Design and synthesis of highly potent and selective human peroxisome proliferator-activated receptor α agonists

Article abstract of DOI:10.1016/j.bmcl.2007.05.066

A combination of benzoxazole, phenoxyalkyl side chain, and phenoxybutyric acids was identified as a highly potent and selective human peroxisome proliferator-activated receptor α (PPARα) agonist. The synthesis, structure-activity relationship (SAR) studies, and in vivo activities of the representative compounds are described.

Full text of DOI:10.1016/j.bmcl.2007.05.066

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4689–4693
Design and synthesis of highly potent and selective
human peroxisome proliferator-activated receptor a agonists
Yukiyoshi Yamazaki,a,* Kazutoyo Abe,a Tsutomu Toma,a Masahiro Nishikawa,a
Hidefumi Ozawa,a Ayumu Okuda,a Takaaki Araki,a Soichi Oda,a Keisuke Inoue,a
Kimiyuki Shibuya,a Bart Staelsb and Jean-Charles Fruchartb
aTokyo New Drug Research Laboratories 1, Pharmaceutical Division, Kowa Co., LTD, 2-17-43, Noguchicho,
Higashimurayam, Tokyo 189-0022, Japan
bInstitut Pasteur de Lille, Departement d’Atherosclerose, Inserm U545, Universite de Lille 2, 1 rue Calmette BP245,
59019 Lille cedex, France
Received 10 April 2007; revised 17 May 2007; accepted 19 May 2007
Available online 24 May 2007
Abstract—A combination of benzoxazole, phenoxyalkyl side chain, and phenoxybutyric acids was identified as a highly potent and
selective human peroxisome proliferator-activated receptor a (PPARa) agonist. The synthesis, structure–activity relationship (SAR)
studies, and in vivo activities of the representative compounds are described.
Ó 2007 Elsevier Ltd. All rights reserved.
Fibrates are used in the treatment of lipid abnormalities
to decrease triglyceride (TG) levels and increase HDL-
cholesterol (HDL-C) level moderately.1 In many inter-
vention studies, fibrates showed beneficial effects in
preventing the progression of atherosclerotic lesions
and cardiovascular events in both non-diabetic and dia-
betic patients.2 These effects are attributed to PPARa
activation, which enhances lipid catabolism, decreases
TG by lowering apoC-III synthesis, and increases
HDL-C by induction of apoA-I, A-II.3 In more recent
studies, PPARa activation was found to exert anti-
inflammatory effects in blood vessels.4
PPARa-selective agonists have been reported previ-
ously.7 The common structure is composed of an aro-
matic ring and carboxylic acid with various spacers, as
shown in Chart 1. The aromatic ring of the ligands
interacts with the hydrophobic binding pocket and
the acidic moiety forms hydrogen bonds with amino
acids on the AF-2 region8 in a roughly Y-shaped
ligand binding site.
On the other hand, some natural fatty acids exhibit dif-
ferent subtype selectivity, which is dependent on the de-
gree of saturation and chain length.
Although fibrates have been shown to have clinical ben-
efits, their PPARa agonistic activity and subtype-selec-
tivity are poor.5
These observations suggest the possibility of identifying
chemical structures capable of exhibiting high affinity
for PPARa by introducing several hydrocarbon chains
in our drug design.
We postulated that highly potent and selective PPARa
agonists would be useful drugs for dyslipidemia and ath-
erosclerosis without body weight gain, edema, or other
adverse effects observed with PPARc activation.6
We designed a Y-shaped structure as described in
Chart 1. We selected the fibric acid as a common
structure in fibrates. To enhance PPARa activity and
selectivity, we selected a 2-aminobenzoxazole ring as
a bioisostere of N-phenylurea in GW9578 and con-
nected several alkyl and alkenyl chains to the nitrogen
atom. In a study to explore adequate side chains, we
found a phenoxyalkyl structure with high PPARa
activity and selectivity.
