Design and synthesis of α-aryloxyphenylacetic acid derivatives: A novel class of PPARα/γ dual agonists with potent antihyperglycemic and lipid modulating activity

Article abstract of DOI:10.1021/jm0502135

The synthesis and structure-activity relationships of novel series of α-aryloxyphenylacetic acids as PPARα/γ dual agonists are reported. The initial search for surrogates of the ester group in the screen lead led first to the optimization of a subseries w

Full text of DOI:10.1021/jm0502135

J. Med. Chem. 2005, 48, 4457-4468
4457
Design and Synthesis of r-Aryloxyphenylacetic Acid Derivatives: A Novel Class
of PPARr/γ Dual Agonists with Potent Antihyperglycemic and Lipid Modulating
Activity
Guo Q. Shi,*,† James F. Dropinski,† Brian M. McKeever,† Shihua Xu,† Joseph W. Becker,† Joel P. Berger,‡
Karen L. MacNaul,‡ Alex Elbrecht,‡ Gaochao Zhou,‡ Thomas W. Doebber,‡ Peiran Wang,§ Yu-Sheng Chao,§
Mike Forrest,# James V. Heck,† David E. Moller,‡ and A. Brian Jones†
Departments of Medicinal Chemistry, Metabolic Disorders, Atherosclerosis and Endocrinology and Animal Pharmacology,
Merck Research Laboratories, P.O Box 2000, Rahway, New Jersey 07065-0900
Received March 8, 2005
The synthesis and structure-activity relationships of novel series of R-aryloxyphenylacetic
acids as PPARR/γ dual agonists are reported. The initial search for surrogates of the ester
group in the screen lead led first to the optimization of a subseries with a ketone moiety. Further
efforts to modify the ketone subseries led to the design and synthesis of two new subseries
containing fused heterocyclic ring systems. All these analogues were characterized by their
“super” PPARR agonist activity and weak or partial agonist activity on PPARγ in PPAR-GAL4
transactivation assays despite their similar binding affinities for both receptors. The cocrystal
structures of compounds 7 and rosiglitazone with PPARγ-LBD were compared, and significant
differences were found in their interactions with the receptor. Select analogues in each subseries
were further evaluated for in vivo efficacy. They all showed excellent anti-hyperglycemic efficacy
in a db/db mouse model and hypolipidemic activity in hamster and dog models without
provoking the typical PPARγ-associated side effects in the rat tolerability assay.
Chart 1. Chemical Structures of Selected PPARR/γ
Dual Agonists
Introduction
The peroxisome proliferator-activated receptors
(PPARs) are ligand-activated transcription factors in the
nuclear hormone receptor superfamily.1,2 Three distinct
PPAR subtypes (PPARγ, PPAR and PPARδ) have been
identified in most mammalian species. The physiological
role of the PPARs as regulators of lipid and glucose
metabolism has been extensively reviewed in recent
years.3-7 PPARγ is well-known at a cellular level for
its role in adipogenesis and has been implicated as the
primary receptor modulating the antidiabetic activity
through insulin sensitization.8,9 PPARγ agonists, such
as TZDs, have proven to be efficacious as insulin
sensitizing agents in the treatment of type 2 diabetes.10-12
Unfortunately, they are also known to cause undesirable
side effects including weight gain (as a result of en-
hanced adipogenisis and increased plasma volume),
edema and anemia in both animal models and humans.
PPARR is known to play a pivotal role in the uptake
and oxidation of fatty acids and also in lipoprotein
metabolism.13 Activation of PPARR is known to be the
predominant mechanism by which fibrates lower trig-
lycerides, elevate HDL and exert insulin sensitizing
effects.14-20 Since the majority of type 2 diabetes pa-
tients suffer from atherogenic lipid abnormalities in
addition to insulin resistance, both of which are mala-
dies associated with the so-called metabolic syndrome,21
the concept of identifying ligands that activate both
PPARR and PPARγ represents a logical continuation in
the field of PPAR research.
Indeed, there have been intense efforts within the
pharmaceutical industry to develop PPARR/γ dual ago-
nists for clinical use (Chart 1).22-33 The first example,
compound 1 (farglitazar),23 a potent PPARγ agonist with
a moderate PPARR component, was developed as a
clinical candidate but was dropped due to the emergence
* To whom correspondence should be addressed. Tel: 732-594-8592;
Fax: 732-5949556; e-mail: guoqiang_shi@merck.com.
† Department of Medicinal Chemistry.
‡ Metabolic Disorders.
§ Atherosclerosis and Endocrinology.
# Animal Pharmacology.
10.1021/jm0502135 CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/04/2005

4458 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Shi et al.
Chart 2
transactivation assay (TA) assay, this compound turned
out to be a “super” agonist for hPPARR and a weak
35
agonist for hPPARγ receptors. The novel structural
features of compound 5 coupled with its interesting in
vitro biological profile prompted us to embark on further
investigation. A preliminary report describing some
early structure-activity studies has appeared.31 This
paper describes the synthesis, the SAR development,
X-ray crystallography study and various in vivo evalu-
ations of a new series of R-aryloxyphenylacetic acids
that culminated in the discovery of compound 6 and 7.
Chemistry
Most compounds in the racemic series were readily
synthesized by SN2 coupling between R-bromophenyl-
acetates and three different types of phenols (Scheme
1). For the synthesis of the enantiomerically pure
analogues, a highly stereoselective synthetic route based
on the use of a lactate-derived chiral auxiliary 9 was
adopted.36 Thus, the readily available phenylglyoxalic
acid 8 was esterified with chiral auxiliary 9, and the
resulting keto ester 10 was converted to the R-bromo
ester 11 as a diastereomeric mixture. Displacement of
the bromide by various lithium phenoxides, generated
from the phenols 12, 13 and 14, proceeded in a highly
stereoselective fashion to give the coupling product 15.
Cleavage of the chiral auxiliary by simple hydrolysis
gave three different subseries of R-phenoxyphenyl acetic
acids, i.e., the ketophenol series (ArOH ) 12), the
benzoxazolones (ArOH ) 13) and the benzisoxazolones
(ArOH ) 14), typically with an enantiomeric excess
greater than 96%.
Each type of phenol used for the coupling reaction was
synthesized by the synthetic routes that allowed for the
incorporation of various R1, R2 and R3 groups in the
indicated positions. In each case, the incorporation of
the R2 and R3 group is generally made possible either
by use of commercially available ortho-substituted phe-
nols or by utilizing various reactions directed by the
ortho phenolic hydroxyl group. For the preparation of
ketophenol 12, a regioselective acylation at the position
para to the phenolic hydroxyl group of 16 was achieved
using various R1-substituted acids in neat triflic acid.
Alternatively, the addition of Grignard agents R1MgX
to the formyl group of 17 at the para position followed
of edema in an advanced stage of development. Two dual
agonists with more substantial PPARR activity, 2
(ragaglitazar or DRF2725)25 and 3 (MK-767 or KRP-
297),24 were also dropped from late clinical development
due to carcinogenicity in rodent toxicity models. So far,
compound 4 (muraglitazar)28,33 is the only dual agonist
candidate that has been successfully advanced to NDA
filing.34 Thus, the development of PPARR/γ dual ago-
nists with novel biological profiles and structural di-
versity remains an attractive field for further explora-
tion.
Our research on PPARR/γ dual agonists was initiated
from the screening lead 5, which possessed a unique
molecular architecture (Chart 2). It was a phenylacetic
acid substituted at the R-position with a phenoxy group,
to which was symmetrically attached two propyl groups
on each side and an ester group at the para position.
This structural feature differentiates compound 5 from
most existing PPARR/γ dual agonists reported in the
literature, which are frequently represented by an
amphiphilic molecular design including an acidic “head-
group” tethered with a chain to a lipophilic “tail group”,
such as those shown in Chart 1. Compound 5 was
discovered as a PPAR ligand with micromolar hPPARγ
and nanomolar hPPARR affinity (PPAR_SPA IC50: R/γ
) 0.082/1.1 µM). In the cell-based PPAR-GAL4 chimera
Scheme 1. General Synthetic Route to R-Aryloxyphenylacetic Acid Derivativesa
a Reagents and conditions: (a) (COCl)2, then R*OH, Et3N, CH2Cl2; (b) NaBH4, THF; (c) PBr3, CH2Cl2; (d) t-BuOLi, THF, 0 °C; (e)
LiOH, H2O2 THF-H2O.

