Novel biphenylcarboxylic acid peroxisome proliferator-activated receptor (PPAR) δ selective antagonists

Article abstract of DOI:10.1016/j.bmcl.2009.10.021

We designed and synthesized novel PPARδ antagonists based on the crystal structure of the PPARδ full agonist TIPP-204 bound to the PPARδ ligand-binding domain, in combination with our nuclear receptor helix 12 folding modification hypothesis. Representati

Full text of DOI:10.1016/j.bmcl.2009.10.021

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6595–6599
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel biphenylcarboxylic acid peroxisome proliferator-activated receptor
(PPAR) d selective antagonists
Jun-ichi Kasuga a, Seiichi Ishida b, Daisuke Yamasaki c, Makoto Makishima d, Takefumi Doi c,
Yuichi Hashimoto a, Hiroyuki Miyachi e,
*
a Institute of Molecular and Cellular Biosciences, The University of Tokyo, 1-1-1, Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan
b Division of Pharmacology, National Institute of Health Sciences, 1-18-1, Kamiyoga, Setagaya-ku, Tokyo 158-8501, Japan
c Graduate School of Pharmaceutical Sciences, Osaka University. 1-6 Yamadaoka, Suita city, Osaka 565-0871, Japan
d Division of Biochemistry, Nihon University School of Medicine, 30-1 Oyaguchi-kamicho, Itabashi-ku, Tokyo 173-8610, Japan
e Division of Pharmaceutical Sciences, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 1-1-1, Tsushima-Naka, Kita-ku,
Okayama 700-8530, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We designed and synthesized novel PPARd antagonists based on the crystal structure of the PPARd full
agonist TIPP-204 bound to the PPARd ligand-binding domain, in combination with our nuclear receptor
helix 12 folding modification hypothesis. Representative compound 3a exhibits PPARd-preferential
antagonistic activity.
Received 16 September 2009
Revised 4 October 2009
Accepted 6 October 2009
Available online 8 October 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
PPAR
PPARd
PPARd antagonist
Biphenylcarboxylic acid
The peroxisome proliferator-activated receptors (PPARs) are li-
gand-dependent transcription factors belonging to the nuclear
ipidemia. In contrast, there was initially limited interest in PPARd,
probablyduetoitsubiquitousdistribution. However, theavailability
of PPARd-knockout animals and selective ligands, especially GW-
501516, developed by GlaxoSmithKline, prompted us to examine
the involvement of PPARd in fatty acid metabolism, insulin resis-
tance, reverse cholesterol transport and other biological path-
ways.6,7 Recently, interest in PPARd biology and/or pharmacology
has increased enormously, and the therapeutic potential for PPARd
ligands now appears to extend well beyond lipid, lipoprotein and
glucose homeostasis. For example, PPARd is reported to play a criti-
cal role in wound healing. After tissue damage resulting from chem-
ical, mechanical, or biological injury, the injured cells release
proinflammatory cytokines.8,9 These stimulate PPARd expression,
coordinating transcriptional up-regulation of integrin-linked kinase
and 3-phosphoinositide-dependent kinase (PDK), and repressing
the expression of phosphatase and tensin homolog 10 (PTEN).10 As
receptor (NR) superfamily. The three subtypes (PPAR
a, PPARd,
and PPAR ) identified to date are differentially expressed in a tis-
c
sue-specific manner, and play pivotal roles in lipid, lipoprotein,
and glucose homeostasis.1
PPARa
is mostly expressed in tissues involved in lipid oxidation,
is expressed in adipose tis-
such as liver, heart, and kidney.2 PPAR
c
sue, macrophages and vascular smooth muscles.3 In contrast to the
specific distribution of the other two PPAR subtypes, PPARd is ex-
pressed ubiquitously, though it is mainly found in skeletal muscle
and adipose tissues.4 PPARs function by heterodimerization with
another cognate nuclear receptor partner, retinoid X receptor
(RXR), and the heterodimers regulate gene expression by binding
to a specific consensus DNA sequence, termed PPRE (peroxisome
proliferator responsive element),5 located in the promoter region
of target genes.
