(Z)-2-(2-bromophenyl)-3-{[4-(1-methyl-piperazine)amino]phenyl}acrylonitrile (DG172): An orally bioavailable PPARβ/δ-selective ligand with inverse agonistic properties

Article abstract of DOI:10.1021/jm2017122

The ligand-regulated nuclear receptor peroxisome proliferator-activated receptor β/δ (PPARβ/δ) is a potential pharmacological target due to its role in disease-related biological processes. We used TR-FRET-based competitive ligand binding and coregulator

Full text of DOI:10.1021/jm2017122

Article
pubs.acs.org/jmc
(Z)-2-(2-Bromophenyl)-3-{[4-(1-methyl-piperazine)amino]phenyl}-
acrylonitrile (DG172): An Orally Bioavailable PPARβ/δ-Selective
Ligand with Inverse Agonistic Properties
Sonja Lieber,†,§ Frithjof Scheer,‡,§ Wolfgang Meissner,† Simone Naruhn,† Till Adhikary,†
Sabine Muller-Brusselbach,† Wibke E. Diederich,*,‡ and Rolf Muller*,†
̈
̈
̈
†Institute of Molecular Biology and Tumor Research (IMT), Philipps University, Emil-Mannkopff-Strasse 2,
35033 Marburg, Germany
‡Institute of Pharmaceutical Chemistry, Philipps-University, Marbacher Weg 6, 35032 Marburg, Germany
S
* Supporting Information
ABSTRACT: The ligand-regulated nuclear receptor peroxisome
proliferator-activated receptor β/δ (PPARβ/δ) is a potential
pharmacological target due to its role in disease-related biological
processes. We used TR-FRET-based competitive ligand binding
and coregulator interaction assays to screen 2693 compounds of
the Open Chemical Repository of the NCI/NIH Developmental
Therapeutics Program for inhibitory PPARβ/δ ligands. One com-
pound, (Z)-3-(4-dimethylamino-phenyl)-2-phenyl-acrylonitrile,
was used for a systematic SAR study. This led to the design of
derivative 37, (Z)-2-(2-bromophenyl)-3-{[4-(1-methyl-piperazine)amino]phenyl}acrylonitrile (DG172), a novel PPARβ/δ-selective
ligand showing high binding affinity (IC50 = 27 nM) and potent inverse agonistic properties. 37 selectively inhibited the agonist-
induced activity of PPARβ/δ, enhanced transcriptional corepressor recruitment, and down-regulated transcription of the PPARβ/δ
target gene Angptl4 in mouse myoblasts (IC50 = 9.5 nM). Importantly, 37 was bioavailable after oral application to mice with peak
plasma levels in the concentration range of its maximal inhibitory potency, suggesting that 37 will be an invaluable tool to elucidate
the functions and therapeutic potential of PPARβ/δ.
and HDAC recruiting factors, protein arginine methyl trans-
ferases, and factors with chromatin remodeling functions. While
the role of histone acetylation in PPAR-mediated transcrip-
tional activation is well established, the exact role of other enzy-
matic modifications and coregulators remains unclear, in particular
for the PPARβ/δ subtype. The mechanisms of PPARβ/δ-mediated
repression by PPRE-bound unliganded receptors are even less
understood. A number of corepressors have been identified,
such as class I HDACs, NCoR/SMRT, and SHARP,10 but their
precise function in the regulation of specific target genes involv-
ing the ordered assembly and disassembly of multiprotein com-
plexes is not known. The complexity of PPARβ/δ-mediated
transcriptional regulation is further complicated by the fact
that distinct regulatory mechanisms govern the expression
of different sets of target genes.11 Thus, repression appears
to represent the major mode of PPARβ/δ-mediated transcrip-
tional regulation, and only a subset of target genes is subject to
an agonist-mediated switch from active repression to activation.
Finally, PPARs can also regulate genes without making direct
DNA contacts by directly interacting with specific transcrip-
tion factors, as exemplified by the repression of BCL-6 by
PPARβ/δ.12
INTRODUCTION
■
Members of the class II subset of nuclear receptors, including
the thyroid hormone receptor, the retinoic acid receptor, and
peroxisome proliferator-activated receptors (PPARs), can actively
repress target genes in the absence of ligand binding but activate
the same genes if bound by an agonistic ligand.1 These activities
are linked to the induction of distinct local chromatin structures
depending on the presence or absence of an agonistic ligand.
The three PPAR subtypes (PPARα, PPARβ/δ, and PPARγ)
regulate their target genes through binding to specific DNA
elements (PPREs) as obligatory heterodimers with the retinoid
X receptor. Certain lipids, fatty acid metabolites, and subtype-
selective synthetic ligands modulate their transcriptional
activity,2−4 suggesting that PPARs act as sensors for both en-
dogenous and exogenous stimuli, which impinge not only on
intermediary metabolism but also on inflammatory pathways.5
In addition to these functions, PPARs figure in development, wound
healing, cell differentiation, proliferation, and apoptosis.6−8
PPRE-bound PPARβ/δ complexes have functions in both
transcriptional repression and transcriptional activation. Ago-
nistic ligands induce a conformational change in PPARs that
favors the association with coactivators and the dissociation of
corepressors.9 Many PPAR-interacting coregulators have been
described, including histone acetyl transferases (HATs) and
HAT-recruiting coregulators, histone deacetylases (HDACs)
Received: January 2, 2012
Published: February 27, 2012
© 2012 American Chemical Society
2858
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
Because of these complexities, the correlation of biological
functions and transcriptional pathways regulated by PPARβ/δ
is difficult. This is exemplified by the genetic disruption of
Ppard genes, which can have opposite effects of individual
PPARβ/δ target genes, depending on their mode of transcrip-
tional regulation, which in turn hampers the assessment of
PPARβ/δ as a potential target for pharmacological inhibition.
While potent synthetic agonists that are bioavailable, selective
for PPARβ/δ, and bind reversibly are available, inhibitory ligands
for PPARβ/δ fulfilling these criteria have not been described to
date. Both 2-(2-methyl-4-((4-methyl-2-(naphthalen-1-yl)thiazol-
5-yl)methylthio)phenoxy)acetic acid (SR13904)13 and 4-chloro-
N-(2-((5-trifluoromethyl-2-pyridyl)sulfonyl-)ethyl)benzamide
(GSK3787)14,15 are not specific for PPARβ/δ, and GSK3787
binds PPARβ/δ irreversibly, which is pharmacologically undesirable.
3-(((2-Methoxy-4-(phenylamino)phenyl)amino)sulfonyl)-2-thio-
phenecarboxylate (GSK0660)16 is PPARβ/δ subtype-specific but is
not bioavailable. This also applies to methyl 3-(N-(4-(hexylamino)-
2-methoxyphenyl)sulfamoyl)thiophene-2-carboxylate (ST247),
a recently developed GSK0660 derivative with greatly improved
affinity.17,18 These ligands are not only competitive antagonists
but exert their inhibitory function as inverse agonists, as indi-
cated by their inhibitory effect on the basal expression of PPARβ/δ
target genes and an increased recruitment of transcriptional
corepressors.15−17 Finally, a biphenylcarboxylic acid-based
antagonist has been described, but its in vivo performance has
not been addressed.19
gene ANGPTL4 (Supporting Information Table S1). Compound
1, which is (Z)-3-[4-(dimethylamino)phenyl]-2-phenylacrylnitrile,
was used as a lead structure for further development, as described
in detail below.
Among the eight compounds identified as inverse agonists is
the clinically important drug (Z)-2-[4-(1,2-diphenylbut-1-enyl)-
phenoxy]-N,N-dimethylethanamine (tamoxifen) (Supporting
Information Table S1). However, in spite of efficient corepressor
recruitment in vitro, no activity was detectable in the cell-based
assays. The same observations were made with three metabo-
lites of tamoxifen, i.e., 4-OH-tamoxifen, N-desmethyl-tamoxifen,
and endoxifen (Supporting Information Table S2). Because these
compounds are able to modulate estrogen receptor-driven gene
expression in intact cells, their failure to affect PPARβ/δ activity
cannot be attributed to a lack of cellular uptake. It is possible that
the subcellular compartmentalization of tamoxifen and its me-
tabolites is a limiting step restricting the accessibility of target
proteins. We also analyzed other commercially available stil-
benes, including the pharmacologically relevant compounds
resveratrol and diethylstilbestrol, but did not observe any sig-
nificant activities (Supporting Information Table S2). These
observations show that binding to PPARβ/δ is not a general
property of stilbenes.
Optimization of the Screening Hit 1. 1 was chosen as
starting point for optimization (Figure 1). We first turned our
In light of the lack of inhibitory PPARβ/δ ligands suitable for
in vivo applications, we have searched for novel chemical struc-
tures that could serve as leads for the development of improved
inverse agonists. Toward this end, we screened a chemical com-
pound library and identified several stilbene-based or -related
inhibitory PPARβ/δ ligands. One of these compounds was chosen
for further development and the establishment of structure−
activity relationships. This finally yielded a compound with the
desired properties, including high affinity, specificity, and bio-
availability after oral application.
