Design, synthesis, and structure-activity relationship of carbamate-tethered aryl propanoic acids as novel PPARα/γ dual agonists

Article abstract of DOI:10.1016/j.bmcl.2007.04.057

We have developed a new class of PPARα/γ dual agonists, which show excellent agonistic activity in PPARα/γ transactivation assay. In particular, (R)-9d was identified as a potent PPARα/γ dual agonist with EC₅₀s of 0.377 μM in PPARα and 0.136 μM in PPARγ, respectively. Interestingly, the structure-activity relationship revealed that the stereochemistry of the identified PPARα/γ dual agonists significantly affects their agonistic activities in PPARα than in PPARγ.

Full text of DOI:10.1016/j.bmcl.2007.04.057

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3595–3598
Design, synthesis, and structure–activity relationship
of carbamate-tethered aryl propanoic acids as novel
PPARa/c dual agonists
Nam-Jung Kim,a Kwang-Ok Lee,a Bon-Woong Koo,a Funan Li,a Ja-Kyung Yoo,b
Hyun-Ju Park,b Kyung-Hoon Min,b Joong In Lim,c Mi Kyung Kim,c
Jin-Kwan Kimc and Young-Ger Suha,*
aCollege of Pharmacy, Seoul National University, San 56-1, Sillim-dong, Gwanak-gu, Seoul 151-742, Republic of Korea
bCollege of Pharmacy, Sungkyunkwan University, Suwon-si, Gyeonggi-do 440-746, Republic of Korea
cDong-a Pharm.Co., Ltd, 47-5, Sanggal-dong, Giheung-gu, Yongin-si, Gyeonggi-do 449-905, Republic of Korea
Received 28 February 2007; revised 16 April 2007; accepted 18 April 2007
Available online 25 April 2007
Abstract—We have developed a new class of PPARa/c dual agonists, which show excellent agonistic activity in PPARa/c transac-
tivation assay. In particular, (R)-9d was identified as a potent PPARa/c dual agonist with EC50s of 0.377 lM in PPARa and
0.136 lM in PPARc, respectively. Interestingly, the structure–activity relationship revealed that the stereochemistry of the identified
PPARa/c dual agonists significantly affects their agonistic activities in PPARa than in PPARc.
Ó 2007 Elsevier Ltd. All rights reserved.
O
As members of the nuclear hormone receptor super fam-
ily of ligand-activated transcription factors, the Peroxi-
NH
N
N
S
some
proliferator-activated
receptors
(PPARs),
O
O
transduce a wide variety of signals to regulate glucose
and lipid homeostasis, sensitivity to insulin, cell prolifer-
ation/differentiation, and innate and adaptive immune
responses. Up to date, three isotypes, which are desig-
nated PPARa, PPARc, and PPARd, have been identi-
fied and cloned.1 Recently, the rapid progress in
functional analysis of these receptors prompted the sci-
entific and pharmaceutical attention to PPARs due to
their biological importance. In particular, the PPAR
agonists have intensively been studied as a therapeutic
target of type 2 diabetes. As rosiglitazone is a known
PPARc agonist as one of the representative insulin sen-
sitizing thiazolidinediones (Fig. 1),2 activation of
PPARc induces adipocyte differentiation, lipid uptake,
and insulin sensitization, while PPARa regulates lipid
homeostasis via control of fatty acid catabolism.3 The
PPARa agonists such as fibrate analogs have been used
to treat dyslipidemia by lowering the free triglyceride
rosiglitazone
O
N
CO2H
N
O
O
O
muraglitazar
OCH3
Figure 1. PPARc agonist (rosiglitazone) and PPARa/c dual agonist
(muraglitazar).
(TG) plasma concentration.4,5 Considering the signifi-
cantly increased risk of complications related to lipid
catabolism among the patients with type 2 diabetes, it
has been postulated that PPARa/c dual agonists, such
as muraglitazar, might provide outstanding therapeutic
effects for the treatment of type 2 diabetes and dyslipide-
mia (Fig. 1).6 We herein report identification of the
Keywords: Type 2 diabetes; PPARs; PPARa/c dual agonists.
