New 2-(aryloxy)-3-phenylpropanoic acids as peroxisome proliferator- activated receptor α/γ dual agonists able to upregulate mitochondrial carnitine shuttle system gene expression

Article abstract of DOI:10.1021/jm301018z

The preparation of a series of 2-(aryloxy)-3-phenylpropanoic acids, resulting from the introduction of different substituents into the biphenyl system of the previously reported peroxisome proliferator-activated receptor α/γ (PPARα/γ) dual agonist 1, allowed the identification of new ligands with higher potency on PPARα and fine-tuned moderate PPARγ activity. For the most promising stereoisomer (S)-16, X-ray and calorimetric studies in PPARγ revealed, at high ligand concentration, the presence of two molecules simultaneously bound to the receptor. On the basis of these results and docking experiments in both receptor subtypes, a molecular explanation was provided for its different behavior as a full and partial agonist of PPARα and PPARγ, respectively. The effects of (S)-16 on mitochondrial acylcarnitine carrier and carnitine-palmitoyl-transferase 1 gene expression, two key components of the carnitine shuttle system, were also investigated, allowing the hypothesis of a more beneficial pharmacological profile of this compound compared to the less potent PPARα agonist fibrates currently used in therapy.

Full text of DOI:10.1021/jm301018z

Article
pubs.acs.org/jmc
New 2‑(Aryloxy)-3-phenylpropanoic Acids as Peroxisome
Proliferator-Activated Receptor α/γ Dual Agonists Able To
Upregulate Mitochondrial Carnitine Shuttle System Gene Expression
A. Laghezza,† G. Pochetti,§ A. Lavecchia,∥ G. Fracchiolla,† S. Faliti,† L. Piemontese,† C. Di Giovanni,∥
V. Iacobazzi,‡,⊥ V. Infantino,‡,# R. Montanari,§ D. Capelli,§ P. Tortorella,† and F. Loiodice*,†
‡
†Dipartimento di Farmacia-Scienze del Farmaco and Laboratorio di Biochimica e Biologia Molecolare, Dipartimento di Bioscienze,
Biotecnologie e Biofarmaceutica, Universita
§Istituto di Cristallografia, Consiglio Nazionale delle Ricerche, Montelibretti, 00015 Monterotondo Stazione, Roma, Italy
∥Dipartimento di Chimica Farmaceutica e Tossicologica, Universita
di Napoli “Federico II”, 80131 Napoli, Italy
̀
degli Studi di Bari “Aldo Moro”, 70126 Bari, Italy
̀
⊥Istituto di Biomembrane e Bioenergetica, Consiglio Nazionale delle Ricerche, 70126 Bari, Italy
#Dipartimento di Chimica, Universita
̀
della Basilicata, 85100 Potenza, Italy
S
* Supporting Information
ABSTRACT: The preparation of a series of 2-(aryloxy)-3-phenylpropanoic acids, resulting
from the introduction of different substituents into the biphenyl system of the previously
reported peroxisome proliferator-activated receptor α/γ (PPARα/γ) dual agonist 1, allowed
the identification of new ligands with higher potency on PPARα and fine-tuned moderate
PPARγ activity. For the most promising stereoisomer (S)-16, X-ray and calorimetric studies in
PPARγ revealed, at high ligand concentration, the presence of two molecules simultaneously
bound to the receptor. On the basis of these results and docking experiments in both receptor
subtypes, a molecular explanation was provided for its different behavior as a full and partial
agonist of PPARα and PPARγ, respectively. The effects of (S)-16 on mitochondrial
acylcarnitine carrier and carnitine-palmitoyl-transferase 1 gene expression, two key
components of the carnitine shuttle system, were also investigated, allowing the hypothesis
of a more beneficial pharmacological profile of this compound compared to the less potent PPARα agonist fibrates currently used
in therapy.
diabetic patients. The success of the hypolipidemic fibrates and
TZD class of insulin sensitizers has prompted pharmaceutical
companies to concentrate on developing more potent and dual
agonists acting on these two subtypes. Dual-acting PPARα/γ
agonists, in fact, are considered a very attractive option in the
treatment of dyslipidemic type 2 diabetes.10−18
One of the key challenges for the development of a dual
agonist is the identification of the optimal receptor subtype
selectivity ratio. PPARγ full agonists, in fact, despite their
proven benefits, have been plagued by certain adverse effects
such as weight gain, higher rate of bone fractures, fluid
accumulation, and pulmonary edema, leading to increased
frequency of congestive heart failure.19−21 However, as
demonstrated from the problems that many dual agonists
have encountered during their development, an ideal PPARα/γ
activity ratio probably does not exist. For a specific series of
strictly related ligands, in fact, this value depends on many
factors such as tissue selectivity and ability to recruit particular
coactivator proteins and activate the transcription of specific
target genes. During the past decade, therefore, new drugs
INTRODUCTION
■
Peroxisome proliferator-activated receptors (PPARs) are
ligand-activated transcription factors belonging to the nuclear
hormone receptor superfamily. They are lipid sensors activated
by specific ligands, which are usually lipophilic small
molecules.1 Binding of these ligands results in conformational
changes of the receptors that facilitate their interaction with
coactivator proteins in the nucleus.2−4 The resulting protein
complexes activate the transcription of specific target genes,
resulting in the induction of intracellular signaling cascades that
mediate the physiological effects of the ligands.5,6
There are three distinct subtypes, PPARα, PPARγ, and
PPARδ, which are encoded by different genes. The different
PPARs have different tissue distributions and modulate
different physiological functions. PPARs play a key role in
various aspects of the regulation of a large number of genes; the
products of these genes are directly or indirectly involved in
energy homeostasis and lipid and carbohydrate metabolism.
The fibrate class of lipid-lowering drugs (e.g., fenofibrate and
gemfibrozil) are PPARα ligands.7,8 The marketed thiazolidine-
dione (TZD) class of antidiabetic agents (rosiglitazone and
pioglitazone) activate PPARγ.9 They enhance insulin sensitivity
in target tissues and lower glucose and fatty acid levels in type 2
Received: July 12, 2012
Published: November 22, 2012
© 2012 American Chemical Society
60
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
Table 1. Activity of the Tested Compounds in Cell-Based Transactivation Assay
PPARα
PPARγ
a
a
compd
config
R
R1
EC50 (μM)
Emax
EC50 (μM)
Emax
1
S
H
H
H
H
H
H
0.19 0.04
0.99 0.08
0.08 0.03
1.13 0.03
2.1 0.5
116
68
4
0.55 0.12
0.70 0.10
0.70 0.04
3.1 0.6
2.04 1.10
2.54 0.37
0.59 0.26
0.46 0.13
1.36 0.26
0.64 0.14
0.69 0.19
1.00 0.14
1.4 0.7
2.30 0.01
1.2 0.4
1.0 0.2
nc
62
34
47
18
13
16
47
43
42
49
37
32
10
21
28
33
12
8
7
4
3
1
1
1
1
4
4
2
6
2
1
6
2
3
8
2
S
3-CH3
8
1
3
S
2-CH3
120
61
9
4
S
3,5-(CH3)2
5
R,S
S
2,6-(CH3)2
50 16
6
H
2′,6′-(CH3)2
4′-CH3
3′-CH3
2′-CH3
2′-CH3
2′-F
2′-CF3
2′-Ph
2′-OCH3
2′-C2H5
2′-iPr
0.23 0.04
0.21 0.09
0.22 0.08
0.067 0.005
0.031 0.002
0.12 0.04
0.022 0.002
0.044 0.028
0.13 0.06
0.012 0.002
0.012 0.007
0.10 0.03
0.24 0.01
0.50 0.05
0.28 0.03
1.56 0.30
ia
123 13
7
S
H
74
69
3
4
8
R,S
S
H
9
H
155 12
10
11
12
13
14
15
S
2-CH3
H
133
68
8
S
6
S
H
126
132
6
5
S
H
S
H
124 14
143
S
H
8
(S)-16
(R)-16
17
S
H
147 20
145 24
R
H
2′-iPr
2′-nPr
2′-nC5H11
2′-CH2CH2Ph
R,S
R,S
R,S
H
131
4
nc
6
18
H
107 20
73 11
100 10
ia
nc
11
7
6
19
H
nc
2
Wy 14,643
rosiglitazone
ia
ia
0.039 0.003
100
9
a
Efficacy values were calculated as a percentage of the maximum obtained fold induction with the reference compounds. ia = inactive at tested
concentrations (up to 10 μM). nc = not computable; the activity, in fact, increases with increasing concentrations up to 10 μM, above which the
activity begins to decrease.
acting as full agonists of PPARα and partial agonists of PPARγ
have been developed with the goal of retaining the beneficial
effects while reducing the adverse effects.
was the identification of a series of ligands which, while
maintaining high potency and efficacy on PPARα, presented a
fine-tuned moderate PPARγ activity. This could allow
investigation, in a series of strictly related compounds, of the
effects due to small differences in PPARγ activity and
identification of the optimal selectivity ratio for this series of
dual agonists. For this purpose, we synthesized the new 2-
(aryloxy)-3-phenylpropanoic acids 2−19 reported in Table 1
and evaluated their PPARα/γ activity. Considering the high
degree of stereoselectivity displayed from these receptors
toward the stereoisomer 1,22 most derivatives were synthesized
only as the S isomers. Noteworthy, we identified new ligands
showing full agonism and higher potency on PPARα while
maintaining basically unchanged potency on PPARγ with
efficacy variable in the 10−50% range. For compound (S)-16,
X-ray and calorimetric studies with PPARγ and docking
experiments with both receptor subtypes were performed to
provide a molecular explanation for its different activity. This
ligand was selected because it was the most potent PPARα full
agonist with fairly good potency and concomitant moderate
efficacy on PPARγ (about 3-fold less effective than
rosiglitazone). Finally, we investigated the effects of (S)-16
on mitochondrial carnitine/acylcarnitine carrier (CAC) and
carnitine-palmitoyl-transferase 1 (CPT1) gene expression.