a consequence, the protein kinase alpha (PKBa) activity, which
Based onthe findingsthat the antidiabetic glitazones, and antidy-
slipidemic fibrates are ligands of PPARc and PPARa, respectively,
great research interest has been focused on these two PPAR subtypes
as therapeutic targets for the treatment of type II diabetes and dysl-
implicated as a controller of apoptosis, is increased and apoptotic
cascades are repressed. The resulting increased resistance to cell
death helps to maintain a sufficient number of viable wound kerat-
inocytes for re-epithelization. Therefore, PPARd is expected to be a
therapeutic target for treatment of tissue injury.
This example clearly indicates that not only PPAR
but also PPARd, have enormous potential as therapeutic agents,
c and PPARa,
* Corresponding author.
E-mail address: miyachi@pharm.okayama-u.ac.jp (H. Miyachi).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.10.021

6596
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6595–6599
O
F
O
O
H
N
F
O
O
COOMe
S
N
H
OH
O
OH
N
H
S
F
F
N
F
F
O
O
H
OMe
F
F
3a-l
1; GSK0660
2; TIPP-204
newly designed PPARδ ligand
Figure 1. Structures of PPARd-selective antagonist GSK0660 (1), our PPARd-selective agonist TIPP-204 (2), and the general formula of the present series of compounds (3a–l).
Figure 2. (A) An overview of the crystal structure of the human PPARd ligand-binding domain (green) complexed with PPARd-selective agonist 2. Helices 1 to 12 are indicated
and the functionally important Helix 12 is illustrated in red. 2 is shown in van der Waals representation. (B) Zoomed view of the interaction between the n-butoxy moiety of 2
and the binding cavity formed by leucine, valine, and isoleucine. Side chains of these three amino acids are shown in van der Waals representation. (C) Zoomed view of the
interaction between the acidic head of 2 and the surrounding hydrogen bond forming amino acids, histidine, serine, and tyrosine. Side chains of these three amino acids are
shown in van der Waals representation.
and the range of possible applications has certainly not yet been
fully explored.
(Fig. 2c). We consider that this network of interactions effectively
locks the receptor into an active conformation permissive for coac-
tivator interactions that promote gene expression.
In order to fully understand the biology and/or pharmacology of
PPARd, both agonist and antagonist ligands are needed as investi-
gative tools. But, most medicinal chemistry research on PPARd
has focused on the discovery of agonists, and few antagonists have
been described. As far as we know, the only PPARd-selective antag-
onist currently available is GSK0660 (1), a 2-methoxycarbonylthi-
ophene derivative, which was identified via high-throughput
screening.11 Although 1 is a valuable chemical tool in vitro, its lack
of in vivo bioavailability limits its usefulness (Fig. 1).
In this Letter, we would like to present our novel PPARd-selec-
tive antagonists. In designing novel PPARd-selective antagonists,
we focused on our previously reported PPARd-selective agonist
TIPP-204 (compound 2), which exhibited extremely potent PPARd
transactivation activity, comparable to that of the known PPARd-
selective full agonist, GW-501516.12 2 is a potent PPARd-selective
full agonist (EC50s of the transactivation activity against PPARd,
Previously, we have reported that nuclear receptor antagonists
can be designed based on the hypothesis that the proper folding of
helix 12, an LBD substructure, is critical for nuclear receptors to
form active conformation, that is, for ligand-dependent activation.
Blocking the proper folding of helix 12 is therefore the key to
antagonist design. Based on this nuclear receptor helix 12 folding
modification hypothesis, we have designed and synthesized a
range of nuclear receptor antagonists, including retinoic acid
receptor (RAR) antagonists, retinoid X receptor (RXR) antagonists,
and farnesoid X receptor (FXR) antagonists.14
In order to create structurally new PPARd antagonists, we fo-
cused on the a-ethyl phenylpropanoic acid structure of 2, since this
moiety might be critical for forming the active conformation of
PPARd LBD, and suitable structural manipulation might decrease
the ability of the ligand-bound PPARd to form transcriptionally ac-
tive conformation. We therefore adopted a biphenylcarboxylic acid
framework as a candidate antagonist template, anticipating that
the conformationally restricted biphenyl structure would interfere
with the appropriate positioning of helix 12 of PPARd, thereby
resulting in partial agonist and/or antagonist activity.
