Substituted 2-[(4-aminomethyl)phenoxy]-2-methylpropionic acid PPARα agonists. 1. Discovery of a novel series of potent HDLc raising agents

Article abstract of DOI:10.1021/jm058056x

The peroxisome proliferator activated receptors PPARα, PPARγ, and PPARδ are ligand-activated transcription factors that play a key role in lipid homeostasis. The fibrates raise circulating levels of high-density lipoprotein cholesterol and lower levels of triglycerides in part through their activity as PPARα agonists; however, the low potency and restricted selectivity of the fibrates may limit their efficacy, and it would be desirable to develop more potent and selective PPARα agonists. Modification of the selective PPARδ agonist 1 (GW501516) so as to incorporate the 2-aryl-2-methylpropionic acid group of the fibrates led to a marked shift in potency and selectivity toward PPARα agonism. Optimization of the series gave 25a, which shows EC₅₀ = 4 nM on PPARα and at least 500-fold selectivity versus PPARγ and PPARγ. Compound 25a (GW590735) has been progressed to clinical trials for the treatment of diseases of lipid imbalance.

Full text of DOI:10.1021/jm058056x

J. Med. Chem. 2007, 50, 685-695
685
Substituted 2-[(4-Aminomethyl)phenoxy]-2-methylpropionic Acid PPARr Agonists. 1. Discovery
of a Novel Series of Potent HDLc Raising Agents
Michael L. Sierra,†,‡ Ve´ronique Beneton,§ Anne-Be´ne´dict Boullay,| Thierry Boyer,‡ Andrew G. Brewster,‡ Fre´de´ric Donche,‡
Marie-Claire Forest,| Marie-He´le`ne Fouchet,‡ Franc¸oise J. Gellibert,‡ Didier A. Grillot,| Millard H. Lambert, Alain Laroze,‡
Christelle Le Grumelec,‡ Jean Michel Linget,§,# Valerie G. Montana, Van-Loc Nguyen,‡ Edwige Nicode`me,| Vipul Patel,X
Annie Penfornis,‡ Olivier Pineau,‡ Danig Pohin,‡ Florent Potvain,| Ge´raldine Poulain,‡,# Ce´cile Bertho Ruault,‡
,x
Michael Saunders,|,O Je´roˆme Toum,‡ H. Eric Xu, Robert X. Xu, and Pascal M. Pianetti*,‡,§
Department of Medicinal Chemistry, Department of Biology, Department of Drug Metabolism and Pharmacokinetics, Laboratoire
GlaxoSmithKline, Centre de Recherches, 25-27 aVenue du Que´bec, 91951 Les Ulis, France, Department of Medicinal Chemistry,
GlaxoSmithKline Research and DeVelopment, Medicines Research Centre, Gunnels Wood Road, SteVenage, Hertfordshire, SG1 2N7,
United Kingdom, and Department of Computational, Analytical and Structural Sciences, GlaxoSmithKline, 5 Moore DriVe,
Research Triangle Park, North Carolina 27709
ReceiVed December 5, 2005
The peroxisome proliferator activated receptors PPARR, PPARγ, and PPARδ are ligand-activated transcription
factors that play a key role in lipid homeostasis. The fibrates raise circulating levels of high-density lipoprotein
cholesterol and lower levels of triglycerides in part through their activity as PPARR agonists; however, the
low potency and restricted selectivity of the fibrates may limit their efficacy, and it would be desirable to
develop more potent and selective PPARR agonists. Modification of the selective PPARδ agonist 1
(GW501516) so as to incorporate the 2-aryl-2-methylpropionic acid group of the fibrates led to a marked
shift in potency and selectivity toward PPARR agonism. Optimization of the series gave 25a, which shows
EC50 ) 4 nM on PPARR and at least 500-fold selectivity versus PPARδ and PPARγ. Compound 25a
(GW590735) has been progressed to clinical trials for the treatment of diseases of lipid imbalance.
Table 1. Human PPARR Agonist Potencies of Fibratesa
Introduction
Elevated circulating levels of low-density lipoprotein cho-
lesterol (LDLc) constitutes a major risk factor for coronary artery
disease, and there exists a wealth of clinical data supporting
the use of the LDLc-lowering statins in an increasingly wide
range of patients.1 More recently, the roles of both low
circulating levels of high-density lipoprotein cholesterol (HDLc)
and high circulating levels of triglycerides (TG) as cardiovas-
cular disease risk factors have come into focus, and there is a
growing level of confidence in the potential of drugs that lower
TG or raise HDLc to play an important part in future
cardiovascular drug therapy.2
The principal agents in current use for raising HDLc and
lowering TG are the fibrates and nicotinic acid. The fibrates
give increases of up to 20% in HDLc and decreases in TG of
up to 40%;2,3 nicotinic acid shows similar efficacy on HDLc
with perhaps a slightly less pronounced effect on TG. It would
be highly desirable to discover drugs that exert greater effects
on HDLc and TG levels.
a Data were taken from ref 10 and were generated using the PPAR-
GAL4 transactivation assay.
to act as agonists at the peroxisome proliferator activated
receptor alpha (PPARR).5,6 The PPARs comprise a family of
ligand-activated transcription factors that play a key role in lipid
homeostasis. There are three members of the family: PPARR,
PPARδ (or â), and PPARγ.7 PPARR is highly expressed in the
liver, heart, and muscle and includes a range of fatty acids
among its natural ligands. In man, activation of PPARR results
in increased clearance of TG-rich very low-density lipoprotein
(VLDL) via a reduction in plasma levels of ApoCIII8 and in
upregulation of ApoA1, the principal lipoprotein component of
HDL.9 As a consequence, the fibrates lower TG and raise HDLc
levels. The fibrates are only weakly potent PPARR agonists,10
as shown in Table 1. It has been proposed that more potent and
subtype-selective PPARR activators might offer enhanced
specificity and reduced side effects compared with that of
existing fibrates,11 and several groups have reported the
While the molecular target of nicotinic acid has only recently
been identified,4 the fibrates have been known for several years
* To whom correspondance should be addressed. Tel.: (33) 1 69 29 60
05. Fax : (33) 1 69 07 48 92. E-mail : pascal.m.pianetti@gsk.com.
† Current address: Leo Pharma AS, 55 Industriparken, DK-2750
Ballerup, Denmark.
‡ Department of Medicinal Chemistry, Les Ulis.
§ Department of Drug Metabolism and Pharmacokinetics.
| Department of Biology.
Computational, Analytical and Structural Sciences.
# Current address : CareX, Bd Sebastien Brant, BP 30442, 47412
ILLKIRCH Cedex.
X Department of Medicinal Chemistry, Stevenage.
O Current address : Bio-Michs sprl, 132 rue Berkendael, 1050 Bruxelles,
Belgium.
x Current address : Laboratory of Structural Sciences, Van Andel
Research Institute, 333 Bostwick Avenue, Grand Rapids, MI 49503.
10.1021/jm058056x CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/23/2007

686 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Sierra et al.
Scheme 1
Scheme 2a
a Reagents and conditions: (a) ethyl-2-chloroacetoacetate, EtOH, reflux; (b) LiAlH4, THF, 0 °C; (c) PCC, CH2Cl2, rt; (d) conc. HCl, petroleum ether,
37% HCHO/H2O; (e) PPh3, toluene, reflux; (f) NaH (60% in mineral oil), anhydrous EtOH, rt; (g) 10% Pd/C, EtOH/dioxan, H2; (h) NaOH (1 N), dioxane,
reflux.