These two molecular components of the carnitine shuttle
system are essential for the mitochondrial oxidation of fatty
Recently, we reported the synthesis and activity of the new
ligand 1 (Table 1), which is a dual PPARα/γ ligand with a
partial agonism profile toward the γ subtype.22 The X-ray
structure of the complex with PPARγ shows that this ligand
occupies a branch, named the “biphenyl pocket”, of a new L-
shaped region of the PPARγ ligand binding domain (LBD)
never sampled before by other known ligands and increases the
stabilization of the helix H3, inducing a conformation of the
LBD less favorable to the recruitment of coactivators required
for full activation of PPARγ.22 As regards PPARα, the ligand
assumes a different orientation in the LBD that better stabilizes
the helices H11 and H12, resulting in a full agonism response.
The identification of these novel ligand−receptor interaction
modalities represents a new hallmark of the partial agonism
behavior of certain ligands and could be exploited for the
design of new antidiabetic agents appropriately targeting the
PPARs. Since 1 occupies only the first branch of the new L-
shaped region, we decided to prepare some analogues
characterized by the presence of a substituent, with different
lengths and stereoelectronic properties, on the aromatic rings of
the biphenyl system that could allow the complete occupation
of the entire cavity and the evaluation of the effects due to a
further stabilization of H11, H12, and the loop 11/12. The aim
61
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of PPAR Agonists 2−16
a
Reagents and conditions: (i) ArOH, DVB−PPh3, DIAD, dry toluene; (ii) ArB(OH)2, Cs2CO3, DVB−Pd(PPh3)4 or Pd(PPh3)4, dry toluene, reflux;
(iii) 1 N NaOH, THF, rt.
acids because they catalyze the entry of fatty acid acyl groups
into the mitochondrial matrix where the enzymes of fatty acid
β-oxidation are located. Recent studies demonstrated that the
transcription of the CAC gene is enhanced by statins and
fibrates, providing, therefore, a novel contribution to the
understanding of their hypolipidemic action.23−25
reaction of 22 with the suitable substrates afforded the
unsaturated intermediates 23a−c, which were reduced under
hydrogen pressure in the presence of the Wilkinson catalyst or
10% palladium on carbon to give esters 24a−c, whose alkaline
hydrolysis led to the target acids 17−19. The attempt to obtain,
by this procedure, the S isomers of 17−19 starting from 20a
failed due to the racemization process occurring in the Wittig
reaction.
The enantiomeric excess (ee) of all optically active isomers
was determined by HPLC on a chiral stationary phase (see the
Experimental Section). For this purpose, small amounts of the
corresponding racemates were also prepared starting from
racemic ethyl phenyllactate. In most cases, the ee was higher
than 94−95%; only acids 10, 11, 12, and 14 showed an ee in
the 75−88% range.
CHEMISTRY
■
The synthesis of compounds 2−16 is depicted in Scheme 1 and
was carried out starting from the ethyl ester of racemic
phenyllactic acid or (R)-phenyllactic acid and the suitably
substituted 4-bromophenol, which were condensed, under
Mitsunobu conditions, to obtain the key intermediates 20a−
g.22 These underwent Suzuki coupling by condensation with
the suitable boronic acids to give esters 21a−r, which were
hydrolyzed to the desired acids. Through this procedure acid
(R)-16 was prepared as well; in this case (S)-ethyl phenyllactate
was used as the starting material. Alternatively, to reduce the
unacceptably long reaction time required for the Suzuki
coupling under normal conditions, the appropriate intermediate
20c was condensed with 2-formylphenylboronic acid in a
microwave reactor to give compound 22 (Scheme 2). Wittig
RESULTS AND DISCUSSION
■
Compounds 2−19 were evaluated for their agonist activity on
the human PPARα (hPPARα) and PPARγ (hPPARγ) subtypes.
For this purpose, GAL4-PPAR chimeric receptors were
expressed in transiently transfected HepG2 cells according to
a previously reported procedure.26 The results obtained (Table
1) were compared with corresponding data for Wy-14,643 and
rosiglitazone used as reference compounds in the PPARα and
PPARγ transactivation assays, respectively. The maximum
induction obtained with the reference agonist was defined as
100%.
a
Scheme 2. Synthesis of PPAR Agonists 17−19
We first set the series of derivatives 2−10 bearing one or two
methyl groups on different positions of the biphenyl system
with the aim to identify the most promising sites of substitution
in terms of potency and efficacy. As shown in Table 1, when at
least a methyl group was located ortho to the biphenyl simple
bond, full agonists of PPARα were obtained with similar (6) or
2−6-fold higher (3, 9, 10) potency as compared to lead
compound 1. It is reasonable to hypothesize that the presence
of a methyl in this position induces the biphenyl system of
these ligands to assume a torsion angle favoring interactions
with the receptor. The only exception was represented from the
2,6-dimethyl derivative 5 exhibiting relatively low potency and
partial agonism; even though this compound was tested as a
racemate, in fact, it must be taken into account that its possible
a
Reagents and conditions: (i) 2-formylphenylboronic acid, Cs2CO3,
Pd(PPh3)4, THF; microwave, 100 °C, 150 W, 1 h; (ii) R2CH2P-
(Ph)3X, DBU, dry CH3CN, N2, reflux; (iii) H2 (10−15 atm),
Wilkinson catalyst, EtOH, rt for esters 24a,b or 10%Pd/C, EtOH, rt
for ester 24c; (iv) 1 N NaOH, THF, rt.
62
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
S eutomer would have at most a 2-fold higher potency than the
racemate with unchanged efficacy. In this case, it is likely that
the concomitant presence of two methyls in these positions of
the proximal benzene ring causes some steric clash, which does
not allow a favorable accommodation of the molecule into the
binding site. The introduction of a methyl in other positions of
the biphenyl nucleus still afforded potent PPARα activators (2,
4, 7, 8) endowed, however, with partial agonist activity. As
regards PPARγ activity, derivatives 2−10 were all partial
agonists with lower potency compared to PPARα. No
correlation was found between activity and the type of
substitution, suggesting less constraining structural require-
ments for these ligands in PPARγ. These analogues showed the
desired variability of efficacy, which ranged from 15% to 50%;
however, with the aim to search for even more potent PPARα
agonists, we decided on an in-depth investigation of the ortho
position to the biphenyl bond, which seemed to be the most
promising site to achieve this result. For this reason, we decided
to synthesize compounds 11−19 in which the methyl group
located in the 2′ site of 9 was replaced by substituents with
different stereoelectronic properties. Most derivatives exhibited
full agonism toward PPARα with submicromolar potency. The
activity was increased about 5-fold by replacing the methyl with
ethyl (15) or a branched alkyl group such as isopropyl (16),
whereas it remained almost unchanged with the introduction of
an n-propyl (17); this last compound, however, was tested as a
racemate as well as 18, in which a further lengthening of the
alkyl chain to pentyl afforded a marked reduction of activity
probably due to the lack of steric requirements. More potent
agonists were also achieved by increasing lipophilicity as shown
from trifluoromethyl and phenyl derivatives 12 and 13,
respectively. When the benzene ring of 13 was moved away
from biphenyl by introduction of an ethyl chain (19), a 6-fold
decrease in potency resulted, suggesting a possible reduced
capacity to form a favorable π−π interaction with some residues
of the receptor and/or a poor accommodation of the phenethyl
group into the binding pocket because of its steric hindrance.
Also in this case, however, it must be taken into account that 19
was tested as a racemate. PPARα activity, on the contrary,
seemed to be less affected by the electronic properties of the
substituents; the presence of an electron-withdrawing fluorine
(11) or an electron-releasing methoxyl (14) afforded, in fact,
ligands with the same potency even though they had different
efficacies. It is possible that, in this set of analogues, the right
size of the substituent is needed to obtain the best PPARα
activity: too small or too large substituents, in fact, seem to not
be appropriate for a favorable interaction with the receptor. As
regards PPARγ, also in this case, all derivatives 11−19 displayed
lower potency and efficacy compared to PPARα. No relevant
difference of activity was observed by changing the electronic
and lipophilic properties of the substituents with the exception
of methoxyl derivative 14, which, differently from PPARα, was
3 times less potent than fluoro derivative 11. Steric require-
ments, also, did not significantly affect activity as shown from
the similar potency and efficacy of 9, 15, and 16. However,
compounds 17−19 with long-chain substituents exhibited poor
activity up to 5−10 μM, suggesting that spatial limitation may
exist for the binding pocket that accommodates this part of the
molecule.
showed the best dual agonist profile on the α and γ subtypes. It
was, in fact, the most potent PPARα full agonist with fairly
good potency and concomitant moderate efficacy on PPARγ
(about 3-fold less effective than rosiglitazone). Actually, this
stereoisomer displayed a singular behavior; it was almost
inactive on PPARγ and about 10 times less potent than the
corresponding S isomer on PPARα. However, it behaved as a
full agonist toward PPARα, showing equipotency with lead
compound 1 and higher potency than reference compound
Wy-14,643, allowing the hypothesis that the presence of a
substituent with appropriate stereoelectronic properties placed
ortho to the biphenyl bond of the ligand is able to
counterbalance the lack of a correct spatial arrangement of
the groups bound to the stereogenic center.
