Design and synthesis of substituted phenylpropanoic acid derivatives as human peroxisome proliferator-activated receptor α/δ dual agonists

Article abstract of DOI:10.1016/j.bmcl.2005.10.049

A series of phenylpropanoic acids was prepared as candidate dual agonists of peroxisome proliferator-activated receptors (PPAR) α and δ. Structure-activity relationship studies indicated that the shape of the linker moiety and the nature of the substituen

Full text of DOI:10.1016/j.bmcl.2005.10.049

Bioorganic & Medicinal Chemistry Letters 16 (2006) 554–558
Design and synthesis of substituted phenylpropanoic acid
derivatives as human peroxisome proliferator-activated
receptor a/d dual agonists
Jun-ichi Kasuga,a Makoto Makishima,b Yuichi Hashimotoa and Hiroyuki Miyachia,*
aInstitute of Molecular and Cellular Biosciences, University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan
bSchool of Medicine, Nihon University, 30-1 Oyaguchi-kamicho, Itabashi-ku, Tokyo 173-8610, Japan
Received 14 September 2005; revised 10 October 2005; accepted 17 October 2005
Available online 3 November 2005
Abstract—A series of phenylpropanoic acids was prepared as candidate dual agonists of peroxisome proliferator-activated receptors
(PPAR) a and d. Structure–activity relationship studies indicated that the shape of the linker moiety and the nature of the substi-
tuent at the distal benzene ring play key roles in determining the potency and selectivity of PPAR subtype transactivation. Optically
active a-ethylphenylpropanoic acid derivatives were identified as potent human PPAR a and d dual agonists with potential for the
treatment of metabolic syndrome.
Ó 2005 Elsevier Ltd. All rights reserved.
Identification of the molecular targets of the transducer
proteins critically involved in metabolic pathways is cru-
cial for understanding energy homeostasis in humans,
and also for developing new therapeutic agents for the
treatment of altered metabolic homeostasis, such as ath-
erosclerosis, diabetes, and obesity. Metabolic nuclear
receptors are particularly attractive target molecules,
since they play a central role in maintaining cellular
and whole-body glucose and lipid homeostasis. These
nuclear receptors are activated by a wide variety of
physiological ligands, including dietary fatty acids and
cholesterol metabolites, and xenobiotic compounds,
thereby modulating the transcriptional network of the
target response genes. Among these receptors, special
attention has been paid for more than a decade to the
members of the peroxisome proliferator-activated recep-
tor (PPAR) family.
expressed in a tissue-specific manner and to play a piv-
otal role in lipid, lipoprotein, and glucose homeostasis.3
PPARa is mostly expressed in the tissues involved in lip-
id oxidation, such as liver, kidney, skeletal, cardiac mus-
cle, and adrenal glands. PPARc is expressed in adipose
tissue, macrophages, and vascular smooth muscles. In
contrast to the specific distribution of PPARa and
PPARc, PPARd is ubiquitously expressed in almost all
mammalian tissues. PPARs heterodimerize with another
nuclear receptor partner, retinoid X receptor (RXR),
and the heterodimers regulate gene expression by bind-
ing to a specific consensus DNA sequence, termed
PPRE (peroxisome proliferator responsive element),
which is a direct repeat of the hexameric AGGTCA rec-
ognition motif separated by a single nucleotide (DR1),
present in the promoter region of the target genes.4
PPARc was first identified as a master regulator of adi-
pocyte differentiation, but more recent molecular-bio-
logical studies have indicated that PPARc activation is
also linked to the expression of many important genes
that affect energy metabolism, such as the TNF-a, lep-
tin, and adiponectin genes.5
PPARs are members of the nuclear receptor superfamily
and are activated by endogenous saturated and unsatu-
rated fatty acids and their metabolites, as well as syn-
thetic ligands.1 PPARs are heterogeneous and three
subtypes have been identified to date; PPARa [officially
NR1C1], PPARd/b [NR1C2], and PPARc [NR1C3].2
Each PPAR subtype appears to be differentially
PPARa regulates the expression of genes encoding for
proteins involved in lipid and lipoprotein homeostasis.6
For example, it regulates genes involved in fatty acid
uptake (such as fatty acid binding protein, FABP), b-ox-
idation (acyl-CoA oxidase), and x-oxidation (cyto-
Keywords: PPAR; PPARa; PPARd; Metabolic syndrome.