PPARR/γ Dual Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4459
Scheme 2. Synthesis of Phenols for the Ketone Seriesa
Scheme 4. Synthesis of Phenols for the
Benzisoxazolone Seriesa
a Reagents: (a) R1CO2H, triflic acid, neat; (b) R1MgX, THF; (c)
TPAP/NMO, 4 Å MS, CH2Cl2.
a Reagents and conditions: (a) AcCl, pyridine; (b) (COCl)2, then
R1NHOH; (c) DEAD, Ph3P, THF; (d) NaOH, MeOH-H2O.
Scheme 3. Synthesis of Phenols for the Benzoxazolone
Seriesa
ester group (the corresponding diacid was completely
inactive on PPAR R/γ receptors), our SAR studies began
with the search for surrogates of the ester group in 5.
To this end, a variety of functional groups have been
chosen to try to mimic the ester group (Chart 3).
However, our initial attempt met with only limited
success. Thus, the replacement of the ester group with
an amide, a sulfone or a lactone moiety (compounds 23,
26, 27) led to the complete loss of activity. The relocation
of the ester group from para to meta position (compound
24) also abolished the activity. The substitution of a
gem-difluoropropyl group for the methyl ester (com-
pound 25) retained much of the binding potency but
suffered a considerable loss of activity in transactivation
assay, suggesting that the presence of a carbonyl group
in this region is required for maintaining functional
activity. Fortunately, an ethylcarbonyl group was found
to be able to imitate the original ester group to give
compound 28, which maintained the same binding
affinities for both R and γ receptors as those of com-
pound 5.
With the successful replacement of a ketone for the
ester residue, the structure-activity relationships in-
volving the R1-R4 groups on the ketone-based scaffold
were investigated in more detail (Table 1). In general,
it was found that most active analogues in this series
have inherited the unique in vitro profile of the lead
compound 5, such that they all behaved as “super
agonists” for PPARR and as weak or partial agonists
for PPARγ in transactivation assays despite of having
comparable binding affinities for both R and γ receptors.
With regard to the R1 group, several lipophilic alkyl
groups, but not aryls (compound 31), seemed to be well
tolerated, though they did not offer any advantage over
the ethyl group in terms of pharmacokinetic properties
or in vivo efficacy. On the other hand, quite restrictive
SAR was seen for the R2 and R3 groups attached to both
sides of the phenoxy ring. Thus, operations involving
the removal of the two n-propyl groups or the substitu-
tion of one or both of the n-propyl groups with other
alkyls or halogens resulted in complete or significant
loss of binding activity (32-36), indicating that there
are distinct structural requirements for the two n-propyl
group to be present at both positions. Most notably, the
installation of a lipophilic group at the R4 position
clearly increases the affinity for PPARγ receptor, but
not on the PPARR (e.g. compound 28 vs 38). To have
a Reagents and conditions: (a) NaNO2, AcOH; (b) EtOCOCl,
Et3N, CH2Cl2; (c) H2, Pd/C; (d) R1-I, Cs2CO3; (e) NaOH, MeOH.
by oxidation has been used for the attachment of the
R1 group (Scheme 2).
The construction of the benzoxazolones 13 started
with the nitrosation of R2,R3-substituted resorcinol 18
followed by reductive cyclization of the nitroso com-
pound 19 to give cyclic intermediate 20. Subsequent
N-alkylation with various alkyl halides allowed the
introduction of various R1 groups (Scheme 3).
To prepare the benzisoxazolone phenol 14, we have
developed a novel method based on the use of the
Mitsunobu reagent for cyclization of o-hydroxybenzo-
hydroxamic acid 22.37 Compounds 22 were readily
prepared from the corresponding hydroxybenzoic acids
21 and various R1-substituted hydroxyamines (Scheme
4).
Results and Discussion
In Vitro SAR Studies. Compounds were evaluated
for in vitro potency and subtype selectivity in PPAR
scintillation proximity assays (SPA) and expressed as
IC50 for displacement of a radiolabeled reference com-
pound. Their agonist responses were measured in the
cell-based transactivation assay (TA) using PPAR-GAL4
chimeric receptors and a reporter gene containing a
GAL-4 response element, and expressed as EC50, de-
fined as the concentration of test compound to produce
50% of maximal reporter activity. The maximal activity
of a given compound was measured against that of a
reference full agonist, being defined as 100%.
Since compound 5 itself was not suitable for in vivo
studies due to the presence of an easily hydrolyzable

4460 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Chart 3. The Search for Ester Surrogates
Shi et al.
Table 1. In Vitro Human PPAR Activity of R-(p-Ketophenoxy)phenylacetic Acid
binding IC50 (µM)a
TA EC50 (µM) (max %)b
compd
R/S
X
R1
Et
R2
R3
R4
PPARR
PPARγ
hPPARR
hPPARγ
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
racemic
racemic
racemic
racemic
racemic
racemic
racemic
racemic
racemic
S
O
n-Pr
n-Pr
n-Pr
n-Pr
H
n-Pr
n-Pr
n-Pr
n-Pr
H
s-Bu
Me
Cl
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
H
0.051
0.16
0.13
0.80
>50
2.76
3.16
2.7
0.10 (1233%)
0.25 (1650%)
0.16 (1455%)
(28%)
O
i-Pr
t-Bu
Ph
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
H
(32%)
O
H
(53%)
O
H
4.68
>50
>50
>50
>50
0.26
0.22
0.21
0.22
0.075
2.98
0.20
0.21
>50
>50
-
-
O
H
-
-
-
-
-
O
s-Bu
Me
H
6.12
3.0
1.6
-
O
H
-
O
Cl
H
-
O
Me
i-Pr
i-Pr
Et
0.16
0.26
0.064
0.058
0.023
3.90
0.043
0.089
>15
0.16 (810%)
0.197 (70%)
0.36 (77%)
0.48 (40%)
0.18 (66%)
0.15 (50%)
-
O
Cl
0.26 (1138%)
0.076 (675%)
0.039 (807%)
0.056 (1067%)
S
O
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
S
O
i-Bu
i-Pr
i-Pr
i-PrO
CF3
i-Pr
i-Pr
S
O
R
O
-
S
O
0.11 (479%)
0.25 (1448%)
(59%)
(56%)
-
-
S
O
racemic
racemic
CH2
NH
-
-
>50
a Mean value of three determinations using scintillation proximity assay (SPA). b TA (transactivation assay). Mean value of three
determinations. EC50s were not calculated for compounds whose activity did not reach plateau at the maximal concentration of 3 µM;
instead their percentage responses at a concentration of 3 µM were listed.
more balanced PPARR/γ activity, an isopropyl group
was chosen as the optimized R4 group since it imparts
the best PPARγ activity without sacrificing PPARR
activity
The effect of the stereochemistry at the R-position of
the carboxylic acid was next examined. The opposite
enantiomer of the isopropyl analogue 40, one of the most
active compounds in the series, was prepared according
to Scheme 1 using the S-enantiomer of compound 9 as
chiral auxiliary. The absolute configuration of each
enantiomer was deduced based on mechanistic consid-
eration36 and was further confirmed in the X-ray co-
crystal structure with PPAR-LBD (vide infra). The
binding activities of the two enantiomers were com-
pared. The results clearly showed that the binding
affinity for both R and γ receptor mainly resided in the
S enanantiomer, while the corresponding R enantiomer,
compound 41, was 160-fold and 40-fold less potent on
PPARR and PPARγ, respectively. It should be noted that
none of the analogues described here demonstrated
significant binding to the PPARδ receptor (IC50 > 15
µM).

PPARR/γ Dual Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4461
Scheme 5. Mechanism for Photoinduced Decomposition
of Compound 40
Chart 4. From Ketones to Benzoxazolones and
Benzisoxazolones
In the ketone series, compound 40 appeared to have
the best overall in vitro profile and more consistent
activity on human and murine PPAR receptors. It was
also one of the most efficacious compounds in our in vivo
models for glucose and lipid lowering (vide infra).
However, during preclinical evaluation of 40, we noticed
that in an aqueous solution it showed some noticeable
decomposition after being stored under ordinary labora-
tory light for 2 weeks. Since the decomposition occurred
only under light and the product consisted mainly of
the phenol 12b and aldehyde 46, we believed that the
photoinduced fragmentation at the relatively weak
benzylic ether bond was caused by the presence of the
keto group at the para position which could well serve
as sensitizer for photo activation (Scheme 5).38
To minimize the photoinstability problem, our first
approach was to replace the ether linkage with a more
robust bond such as a carbon-carbon or a carbon-
nitrogen bond. Unfortunately and somewhat surpris-
ingly, none of them was able to imitate the role of the
C-O bond and the corresponding carbo analogue 44
(X ) CH2) and nitrogen analogue 45 (X ) NH) were
found to be devoid of PPAR activity. Thus the benzylic
ether linkage in the current class of compounds ap-
peared be essential for PPAR binding activity.
Our next approach to deal with the stability problem
was to eliminate the photosensitizing keto group in the
ketone series. Given that ketone appeared to be the only
suitable surrogate for the ester residue among several
promising noncyclic structural candidates, as described
previously in Chart 3, our next effort was focused on
the use of some cyclic structures as possible replace-
ments for the ketone moiety. Gratifyingly, we have
found that two alkyl substituted heterocyclic motifs,i.e.,
oxazolones and isoxazolones, when fused into the aro-
matic ring, could mimic efficiently the alkyl ketone
moiety (Chart 4).
Table 2 details the SAR study that led to the optimi-
zation of compound 6 and 7. The R1 substituent on the
nitrogen of each heterocyclic scaffold was investigated
while keeping the R2, R3 and R4 groups the same as
those optimized in the ketone series. The data suggested
that a lipophilic alkyl group at the R1 position was well
tolerated in the benzoxazolone series, though there was
little to be gained from the additional steric bulk (6, 47-
51). On the other hand, increasing the size of R1
significantly eroded the binding activity in the benzisox-
azolone case (56 and 57), and a methyl group was found
to be the best on both PPARR and γ (7). The structure-
activity relationships regarding the R2, R3 and R4 groups
on the two heterocyclic scaffolds were found to be very
much the same as those found in the ketone series.
Thus, the replacement of the n-propyl group at R3 with
chlorine or removal of the alkyl group at R4 reduced
significantly the binding activity. Like the ketone
analogues, compounds in these two heterocyclic series
also exhibited “super” agonist activity for PPARR and
weak or partial agonist activity for PPARγ. On the basis
of their overall in vitro profiles, compound 6 and 7, each
representing the oxazolone and isoxazolone analogues,
were chosen for further in vivo evaluation.
X-ray Crystallography Studies. One of the inter-
esting features displayed by the current class of com-
pounds is that they all showed weak or partial agonist
response toward the hPPARγ receptor despite their
relatively high binding affinity for the receptor and, as
a indication of good cell permeability, a good match of
potency in binding and transactivation assays. To
further understand this interesting phenomenon, com-
pound 7 was chosen to cocrystallize with PPARγ-LDB
(residues Gln203 to Tyr477) as a binary complex with-
out coactivator/corepressor peptides. In this crystal
form, there were two PPARγ molecules in the crystal-
lographic asymmetric unit, one in the activated confor-
mation and one in an unactivated conformation. The
structure was solved at a nominal resolution of 2.50 Å
with an R-factor of 24.4% and R-free of 30.2%. There
was clear electron density for compound 7 in the
activated molecule, but the inactivated binding site was
empty. For the purpose of comparison, the structure
obtained was superimposed on the cocrystal structure
of rosiglitazone with PPARγ-LBD, as shown in Figure
1. Compound 7 binds to the activated receptor with its
carboxyl group forming a hydrogen bonding network
involving Ser289 (helix 3), His323 (helix 6), His449
(helix 11) and Tyr473 (helix12). The isoxazolone ring
sits in a hydrophobic pocket bounded by Ile281, Phe282,
Cys285, Phe363 and Met364 in PPARγ. This site leads
into a larger chamber in PPARγ which has been seen
to accommodate groups the size of benzophenone (data
unpublished). The p-isopropyl-substituted aromatic ring
on 7 points along the path that wraps around behind