With the aim to provide a molecular explanation for the
different behavior of (R)- and (S)-16 compared to 1, X-ray
crystal studies in PPARγ-LBD have been performed for both
enantiomers.
Binding of (S)-16 in the PPARγ-LBD. The final simulated
annealing omit map (Figure 1 of the Supporting Information)
clearly shows two molecules of (S)-16 simultaneously bound to
PPARγ-LBD. One molecule (mol A) is accommodated into the
biphenyl pocket (site A), as its analogue 1 lacking the isopropyl
group,22 and interacts with the triad H323, H449, and Y473.
The second molecule (mol B) binds in the typical position of
the pure partial agonists of PPARγ,22,27 facing the β-sheet and
interacting with it through H-bonds (site B). Moreover, the two
molecules interact with each other, making short van der Waals
(vdW) contacts through the isopropyl group of mol B and the
benzylic aromatic ring of mol A (Figure 1). The presence of
Figure 1. (S)-16 in the complex with PPARγ: mol A (cyan) positioned
into the biphenyl pocket and mol B (yellow) in front of the β-sheet.
two molecules of the ligand is also confirmed by indirect
evidence; in fact, the side chains of R288 and E291 are
dramatically rearranged for the presence of the ligand in site B,
and the side chain of F282 is shifted from the t to the folded g*
conformation for the presence of a second molecule in site A.
However, the weak electron density on the side chain of F282
denotes a probable disorder of this residue on site A. This could
be attributed to the steric hindrance of the bulky isopropyl-
substituted biphenyl moiety into the narrow biphenyl pocket.
This is also confirmed by the weak electron density on the
At this point, with the aim to confirm the importance of
stereochemistry in PPAR activation, we decided to prepare the
R configurational isomer of the isopropyl derivative 16. The
preference went to this compound because it apparently
63
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
isopropyl group that interacts with very short vdW contacts
with the side chain of F282.
Calorimetric Measurements. With the aim to evaluate the
binding characteristics of (S)-16 and those of the correspond-
ing enantiomer (R)-16, we decided to perform isothermal
titration calorimetry (ITC) experiments. This analysis directly
gives the standard enthalpy change ΔH associated with the
formation of the receptor−ligand complex, the estimation of
ΔG, and the host−guest stoichiometry (n) by titration curve
fitting. The entropy change ΔS can also be calculated from the
Gibbs−Helmholtz equation. The ITC measurements were
carried out at 298 K in 20 mM Hepes buffer at pH 8. Figure 4a
shows the calorimetric data (raw and integration data) obtained
in the titration of PPARγ with (S)-16. The same experiment
was also conducted for 1 and (R)-16 (Figures 3 and 4 of the
Supporting Information, respectively). After subtraction of the
heat of dilution from the raw data and linear curve fitting under
the “one set of sites model”, the thermodynamic parameters
shown in Table 2 were obtained. For (S)-16 the inflection
point in the calorimetric isotherm occurs near the molar ratio of
1.0, which corresponds to a 1:1 binding mode (Ka = 3.60 × 105
M−1). To check if the ligand had multiple binding sites, as
found in the X-ray experiments, a reverse titration was
performed in which a 400 μM solution of protein was injected
into the cell containing a diluted solution of the ligand (40
μM). For 1:1 bimolecular reactions, the measured thermody-
namic parameters are expected to remain unchanged when the
experiment is conducted in reverse order. On the contrary, the
reverse titration (shown in Figure 4b) has the inflection point
near the molar ratio of 0.5 (i.e., two molecules of ligand for one
of protein), indicating that when the ligand is in large excess
(10:1 ratio at the beginning of the experiment), the prevailing
binding mode is 2:1. To establish the affinity of (S)-16 toward
each site, a third experiment was performed in which the
protein was pre-equilibrated in a 1:1 ratio with lead compound
1, which occupies only site A, as observed in the crystal
structure of the complex PPARγ/1.22 After pre-equilibration,
the protein (47 μM) was titrated with (S)-16 (470 μM)
(Figure 4c). Given the higher affinity for site A of 1 with
respect to (S)-16, (Ka = 4.59 × 105 M−1 vs 3.60 × 105 M−1,
respectively), (S)-16 should not be able to displace 1 from site
A, unless it is in large excess. Thus, in this case, binding could
be mainly referred to the second binding site (site B). As
expected, the titration gave a lower affinity constant (Ka = 1.87
× 105 M−1) and a stoichiometry of 0.7.
On site B, the carboxylate oxygens of mol B form H-bonds
with the NH of S342 and CO of L340 on the β-sheet, the last
one mediated by a water molecule. Moreover, several vdW
contacts are formed by mol B with the side chains of F264,
C285, M364, and L330 (Figure 2).
Figure 2. H-bond network and vdW interactions of mol B of (S)-16
(yellow) in site B of PPARγ-LBD.
Binding of (R)-16 in the PPARγ-LBD. In this case the
occupation of the only site B is observed. Unlike mol B of its S
enantiomer, (R)-16 is shifted more toward the entrance of the
LBD (Figure 2 of the Supporting Information). Site A is empty
as confirmed also by the clear density on the F282 side chain in
the trans conformation to close the entrance to the biphenyl
pocket. One water molecule coordinates the triad H323, H449,
and Y473. Also in this case R288 and E291 side chains are
dramatically rearranged to allow the occupation of site B by the
ligand. The carboxylate group forms H-bonds with the NH of
S342, and the benzylic aromatic ring of the ligand is
sandwiched between those of F287 and H266, giving rise to
strong vdW interactions (Figure 3).
The analysis of these ITC experiments shows that, at ligand/
protein ratios not higher than 2:1 (final conditions of the
experiment shown in Figure 4a), (S)-16 binds to the higher
affinity binding site (Ka = 3.60 × 105 M−1); a similar affinity was
measured for the binding of lead compound 1 in the biphenyl
pocket (site A, Ka = 4.59 × 105 M−1) (Figure 3 of the
Supporting Information). The similarity between these Ka
values seems to confirm that site A is the one with the higher
affinity. When the protein is pre-equilibrated with 1 in a 1:1
ratio, site A is already occupied and the titration with (S)-16
leads new molecules to preferentially bind to the available site
B, resulting in a protein with both sites occupied, although site
B seems to be engaged only at 70% (n = 0.70). In this case the
titration affords the Ka (1.87 × 105 M−1) and the other
thermodynamic parameters for site B (Table 2). This Ka value
is close to that obtained by titrating (R)-16 (1.34 × 105 M−1;
see below), corresponding to the occupation of the only site B.
This similarity indirectly confirms that this Ka value is an
effective measure of (S)-16 for site B.
Figure 3. H-bond network and vdW interactions of (R)-16 in the LBD
of PPARγ.
64
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
Figure 4. (a) Titration of PPARγ (40 μM) with (S)-16 (400 μM). (b) Reverse titration of (S)-16 (40 μM) with PPARγ (400 μM). (c) Titration
with (S)-16 (470 μM) of PPARγ (47 μM) after pre-equilibration with 1 (ratio 1:1).
As noted in Table 2, the enthalpy change in the reverse
titration (site A + site B, ΔH = −11.26 kcal/mol)
approximately corresponds to the sum of ΔH values measured
for the single occupation of site A and site B (−6.29 and −4.15
kcal/mol, respectively) plus an additional amount probably due
to the direct interaction of the two molecules in the LBD. The
Ka value measured in this experiment (3.10 × 105 M−1) is
within those calculated for each site (3.60 × 105 and 1.87 × 105
M−1). Despite the favorable enthalpic contribution due to the
binding of two molecules (ΔH = −11.26 kcal/mol), a very
unfavorable entropic term (−TΔS = 3.79 kcal/mol) leads the
net free energy to a value slightly lower than that of the binding
on site A (−7.47 vs−7.58 kcal/mol, respectively). This could
suggest a conformational rearrangement of the protein upon
the binding of two molecules, which may act as an energy
barrier.
The ITC titration of the ligand (R)-16 (600 μM) into the
cell containing PPARγ (60 μM) is shown in Figure 4 of the
Supporting Information. The binding mode of the ligand is 1:1
(n = 0.99), and the Ka corresponding to the occupation of site
B is 1.34 × 105 M−1. The reverse titration performed for (R)-16
(Figure 5 of the Supporting Information) shows a stoichiom-
etry n value of 0.44, indicating also for this ligand a 2:1 binding
mode with both sites of similar affinity. The Ka value of 1.30 ×
105 M−1 obtained from this experiment is almost identical to
that of site B (Ka = 1.34 × 105 M−1), suggesting the existence of
a second low-affinity site (site C), which, however, was not
observed in the crystal complex of PPARγ/(R)-16.