Figure 1. Strategy for optimization of the initial screening hit 1.
RESULTS AND DISCUSSION
■
attention toward the central acrylonitrile moiety. However,
modification at this position, e.g., by hydrogenation 2, removal
3 or alteration of the position of the nitrile functionality 4, or
elongation leading to the 1,3-butadiensystem 5 resulted in a
complete loss of activity (Supporting Information Figure S1).
Therefore, the acrylonitrile moiety seems to be crucial for activity.
We then examined the effect of the para-dimethylamino-substituent
present in 1. Removal (6) or replacement by a variety of either
electron-withdrawing or electron-donating functional groups
(7−12) again led to a significant drop in affinity. The only
exception turned out to be 13 bearing a primary amino func-
tionality in para-position, indicating that the existence of an
electron-related push−pull system is essential for activity
(Supporting Information Figure S1). Consequently, introduc-
tion of a dimethylaminomethylene substituent in para-position
14 (Figure 2) and thus disruption of the conjugated push−pull
system also diminished the binding affinity toward the PPARβ/
δ-LDB significantly. Because the para-dimethylamino derivative
1 possessed a higher binding affinity than the unsubstituted
para-amino-representative 13, we focused our attention on the
substitution pattern of this essential amino group to achieve a
further increase in binding affinity (Figure 2). Besides tertiary
amines of varying ring sizes, such as in pyrrolidine- (15),
Screening for Inhibitory PPARβ/δ Ligands. A TR-FRET-
based competitive ligand-binding assay was used to screen 2693
compounds of the Open Chemical Repository of the NCI/NIH
Developmental Therapeutics Program for PPARβ/δ ligands. In
this assay, the terbium-labeled PPARβ/δ LBD interacts with the
fluorescent PPAR ligand Fluormone Pan-PPAR Green, which
produces FRET from terbium (495 nm) to Pan-PPAR Green
(520 nm). Displacement of the fluorescent ligand by an un-
labeled test compound results in a quantifiable attenuation of
FRET. Out of 191 identified compounds, 10 disrupted the in-
teraction of the PPARβ/δ LBD with a coactivator peptide in a
TR-FRET-based assay (Supporting Information Table S1). Four
of these compounds possess a stilbene-based or -related core
structure. In this assay, interaction of the PPARβ/δ LBD
(indirectly labeled by terbium) with the fluorescein-labeled
coactivator peptide C33 is determined. The data therefore indi-
cates that these 10 compounds act as inhibitory ligands. Eight
of these ligands were also able to trigger the association with
the SMRT-ID2 peptide, derived from the interaction domain
2 of the corepressor SMRT, which qualifies these compounds
as inverse agonists. Two of these ligands, NSC667251 and com-
pound 1 (NSC636948), also showed efficacy in cell-based assays,
i.e., repression of agonist-induced transcription in a luciferase
reporter assay and repression of the endogenous PPARβ/δ target
2859
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
Figure 2. Activity of 1 and the indicated derivatives as PPARβ/δ ligands determined in vitro by competitive ligand binding assay. Displacement
of a fluorescent PPAR ligand (Fluormone Pan-PPAR Green) from recombinant GST-PPARβ/δ by the indicated compounds was determined by
TR-FRET. Each compound was tested at a concentration of 1 μM. Results are expressed as the ratio of fluorescence intensity at 520 nm (fluorescein
emission excitated by terbium emission) and 495 nm (terbium emission). All data points represent averages of triplicates ( SD). ***, **, and *:
significant difference to compound 1 by t test (P < 0.001, P < 0.01, and P < 0.05, respectively).
Figure 3. Activity of compound 1 and the indicated derivatives as PPARβ/δ ligands. All experimental details were as in Figure 2.
piperidine- (16), or azepane- (17) substituted structures present,
we also tested two secondary amines (18, 19).
Although the competitive TR-FRET assay showed only slight
differences between these compounds, the piperidine analogue
gave the best results in a cell-based luciferase reporter assay
(data not shown). Hence, further compounds bearing six-
membered heterocycles were synthesized. Introduction of a
4-methylpiperidino (20), a morpholino (21), and a piperazino
(22) moiety, respectively, led to a significant gain in affinity.
The best compound within this series was found to be 23
2860
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
equipped with a 4-methylpiperazino substituent. The two sec-
ondary amines, the aniline- (19) as well as the cyclohexylamine-
derivative (18), also possessed a higher binding affinity compared
to 1 but could still not compete with 23.
We then turned our attention to the second aromatic portion
within this compound class (Figure 3). The initial screening hit
1 was likewise used as reference. However, any tested substituent
introduced in para-position of this phenyl substituent led to a
decrease in binding affinity, indicating that there might only
be limited space available within the respective binding pocket
(24−27). On the contrary, introducing a chlorine substituent in
meta-position 28 gave a significant improvement in binding
affinity. This effect was even more pronounced for this
substituent in ortho-position as in compound 29 (DG138).
Iodine as ortho-substituent 30 performed equally well while
compound 31, equipped with a bromine in this position, turned
out to be the most potent ligand within this series. Introduction
of other substituents with a stronger -I-effect such as 32, 33, and
34 only led to a slight increase in comparison to 1 or even
resulted in a decrease in binding affinity when strong electron
withdrawing groups (35, 36) were introduced.
Combination of the substitution patterns of the most active
compounds of both series, i.e., halogenation in the ortho-position
and the introduction of a 4-methylpiperazine, finally led to deriv-
ative 37 (DG172) (see Scheme 1), analyzed in detail below.
a
Scheme 1. Synthesis of 37
Figure 4. In vitro binding and interaction properties of compound 1 and
its derivatives 29 and 37. (A) FRET-based competitive ligand binding
assay as in Figure 2. GSK0660 is included for comparison. *Measure-
ment of 1 at 10 μM was not possible due to a lack of solubility. (B)
Comparison of 29- and 37-induced binding of a corepressor-derived
peptide to the PPARβ/δ LBD. Interaction of SMRT-ID2 peptide
(fluorescein labeled) and recombinant GST-PPARβ/δ (labeled by a
terbium-coupled anti-GST antibody) was measured by TR-FRET. In
both panels, results are expressed as the ratio of fluorescence intensity at
520 nm (fluorescein emission excitated by terbium emission) and 495
nm (terbium emission). All data points represent averages of triplicates
( SD). ***, **, and *: significant difference by t test compared to
untreated sample (P < 0.001, P < 0.01, and P < 0.05, respectively).
a
Reagents and conditions: (a) K2CO3, DMSO, 100 °C, 78% (b) 2-
bromophenylacetonitrile, pyrrolidine, MeOH, 60 °C, 79%.
an IC50 value of 26.9 nM, compared to ∼180 nM for 29 and
>300 nM for GSK0660 (values are averages from three inde-
pendent experiments each analyzing five different concentrations
as triplicates). The latter two values cannot be accurately deter-
mined due to a lack of solubility at high concentrations.
To evaluate the inhibitory properties of 29 and 37, we inves-
tigated the effect of these compounds on the interaction of
PPARβ/δ with the synthetic corepressor peptide SMRT-ID2 by
TR-FRET. The data obtained by this assay (Figure 4B) show a
clearly enhanced interaction for 37 compared to 29 and thus
closely mirror the results obtained by the competitive binding
assay (Figure 4A). The data also confirm both ligands as inverse
agonists.
Specificity for PPARβ/δ. The PPAR subtype specificity of
29 and 37 was addressed by a competitive TR-FRET assay. The
data in Figure 5 show that at 1 μM both compounds selectively
competed for binding to PPARβ/δ. Competition for binding to
PPARα or PPARγ was extremely low or undetectable. In contrast,
the PPARα agonist GW7647, the PPARβ/δ agonist GW501516,
and the PPARγ agonist GW1929 strongly interacted with the
The compounds described above are easily accessible via a
Knoevenagel condensation, which exclusively yield the (Z)-isomers
(for example, 3J(H,C) = 14.4 Hz for 1), employing the corresponding
aldehydes and phenylacetonitriles under basic conditions. For the
preparation of several of the amino-derivatives, 4-bromophenylalde-
hyde was employed in the Knoevenagel reaction, followed by a
Buchwald−Hartwig reaction20−22 to introduce the respective
amino substituent. In case of 37, 4-fluorobenzaldehyde 38 was
first reacted with 4-methylpiperazine 39 to 40, followed by a
knoevenagel condensation employing 2-bromophenylacetonitrile,
as outlined in Scheme 1.