*
Corresponding author. e-mail: miyachi@iam.u-tokyo.ac.jp
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.10.049

J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 554–558
555
chrome P450). It down-regulates apolipoprotein
C-III, a protein that inhibits triglyceride hydrolysis by
lipoprotein lipase,6 and it also regulates genes involved
in reverse cholesterol transport, such as apolipoprotein
A-I and apolipoprotein A-II.6
and alkaline hydrolysis to afford the desired products
(Scheme 1).
Optically active a-ethyl-substituted phenylpropanoic
acid derivatives (16a, 16b) were prepared by means of
EvansÕ asymmetric alkylation procedure as a key step.
Acylation of 4-(R)-benzyloxazolidinone provided 11,
which was treated with benzyl 5-bromomethyl-2-meth-
oxybenzoate under EvansÕs asymmetric alkylation pro-
tocol, followed by hydrogenolysis to afford the benzoic
acid intermediate 12. This was reduced with BH3, then
protection of the hydroxyl group with a tert-butyldim-
ethylsilyl group (TBDMS), treatment with benzyl lithi-
um and deprotection of the TBDMS, afforded 13. 13
was brominated, and treated with potassium phthali-
mide, then deprotection of the phthaloyl group afforded
the key aminomethyl intermediate 14. 14 was reacted
with benzoyl chloride derivatives, and subsequent
removal of the chiral oxazolidinone moiety afforded
the desired (S) form of compounds 16a and 16b (97%
ee) (Scheme 2). The antipodal (S)-isomer (19) was pre-
pared by the use of 4-(S)-benzyloxazolidinone as a start-
ing material. Racemic 2, 17, and 18 were prepared by the
reported methods.9
Based on the findings that antidiabetic thiazolidine-2,4-
diones (glitazones) and antidyslipidemic fibrates are
ligands of PPARc and PPARa, respectively, much
research has been focused on these metabolic nuclear
receptor subtypes as therapeutic targets for the treat-
ment of metabolic syndrome (also known as insulin
resistance syndrome, visceral fat syndrome, syndrome
X, and deadly quartet), such as GW-9578 (1) and
KCL (2) (Fig. 1).
Although PPARd is ubiquitously distributed in a wide
range of tissues and cells, research interest in PPARd
has been limited. However, after 2001, the availability
of PPARd-knockout animals and selective ligands, such
as GW-501516 (4) and L-165041 (5), prompted us to
examine the roles of PPARd in fatty acid metabolism, re-
verse cholesterol transport, and other disease states.7,8
Considering the positive contributions of both PPARa
and PPARd to lipid, lipoprotein, and cholesterol homeo-
stasis, dual agonists of both PPARa and PPARd might be
other candidates for the treatment of metabolic disease,
but there have been very limited reports on such agonists,
such as GW-2433 (3) (Fig. 1), and the activity is moderate.
The transactivation activity of the present series of com-
pounds toward PPARs is summarized in Table 1,
together with the results for racemic KCL, a human
PPARa-selective agonist. First, we investigated the ef-
fect of the introduction of a fluorine atom in racemic
KCL, because we expected that a fluorinated compound
would bind more tightly to PPAR by effectively exclud-
ing the water molecule located in the hydrophobic bind-
ing pocket of PPAR. Compound (10a), which has a
fluorine atom on the central benzene ring, instead of a
methoxyl group, showed considerably decreased
PPARa, PPARd, and PPARc transactivation activity
as compared with racemic KCL, that is, this compound
did not show apparent activity at the concentration of
1 lM. This result clearly indicated the importance of
the methoxyl group located at the ortho position of
the amide carbonyl group; it may be involved in a
hydrogen bonding interaction that restricts the confor-
mation of the central benzene part of the molecule. On
the other hand, introduction of a fluorine atom at the
distal benzene ring (10b, 10c) enhanced the PPARa
transactivation activity to some extent.