Only a few examples of crystalline structures reporting the
simultaneous binding of two molecules of a ligand to PPARγ-
LBD are known in the literature.28−30 Among these, the
antidiabetic agent 25 (T2384) can adopt two distinct binding
65
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
orientations different from the one seen in the crystal structure
at site B (data not shown), indicating the outward movement of
the ligand as a consequence of negative cooperativity between
two sites. The docking scores of (S)-16 were higher for site A
(47.2 kJ/mol) than site B (30.9 kJ/mol), implying that this
molecule has extremely low binding affinity toward site B.
Together, these results confirm that there is a preference of site
A to be the primary site for (S)-16 binding, corroborating the
hypothesis that, as observed in ITC experiments, PPARγ can
host two molecules of (S)-16 only in presence of a high
concentration of this ligand.
Table 2. Thermodynamic Parameters Relating to the
Formation of the Complexes PPARγ-LBD/Ligand
Determined by the ITC Assay
ΔG
ΔH
−TΔS
Ka × 105
(kcal
(kcal
(kcal
a
(M−1
)
n
M−1
)
M−1
)
M−1
)
1
4.59
3.60
3.10
0.91
1.09
0.54
−7.72
−7.58
−7.47
−8.61
−6.29
−11.26
0.89
−1.29
3.79
(S)-16, site A
(S)-16, reverse
titration (A + B)
(S)-16, pre-
equilibrated with 1
(B)
1.87
0.70
−7.19
−4.15
−3.04
To help interpretation of structure−activity relationship
(SAR) data and to gain a better understanding of how the
described (aryloxy)phenylpropanoic acids might bind to
PPARα, we performed docking experiments of the most active
12, 13, 15, and (S)-16 into the ligand binding site of the
previously reported cocrystal structure of PPARα bound to the
α/γ dual agonist 28 (BMS-631707) (PDB code 2REW).34
When 12, 15, and (S)-16 were docked within the PPARα
binding site, about 80% of the conformations generated by
GOLD adopted only one highly conserved orientation. Figure
5 depicts the binding mode of the representative compound
(S)-16 into the PPARα binding site. Interestingly, the ligands
assume a position very similar to that of the cocrystallized 28
and are stabilized by a combination of H-bonds and
hydrophobic interactions. The carboxylate group forms the
well-recognized H-bonding network with S280 (H3), Y314
(H5), H440 (H11), and Y464 (H12), which is believed to be
critical for the functional activity of PPAR ligands. The
remaining interactions between ligands and protein are
hydrophobic in nature. In particular, the substituted biphenyl
moiety lodges into a hydrophobic pocket (pocket I),
corresponding to the “biphenyl pocket” in PPARγ, and interacts
with F351 and I354 of H7, V444 and I447 of H11, A454 and
L456 of the loop 11/12, and F273, C276, and Q277 of H3.
Notably, F273 forms a face−edge−face aromatic-stacking
interaction with the first aromatic ring of the substituted
biphenyl moiety of the ligands. The isopropyl group of (S)-16
establishes hydrophobic interactions with F273, F351, I354,
and I447. These results are consistent with the SAR data
showing that compounds 12 and 15, which bear a lipophilic
group at the 2′ position of the biphenyl moiety (ethyl and
trifluoromethyl, respectively), are highly potent PPARα
(R)-16, site B
1.34
1.30
0.99
0.44
−6.99
−6.97
−2.37
−6.23
−4.62
−0.74
(R)-16, reverse
titration (B + C)
a
n = molar binding ratio of the ligand−protein interaction (observed
stoichiometry).
modes in the LBD of PPARγ, each of which is able to induce
distinct patterns of coregulatory protein interaction with
PPARγ in vitro displaying unique receptor functions in cell-
based activity assays.28
In Figures 6−9 of the Supporting Information the super-
positions of (S)-16 with 25, 26 (HODE-9),29 and 27
(AF3369)30 all simultaneously bound with two molecules to
the LBD of PPARγ are shown. Although the superpositions are
not identical, it can be noted that in all cases one molecule
occupies the region in front of the β-sheet whereas the second
molecule lies in a region closer to helix 12.
Molecular Modeling. To test whether the ligand binding
pocket of PPARγ can host two molecules of (S)-16
simultaneously, we carried out docking calculations on the
above-reported two-ligand-bound PPARγ crystal structure.
With this purpose, we examined the relative positions of two
docked (S)-16 molecules with respect to the bound
conformations of (S)-16 in the X-ray crystal structure at site
A and site B. The two (S)-16 molecules were sequentially
docked through the automated docking software GOLD
5.0.1,31,32 using the GoldScore-CS docking protocol33 (see
the Experimental Section for details). The first docked (S)-16
conformation was similar to the one seen in site A of the crystal
structure (rmsd = 0.7 Å) with the same contact residues. The
second (S)-16 molecule was then docked and assumed
Figure 5. Compounds (S)-16 (a, yellow) and 13 (b, green) docked into the PPARα binding site represented as a blue ribbon model. Only amino
acids located within 4 Å of the bound ligand are displayed (white) and labeled. H-bonds discussed in the text are depicted as dashed black lines.
66
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
agonists. In addition, the benzyl ring of the ligands binds in a
second hydrophobic pocket (pocket II), making lipophilic
interactions with residues C276, T279 of H3, I317, F318, and
L321 of H4, M330 of β-sheet, and M355 of H7. It is known
that hydrophobic sulfur and especially sulfur−arene interactions
are strongly attractive.35,36 Figure 5 shows that the environment
of the benzyl moiety is very rich in S-containing protein side
chains (C276, M330, M355), so hydrophobic intermolecular
contacts with the ligands are formed, further contributing to the
complex stability. Moreover, the stabilization of H11, H3, the
loop 11/12, and H12 upon ligand binding could explain the
nanomolar activity of 12, 15, and (S)-16 and their character of
full agonists toward PPARα. Curiously, when compound 13
was docked into the PPARα binding cleft, only 10% of the
generated conformations adopted the above-described binding
mode, whereas 90% were in a different orientation that
positioned both benzyl and substituted biphenyl moieties in a
reverse mode within the two above-mentioned hydrophobic
pockets. It is likely that the much larger phenyl ring at the 2′
position of the biphenyl moiety does not find sufficient room to
be accommodated inside the hydrophobic pocket I, forcing the
ligand to adopt a different pose into PPARα. The carboxylate
group still engages H-bonds with S280, Y314, and Y464. In
addition, an additional H-bond between the phenoxylic oxygen
atom of 13 and the H440 NεH hydrogen is formed. The benzyl
substituent occupies, but does not completely fill, the
hydrophobic pocket I formed by helices 3, 7, and 11 and the
loop 11/12. The remainder of the ligand wraps around helix 3
and buries the phenyl-substituted biphenyl moiety into the
lipophilic pocket II formed by helices 3, 4, and 7 and the β
sheet.
Figure 6. (A) Total RNA extracted from HepG2 cells, treated with or
without 1 μM 29, 50 μM Wy 14,643, or 1 μM (S)-16 for 24 h, was
used to quantify CAC mRNA by real-time PCR. (B) CAC and β-actin
of HepG2 cells, treated for 48 h as described for (A), were
immunodecorated with specific antibodies. In (A) means
SD of
three duplicate independent experiments are shown. In (A),
differences between the samples and control (without any addition)
were significant (*, P < 0.05; **, P < 0.01; ***, P < 0.001; one-way
ANOVA, Bonferroni test).
Effects of (S)-16 on the Carnitine Shuttle System.
Oxidation of fatty acids in mitochondria coupled to oxidative
phosphorylation is the most important pathway for the
production of metabolic energy during fasting. This process
occurs in the mitochondrial matrix where the enzymes of fatty
acid β-oxidation are located. Fatty acyl groups are transported
from the cytosol into the mitochondrial matrix by means of the
carnitine shuttle system. In the cytosol, fatty acid units are
transferred from acyl-CoAs to carnitine by the action of CPT1,
which is located on the external surface of the outer
mitochondrial membrane;37,38 the formed acylcarnitines cross
the outer membrane, which is permeable to small molecules,39
and are translocated through the inner mitochondrial
membrane by the CAC. CAC is an integral protein that
transports acylcarnitine esters (in exchange for free carnitine)
into mitochondria where the acyl groups are oxidized.