4462 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Shi et al.
Table 2. In Vitro Human PPAR Activity of Benzoxazolone and Benzisoxazolone Analogues
IC50 (µM)
EC50 (µM) (max %)
compd
ring
R1
R2
R3
R4
PPARR
PPARγ
hPPARR
(680%)
0.29 (440%)
(218%)
0.14 (508%)
(63%)
0.31 (123%)
(1221%)
(861%)
(571%)
hPPARγ
47
6
oxz
oxz
oxz
oxz
oxz
oxz
oxz
oxz
oxz
ixz
ixz
ixz
ixz
ixz
CH3
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
Cl
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
H
Cl
i-Pr
H
i-Pr
i-Pr
i-Pr
i-Bu
0.393
0.074
0.15
0.085
0.099
0.087
0.051
0.12
0.11
1.79
0.22
0.52
3.8
0.21
0.16
0.44
0.26
0.64 (42%)
0.12 (36%)
0.12 (54%)
0.14(38%)
0.33 (36%)
0.14 (38%)
(27%)
CH3CH2
CH3CH2CH2
(CH3)2CH
cyclopropylmethyl
CF3CH2
i-Pr
CH3CH2
CH3CH2
CH3
48
49
50
51
52
53
54
55
7
0.072
0.13
0.10
0.55
0.26
(57%)
1.41
0.36 (48%)
0.39 (41%)
0.39 (41%)
0.43 (38%)
(23%)
0.85
0.67 (1005%)
0.67 (1005%)
0.27 (889%)
(20%)
CH3
0.074
0.23
56
57
58
(CH3)2CH
cyclohexyl
i-Pr
>15
0.79
(382%)
0.37 (35%)
a Mean value of three determinations using scintillation proximity assay (SPA). b TA (transactivation assay). Mean value of three
determinations. EC50s were not calculated for compounds whose activity did not reach plateau at the maximal concentration of 3 µM;
instead their percentage responses at a concentration of 3 µM were listed.
Table 3. Pharmacokinetic Properties of Compound 40, 6 and 7
compds species bioavailability (%) AUCa (µM h) t1/2 (h)b
40
rat
dog
rat
dog
rat
dog
65
85
72
35
73
73
153.7
70.5
127.2
9.1
67.5
26.9
6.1
2.5
6.0
1.2
7.2
1.6
6
7
a Dosed at 2 mg/kg po. b Dose 0.5 mg/kg iv. See Experimental
Section for details.
Pharmacokinetic and in Vivo Studies. Three
representative compounds (40, 6, and 7) have been
chosen for extensive in vivo evaluation. First, their
pharmacokinectic parameters were examined in rats
and dogs. As shown in Table 3, when dosed orally, all
three compounds exhibited good exposure with excellent
bioavailability and a good half-life.
Figure 1. Superimposed crystal structures of the PPARγ-LBD
in complex with compound 7 (grey) and rosiglitazone (green).
Protein is depicted as red helices, cyan beta sheet and white
random coil or turns. All residues within 3 Å of the ligand are
shown.
Next, the in vivo pharmacology of compounds 6 and
7 was studied in a well-established rodent model of
diabetes, using male db/db mice (10-11 week-old
C57BLKS/J-m +/+Leprdb mice), which is characterized
by severe insulin resistance, hyperglycemia and marked
hyper-triglyceridemia. As depicted in Figure 2, when
dosed orally at 3 and 10 mg/kg by gavage once daily for
11 days, the two compounds demonstrated robust dose-
related serum glucose-lowering efficacy (-69%, -95%
for 6 at 3, 10 mpk and -58% for 7 at 10 mpk) that is
either superior or comparable to the marketed PPARγ
agonist rosiglitazone dosed at 10 mg/kg (-70% glucose
reduction). As described above, both compounds 6 and
7 had PPARγ binding potency similar to that of rosigli-
tazone (PPARγ IC50 ∼ 0.20 µM) but weaker agonist
responses in the transactivation assay. Although it is
not clear what fraction of the observed antihyperglyce-
mic activity of these compounds is contributed by their
PPARγ activity, the present results and the recent
report of PPARR-mediated insulin sensitization19,20 sug-
helix 3, used by nearly all of the agonists whose complex
structures have been solved.27,39-41 However, unlike
rosiglitazone, no part of 7 extends past the Cys285/
Met364 boundary. Overall, these results suggested that
simply forming a direct link to helix 12, although
necessary, is not sufficient for full agonism. There may
be some other aspect of compound 7, such as compound
size, shape and position in the ligand binding pocket
that is unable to affect PPARγ in the same manner as
rosiglitazone.
Although we do not have the corresponding structure
of compound 7 with the PPARR receptor at this moment,
it can be certain that compound 7 will also bind to the
PPARR receptor in somewhat different fashion from a
typical full agonist because of its relatively compact
structure. In this context, the interesting question of
why its interaction with the R receptor elicited an
unusually higher agonist response in the PPARR-GAL4
transactivation assay remains to be answered.

PPARR/γ Dual Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4463
Table 5. Effects of Compounds 40 and 7 on Serum Cholesterol
in Dogsa
compound
dose (mpk)
change (%)b
fenofibrate
20
-4.7
-23.5
-17.5
-17.3
fenofibrate
100
10
40
7
10
a Male dogs being fed a normal chow diet were orally dosed daily
with the test compounds at the indicated doses for 15 days. See
Experimental Section for details. b Mean value (n ) 5). c p < 0.05
vs vehicle control.
to lower cholesterol by 16% when dosed at 4 mg/kg.42
These results clearly indicated that our PPARR/γ dual
agonists can provide comparable cholesterol lowering
efficacy to that achieved by fibrate drugs at much high
doses or by statin drugs at similar dose levels.
Figure 2. Effects of test compound on plasma glucose levels
in db/db mice. Male db/db mice and lean mice were dosed daily
for 11 days by gavage with vehicle or the indicated doses of
test compounds. Each data point represents the mean value
((SD) of seven individual mice. Data for compound 6 at 3 mpk
were taken from a separate study.
To assess the typical PPARγ agonist-mediated side
effects, compounds 6 and 7 were tested in a two week
normal Sprague-Dawley (SD) rat tolerability study
using the PPARγ full agonist rosiglitazone as a positive
control. As shown in Table 6, the typical effects that
include brown adipose tissue (BAT) proliferation, heart
weight increases and hematocrit decrease were not seen
with the current three compounds. The only major
finding was an expected PPARR driven liver weight
increase and some body weight decrease. In contrast,
rosiglitazone increased heart weight by 22% and BAT
depot mass 260% and decreased hematocrit by 12.5%.
The lack of these PPARγ-related side effects with
compounds 6 and 7 could well be due to the fact that
these compounds possessed only partial PPARγ agonist
activity, thereby raising the possibility that such agents
may exert their PPARγ-mediated beneficial in vivo
metabolic effects without provoking the undesirable side
effects. Alternatively, the lack of PPARγ-mediated side
effects may suggest that their in vivo pharmacology may
have been largely driven by their PPARR activity.
Table 4. Effects of Compounds 40, 6 and 7 on Serum
Cholesterol and Triglyceride in Hamstera
dose
(mpk)
cholesterol change
TG
change
(%)c
compd
(mg/dL)b
(%)c
(mg/dL)b
vehicle
128
91
331
244
259
236
-
fenofibrate
100
3
10
10
10
-29
-32
-33
-25
-17
-26
-22
-29
-6
40
40
6d
7
88
86
-
106
303
-9
a Male Golden Syrian hamsters fed a normal rodent chow were
orally dosed for 9 days with the test compounds. See Experimental
Section for details. b Mean value (n ) 10). c p < 0.01 vs vehicle
control. d The data for this compound is obtained from a separate
study where fenofibrate at 100 mpk showed -41% and -23%
reductions in cholesterol and TG, respectively.
gest that a dual PPARR/γ agonist can provide better
glucose control than currently marketed selective PPARγ
agents.
To determine the ability of the current class of
compounds to impact lipid homeostasis, male Syrian
hamster and male Beagle dogs were used to investigate
the effects of compounds 40, 6 and 7 on the serum levels
of triglyceride and total cholesterol. The former pre-
clinical animal model faithfully predicts the clinical lipid
modulating efficacy of fibric acid derivatives and the
latter of fibric acid derivatives and statins.42 As shown
in Table 4, when administered orally to hamsters (N )
6) daily for 14 days compound 40 achieved superior
cholesterol and triglyceride lowering compared to fenofi-
brate even at significantly lower dose. The apparent lack
of dose response regarding the cholesterol reduction by
40 may suggest either a saturation of the response or a
similar drug exposure at two doses. Further pharma-
cokinetic measurement should help to clarify this. The
less robust lipid decrease by compound 6 and 7 may
partly be due to their less potent hamster PPARR
activity (6: EC50 ) 0.96 µM; 7: EC50 ) 0.45 µM)
compared with that of compound 40 (EC50 ) 0.085 µM).
In male Beagle dogs, good lipid lowering efficacy was
again achieved by both compound 40 and 7 (Table 5),
though 40 was somewhat more potent (EC50 ) 0.13 µM)
than 7 (EC50 ) 0.60 µM) in dog PPARR transactivation
assay. It is worth noting that in the same dog model,
the HMG CoA reductase inhibitor Simvastatin was able
Conclusion
In summary, we have described the synthesis and
structure-activity studies of new series of R-aryloxy-
phenylacetic acid derivatives as a novel class of PPARR/γ
dual agonists. Unlike most PPARR/γ dual agonists
disclosed in the literature, these compounds are char-
acterized by their balanced binding affinity for both
PPARR and γ receptors along with a “super” agonist
activity in PPARR-GAL4 TA assay and a weak or
partial agonist activity toward PPARγ. The X-ray
crystallographic analysis of cocrystal structures of
PPARγ-LBD complexed respectively with compound 7
and rosiglitazone revealed a similarity as well as some
significant differences in their interactions with the
receptor. These differences were used to explain why
most compounds described here elicited a partial agonist
response toward PPARγ despite of their relatively high
affinity for the receptor. Most notably, the three com-
pounds chosen for in vivo evaluation demonstrated
excellent anti-hyperglycemic efficacy in the db/db mouse
and hypolipidemic activity in hamster and dog models,
without provoking the typical PPARγ-mediate toxicity
in rat tolerability model. These results suggest that
further efforts to develop these novel PPARR/γ dual
agonists as improved therapies for type 2 diabetes,