Binding Affinities and Inhibitory Properties of 29 and
37 in vitro. We next analyzed 37 in further detail with respect
to its binding affinity, inhibitory properties, and specificity. First,
37 was compared to both 29 (harboring the ortho-halogenation
but lacking the 4-methylpiperazine) and its parent molecule 1 in
a competitive ligand binding assay. The data in Figure 4A shows
that 29 possesses a significantly enhanced affinity compared to 1
and performed similarly as a published reference compound,
GSK0660. As expected, 37 was the most potent compound with
2861
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
Figure 5. PPAR subtype binding specificity. Competition of 29 (A) or 37 (B) with Fluormone Pan-PPAR Green for binding to PPARα, PPARβ/δ,
and PPARγ compared to the PPARα agonist GW7647 (top), the PPARβ/δ agonist L165,041 (middle), the PPARγ agonist GW1929 (bottom), or
solvent (DMSO) only. Experimental details are described in Figure 4.
respective PPAR subtype (Figure 5), thus confirming the validity
of the assay.
that the observed effect on Angptl4 expression is mediated though
PPARβ/δ. This is strongly supported by our observation that the
inhibition of Angptl4 expression by 37 was dependent on the
presence of wild-type PPARβ/δ alleles (Figure 7B).
We next analyzed the effect of both compounds (and of
GSK0660 for comparison) on the agonist-induced transcrip-
tional activity of PPARα, PPARβ/δ, and PPARγ in a cell-based
assay. As shown in Figure 6, treatment with subtype-selective
agonists resulted in a 3−7.5-fold activation of the respective
PPAR subtype in luciferase reporter assays. Whereas 29 and
37 had no significant effect on PPARα- or PPARγ-driven
transcription, they both efficiently antagonized ligand activation
of PPARβ/δ, which is consistent with the results of the in vitro
ligand-binding assay described above.
Inhibition of Endogenous PPARβ/δ Target Gene
Expression. The inverse agonistic properties of 29 and 37
were tested in an endogenous cellular context by investigating their
effect on the established PPARβ/δ target gene Angptl4.23,24 Toward
this end, we performed titration experiments to determine the IC50
values for 29 and 37 in C2C12 mouse myoblasts (Figure 7A). The
parent compound 1 and GSK0660 were included in this study for
comparison. This analysis clearly revealed the superior effect of 37
(IC50 = 9.5 nM) compared to the other compounds, which
showed IC50 values of 52 nM (29), >500 nM (1), and 48 nM
(GSK0660), respectively (values are averages from three inde-
pendent experiments each analyzing six different concentrations
as triplicates). Because the tested compounds had no detectable
effect on PPARα and PPARγ (Figures 5 and 6), it is very likely
Effect on Corepressor Recruitment to Chromatin-
Bound PPARβ/δ. To investigate the effect of 37 on the assembly
of chromatin-associated corepressor complexes, we performed chro-
matin immune precipitation (ChIP) analyses of HDAC3 recruit-
ment to the ANGPTL4 gene in WPMY-1 human myofibroblasts. As
can be seen in Figure 8A, 37 induced an enhanced recruitment of
HDAC3 compared to solvent-treated cells (DMSO). The
specificity of the ChIP assay was shown by the lack of antibody
binding to an irrelevant region of the PDK4 gene (Figure 8B)
and by the lack of any detectable effect on HDAC3 binding
(Figure 8A) of reference compound 41, which is a pure PPARβ/δ
antagonist and therefore unable to enhance corepressor recruit-
ment.17 The data in Figure 8A also show that GSK0660 and 37
have similar effects, which do not correlate with the higher po-
tency of 37 to repress ANGPTL4 transcription (Figure 7A). We
attribute this to the possibility that other corepressors are instru-
mental in 37-mediated repression, as suggested by the multitude
of coregulators interacting with repressive PPAR complexes.
Pharmacokinetics in Mice. Finally, to determine the po-
tential suitability of 29 and 37 for in vivo applications, phar-
macokinetic studies were carried out in mice. 29 and 37 were
administered intravenously (1 mg/kg) and orally (5 mg/kg),
2862
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
Figure 7. Impact on expression of the endogenous PPARβ/δ target
gene Angptl4. (A) C2C12 mouse myoblasts were treated for 24 h with
1, 29, and 37 at the indicated concentration, and RNA was analyzed by
RT-qPCR. GSK0660 is included for comparison. (B) Dependence on
PPARβ/δ. Macrophages from Ppard wild-type (WT) and null (KO)
mice were treated with the agonist L165,041 (500 nM), 37 (1 μM),
GSK0660 (1 μM), or with solvent only (DMSO) for 6 h, and the
expression of Angptl4 was determined by RT-qPCR. Statistical analysis
was performed as in Figure 4.
(CL = 176 mL/min/kg; Vss = 6.2 L/kg), 29 was detectable in the
blood at very low levels (≤6 ng/mL) and for a short time following
oral application (≤30 min), indicating a lack of bioavailability.
Figure 6. Effects on the agonist-induced transcriptional activity of
LexA-PPARα (A), LexA-PPARβ/δ (B), and LexA-PPARγ (C).
NIH3T3 cells were transiently transfected with a luciferase reporter
plasmid containing multiple LexA binding sites. Four hours post-
transfection, the cells were treated with the indicated inhibitory ligands
(1 μM) for 48 h, followed by 300 nM of the PPARα agonist GW7647,
1 μM of the PPARβ/δ agonist L165,041, or 300 nM of the PPARγ
agonist GW1929 or agonist solvent. GSK0660 (1 μM) is included for
comparison. Induction values represent luciferase activities of agonist-
treated cells relative to cells treated with agonist solvent. Statistical
analysis was performed as in Figure 4.
CONCLUSIONS
■
By screening a chemical compound library, we identified (Z)-3-
[4-(dimethylamino)phenyl]-2-phenyloacrylnitrile (1) as an inhibitory
PPARβ/δ ligand. A comprehensive SAR study revealed two
modifications, ortho-halogenation and introduction of an N-4-
methylpiperazine moiety, that greatly improved the binding
affinity for PPARβ/δ and the efficiency of corepressors. The
combination of these two critical modifications led to the dis-
covery of (Z)-2-(2-bromophenyl)-3-{[4-(1-methyl-piperazine)-
amino]phenyl}acrylonitrile (37), which is the most potent inverse
agonist for PPARβ/δ known to date. 37 is PPAR-subtype selective
and inhibits both agonist-induced and basal level PPRE-dependent
transcription in cells. Most importantly, 37 has good oral phar-
macokinetic properties, making it the first bioavailable PPARβ/
δ-selective inverse agonist described to date. 37 therefore rep-
resents a useful novel tool to investigate the biological and
pathophysiological functions of PPARβ/δ and to clarify its
potential as a target for drug development.
blood samples were analyzed 10 min to 12 h post-treatment by
HPLC-MS (Figure 9), and basic pharmacokinetic parameters
were determined. After intravenous administration of 37, a plasma
half-life of 76 min was measured, the mean clearance (CL) was
121 mL/min/kg, and the volume of distribution at steady state
(Vss) 12.5 L/kg. Oral administration yielded a good exposure
with an AUCinf of 8239 min·ng/mL and a peak plasma level
(Cmax) of 94 ng/mL (207 nM), which is clearly within the
concentration range of maximal activity determined in vitro (IC50 =
23 nM; Figure 4A) or in cell culture (IC50 = 6.5 nM for C2C12
cells; Figure 7A). Furthermore, half-life (634 min) and bio-
availability (72%) were in the desired range. This pharmaco-
kinetic data set suggests that 37 is suitable for in vivo applications
in mice, including its peroral administration. In contrast, despite an
acceptable plasma half-life after intravenous injection of 76 min
EXPERIMENTAL SECTION
Ligands. {2-Methyl-4-[({4-methyl-2-[4-(trifluoromethyl)phenyl]-
1,3-thiazol-5-yl}methyl)sulfanyl]phenoxy}acetic acid (GW501516) was
■
2863
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
and 4-[3-(2-propyl-3-hydroxy-4-acetyl)phenoxy]propyloxyphenoxy-
acetic acid (L165,041) from Biozol (Eching, Germany), and 2-(4-{2-[4-
cyclohexylbutyl(cyclohexylcarbamoyl)amino]ethyl}phenyl)sulfanyl-2-meth-
ylpropanoic acid (GW7647) from Sigma-Aldrich (Steinheim, Germany).
Synthesis of GSK0660 and compound 41, 3-{N-[4-(tert-butylamino)-2-
methoxyphenyl]sulfamoyl}-
thiophene-2-carboxylate (PT-S58), has been reported previously.19
Chemistry. Reagents and solvents that are commercially available
were used without further purification. Thin layer chromatography was
performed on precoated plates silica gel 60 F254, Merck. Flash column
chromatography was performed on prepacked flash chromatography
columns (PF 30-SIHP-JP/12G) purchased from Interchim and using a
Buchi separation system. Cyclohexane was purchased in pa quality
̈
from Grussing and distilled prior to use, and iso-hexane was purchased
̈
in technical quality and distilled prior to use.