Scheme 3a
a Reagents and conditions: (a) 4-methoxybenzyltriphenylphosphonium chloride, NaH (60% in mineral oil), CH2Cl2; (b) 10% Pd/C, EtOH/AcOH, H2; (c)
BBr3, CH2Cl2, -40 °C; (d) 2-trichloromethyl-2-propanol, acetone, NaOH (conc), rt.
discovery of potent PPARR agonists and their effects in animal
models of dyslipidemia.12
LiAlH4 and oxidized with PCC to give the aldehyde 5.
Compound 7 was prepared in two steps from ethyl 2-methyl-
phenoxyacetate via reaction with formaldehyde and HCl to give
6 followed by substitution with triphenylphosphine to give 7.
Wittig condensation of 5 with 7 gave the alkene 8, which was
hydrogenated over 10% Pd on carbon to give 9. Hydrolysis
using NaOH in dioxane furnished compound 10.
We describe below how, starting from the recently published13
selective PPARδ agonist 1 (GW501516) which showed weak
potency on PPARR, we have first incorporated the classical
fibrate “head group” to improve selectivity for PPARR and
subsequently optimized potency and selectivity. We have thus
developed a new series of highly potent and selective PPARR
agonists, one of which, 25a, is currently in clinical trials for
the treatment of dyslipidemia (Scheme 1).
Compound 14 was synthesized according to the procedure
depicted in Scheme 3. Wittig condensation of 5 with 4-meth-
oxybenzyltriphenylphosphonium chloride gave 11, which was
hydrogenated over 10% Pd on carbon to give 12. Demethylation
using boron tribromide gave 13, and coupling with 2-trichloro-
methyl-2-propanol and hydrolysis afforded 14.
Chemistry
Compound 10 was prepared as shown in Scheme 2. Con-
densation of 4-trifluoromethylthiobenzamide 2a with ethyl-2-
chloroacetoacetate gave the thiazole 3a, which was reduced with
Alkylation of 4-nitrophenol followed by hydrogenolysis gave
16, which was coupled with the acid 17a, prepared by hydrolysis

Series of Potent HDLc Raising Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 687
Scheme 4a
a Reagents and conditions: (a) ethyl 2-bromo-2-methylpropionate, DMF, K2CO3, 40 °C; (b) 5% Pd/C, EtOH, 30 °C, H2; (c)NaOH (aq), EtOH, reflux; (d)
HOBT, EDC, Et3N, DMF, rt; (e) NaOH (1 N), EtOH, 40°.
Scheme 5a
The reversed amide 34 was prepared as shown in Scheme 9.
Bromination of 4 using PBr3 gave 28. Substitution by potassium
phthalimide gave 29, which was cleaved by hydrazine to give
the amine 30. Compound 30 was coupled with 32, in turn
prepared by alkylation of 4-hydroxybenzaldehyde followed by
oxidation using sodium chlorite and sodium hydrogen phosphate,
giving 33, which was then hydrolyzed to give 34.
The homologated amide 37 was synthesized via HOBT
coupling of 17a with 4-hydroxyphenethylamine followed by
alkylation and hydrolysis, as shown in Scheme 10.
Compound 25k was prepared by hydrogenolysis of 24e
followed by hydrolysis (Scheme 11).
The 4-methanesulfonyloxy thiazole carboxylic ester 3l was
prepared from 4-hydroxythiobenzamide 2k using the procedure
described above, followed by mesylation using methane sulfonyl
chloride and triethylamine. LiOH mediated hydrolysis gave the
carboxylic acid 17k, which was alkylated to give ester 24l.
Hydrolysis under mild conditions (1 equiv of LiOH, rt) of 24l
gave product 25l, whereas more vigorous hydrolysis of 24l
(NaOH, THF/EtOH, 70°) gave the phenol 25m (Scheme 12).
The regioisomeric thiazole 41 was prepared in a similar
manner to 25a via the carboxylic acid 39, obtained from
condensation of 2a in refluxing ethanol with the intermediate
obtained from the reaction of 2-ketobutyric acid with bromine
to give 38, followed by hydrolysis. Coupling of 39 with 21
followed by standard hydrolysis gave 41, as depicted in Scheme
13.
a Reagents and conditions: (a) NaH (60% in mineral oil), DMF, ethyl-
2-bromo-2-methylpropionate, rt then reflux; (b) 10% Pd/C, AcOH/EtOH,
H2; (c) 5, NaBH(OAc)3, dichloroethane/AcOH, rt; (d) NaOH (1 N), EtOH,
40°.
Scheme 6a
The trifluoromethyl-substituted thiazole 45 was prepared
using a similar route to that used for 25a, substituting ethyl-2-
chloro-4,4,4-trifluoroacetate for the methyl analogue (Scheme
14).
Results and Discussion
a Reagents and conditions: (a) ethyl-2-chloroacetoacetate, EtOH, reflux;
(b) NaOH (1 N), EtOH, reflux.
The recently published 1 is a potent agonist at PPARδ with
around 1000-fold selectivity with respect to PPARR, as mea-
sured using cell-based transient-transfection assays14 (Table 2).
Replacement of the thiomethylene chain of 1 by ethylene gave
10, which showed equally low potency against PPARR. PPARδ
potency was somewhat decreased (EC50 ) 4 nM) relative to 1,
while PPARγ activity remained very weak. Encouragingly
however, replacement of the o-methyl phenoxyacetate group
by the classical “fibrate head group” shared by all of the drugs
shown in Table 1 gave 14, which regained potency against
PPARR (EC50 ) 210 nM).
We further explored the effect of modifying the chain linking
the head group to the biaryl unit (Table 3). Replacement of the
ethylene of 14 by an amide link gave 19, which showed almost
complete loss of activity. However, increasing the length of the
linking chain proved beneficial for PPARR potency. The
of ester 3a, to give 18. Hydrolysis using NaOH in EtOH gave
19, as shown in Scheme 4.
Compound 23 was prepared from 4-cyanophenol via alkyl-
ation and hydrogenation to give 21, which was reacted with
aldehyde 5 under reductive amination conditions. The standard
hydrolysis gave 23 (Scheme 5).
Thiazole carboxylic acids 17b-j were prepared in the same
manner as 17a, as indicated in Scheme 6. The acids 17a-j were
then elaborated to the amides 25a-j via coupling using either
HOBT/EDC/Et3N or SOCl2, followed by hydrolysis (Scheme
7).
N-Methylamide 27 was obtained by methylation of 24a using
NaH and methyl iodide followed by hydrolysis, as shown in
Scheme 8.

688 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Sierra et al.
Scheme 7a
a Reagents and conditions: Method A, SOCl2, 21, Et3N, CH2Cl2; Method B, 21, HOBT, EDC, Et3N, DMF, rt; (a) NaOH (1 N), THF, reflux.
Scheme 8a
shown in Figure 1, 25a fits well into the PPARR binding site.
The carboxylate moiety of the head group forms four hy-
drogen bonds with Ser-280, Tyr-314, Tyr-464, and His-440, as
expected from the previous PPARR X-ray crystal structures of
GW40954415 and AZ242.16 The gem-dimethyl substituents are
directed into a lipophilic pocket bounded by Phe-273, Gln-277,
Val-444, and Leu-456, a region at the top end of the so-called
“benzophenone” pocket.17 These residues are conserved in
PPARγ and PPARδ except for Val-444, which is leucine in
PPARγ and methionine in PPARδ. This side chain is smallest
and most able to accommodate the fibrate headgroup in PPARR
and is largest and least able to accommodate the fibrate
headgroup, in PPARδ.14,18 This might explain why 14 gains
potency versus PPARR and PPARγ while maintaining a similar
level of PPARδ potency compared with 10.