*
Corresponding author. Tel.: +82 2 880 7875; fax: +82 2 888 0649;
e-mail: ygsuh@snu.ac.kr
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.057

3596
N.-J. Kim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3595–3598
novel and potent PPARa/c dual agonists as well as their
structure–activity relationship.
was synthesized from the commercially available 4-
hydroxybenzaldehyde 2. Sequential O-benzylation of
phenol and Wittig olefination of aldehyde, followed by
hydrogenation of the resulting olefin, provided 3, which
was readily converted to the acid 4 by benzylation and
ester hydrolysis. The racemic 4 was transformed into
the optically pure ester 5 via a sequence of (R)-2-phenyl-
glycinol coupling, separation of the resulting two diaste-
reomers, and amide hydrolysis, followed by ethyl ester
formation. The absolute configuration of the ester 5
was confirmed by comparison of its optical rotation
with that of the known compound.9 Triflation of 5,
Sonogashira coupling of the resulting triflate 6 with
the racemic or optically pure propargyl alcohol 7,10
and then hydrogenation of the triple bond afforded the
corresponding optically active or racemic intermediate
8 in optically pure form or as a mixture of two diastereo-
mers, respectively. The analogs 9a–h, 11 were prepared
by coupling of the corresponding intermediate 8 with
the appropriate isothiocyanate or isocyanate, followed
by ester hydrolysis, while the acid 10 was prepared by di-
rect hydrolysis of the hydroxyl ester 8. Our design and
synthesis of the analogs of 9a was focused on the thioc-
arbamate analogs rather than the carbamate because
our preliminary studies revealed that the thiocarbamates
are much more active than the carbamates for PPARa/c
as shown in the activities of 9a and 11 of Table 1. The
optically active thiocarbamates could also be prepared
by employing (S)- or (R)-7 by the same procedure.
The analog 14 was synthesized from propargyl alcohol
by analogy to 9a–h as depicted in Scheme 2. The thioc-
arbamate 18 was synthesized from commercially avail-
able racemic 1-phenyl-1, 2-ethanediol (15). Selective
tosylation of the primary alcohol,11 TBS protection of
the remaining secondary alcohol, and then alkylation
In order to efficiently develop the novel PPARa/c dual
agonists, we initially attempted virtual screening of our
chemical library for PPARc (PDB number: 1K74) and
then identified the compound 1 with weak PPARc ago-
nistic activity through a cell based transactivation assay
of the virtual hits. In addition, replacement of benzoic
acid moiety of 1 with b-aryl a-(S)-ethoxy propanoic
acid, a well-known moiety of high binding affinity to
PPARs,7 has led us to discover the lead compound 9a
(Fig. 2) as a potent PPARa/c dual agonist. Thus, we
commenced our studies on development of novel
PPARa/c dual agonists based on the structure of 9a as
well as their structure–activity relationship. The novel
series of 9a consisting of hydrogen bonding donor/
acceptor (A), linker (B), and hydrophobic part (C),8
which are the typical functional regions of the known
PPARa/c dual agonists, were designed and synthesized
for the detail investigation of each region.
The compounds 9a–h, 10, 11 were conveniently synthe-
sized as outlined in Scheme 1. The key intermediate 6
C
NH
NH
O
S
O
S
CO2H
OEt
CO2H
1
9a
B
A
Figure 2. Conversion of hit 1 to lead compound 9a.