1H NMR and 13C NMR spectra were recorded on a Jeol ECX-400
or on a Jeol ECA-500 spectrometer. Chemical shifts (δ) are given in
ppm with the residual solvent signal used as reference (CDCl3: s, 7.26
ppm [1H] and t, 77.1 ppm [13C]; DMSO-d6: quint, 2.50 ppm [1H]
and septet, 40.1 ppm [13C]). Unless otherwise noted, spectra with
CDCl3 as solvent were recorded at 20 °C while spectra with DMSO-d6
as solvent were recorded at 30.0 °C. Peak patterns were described as
folows: s (singlet), d (doublet), dd (double doublet), ddd (doublet of
doublet of doublet), t (triplet), m (multiplet), sm (symmetric multiplet),
bs (broad singlet), psd (pseudo doublet). Mass spectra were recorded
on a double-focusing sector field spectrometer type 70-70H (Vacuum
Generators) or on a double-focusing sector field spectrometer type
AutoSpec (Micromass). Elemental combustion analyses were performed
on a Vario MICRO cube (Elementar Analysensysteme GmbH, Hanau,
Germany). Melting points were determined using a melting point meter
Figure 8. Corepressor binding to PPARβ/δ. The impact of 37 on
recruitment of HDAC3 to the ANGPTL4 promoter in WPMY-1
myofibroblasts was determined by ChIP. Compound 41 does not
induce corepressor recruitment17 and was used as a negative control.
Cells were treated with the indicated compounds (1 μM) for 30 min.
ChIP was carried out using antibodies against HDAC3 or a nonspecific
rabbit IgG pool (negative control). DNA was amplified with primers
encompassing the ANGPTL4 PPREs (A) or a control region (B).
Relative amounts of amplified DNA in immunoprecipitates were
calculated by comparison with 1% of input DNA. Results are expressed
as % input. Statistical analysis was performed as in Figure 4.
KSP1N (A. Kruss Optronic GmbH, Hamburg, Germany) and are
̈
uncorrected.
All tested compounds were at least 95% pure as a single isomer,
determined by NMR and combustion analysis.
Procedure A: To a solution of the respective phenylacetonitrile
(1 equiv) and the corresponding benzaldehyde (1 equiv) in methanol
(0.6 M) was added potassium hydroxide, and the reaction mixture was
stirred at RT until TLC indicated full conversion of the starting
material. The precipitate was collected, washed with water and hexane,
and dried in vacuo.
Procedure B: To a solution of the respective phenylacetonitrile
(1 equiv) and the corresponding benzaldehyde (1 equiv) in methanol
(0.6 M) was added pyrrolidine, and the reaction mixture was stirred
until TLC indicated full conversion of the starting material. The pre-
cipitate was collected, washed with water and hexane, and dried in vacuo.
Procedure C: (Z)-3-(4-Bromophenyl)-2-phenylacrylonitrile (1 equiv,
prepared following procedure A) was dissolved in dry toluene (0.7 M)
under argon atmosphere. ( )-BINAP (0.075 equiv), Pd2(dba)3 (0.05
equiv), sodium tert-butoxide (1.5 equiv), and the corresponding amine
(2 equiv) were added, and the suspension was stirred at 80 °C until thin
layer chromatography indicated full conversion of the starting material.
The reaction mixture was diluted with DCM, filtered through a pad of
Celite, absorbed on silica gel, and purified by flash chromatography.
(Z)-3-{4-[(Dimethylamino)methyl]phenyl}-2-phenylacrylonitrile
Hydrochloride (14). To a solution of 4-[(dimethylamino)methyl]-
benzaldehyde (105 mg, 0.90 mmol) and phenylacetonitrile (105 mg,
0.90 mmol) in methanol (2 mL) was added potassium hydroxide
(50 mg, 0.90 mmol), and the reaction mixture was stirred for 24 h at
RT. The reaction mixture was diluted with EtOAc, and the organic
phase was washed with water, saturated potassium hydrogencarbonate
solution and brine, dried over MgSO4, filtered, and concentrated in
vacuo. The free base was obtained by flash chromatography (hexane/
EtOAc, gradient from 0 to 50% in 15 min) and was afterward con-
verted to the hydrochloride salt 14 (120 mg, 0.40 mmol, 45%) by
precipitation from EtOAc with HCl (5−6 M in i-PrOH); mp above
Figure 9. Pharmacokinetics in mice. 29 and 37 were administered
either intravenously at a dose of 1 mg/kg (A) or orally at 5 mg/kg (B),
and blood samples were analyzed by HPLC-MS/MS at the indicated
time points post-treatment. Results represent averages of biological
triplicates ( SD). Both compounds were undetectable at 24 h.
1
decomposition temperature. H NMR (DMSO-d6) δ 11.04 (bs, 1H),
8.07 (s, 1H), 7.97 (psd, J = 8.2 Hz, 2H), 7.78−7.71 (m, 4H), 7.54−
7.48 (m, 2H), 7.47−7.42 (sm, 1H), 4.31 (s, 2H), 2.68 (s, 6H). 13C
NMR (DMSO-d6) δ 142.6, 135.2, 134.1, 133.3, 132.1, 130.1, 129.9,
129.8, 126.4, 118.3, 111.9, 59.4, 42.1. HRMS (EI) calcd for C18H18N2
purchased from Axxora (Lorrach, Germany), N-(2-benzoylphenyl)-O-
̈
[2-(methyl-2-pyridinylamino)ethyl]-L-tyrosine hydrochloride (GW1929)
2864
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
[M]+ 262.146999; found 262.145737. Anal. Calcd for C18H19ClN2: C,
72.35; H, 6.41; N, 9.37. Found: C, 71.83; H, 6.52; N, 9.21.
7.37−7.31 (sm, 1H), 6.99 (psd, J = 9.2 Hz, 2H), 3.23−3.19 (sm, 4H),
2.81−2.77 (sm, 4H). 13C NMR (CDCl3) δ 152.8, 142.2, 135.3, 131.2,
129.0, 128.4, 125.7, 124.1, 119.1, 114.5, 106.5, 48.8, 45.9. HRMS (EI)
calcd for C19H19N3 [M]+ 289.157898; found 289.155945.
(Z)-2-Phenyl-3-[4-(piperidin-1-yl)phenyl]acrylonitrile (16). Accord-
ing to procedure B, employment of 4-(piperidin-1-yl)benzaldehyde
(492 mg, 2.60 mmol), benzyl cyanide (305 mg, 2.60 mmol), and
pyrrolidine (185 mg, 2.60 mmol) gave rise to 16 as a yellow solid
(Z)-3-[4-(4-Methylpiperazin-1-yl)phenyl]-2-phenylacrylonitrile
(23). According to procedure B, employment of 4-(4-methylpiperazin-
1-yl)benzaldehyde (265 mg, 1.30 mmol), benzyl cyanide (152 mg,
1.30 mmol), and pyrrolidine (92 mg, 1.30 mmol) furnished 23 as a
1
(150 mg, 0.52 mmol, 20%); mp 128 °C. H NMR (CDCl3) δ 7.84
(psd, J = 8.9 Hz, 2H), 7.66−7.61 (m, 2H), 7.44−7.38 (m, 3H), 7.35−7.30
(m, 1H), 6.92 (psd, J = 7.8 Hz, 2H), 3.37−3.31 (m, 4H), 1.76−1.60
(m, 6H). 13C NMR (CDCl3) δ 152.8, 142.4, 135.5, 131.4, 129.0, 128.3,
125.6, 123.2, 119.4, 114.5, 105.6, 48.9, 25.5, 24.5. HRMS (EI) calcd
for C20H20N2 [M]+ 288.162649; found 288.164001. Anal. Calcd for
C20H20N2: C, 83.30; H, 6.99; N, 9.71. Found: C, 83.23; H, 7.14; N, 9.81.
(Z)-3-[4-(Cyclohexylamino)phenyl]-2-phenylacrylonitrile (18).
According to procedure C, utilization of (Z)-3-(4-bromophenyl)-2-
phenylacrylonitrile (200 mg, 0.70 mmol), ( )-BINAP (32.9 mg, 0.053
mmol), Pd2(dba)3 (32.2 mg, 0.035 mmol), sodium tert-butoxide
(102 mg, 1.06 mmol), and cyclohexylamine (140 mg, 1.41 mmol)
yielded, after purification by flash chromatography (iso-hexane/EtOAc,
gradient from 0 to 25% in 12 min), 18 as a yellow solid (94 mg, 0.31
1
yellow solid (268 mg, 0.88 mmol, 68%); mp 143 °C. H NMR
(DMSO-d6) δ 7.84 (psd, J = 8.9 Hz, 2H), 7.81 (s, 1H), 7.69−7.64
(m, 2H), 7.47−7.41 (m, 2H), 7.37−7.32 (sm, 1H), 7.02 (psd, J =
9.2 Hz, 2H), 3.30 (t, J = 5.0 Hz, 4H), 2.41 (t, J = 5.0 Hz, 4H), 2.19
(s, 3H). 13C NMR (DMSO-d6) δ 152.7, 143.2, 135.2, 131.5, 129.6,
128.8, 125.7, 123.5, 119.5, 114.5, 104.7, 54.9, 47.2, 46.3. HRMS
(EI) calcd for C20H21N3 [M]+ 303.173548; found 303.171852.