Therefore, the molecular components of the carnitine shuttle
system are essential for the mitochondrial oxidation of fatty
acids and, hence, for life.40
It has been previously demonstrated that the two fibrates Wy
14,643 and 29 (GW7647) increase the CAC transcript level in
primary and secondary cell lines, providing, therefore, a novel
contribution to the understanding of their hypolipidemic
action.23 On this basis, we decided to test (S)-16 to evaluate
the possibility that this fibratelike drug was able, in the same
way, to upregulate mitochondrial CAC as well as CPT1 gene
expression. The effects of in vitro application of this compound
were investigated on HepG2 cells which were incubated for 24
h with a 1 μM concentration of (S)-16. Wy 14,643 (50 μM)
and 29 (1 μM) were used as reference compounds. After
incubation, total mRNA was extracted and used to determine
CAC and CPT1 transcript levels. As shown in Figures 6A and
7A, (S)-16 caused a CAC and CPT1 mRNA increase of about
1.5- and 2.3-fold in HepG2 cells, respectively. Consistently,
Western blotting analysis confirmed markedly increased CAC
and CPT1 protein levels (Figures 6B and 7B) in cells treated
with (S)-16 as compared to untreated cells. The observed
effects of (S)-16 on both CAC and CPT1 gene expression
levels turned out to be lower than those obtained with 29 and
comparable to those of Wy 14,643. However, the concentration
of Wy 14,643 needed to activate both genes is 50-fold higher
(50 μM) as compared to that of (S)-16 (1 μM). These results
clearly indicate that (S)-16 is a potent upregulator of gene
expression of CAC and CPT1, two key proteins of the
mitochondrial fatty acid oxidation pathway. Moreover, even
though less potent than the selective PPARα ligand 29, (S)-16
shows a dual full PPARα and partial PPARγ agonist activity that
is currently considered as a very attractive option in the
treatment of dyslipidemic type 2 diabetes.
CONCLUSION
■
In the present study, we have investigated the effects resulting
from the introduction of different substituents into the biphenyl
system of compound 1, previously reported as a PPARα/γ dual
agonist. This modification allowed the identification of new
ligands with higher potency on PPARα and fine adjusted
moderate activity on PPARγ. For the most promising
stereoisomer (S)-16, X-ray and calorimetric studies in PPARγ
and docking experiments in both receptor subtypes provided a
molecular explanation for its different behavior as a full and
partial agonist of PPARα and PPARγ, respectively. Interest-
ingly, it has been observed that PPARγ can host two molecules
of (S)-16 in the presence of a high concentration of this ligand,
67
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
EXPERIMENTAL SECTION
■
Chemical Methods. Column chromatography was performed on
ICN Silica Gel 60 Å (63−200 μm) as a stationary phase. Melting
points were determined in open capillaries on a Gallenkamp
electrothermal apparatus and are uncorrected. Mass spectra were
recorded on an HP GC/MS 6890-5973 MSD spectrometer, electron
impact 70 eV, equipped with an HP chemstation or an Agilent LC/MS
1100 series LC/MSD Trap System VL spectrometer, electrospray
ionization (ESI). 1H NMR spectra were recorded in CDCl3 (the use of
other solvents is specified) on a Varian-Mercury 300 (300 MHz)
spectrometer at room temperature. Chemical shifts are expressed as
parts per million (δ). The purity of all tested compounds was >95%, as
confirmed by combustion analysis carried out with a Eurovector Euro
EA 3000 model analyzer.
Optical rotations were measured with a Perkin-Elmer 341
polarimeter at room temperature (20 °C): concentrations are
expressed as grams per 100 mL. The enantiomeric excesses of the
final acids were determined by HPLC analysis of acids or their methyl
esters, obtained by reaction with an ethereal solution of diazomethane,
on a Chiralcel OD or AD column (4.6 mm i.d. × 250 mm, Daicel
Chemical Industries, Ltd., Tokyo, Japan). Analytical liquid chromatog-
raphy was performed on a PE chromatograph equipped with a
Rheodyne 7725i model injector, a 785A model UV/vis detector, a
series 200 model pump, and an NCI 900 model interface. Chemicals
were from Aldrich (Milan, Italy), AlfaAesar (Karlsruhe, Germany), or
Acros (Milan, Italy) and were used without any further purification.
Preparation of Ethyl 2-(Aryloxy)-3-phenylpropanoates 20a−
g: General Procedure. To an ice-bath-cooled suspension of PPh3−
DVB22 (2% DVB, 3 mmol g−1, 13.8 mmol) in anhydrous toluene (35
mL) was added dropwise a solution of diisopropyl azodicarboxylate
(DIAD; 13.5 mmol) in anhydrous toluene (35 mL). The resulting
mixture was stirred for 0.5 h. A solution of the suitable R or S
enantiomer or racemic ethyl phenyllactate (11 mmol) in anhydrous
toluene (50 mL) and suitable 4-bromophenol (10 mmol) was added at
rt. The reaction mixture was stirred overnight at rt under a N2
atmosphere. The solid was filtered off, and the filtrate was evaporated
to dryness. The residue was chromatographed on a silica gel column
(petroleum ether/ethyl acetate, 9:1 or 8:2, as eluent), affording the
desired compounds as pale yellow oils in 54−82% yields.
Preparation of Ethyl 2-(Biaryl-4-yloxy)-3-phenylpropa-
noates 21a−r: General Procedure. The suitable arylboronic acid
(3.64 mmol) and Cs2CO3 (2.73 mmol) were added, under a N2
atmosphere, to a stirred solution of the corresponding compounds
20a−g synthesized as above (1.82 mmol) in anhydrous toluene (35
mL); after 0.5 h, Pd(PPh3)4 (0.11 mmol) or Pd(PPh3)4−DVB (2%
DVB, 0.5 mmol g−1, 0.055 mmol) was added. The reaction mixture
was stirred overnight at 95 °C and then quenched with 1 N HCl (7.5
mL) and ethyl acetate (7.5 mL). The suspension was filtered through a
Celite pad to remove the catalyst, and the filtrate was washed with
NaHCO3 saturated solution and brine. After the filtrate was dried over
Na2SO4, the solvent was evaporated to dryness to give a dark oily
residue which was chromatographed on a silica gel column (petroleum
ether/ethyl acetate, 98:2 or 95:5, as eluent), affording the desired
compounds as colorless or yellow oils in 25−95% yields.
Figure 7. (A) Total RNA extracted from HepG2 cells, incubated with
or without 1 μM 29, 50 μM Wy 14,643, or 1 μM (S)-16 for 24 h, was
used to quantify CPT1 mRNA by real-time PCR. (B) CPT1 and β-
actin of HepG2 cells, treated for 48 h as described for (A), were
immunodecorated with specific antibodies. In (A) means
SD of
three duplicate independent experiments are shown. In (A),
differences between the samples and control (without any addition)
were significant (**, P < 0.01; ***, P < 0.001; one-way ANOVA,
Bonferroni test).
and its relative affinities for the two sites were determined.
Calorimetric data and docking simulations show that at lower
ligand concentration the occupation of the only site A is
preferred. In addition, the adaptability of PPARγ-LBD has been
demonstrated through the rearrangement of several side chains
of the internal hydrophobic pocket (F282, R288, E291, etc.),
improving the interaction with the ligands. All these features
ensure that different ligands, depending on the concentration
and the occupied LBD regions, can recruit different classes of
coregulators eliciting distinct biological responses. For this
reason, even small modifications of the chemical structure of
known PPARγ ligands could be able to fine-tune the agonist
activity, allowing identification, in a series of strictly related
compounds, of the optimal PPARα/PPARγ selectivity ratio.
Moreover, the effects of (S)-16 on the transcription of the
mitochondrial CAC and CPT1 genes, two key components of
the carnitine shuttle system that is essential for the
mitochondrial oxidation of fatty acids, allow the hypothesis of
a more beneficial pharmacological profile of this compound in
comparison with the less potent PPARα agonist fibrates
currently used in therapy. Further studies are in progress to
identify the optimal PPARα/γ selectivity ratio in this set of
compounds by cofactor recruitment FRET assay for each
ligand. In the next step, in vivo experiments will be carried out
with the ligands able to recruit cofactors similar to those
recruited by selective PPARγ modulators (SPPARγMs) or
PPARγ partial agonists currently under clinical investigation to
gain insight into the properties and biological action of these
molecules, which may represent the potential leads of a new
class of drugs with better therapeutic effects in the treatment of
dyslipidemic type 2 diabetes.
Synthesis of Ethyl 2-[(2′-Formylbiphenyl-4-yl)oxy]-3-phe-
nylpropanoate (22). In a suitable vessel, a mixture of 20c (2.87
mmol), 2-formylbenzeneboronic acid (4.29 mmol), and Cs2CO3 (5.73
mmol) was dissolved in 45 mL of THF. After 20 min Pd(PPh3)4 (0.17
mmol) was added, and the resulting suspension was vigorously stirred
in a microwave reactor (1 h, 100 °C, 150 W). The resulting crude was
filtered through a Celite pad to remove the catalyst, and the filtrate was
washed with NaHCO3 saturated solution and brine. After the filtrate
was dried over Na2SO4, the solvent was evaporated to dryness to give a
dark oily residue which was chromatographed on a silica gel column
(n-hexane/ethyl acetate, 98:2, as eluent), affording the desired
compound as a yellow oil.
Synthesis of Ethyl 2-[(2′-Substituted biphenyl-4-yl)oxy]-3-
phenylpropanoates 23a−c: General Procedure. A solution of 22
(2.5 mmol) in anhydrous CH3CN (20 mL) was carefully added, under
a N2 atmosphere and during 0.5 h, to a stirred solution of the
68
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
appropriate alkyl- or (arylalkyl)triphenylphosphonium bromide (5
mmol) in anhydrous CH3CN (15 mL). After 0.5 h a solution of DBU
(5 mmol) in dry CH3CN (15 mL) was added to the mixture. The
reaction mixture was stirred and heated under reflux for 24 h, the
solvent was removed under reduced pressure, and the residue was
dissolved in ethyl acetate. The organic phase was washed with NH4Cl
saturated solution and twice with brine, then dried over Na2SO4, and
filtered. The solvent was evaporated to dryness, affording crude oils
which were chromatographed on a silica gel column (petroleum ether/
ethyl acetate, 95:5, as eluent) to obtain colorless oils in 17−54% yields.