Previously, we have reported the design and synthesis of
substituted phenylpropanoic acid derivatives as human
PPARa-selective agonists.9 As part of our continuing re-
search directed toward the structural development of
subtype-selective PPAR agonists, we report here potent
novel substituted phenylpropanoic acid dual PPARa/d
agonists. To develop structurally new human PPARa/d
dual agonists, we selected 2 (KCL) (Fig. 1), a human
PPARa-selective agonist, as a lead compound,9 and per-
formed further structural modification focusing on the
distal hydrophobic tail part and the linker moiety.
Synthetic routes to the phenylpropanoic acids prepared
in this study are outlined in Schemes 1 and 2.
Compounds (10a–10d) were prepared from a benzalde-
hyde derivative (6a or 6b) in 4 steps. Compound 6a or
6b, which was prepared from 5-formylsalicylic acid,
was treated with Emmons reagents, followed by hydrog-
enolysis, condensation with substituted benzylamine,
As regards PPARd transactivation activity, introduction
of a fluorine atom at the distal benzene ring (10b, 10c)
Figure 1. Structures of representative PPARa (1 (GW-9578), 2 (KCL)), PPARd (4 (GW-501516), 5 (L-165041)), and dual PPARa/d (3 (GW-2433))
agonists.

556
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 554–558
O
X
O
O
X
O
O
X
O
O
X
O
OEt
N
H
a
b
c
BnO
H
BnO
OEt
HO
OEt
z
Y
6a: X = F
6b: X = OMe
7a
7b
8a
8b
9a: X = F, Y = H, , Z = 4-CF3
9b: X = OMe, Y = 2-F, Z = 4-CF3
9c: X = OMe, Y = 3-F, Z = 4-CF3
9d: X = OMe, Y = 4-F, Z = 3-CF3
O
X
O
OH
N
H
d
z
Y
10a: X = F, Y = H, , Z = 4-CF3
10b: X = OMe, Y = 2-F, Z = 4-CF3
10c: X = OMe, Y = 3-F, Z = 4-CF 3
10d: X = OMe, Y = 4-F, Z = 3-CF3
Scheme 1. Reagents: (a) (EtO)2POCH(Et)CO2Et, NaH, THF; (b) H2, 10% Pd-C, EtOH; (c) substituted benzylamine, ClCO2Et, TEA, CH2Cl2; (d) aq
NaOH, EtOH.
O
O
O
O
O
O
O
O
O
HO
MeO
N
N
e
f
g
O
O
HO
H2N
MeO
MeO
11
12
13
14
O
O
R
O
O
R
O
OH
N
N
H
MeO
h
i
H
F3C
F3C
MeO
15a: R = 2-F
15b: R = 3-F
16a: R = 2-F
16b: R = 3-F
Scheme 2. Reagents and conditions: (e) 1—LiHMDS, THF, benzyl 5-bromomethyl-2-methoxybenzoate, À40 °C–rt, 2—H2, 10% Pd-C, AcOEt, rt; (f)
1—BH3, THF, À10 °C, 2—TBDMSCl, imidazole, THF, rt, 3—benzyl alcohol, n-BuLi, THF, À40 °C–rt, 4—TBAF, THF, rt; (g) 1—P(Ph)3-polymer,
CBr4, rt, 2—potassium phthalimide, DMF, 3—NaBH4, iPrOH-H2O, rt then AcOH, 80 °C; (h) RPhCOCl, TEA, CH2Cl2, rt; (i) H2, 10% Pd-C,
AcOEt, rt.
Table 1. In vitro functional PPAR transactivation activity of substituted phenylpropanoic acids
Y
5
3
6
O
L
X
4
OH
z
2
Compound
X
Y
Z
L
Stereo
EC50 (nM)a
PPARa
PPARd
PPARc
Racemic 2
10a
10b
MeO
F
H
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
3-CF3
4-CF3
4-CF3
4-CF3
CH2NHCO
CH2NHCO
CH2NHCO
CH2NHCO
CONHCH2
CONHCH2
CONHCH2
CONHCH2
CONHCH2
Racemic
Racemic
Racemic
Racemic
Racemic
Racemic
S
70
iab
20
700
iab
3000
iab
500
H
MeO
MeO
MeO
MeO
MeO
MeO
MeO
2-F
3-F
H
1000
500
130
2000
12
10c
17
25
10
1600
2000
18
16a
4-F
2-F
3-F
3-F
250
10
>10,000
1900
4900
16b
19
S
12
180
23
700
R
>10,000
a Compounds were screened for agonist activity on PPAR-GAL4 chimeric receptors in transiently transfected HEK-293 cells as described. EC50 value
is the molar concentration of the test compound that affords 50% of the maximal reporter activity. n = 3.