Anal. Calcd for C20H21N3: C, 79.17; H, 6.98; N, 13.85. Found: C,
78.95; H, 7.01; N, 13.86.
(Z)-2-(4-Chlorophenyl)-3-[4-(dimethylamino)phenyl]acrylonitrile
(27). According to procedure A, usage of 4-(dimethylamino)benzaldehyde
(351 mg, 2.35 mmol), 2-(4-chlorophenyl)acetonitrile (357 mg, 2.35 mmol),
and potassium hydroxide (132 mg, 2.35 mmol) furnished 27 as a yellow
1
mmol, 44%); mp 122 °C. H NMR (DMSO-d6) δ 7.74 (psd, J = 8.7
1
solid (326 mg, 1.15 mmol, 49%); mp 193 °C. H NMR (CDCl3) δ 7.85
Hz, 2H), 7.69 (s, 1H), 7.64−7.60 (m, 2H), 7.45−7.39 (m, 2H), 7.33−
7.28 (sm, 1H), 6.64 (psd, J = 8.7 Hz, 2H), 6.40 (d, J = 7.8 Hz, 1H),
3.33−3.23 (sm, 1H), 1.93−1.86 (sm, 2H), 1.74−1.65 (sm, 2H), 1.61−
1.53 (sm, 1H), 1.39−1.27 (sm, 2H), 1.21−1.10 (m, 3H). 13C NMR
(CDCl3) δ 149.5, 142.8, 135.7, 131.7, 129.0, 128.1, 125.6, 122.3, 119.6,
112.6, 104.5, 51.4, 33.3, 25.8, 25.0. HRMS (EI) calcd for C21H22N2
[M]+ 302.178299; found 302.178004. Anal. Calcd for C21H22N2: C,
83.40; H, 7.33; N, 9.26. Found: C, 83.27; H, 7.26; N, 9.10.
(psd, J = 8.9 Hz, 2H), 7.55 (psd, J = 8.9 Hz, 2H), 7.38−7.35 (m, 3H), 6.74
(psd, J = 8.9 Hz, 2H), 3.06 (s, 6H). 13C NMR (CDCl3) δ 152.0, 142.9,
134.2, 133.8, 131.5, 129.1, 126.7, 121.4, 119.3, 111.7, 103.2, 40.2. HRMS
(EI) calcd for C17H15ClN2 [M]+ 282.092376; found 282.093166. Anal.
Calcd for C17H15ClN2: C, 72.21; H, 5.35; N, 9.91. Found: C, 72.06; H,
5.37; N, 9.85.
(Z)-2-(3-Chlorophenyl)-3-[4-(dimethylamino)phenyl]acrylonitrile
(28). According to procedure A, employment of 4-(dimethylamino)-
benzaldehyde (585 mg, 3.92 mmol), 2-(3-chlorophenyl)acetonitrile
(595 mg, 3.92 mmol), and potassium hydroxide (220 mg, 3.92 mmol)
gave rise to 28 as a yellow solid (710 mg, 2.51 mmol, 49%); mp
(Z)-2-Phenyl-3-[4-(phenylamino)phenyl]acrylonitrile (19). Follow-
ing procedure C, usage of (Z)-3-(4-bromophenyl)-2-phenylacryloni-
trile (200 mg, 0.70 mmol), ( )-BINAP (32.9 mg, 0.053 mmol),
Pd2(dba)3 (32.2 mg, 0.035 mmol), sodium tert-butoxide (102 mg, 1.06
mmol), and aniline (131 mg, 1.41 mmol) yielded, after purification by
flash chromatography (iso-hexane/EtOAc, gradient from 0 to 25% in
12 min), 19 as a yellow solid (110 mg, 0.37 mmol, 53%); mp 162 °C.
1H NMR (CDCl3) δ 7.85 (psd, J = 8.7 Hz, 2H), 7.67−7.63 (m, 2H),
7.45−7.44 (m, 3H), 7.38−7.32 (m, 3H), 7.21−7.17 (m, 2H), 7.11−
7.04 (m, 3H). 13C NMR (CDCl3) δ 146.1, 142.1, 141.0, 135.2, 131.4,
129.6, 129.2, 128.6, 125.8, 125.5, 123.0, 120.2, 119.1, 115.7, 107.0.
HRMS (EI) calcd for C21H16N2 [M]+ 296.131349; found 296.129489.
Anal. Calcd for C21H16N2: C, 85.11; H, 5.44; N, 9.45. Found: C, 84.78;
H, 5.66; N, 9.19.
1
132 °C. H NMR (CDCl3) δ 7.85 (psd, J = 8.9 Hz, 2H), 7.60 (t, J =
1.9 Hz, 1H), 7.50 (sm, 1H), 7.38 (s, 1H), 7.33 (t, J = 7.8 Hz, 1H), 7.26
(sm, 1H), 6.71 (psd, J = 9.1 Hz, 2H), 3.06 (s, 6H). 13C NMR (CDCl3)
δ 152.0, 143.6, 137.6, 135.0, 131.7, 130.2, 127.9, 125.4, 123.7, 121.2,
119.2, 111.7, 102.8, 40.1. MS (EI) m/z (%) 282.1 (100) [M]+. Anal.
Calcd for C17H15ClN2: C, 72.21; H, 5.35; N, 9.91. Found: C, 72.35; H,
5.51; N, 9.82.
(Z)-2-(2-Chlorophenyl)-3-[4-(dimethylamino)phenyl]acrylonitrile
(29). Following procedure B, usage of 4-(dimethylamino)benzaldehyde
(351 mg, 2.35 mmol), 2-(2-chlorophenyl)acetonitrile (357 mg, 2.35 mmol),
and pyrrolidine (167 mg, 2.35 mmol) at 60 °C furnished 29 as a
yellow solid (326 mg, 1.15 mmol, 49%); mp 99 °C. 1H NMR (CDCl3)
δ 7.85 (psd, J = 8.9 Hz, 2H), 7.46−7.40 (m, 2H), 7.33−7.26 (m, 2H),
7.12 (s, 1H), 6.73 (psd, J = 8.9 Hz, 2H), 3.06 (s, 6H). 13C NMR
(CDCl3) δ 152.0, 148.4, 135.6, 133.2, 131.5, 131.0, 130.4, 129.7, 127.4,
121.2, 119.0, 111.7, 101.8, 40.2. HRMS (EI) calcd for C17H15ClN2
[M]+ 282.092376; found 282.094431. Anal. Calcd for C17H15ClN2: C,
72.21; H, 5.35; N, 9.91. Found: C, 72.43; H, 5.53; N, 10.00.
(Z)-3-[4-(Dimethylamino)phenyl]-2-(2-iodophenyl)acrylonitrile
(30). To a solution of 4-dimethylaminobenzaldehyde (161 mg, 1.08 mmol)
and 2-(2-iodophenyl)acetonitrile (263 mg, 1.08 mmol) in methanol (2 mL)
was added pyrrolidine (145 mg, 1.08 mmol), and the reaction mixture
was stirred for 18 h at 60 °C. The reaction mixture was diluted with
EtOAc, and the organic phase was washed with water, saturated
potassium hydrogencarbonate solution, and brine, dried over MgSO4,
filtered, and concentrated in vacuo. Flash chromatography (cyclo-
hexane/EtOAc, gradient from 0 to 30% in 12 min) furnished 30 as a
yellow solid (185 mg, 0.49 mmol, 46%); mp 134 °C. 1H NMR (CDCl3)
δ 7.94−7.91 (m, 1H), 7.85 (psd, J = 8.9 Hz, 2H), 7.42−7.36 (m, 2H),
7.04 (ddd, J = 7.7, 6.6, 2.5 Hz, 1H), 6.99 (s, 1H), 6.73 (psd, J = 9.2 Hz,
2H), 3.06 (s, 6H). 13C NMR (CDCl3) δ 152.1, 148.5, 141.1, 140.1,
131.4, 130.5, 129.9, 128.7, 121.0, 118.8, 111.7, 106.6, 98.7, 40.2. HRMS
(EI) calcd for C17H15IN2 [M]+ 374.028001; found 374.024834. Anal.
Calcd for C17H15IN2: C, 54.56; H, 4.04; N, 7.49. Found: C, 54.82; H,
4.16; N, 7.45.