Synthesis of Ethyl 2-[(2′-Substituted biphenyl-4-yl)oxy]-3-
phenylpropanoates 24a−c: General Procedure. The oily
mixtures of cis and trans isomers 23a−c (1 mmol) were dissolved in
absolute EtOH (90 mL) and stirred at rt under a H2 atmosphere (10−
15 atm) in the presence of the Wilkinson catalyst or 10% Pd/C (0.05
mmol). After 24−48 h the suspension was filtered through a Celite
pad to remove the catalyst, and the filtrate was concentrated under
reduced pressure to give dark solids which were chromatographed on a
silica gel column (Et2O/CH2Cl2, 90:10 or 80:20, or petroleum ether/
ethyl acetate, 80:20, as eluent), affording the title compounds as
colorless oils in 84−93% yields.
Preparation of the Final Acids 2−19: General Procedure. To
a stirred solution of the corresponding ethyl ester (1 mmol) in THF
(10 mL) was added 1 or 2 N NaOH (10 mL). The reaction mixture
was stirred for 4−10 h. THF was evaporated in vacuo, and the aqueous
phase was acidified with 2 N HCl and extracted with Et2O. The
combined organic layers were washed twice with brine, dried over
Na2SO4, and evaporated to dryness, affording the title acids in
quantitative yields as white solids or colorless oils. The oily acids were
dissolved in 96% EtOH (10 mL) and added to a solution of NaHCO3
(1 equiv) in H2O (1 mL). The reaction mixture was stirred for 2 h,
and then the solvent was evaporated to dryness, affording the
corresponding sodium salts in quantitative yields as white solids which
were washed or crystallized from the suitable solvents.
Western Blot Analysis. HepG2 cells were incubated for 48 h with
Wy 14,643 (50 μM), 29 (1 μM), and (S)-16 (1 μM), starting 24 h
after having been depleted of serum. Proteins were electroblotted onto
nitrocellulose membranes (Bio-Rad) and subsequently treated with
anti-CPT1 (ARP44796_P050, Aviva Systems Biology), anti-CAC,23
and anti-β-actin (sc-58619, Santa Cruz) antibodies. The immuno-
reaction was detected by the ECL plus system (Amersham).
Protein Expression, Purification, and Crystallization. The
LBD of PPARγ was expressed as an N-terminal His-tagged protein
using a peT28 vector and purified onto a Ni2+−nitriloacetic acid
column (GE Healthcare) as previously described.45 Crystals of apo-
PPARγ were obtained by the vapor diffusion method at 18 °C using a
sitting drop made by mixing 2 μL of protein solution (12 mg/mL, in
20 mM Tris and 1 mM TCEP, pH 8.0) with 2 μL of reservoir solution
(0.8 M sodium citrate and 0.15 M Tris, pH 8.0). The crystals were
soaked for one week in a storage solution (1.2 M sodium citrate and
0.15 M Tris, pH 8.0) containing the ligand (0.5 mM). The ligand
dissolved in DMSO was diluted in the storage solution so that the final
concentration of DMSO was 0.5%. The storage solution with glycerol
[20% (v/v)] was used as a cryoprotectant. Crystals belong to the space
group C121 with cell parameters shown in Table 1 of the Supporting
Information.
Structure Determination. X-ray data were collected at 100 K
under a nitrogen stream using synchrotron radiation (beamline ID 23-
2 at ESRF, Grenoble, France). The diffracted intensities were
processed using the programs MOSFLM and SCALA.46 Structure
solution was performed with AMoRe,47 using the coordinates of
PPARγ/1 (PDB code 3B3K) as the starting model. The coordinates
were then refined with CNS.48 All data between 8 and 2.5 Å were
included. The statistics of crystallographic data and refinement are
summarized in Table 1 of the Supporting Information. The
coordinates of PPARγ/(S)-16 and PPARγ/(R)-16 have been
deposited in the Brookhaven Protein Data Bank (PDB) with the
codes 4E4K and 4E4Q, respectively.
Biological Methods. Reference compounds, the medium, and
other cell culture reagents were purchased from Sigma-Aldrich (Milan,
Italy).
Plasmids. The expression vectors expressing the chimeric receptor
containing the yeast Gal4-DNA binding domain fused to the human
PPARα- or PPARγ-LBD and the reporter plasmid for these Gal4
chimeric receptors (pGal5TKpGL3) containing five repeats of the
Gal4 response elements upstream of a minimal thymidine kinase
promoter that is adjacent to the luciferase gene were described
previously.41
Isothermal Titration Calorimetry. ITC experiments were
performed at 25 °C using a MicroCal ITC200 microcalorimeter
(MicroCal Inc., Northampton, MA). PPARγ was extensively dialyzed
against the buffer Hepes (20 mM, pH 8.0), TCEP (1 mM), with
Amicon Ultra filters, and the final exchange buffer was then used to
dilute the ligand stock solutions (50 mM in DMSO). DMSO was
added to the protein solution at the same percentage of the ligand
solution (below 1%). Samples were centrifuged before the experiments
to eliminate possible aggregates. Protein and ligand solutions were
degassed before use. The protein solution (40 μM) was placed in the
sample cell, and the ligand solution (400 μM) was loaded into the
syringe injector. In the reverse titration the protein solution (400 μM)
was injected into the cell containing the ligand (40 μM). The titrations
involved 19 injections of 2 μL at 180 s intervals. The syringe stirring
speed was set at 1000 rpm. Reference titrations of ligands into buffer
were used to correct for heats of dilution. The thermodynamic data
were processed with Origin 7.0 software provided by MicroCal. The
ΔH values were measured for each titration, and fitting the binding
isotherms with a one-site binding model yielded the values of the
association constant (Ka). From the Gibbs−Helmholtz equation is also
calculated the change of entropy (ΔS). The inflection point in the
calorimetric isotherm gives the stoichiometry value n, indicating the
ligand/protein ratio of the binding. To correct for any discrepancies in
the baseline outlined by the software, a manual adjustment was
performed. In one experiment the protein was preliminarily
equilibrated with the ligand for 3 h.
Cell Culture and Transfections. Human hepatoblastoma cell line
HepG2 (Interlab Cell Line Collection, Genoa, Italy) was cultured in
minimum essential medium (MEM) containing 10% heat-inactivated
fetal bovine serum, 100 U of penicillin G mL−1, and 100 μg of
streptomycin sulfate mL−1 at 37 °C in a humidified atmosphere of 5%
CO2. For transactivation assays, 105 cells per well were seeded in a 24-
well plate and transfections were performed after 24 h with CAPHOS,
a calcium phosphate method, according to the manufacturer’s
guidelines. Cells were transfected with expression plasmids encoding
the fusion protein Gal4−PPARα-LBD or Gal4−PPARγ-LBD (30 ng),
pGal5TKpGL3 (100 ng), and pCMVβgal (250 ng). Four hours after
transfection, cells were treated for 20 h with the indicated ligands in
triplicate. Luciferase activity in cell extracts was then determined by a
luminometer (VICTOR3 V Multilabel Plate Reader, PerkinElmer). β-
Galactosidase activity was determined using ortho-nitro-phenyl-β-D-
galactopyranoside as described previously.42 All transfection experi-
ments were repeated at least twice.
Real-Time PCR. HepG2 cells were incubated for 24 h with Wy
14,643 (50 μM), 29 (1 μM), and (S)-16 (1 μM), starting 24 h after
having been depleted of serum. Total RNA was extracted from 1 × 106
cells, and reverse transcription was performed as reported.43 Real-time
PCR was carried out as described previously.44 Assays-on-demand for
human CPT1, CAC, and human actin (catalog nos. Hs00912671_m1,
Hs01088810_g1, and Hs00357333_g1, respectively) were purchased
from Applied Biosystems. All transcript levels were normalized against
the β-actin expression levels.
Computational Chemistry. Molecular modeling and graphics
manipulations were performed using the molecular operating
environment (MOE)49 and UCSF-Chimera software packages,50
running on an E4 Computer Engineering E1080 workstation provided
with an Intel Core i7-930 Quad-Core processor. GOLD 5.0.131 was
used for all docking calculations. Figures were generated using Pymol
1.0.51
Ligand and Protein Setup. The core structures of compounds
12, 13, 15, and (S)-16 were constructed using standard bond lengths
and bond angles of the MOE fragment library. The carboxylate group
69
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
was taken as dissociated. Geometry optimizations were accomplished
with the MMFF94X force field, available within MOE. The
coordinates of PPARγ/(S)-16 (PDB code 4E4K) and PPARα/28
(PDB code 2REW),34 recovered from the Brookhaven Protein
Database,52 were used in the docking experiments. Bound ligands
and water molecules were removed. A correct atom assignment for
Asn, Gln, and His residues was done, and hydrogen atoms were added
using standard MOE geometries. Partial atomic charges were
computed by MOE using the AMBER99 force field. All heavy atoms
were then fixed, and hydrogen atoms were minimized using the
AMBER99 force field and a constant dielectric of 1, terminating at a
gradient of 0.001 kcal mol−1 Å−1.