CO2Et
OEt
CO2H
OEt
CO2Et
OEt
CHO
CO2Et
OEt
f,g,h
i
a, b, c
d, e
TfO
BnO
HO
HO
HO
2
3
4
5
6
R
9 a R=benzyl (24% from 8)
9 b R=phenyl (41% from 8)
NH
OH
9 c R=phenethyl (60% from 8)
O
S
j, k
l, n
9 d R=4-t-buytlbenzyl (22% from 8)
OEt
CO2Et
OEt
9 e R=ethyl (17% from 8)
9 f R=propyl (83% from 8)
OH
8
9a-h
CO2H 9 g R=n-butyl (8% from 8)
9 h R=cyclohexylmethyl (25% from 8)
7
OH
m, n
OH
n
NH
(S) -
7
(R) - 7
OH
O
O
OEt
CO2H
OEt
CO2H
10 (100%)
11 (82%)
Scheme 1. Reagents and conditions: (a) benzylchloride, K2CO3, DMF, 80 °C, 93%; (b) (1,2-diethoxy-2-oxyethyl)triphenylphosphoniumchloride,
tetramethylguanidine, CH2Cl2, 0 °C to rt, 93%; (c) H2, Pd/C, MeOH, 95%; (d) benzylchloride, K2CO3, DMF, 120 °C, 75%; (e) LiOHÆH2O, MeOH/
THF/H2O, 94%; (f) EDCI, HOBT, DIEA, CH2Cl2, then (R)-phenylglycinol, 0 °C to rt, 30% (desired diastereomer); (g) concd-H2SO4, H2O/1,4-
dioxane, reflux; (h) TMSCl, EtOH, 45% for two steps; (i) Tf2O, Et3N, CH2Cl2, 0 °C to rt, 92%; (j) (S)- or (R)-7 for 9a, 9c, and 9d, and 7 for 9b, 9e–h,
10, 11, Pd(PPh3)2Cl2, CuI, Et3N, TBAI, DMF, 50 °C, 83%; (k) H2, Pd/C, MeOH, 100%; (l) NaH, alkyl or arylisothiocyanate, THF; (m) BF3ÆOEt2,
benzylisocyanate, toluene; (n) LiOHÆH2O, MeOH/THF/H2O.

N.-J. Kim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3595–3598
3597
Table 1. In vitro activities of the synthesized compounds in cell-based
GAL4-PPARa/c transactivation assay
PPARc with EC50 of 0.14 lM. These results might im-
ply that the bulky substituents in the lipophilic region
are beneficial for the high PPARc activation and the
similar results were observed for the alkyl thiocarba-
mates as the analog 9h with cyclohexylmethyl substitu-
ent exhibited the most potent activity for PPARc. Our
preliminary work revealed that the thiocarbamates are
much more active than the carbamates for PPARa/c,
as it was confirmed by the activities of 9a (EC50: 0.26–
0.58 lM) and 11 (EC50: 1.42–6.50 lM). Replacement
of the carbon linker with an ether linkage decreased
the activity as shown in the activities of 9a and 18.
Importantly, the agonistic activity of the thiocarbamate
analogs sensitively changes according to the stereochem-
istry at the benzylic carbon, particularly for PPARa.
The (S)-isomer of 9a was more potent than the corre-
sponding (R)-isomer for PPARa, while both exhibited
the similar activity for PPARc. The (S)-isomer of 9c is
slightly more potent than the (R)-isomer of 9c in both
PPARa and c. In case of the compound 9d showing
the most potent activity for PPARc, the (R)-isomer of
9d was 10-fold more potent than the (S)-isomer for
PPARa. The docking studies with Surflex–DockTM,13
shown in Fig. 3, provided a rationale in part for the ste-
reochemistry/substituent-dependent selectivity for the
PPAR isotypes. The benzylthiocarbamoyl moiety of
(S)-9a seems to fit well to the lipophilic region of the ac-
tive site of PPARa, while the corresponding moiety of
(R)-9a orients toward the hydrophilic region instead of
the lipophilic region, which is enforced by the stereo-
chemistry. In addition, the calculated binding energies
revealed the lower binding energy for (S)-9a compared
to that for (R)-9a. However, p-tert-butyl benzyl group
of 9d appears to be too bulky to fit to the lipophilic
pocket of PPARa in both (R)- and (S)-isomers and sub-
sequently the bulky lipophilic moiety to orient toward
the spacious hydrophilic pocket. As the results, the
activities of (R)-9a, (S)-9d, and (R)-9d seem to be re-
duced by the inadequate orientations of their lipophilic
moiety. Removal of the phenyl moiety of the B-region
of 9a significantly drops the activity for both PPARa
and PPARc as shown in the activities of 14. This would
be understood in terms of formation of the favorable
conformation for the high activity, which is induced
Compounda
Human PPARa
EC50
Efficacyb
(lM)
3.46
Human PPARc
EC50
Efficacyb
(lM)
0.03
(%)
(%)
Rosiglitazone
56
100
101
234
154
112
120
71
100
79
Gemfibrozil
(R)-9a
(S)-9a
9b
193.30
0.26
0.07
5.25
2.77
0.88
0.38
3.27
20.40
3.94
7.26
10.70
1.51
6.05
1.47
6.21
147.80
0.35
0.58
4.58
1.05
0.49
0.14
0.16
5.39
3.48
3.60
1.06
0.30
1.42
1.40
7.17
113
159
204
190
177
197
187
182
152
118
145
119
160
117
138
(R)-9c
(S)-9c
(R)-9d
(S)-9d
9e
66
130
128
104
117
108
164
118
136
9f
9g
9h
10
11
14
18
a (R) or (S)-9a, 9c and 9d as well as 14 are optically active, while 9b, 9e–
h, 10, 11, and 18 are 1:1 diastereomeric mixtures.