(Z)-3-[4-(4-Methylpiperidin-1-yl)phenyl]-2-phenylacrylonitrile
(20). According to procedure C, utilization of (Z)-3-(4-bromophenyl)-
2-phenylacrylonitrile (200 mg, 0.70 mmol), ( )-BINAP (32.9 mg,
0.053 mmol), Pd2(dba)3 (32.2 mg, 0.035 mmol), sodium tert-butoxide
(102 mg, 1.06 mmol), and 4-methylpiperidine (140 mg, 1.41 mmol)
rendered, after purification by flash chromatography (iso-hexane/DCM,
1
5:2), 20 as a yellow solid (194 mg, 0.64 mmol, 91%); mp 120 °C. H
NMR (DMSO-d6) δ 7.83 (psd, J = 8.9 Hz, 2H), 7.78 (s, 1H), 7.68−7.63
(m, 2H), 7.46−7.41 (m, 2H), 7.36−7.31 (sm, 1H), 6.99 (psd, J = 9.2
Hz, 2H), 3.92−3.85 (sm, 2H), 2.84−2.76 (sm, 2H), 1.69−1.62 (sm,
2H), 1.63−1.50 (sm, 1H), 1.21−1.08 (sm, 2H), 0.90 (d, J = 6.4 Hz,
3H). 13C NMR (CDCl3) δ 152.6, 142.4, 135.5, 131.3, 129.0, 128.3,
125.6, 123.2, 119.4, 114.5, 105.6, 48.2, 33.7, 31.0, 22.0. HRMS (EI)
calcd for C21H22N2 [M]+ 302.178299; found 302.178744. Anal. Calcd
for C21H22N2: C, 83.40; H, 7.33; N, 9.26. Found: C, 83.20; H, 7.30; N,
8.81.
(Z)-2-Phenyl-3-[4-(piperazin-1-yl)phenyl]acrylonitrile (22). (Z)-3-
(4-Bromophenyl)-2-phenylacrylonitrile (100 mg, 0.35 mmol) was
dissolved in dry toluene (2 mL) under an argon atmosphere. Tri-tert-
butylphosphine (14.2 mg, 0.070 mmol), Pd2(dba)3 (16.1 mg, 0.018
mmol), sodium tert-butoxide (101 mg, 1.06 mmol), and piperazine
(182 mg, 2.11 mmol) were added, and the suspension was stirred at
120 °C for 15 h. The reaction mixture was diluted with DCM, filtered
through a pad of Celite, absorbed on silica gel, and purified by flash
chromatography (DCM/methanol, 50:1), giving rise to 22 as a yellow
1
wax (53.1 mg, 0.18 mmol, 52%). H NMR (DMSO-d6) δ 7.84 (psd,
(Z)-2-(2-Bromophenyl)-3-[4-(dimethylamino)phenyl]acrylonitrile
(31). A solution of 2-bromophenylacetonitrile (376 mg, 1.93 mmol)
J = 9.2 Hz, 2H), 7.80 (s, 1H), 7.68−7.64 (m, 2H), 7.47−7.41 (m, 2H),
2865
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
and 4-dimethylaminobenzaldehyde (288 mg, 1.93 mmol) in morpho-
line (2 mL) was stirred for 12 h at 120 °C. The reaction mixture was
absorbed onto silica, and flash chromatography (iso-hexane/EtOAc/
DCM, 18:1:1) gave rise to 31 as yellow solid (184 mg, 0.56 mmol,
7.37−7.33 (sm, 1H), 7.32 (s, 1H), 7.10 (psd, J = 8.9 Hz, 2H), 4.05−
3.95 (m, 2H), 3.49−3.29 (m, 4H), 3.18−3.05 (m, 2H), 2.77 (s, 3H).
13C NMR (DMSO-d6) δ 151.7, 148.6, 136.9, 133.7, 132.1, 131.4,
131.3, 129.1, 124.2, 122.8, 118.5, 115.3, 105.3, 52.1, 44.6, 42.4. HRMS
(EI) calcd for C20H20BrN3 [M]+ 381.084059; found 381.087401. Anal.
Calcd for C20H22BrCl2N3: C, 52.77; H, 4.87; N, 9.23. Found: C, 52.68;
H, 4.97; N, 9.18.
Time-Resolved Fluorescence Resonance Energy Transfer
(TR-FRET) Assays in Vitro. Ligand binding was determined by
TR-FRET in vitro25 using the Lanthascreen TR-FRET PPARβ/δ
competitive binding assay as described.26,27 The interaction of the
PPARβ/δ LBD with a fluorescein-labeled corepressor peptide derived
from the silencing mediator for retinoid and thyroid hormone recep-
tors interaction domain 2 (SMRT-ID2) was determined using the
Lanthascreen TR-FRET PPARβ/δ coregulator assay.27 Assays were
carried out and evaluated as described.
Chemical Compound Library Screening. The Open Chemical
Repository of the NCI/NIH Developmental Therapeutics Program
consisting of the Approved Oncology Drugs Set III (97 compounds),
the Diversity Set III (1597 compounds), the Mechanistic Set (879
compounds), and the Natural Product Set II (120 compounds) was
initially screened for compounds binding to the PPARβ/δ LBD using
the competitive TR-FRET assay described above. Compounds showing
significant competition (n = 129) were subsequently validated in tripli-
cates using TR-FRET-based coactivator and corepressor peptide recruit-
ment assays (see above).27
1
29%); mp 140 °C. H NMR (CDCl3) δ 7.85 (psd, J = 8.9 Hz, 2H),
7.64 (dd, J = 8.0, 1.1 Hz, 1H), 7.41 (dd, J = 7.6, 1.8 Hz, 1H), 7.35 (td,
J = 7.6, 1.4 Hz, 1H), 7.21 (ddd, J = 7.6, 1.8, 0.5 Hz, 1H), 7.07 (s, 1H),
6.72 (psd, J = 9.2 Hz, 2H), 3.06 (s, 6H). 13C NMR (CDCl3) δ 152.1,
148.4, 137.5, 133.6, 131.4, 131.2, 129.8, 127.9, 123.2, 121.1, 118.9,
111.7, 103.5, 40.2. HRMS (EI) calcd for C17H15BrN2 [M]+
326.041860; found 326.042488. Anal. Calcd for C17H15BrN2: C,
62.40; H, 4.62; N, 8.56. Found: C, 62.34; H, 4.79; N, 8.44.
(Z)-3-[4-(Dimethylamino)phenyl]-2-(2-methoxyphenyl)-
acrylonitrile (32). To a solution of 4-dimethylaminobenzaldehyd (304 mg,
2.04 mmol) and 2-(2-methoxyphenyl)acetonitrile (300 mg, 2.04 mmol)
in methanol (4 mL) was added pyrrolidine (145 mg, 2.04 mmol), and
the reaction mixture was stirred for 48 h at 60 °C. The reaction mixture
was diluted with EtOAc, and the organic phase was washed with water,
saturated potassium hydrogencarbonate solution, and brine, dried over
MgSO4, filtered, and concentrated in vacuo. 32 was obtained after flash
chromatography (cyclohexane/EtOAc/DCM, 8:1:1) as a yellow solid
(111 mg, 0.40 mmol, 20%); mp 97 °C. 1H NMR (CDCl3) δ 7.84 (psd,
J = 8.7 Hz, 2H), 7.38 (dd, J = 7.6, 1.6 Hz, 1H), 7.32 (ddd, J = 8.2, 7.4,
1.6 Hz, 1H), 7.28 (s, 1H), 6.99 (td, J = 7.6, 1.1 Hz, 1H), 6.94 (dd, J =
8.2, 0.7 Hz, 1H), 6.76 (psd, J = 7.1 Hz, 2H), 3.91 (s, 3H), 3.05 (s, 6H).
13C NMR (CDCl3) δ 157.0, 151.6, 146.5, 131.2, 129.9, 129.7, 125.9,
Cell Culture. WPMY-1 human myofibroblasts28 (ATCC, CRL-2854),
C2C12 murine myoblasts29 (kindly provided by Dr. Thomas Braun,
Bad Nauheim, Germany), and NIH3T3 cells were maintained in
DMEM supplemented with 10% fetal bovine serum, 100 U/mL
penicillin, and 100 μg/mL streptomycin in a humidified incubator at
37 °C and 5% CO2.
Transcription, Gene Expression, and Chromatin Analyses.
Luciferase reporter assays were performed and evaluated as reported
previously. LexA-PPAR expression plasmids and the 7 L-TATAi luciferase
reporter construct have been described elsewhere.30,31 RT-qPCR analyses
of endogenous Angptl4 expression and statistical analyses were carried out
as described,17 using L27 as the normalizer. ChIP analysis was performed
as reported elsewhere.24,32
Pharmacokinetics in Mice. In vivo pharmacokinetic studies were
performed by Cerep, Redmond, WA. Briefly, compounds were for-
mulated in DMSO/Solutol HS 15/PBS, pH 7.4 (5/5/90, v/v/v) and
administered iv (1 mg/kg) and po (5 mg/kg) to male CD-1 mice by
tail vein injection and gastric gavage, respectively. Blood samples were
taken at eight time points post injection by parallel sampling (three
mice each; see Figure 9 for details). Plasma samples were processed by
acetonitrile precipitation and analyzed by HPLS-MS/MS following
standard procedures.