The amide group of 25a adopts a trans-conformation in which
the carbonyl oxygen is directed toward Thr-279 and nearby
water molecules. Although the carbonyl oxygen of 25a lies near
Thr-279, the threonine side chain appears to adopt a conforma-
tion that turns its hydroxyl group away from the carbonyl
oxygen. Instead of hydrogen bonding with this threonine, the
carbonyl oxygen hydrogen bonds indirectly with Ser-280 and
Thr-283 via two water molecules (Figure 2a). Thr-279 is
hydrogen bonded to the backbone NH of Ala-333 via a third
water molecule and oriented toward the sulfur of the thiazole
in a manner likely to give a favorable interaction, even though
sulfur does not form strong hydrogen bonds.
The amide NH is positioned close to the methyl group of the
thiazole and surrounded by three sulfur-containing amino acids,
Cys-276, Met-355, and Met-330, with the sulfur atoms directed
toward the NH proton at distances of 4.38 Å, 2.92 Å, and 3.33
Å, respectively (Figure 2b). This relatively polar environment
stabilizes the partial positive charge of the NH, thereby favoring
the observed conformation of 25a. Overall, the linking amide
group of 25a appears to be well stabilized by the protein
environment with the carbonyl surrounded by a very strong
hydrogen bond network and the NH by a triad of polar atoms.
This remarkable environment might explain the potency and
selectivity of the amide linker compound in the PPARR pocket,
which is more lipophilic and less solvent-exposed than the
corresponding pockets of either PPARγ or PPARδ.15
a Reagents and conditions: (a) NaH (60% in mineral oil), MeI, DMF,
40 °C; (b) NaOH (1 N), EtOH, reflux.
methyleneaminomethylene derivative 23 gained 3-fold in po-
tency, and the amide 25a showed a 50-fold increase in potency.
N-Methylation (27) of the amide 25a was detrimental to PPARR
potency, and the inverse amide 34 and the chain-extended amide
37 were also less potent.
Replacement of the terminal trifluoromethyl substituent by a
range of groups showed that substitution in this position was
largely well tolerated and that potency correlated well with the
lipophilicity of the substituent (Table 4). Thus the tert-butyl,
trifluoromethyl, and trifluoromethoxy groups gave the highest
potency, while fluoro, hydrogen, and amino substituents gave
a progressive decrease in potency. All compounds demonstrated
considerable selectivity for PPARR.
The m-trifluoromethyl analogue 25i was equipotent with 25a
versus PPARR, whereas the o-trifluoromethyl analogue 25j lost
both PPARR potency and selectivity versus PPARδ. The
p-hydroxy 25m and p-methanesulfonyloxy 25l substituted
compounds were inactive.
We next investigated the effect of modifying the central
heterocycle fragment (Table 5). The thiazole regioisomer 41
showed a 500-fold decrease in PPARR potency compared with
25a, whereas PPARδ potency increased. The trifluoromethyl
analogue 45 showed similar PPARR agonist potency to that of
25a but was less selective versus PPARγ.
The precise binding network around the amide linker might
also account for the observation that any modification resulted
in a decrease in PPARR potency. N-Methylation in 27 is not
well tolerated due to the proximity of the sulfur atoms, and the
reverse amide of 34 binds less favorably because its carbonyl
Binding Mode of 25a in the PPARR Ligand Binding
Domain. To determine with precision the binding mode of this
new series of ligands, 25a was cocrystallized with PPARR. As

Series of Potent HDLc Raising Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 689
Scheme 9a
a Reagents and conditions: (a) PBr3, CH2Cl2, 0 °C; (b) potassium phthalimidate, DMF, 70 °C; (c) hydrazine hydrate, EtOH, 90 °C; (d) ethyl-2-bromo-
2-methylpropionate, K2CO3, DMF, 40 °C; (e) NaClO2, NaHPO3, H2O, t-BuOH, rt; (f) SOCl2, Et3N, CH2Cl2; (g) NaOH (1 N), THF, reflux.
Scheme 10a
group fits well into a pocket that is connected with the bottom
end of the benzophenone pocket. This pocket is bounded by
Cys-276, Met-330, Leu-344, and Met-355 residues, which are
conserved in PPARγ and PPARδ with the exception of Met-
330, which is valine in PPARγ and leucine in PPARδ, and Met-
355, which is isoleucine in PPARδ. The smaller valine side-
chain in PPARγ could accommodate slightly larger substituents
at this position on the thiazole, while the branched isoleucine
in PPARδ might crowd larger substituents. In accordance with
this hypothesis, replacement of the methyl by a trifluoromethyl
group in 45 maintains PPARR but increases PPARγ potency
and decreases PPARδ potency.
The phenyl-trifluoromethyl moiety makes a good fit into the
lower pocket of the binding site and would be expected to
contribute favorably to overall binding through van der Waals
interactions. This observation is compatible with SAR data that
show a relationship between size and lipophilicity of the para-
substituent and PPARR potency. The decrease in potency seen
with compounds bearing polar para-substituents (25k and 25m)
may be explained by the hydrophobic nature of the lower pocket.
The inactivity of the methylsulfonyl analogue 25l may be due
to a combination of polarity and steric constraints. meta-
Substitution is well tolerated as evidenced by the fact that 25i
is equipotent with 25a. Although the X-ray structure of 25a
shows that one meta-position is directed toward Leu-247 and
Ile-241, the other meta-position points into an open space and
could easily accept a substituent. On the other hand, ortho-
substitution would force the phenyl group out of plane with
the thiazole and likely lead to a clash with the wall of the narrow
pocket, a conjecture that is consistent with the decrease in
potency seen with 25j.
a Reagents and conditions: (a) 17a, HOBT, EDC, DMF, Et3N, rt; (b)
ethyl-2-bromo-2-methylpropionate, K2CO3, DMF, 80 °C; (c) NaOH (1 N),
THF, reflux.
Scheme 11a
Pharmacokinetics in Rat and Dog (Table 6). Following
intravenous administration of compound 25a (2.7 mg/kg) to the
rat, distribution to the tissues was limited with the volume of
distribution (1 L/kg) similar to that of total body water (0.6
L/kg). Total plasma clearance was low (5 mL/min/kg), repre-
senting about 6% of rat hepatic blood flow. The low clearance
and moderate volume of distribution resulted in a plasma half-
life of 2.4 h. Following a single oral dose of compound 25a at
3 mg/kg, the maximum concentration of compound in the
plasma was 1461 ng/mL after 1.5 h. The bioavailability was
high (47%).
a Reagents and conditions: (a) 10% Pd/C, EtOH, rt; (b) NaOH (1 N),
THF, reflux.
oxygen is probably not hydrogen bonded and is instead
surrounded by three lipophilic amino acids.
The nitrogen atom of the thiazole ring is not involved in any
discernible interaction with the binding site, whereas the methyl
Following intravenous administration of compound 25a to
the dog at 2 mg/kg, distribution to the tissues was limited with
the volume of distribution (2.8 L/kg) being greater than that of

690 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Sierra et al.
Scheme 12a
a Reagents and conditions: (a) ethyl-2-chloroacetoacetate, EtOH, reflux; (b) MeSO2Cl, Et3N, toluene/acetone; (c) LiOH (1.5 equiv), rt; (d) 21, HOBT,
EDC, DMF, Et3N, rt; (e) LiOH (1 equiv), THF/H2O, rt; (f) NaOH (1 N), EtOH, THF, 70°.
Scheme 13a
Scheme 14a
a Reagents and conditions: (a) ethyl-2-chloro-4,4,4-trifluoroacetoacetate,
DMF, 100 °C; (b) LiOH (1 N), EtOH, rt; (c) SOCl2, 21, NEt3, CH2Cl2; (d)
NaOH (1 N), THF, reflux.
Table 2
a Reagents and conditions: (a) Br2; (b) EtOH, reflux; (c) NaOH (1 N),
EtOH, reflux; (d) 21, HOBT, EDC, DMF, Et3N, rt; (e) NaOH (1 N), THF,
reflux.
total body water (0.6 L/kg). Total plasma clearance was
moderate (13 mL/min/kg), representing about 35% of hepatic
blood flow in the dog. The moderate clearance and low volume
of distribution resulted in a plasma half-life of 2.6 h. Following
a single oral dose of compound 25a at 3 mg/kg, the maximum
concentration of compound in the plasma was 1449 ng/mL. The
bioavailability was high (85%).