b The relative maximum efficacy to the percentage of the standards.
of the intermediate 5 with the resulting tosylate 16 gave
the ether 17 after desilylation. Reaction of 17 with ben-
zylisothiocyanate and ester hydrolysis provided 18.
The in vitro activities of the synthesized analogs are
summarized in Table 1.12 Generally, the aryl thiocarba-
mates (9a, 9c, and 9d) exhibited higher agonistic activity
than the alkyl thiocarbamates (9e–h) in both PPARa
and PPARc transactivation assays. In particular, the
analog 9a possessing benzyl thiocarbamate moiety dis-
played potent activities for PPARa/c regardless of the
stereochemistry. However, it decreased as the chain
length of C-region decreased (9b) or increased (9c).
Interestingly, the similar trend was observed for the al-
kyl thiocarbamates and the propyl turned out to be a
linker of optimal length. It is noteworthy that (R)-9d
possessing the bulkiest substituent in the lipophilic moi-
ety (C-region) showed the most potent activity for
S
HO
OEt
a,b
c, d
N
O
OEt
HO
H
OEt
OH
O
O
12
13
14
NH
OH
OTBS
OTs
OH
e, f
g, h
O
i, j
O
S
OH
OEt
CO2Et
O
OEt
CO2H
15
16
17
18
Scheme 2. Reagents and conditions: (a) 6, Pd(PPh3)2Cl2, CuI, Et3N, TBAI, DMF, 60 °C, 45%; (b) H2, Pd/C, MeOH, 76 %; (c) NaH,
benzylisothiocyanate, THF, 0 °C to rt; (d) LiOHÆH2O, MeOH/THF/H2O, 65% for two steps; (e) Bu2SnO, p-toluenesulfonylchloride, Et3N, CH2Cl2,
0 °C to rt, 94%; (f) TBSCl, imidazole, DMAP, DMF, 65 °C, 85%; (g) 5, K2CO3, DMF, 130 °C, (h) 2 N-HCl, THF, 50 °C, 40% for two steps; (i) NaH,
benzylisothiocyanate, THF; (j) LiOHÆH2O, MeOH/THF/ H2O, 60% for two steps.

3598
N.-J. Kim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3595–3598
References and notes
1. Dreyer, C.; Krey, G.; Keller, H.; Givel, F.; Helftenbein,
G.; Wahli, W. Cell 1992, 68, 879.
2. Lehmann, J. M.; Moore, L. B.; Smith-Oliver, T. A.;
Wilkinson, W. O.; Willson, T. M.; Kliewer, S. A. J. Biol.
Chem. 1995, 270, 12953.
3. Staels, B.; Auwerx, J. Curr. Pharm. Des. 1997, 3, 1.
4. Chong, P. H.; Bachenheimer, B. S. Drugs 2000, 60, 55.
5. Milionis, H. J.; Elisaf, M. S.; Mikhailidis, D. P. Curr.
Med. Res. Opin. 2000, 16, 21.
6. Devasthale, P. V.; Chen, S.; Jeon, Y.; Qu, F.; Shao, C.;
Wang, W.; Zhang, H.; Farrelly, D.; Golla, R.; Grover, G.;
Harrity, T.; Ma, Z.; Moore, L.; Ren, J.; Seethala, R.;
Cheng, L.; Sleph, P.; Sun, W.; Tieman, A.; Wetterau, J.
R.; Doweyko, A.; Chandrasena, G.; Chang, S. Y.;
Humphreys, W. G.; Sasseville, V. G.; Biller, S. A.; Ryono,
D. E.; Selan, F.; Hariharan, N.; Cheng, P. T. W. J. Med.
Chem. 2005, 48, 2248.