ACKNOWLEDGMENTS
■
This work was accomplished thanks to the financial support of
the Universita
2010).
̀
degli Studi di Bari “Aldo Moro” (Research Fund
ABBREVIATIONS USED
■
PPAR, peroxisome proliferator-activated receptor; LBD, ligand
binding domain; TZDs, thiazolidinediones; CAC, carnitine/
acylcarnitine carrier; CPT1, carnitine-palmitoyl-transferase 1;
ITC, isothermal titration calorimetry
Docking Simulations. Docking of (S)-16 to PPARγ and 12, 13,
15, and (S)-16 to PPARα was performed with the GOLD program,
which uses a genetic algorithm for determining the docking modes of
ligands and proteins. In the present study, GOLD was allowed to
calculate interaction energies within a sphere of a 7 Å radius centered
around the two bound (S)-16 molecules for site A and site B of
PPARγ and a 13 Å sphere radius centered on the cocrystallized ligand
28 of PPARα. The Goldscore-CS docking protocol33 was adopted in
this study. In this protocol, the poses obtained with the original
GoldScore function are rescored and reranked with the GOLD
implementation of the ChemScore function.33,53 To perform a
thorough and unbiased search of the conformation space, each
docking run was allowed to produce 200 poses without the option of
early termination, using standard default settings. The docking
procedure employed for initial exploration of multiple ligand binding
to PPARγ was analogous to one that was explored by a number of
groups whose aim was to probe the structural basis of heterotropic/
homotropic activation or inhibition.54,55 In this approach, two ligands
were docked in a sequential fashion. The first ligand was docked using
a pre-evaluated interaction grid based on interactions with atoms in
the protein alone. The second ligand was then docked employing an
interaction grid including interactions with the protein and bound
configurations of the first ligand. In this way, the energy landscape for
binding of the second ligand includes interactions with the first ligand.
It is plausible that the binding of multiple ligands to PPARγ involves
such a sequential process. Cobound configurations of two (S)-16
molecules were searched by first determining the unique sets of
clustered configurations of one of the two ligands, using an rmsd
criterion of 1.5 Å, followed by docking the second ligand to a
representative structure for clusters of the first docked ligand.
REFERENCES
■
(1) Kliewer, S. A.; Sundseth, S. S.; Jones, S. A.; Brown, P. J.; Wisely,
G. B.; Koble, C. S.; Devchand, P.; Wahli, W.; Willson, T. M.; Lenhard,
J. M.; Lehmann, J. M. Fatty acids and eicosanoids regulate gene
expression through direct interactions with peroxisome proliferator-
activated receptors α and γ. Proc. Natl. Acad. Sci. U.S.A. 1997, 94,
4318−4323.
(2) Berger, J. P.; Akiyama, T. E.; Meinke, P. T. PPARs: therapeutic
targets for metabolic disease. Trends Pharmacol. Sci. 2005, 26, 244−
251.
(3) Nagy, L.; Schwabe, J. W. Mechanism of the nuclear receptor
molecular switch. Trends Biochem. Sci. 2004, 29, 317−324.
(4) Nettles, K. W.; Greene, G. L. Ligand control of coregulator
recruitment to nuclear receptors. Annu. Rev. Physiol. 2005, 67, 309−
333.
(5) Renaud, J. P.; Moras, D. Structural studies on nuclear receptors.
Cell. Mol. Life Sci. 2000, 57, 1748−1769.
(6) Desvergne, B.; Wahli, W. Peroxisome proliferator-activated
receptors: nuclear control of metabolism. Endocr. Rev. 1999, 20,
649−688.
(7) Elisaf, M. Effects of fibrates on serum metabolic parameters. Curr.
Med. Res. Opin. 2002, 18, 269−276.
(8) 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.
(9) Campbell, I. W. The clinical significance of PPAR gamma
agonism. Curr. Mol. Med. 2005, 5, 349−363.
(10) Reifel Miller, A. Today’s challenges and tomorrow’s
opportunities: ligands to peroxisome proliferator-activated receptors
as therapies for type 2 diabetes and metabolic syndrome. Drug Dev.
Res. 2006, 67, 574−578.
(11) Henke, B. R. Peroxisome proliferator-activated receptor α/γ
dual agonists for the treatment of type 2 diabetes. J. Med. Chem. 2004,
47, 4118−4127.
(12) 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.; Rajagopalan, R. (−)3-[4-[2-(Phenoxazin-10-
yl)ethoxy]phenyl]-2-ethoxypropanoic acid [(−)DRF 2725]: a dual
PPAR agonist with potent antihyperglycemic and lipid modulating
activity. J. Med. Chem. 2001, 44, 2675−2678.
(13) Sauerberg, P.; Pettersson, I.; Jeppesen, L.; Bury, P. S.;
Mogensen, J. P.; Wassermann, K.; Brand, C. L.; Sturis, J.; Woldike,
H. F.; Fleckner, J.; Andersen, A. S.; 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-α-
alkyloxyphenylpropionic acids: dual PPARα/γ agonists with hypolipi-
demic and antidiabetic activity. J. Med. Chem. 2002, 45, 789−804.
(14) Ebdrup, S.; Pettersson, I.; Rasmussen, H. B.; Deussen, H. J.;
Frost Jensen, A.; Mortensen, S. B.; Fleckner, J.; Pridal, L.; Nygaard, L.;
Sauerberg, P. Synthesis and biological and structural characterization
of the dual-acting peroxisome proliferator-activated receptor α/γ
agonist ragaglitazar. J. Med. Chem. 2003, 46, 1306−1317.
(15) 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.
ASSOCIATED CONTENT
■
S
* Supporting Information
Physicochemical properties and spectroscopic data for inter-
mediates and final compounds, statistics of crystallographic data
and refinement, superposition of (R)-16 and Mol B of (S)-16
into the LBD of PPARγ, titration curve of PPARγ with 1 and
(R)-16, reverse titration curve of (R)-16 with PPARγ, and
superposition of (S)-16 and PPARγ partial agonists 25, 26, and
27 into the PPARγ-LBD. This material is available free of
charge via the Internet at http://pubs.acs.org.
Accession Codes
PDB IDs 4E4K, 4E4Q, 3B3K, and 2REW.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +39 080-5442778. Fax: +39 080-5442231. E-mail:
fulvio.loiodice@uniba.it.
Notes
The authors declare no competing financial interest.
70
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
Y.; Humphreys, W. G.; Sasseville, V. G.; Biller, S. A.; Ryono, D. E.;
Selan, F.; Hariharan, N.; Cheng, P. T. 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 α/γ dual agonist with
efficacious glucose and lipid-lowering activities. J. Med. Chem. 2005,
48, 2248−2250.
(16) 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 PPARα/
γ dual agonists as antihyperglycemic and hypolipidemic agents. J. Med.
Chem. 2004, 47, 3255−3263.
(17) Shi, G. Q.; Dropinski, J. F.; McKeever, B. M.; Xu, S.; Becker, J.
W.; Berger, J. P.; MacNaul, K. L.; Elbrecht, A.; Zhou, G.; Doebber, T.
W.; Wang, P.; Chao, Y. S.; Forrest, M.; Heck, J. V.; Moller, D. E.;
Jones, A. B. Design and synthesis of α-aryloxyphenylacetic acid
derivatives: a novel class of PPARα/γ dual agonists with potent
antihyperglycemic and lipid modulating activity. J. Med. Chem. 2005,
48, 4457−4468.
(18) Ljung, B.; Bamberg, K.; Dahllof, B.; Kjellstedt, A.; Oakes, N. D.;
Ostling, J.; Svensson, L.; Camejo, G. AZ 242, a novel PPARα/γ agonist
with beneficial effects on insulin resistance and carbohydrate and lipid
metabolism in ob/ob mice and obese Zucker rats. J. Lipid Res. 2002,
43, 1855−1863.
(19) Grey, A. Skeletal consequences of thiazolidinedione therapy.
Osteoporosis Int. 2008, 19, 129−137.
(20) Rubenstrunk, A.; Hanf, R.; Hum, D. W.; Fruchart, J.-C.; Staels,
B. Safety issues and prospects for future generations of PPAR
modulators. Biochim. Biophys. Acta 2007, 1771, 1065−1081.
(21) Shearer, B. G.; Billin, A. N. The next generation of PPAR drugs:
do we have the tools to find them? Biochim. Biophys. Acta 2007, 1771,
1082−1093.
(22) Montanari, R.; Saccoccia, F.; Scotti, E.; Crestani, M.; Godio, C.;
Gilardi, F.; Loiodice, F.; Fracchiolla, G.; Laghezza, A.; Tortorella, P.;
Lavecchia, A.; Novellino, E.; Mazza, F.; Aschi, M.; Pochetti, G. Crystal
structure of the peroxisome proliferator-activated receptor γ (PPARγ)
ligand binding domain complexed with a novel partial agonist: a new
region of the hydrophobic pocket could be exploited for drug design. J.
Med. Chem. 2008, 51, 7768−7776.
(29) Itoh, T.; Fairall, L.; Amin, K.; Inaba, Y.; Szanto, A.; Balint, B. L.;
Nagy, L.; Yamamoto, K.; Schwabe, J. W. R. Structural basis for the
activation of PPARγ by oxidized fatty acids. Nat. Struct. Mol. Biol.