Keywords: PPARa; Design; Synthesis; Structure–activity relationship;
PPARa-selective; Fibrates; Lipid abnormalities; Triglyceride; HDL-
cholesterol; Nuclear receptor.
*
Corresponding author. Tel.: +81 0 42 391 6211; fax: +81 0 42 395
0312; e-mail: y-yamazk@kowa.co.jp
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.05.066

4690
Y. Yamazaki et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4689–4693
O
O
O
aromatic ring
O
O
N
H
OH
O
O
R3
N
NH
OH
spacer
O
N
F3C
F
OH
N
O
R1 R2
N
(CH2)m
KCL1998001079
LY-674
O
(CH2)n
fibric acid
S
OH
O
O
O
R4
side chain
O
O
N
H
N
N
Drug design
F
NS-220
O
OH
GW9578
Chart 1. Structure of PPARa agonists and our drug design.
The syntheses of phenoxyalkylamine 3 and 2-chloro-
benzoxazole 6 are illustrated in Scheme 1. Phenol 1
was alkylated with bromoacetonitrile (n = 2) or acrylo-
nitrile (n = 3) to give phenoxyalkylcyanide 2, followed
by reduction with borane-THF to provide pheno-
xyalkylamine 3. Substituted 2-aminophenol 4 was
cyclized by potassium xanthogenate to give 2-mercapto-
benzoxazole 5, followed by chlorination to give 2-chlo-
robenzoxazole 6.
benzaldehyde 7 was alkylated with 2-bromoacetate to
give 8, followed by reductive amination, which afforded
the secondary amine 9. When the number of methylene
carbons of the spacer was m = 2 and 3, the hydroxyphe-
nylalkanoic acid 10 was converted to amide 11, then by
alkylation with 2-bromoacetate to give 12, followed by
reduction, which afforded secondary amine 13. The
resulting secondary amines 9 and 13 were reacted with
2-chlorobenzoxazole 6 to afford compound 14, which
was converted to the acid 15.
The synthetic routes of all compounds were planned to
allow flexibility of substituents on the benzoxazole ring,
length of spacer, and side chain in the preparation of
analogs for SAR studies (Scheme 2). When the number
of methylene carbons in the spacer was m = 1, hydroxy-
The obtained compounds were evaluated for in vitro po-
tency and subtype selectivity in cell-based transactiva-
tion assays9 using hPPAR-GAL-4 chimeric receptors
and the results are expressed as EC50, defined as the con-
R4= H, F, MeO, OCH2O, Me,
4
a
4
R4
b
R
R
R
NH
CN
(CH )
(CH )
2
2 n
2 n-1
OH
O
O
NO2, OH, MeSO3.
n= 2, 3.
1
2
3
NH
OH
N
O
N
O
2
c
d
3
3
3
R3= H, F, Cl, CF3, MeO, OCH2O, Me, OH.
Cl
R
SH
R
4
5
6
Scheme 1. Reagents: (a) n = 1: bromoacetonitrile, K2CO3, DMF or MeCN; n = 2: acrylonitrile, Triton B, MeCN; (b) BH3–THF, THF; (c) potassium
xanthogenate, EtOH; (d) SOCl2, DMF, toluene.
O
O
R1 R2
OR5
R1, R2= H, Me, Et.
R5= Me, Et, tBu.
O
OH
O
R1
OR5
HN
a
b
R2
OHC
OHC
OH
(CH2)n
m= 1 - 3.
7
8
9
O
R4
O
O
R1
OH
OR5
O
O
R2
(CH2)m-1
(CH2)m-1
HN
HN
O
c
d
(CH2)m-1
(CH2)n
(CH2)n
HO
11
O
10
12
O
R4
R4
O
O
O
O
R1
R3
R3
N
N
O
R1
O
R1
OR5
OR5
OH
R2
O
R2
R2
f
(CH2)m
O
N
(CH2)m
(CH2)m
e
g
HN
N
(CH2)n
(CH2)n
(CH2)n
O
O
O
15
R4
13
14
R4
R4
Scheme 2. Reagents: (a) BrR1R2CCO2R5, K2CO3, DMF or MeCN; (b) 3, NaBH(OAc)3, AcOH, 1,2-dichloroethane; (c) 3, WSC–HCl, CH2Cl2; (d)
BrR1R2CCO2R5, K2CO3, DMF or MeCN; (e) BH3–THF , THF; (f) 6, iPr2NEt, MeCN; (g) R5 = Me or Et; NaOH, MeOH, R5 = tBu; TFA, CH2Cl2.