Synthesis of biphenylcarboxylic acids 3a–l is depicted in
Scheme 1. 4-n-Butoxybromobenzene (compound 4) was formylat-
ed, followed by reductive amide alkylation to afford 6. Compound 6
was coupled with the appropriate boronic acids by means of Suzu-
ki coupling reaction to afford 3a, 3b, 8, and 9, or with the appropri-
PPARa and PPARc are 0.72 nM, 240 nM, and 1400 nM, respec-
tively), and the crystal structure of its complex with the human
PPARd ligand-binding domain (LBD) at 2.7 Å resolution has been
reported (Fig. 2a).13 The PPARd-selectivity of 2 is mainly attributed
to the chain length of the side chain alkoxy group at the center of
the ligand molecule. In the complex, the longer n-butoxy group of
2 has a hydrophobic interaction with the cavity formed by Val334,
L339, and I364 (Fig. 2b). Further, the carboxylate group of 2 was
found to form an intricate network of hydrogen bonds with histi-
dine, tyrosine, and serine residues in helices 3, 4, 11, and 12

J. Kasuga et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6595–6599
6597
O
O
F
O
F
O
Br
Br
Br
Ar
H
N
H
N
H
(a)
(b)
(d) or (e)
O
F
F
F
F
O
O
F
F
4
5
6
3a;
3b;
Ar = 4-COOH-Ph
Ar = 3-COOH-Ph
8; Ar = 2-COOMe-Ph
3c; Ar = 2-COOH-Ph
(f)
(g)
(h)
9; Ar = 5-CHO-furan-2-yl
3d;
Ar = 5-COOH-furan-2-yl
Ar = 2-Me-4-COOMe-Ph
Ar = 2-Me-4-COOH-Ph
(c)
10;
3h;
F
O
F
O
O
B
Ar
N
H
O
(i)
N
H
F
F
F
F
O
O
7
F
F
3e; Ar = 5-COOH-pyridin-2-yl
3f;
3g;
3i;
Ar = 6-COOH-pyridin-3-yl
Ar = 2-F-4-COOH-Ph
Ar = 3-Me-4-COOH-Ph
Ar = 2-Me-3-COOH-Ph
3j;
3k; Ar = 4-Me-3-COOH-Ph
3l; Ar = 2-Me-5-COOH-Ph
Scheme 1. Synthetic routes to the present series of compounds, 3a–l: Reagents and conditions: (a) TiCl4, CHCl2OCH3, DCM, À78 °C, 99%; (b) 2-F-4-CF3-benzamide, Et3SiH,
TFA, toluene, reflux, 61%; (c) bis(pinacolato)diboron, PdCl2(PPh3)2, KOAc, dioxane, 100 °C, 87%; (d) the appropriate boronic acid, PdCl2(PPh3)2, aq Na2CO3, DME, ethanol, 60 °C,
28–93%; (e) the appropriate boronic acid pinacol ester, Pd(OAc)2, K3PO4, X-Phos, THF, rt, 95%; (f) aq NaOH, dioxane, 70 °C, 83%; (g) 2-methyl-2-butene, NaClO2, NaH2PO4,
t-BuOH, H2O, 80 °C, 84%; (h) 6 mol/L HCl, AcOH, 100 °C, 96%; (d) the appropriate aryl bromide, PdCl2(PPh3)2, aq Na2CO3, DME, ethanol, 60 °C, 28–93%.
ate boronic acid pinacol ester to afford 10. Compounds 8 and 10
were hydrolyzed to afford 3c and 3h, while 9 was oxidized to af-
ford 3d. Compound 6 was treated with bis(pinacolato)diboron to
afford 7, which was coupled with the appropriate aryl bromides
by means of Suzuki coupling reaction to afford 3e–g, and 3i–l.