122.1, 121.0, 119.6, 111.6, 111.5, 102.0, 55.9, 40.2. HRMS (EI) calcd
for C18H18N2O [M]+ 278.141913; found 278.140550. Anal. Calcd for
C18H18N2O: C, 77.67; H, 6.52; N, 10.06. Found: C, 77.25; H, 6.61; N,
9.74.
(Z)-3-[4-(Dimethylamino)phenyl]-2-(2-(trifluoromethyl)phenyl)-
acrylonitrile (33). A solution of 2-(2-trifluoromethylphenyl)acetoni-
trile (200 mg, 1.08 mmol) and 4-dimethylaminobenzaldehyde (161 mg,
1.08 mmol) in morpholine (2 mL) was stirred for 24 h at 120 °C and
subsequently absorbed onto silica gel. Flash chromatography (iso-hexane/
EtOAc, 5:1) gave rise to 33 as a yellow solid (67 mg, 0.21 mmol, 20%); mp
1
110 °C. H NMR (CDCl3) δ 7.82 (psd, J = 8.9 Hz, 2H), 7.76−7.72 (m,
1H), 7.61−7.56 (m, 1H), 7.52−7.46 (m, 2H), 6.98 (s, 1H), 6.80 (psd, J =
8.2 Hz, 2H), 3.07 (s, 6H). 13C NMR (CDCl3) δ 152.1, 148.3, 147.1, 135.7,
132.2, 132.0, 131.3, 129.3, 128.7, 124.0 (d, JC,F = 274.0 Hz), 121.0, 119.2,
111.6, 100.7, 40.0. HRMS (EI) calcd for C18H15F3N2 [M]+ 316.118733;
found 316.117731. Anal. Calcd for C18H15F3N2: C, 68.35; H, 4.78; N, 8.86.
Found: C, 68.58; H, 5.18; N, 8.75.
(Z)-3-[4-(Dimethylamino)phenyl]-2-(2-fluorophenyl)acrylonitrile
(34). Following procedure B, usage of 4-(dimethylamino)benzaldehyde
(442 mg, 2.96 mmol), 2-(2-fluorophenyl)acetonitrile (400 mg, 2.96
mmol), and pyrrolidine (463 mg, 6.51 mmol) gave rise to 34 as a
yellow solid (583 mg, 2.19 mmol, 74%); mp 106 °C. 1H NMR
(CDCl3) δ 7.86 (psd, J = 8.9 Hz, 2H), 7.54 (td, J = 7.8, 1.8 Hz, 1H),
7.42 (s, 1H), 7.32−7.26 (sm, 1H), 7.19 (td, J = 7.6, 1.1 Hz, 1H), 7.12
(ddd, J = 11.2, 8.2, 1.1 Hz, 1H), 6.71 (psd, J = 9.2 Hz, 2H), 3.06 (s,
6H). 13C NMR (CDCl3) δ 159.8 (d, JC,F = 250.4 Hz), 152.0, 147.4 (d,
JC,F = 7.8 Hz), 131.6, 129.7 (d, JC,F = 3.0 Hz), 129.6 (d, JC,F = 8.7 Hz),
124.6 (d, JC,F = 3.0 Hz), 124.2 (d, JC,F = 11.6 Hz), 121.5, 119.3, 116.5
(d, JC,F = 23.1 Hz), 111.6, 98.7 (d, JC,F = 1.9 Hz), 40.1. HRMS (EI)
calcd for C17H15FN2 [M]+ 266.121927; found 266.123324. Anal. Calcd
for C17H15FN2: C, 76.67; H, 5.68; N, 10.52. Found: C, 76.57; H, 5.73;
N, 10.50.
ASSOCIATED CONTENT
* Supporting Information
■
S
Properties of compounds identified by library screening as
PPARβ/δ ligands, Analysis of stilbene derivatives as potential
PPARβ/δ ligands, Analysis of compound 1 (NSC636948) and
derivatives by competitive in vitro ligand binding assay, and
experimental procedures for further compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
(Z)-2-(2-Bromophenyl)-3-[4-(4-methylpiperazin-1-yl)phenyl]-
acrylonitrile Dihydrochloride (37). To a solution of 2-(2-bromophenyl)-
acetonitrile (480 mg, 2.45 mmol) and 4-(4-methylpiperazino)benzaldehyde
(500 mg, 2.45 mmol) in methanol (4 mL) was added pyrrolidine (174 mg,
2.45 mmol). The reaction mixture was stirred for 48 h at 60 °C and
subsequently absorbed onto silica gel. The free base was obtained by
flash chromatography (DCM/MeOH, 49:1) and was afterward con-
verted to the dihydrochloride salt 37 (806 mg, 1.93 mmol, 79% yield)
by precipitation from EtOAc with HCl (5−6 M in iPrOH); mp above
decomposition temperature. 1H NMR (DMSO-d6, 80 °C, 500 MHz) δ
11.46 (bs, 1H), 9.96 (bs, 1H), 7.85 (psd, J = 8.9 Hz, 2H), 7.71 (d, J =
8.0 Hz, 1H), 7.51 (dd, J = 7.7, 1.7 Hz, 1H), 7.48−7.45 (sm, 1H),
AUTHOR INFORMATION
Corresponding Author
*For W.D.: phone, +49 6421 2825810; E-mail, wibke.diederich@
staff.uni-marburg.de. For R.M.: phone, +49 6421 2866236; E-mail,
rmueller@imt.uni-marburg.de.
■
Author Contributions
§The first two authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
2866
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
(12) Lee, C. H.; Chawla, A.; Urbiztondo, N.; Liao, D.; Boisvert, W. A.;
Evans, R. M.; Curtiss, L. K. Transcriptional repression of atherogenic
inflammation: modulation by PPARdelta. Science 2003, 302, 453−457.
(13) Zaveri, N. T.; Sato, B. G.; Jiang, F.; Calaoagan, J.; Laderoute,
K. R.; Murphy, B. J. A novel peroxisome proliferator-activated receptor
delta antagonist, SR13904, has anti-proliferative activity in human
cancer cells. Cancer Biol. Ther. 2009, 8, 1252−1261.
ACKNOWLEDGMENTS
■
We thank Klaus Weber, Margitta Alt, and Julia Dick for expert
technical assistance. This work is supported by grants to R.M.
from the Deutsche Forschungsgemeinschaft (SFB-TR17/A3)
and the LOEWE-Schwerpunkt “Tumor and Inflammation” of
the state of Hesse.
(14) Shearer, B. G.; Wiethe, R. W.; Ashe, A.; Billin, A. N.; Way, J. M.;
Stanley, T. B.; Wagner, C. D.; Xu, R. X.; Leesnitzer, L. M.; Merrihew,
R. V.; Shearer, T. W.; Jeune, M. R.; Ulrich, J. C.; Willson, T. M.
Identification and characterization of 4-chloro-N-(2-{[5-trifluorometh-
yl)-2-pyridyl]sulfonyl}ethyl)benzamide (GSK3787), a selective and
irreversible peroxisome proliferator-activated receptor delta (PPAR-
delta) antagonist. J. Med. Chem. 2010, 53, 1857−1861.
ABBREVIATIONS USED
■
ANGPTL4, angiopoietin-like 4 protein; ANGPTL4, angiopoie-
tin-like 4 gene (human); Angptl4, angiopoietin-like 4 gene
(mouse); BCL-6, B-cell chronic lymphatic leukemia/lymphoma
6 protein; ChIP, chromatin immune precipitation; CL, mean
clearance; DBD, DNA binding domain; FRET, fluorescence
resonance energy transfer; GST, gluthatione S-transferase;
HAT, acetyl transferase; HDAC, histone deacetylase; LBD,
ligand binding domain; NCoR, nuclear receptor corepressor;
PDK4, pyruvate dehydrogenase kinase 4 gene; PPAR,
peroxisome proliferator-activated receptor; PPRE, peroxisome
proliferator responsive element; RT-qPCR, real-time quantita-
tive polymerase chain reaction; RXR, retinoid X receptor; SAR,
structure−activity relationship; SMRT, silencing mediator for
retinoid and thyroid hormone receptors; SMRT-ID2, SMRT
interaction domain 2; SHARP, SMRT and HDAC-associated
repressor protein; TR-FRET, time-resolved fluorescence
resonance energy transfer; Vss, volume of distribution at steady
state
(15) Palkar, P. S.; Borland, M. G.; Naruhn, S.; Ferry, C. H.; Lee, C.;
Sk, U. H.; Sharma, A. K.; Amin, S.; Murray, I. A.; Anderson, C. R.;
Perdew, G. H.; Gonzalez, F. J.; Muller, R.; Peters, J. M. Cellular and
̈
pharmacological selectivity of the peroxisome proliferator-activated
receptor-beta/delta antagonist GSK3787. Mol. Pharmacol. 2010, 78,
419−430.