In Vivo Pharmacology. Several compounds in this series
demonstrated profound in vivo activity in animal models of
dyslipidemia, as illustrated by compound 25a. The human Apo-
A-I-transgenic mouse model has been proposed to be potentially
relevant to human disease, because in this model, fibrates give
upregulation rather than the repression of Apo-A-I seen in other
rodent models.19 Compound 25a shows similar PPARR agonist
potency and selectivity versus murine and human PPARs
(murine PPAR EC50, R ) 15 nM; δ ) 1000 nM; γ > 10 000
nM). When administered orally twice a day for 5 days, 25a,
prepared as a suspension in 0.5% HPMC 100/1% Tween80 at
pH ) 7.0, gave dose-related decreases in circulating TG,
PPARRa
EC50b (µM)
PPARδa
EC50b (µM)
PPARγa
EC50b (µM)
cmpd
(% activation)b,c
(% activation)b,c
(% activation)b,c
1
1.888 ( 0.1
(108 ( 1%)
2.0d
(85%)
0.21 ( 0.08
(138 ( 61%)
0.001 ( 0.000 09
(97 ( 1%)
0.004d
(117%)
0.02 ( 0.07
(90 ( 10%)
8.900 ( 0.27
(80 ( 1%)
10.0d
(74%)
1.87 ( 0.79
(76 ( 14.9%)
10
14
a Data generated using cell based transient transfection assays described
in ref 13. b (SD. c Percent of maximal activation of all compounds was
compared to reference compounds normalized to 100%. For the PPARR
activity, the reference compound was 25a. For the PPARδ and PPARγ
activities, the reference compounds were 1 and rosiglitazone, respectively.
d Results of a single experiment.
VLDLc, and LDLc and concomitant increases in HDLc (Table
7). The ED50 for the HDL effect was approximately 1 mg/kg.
Circulating levels of Apo-A-I were also increased by the

Series of Potent HDLc Raising Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 691
Table 3
Table 5
PPARRa
PPARδa
PPARγa
EC50b (µM)
EC50b (µM)
EC50b (µM)
cmpd
chain
(% activation)b,c (% activation)b,c (% activation)b,c
14 -CH2CH2-
0.21 ( 0.08
(138 ( 61%)
9.5 ( 0.5
(69(0%)
0.07 ( 0.01
(110 ( 4%)
0.004 ( 0.002
(95 ( 14%)
0.24d
0.02 ( 0.07
(90 ( 10%)
>10
1.87 ( 0.79
(76 ( 14.9%)
>10
19 -NHCO-
23 -CH2NHCH2-
25a -CH2NHCO-
27 -CH2NMeCO-
34 -CONHCH2-
4.05 ( 1.1
(61 ( 2%)
2.83 ( 1.18
(82 ( 21%)
>10
>10
>10
>10
>10
>10
(84%)
1.32d
(109%)
6.26d
(62%)
>10
37 -CH2CH2NHCO- 0.11 ( 0.05
(129 ( 58%)
a Data generated using cell based transient transfection assays described
in ref 13. b (SD. c Percent of maximal activation of all compounds was
compared to reference compounds normalized to 100%. For the PPARR
activity, the reference compound was 25a. For the PPARδ and PPARγ
activities, the reference compounds were 1 and rosiglitazone, respectively.
d Results of a single experiment.
a Data generated using cell based transient transfection assays described
in ref 13. b (SD. c Percent of maximal activation of all compounds was
compared to reference compounds normalized to 100%. For the PPARR
activity, the reference compound was 25a. For the PPARδ and PPARγ
activities, the reference compounds were 1 and rosiglitazone, respectively.
d Results of a single experiment.
Table 4
PPARRa
PPARδa
PPARγa
EC50b (µM)
EC50b (µM)
EC50b (µM)
cmpd
X
(% activation)b,c (% activation)b,c (% activation)b,c
25a
p-CF3
0.004 ( 0.002
(95 ( 14%)
0.005 ( 0.000
(66 ( 6%)
0.004 ( 0.000
(116 ( 5%)
0.01 ( 0.003
(94 ( 7%)
0.03 ( 0.01
(111 ( 17%)
0.01 ( 0.01
(103 ( 8%)
0.04d
2.83 ( 1.18
(82 ( 21%)
>10
>10
25b
25c
25d
25e
25f
25g
25h
25i
p-t-butyl
p-OCF3
p-Cl
0.86 ( 0.01
(78 ( 0%)
>10
4.6 ( 0.5
(73 ( 3%)
5.4 ( 0.9
(85 ( 9%)
>10
>10
>10
>10
>10
>10
>25
>25
>10
Figure 1. X-ray crystal structure of 25a complexed with the PPARR
ligand binding domain. The molecular surface of the binding site is
represented by blue dots. Interactions of the head group amino acids
with the carboxylate moiety of the ligand are shown as white dotted
lines. The hydrogens shown here and in Figure 2a,b were not visible
in the electron density and were instead positioned to optimize their
hydrogen bonding interactions.
p-NO2
p-OMe
p-F
>10
>10
>10
>25
(118%)
H
0.12 ( 0.02
(107 ( 5%)
0.004d
m-CF3
o-CF3
p-NH2
(106%)
in HDL cholesterol (+26%). The finding that 25a is able to
lower LDLc and TG and increase HDL cholesterol in the Apo-
A-I-transgenic mouse model suggests that it will deliver
significant therapeutic benefit in the treatment of dyslipidemia
and hypertriglyceridemia. Compound 25a (GW590735) has been
progressed to clinical trials for the treatment of diseases of lipid
imbalance.
25j
25k
0.465d
0.808d
(74%)
>10
(82%)
5.64 ( 0.29
(67 ( 4%)
>10
25l
25m
p-OSO2Me
p-OH
>10
>10
>10
>10
>10
a Data generated using cell based transient transfection assays described
in ref. 13. b (SD. c Percent of maximal activation of all compounds was
compared to reference compounds normalized to 100%. For the PPARR
activity, the reference compound was 25a. For the PPARδ and PPARγ
activities, the reference compounds were 1 and rosiglitazone, respectively.
d Results of a single experiment.
Conclusions
Starting from the selective PPARδ agonist 1, which showed
weak potency on PPARR, we have developed a new series of
highly potent and selective PPARR agonists, one of which,
25a, has progressed to clinical evaluation. The superior potency
and PPAR subtype selectivity of 25a suggest that this com-
pound offers the potential to deliver significantly improved
therapeutic benefits over the fibrates in dyslipidemia and
hypertriglyceridemia.
treatment (data not shown), consistent with the mechanism of
action. As a reference, Fenofibrate was tested in human Apo-
A-I transgenic mice at 50 mg/kg. It produced the expected
profile with a decrease in plasma TG (-43%), VLDL cholesterol
(-64%), and LDL cholesterol (-78%), as well as an increase

692 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Sierra et al.
Splitting patterns are designed as s, singlet; d, doublet; t, triplet; q,
quartet; m, multiplet; bs, broad singlet. Low-resolution mass spectra
(MS) were recorded on a Micromass platform LC or a Agilent GC/
MS spectrometer. High-resolution mass spectra were recorded on
a Micromass LCT (TOF) spectrometer. Mass spectra were acquired
in either positive or negative ion mode under electrospray ionization
(ESI) or atmospheric pressure chemical ionization (APCI) methods.