Compounds 3a–l were initially screened in a cell-based GAL4-PPAR
chimera receptor agonist assay system.
The chemical structures of 3a–l and the agonistic activities to-
wards PPARs are shown in Figure 3. First, we synthesized biphenyl
compounds 3a, 3b, and 3c to investigate the effect of the position
F
O
Ar
N
H
F
O
F
F
EC50 for PPARs (%efficacy)
EC50 for PPARs (%efficacy)
Compound: Ar
Compound: Ar
PPARα
PPARδ
PPARγ
PPARα
PPARδ
PPARγ
COOH
COOH
COOH
COOH
N.A.a
N.A.a
N.A.a
N.A.a
53 nM (11%)
3g :
3a:
170 nM (8%)
F
N.A.a
N.A.a
11 nM (9%)
45 nM (5%)
3h :
N.A.a
N.A.a
3b:
3c:
790 nM (100%) 29 nM (48%)
Me
COOH
Me
N.A.a
N.A.a
3i :
3000 nM (100%)
N.A.a
N.A.a
COOH
COOH
COOH
3j :
130 nM (100%) 1.6 nM (40%)
3d :
N.A.a
N.A.a
N.A.a
N.A.a
N.A.a
N.A.a
N.A.a
N.A.a
COOH
O
Me
Me
3k :
14 nM (68%)
4.9 nM (59%)
N.A.a
N.A.a
120 nM (100%)
570 nM (100%)
3e :
COOH
N
Me
COOH
3l :
400 nM (9%)
3f :
N
COOH
Figure 3. PPARs transactivation activities of the compounds. EC50 values and % efficacy relative to the positive control, GW501516 for PPARd, fenofibric acid for PPAR
a, and
ciglitazone for PPAR
antagonist.
c, are given. In this assay, HEK293 cells transfected with GAL4-PPAR chimeric receptor were used. (a) N.A. means not active at 1 lM as an agonist or

6598
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6595–6599
of the carboxylic acid moiety on the PPARs agonistic activity. As ex-
pected, all the active compounds exhibited PPARs agonistic activ-
ity, with preference for PPARd. Interestingly, the PPAR activity
profile was highly dependent on the position of the carboxylic acid
moiety. The 3-biphenylcarboxylic acid derivative (3b) exhibited
potent partial PPARd agonistic activity (EC50 = 29 nM, efficacy of
48% compared to full agonist GW501516). The 4-biphenylcarboxy-
lic acid derivative (3a) exhibited very weak PPARd agonistic activ-
ity (EC50 = 170 nM, efficacy of 8%), while the 2-biphenylcarboxylic
acid derivative (3c) lacked apparent PPARd agonistic activity even
at the concentration of 10
tic activity towards PPAR
10 M. These results indicated that the small modification of the
-ethyl phenylpropanoic acid structure of 2 to biphenylcarboxylic
l
M. Neither 3a nor 3c exhibited agonis-
a
, and PPAR at concentrations of up to
c
l
a
acid resulted in retention of PPARd-selectivity, when the COOH
group was present at the 3- or 4-position. The efficacy for PPARd
decreased in the order of 3-biphenylcarboxylic acid >4-biphenyl-
carboxylic acid >2-biphenylcarboxylic acid. The dose–response
relationships of 3a and 3b, as well as GW501516, are illustrated
in Figure 4.
Based on the above results, we next designed and synthesized
heterocycle-containing compounds 3d–f and 3-fluoro- or 3-
methyl-substituent-containing compounds 3g and 3h. The 2-pyri-
dine derivative 3f exhibited somewhat decreased activity, while
efficacy was maintained compared with 3a, but the furan deriva-
tive 3d and 3-pyridine derivative 3e did not exhibit activity at
Figure 4. Concentration-dependent activation of PPARd by compounds 3a, 3b, and
GW501516.
pounds 3i–l were designed and synthesized. As expected, 3j–l
exhibited more potent activity than the corresponding 3-biphenyl-
carboxylic acid derivative 3b, while retaining partial PPARd agonis-
tic activity, and 3i exhibited more potent activity than the
corresponding 4-biphenylcarboxylic acid derivative 3a.