Y. Yamazaki et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4689–4693
4691
centration of test compounds to produce 50% of maxi-
mal reporter activity.
ortho-positions (16) clearly lost PPARa activity and
did not activate any subtype at 1 lM. A two-methylene
spacer (m = 2, 19) showed the highest PPARa activity
and subtype selectivity when R1 and R2 were methyl
groups.
These results are summarized in Tables 1 and 2. First,
we studied the effects of position and length of spacer,
length of phenoxyalkyl chain, and a-substituents of car-
boxylic acid (Table 1). With regard to the effect of the
position of the spacer, the meta-position (18) showed
high PPARa activity. Surprisingly, para-(17) and
Although a three-methylene spacer (m = 3, 20) showed
higher PPARa activity than the one-methylene (m = 1,
18) spacer, there was a tendency to decrease PPARa
Table 1. In vitro h-PPAR transactivation activity of 16–26
O
N
O
R1
OH
O
N
R2
m
(CH2)n
O
o, m, p
Compound
R1
R2
m
n
o, m, p
h-a EC50 (lM)
h-c EC50 (lM)
h-d EC50 (lM)
16
17
18
19
20
21
22
23
24
25
26
Me
Me
Me
Me
Me
Me
H
Me
Me
Me
Me
Me
Me
H
1
1
1
2
3
1
1
1
1
2
2
2
2
2
2
2
3
3
3
3
3
3
o
i.a.
i.a.
i.a.
i.a.
2.1
1.1
0.1
5
i.a.
i.a.
n.d.
n.d.
0.8
1.4
n.d.
2.7
2.0
2.3
1.4
p
m
m
m
m
m
m
m
m
m
0.63
0.03
0.1
0.19
1.2
n.d.
1.2
>10
1.8
1.5
Me
Et
H
0.01
0.009
0.01
0.05
H
Me
Et
H
H
Table 2. In vitro h-PPAR transactivation activities of 24, 27–45
3
R
N
O
O
O
N
OH
4
R
O
Compound
R3
R4
h-a EC50 (lM)
h-c EC50 (lM)
h-d EC50 (lM)
24
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
H
5-F
H
0.009
0.003
0.005
0.002
0.023
0.10
>10
1.0
2.0
1.1
H
6-F
7-F
H
1.1
1.9
8.0
H
0.53
0.40
>10
1.6
5-CF3
5-Me
5,6-OCH2O
H
0.16
5.0
H
H
0.005
0.16
0.18
1.1
1.9
4-OH
5-OH
H
H
2.0
H
4-F
>10
>10
>10
1.8
0.14
1.6
0.008
0.001
0.007
0.024
0.12
H
4-MeO
3,4-OCH2O
3-NO2
4-MeSO3
4-Me
2,4-diMe
3,4-diMe
2,4-diF
3,4-diF
4-OH
1.7
1.7
H
H
>10
>10
0.21
0.13
0.75
2.1
1.4
H
n.d.
0.12
0.11
0.83
6.5
H
0.001
0.001
0.001
0.011
0.002
0.26
H
H
H
H
1.8
1.4
1.3
0.003
H

4692
Y. Yamazaki et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4689–4693
selectivity. With regard to the length of the pheno-
xyalkyl chain, a three-methylene linkage (n = 3, 21)
showed 3-fold higher PPARa activity than the shorter
linkage (n = 2, 18). Next, we examined the effects of a-
substituent based on the structure of 21. Interestingly,
mono-alkyl substitution (Me, 23; Et, 24) showed
remarkable increases in PPARa activity and subtype
selectivity. On the other hand, no-substitution at the
a-position (22) showed weak PPARa activity. With re-
gard to the effects of two-methylene spacer (m = 2),
compounds (Me, 25; Et, 26) exhibited high PPARa
activity but lost the selectivity in comparison with 24.
