amino acids interacted with the docked
ligand are displayed and labeled. Several hydrogen bonds discussed in the text are
depicted as dashed red lines.
studies suggested that there might be different modes for the bind-
ing of our indole analogs, especially N-benzyl substituted analogs,
in the active pocket of PPAR
c.
Recently, crystallographic analyses of PPAR
c
co-crystallized
compound 3a interacted with His323 and Tyr473. In contrast,
introduction of N-benzyl substituent turned the indole ring over
and revealed similar binding mode with indomethacin. Benzyl
group of compound 3b occupied hydrophobic cavity covered with
Phe282, Phe360 and Phe363 in a similar way to p-Cl-benzoyl group
of indomethacin. This result suggested that the additional hydro-
phobic interaction of benzyl group in the hydrophobic pocket of
the binding site might be responsible to higher activity of N-benzy-
lated indole analogs than those of non-benzylated analogs. These
docking results guide further rational drug design of N-substituted
indolylacetic acid analogs as a novel PPAR agonist.
with serotonin-metabolites and indomethacin, which contain
indole acetate group as a common moiety, and fatty acid was
reported (PDB code: 3ADX).17 They provided two distinct sub-
pockets; one is namely AF-2 pocket adjacent to helix H12 for bind-
ing of indole acetate-containing ligands and the other is
X pocket
for the binding of fatty acid metabolites. We employed this crystal
structure of LBD complex with indole acetate-containing ligand
and fatty acid ligand for the docking analysis of our indole com-
pounds. Docking result indicated that indolylacetic acid moiety
occupied AF-2 pocket with the polar interaction and the lipophilic
tail occupied
X pocket (Fig 5).
In summary, we designed and synthesized carbazole or phenox-
azine containing 5- or 6-alkoxyindolyl acetic acid analogs as
PPARcd dual agonist. In general, 5-alkoxyindole analogs were more
active than the corresponding 6-alkoxyindole and phenoxazine
was more favorable for hydrophobic moiety than carbazole. We
propose two putative binding modes of our indole analogs by
The carboxyl groups of both compounds 3a and 3b interacted
with Gln286 and compound 3b additionally with His449 in AF-2
pocket and the hydrophobic tail phenoxazine occupied similarly
X
pocket where fatty acid bound. In comparison of non-benzylated
indole analog 3a to benzylated analog 3b, the NH of indole ring of

H. J. Gim et al. / Bioorg. Med. Chem. Lett. 23 (2013) 513–517
517
docking analysis employing two crystal structures of PPAR
crystallized with rosiglitazone or indole acetate-containing li-
gands. N-benzyl substitution at indole increased PPAR d activity
and docking analysis indicated that our indole analog might have
a different binding mode in LBD of PPAR as compared to the well
known PPAR lignad such as rosiglitazone. We are exploring the
further SAR studies and development of PPAR d agonists by utiliz-
c co-
References and notes
1. Issemann, I.; Green, S. Nature 1990, 347, 645.
2. Mangelsdorf, D. J.; Evans, R. M. Cell 1995, 83, 841.
3. Berger, J.; Moller, D. E. Annu. Rev. Med. 2002, 53, 409.
4. Lehmann, J. M.; Moore, L. B.; Smith-Oliver, T. A.; Wilkison, W. O.; Willson, T. M.;
Kliewer, S. A. J. Biol. Chem. 1995, 270, 12953.
c
c
c
5. Chaput, E.; Saladin, R.; Silvestre, M.; Edgar, A. D. Biochem. Biophys. Res. Commun.
2000, 271, 445.
c
ing alkoxyindolylacetic acid template. We suggest that alkoxyind-
olylacetic acid might be useful not only as chemical tools to study
PPAR function as well as development of a drug candidate for the
treatment of metabolic disease.
6. Nissen, S. E.; Wolski, K.; Topol, E. J. JAMA 2005, 294, 2581.
7. Oliver, W. R., Jr.; Shenk, J. L.; Snaith, M. R.; Russell, C. S.; Plunket, K. D.; Bodkin,
N. L.; Lewis, M. C.; Winegar, D. A.; Sznaidman, M. L.; Lambert, M. H.; Xu, H. E.;
Sternbach, D. D.; Kliewer, S. A.; Hansen, B. C.; Willson, T. M. Proc. Natl. Acad. Sci.
U.S.A. 2001, 98, 5306.
8. Luquet, S.; Gaudel, C.; Holst, D.; Lopez-Soriano, J.; Jehl-Pietri, C.; Fredenrich, A.;
Grimaldi, P. A. Biochim. Biophys. Acta 2005, 1740, 313.
Acknowledgments
9. Takahashi, S.; Tanaka, T.; Kodama, T.; Sakai, J. Pharmacol. Res. 2006, 53, 501.
10. Evans, J. L.; Lin, J. J.; Goldfine, I. D. Curr. Diabetes Rev. 2005, 1, 299.
11. Berger, J. P.; Akiyama, T. E.; Meinke, P. T. Trends Pharmacol. Sci. 2005, 26, 244.
12. Nissen, S. E.; Wolski, K. N. Eng. J. Med. 2007, 356, 2457.
13. Balakumar, P.; Rose, M.; Singh, M. Pharmacology 2007, 80, 1.
14. Gim, H. J.; Cheon, Y. J.; Ryu, J. H.; Jeon, R. Bioorg. Med. Chem. Lett. 2011, 21, 3057.
15. Fukuyama, T.; Chen, X.; Peng, G. J. Am. Chem. Soc. 1994, 116, 3127.
16. Nolte, R. T.; Wisely, G. B.; Westin, S.; Cobb, J. E.; Lambert, M. H.; Kurokawa, R.;
Rosenfeld, M. G.; Willson, T. M.; Glass, C. K.; Milburn, M. V. Nature 1998, 395,
137.
This work was supported by the National Research Foundation
of Korea Grant funded by the Korea government (Project No 2011-
0030699) and Sookmyung Women’s University Research Grants
(2010).
Supplementary data
17. Waku, T.; Shiraki, T.; Oyama, T.; Maebara, K.; Nakamori, R.; Morikawa, K. EMBO
J. 2010, 29, 3395.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.11.
033.