4464 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Shi et al.
Table 6. PPARγ-Mediated Effects of Rosiglitazone and Compounds 6 and 7
vehicle
rosiglitazone
compound 6
compound 7
150
0.24 ( 0.02
1.14 ( 0.08
20.3 ( 2.37
45.7 ( 2.6
278 (13
dose (mg/kg)
-
150
150
BAT (g)b
0.28 ( 0.05
1.19 ( 0.07
15.0 ( 0.94
47.2 ( 1.2
321 ( 13
1.03 ( 0.32
1.46 ( 0.07
16.6 ( 1.66
41.3 ( 2.4
322 ( 13
0.24 ( 0.04
1.10 ( 0.08
22.4 ( 3.13
45.6 ( 3.2
283 ( 19
heart weight (g)b
liver weight (g)
hematocrit (%)b
body weight (g)b
a Male normal rat fed a rodent chow with test compounds at the indicated doses for 2 weeks. For details, see the Experimental Section.
b Mean value ( SD. Values in bold are those that are statistically different versus vehicle control.
by the addition of 0.5 M formate buffer, pH 3.0, and stored at
-70 °C. Quantitative analysis was carried out with LC-MS/
MS using the PE Sciex API 3000 triplex quadruple mass
spectrometer.
dislipidemia and other aspects of metabolic syndrome
are indeed warranted.
Experimental Section
Synthetic Materials and Methods. Reagents and sol-
vents were obtained from commercial suppliers and used
without further purification. Flash chromatography was per-
formed using E. Merck silica gel (230-400 mesh). 1H NMR
nuclear magnetic resonance (NMR) spectra were recorded in
the deuterated solvents specified on a Varian Unity INOVA
500 MHz instrument. Chemical shifts are reported in ppm
from the tetramethylsilane resonance in the indicated solvent
(TMS: 0.0 ppm). The mass spectrum was measured using
HP1100 and Micromass ZQ instruments (LC-MS system; a 4.6
× 50 mm RP-18 column and a gradient solvent system of 10-
100% acetonitrile in water with 0.1% trifluoroacetic acid were
used for HPLC and a positive electrospray ionization mode
(ESI) for MS). Elemental analyses were obtained from Rob-
ertson Microlit Laboratories (Madison, NJ).
In Vitro Assays. Activities of compounds were evaluated
for both binding affinity and functional activity. First, binding
affinities for the PPARs were measured in a scintillation
proximity assay (SPA).43 Second, potencies of PPAR gene
activation were evaluated in cell-based transcription assays
using GAL4-PPAR chimeric receptors as previous described.44
All results were produced in triplicate, and mean values are
reported. Generally, compound synthesis was driven by the
SAR developed from these two assays.
In Vivo db/db Mouse Studies. db/db (C57BLKS/J-m
+/+Leprdb) mice were used as a type 2 diabetic animal model.
Male db/db mice at 12-13 weeks of age and nondiabetic db/+
(lean) mice from Jackson Laboratories were housed seven mice
per cage and fed a rodent chow (Purina no. 5001) ad libitum
with free access to water. Mice (seven per group) received a
once daily oral dosing of test compounds with vehicle (0.25%
methylcellulose) by oral gavage for 11 days. Blood was collected
from the tail immediately prior to the next dosing at days 0,
4, 7 and 11 for measurement of the plasma glucose levels.
Hamster Lipid Studies. Golden Syrian hamsters weighing
between 120 and 150 g were purchased from Charles River
Laboratories and used for the experiments. Hamsters were
housed in boxes (five per box) and fed a normal rodent chow
ad libitum with free access to water. Hamsters (10 for each
group) were orally dosed with compounds (suspended in 0.5%
methylcellulose) for 9 days. On the morning of the 10th day,
hamsters were euthanized with carbon dioxide, and blood
samples were obtained via heart puncture. Serum cholesterol
and triglyceride levels were determined from the samples.
Dog Lipid Studies. Mature male Beagle dogs weighing
between 12 and 18 kg were purchased from Marshall Farm,
PA. They were housed individually and fed a cholesterol-free
chow diet ad libitum with free access to water. Prior to starting
experiments, the dogs (five for each group) were bled weekly
from the jugular vein and their serum cholesterol levels were
determined. Test compounds were suspended in 0.5% meth-
ylcellulose and gavaged daily to the dogs for 2 weeks. Blood
samples were taken during and after the dosing periods, and
serum cholesterol levels were determined.
Rat Tolerability Studies. Male Sprague-Dawley rats
weighing about 200 g from Charles River were house 3 rats
per cage and provided free access to water and rodent chow.
Rats (n ) 6) were orally dosed once daily by gavage with test
compounds at indicated doses in 0.5% methylcellulose. Body
weights were measured prior to each dosing on days 1, 4, 6 8,
11 and 14. Animals were euthanized via carbon dioxide 24 h
following the last dose and the indicated tissue and organs
were removed and weighed. Blood samples were drawn via
heart puncture and assayed for serum chemistry.
Pharmacokinetic Studies. Male Sprague-Dawley rats (n
) 3), male adult Beagle dogs (n ) 4), and male adult Rhesus
monkeys (n ) 3) that had been fasted overnight received an
oral gavage dose of 2 mg/kg, or an intravenous dose of 0.5 mg/
kg by bolus injection. Blood samples were obtained from the
femoral arterial cannula for rat, and the jugular vein for dog,
at designated time points into heparin-containing tubes. The
plasma was prepared immediately by centrifugation, acidified
1-(4-Hydroxy-3,5-dipropylphenyl)propan-1-one (12b).
A suspension of sodium propanoate (1.92 g, 20 mmol) and 2,6-
di-propylphenol (1.78 g, 10 mmol) in neat triflic acid (10 mL)
was heated at 50 °C for 1 h. The reaction mixture was poured
into ice water and extracted with ethyl acetate (2 × 30 mL).
The extract was washed with brine (2 × 30 mL) and aqueous
sodium bicarbonate (30 mL), dried over MgSO4 and concen-
trated. The residue was purified by chromatography on silica
gel eluting with a 1:9 mixture of ethyl acetate:hexanes to give
1
1.98 g (85% yield) of 12b as a white solid. H NMR (CDCl3,
500 MHz) δ 7.66 (s, 2H), 5.19 (br.s, 1H), 2.96 (q, J ) 7.3 Hz,
2H), 2.62 (t, J ) 7.7 Hz, 4H), 1.68 (m, 4H), 1.22 (t, J ) 7.3 Hz,
3H), 1.01 (t, J ) 7.3 Hz, 6H). MS (ESI) 235.11 (MH+).
3-Ethyl-6-hydroxy-5,7-dipropyl-1,3-benzoxazol-2(3H)-
one (13a). Step 1. Preparation of Ethyl 4-nitroso-2,6-
dipropyl-1,3-phenylene Biscarbonate. To a solution of 2,6-
dipropylresorsinol (9.7 g, 5.0 mmol) in acetic acid (50 mL) and
ethanol (250 mL) cooled at -20 °C was added sodium nitrite
(8.6 g, 125 mmol). The resulting suspension was stirred and
warmed to room temperature overnight. The reaction mixture
was diluted with ethyl acetate (500 mL) and washed succes-
sively with brine and a saturated solution of sodium bicarbon-
ate. The organic phase was dried over MgSO4 and filtered. The
filtrate was treated with ethyl chloroformate (10.1 mL, 100
mmol) and triethylamine (22 mL, 150 mmol) at room temper-
ature for 1 h and then concentrated to a small volume. The
precipitated salt was removed, and the residue was purified
by chromatography on silica gel eluting with a 8:2 mixture
hexane:ethyl acetate to give 16.5 g (90% yield) of the title
compound as a yellowish oil. 1H NMR (CDCl3, 400 MHz) δ 7.39
(s, 1H), 4.38 (m, 4H), 2.42-2.49 (m, 4H), 1.40-1.60 (m, 4H),
1.42 (t, J ) 7.5 Hz, 3H), 1.41 (t, J ) 7.5 Hz, 3H), 0.98 (t, J )
7.5 Hz, 3H), 0.95 (t, J ) 7.5 Hz, 3H). MS (ESI) 368.31 (MH+).
Step 2. 6-Hydroxy-5,7-dipropyl-1,3-benzoxazol-2(3H)-
one. A solution of the product from Step 1 (16.5 g, 45 mmol)
in ethyl acetate (250 mL) and 10% palladium on carbon (1.7
g) was stirred under hydrogen (1 atm) for 16 h. The reaction
mixture was filtered through a pad of Celite, and the filtrate
was concentrated. The residue was dissolved in DMF (250 mL)
and stirred at 25 °C with Cs2CO3 (15 g, 45 mmol) for 4 h. The
mixture was poured into water and extracted with ethyl
acetate. The organic phase was washed with water, dried over
MgSO4 and concentrated. The residue was subjected to chro-