We selected 24 with higher PPARa activity (h-
aEC50 = 0.009 lM) and subtype selectivity (more than
200-fold by EC50), and further explored the SAR of a
series of compounds.
0.001 lM, h-cEC50 = 1.10 lM, h-dEC50 = 1.58 lM,
>1000-fold subtype selectivity). The stereochemistry of
the a-substituent of carboxylic acid is identical to that
in KCL1998001079, tesaglitazar12, and ragaglitazar.13
This suggests that the common steric feature can make
favorable hydrogen bond interactions on the AF-2 re-
gion in PPARa ligand binding domain.
The sodium salts of (R)-24, (R)-35, and (R)-36 were
evaluated for TG lowering effects in normal rats after
1 week of treatment (Table 4). Fenofibrate was used as
a reference compound. These R-enantiomers at 1 mg/
kg showed much better TG lowering effects than fenof-
ibrate at 100 mg/kg. In rats, PPARa agonists increase li-
ver weight due to peroxisomal proliferation,
a
phenomenon that is well known in rats as the response
to PPARa activation. Fenofibrate showed a 44% de-
crease in TG and 44% increase in liver weight. Interest-
ingly, the R-enantiomers showed relatively small
changes in liver weight despite high TG lowering effects.
Second, we studied the effects of substituents on the
benzoxazole ring (Table 2). Although there were slight
differences in the activity, trifluoromethyl (30) as an elec-
tron-withdrawing group and a fluorine atom (27, 28, 29)
yielded higher PPARa activity than electron-donating
groups (Me, 31; OCH2O, 32).
Next, we evaluated (R)-24, (R)-35, and (R)-36 in human
apoA-I (h-apoA-I) transgenic mice for the ability to in-
crease HDL-C.14 The results of plasma h-apoA-I
changes are summarized in Table 5. (R)-35 at 3 mg/kg
increased plasma h-apoA-I to an extent similar to fenof-
ibrate at 300 mg/kg. Interestingly, (R)-24 at 3 mg/kg and
(R)-36 at 1 mg/kg showed more pronounced increases in
plasma h-apoA-I than fenofibrate at 300 mg/kg. The
Third, we studied the effects of substituents on the ben-
zene ring of the side chain. A fluorine atom (35, 43, 44)
and electron-donating groups (Me, 40; MeO, 36;
OCH2O, 37) showed high PPARa activity. There were
no significant differences in PPARa activity related to
the degree of substituents (di-Me, 41, 42; di-F, 43, 44).
Electron-withdrawing groups showed slight decreases
in PPARa activity (NO2, 38; MeSO3, 39). Notably, the
derivative with a hydroxy group (45) showed the lowest
PPARa activity. Similar inactivation was observed in 33
and 34. Due to the excellent PPARa activity and selec-
tivity, we selected compounds 24, 35, and 36 for further
evaluation.
Table 4. In vivo efficacy of (R )-24, (R)-35, and (R)-36 in normal rats
Compound
Dose
(mg/kg)
Triglyceridea
(%)
Liver weightb
(%)
(R)-24
(R)-35
1
À51 9c
À46 9c
À49 15c
À44 16c
19 4c
20 9c
25 8c
44 8c
1
(R)-36
Fenofibrate
1
100
Next, we performed optical resolution to investigate the
enantiodependency of the activities.10 The results of
in vitro experiments are summarized in Table 3. All
R-enantiomers11 showed more than 10-fold higher
PPARa activity than the corresponding S-enantiomers.