7. Buckle, D. R.; Cantello, B. C. C.; Cawthorne, M. A.;
Coyle, P. J.; Dean, D. K.; Faller, A.; Haigh, D.; Hindley,
R. M.; Jefcott, L. J.; Lister, C. A.; Pinto, I. L.; Rami, H.
K.; Smith, S. A. Bioorg. Med. Chem. Lett. 1996, 6, 2127.
8. Wei, Z.; Petukhov, P. A.; Bizik, F.; Teixeira, J. C.;
Mercola, M.; Volpe, E. A.; Glazer, R. I.; Willson, T. M.;
Kozikowski, A. P. J. Am. Chem. Soc. 2004, 126, 16714.
9. Ebdrup, S.; Pettersson, I.; Rasmussen, H. B.; Deussen, H.-
J.; Jensen, A. F.; Mortensen, S. B.; Fleckner, J.; Pridal, L.;
Nygaard, L.; Sauerberg, P. J. Med. Chem. 2003, 46, 1306.
10. Powell, N. A.; Rychnovsky, S. D. Tetrahedron Lett. 1996,
37, 7901.
Figure 3. Docking model of (S)-9a (white), (R)-9a (magenta), (S)-9d
(cyan), (R)-9d (orange), and the co-crystallized ligand as a reference
molecule (yellow) with PPARa (PDB No. 1K7L). LP, lipophilicity;
dotted line, hydrogen bonding. Note. *Log value (kcal/mol).
by the phenyl substituent. On the other hand, elimina-
tion of the thiocarbamate moiety of the thiocarbamate
9a significantly reduced the activity for PPARa as
shown in the activity of the alcohol 10 while the activity
for PPARc remained. This result obviously supports
that the thiocarbamate moiety is essential for the high
activity of our series for PPARa.
11. Martinelli, M. J.; Vaidyanathan, R.; Pawlak, J. M.;
Nayyar, N. K.; Dhokte, U. P.; Doeke, C. W.; Zollars,
ˇ
L. M. H.; Moher, E. D.; Khau, V. V.; Kosmrlj, B. J. Am.
Chem. Soc. 2002, 124, 3578.
12. In vitro transactivation assays: the ligand binding domains
(LBD) of the human PPARa/c receptors were fused to the
DNA binding domain (DBD) of the yeast transcription
factor GAL4. CV-1 cells were transiently transfected with
an expression vector for the respective PPAR chimera
along with a reporter construct containing five copies of
the GAL4DNA binding site and pRL-TK as a control
vector (Promega). The test compounds were dissolved in
DMSO and diluted 1:1000 in media. Cells were treated
with the compounds at seven concentrations ranging from
0.03 to 100 lM for 24 h followed by dual luciferase assay
using Dual-Glo luciferase reagent (promega). EC50 values
were calculated by nonlinear regression using SigmaPlot
4.0 (SPSS).
13. The X-ray structures of PPARa and PPARc complexed
with the ligand were taken from the RCSB Protein Data
Bank (PDB code: 1K7L for PPARa/1K74 for PPARc).
The protein was prepared for docking, using the protein
preparation and refinement utility provided by Surflex–
DockTM suite in SYBYL 7.0. Water molecules of the crystal
structure were removed from the complexes, and the
hydrogen atoms were added computationally at appro-
priate positions. Calculations for the docking studies
between the prepared PPARa protein and the selected
compounds were performed using SYBYL 7.0 software.
In summary, we have identified the novel and potent
PPARa/c dual agonists and established SAR of the car-
bamate-tethered propanoic acids through the computa-
tional and synthetic chemistry. The stereochemistry/
substituent-dependent activation of the particular iso-
type of PPARs by our carbamate series was elucidated
on the basis of the binding mode of the ligands in the ac-
tive site. Further work for the development of the ther-
apeutically useful PPARa/c dual agonists based on our
current results is in good progress.
Acknowledgments
This research work was supported by the grant from R&
D Center for Antidiabetic Drugs, by Ministry of Health
& Welfare, Republic of Korea (A020600), and in part
from Center for Bioactive Molecular Hybrids, Yonsei
University.