PPARR/γ Dual Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4465
matography on silica gel eluting with a 7:3 mixture of hexane:
ethyl acetate to give 9.0 g (80% yield) of the title compound as
carbonate (200 mL). After 30 min of vigorous stirring, the
reaction mixture was acidified with 2 N hydrochloric acid to
pH 2 and extracted with ethyl acetate. The organic layer was
washed with brine, dried over MgSO4 and concentrated. The
residue was purified by chromatography on silica gel eluting
with a 7:3 mixture of hexane:ethyl acetate to give 20 g (65%
1
a solid. H NMR (CDCl3, 400 MHz) δ 8.70 (br.s, 1H), 6.70 (s,
1H), 4.60 (s, 1H), 2.78 (t, J ) 7.5 Hz, 2H), 2.60 (t, J ) 7.5 Hz,
2H), 1.50-1.70 (m, 4H), 0.99 (t, J ) 7.5 Hz, 3H), 0.97 (t, J )
7.5 Hz, 3H). MS (ESI) 236.30 (MH+).
1
yield) of the title compound as a brown oil. H NMR(CDCl3,
Step 3. 3-Ethyl-6-hydroxy-5,7-dipropyl-1,3-benzoxazol-
2(3H)-one (13a). To a solution of the product from Step 2 (8.2
g, 35 mmol) and pyridine (4.2 mL) in dichloromethane (150
mL) cooled at -30 °C was added dropwise acetyl chloride (2.6
mL, 36.5 mmol). The reaction was warmed to room tempera-
ture over 1 h and poured into 0.5 N hydrochloric acid. The
organic phase was separated, washed with brine, dried over
MgSO4 and concentrated. The residue was dissolved in DMF
(150 mL) and stirred with Cs2CO3 (15 g, 46.5 mmol) at room
temperature for 30 min. Ethyl iodide (5.6 mL, 70 mmol) was
added, and stirring was continued for additional 1 h. The
mixture was poured into water and extracted with ethyl
acetate. The organic phase was washed with water, dried over
MgSO4 and concentrated. The residue was taken up in
methanol (100 mL) and treated with 2 N KOH (20 mL) at room
temperature for 30 min. The mixture was neutralized with
concentrated hydrochloric acid and concentrated. The residue
was purified by chromatography on silica gel eluting with a
8:2 mixture of hexane:ethyl acetate to give 6.9 g (75% yield)
400 MHz): δ 7.05 (s, 1H), 3.59 (s, 3H), 2.46-2.60 (m, 2H),
2.30-2.41 (m, 2H), 2.38 (s, 3H), 1.41-1.56 (m, 4H), 0.98 (t, J
) 7.2 Hz, 3H), 0.96 (t, J ) 7.2 Hz, 3H). MS (ESI): 310.22
(MH+).
Step 4. 6-Hydroxy-2-methyl-5,7-dipropyl-1,2-benzisox-
azol-3(2H)-one (14a). The product from Step 3 (3.1 g, 10
mmol) and triphenylphosphine (3.9 g, 15 mmol) were dissolved
in dry THF (100 mL). To the resulting solution was added
dropwise diethyl azodicarboxylate (2.6 g, 15 mmol). After being
stirred at 25 °C for 30 min, the mixture was quenched with a
1:1 mixture of methanol:acetic acid (1 mL) and concentrated.
The residue was triturated with 1:1 hexane:diethyl ether and
filtered through a pad of silica gel. The filtrate was concen-
trated, and the residue was dissolved in methanol (50 mL) and
treated with 2 N NaOH (10 mL) for 30 min. The solution was
neutralized with 2 N hydrochloric acid and evaporated to
dryness under reduced pressure. The residue was purified by
chromatography on silica gel eluting with a 7:3 mixture of
hexane:ethyl acetate to give 2.1 g (86% yield) of the title
1
of the title product as a solid. H NMR (CDCl3, 500 MHz) δ
1
6.59 (s, 1H), 4.81(br.s, 1H), 3.82(q, J ) 7.5 Hz, 2H), 2.77 (t, J
) 7.5 Hz, 2H), 2.60 (t, J ) 7.5 Hz, 2H), 1.60-1.70 (m, 4H),
1.36 (t, J ) 7.5 Hz, 3H), 0.99 (t, J ) 7.5 Hz, 3H), 0.95(t, J )
7.5 Hz, 3H). MS (ESI) 264.23 (MH+).
compound. H NMR (CDCl3, 400 MHz): δ 7.43 (s, 1H), 6.50
(s, 1H), 0.3.63 (s, 3H), 2.77 (t, J ) 7.1 Hz, 2H), 2.63 (t, J ) 7.1
Hz, 2H), 1.65 (m, 4H), 0.99 (t, J ) 7.5 Hz, 3H), 0.97 (t, J ) 7.5
Hz, 3H). MS (ESI) 250.13 (MH+).
6-Hydroxy-2-methyl-5,7-dipropyl-1,2-benzisoxazol-
3(2H)-one (14a). Step 1. Preparation of Methyl 2,4-
Dihydroxy-3,5-diallylbenzoate. Methyl 2,4-bis(allyloxy)-
benzoate (37.2 g, 150 mmol), obtained from allylation of methyl
2,4-dihydroxybenzoate with allyl bromide, was dissolved in
1,2,4-trichlorobenzene (150 mL). The solution was heated
under reflux for 7 h and then cooled to 25 °C. The reaction
mixture was poured on the top of a column of silica gel and
eluted first with hexane and then with 1:9 mixture of hexane:
ethyl acetate to give 30 g (81%yield) of the title compound as
Phenyl(4-propionyl-2,6-dipropylphenoxy)acetic Acid
(28). A mixture of the phenol 12b (1.7 g, 5.0 mmol), methyl
R-bromophenylacetate (1.4 g, 6.0 mmol) and CsCO3 (4.9 g, 7.5
mmol) in DMF (25 mL) was stirred at room temperature for 2
h. The reaction mixture was poured into water and extracted
with ethyl acetate. The organic phase was washed with water,
dried over MgSO4 and concentrated. The residue was taken
up in methanol (25 mL) and treated with 2 N NaOH (5 mL)
for 30 min. The mixture was acidified with 2 N hydrochloric
acid and concentrated under reduced pressure. The residue
was purified by chromatography on silica gel eluting with a
7:3 mixture of hexane:ethyl acetate containing 0.1% acetic acid
1
an oil. H NMR (CDCl3, 400 MHz): 11.0 (s, 1H), 7.58 (s, 1H),
5.93-6.06 (m, 2H), 5.75 (s, 1H), 5.10-5.21 (m, 4H), 3.96 (s,
3H), 3.52 (m, 2H), 3.36 (m, 2H).
1
to give 1.56 g (85% yield) of compound 28. H NMR (CD3OD,
500 MHz) δ 7.65 (s, 2H), 7.49 (m, 2H), 7.40 (m, 3H) 5.18 (s,
1H) 2.99 (q, 2H, J ) 7.3 Hz), 2.44 (m, 4H), 1.51 (m, 4H) 1.15
(t, 3H, J ) 7.3 Hz) 0.83 (t, 6H, J ) 7.3 Hz). MS (ESI) 369.10
(MH+).
Step 2. Preparation of 2,4-Dihydroxy-3,5-dipropylben-
zoic Acid. A solution of the product from Step 1 (30 g, 121
mmol) in ethyl acetate (500 mL) and palladium on carbon
(10%, 1.5 g) was stirred under hydrogen (1 atm) for 6 h. The
catalyst was filtered off through a pad of Celite, and the filtrate
was concentrated. The residue was dissolved in methanol (500
mL) and treated with 3 N NaOH (80 mL) under reflux for 16
h. The reaction mixture was concentrated and the residual
aqueous phase was neutralized with 2 N HCl to pH 2. The
precipitated solid was collected by filtration, washed briefly
with water and cold methanol, and dried under vacuum to give
24 g (83% yield) of the title compound as a white solid. 1H
NMR (CDCl3, 400 MHz): 7.01 (s, 1H), 5.2 (s, 1H), 2.68 (t, J )
7.5 Hz, 2H), 2.52 (t, J ) 7.5 Hz, 2H), 1.60-1.70 (m, 4H), 0.99
(t, J ) 7.5 Hz, 3H), 0.96 (t, J ) 7.5 Hz, 3H). MS (ESI) 239.20
(MH+).
Step 3. N-Methyl-4-acetoxy-3,5-dipropyl-6-hydroxy-
benzohydroxamic Acid. The acid from Step 2 (24 g, 100
mmol) and acetyl chloride (21 mL, 300 mmol) were dissolved
in dichloromethane (250 mL). Pyridine (35.0 mL, 432 mmol)
was added dropwise at 0 °C. The reaction mixture was warmed
to 25 °C over 1 h and poured into 0.5 N hydrochloric acid (250
mL). The separated organic layer was washed with brine, dried
over MgSO4 and concentrated. The residue was dissolved in
dichloromethane (50 mL) and treated with oxalyl chloride (16.4
mL, 200 mmol) and 2 drops of DMF under reflux for 1 h. All
the volatiles were then removed by azeotroping with toluene
(2 × 50 mL). The residue was redissolved in dichloromethane
(400 mL) and added to a vigorously stirred biphasic mixture
containing N-methylhydroxyamine hydrochloride (25 g, 300
mmol), diethyl ether (200 mL) and 2 N aqueous sodium
(2R)-1-Oxo-1-pyrrolidin-1-ylpropan-2-ol (9). A mixture
of isobutyl (R)-(+)-lactate (25 mL, 166 mmol) and pyrrolidine
(27.7 mL, 332 mmol) were kept at 25 °C for 3 days. The
volatiles was then removed in a rotavapor, and the residue
was azeotroped with toluene (2 × 100 mL) to give essentially
1
pure compound 5 as a brown oil. H NMR (CDCl3, 500 MHz)
δ 4.90 (q, J ) 6.9 Hz, 1H), 3.61-3.72 (m, 2H) 3.35-3.50 (m,
2H), 1.53 (m, 2H) 1.42 (m, 2H), 1.32 (d, J ) 6.9 Hz, 3H).
(4-Isopropylphenyl)(oxo)acetic Acid (8, R4 ) i-Pr). At
0 °C, aluminum trichloride (26.7 g, 0.20 mmol) was added to
a solution of isopropylbenzene (24.0 g, 0.20 mole) and ethyl
chlorooxoacetate (41.0 g, 0.30 mol) in dichloromethane (0.50
L). The resulting mixture was stirred at 25 °C for 2 h and then
poured into 0.5 N hydrochloric acid (0.50 L). The organic layer
was separated, and the aqueous phase was extracted with
dichloromethane (2 × 200 mL). The combined organic layers
were washed successively with brine (300 mL) and saturated
aqueous sodium bicarbonate (300 mL), dried over MgSO4 and
concentrated. The residue was taken up in methanol (500 mL)
and treated with 2 N NaOH (83 mL) for 30 min. Methanol
was removed under reduced pressure, and the remaining
aqueous phase was extracted with diethyl ether and acidified
with 2 N hydrochloric acid to pH 2. The precipitated product
was collected, washed with water and air-dried to give 22 g of
1
crude compound 8 (R ) i-Pr). H NMR (CDCl3, 400 MHz) δ
8.1 (d, J ) 8.0 Hz, 2H), 7.31 (d, J ) 8.0 Hz, 2H), 3.20 (m, 1H),
1.25 (d, J ) 7.2 Hz, 6H).