Analytical high performance liquid chromatography (HPLC;
system A) was conducted on a Chromolith Performance RP 18
column (100 × 4.6 mm id) eluting with 0.01 M ammonium acetate
in water and 100% acetonitrile (CH3CN), using the following elution
gradient: 0 to 100% CH3CN over 4 min and 100% CH3CN over
1 min at 5 mL/min at a temperature of 30 °C.
Analytical HPLC (system B) was conducted on a XTERRA-
MS C18 column (30 × 3 mm id, 2.5 µm) eluting with 0.01 M
ammonium acetate in water and 100% acetonitrile (CH3CN), using
the following elution gradient: 0 to 100% CH3CN over 4 min and
100% CH3CN over 1 min at 1.1 mL/min at a temperature of 40
°C.
Analytical HPLC (system C) was conducted on a Uptisphere-
hsc column (3 µm 33 × 3 mm id) eluting with 0.01 M ammonium
acetate in water and 100% acetonitrile (CH3CN) using the following
elution gradient: 0-5% CH3CN over 0.5 min, gradient 5-100%
CH3CN over 3.25 min, and 100% CH3CN over 0.75 min at 1.3
mL/mn at a temperature of 40 °C.
Analytical HPLC (system D) was conducted on a Uptisphere-
hsc column (3 µm 33 × 3 mm id) eluting with 0.01 M ammonium
acetate in water and 100% acetonitrile (CH3CN) using the following
elution gradient: 0-5% CH3CN over 0.5 min, gradient 5-100%
CH3CN over 6.25 min, and 100% CH3CN over 0.75 min at 1.3
mL/mn at a temperature of 40 °C.
Figure 2. (a) Environment of the linking amide group of 25a:
hydrogen bond network surrounding the carbonyl. Water molecules
are shown as white spheres; hydrogen bond interactions are represented
as green dotted lines; distances are indicated in Å and represented as
yellow dotted lines; for clarity of the figure, the amino acid backbone
was deleted for Ser-280 and Thr-279. (b) Environment of the linking
amide group of 25a: triad of sulfur containing amino acids surrounding
the amide NH. NH-S distances are indicated in Å and represented as
yellow dotted lines.
The purity of key compounds was determined on two analytic
HPLC systems using UV detection.
Pharmacokinetics in Rat and Dog. Compound 25a was
administered to Wistar rats (n ) 15) by oral gavage at dose of 3
mg/kg in pH 7 buffer, 0.1% Tween80 and by intravenous injection
via the penis vein (n ) 30) at a dose of 2.7 mg/kg in 10% DMSO
and PEG200. Compound 25a was administered orally to male
beagle dogs (n ) 3) by stomach intubation at a dose of 3 mg/kg in
pH 7 buffer, 0.1% Tween80 and by intravenous injection via the
cephalic vein at a dose of 2 mg/kg in 10% DMSO and PEG200.
Blood samples were placed on wet ice, and plasma was collected
after centrifugation. Plasma samples were stored frozen at -20 °C
until time of analysis. Plasma samples (0.5 mL) were diluted with
1:1 buffer (NaH2PO4, 0.1 M, pH 4) and then extracted with ethyl
acetate (5 mL). The ethyl acetate was evaporated, and the residue
was resuspended in 200 µL of mobile phase (water/acetonitrile/
TFA; 30v/70v/0.1%). Samples were analyzed by high-performance
liquid chromatography spectrometric analysis (LC/MS/MS). Phar-
macokinetic parameters were determined by SIPHAR.
Table 6. Pharmacokinetic Data for Compound 25a
PK parameter
rat
dog
Cl (mL/min/kg)
Vd (L/kg)
5
1
2.4
47
13
2.8
2.6
85
T1/2 (h)
F (%)
Table 7. Effect of 25a in the Human Apo-A-I Transgenic Mouse
Modela
oral dose
(mg/kg, b.i.d.)
VLDL chol
(g/L) %
LDL chol
(g/L) %
HDL chol
(g/L) %
TG
(g/L) %
Apo-A-I Transgenic Mouse Model. Male C57BL/6 mice
transgenic for human ApoA-I were obtained from Charles River
Laboratories (L’Arbresle, France) and randomized into treatment
groups of n ) 5 animals. Twice a day oral administration of vehicle
(0.5% HPMC/1% Tween80, pH ) 7.0) or indicated doses of
compound as a suspension began when animals were nine weeks
old and lasted for 5 days. Animals were fasted overnight before
blood samples were taken by intracardiac puncture. Whole liver
was collected and weighed. Blood samples were left for 30 min at
37 °C to coagulate and centrifuged 10 min at 10 000 rpm. Total
serum fraction was then collected and frozen at -20 °C until use.
Total cholesterol and total TG were dosed using kits 61219 and
61236, respectively, following manufacturer instructions (Bio
Me´rieux, Marcy-l’e´toile, France). After 10 min of incubation at 37
°C, the colorimetric reaction was read at 492 nm with an iEMS
reader (ThermoLife Sciences, Cergy-Pontoise, France). Cholesterol
HDL, LDL, and VLDL fractions were separated by HPLC. Samples
were diluted 1/5 in phosphate buffer (Ca++ and Mg- free) and
filtered on 0.45 µm to remove excess proteins before HPLC. All
changes reported with an asterisk are statistically significant (p <
0.05) as determined by one-way ANOVA analysis.
0.5
1.0
5.0
-48 ( 8
-63 ( 10
-86 ( 5
-61 ( 2
-65 ( 3
-69 ( 5
+43 ( 4
+51 ( 19
+86 ( 18
-37 ( 6
-48 ( 11
-65 ( 6
a All changes are statistically significant (p < 0.05), as determined by
one-way ANOVA analysis.
Experimental Section
General Methods. All commercial chemicals and solvents are
reagent grade and were used without further purification unless
otherwise specified. All reactions except those in aqueous media
were carried out with the use of standard techniques for the
exclusion of moisture. Reactions were monitored by thin-layer
chromatography on 0.25 mm silica gel plates (60F-254, E. Merck)
and visualized with UV light or 5% phosphomolybdic acid in 95%
ethanol.
Final compounds were typically purified either by flash chro-
matography on silica gel (E. Merck, 40-63 mm) or by recrystal-
lization. H NMR spectra were recorded on either a Brucker 300
MHz Avance DPX or a Bruker 400 MHz Avance DRX instrument.
Chemical shifts are reported in parts per million (ppm, d units).
1

Series of Potent HDLc Raising Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 693
Chemicals. Substituted Thiobenzamides 2a-j. Substituted
thiobenzamides were either purchased from commercial suppliers
(2a, 2d, 2f, 2h Lancaster; 2c Fluorochem; 2g Acros) or were
prepared using one of the routes a or b described below.
reaction was quenched by addition of water (200 mL) and was
stirred for 5 min. The aqueous layer was extracted with dichloro-
methane (2 × 200 mL). The combined organic layers were washed
with hydrochloric acid (1 N, 200 mL), water (200 mL), saturated
aqueous sodium carbonate (200 mL), and brine (200 mL). After
drying over magnesium sulfate, filtration, and concentration to
dryness, the crude material was suspended in isopropyl ether (200
mL), triturated, filtered, and dried to give 24a (47.6 g, 69.7%) as
a white powder: 1H NMR (DMSO-d6): δ 8.87 (t, J ) 5.6 Hz,
1H), 8.14 (d, J ) 8.1 Hz, 2H), 7.87 (d, J ) 8.5 Hz, 2H), 7.23 (d,
J ) 8.7 Hz, 2H), 6.75 (d, J ) 8.7 Hz, 2H), 4.37 (d, J ) 5.8 Hz,
2H), 4.15 (q, J ) 7.1 Hz, 2H), 2.63 (s, 3H), 1.50 (s, 6H), 1.16 (t,
J ) 7.1 Hz, 3H).