(16) Shearer, B. G.; Steger, D. J.; Way, J. M.; Stanley, T. B.; Lobe,
D. C.; Grillot, D. A.; Iannone, M. A.; Lazar, M. A.; Willson, T. M.; Billin,
A. N. Identification and characterization of a selective peroxisome
proliferator-activated receptor beta/delta (NR1C2) antagonist. Mol.
Endocrinol. 2008, 22, 523−529.
(17) Naruhn, S.; Toth, P. M.; Adhikary, T.; Kaddatz, K.; Pape, V.;
Dorr, S.; Klebe, G.; Muller-Brusselbach, S.; Diederich, W. E.; Muller,
̈
̈
̈
̈
R. High-affinity peroxisome proliferator-activated receptor beta/delta-
specific ligands with pure antagonistic or inverse agonistic properties.
Mol. Pharmacol. 2011, 80, 828−838.
(18) Toth, P. M.; Naruhn, S.; Pape, V. F.; Dorr, S. M.; Klebe, G.;
̈
REFERENCES
■
Muller, R.; Diederich, W. E. Development of Improved PPARbeta/
̈
(1) Lonard, D. M.; O’Malley, B. W. Nuclear receptor coregulators:
judges, juries, and executioners of cellular regulation. Mol. Cell 2007,
27, 691−700.
(2) Desvergne, B.; Michalik, L.; Wahli, W. Transcriptional regulation
of metabolism. Physiol. Rev. 2006, 86, 465−514.
(3) Zoete, V.; Grosdidier, A.; Michielin, O. Peroxisome proliferator-
activated receptor structures: ligand specificity, molecular switch and
interactions with regulators. Biochim. Biophys. Acta 2007, 1771, 915−
925.
(4) Peraza, M. A.; Burdick, A. D.; Marin, H. E.; Gonzalez, F. J.;
Peters, J. M. The toxicology of ligands for peroxisome proliferator-
activated receptors (PPAR). Toxicol. Sci. 2006, 90, 269−295.
(5) Glass, C. K.; Saijo, K. Nuclear receptor transrepression pathways
that regulate inflammation in macrophages and T cells. Nature Rev.
Immunol. 2010, 10, 365−376.
(6) Michalik, L.; Wahli, W. Peroxisome proliferator-activated
receptors (PPARs) in skin health, repair and disease. Biochim. Biophys.
Acta 2007, 1771, 991−998.
(7) Peters, J. M.; Gonzalez, F. J. Sorting out the functional role(s) of
peroxisome proliferator-activated receptor-beta/delta (PPARbeta/
delta) in cell proliferation and cancer. Biochim. Biophys. Acta 2009,
1796, 230−241.
(8) Montagner, A.; Rando, G.; Degueurce, G.; Leuenberger, N.;
Michalik, L.; Wahli, W. New insights into the role of PPARs.
Prostaglandins, Leukotrienes Essent. Fatty Acids 2011, 85, 235−243.
(9) Yu, S.; Reddy, J. K. Transcription coactivators for peroxisome
proliferator-activated receptors. Biochim. Biophys. Acta 2007, 1771,
936−951.
(10) Shi, Y.; Hon, M.; Evans, R. M. The peroxisome proliferator-
activated receptor delta, an integrator of transcriptional repression and
nuclear receptor signaling. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2613−
2618.
delta Inhibitors. ChemMedChem 2012, 7, 159−170.
(19) Kasuga, J.; Ishida, S.; Yamasaki, D.; Makishima, M.; Doi, T.;
Hashimoto, Y.; Miyachi, H. Novel biphenylcarboxylic acid peroxisome
proliferator-activated receptor (PPAR) delta selective antagonists.
Bioorg. Med. Chem. Lett. 2009, 19, 6595−6599.
(20) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. A Simple Catalytic
Method for the Conversion of Aryl Bromides to Arylamines. Angew.
Chem., Int. Ed. 1995, 34, 1348−1350.
(21) Jiang, L.; Buchwald, S. L. Palladium-Catalyzed Aromatic
Carbon−Nitrogen Bond Formation; WILEY-VCH Verlag GmbH &
Co: Weinheim, Germany, 2004; Vol. 2, pp 699−760.
(22) Louie, J.; Hartwig, J. F. Palladium-catalyzed synthesis of
arylamines from aryl halides. Mechanistic studies lead to coupling in
the absence of tin reagents. Tetrahedron Lett. 1995, 36, 3609−3612.
(23) Mandard, S.; Zandbergen, F.; Tan, N. S.; Escher, P.; Patsouris,
D.; Koenig, W.; Kleemann, R.; Bakker, A.; Veenman, F.; Wahli, W.;
Muller, M.; Kersten, S. The direct peroxisome proliferator-activated
receptor target fasting-induced adipose factor (FIAF/PGAR/
ANGPTL4) is present in blood plasma as a truncated protein that is
increased by fenofibrate treatment. J. Biol. Chem. 2004, 279, 34411−
34420.
(24) Kaddatz, K.; Adhikary, T.; Finkernagel, F.; Meissner, W.;
Muller-Brusselbach, S.; Muller, R. Transcriptional profiling identifies
̈
̈
̈
functional interactions of TGFβ and PPARβ/δ signaling: synergistic
induction of ANGPTL4 transcription. J. Biol. Chem. 2010, 285,
29469−29479.
(25) Stafslien, D. K.; Vedvik, K. L.; De Rosier, T.; Ozers, M. S.
Analysis of ligand-dependent recruitment of coactivator peptides to
RXRbeta in a time-resolved fluorescence resonance energy transfer
assay. Mol. Cell. Endocrinol. 2007, 264, 82−89.
(26) Rieck, M.; Meissner, W.; Ries, S.; Muller-Brusselbach, S.; Muller, R.
̈
̈
̈
Ligand-mediated regulation of peroxisome proliferator-activated receptor
(PPAR) beta/delta: a comparative analysis of PPAR-selective agonists and
all-trans retinoic acid. Mol. Pharmacol. 2008, 74, 1269−1277.
(11) Adhikary, T.; Kaddatz, K.; Finkernagel, F.; Schonbauer, A.; Meissner,
̈
W.; Scharfe, M.; Jarek, M.; Blocker, H.; Muller-Brusselbach, S.; Muller, R.
̈
̈
̈
̈
Genomewide analyses define different modes of transcriptional regulation
by peroxisome proliferator-activated receptor-beta/delta (PPARbeta/delta).
PLoS One 2011, 6, e16344.
(27) Naruhn, S.; Meissner, W.; Adhikary, T.; Kaddatz, K.; Klein, T.;
Watzer, B.; Muller-Brusselbach, S.; Muller, R. 15-Hydroxyeicosatetraenoic
̈
̈
̈
2867
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868

Journal of Medicinal Chemistry
Article
acid is a preferential peroxisome proliferator-activated receptor β/δ agonist.
Mol. Pharmacol. 2010, 77, 171−184.
(28) Webber, M. M.; Trakul, N.; Thraves, P. S.; Bello-DeOcampo, D.;
Chu, W. W.; Storto, P. D.; Huard, T. K.; Rhim, J. S.; Williams, D. E. A
human prostatic stromal myofibroblast cell line WPMY-1: a model for
stromal−epithelial interactions in prostatic neoplasia. Carcinogenesis
1999, 20, 1185−1192.
(29) Yaffe, D.; Saxel, O. Serial passaging and differentiation of
myogenic cells isolated from dystrophic mouse muscle. Nature 1977,
270, 725−727.
(30) Fauti, T.; Muller-Brusselbach, S.; Kreutzer, M.; Rieck, M.;
̈
̈
Meissner, W.; Rapp, U.; Schweer, H.; Komhoff, M.; Muller, R. Induction
̈
̈
of PPARbeta and prostacyclin (PGI2) synthesis by Raf signaling: failure
of PGI2 to activate PPARbeta. FEBS J. 2006, 273, 170−179.
́ ̂
(31) Jerome, V.; Muller, R. Tissue-specific, cell cycle-regulated
̈
chimeric transcription factors for the targeting of gene expression to
tumor cells. Hum. Gene Ther. 1998, 9, 2653−2659.
(32) Stockert, J.; Adhikary, T.; Kaddatz, K.; Finkernagel, F.;
Meissner, W.; Muller-Brusselbach, S.; Muller, R. Reverse crosstalk of
̈
̈
̈
TGFβ and PPARβ/δ signaling identified by transcriptional profiling.
Nucleic Acids Res. 2011, 39, 119−131.
2868
dx.doi.org/10.1021/jm2017122 | J. Med. Chem. 2012, 55, 2858−2868