4466 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Shi et al.
(1R)-1-Methyl-2-oxo-2-pyrrolidin-1-ylethyl (4-isoprop-
ylphenyl)(oxo)acetate (10, R4 ) i-Pr). A solution of the
crude keto acid 8 (22.0 g) and oxalyl chloride (19.9 mL, 228
mmol) in dichloromethane (100 mL) was treated with 2 drops
of DMF and refluxed for 30 min. All the volatiles were then
distilled off, and the residue was azeotroped with toluene (100
mL) under reduced pressure. The residue (27.5 g) was mixed
with the (R)-lactamide 9 (16.3 g, 114 mmol) in dichloromethane
(200 mL) at 0 °C and treated with triethylamine (31.0 mL,
228 mmol). The reaction mixture was warmed to 25 °C over
30 min and poured into 0.5 N hydrochloric acid (300 mL). The
organic layer was separated, and the aqueous phase was
extracted with dichloromethane (100 mL). The combined
organic phases were washed with brine, dried over MgSO4 and
concentrated. The residue was subjected to chromatography
on silica gel eluting with a 1:1 mixture of hexane:ethyl acetate
to give 28.9 g (80% yield) of compound 10 (R ) i-Pr) as a white
Hz, 2H), 6.60 (s, 1H),5.29 (q, J ) 7.5 Hz, 1H) 5.16 (s, 1H),
3.85 (q, J ) 7.5 Hz, 2H), 3.52-3.63 (m, 2H), 3.45 (m, 1H), 3.35
(m, 1H), 2.98 (m, 1H), 2.35-2.56 (m, 4H) 1.82-2.0 (m, 4H),
1.40-1.60 (m, 4H), 1.38 (t, J ) 7.5 Hz, 3H), 1.36 (d, J ) 7.5
Hz, 3H), 1.28 (d, J ) 7.5 Hz, 6H), 0.85 (t, J ) 7.5 Hz, 3H),
0.78 (t, J ) 7.5 Hz, 3H). MS (ESI): 565.7 (MH+).
(1R)-1-Methyl-2-oxo-2-pyrrolidin-1-ylethyl (2S)-(4-iso-
propylphenyl)[(2-methyl-3-oxo-5,7-dipropyl-2,3-dihydro-
1,2-benzisoxazol-6-yl)oxy]acetate (15c: R4 ) i-Pr; ArOH
) 14a): 95% yield using compound 14a as the phenol. 1H NMR
(400 MHz, CDCl3): δ 7.45 (s, 1H), 7.40 (d, J ) 7.5 Hz, 2H),
7.25 (d, J ) 7.5 Hz, 2H), 5.28 (s, 1H), 5.27 (q, J ) 7.1 Hz, 1H),
3.62 (s, 3H), 3.40-3.56 (m, 4H), 2.95 (m, 1 H), 2.36-2.54 (m,
4H), 1.82-1.95 (m, 4H), 1.40-1.60 (m, 5H) 1.37 (d, J ) 7.1
Hz, 3H), 1.24 (d, J ) 7.2 Hz, 6H), 0.83 (t, J ) 7.5 Hz, 3H),
0.81(t, J ) 7.5 Hz, 3H). MS (ESI) 551.4 (MH+).
General Procedure B: Hydrolysis of the Coupling
Products 15. The coupling product obtained by General
Procedure A (3.3 mmol) in THF (5.0 mL) was added to a
mixture of 1 N LiOH (6.9 mL, 6.9 mmol) and 30% hydrogen
peroxide (5.0 mL) in THF (20 mL) cooled at 0 °C. The reaction
mixture was stirred at 0 °C for 2 h and then acidified with 2
N hydrochloric acid to pH 2. The organic layer was separated,
and the aqueous layer was extracted with ethyl acetate (30
mL). The combined organic phase was washed with a 2 N
solution of Na2SO3, dried over MgSO4 and concentrated. The
residue was purified by chromatography on silica gel using a
3:7 mixture of ethyl acetate:hexanes containing 1% of acetic
acid as the eluent. The enantiomeric purity was determined
by HPLC using a Cyclobond 2000 column (4.6 × 250 mm) and
a solvent system of methanol:acetonitrile:acetic acid (20: 80:
1) at a flow rate of 1.5 mL/min.
1
solid. H NMR (CDCl3, 400 MHz) δ 8.1 (d, J ) 8.0 Hz, 2H),
7.31(d, J ) 8.0 Hz, 2H), 5.39 (q, J ) 7.0 Hz, 1H), 3.60-3.79
(m, 2H), 3.40-3.55 (m, 2H), 3.20 (m, 1H), 1.90-2.12 (m, 4H),
1.60 (d, J ) 7.0 Hz. 3H), 1.26 (d, J ) 7.2 Hz, 6H). MS (ESI)
318.20 (MH+).
(1R)-1-Methyl-2-oxo-2-pyrrolidin-1-ylethyl Bromo(4-
isopropylphenyl)acetate (11, R ) i-Pr). To a solution of
the keto ester 10 (R4 ) i-Pr, 5.08 g, 16.0 mmol) in dry THF
(100 mL) cooled at 0 °C was added sodium borohydride (0.302
g, 8.0 mmol). The reaction mixture was stirred at 0 °C for 30
min and then poured into a cold mixture of brine (50 mL) and
2 N hydrochloric acid (4 mL). The organic layer was separated,
and the aqueous phase was extracted with ethyl acetate. The
combined organic layers were washed with brine, dried over
MgSO4 and concentrated. The residue (4.6 g) was dissolved in
dichloromethane (50 mL) at 25 °C and treated with phosphorus
tribromide (1.37 mL, 14.5 mmol) at 25 °C. After 30 min, the
mixture was poured into brine (100 mL), and the separated
organic phase was washed with brine, dried over MgSO4 and
concentrated. The residue was purified by chromatography on
silica gel eluting with a 1:1 mixture of hexane:ethyl acetate
to give 5.0 g (82% yield) of compound 11 (R ) i-Pr) as a 1:1
mixture of diastereomers. 1H NMR (CDCl3, 500 MHz): δ? 7.50
(d, J ) 8.5 Hz, 0.5 × 2 H), 7.49 (d, J ) 8.5 Hz, 0.5 × 2 H),
7.20-7.26 (m, 2H), 5.42 (s, 1 H), 5.29 (q, J ) 7.0 Hz, 0.5 × 1
H), 5.27 (q, J ) 7.0 Hz, 0.5 × 1 H), 3.30-3.60 (m, 4H), 2.9 (m,
1 H), 1.80-2.0 (m, 4H), 1.50 (d, J ) 7.0 Hz, 0.5 × 3H), 1.45 (d,
J ) 7.0 Hz, 0.5 × 3H), 1.25 (d, J ) 7.2 Hz, 0.5 × 6 H), 1.24 (d,
J ) 7.2 Hz, 0.5 × 6H). MS (ESI) 382.30, 384.21 (MH+).
(2S)-(4-Isopropylphenyl)(4-propionyl-2,6-dipropyl-
phenoxy)acetic acid (40): 92% yield and 96% ee from
1
hydrolysis of compound 15a. H NMR (500 MHz, CD3OD) δ
7.64 (s, 2H), 7.38 (d, J ) 8.2 Hz, 2H), 7.27 (d, J ) 8.2 Hz, 2H),
5.13 (s, 1H), 2.99 (q, J ) 7.2 Hz, 2H), 2.92 (m, 1H) 2.38-2.50
(m, 4H), 1.38-1.60 (m, 4H), 1.26 (d, J ) 6.8 Hz, 6H), 1.15 (t,
J ) 7.2 Hz, 3H), 0.82 (t, J ) 7.3 Hz, 6H). MS (ESI): 411.20
(MH+). Anal. (C26H34O4) C, H.
(2S)-[(3-Ethyl-2-oxo-5,7-dipropyl-2,3-dihydro-1,3-benz-
oxazol-6-yl)oxy](4-isopropylphenyl)acetic acid (6): 90%
1
yield and 97% ee from hydrolysis of compound 15b. H NMR
(CD3OD, 500 MHz): δ 7.38 (d, J ) 8.5 Hz, 2H), 7.28 (d, J )
8.5 Hz, 2H), 6.84 (s,1H), 5.03 (s, 1H), 3.84 (q, J ) 7.0 Hz, 2H),
2.95 (m, 1 H), 2.38-2.55 (m, 4H), 1.35-1.60 (m, 4H), 1.31 (t,
J ) 7.0 Hz, 3H), 1.26 (d, J ) 7.5 Hz, 6H), 0.82 (t, J ) 7.5 Hz,
3H), 0.80 (t, J ) 7.5, 3H). MS (ESI) 440.24 (MH+). Anal.
(C26H33NO5) C, H, N.
General Procedure A: Coupling of the Bromide 11
with Phenols 12-14. To a solution of the phenol (5.0 mmol)
in THF (50 mL) was added a solution of lithium tert-butoxide
in THF (1.0 M, 4.8 mL, 4.8 mmol). The resulting solution was
cooled to 0 °C, and a solution of the chiral bromide 11 (1.9 g,
5.0 mmol) in THF (5.0 mL) was added. The reaction mixture
was stirred at 0 °C overnight, quenched with acetic acid (0.5
mL) and poured into water (50 mL). The organic phase was
separated, and the aqueous phase was extracted with ethyl
acetate (30 mL). The combined organic phase was washed with
brine, dried over MgSO4 and concentrated. The residue was
purified by chromatography on silica gel using a mixture of
hexane:ethyl acetate as the eluent.
(1R)-1-Methyl-2-oxo-2-pyrrolidin-1-ylethyl (2S)-(4-iso-
propylphenyl)(4-propionyl-2,6-dipropylphenoxy)-
acetate (15a: R4 ) i-Pr; ArOH ) 12b): 92% yield using
compound 12b as the phenol. 1H NMR 500 MHz (CD3OD); 7.64
(s, 2H), 7.39 (d, J ) 8.0 Hz, 2H), 7.28 (d, J ) 8.1 Hz, 2H) 5.32
(s, 1H), 5.27 (q, J ) 6.9 Hz, 1H), 3.72 (m, 1H), 3.61 (m, 1H)
3.35-3.50 (m, 4H), 2.98 (q, J ) 7.2 Hz, 2H), 2.93 (m, 1H) 2.41
(m, 4H), 1.93 (m, 2H) 1.86 (m, 3H) 1.53 (m, 2H) 1.42 (m, 2H),
1.32 (d, J ) 6.9 Hz, 3H), 1.25 (d, J ) 6.9 Hz, 6H), 1.15 (t, J )
7.3 Hz, 3H), 0.80 (t, J ) 7.3 Hz, 6H). MS (ESI): 536.5 (MH+).
(2S)-(4-Isopropylphenyl)[(2-methyl-3-oxo-5,7-dipropyl-
2,3-dihydro-1,2-benzisoxazol-6-yl)oxy]acetic acid (7): 93%
1
yeild and 96% ee from hydrolysis of compound 15c. H NMR
(CD3OD, 500 MHz) δ 7.45 (d, J ) 8.0 Hz, 2H), 7.43 (s, 1H),
7.40 (d, J ) 8.0 Hz, 2H), 5.17 (s, 1H), 3.62 (s, 3H), 2.86 (m, 1
H), 2.36-2.50 (m, 4H), 1.34-1.58 (m, 4H), 1.24 (d, J ) 7.2
Hz, 6H), 0.82 (t, J ) 7.5 Hz, 3H), 0.81 (t, J ) 7.5, 3H). MS
(ESI) 426.21 (MH+). Anal. (C25H31NO5) C, H, N.
X-ray Crystallography. Purified PPARgamma-LBD (resi-
dues Gln203 to Tyr477) at 10-15 mg/mL was mixed with
compound 7 at a 1.6:1 ratio of compound:protein on ice and
allowed to stand at 4 °C overnight. Crystals were grown by
vapor diffusion at room temperature in 2 µL “sitting” drops
that contained equal volumes of the protein complex solution
and a reservoir solution consisting of 100 mM Tris-HCl, pH
8.0, 0.65-0.90 M Na3 citrate, 1 mM TCEP, against 0.5 to 1.0
mL reservoir solutions in a Cryschem MVD-24 crystallization
tray. Crystals were transferred to 100 mM Tris-HCl, pH 8.0,
1.44 M Na3 citrate, and 1 mM TCEP and vitrified by plunging
the nylon-loop-captured crystals into liquid nitrogen. A crystal
was put into a cold, -170 °C nitrogen gas stream generated
by a CryoStream model 600 (Oxford Cryosystems Inc.) for the
duration of the diffraction experiment. X-ray diffraction data
were collected in-house on a Rigaku FR-D X-ray generator
operating at 50 kV, 100 mA, 20% bias voltage with the Cu KR
(1R)-1-Methyl-2-oxo-2-pyrrolidin-1-ylethyl (2S)-(4-iso-
propylphenyl)[(3-ethyl-2-oxo-5,7-dipropyl-2,3-dihydro-
1,3-benzoxazol-6-yl)oxy]acetate (15b: R4 ) i-Pr; ArOH )
13a): 95% yield using compound 13a as the phenol. 1H NMR
(400 MHz, CDCl3): δ 7.40 (d, J ) 8.0 Hz, 2H), 7.18 (d, J ) 8.0