Procedure a for the Preparation of 2b. To a solution of P4S10
(0.2 mmol) in toluene (100 mL) was added NaHCO3 (2 mmol),
and the mixture was heated to reflux for about 30 min. The
substituted benzamide (1 mmol) was added, and the reaction was
stirred at 90 °C for 1 h. The reaction was then evaporated to dryness,
treated with brine (100 mL), and extracted with CH2Cl2 (2 × 50
mL). The organic phase was dried, filtered, and evaporated to afford
the final product. In this manner were obtained products 2b (orange
solid; 49%). 1H NMR (CDCl3): δ 7.7 (d, 2H), 7.4 (bs, 1H), 7.3 (d,
2H), 7.0 (bs, 1H), 1.2 (s, 9H).
2-Methyl-2-[4-{[(4-methyl-2-[4-trifluoromethylphenyl]-thia-
zol-5-ylcarbonyl)amino]methyl}phenoxy]propionic Acid (25a).
To a solution of 24a (230.8 g, 0.46 mol) in 1.2 L (5 vol) of
tetrahydrofuran was added aqueous sodium hydroxide (1 N, 480
mL, 1.05 equiv). The solution was stirred under reflux for 18 h.
After removal of THF under reduced pressure, 1 N NaOH (500
mL) and methyl alcohol (100 mL) were added. The aqueous layer
was extracted with dichloromethane (2 × 400 mL) and acidified
to pH ) 1 with concentrated aqueous hydrochloric acid. The oily
residue was extracted with dichloromethane (3 × 400 mL), and
the combined organic layers were washed with brine (600 mL).
After drying over magnesium sulfate, filtration, and concentration
to dryness, the oily residue was triturated with pentane (500 mL),
filtered, and washed with pentane (2 × 250 mL) to give, after
drying, crude compound 25a (207.2 g) as a white powder. The
solid material was dissolved in 310 mL (1.5 vol) of boiling toluene.
After filtration of the hot solution and return to room temperature,
the crystallized material was filtered, washed with toluene (2 ×
200 mL), and dried in vacuo to give 25a (196.3 g, 90%), as a white
Procedure b for the Preparation of Substituted Thiobenz-
amides 2e, 2i, and 2j. To the substituted benzonitrile (1 mmol) in
DMF (30 mL) was added dropwise DMF (21 mL) saturated with
HCl(g) over a period of 1 min. Thioacetamide (2 mmol) was then
added, and the reaction was heated to 100 °C for 1 h. HCl(g) was
bubbled in for about 1 min, and stirring was continued at 100 °C
for another 18 h. The reaction was cooled to rt, treated with ice,
and extracted with Et2O (3 × 250 mL). The organic phase was
washed with H2O (3 × 300 mL), dried over Na2SO4, filtered, and
evaporated to dryness. The residue was washed with a mixture of
isopropyl ether/pentane (1:3) to afford the final product. In this
1
manner were obtained products 2e (orange solid; 83%), H NMR
(DMSO-d6) δ 10.1 (bs, 1H), 9.7 (bs, 1H), 8.1 (d, 2H), 7.9 (d, 2H);
1
2i (light yellow solid; 88%), H NMR (DMSO) δ 8.20 (bs, 1H),
8.08 (s, 1H), 8.02 (d, J) 7.9 Hz, 1H), 7.91 (d, J) 7.9 Hz, 1H),
7.76 (t, J) 7.9 Hz,1H), 7.34 (bs, 1H); and 2j (light yellow solid;
1
47%), H NMR (DMSO) δ 7.9-7.4 (m, 4H), 7.8 (bs, 1H), 7.06
(bs, 1H).
Ethyl 4-Methyl-2-[4-(trifluoromethyl)phenyl]-1,3-thiazole-5-
carboxylate (3a). To a suspension of of 4-(trifluoromethyl)-
thiobenzamide 2a (302.4 g, 1.47mol.) in ethyl alcohol (1.5L, 5 vol)
was added at room temperature ethyl 2-chloroacetoacetate (203.8
mL, 1 equiv). The solution was refluxed for 24 h, and then the
solvent was removed under reduced pressure. The solid material
was stirred with cooled hexane (500 mL) for 30 min, filtered, and
washed with hexane (2 × 150 mL). Drying gave crude compound
3a (352.9 g). A second crop of 25.7 g was obtained by concentration
of the hexane filtrates to 50 mL, giving an overall yield of 378.6
g (81.5%). 1H NMR (CDCl3): δ 8.06 (d, J)8.1 Hz, 2H), 7.69 (d,
J)8.2 Hz, 2H), 7.36 (q, J)7.1 Hz, 2H), 2.78 (s, 3H), 1.39 (t, J)7.1
Hz, 3H).
4-Methyl-2-[4-(trifluoromethyl)phenyl]-1,3-thiazole-5-car-
boxylic Acid (17a). To a cooled solution of 3a (378.6 g, 1.2 mol)
in ethyl alcohol (2 L, 5 vol) was added a solution of sodium
hydroxide (96.15 g, 2 equiv) in 2 L of water. The solution was
heated at 85 °C for 1.5 h. After evaporation of the ethyl alcohol,
the aqueous solution was diluted with 2 L of water and acidified
to pH ) 1 with concentrated aqueous hydrochloric acid. The solid
material was filtered and washed twice with 1 L of water and 1 L
of dichloromethane. After drying in a vacuum oven, 17a (267.2 g)
was obtained as an off-white powder. A second crop of 25.7 g was
obtained by concentration of the dichloromethane and triturating
with pentane, giving an overall yield of 17a of 292.9 g (85%). 1H
NMR (DMSO-d6): δ 13.51 (bs, 1H), 8.15 (d, J)8.3 Hz, 2H), 7.84
(d, J)8.7 Hz, 2H), 2.68 (s, 3H).
2-Methyl-2-[4-{[(4-methyl-2-[4-trifluoromethylphenyl]thiazol-
5-ylcarbonyl)amino]methyl}phenoxy]propionic Acid Ethyl Ester
(24a). A suspension of crude 17a (38.7 g, 0.13 mol) in thionyl
chloride (200 mL, 5 vol) was refluxed for 3 h. After return to room
temperature, the thionyl chloride was removed under reduced
pressure, and the residue was washed twice with toluene (100 mL)
and evaporated to dryness. The crude acid chloride obtained (off-
white solid) was used without purification. To a mixture of crude
21 (35.5 g; 1 equiv/LC-MS purity: 90%) and triethylamine (20.62
mL, 1.1 equiv) in dichloromethane (350 mL, 10 vol), maintained
at 10 °C, was added portionwise the acid chloride over 20 min,
and the mixture was stirred at room temperature overnight. The
1
powder; mp 130-131 °C; H NMR (DMSO-d6) δ 13.0 (bs, 1H),
8.87 (t, J ) 5.7 Hz, 1H), 8.14 (d, J ) 8.1 Hz, 2H), 7.87 (d, J ) 8.3
Hz, 2H), 7.22 (d, J ) 8.7 Hz, 2H), 6.78 (d, J ) 8.7 Hz, 2H), 4.37
(d, J ) 5.9 Hz, 2H), 2.63 (s, 3H), 1.48 (s, 6H). MS (APCI) m/z
479 (M + H)+. HRMS calcd for C23H21F3N2O4S (M + H),
479.1252; found, 479.1236. Analytical HPLC tR ) 1.85 min, 100%
pure (A); tR ) 2.16 min, 100% pure (C).