The selected compounds exhibited enantiodependent
PPARa activity. Especially, (R)-36 showed superior
Male Sprague–Dawley rats (6 weeks old, n = 6) were fed normal chow
with free access to water and received oral dosing with the sodium salts
of the test compounds and fenofibrate by gavage with vehicle (0.5%
methylcellulose) once a day for 7 days.
a Values are means SD of TG change.
b Values are means SD of liver weight (g/100 g body weight) change.
c P < 0.05 versus vehicle control by Student’s t test.
PPARa activity and subtype selectivity (h-aEC50
=
Table 3. In vitro h-PPAR transactivation activity of chiral compounds
Table 5. In vivo efficacy of (R)-24, (R)-35, and (R)-36 in human apoA-I
transgenic mice
N
O
O
*
O
O
N
OH
h- apoA-I (%)a
4
Compound
Dose (mg/kg)
R
(R)-24
(R )-35
3
201 30*
147 43*
194 19*
145 26*
3
(R )-36
Fenofibrate
1
300
Compound
R/S
R4
h-a EC50
(lM)
h-c EC50
(lM)
h-d EC50
(lM)
Heterozygous human apoA-I transgenic mice (n = 5–6) were fed nor-
mal chow with free access to water and received oral dosing of the
sodium salts of the test compounds and fenofibrate by gavage with
vehicle (0.5% methylcellulose) once a day for 14 days.
a Values are means SD of h-apoA-I change.
24
24
35
35
36
36
R
S
R
S
R
S
H
H
F
0.002
0.13
0.91
0.87
0.94
1.19
1.10
1.31
1.61
1.66
1.19
1.57
1.58
1.80
0.003
0.02
F
MeO
MeO
0.001
0.01
* P < 0.05 versus vehicle control by Dunnett’s multiple comparison
test.

Y. Yamazaki et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4689–4693
4693
7. (a) Miyachi, H. Expert Opin. Ther. Patents 2004, 14, 607;
(b) Miyachi, H. Expert Opin. Ther. Patents 2005, 15, 1521;
(c) Brown, P. J.; Winegar, D. A.; Plunket, K. D.; Moore,
L. B.; Lewis, M. C.; Wilson, J. G.; Sundseth, S. S.; Koble,
C. S.; Wu, Z.; Chapman, J. M.; Lehmann, J. M.; Kliewer,
S. A.; Willson, T. M. J. Med. Chem. 1999, 42, 3785; (d)
Xu, Y.; Mayhugh, D.; Saeed, A.; Wang, X.; Thompson,
R. C.; Dominianni, S. J.; Kauffman, R. F.; Singh, J.; Bean,
J. S.; Bensch, W. R.; Barr, R. J.; Osborne, J.; Montrose-
Rafizadeh, C.; Zink, R. W.; Yumibe, N. P.; Huang, N.;
Luffer-Atlas, D.; Rungta, D.; Maise, D. E.; Mantlo, N. B.
J. Med. Chem. 2003, 46, 5121; (e) Nomura, M.; Tanase,
T.; Ide, T.; Tsunoda, M.; Uchiki, H.; Murakami, K.;
Miyachi, H. J. Med. Chem. 2003, 46, 3581; (f) Kuwabara,
K.; Murakami, K.; Todo, M.; Aoki, T.; Asaki, T.; Murai,
M.; Yana, J. J. Pharmacol. Exp. Ther. 2004, 309, 970.
8. (a) Xu, H. E.; Lambert, M. H.; Montana, V. G.; Plunket,
K. D.; Moore, L. B.; Collins, J. L.; Oplinger, J. A.;
Kliewer, S. A.; Gampe, R. T., Jr.; McKee, D. D.; Moore,
J. T.; Willson, T. M. Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 13919; (b) Xu, H. E.; Lambert, M. H.; Montana, V.
G.; Parks, D. J.; Blanchard, S. G.; Brown, P. J.;
Sternbach, D. D.; Lehmann, J. M.; Wisely, G. B.; Willson,
T. M.; Kliewer, S. A.; Milburn, M. V. Mol. Cell 1999, 3,
397.
changes in HDL-C were correlated with those of plasma
h-apo-AI (data not shown).