PPARR/γ Dual Agonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4467
Trial with Gemfibrozil in Middle-Aged Men with Dyslipidemia.
N. Engl. J. Med. 1986, 314, 488.
beam focused with Osmic mirrors (“purple” optics). The
diffraction data were collected as a series of 540 0.5° rotation
images using an ADSC Quantum 4R 2 × 2 CCD area detector
mounted on a Crystal Logics goniometer. Exposure time was
set to keep low resolution reflection overloads at a minimum
without significantly impacting the quality of the high-
resolution data. Data were measured from a single crystal
exhibiting C2 symmetry to a resolution of 2.5 Å. Measured
intensities were reduced and scaled with DENZO and
SCALEPACK from the HKL suite45 that yielded an average
redundancy on measurements of approximately 5 and an Rsym
of 0.077 (see Supporting Information). The structure was
solved using the coordinates for a previously elucidated
selenomethionine PPARγ-LBD complex (unpulished results).
Following rigid-body refinement of the starting structure,
compound 7 was incorporated between rounds of model build-
ing using CHAIN46 or XTALVIEW47 and refinement using
CNX (version 2002 Accelrys, Inc.)48,49 The coordinates may be
found in the Protein Data Bank under ID code 1ZEO.
(15) Rubins, H. B.; Robins, S. J.; Collins, D.; Fye, C. L.; Anderson, J.
W.; Elam, M. B.; Faas, F. H.; Linares, E.; Shaefer, E. J.;
Shectman, G.; Wilt, T. J.; Wittes, J. Gemfibrozil for the second-
ary prevention of coronary heart disease in men with low level
of high-density lipoprotein cholesterol. N. Engl. J. Med. 1999,
341, 410-418.
(16) Robins, S. J.; Collins, D.; Wittes, J. T.; Papademetrious, V.;
Deedwania, P.; Schaefer, E.; McNamara, J. R.; Kashyap, M. L.;
Hershman, J. M.; Wexler, L. F.; Rubins, H. B. Relation of
gemfibrozil treatment and lipid levels with major coronary
events. J. Am. Med. Assoc. 2001, 285, 1585-1591.
(17) Diabetes Atherosclerosis Intervention Study Investigators. Effect
of fenofibrate on progression of coronary-artery disease in type
2 diabetes: The Diabetes Atherosclerosis Intervention Study, a
randomized study. Lancet 2001, 357, 905-910.
(18) Naderali, E. K.; Fatani, S.; Williams, G. Fenofibrate lowers
adiposity and corrects metabolic abnormalities, but only partially
restores endothelial function in dietary obese rats. Atheroscle-
rosis 2004, 177, 307-312.
(19) Guerre-Millo, M, Gervois, P.; Raspe, E.; Madsen, L.; Poulain,
P.; Derudas, B.; Herbert, J.; Winegar, D.; Willson, T.; Fruchart,
J.; Staels, B. Peroxisome proliferator-activated receptor R activa-
tors improve insulin sensitivity and reduce adiposity. J Biol.
Chem. 2000, 275, 16638-16642.
(20) Ye, J.; Doyle, P.; Iglesias, M.; Watson, D.; Cooney, G.; Kraegen,
E. Peroxisome proliferator-activated receptor-R activation lowers
muscle lipids and improves insulin sensitivity in high fat-fed
rats. Diabetes 2001, 50, 411-417.
(21) Expert Panel on Detection, Evaluation and Treatment of High
Blood Cholesterol in Adults. Executive summary of the third
report of the National Cholesterol Education Program (NCEP)
Expert Panel on Detection, Evaluation, and Treatment of High
Blood Cholesterol in Adults (Adult Treatment Panel III). J. Am.
Med. Assoc. 2001, 285, 2486-2497.
(22) Henke, B. R. Peroxisome Proliferator-Activated Receptor R/γ
Dual Agonists for the Treatment of Type 2 Diabetes. J. Med.
Chem. 2004, 47, 4118-4127.
Acknowledgment. We gratefully acknowledge tech-
nical assistance from Margaret Wu, Neelam Sharma,
John Ventre, Roger Meurer, Chhabi Biswas and the
pharmacokinetic measurements by Dr. Kwan Leung
and Raul Alvaro. We also thank Dr. Peter Meinke for
helpful discussions and reviewing of the manuscript.
Supporting Information Available: Additional experi-
mental data. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(23) Henke, B. R.; Blanchard, S. G.; Brackeen, M. F.; Brown, K. K.;
Cobb, J. E.; Collins, J. L.; Harrington, W. W., Jr.; Hashim, M.
A.; Hull-Ryde, E. A.; Kaldor, I.; Kliewer, S. A.; Lake, D. H.;
Leesnitzer, L. M.; Lehmann, J. M.; Lenhard, J. M.; Orband-
Miller, L. A.; Miller, J. F.; Mook, R. A.; Noble, S. A.; Oliver, W.;
Parks, D. J.; Plunket, K. D.; Szewczyk, J. R.; Willson, T. M. (2-
Benzoylphenyl)-L-tyrosine PPARγ Agonists. 1. Discovery of a
Novel Series of Potent Antihyperglycemic and Antihyperlipi-
demic Agents. J. Med. Chem. 1998, 41, 5020.
(1) Issemann, I.; Green, S. Activation of a member of the steroid
hormone receptor superfamily by peroxisome proliferators.
Nature 1990, 347, 645-650.
(2) Mangelsdorf, D. J.; Evans, R. M. The RXR heterodimers and
orphan receptors. Cell 1995, 83, 841-850.
(3) Willson, T. M.; Brown, P. T.; Sternbach, D. D.; Henke, B. R. The
PPARs: From orphan receptors to drug discovery. J. Med. Chem.
2000, 43, 527-550.
(4) Kersten, S.; Desvergne, B.; Wahli, W. Roles of PPARs in health
and disease. Nature 2000, 405, 421-424.
(5) Chawla A.; Repa, J. J.; Evans, R. M.; Mangelsdorf, D. J. Nuclear
receptors and lipid physiology: opening the X-files. Science 2001,
294, 1866-1870.
(6) Berger, J. P.; Moller, D. E. The mechanisms of action of PPARs.
Annu. Rev. Med. 2002, 53, 409-435.
(7) Knouff, C.; Auwerx, J. Peroxisome proliferator-activated recep-
tor-gamma calls for activation in moderation: lessons from
genetics and pharmacology. Endocr. Rev. 2004, 25, 899-918.
(8) Martin, G.; Schoonjans, K.; Staels, B.; Auwerx, J. PPARγ
activators improve glucose homeostasis by stimulating fatty acid
uptake in the adipocytes. Atherosclerosis 1998, 137 (Suppl.),
S75-S80.
(24) Murakami, K.; Tobe, K.; Ide, T.; Mochizuki, T.; Ohashi, M.;
Akanuma, Y.; Yazaki, Y.; Kadowaki, T. A novel insulin sensitizer
acts as coligand for peroxisome proliferator-activated receptor
R
(PPAR-R) and PPAR-γ: effect of PPAR-R activation on
abnormal lipid metabolism in liver of Zucker fatty rat. Diabetes
1998, 47, 1841-1847.
(25) Lohray, B. B.; Lohray, V. B.; Bajji, A. C.; Kalchar, S.; Poondra,
R. R.; Padakanti, S.; Chakrabarti, R.; Vikramadithyan, R. K.;
Misra, P.; Juluri, S.; Mamidi, N. V. S. R.; Rajagopalan, R. (-)3-
[4-[2-Phenoxazin-10-yl)ethoxy]phenyl]-2-ethoxypropanoic Acid-
[(-)DRF2725]: A dual PPAR agonist with potent antihyper-
glycemic and lipid modulating activity. J. Med. Chem. 2001, 44,
2675-2678.
(26) Brooks, D. A.; Etgen, G. J.; Rito, C. J.; Shuker, A. J.; Dominianni,
S. J.; Warshawsky, A. W.; Ardecky, R.; Paterniti, J. R.; Tyhonas,
J.; Broderick, C. L.; Oldham, B. A.; Montrose-Rafizadeh, C.;
Winneroski, L. L.; Faul, M. M.; McCarthy, J. R. Design and
synthesis of 2-methyl-2-{4-[2-(5-methyl-2-aryloxazol-4-yl)ethoxy]-
phenoxy}propionic acids: A new class of dual PPARR/γ agonists.
J. Med. Chem. 2001, 44, 2061-2064.
(27) Sauerberg, P.; Pettersson, I.; Jeppesen, L.; Bury, P. S.; Mor-
gensen, J. P.; Wassermann, K.; Brand, C. L.; Sturis, J.; Woldike,
H. F.; Fleckner, J.; Andersen, A. T.; Mortensen, S. B.; Svensson,
l. A.; Rasmussen, H. B.; Lehmann, S. V.; Polivka, Z.; Sindelar,
K.; Panajotova, V.; Ynddal, L.; Wulff, E. M. Novel tricyclic-R-
alkoxyphenylpropionic acids: Dual PPARR/γ agonists with
hypolipidemic and antidiabetic Activity. J. Med. Chem. 2002,
45, 789-804.
(28) Cheng, P. T.; Chandrasena, G.; Chen, S. BMS-298585 is a novel,
uniquely balanced dual activator of peroxisome proliferator-
activated receptors (PPAR) alpha and gamma, with an excellent
ADME profile. Diabetes 2002, 51 (Suppl. 2), A94.
(29) Desai, R. C.; Gratale, D. F.; Han, W.; Koyama, H.; Metzger, E.;
Lombardo, V. K.; MacNaul, K. L.; Doebber, T. W.; Berger, J. P.;
Leung, K.; Franklin, R.; Moller, D. E.; Heck, J. V.; Sahoo, S. P.
Aryloxazolidinediones: identification of potent orally active
(9) Rosen, E. D.; Spiegelman, B. M. PPARγ: a nuclear regulator of
metabolism, differentiation, and cell growth. J. Biol. Chem. 2001,
276, 37731-37734.
(10) Lehmann, J. M.; Moore, L. B.; Smith-Oliver, T. A.; Wilkinson,
W. O.; Willson, T. M.; Kliewer, S. A. An antidiabetic thiazo-
lidinedione is a high affinity ligand for peroxisome proliferator-
activated receptor (PPARγ). J. Biol. Chem. 1995, 270, 12953-
12956.
(11) Hulin, B.; McCarthy, P. A.; Gibbs, E. M. The glitazone family of
antidiabetic agents. Curr. Pharm. Des. 1996, 2, 85-102.
(12) Ibrahimi, A.; Teboul, L.; Gaillard, D.; Amri, E. Z.; Ailhaud, G.;
Young, P.; Cawthorne, M. A.; Grimaldi, P. A. Evidence for a
common mechanism of action for fatty acids and thiazolidinedi-
one antidiabetic agents on gene expression in preadipose cells.
Mol. Pharmacol. 1994, 46, 1070-1076.
(13) Torra, I. P.; Gervois, P.; Staels, B. Peroxisome Proliferator-
Activated Receptor alpha in Metabolic Disease, Inflammation,
Atherosclerosis and Aging. Curr. Opin. Lipidol. 1999, 10, 151-
159.
(14) Frick, M. H.; Elo, O.; Haapa, K.; Heinonen, O. P.; Heinsalmi,
P.; Helo, P.; Huttunen, J. K.; Kaitaniemi, P.; Koskinen, P.;
Manninen, V.; Maenpaa, H.; Malkonen, M.; Manttari, M.;
Norola, S.; Pasternack, A.; Pikkarainen, J.; Romo, M.; Sjoblom,
T.; Nikkila, E. A. Helsinki Heart Study: Primary-Prevention