Considering the very low efficacy of 4-biphenylcarboxylic acid
derivatives, we then performed a cell-based GAL4-PPAR chimera
receptor antagonist assay of the representative 4-biphenylcarb-
oxylic acid derivative 3a, adding 100 nM PPARd-selective
the concentration of 10 lM, although the reason for this is not
yet clear. On the other hand, 3g and 3h exhibited more potent
activity than 3a (EC50 values for PPARd; 3a 170 nM, 3g 53 nM, 3h
11 nM), though efficacy was very low. Since the introduction of a
methyl group increased the activity, other methyl-containing com-
GW501516 or 100 nM PPARadc
-pan full agonist TIPP-70315 to
the cells, followed by 3a. In this assay, 3a antagonized only the
PPARd activity elicited by GW501516; its effect was dose-depen-
(b) PPAR alpha
(a) PPAR delta
10000
7500
5000
2500
0
10000
7500
5000
2500
0
(c) PPAR gamma
15000
10000
5000
0
Figure 5. PPARd-selective antagonistic activity of 3a. Gal4-PPARs chimeras were treated with the positive control, with or without the addition of 0.1 lM, 1 lM, or 10 lM 3a.
6
5
4
3
2
1
0
3
2
1
0
CPT1A
HMGCS2
DMSO
GW
GW+3a
DMSO
GW
GW+3a
Figure 6. Gene repression profiles of 3a. Huh-7 cells were treated with 100 nM GW501516, with or without the addition of 1
lM 3a, and the expression of PPARd target genes
was determined.

J. Kasuga et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6595–6599
6599
dent in the range from 0.1
culated to be 68 nM. It is interesting to note that no apparent dose-
dependent antagonistic activity was seen in the case of PPAR or
PPAR activity elicited by TIPP-703 at up to 10 M (Fig. 5). This
l
M to 10
l
M, and the IC50 value was cal-
genes induced by 100 nM GW501516 to a level similar to that in
the vehicle control. These results indicate that the representive
compound 3a is an effective PPARd antagonist and can repress
these PPARd-regulated genes at the cellular level.
a
c
l
PPARd-selective antagonistic activity of 3a led us to speculate that
the specific interaction of the hydrophobic tail part of 3a might be
efficiently retained, even though the acidic head structure had
In summary, we have developed a novel PPARd-selective antag-
onist with 4-biphenylcarboxylic acid structure. We are currently
investigating the X-ray crystal structure of the complex of 4-biphe-
nylcarboxylic acid PPARd-selective antagonist with the PPARd LBD,
and also examining the mRNA repression profile in detail.
been changed from
a-ethyl phenylpropanoic acid structure (2) to
biphenylcarboxylic acid structure. The PPARd antagonistic activity
of 3c, 3d, 3e, and 3h was also assayed. Compounds 3c, 3d, and 3e
did not exhibit apparent antagonistic activity, while 3h exhibited
fairly potent antagonistic activity (IC50 = 10 nM).
References and notes
1. Gross, B.; Staels, B. Best Pract. Res. Clin. Endocrinol. Metab. 2007, 21, 687.
2. Mukherjee, R.; Jow, L.; Noonan, D.; McDonnell, D. P. J. Steroid Biochem. Mol. Biol.
1994, 51, 157.
3. Okuno, A.; Tamemoto, H.; Tobe, K.; Ueki, K.; Mori, Y.; Iwamoto, K.; Umesono,
K.; Akanuma, Y.; Fujiwara, T.; Horikoshi, H.; Yazaki, Y.; Kadowaki, T. J. Clin.
Invest. 1998, 101, 1354.
4. Braissant, O.; Wahli, W. Endocrinology 1998, 39, 2748.
5. Osada, S.; Tsukamoto, T.; Takiguchi, M.; Mori, M.; Osumi, T. Genes Cells 1997, 2,
315.