In summary, we have identified highly potent and selec-
tive human PPARa agonists and demonstrated the
excellent pharmacological effects of (R)-36. Based on
the results of in vitro and in vivo studies, we expect that
(R)-36 would exhibit higher efficacy in patients with
dyslipidemia than currently available fibrates.
Acknowledgments
We thank Dr. K. Sawanobori, Director of Tokyo Re-
search Laboratories I, Kowa Co., Ltd, for his encourage-
ment and support. We are also grateful to Dr. H. Sawada
and Miss M. Isuge of Kowa Tokyo New Drug Research
Laboratories 1 for in vivo evaluation, and Mr. S. Inoue
of Kowa Co., Ltd, for pharmacokinetic evaluation.
References and notes
9. Raspe, E.; Madsen, L.; Lefebvre, A. M.; Leitersdorf, I.;
Gelman, L.; Peinao-Onsurbe, J.; Dallongeville, J.; Fru-
chart, J. C.; Berge, R.; Staels, B. J. Lipid Res. 1999, 40,
2099.
10. Optical resolution was carried out in Daicel Chemical
Industries, Ltd.
11. HPLC analyses were performed on CHIRALPAK AD
(flow rate 1.0 mL/min., UV 254 nm, n-hexane/i-PrOH/
TFA = 100/10/0.1 as a eluent). The enantiomeric excess of
those compounds was more than 99% ee.
1. (a) Staels, B.; Auwerx, J. Curr. Pharm. Des. 1997, 3, 1; (b)
Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, K.;
Leitersdorf, E.; Fruchart, J. C. Circulation 1998, 89, 2008.
2. (a) Vakkilainen, J.; Steiner, G.; Ansquer, J. C.; Aubin, F.;
Rattier, S.; Foucher, C.; Hamsten, A.; Taskinen, M. R.on
behalf of the DAIS Group Circulation 2003, 107, 1733; (b)
Robins, S. J.; Bloomfield, H. E. Curr. Opin. Lipidol. 2006,
17, 431.
3. Fruchart, J. C.; Duriez, P.; Staels, B. Curr. Opin. Lipidol.
1999, 10, 245.
12. Cronet, P.; Petersen, J. F.; Folmer, R.; Blomberg, N.;
Sjoblom, K.; Karlsson, U.; Lindstedt, E. L.; Bamberg, K.
Structure 2001, 9, 699.
13. Ebdrup, S.; Pettersson, I.; Rasmussen, H. B.; Deussen, H.
J.; Frost Jensen, A.; Mortensen, S. B.; Fleckner, J.; Pridal,
L.; Nygaard, L.; Sauerberg, P. J. Med. Chem. 2003, 46,
1306.
4. (a) Staels, B.; Koenig, W.; Habib, A.; Merval, R.; Lebret,
M. Y.; Torra, I. P.; Delerive, P.; Fadel, A.; Chinetti, G.;
Fruchart, J. C.; Najib, J.; Maclouf, J.; Tedgui, A. Nature
1998, 393, 790; (b) Marx, N.; Sukhova, G. K.; Collins, T.;
Libby, P.; Plutzky, J. Circulation 1999, 99, 3125.
5. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
14. Berthou, L.; Duverger, N.; Emmanuel, F.; Langouet, S.;
Auwerx, J.; Guillouzo, A.; Fruchart, J. C.; Rubin, E.;
Denefle, P.; Staels, B.; Branellec, D. J. Clin. Invest. 1996,
97, 2408.
6. (a) Staels, B. Cell Metab. 2005, 2, 77; (b) Vosper, H.; Patel,
L.; Graham, T. L.; Khoudoli, G. A.; Hill, A.; Macphee, C.
H.; Pinto, I.; Smith, S. A.; Suckling, K. E.; Wolf, C. R.;
Palmer, C. N. A. J. Biol. Chem. 2001, 276, 44258.