4468 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Shi et al.
PPAR dual R/γ agonists. Bioorg. Med. Chem. Lett. 2003, 13,
3541-3544.
(39) 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. Ligand binding and co-activator assembly of
the peroxisome proliferator-activated receptor-γ. Nature 1998,
395, 137-143.
(40) 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.
Structural determinants of ligand binding selectivity between
the peroxisome proliferator-activated receptors. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 13919-13924.
(30) Desai, R. C.; Han, W.; Metzger, E. J.; Bergman, J. P.; Gratale,
D. F.; MacNaul, K. L.; Berger, J. P.; Doebber, T. W.; Leung, K.;
Moller, D. E.; Heck, J. V.; Sahoo, S. P. 5-Aryl thiazolidine-2,4-
diones: discovery of PPAR dual R/γ agonists as antidiabetic
agents, Bioorg. Med. Chem. Lett. 2003, 13, 2795-2798.
(31) Adams, A. D.; Hu, Z.; von Langen, D.; Dadiz, A.; Elbrecht, A.;
MacNaul, K. L.; Berger, J. P.; Doebber, T. W.; Meurer, R.;
Forrest, M. J.; Moller, D. E.; Jones, A. B. O-Arylmandelic acids
as highly selective human PPARR/γ agonists. Bioorg. Med.
Chem. Lett. 2003, 13, 3185-3190.
(41) Cronet, P.; Petersen, J. F.; Folmer, R.; Blomberg, N.; Sjoblom,
K.; Karlsson, U.; Lindstedt, E. L.; Bamberg, K. Structure of the
PPARalpha and -gamma ligand binding domain in complex with
AZ 242; ligand selectivity and agonist activation in the PPAR
family. Structure 2001, 9, 699-706.
(32) Koyama, H.; Miller, D. J.; Boueres, J. K.; Desai, R. C.; Jones, A.
B.; Berger, J. P.; MacNaul, K. L.; Kelly, L. J.; Doebber, T. W.;
Wu, M. S.; Zhou, G.; Wang, P. R.; Ippolito, M. C.; Chao, Y. S.;
Agrawal, A. K.; Franklin, R.; Heck, J. V.; Wright, S. D.; Moller,
D. E.; Sahoo, S. P. (2R)-2-ethylchromane-2-carboxylic acids:
discovery of novel PPARalpha/gamma dual agonists as antihy-
perglycemic and hypolipidemic agents. J. Med. Chem. 2004, 47,
3255-63.
(33) Devasthale, P. V.; Chen, S.; Jeon, Y.; Qu, F.; Shao, C.; Wang,
W.; Zhang, H.; Cap, M.; 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. Design and Synthesis of N-[(4-Methoxyphenoxy)-
carbonyl]-N-[[4-[2-(5-methyl-2-phenyl-4-oxazolyl)ethoxy]phenyl]-
methyl]glycine [Muraglitazar/BMS-298585], a Novel Peroxisome
Proliferator-Activated Receptor R/γ Dual Agonist with Effica-
cious Glucose and Lipid-Lowering Activities. J. Med. Chem.
2004, 48, 2248-2250.
(34) Bristol-Myers Squibb Company Press Release: Bristol-Myers
Squibb Provides Regulatory Update for Muraglitazar and Abata-
cept. December 23, 2004.
(35) The definition of the term “super” agonist is somewhat arbitrary.
We use this term to characterize the compounds described in
this paper whose maximal agonist responses in the cell-based
transactivation assay are generally 3-20 times higher than a
typical reference full agonist.
(42) Chao, Y. S.; et al. Unpublished results.
(43) Berger, J. P.; Petro, A. E.; Macnaul, K. L.; Kelly, L. J.; Zhang,
B.; Richards, K.; Elbrecht, A.; Johnson, B. A.; Zhou, G.; Doebber,
T. W.; Biswas, C.; Parikh, M.; Sharma, N.; Tanen, M. R.;
Thompson, G. M.; Ventre, J.; Adams, A. D.; Mosley, R.; Surwit,
R. S.; Moller, D. E. Distinct Properties and Advantages of a
Novel Peroxisome Proliferator-Activated Protein γ Selective
Modulator. Mol. Endocrinol. 2003, 17, 662-676.
(44) Berger, J. P.; Leibowitz, M. D.; Doebber, T. W.; Elbrecht, A.;
Zhang, B.; Zhou, G.; Biswas, C.; Cullinan, C. A.; Hayes, N. S.;
Li, Y.; Tanen, M.; Ventre, J.; Wu, M.; Berger, G. D.; Mosley, R.;
Marquis, R.; Santini, C.; Sahoo, S. P.; Tolman, R. L.; Smith, R.
G.; Moller, D. E. Novel Peroxisome Proliferator-activated Recep-
tor (PPAR) γ and PPAR δ Ligands Produce Distinct Biological
Effects. J. Biol. Chem. 1999, 274, 6718.
(45) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307-
326.
(46) Sack, J. CHAINsa crystallographic modeling program. J. Mol.
Graphics 1988, 6, 224-235.
(47) McRee, D. E. XtalView/XfitsA versatile program for manipulat-
ing atomic coordinates and electron density. J. Struct. Biol. 1999,
125, 156-165.
(36) Devine, P. N.; Dolling, U.-H.; Heid, R. M.; Tschaen, D. M.
Stereoselective synthesis of 2-aryloxy acids from lactamide
derived esters of racemic R-halo carboxylic acids. Tetrahedron
Lett. 1996, 37, 2683-2686.
(37) Shi, G. The first general synthesis of N-substituted 1,2-ben-
zisoxazolin-3-ones. Tetrahedron Lett. 2000, 41, 2295.
(38) Albini, A. Photosensitization in Organic Synthesis. Synthesis.
1981, 249-264.
(48) Bru¨nger, A. T,; Kuriyan, J.; Karplus, M. Crystallographic
R-factor refinement by molecular dynamics. Science 1987, 235,
458-460.
(49) Bru¨nger, A. T.; et al. Crystallography and NMR system (CNS):
A new software system for macromolecular structure determi-
nation. Acta Crystallogr. 1998 D54, 905-921.
JM0502135