6. Tanaka, T.; Yamamoto, J.; Iwasaki, S.; Asaba, H.; Hamura, H.; Ikeda, Y.;
Watanabe, M.; Magoori, K.; Ioka, R. X.; Tachibana, K.; Watanabe, Y.; Uchiyama,
Y.; Sumi, K.; Iguchi, H.; Ito, S.; Doi, T.; Hamakubo, T.; Naito, M.; Auwerx, J.;
Yanagisawa, M.; Kodama, T.; Sakai, J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
15924.
7. Lee, C. H.; Olson, P.; Hevener, A.; Mehl, I.; Chong, L. W.; Olefsky, J. M.; Gonzalez,
F. J.; Ham, J.; Kang, H.; Peters, J. M.; Evans, R. M. Proc. Natl. Acad. Sci. U.S.A. 2006,
103, 3444.
8. Tan, N. S.; Michalik, L.; Desvergne, B.; Wahli, W. J. Biol. Chem. 2005, 280, 18163.
9. Tan, N. S.; Michalik, L.; Di-Po, N.; Ng, C. Y.; Mermod, N.; Roberts, A. B.;
Desvergne, B.; Wahli, W. EMBO J. 2004, 23, 4211.
10. Burdick, A. D.; Bility, M. T.; Girroir, E. E. Cell. Signalling 2007, 19, 1163.
11. Shearer, B. G.; Steger, D. J.; Way, J. M.; Stanley, T. B.; Lobe, D. C.; Grillot, D. A.;
Iannone, M. A.; Lazar, M. A.; Willson, T. M.; Billin, A. N. Mol. Endocrinol. 2008,
22, 523.
12. Kasuga, J.; Nakagome, I.; Aoyama, A.; Sako, K.; Ishizawa, M.; Ogura, M.;
Makishima, M.; Hirono, S.; Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2007,
15, 5177.
To further characterize the 4-biphenylcarboxylic acid derivative
as a PPARd-selective antagonist, we examined its repressive effect
on representative PPARd-responsive genes having a peroxisome
proliferator responsive element (PPRE) in the promoter region at
the cellular level, using human hepatocellular carcinoma Huh-7
cells (Fig. 6). Carnitine palmitoyl acyl-CoA transferase 1A (CPT1A),
and HMG-CoA synthase 2 (HMGCS2) were selected for monitoring,
as the human genes were reported to possess PPRE in the promoter
region and to be regulated by PPARd.15 CPT1A is the key enzyme in
carnitine-dependent transport across the mitochondrial inner
membrane and deficiency results in a decreased rate of fatty acid
b-oxidation. HMGCS2 is a potential regulatory enzyme in the path-
way that converts acetyl-CoA to ketone bodies. This enzyme con-
denses acetyl-CoA with acetoacetyl-CoA to form HMG-CoA,
which is the substrate for HMG-CoA reductase. Defects in HMGCS2
cause HMG-CoA synthase deficiency, which leads to severe
hypoketotic hypoglycemia, mild hepatomegaly, or fatty liver.
We first investigated the effects of GW501516. As indicated in
Figure 6, when Huh-7 cells were treated with 100 nM
GW501516, a sufficient concentration to induce subtype-selective
transactivation activity, expression of the two mRNAs was aug-
mented. These results indicated that PPARd is functionally active
in Huh-7.
13. Oyama, T.; Toyota, K.; Waku, T.; Hirakawa, Y.; Nagasawa, N.; Kasuga, J.;
Hashimoto, Y.; Miyachi, H.; Morikawa, K. Acta Crystallogr., Sect. D 2009, 65, 786.
14. Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2005, 13, 5080.
15. Kasuga, J.; Yamasaki, D.; Ogura, K.; Shimizu, M.; Sato, M.; Makishima, M.; Doi,
T. Y.; Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. Lett. 2008, 18, 1110.
Treatment with 1
lM 3a, a concentration sufficient to inhibit
PPARd, attenuated the expression of both CPT1A and HMGCS2