b ÔiaÕ means inactive (no apparent activity) at the concentration of 1 lM.
apparently did not influence the activity, but in the case
of PPARc, introduction of a fluorine atom at the 2-po-
sition increased the transactivation activity to some ex-
tent. These results might indicate that the shape and
environment of the hydrophobic cavity hosting the dis-
tal benzene ring differ somewhat among the three PPAR
isoforms.
Previously, we found that the 3-atom unit linker; –CH2–
NH–CO– was very potent and PPARa selective, but
shortening or lengthening of the linking group consider-
ably decreased the activity.9 We also found that a flexi-
ble linker, such as –CH2–CH2–CH2– or –CH2–CH2–O–,
decreased PPARa transactivation activity, but caused
PPARd transactivation activity to appear.9 Therefore,

J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 554–558
557
Table 2. In vitro functional PPAR transactivation activity of 16b on
hPPARa, mPPARa, hl272F, hPPARd, and mPPARd
we reinvestigated the effect of the –CO–NH–CH2– linker
(17) and found that this linker increased both PPARa
and PPARd transactivation activities, as compared to
those of the –CH2–NH–CO– linker compound (racemic
2). We speculated that the introduction of a conforma-
tionally flexible group and/or atom, such as –CH2– or
–O–, next to the central benzene ring might favor both
PPARa and PPARd activity, while introduction of a
more conformationally restricted group (–CO–) next to
the central benzene ring might be unfavorable for
PPARd activity. These results might reflect differences
of the shape and environment of the hydrophobic cavity
hosting the distal benzene ring between PPARa and
PPARd.
EC50 (nM)a
hPPARa
mPPARa
hI272F
630
hPPARd
mPPARd
10
1000
40
130
a Compounds were screened for agonist activity on PPAR-GAL4
chimeric receptors in transiently transfected HEK-293 cells as
described. EC50 value is the molar concentration of the test com-
pound that causes 50% of the maximal reporter activity. n = 3.
amino acid residue 272, isoleucine (Ile272), which is
located in the helix 3 region of the human PPARa
LBD.11,12 Therefore, we investigated the activity with
I272F mutant. As can be seen from Table 2, 16b
exhibited about 100-fold less activity, which is compa-
rable to that obtained with mouse PPARa. This result
indicates that although the shape of the linking group
is somewhat different from that of the normal amide
linker derivative, such as KCL, the hydrophobic tail
part of 16b also interacts specifically with amino acid
272, isoleucine.
The position of the substituents introduced onto the dis-
tal benzene ring is very important; compound 18, which
has a trifluoromethyl group and a fluorine atom at the 3-
and 4-positions, exhibited a 10-fold lower PPARa trans-
activation activity. This is consistent with the previously
obtained SAR result that steric bulkiness at the 4-posi-
tion is an important factor for potent PPARa transacti-
vation activity.
Considering the results obtained above, we then pre-
pared optically active derivatives, 16a, 16b, and 19.
As can be seen from Table 1, a clear enantio-depen-
dence of the transactivation activity toward PPARa
and PPARd isoforms was found. 16b, which has (S)
configuration, exhibited much more potent transacti-
vation activity on both PPARa and PPARd, while
the antipodal (R) isomer 19 exhibited less potency.
Therefore, we concluded that the activity resides al-
most exclusively in the (S)-enantiomer, and both 16a
and 16b showed dual-agonist activity toward PPARa
and PPARd.