2-Methyl-2-{[4-({[(4-methyl-2-{4-[(methylsulfonyl)oxy]phen-
yl}-1,3-thiazol-5-yl)carbonyl]amino}methyl)phenyl]oxy}propanoic
Acid (25l). To a solution of 24l (0.34 g, 0.64 mmol) in THF (10
mL) was added a solution of LiOH in water (1 equiv, 2.5 mL),
and the mixture was stirred at rt for 24 h. The THF was evaporated,
and the aqueous phase was acidified with HCl (1 N) and then
extracted with ethyl acetate (50 mL). The organic layer was dried
over Na2SO4, filtered, and evaporated to dryness. The residue was
chromatographed, eluting with CH2Cl2/MeOH 95/5 then 90/10 to
1
afford 25l as a white solid (50 mg, 15%). H NMR (DMSO-d6) δ
8.78 (t, J ) 5.8 Hz, 1H), 8.03 (d, J ) 8.9 Hz, 2H), 7.48 (d, J ) 8.7
Hz, 2H), 7.13 (d, J ) 8.7 Hz, 2H), 6.80 (d, J ) 8.5 Hz, 2H), 4.33
(d, J ) 5.7 Hz, 2H), 3.44 (s, 3H), 2.61 (s, 3H), 1.40 (s, 6H). MS
(APCI) m/z 505 (M + H)+. HRMS calcd for C23H24N2O7S2 (M +
H), 505.1103; found, 505.1088. Analytical HPLC tR ) 1.59 min,
100% pure (A); tR ) 1.93 min, 100% pure (C).
2-({4-[({[2-(4-Hydroxyphenyl)-4-methyl-1,3-thiazol-5-yl]-
carbonyl}amino)methyl]phenyl}oxy)-2-methylpropanoic Acid
(25m). To a solution of 24l (0.20 g, 0.38 mmol) in THF (30 mL)
and EtOH (5 mL) was added NaOH solution (1 N, 1.85 mL, 5
equiv), and the mixture was stirred at 70 °C for 4 h. The mixture
was cooled to room temperature, and the solution was acidified
with HCl (1 N) and extracted with ethyl acetate (3 × 50 mL). The
combined organic layers were washed with H2O, dried over Na2SO4,
filtered, and evaporated to dryness. The residue was crystalized in
a mixture of CH2Cl2 and THF to afford 25m (0.12 g, 74%) as white
crystals; mp 151-152 °C. 1H NMR (DMSO-d6): δ 13.00 (s, 1H),
10.13 (s, 1H), 8.67 (s, 1H), 7.35 (d, J ) 8.7 Hz, 2H), 7.21 (d, J )
8.7 Hz, 2H), 6.86 (d, J ) 8.7 Hz, 2H), 6.77 (d, J ) 8.6 Hz, 2H),
4.33 (d, J ) 5.6 Hz, 2H), 2.57 (s, 3H), 1.48 (s, 6H). MS (APCI)
m/z 427 (M + H)+. HRMS calcd for C22H22N2O5S (M + H),

694 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Table 8. Data Collection and Refinement Statistics
Sierra et al.
The hanging drops contained 1 µL of the above protein-ligand
solution and 1 µL of well buffer comprising 7% PEG 3350, 200
mM NaF, and 12% 2,5-hexanediol. Crystals formed in the space
group P21212 with cell dimensions a ) 61.3 Å, b ) 103.5 Å, and
c ) 49.9 Å. Each asymmetric unit contained a single LBD complex
with 45% solvent content. X-ray data was collected at the IMCA
17-ID line (Argonne, IL) and processed with HKL2000.20 Statistics
are summarized in Table 8.
Structure Determination and Refinement. The structure was
determined by molecular replacement methods with the program
AmoRe,21 as implemented in the CCP4 suite.22 The structure of
the PPARR LBD,23 residues 167-441, was used as the initial
model. The best fitting solution gave a correlation coefficient of
70% and an R-factor of 33%. Model building was performed with
the software program QUANTA, and structure refinement was
carried out with the CNS software program.24 Refinement statistics
are summarized in Table 8.
crystallization
PDB code
PPARR/SRC1 peptide 25a
to be deposited
unit cell dimensions
a (Å)
b (Å)
c (Å)
space group
molecules per asymmetric unit
resolution (Å)
number of unique reflections
data redundancy
completeness (%)
61.3
103.5
49.9
P21212
1
50-1.8
30 145
7.1
99.4
4.5
Rsymm (%)
I/σ
39.4
Refinement
R
Rfree
20.4
22.7
r.m.s. deviation from ideality
bond length (Å)
bond angle (degrees)
average B-factor (Å2)
all atoms
protein atoms
water molecules
ligand
Acknowledgment. The authors thank Melanie Barnathan
and Vanessa Cheminade for valuable technical assistance.
0.005
1.288
Supporting Information Available: Experimental details and
data for compounds 3b-k, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
16, 17b-k, 18, 19, 20, 21, 22, 23, 24b-j, 25b-j, 26, 27, 28, 29,
30, 31, 32, 33, 34, 35, 36, 37, 24k, 25k, 24l, 39, 40, 41, 42, 43, 44,
and 45 plus a table of analytical data for target compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org
27.0
25.9
37.5
19.7
427.1328; found, 427.1319. Analytical HPLC tR ) 2.06 min, 100%
pure (B); tR ) 1.78 min, 98.9% pure (C).
Ethyl 5-Methyl-2-[4-(trifluoromethyl)phenyl]-1,3-thiazole-4-
carboxylate (38). To a solution of 2-ketobutyric acid (5.0 g,
49mmol) in CH2Cl2 (10 mL) was slowly added bromine (1 equiv;
2.5 mL), and the mixture was stirred at rt for 15 min. The solvent
was evaporated. Toluene was added and the solvent was evaporated
again. The residue was dissolved in EtOH, then 1a (8.5 g; 0.85
equiv) was added, and the mixture was stirred at 80 °C for 18 h.
The solvent was evaporated, and the crude product was chromato-
graphed, eluting with CH2Cl2 to afford 38 (4.56 g, 35%) as a white
solid. 1H NMR (CDCl3): δ 8.04 (d, J ) 8.4 Hz, 2H), 7.68 (d, J )
8.1 Hz, 2H), 4.45 (q, J ) 7.2 Hz, 2H), 2.82 (s, 3H), 1.43 (t, J )
7.2 Hz, 3H).
Protein Preparation. The PPARR ligand binding domain (amino
acids 192-468) with an N-terminal 6xHis tag was expressed using
the T7 promoter of plasmid vector pRSETA. BL21(DE3) E. coli
cells transformed with this expression vector were grown at 24 °C
in shaker flasks for 66 h. The cells were harvested, resuspended,
and lysed. The lysed cells were centrifuged, and the supernatant
was loaded on a Ni-agarose column. The column was washed with
150 mL of buffer A (10% glycerol, 20 mM HEPES pH 7.5, 25
mM imidazole), and the protein was eluted with a 450 mL gradient
of buffer B (10% glycerol, 20 mM HEPES pH 7.5, 500 mM
imidazole). The protein, which eluted at 20% buffer B, was diluted
with one volume of buffer C (20 mM HEPES, pH 7.5, 1 mM
EDTA) and loaded on a 100 mL S-Sepharose (Pharmacia, Peapack,
New Jersey) column. The column was washed with 100 mL buffer
C, and the PPARR LBD protein was eluted with a 200 mL gradient
of buffer D (20 mM HEPES, pH 7.5, 10 mM DTT, 1 M ammonium
acetate). The PPARR LBD eluted from the column at 43% buffer
D. The protein yield was 9 mg/L of cells grown and was >95%
pure, as determined by SDS-PAGE analysis.
References
(1) Tobert, J. A. Lovastatin and Beyond: The History of the HMG-
CoA Reductase Inhibitors. Nat. ReV. Drug DiscoVery 2003, 2, 517-
526.
(2) Sprecher, D. L. Raising High-Density Lipoprotein Cholesterol with
Niacin and Fibrates: A Comparative Review. Am. J. Cardiol. 2000,
86 (suppl), 46L-50L.