2008, 15, 924−931.
(30) Communication to Worldpharma 2010 (16th IUPHAR World
Congress of Basis and Clinical Pharmacology, Copenhagen, Denmark,
July 17−23, 2010; Paper 2908, p 434.
(31) GOLD, version 5.0.1; CCDC Software Ltd.: Cambridge, U.K.,
2008.
(32) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727−748.
(33) Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.;
Taylor, R. D. Improved protein-ligand docking using GOLD. Proteins:
Struct., Funct., Genet. 2003, 52, 609−623.
(34) Wang, W.; Devasthale, P.; Farrelly, D.; Gu, L.; Harrity, T.; Cap,
M.; Chu, C.; Kunselman, L.; Morgan, N.; Ponticiello, R.; Zebo, R.;
Zhang, L.; Locke, K.; Lippy, J.; O’Malley, K.; Hosagrahara, V.; Zhang,
L.; Kadiyala, P.; Chang, C.; Muckelbauer, J.; Doweyko, A. M.; Zahler,
R.; Ryono, D.; Hariharan, N.; Cheng, P. T. Discovery of azetidinone
acids as conformationally-constrained dual PPARα/γ agonists. Bioorg.
Med. Chem. Lett. 2008, 18, 1939−1944.
(35) Meyer, E. A.; Castellano, R. K.; Diederich, F. Interactions with
aromatic rings in chemical and biological recognition. Angew. Chem.,
Int. Ed. 2003, 42, 1210−1250.
(36) Kawatkar, S. P.; Kuntz, D. A.; Woods, R. J.; Rose, D. R.; Boons,
G. J. Structural basis of the inhibition of Golgi α-mannosidase II by
mannostatin A and the role of the thiomethyl moiety in ligand−
protein interactions. J. Am. Chem. Soc. 2006, 128, 8310−8319.
(37) Lee, K.; Kerner, J.; Hoppel, C. L. Mitochondrial carnitine
palmitoyltransferase 1a (CPT1a) is part of an outer membrane fatty
acid transfer complex. J. Biol. Chem. 2011, 286, 25655−25662.
(38) Rufer, A. C.; Thoma, R.; Hennig, M. Structural insight into
function and regulation of carnitine palmitoyltransferase. Cell. Mol. Life
Sci. 2009, 66, 2489−2501.
(39) Zeth, K.; Thein, M. Porins in prokaryotes and eukaryotes:
common themes and variations. Biochem. J. 2010, 431, 13−22.
(40) Ramsay, R. R.; Gandour, R. D.; van der Leij, F. R. Molecular
enzymology of carnitine transfer and transport. Biochim. Biophys. Acta
2001, 1546, 21−43.
(41) Raspe, E.; Madsen, L.; Lefebvre, A. M.; Leitersdorf, I.; Gelman,
L.; Peinado-Onsurbe, J.; Dallongeville, J.; Fruchart, J.-C.; Berge, R.;
Staels, B. Modulation of rat liver apolipoprotein gene expression and
serum lipid levels by tetradecylthioacetic acid (TTA) via PPARα
activation. J. Lipid Res. 1999, 40, 2099−2110.
(42) Hollon, T.; Yoshimura, F. K. Variation in enzymatic transient
gene expression assays. Anal. Biochem. 1989, 182, 411−418.
(43) Iacobazzi, V.; Infantino, V.; Convertini, P.; Vozza, A.; Agrimi,
G.; Palmieri, F. Transcription of the mitochondrial citrate carrier gene:
identification of a silencer and its binding protein ZNF224. Biochem.
Biophys. Res. Commun. 2009, 386, 186−191.
(23) Iacobazzi, V.; Convertini, P.; Infantino, V.; Scarcia, P.; Todisco,
S.; Palmieri, F. Statins, fibrates and retinoic acid upregulate
mitochondrial acylcarnitine carrier gene expression. Biochem. Biophys.
Res. Commun. 2009, 388, 643−647.
(24) Indiveri, C.; Iacobazzi, V.; Tonazzi, A.; Giangregorio, N.;
Infantino, V.; Convertini, P.; Console, L.; Palmieri, F. The
mitochondrial carnitine/acylcarnitine carrier: function, structure and
physiopathology. Mol. Aspects Med. 2011, 32, 223−233.
(25) Gutgesell, A.; Wen, G.; Konig, B.; Koch, A.; Spielmann, J.;
̈
Stangl, G. I.; Eder, K.; Ringseis, R. Mouse carnitine-acylcarnitine
translocase (CACT) is transcriptionally regulated by PPARα and
PPARδ in liver cells. Biochim. Biophys. Acta, Gen. Subj. 2009, 1790,
1206−16.
(44) Convertini, P.; Infantino, V.; Bisaccia, F.; Palmieri, F.; Iacobazzi,
V. Role of FOXA and Sp1 in mitochondrial acylcarnitine carrier gene
expression in different cell lines. Biochem. Biophys. Res. Commun. 2011,
404, 376−381.
(26) Pinelli, A.; Godio, C.; Laghezza, A.; Mitro, N.; Fracchiolla, G.;
Tortorella, V.; Lavecchia, A.; Novellino, E.; Fruchart, J. C.; Staels, B.;
Crestani, M.; Loiodice, F. Synthesis, biological evaluation, and
molecular modeling investigation of new chiral fibrates with PPARα
and PPARγ agonist activity. J. Med. Chem. 2005, 48, 5509−5519.
(27) Bruning, J. B.; Chalmers, M. J.; Prasad, S.; Busby, S. A.;
Kamenecka, T. M.; He, Y.; Nettles, K. W.; Griffin, P. R. Partial agonists
activate PPARγ using a helix 12 independent mechanism. J. Med.
Chem. 2007, 15, 1258−1271.
(28) Li, Y.; Wang, Z.; Furukawa, N.; Escaron, P.; Weiszmann, J.; Lee,
G.; Lindstrom, M.; Liu, J.; Liu, X.; Xu, H.; Plotnikova, O.; Prasad, V.;
Walker, N.; Learned, R. M.; Chen, J. L. T2384, a novel antidiabetic
agent with unique peroxisome proliferator-activated receptor γ binding
properties. J. Biol. Chem. 2008, 283, 9168−9176.
(45) Pochetti, G.; Godio, C.; Mitro, N.; Caruso, D.; Galmozzi, A.;
Scurati, S.; Loiodice, F.; Fracchiolla, G.; Tortorella, P.; Laghezza, A.;
Lavecchia, A.; Novellino, E.; Mazza, F.; Crestani, M. Insights into the
mechanism of partial agonism: crystal structures of the peroxisome
proliferator-activated receptor γ ligand-binding domain in the complex
with two enantiomeric ligands. J. Biol. Chem. 2007, 282, 17314−17324.
(46) Leslie, A. G. W. Recent changes to the MOSFLM package for
processing film and image plate data. Jt. CCP4 ESF-EACMB Newsl.
Protein Crystallogr. 1992, 26, 27−33.
(47) Navaza, J. AMoRe: an automated package for molecular
replacement. Acta Crystallogr., Sect. A 1994, 50, 157−163.
(48) Brunger, 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.
71
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72

Journal of Medicinal Chemistry
Article
Crystallography NMR system: a new software suite for macro-
molecular structure determination. Acta Crystallogr., Sect. D 1998, 54,
905−921.
(49) Molecular Operating Environment (MOE), version 2010.10;
Chemical Computing Group, Inc.: Montreal, Canada, 2010; www.
chemcomp.com (accessed February 15, 2012).
(50) Huang, C. C.; Couch, G. S.; Pettersen, E. F.; Ferrin, T. E.
Chimera: an extensible molecular modeling application constructed
using standard components. Pac. Symp. Biocomput. 1996, 1, 724. Also
see the Chimera Home Page. http://www.cgl.ucsf.edu/chimera
(accessed February 10, 2012).
(51) DeLano, W. L. The PyMOL Molecular Graphics System; DeLano
Scientific LLC: San Carlos, CA; http://www.pymol.org/ (accessed
April 20, 2012).
(52) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J. B.; Meyer, E. F.,
Jr.; Brice, M. D.; Rodgers, J. R.; Kennard, O.; Shimanouchi, T.;
Tasumi, T. The Protein Data Bank: a computer based archival file for
macromolecular structures. J. Mol. Biol. 1977, 112, 535−542.
(53) Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G. V.;
Mee, R. P. Empirical scoring functions: I. The development of a fast
empirical scoring function to estimate the binding affinity of ligands in
receptor complexes. J. Comput.-Aided Mol. Des. 1997, 11, 425−445.
(54) Muralidhara, B. K.; Negi, S. S.; Halpert, J. R. Dissecting the
thermodynamics and cooperativity of ligand binding in cytochrome
P450eryF. J. Am. Chem. Soc. 2007, 129, 2015−2024.
(55) Locuson, C. W.; Gannett, P. M.; Ayscue, R.; Tracy, T. S. Use of
simple docking methods to screen a virtual library for heteroactivators
of cytochrome P450 2C9. J. Med. Chem. 2007, 50, 1158−1165.
72
dx.doi.org/10.1021/jm301018z | J. Med. Chem. 2013, 56, 60−72