In contrast to PPARa, there is no report about species
selectivity of PPARd agonist. Therefore, we also evalu-
ated the species-selectivity profile of PPARd transactiva-
tion by 16b. Table 2 also indicates that species selectivity
of 16b for the PPARd subtype is not so marked; 16b
shows EC50 values against human and mouse PPARd
of 40 and 130 nM, respectively. In the case of PPARd,
14 amino acids are different between the human and
mouse PPARd ligand binding domain, that is, 205Ala,
208Thr, 239Lys, 251Cys, 266Ser, 269Ser, 314Arg,
369Arg, 387Ala, 392Ala, 419Arg, 420Ile, 425Thr, and
428Ser of human PPARd are replaced with 204Ser,
207Asn, 238Asn, 250Ser, 265Asn, 268Asn, 313His,
368Gln, 386Val, 391Ser, 418Trp, 419Leu, 424Ser, and
427Leu in the case of mouse PPARd. But, based on a
consideration of the X-ray crystallographic study of hu-
man PPARd and GW-2433 complex and/or human
PPARd and eicosapentaenoic acid complex, none of
these amino acids appears to be involved in specific con-
tact with the agonist ligand. Therefore, compound 16b
did not show a large species selectivity between human
and mouse PPARd. As far as the authors know, this is
the first report that discusses the species selectivity of
PPARd agonist.
As previously described, some peroxisome proliferators
show species-dependent transactivation characteristics
for PPARa.10 The classical PPARa agonist WY-14643
is more effective on rodent PPARa than on human
PPARa. 5,8,11,14-Eicosatetraenoic acid (ETYA)
showed the reverse preference, that is, it is 10-fold more
effective on human PPARa than rodent PPARa. On the
other hand, fenofibric acid, an active metabolite of the
fibrate class antihyperlipidemic agent fenofibrate, did
not show clear species differences.
Therefore, we evaluated the species-selectivity profile of
PPARa transactivation by 16b.
In summary, we have developed the potent human dual
PPARa/d agonists 16a and 16b. Further cell-based assay
and in vivo pharmacological evaluation of these com-
pounds are under way.
As can be seen from Table 2, 16b activated human and
mouse PPARa with EC50 values of 10 and 1000 nM,
respectively. Thus, 16b showed species preference for
humans, and the transactivation activity of 16b for
PPARa was approximately 100-fold less potent in mice
than that in humans.
Acknowledgments
The work described in this paper was partially support-
ed by Grants-in-Aid for Scientific Research from The
Ministry of Education, Culture, Sports, Science and
Technology, Japan, and the Japan Society for the Pro-
motion of Science.
Previously we have demonstrated that a single amino
acid residue is responsible for this human-selective
PPARa activation of 2, that is, the human selectivity
of KCL is primarily mediated through the specific
contact of the hydrophobic tail part of KCL with

558
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 554–558
7. Lim, H.; Gupta, R. A.; Ma, W. G.; Paria, B. C.; Moller,
References and notes
D. E.; Morrow, J. D.; DuBois, R. N.; Trzaskos, J. M.;
Dey, S. K. Genes Dev. 1999, 13, 1561.
1. Banner, C. D.; Gottlicher, M.; Widmark, E.; Sjovall, J.;
Rafter, J. J.; Gustafsson, J. A. J. Lipid Res. 1993, 34, 1583.
2. Nuclear Receptor Nomenclature Committee, Cell 1999,
97, 161.
3. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
8. Sznaidman, M. L.; Haffner, C. D.; Maloney, P. R.;
Fivush, A.; Chao, E.; Goreham, D.; Sierra, M. L.;
LeGrumelec, C.; Xu, H. E.; Montana, V. G.; Lambert,
M. H.; Willson, T. M.; Oliver, W. R., Jr.; Sternbach, D. D.
Bioorg. Med. Chem. Lett. 2003, 13, 1517.
9. Nomura, M.; Tanase, T.; Ide, T.; Tsunoda, M.; Suzuki,
M.; Uchiki, H.; Murakami, K.; Miyachi, H. J. Med.
Chem. 2003, 46, 3581.
4. Keller, H.; Dreyer, C.; Medin, J.; Mahfoudi, A.; Ozato,
K.; Wahli, W. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
2160.
10. Keller, H.; Devchand, P. R.; Perround, M.; Wali, W. Biol.
Chem. 1997, 378, 651.
11. Miyachi, H.; Uchiki, H. Bioorg. Med. Chem. Lett. 2003,
13, 3145.
5. 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.
12. Uchiki, H.; Miyachi, H. Chem. Pharm. Bull. (Tokyo)
2004, 52, 365.
6. Mukherjee, R.; Jow, L.; Noonan, D.; McDonnell, D. P. J.
Steroid Biochem. Mol. Biol. 1994, 51, 157.