(3) Rader, D. J. Effects of Nonstatin Lipid Drug Therapy on High-Density
Lipoprotein Metabolism. Am. J. Cardiol. 2003, 91 (suppl), 18E-
23E.
(4) Wise, A.; Foord, S. M.; Fraser, N. J.; Barnes, A. A.; Elshourbagy,
N.; Eilert, M.; Ignar, D. M.; Murdock, P. R.; Steplewski, K.; Green,
A.; Brown, A. J.; Dowell, S. J.; Szekeres, P. G.; Hassall, D. G.;
Marshall, F. H.; Wilson, S.; Pike, N. B.; Molecular Identification of
High and Low Affinity Receptors for Nicotinic Acid. J. Biol. Chem.
2003, 278, 9869-9874.
(5) Issemann, I.; Green, S.; Activation of a member of the steroid
hormone receptor superfamily by peroxisome proliferators. Nature
1990, 347, 645-650.
(6) Forman, B. M.; Chen, J.; Evans, R. M.; Hypolipidemic drugs,
polyunsaturated fatty acids, and eicosanoids are ligands for peroxi-
some proliferator-activated receptors R and δ. Proc. Natl. Acad. Sci.
U.S.A. 1997, 94, 4312-4317.
(7) 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.
(8) Hertz, R.; Bishara-Shieban, J.; Bar-Tana, J.; Mode of Action of
Peroxisome Proliferators as Hypolipidemic Drugs. Suppression of
Apolipoprotein C-III. J. Biol. Chem. 1995, 270, 13470-13475.
(9) Berthou, L.; Duverger, N.; Emmanuel, F.; Langouet, S.; Auwerx, J.;
Guillouzo, A.; Fruchart, J.-C.; Rubin, E.; Dene`fle, P.; Staels, B.;
Branellec, D.; Opposite Regulation of Human Versus Mouse Apo-
lipoprotein A-1 by Fibrates in Human Apolipoprotein A-1 Transgenic
Mice. J. Clin. InVest. 1996, 97, 2408-2416.
(10) Willson, T.; Brown, P. J.; Sternbach, D. D.; Henke, B. R. The
PPARs: From Orphan Receptors to Drug Discovery. J. Med. Chem.
2000, 43, 527-550.
The protein was then diluted to 1 mg/mL with buffer C such
that the final buffer composition was 220 mM ammonium acetate,
20 mM HEPES pH 7.5, 1 mM EDTA, and 1 mM DTT. The peptide
SRC116 was added in a molar ratio of 1.5 as a 2 mg/100 µL DMSO
stock. The ligand was then added in a 5:1 molar ratio as a 2 mg/
100 µL DMSO stock and spun at 4 K for 20 min to clarify the
solution before concentrating in Centriprep 10 filtration units
(Millipore, Bedford, Massachusetts). The solution containing the
PPARR LBD-SRC1 complexes was concentrated to approximately
10 mg/mL with 80% yield.
(11) Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, K.; Leitersdorf,
E.; Fruchart, J.-C. Mechanism of Action of Fibrates on Lipid and
Lipoprotein Metabolism. Circulation 1998, 98, 2088-2093.
(12) (a) 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. Design and Synthesis of a Potent and Selective
Triazolone-Based Peroxisome Proliferator-Activated Receptor R
Agonist. J. Med. Chem. 2003, 46, 5121-5124. (b) Nomura, M.;
Tanase, T.; Ide, T.; Tsunoda, M.; Suzuki, M.; Uchiki, H.; Murakami,
K.; Miyachi, H.; Design, Synthesis, and Evaluation of Substituted
Crystallization and Data Collection. The crystals were grown
at room temperature using the hanging drop vapor diffusion method.

Series of Potent HDLc Raising Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 695
Phenylpropanoic Acid Derivatives as Human Peroxisome Proliferator
Activated Receptor Activators. Discovery of Potent and Human
Peroxisome Proliferator Activated Receptor R Subtype-Selective
Activators. J. Med. Chem. 2003, 46, 3581-3599. (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. A Ureido-
Thioisobutyric Acid (GW9578) Is a Subtype-Selective PPARR
Agonist with Potent Lipid-Lowering Activity. J. Med. Chem. 1999,
42, 3785-3788.
(17) Gampe, R. T., Jr.; Montana, V. G.; Lambert, M. H.; Miller, A. B.;
Bledsoe, R. K.; Milburn, M. V.; Kliewer, S. A.; Willson, T. M.; Xu,
H. E. Asymmetry in the PPARγ/RXRR Crystal Structure Reveals
the Molecular Basis of Heterodimerization among Nuclear Receptors.
Mol. Cell 2000, 5, 545-555.
(18) Takada, I.; Yu, R. T.; Xu, H. E.; Lambert, M. H.; Montana, V. G.;
Kliewer, S. A.; Evans, R. M.; Umesono, K. Alteration of a Single
Amino Acid in Peroxisome Proliferator-Activated Receptor-R (PPARR)
Generates a PPARδ Phenotype. Mol. Endocrinol. 2000, 14, 733-
740.
(19) Berthou, L.; Duverger, N.; Emmanuel, F.; Langouet, S.; Auwerx, J.;
Guillouzo, A.; Fruchart, J. C.; Rubin, E.; Denefle, P.; Staels, B.;
Branellex, D. Opposite Regulation of Human Versus Mouse Apo-
lipoprotein A-I by Fibrates in Human Apolipoprotein A-I Transgenic
Mice. J. Clin. InVest. 1996, 97, 2408-2416.
(20) Otwinowski, Z.; Minor, W. Macromolecular Crystallography, Part
A. In Methods in Enzymology; Carter, C. W., Jr., Sweet, R. M., Eds.;
Academic Press: New York, 1997; Vol. 276, pp 307-326.
(21) Navaza, J. AMoRe: An Automated Package for Molecular Replace-
ment. Acta Crystallogr. 1994, A50, 157-163.
(22) Collaborative Computational Project, Number 4. “The CCP4 Suite:
Programs for Protein Crystallography”. Acta Crystallogr. 1994, D50,
760-763.
(23) 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.; Kleiwer, S. A.; Milburn, M. V. Molecular
Recognition of Fatty Acids by Peroxisome Proliferator-Activated
Receptors. Mol. Cell 1999, 3, 397-403.
(24) Bru¨nger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros,
P.; Grosse-Kunstleve, R. W.; Jiang, J.-S.; Kuszewski, J.; Nilges, M.;
Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography and NMR System: A New Software Suite for
Macromolecular Structure Determination. Acta Crystallogr. 1998,
D54, 905-921.
(13) Oliver, W. R.; Shenk, J. L.; Snaith, M. R.; Russell, C. S.; Plunket,
L. D.; Bodkin, N. L.; Lewis, M. C.; Winegar, D. A.; Sznaidman, M.
L.; Lambert, M. H.: Xu, E.; Sternbach, D. D.; Kliewer, S. A.;
Willson, T. M. A Selective Peroxisome Proliferator-Activated
Receptor δ Agonist Promotes Reverse Cholesterol Transport. Proc.
Natl. Acad. Sci. U.S.A. 2001, 98, 5306-5311.
(14) 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. N-(2-Benzoylphenyl)-L-
tyrosine PPARγ Agonists. 1. Discovery of a Novel Series of Potent
Antihyperglycemic and Antihyperlipidemic Agents. J. Med. Chem.
1998, 41, 5020-5036.
(15) 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.
(16) Cronet, P.; Petersen, J. F. W.; Folmer, R.; Blomberg, N.; Sjoblom,
K.; Karlsson, U.; Lindstedt, E.-L.; Bamberg, K. Structure of the
PPARR and -γ Ligand Binding Domain in Complex with AZ 242;
Ligand Selectivity and Agonist Activation in the PPAR Family.
Structure 2001, 9, 699-706.
JM058056X