Isoxazolyl-serine-based agonists of peroxisome proliferator-activated receptor: Design, synthesis, and effects on cardiomyocyte differentiation

Article abstract of DOI:10.1021/ja046386l

The peroxisome proliferator-activated receptors (PPARs) are important molecular targets for the development of drugs for the treatment of human metabolic diseases, inflammation, and cancer. They are known to be activated by a variety of structurally diver

Full text of DOI:10.1021/ja046386l

Published on Web 12/04/2004
Isoxazolyl-Serine-Based Agonists of Peroxisome Proliferator-Activated
Receptor: Design, Synthesis, and Effects on Cardiomyocyte Differentiation
Zhi-Liang Wei,† Pavel A. Petukhov,† Fero Bizik,† Joaquim Cabral Teixeira,‡ Mark Mercola,‡
Eugene A. Volpe,§ Robert I. Glazer,§ Timothy M. Willson,| and Alan P. Kozikowski*,†
Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, UniVersity of Illinois at Chicago,
833 South Wood Street, Chicago, Illinois 60612, Stem Cell and Regeneration Program, The Burnham Institute,
10901 North Torrey Pines Road, La Jolla, California 92037, Department of Oncology, Georgetown UniVersity
Medical Center, 3900 ReserVoir Road, NW, Washington, D.C. 20057, and GlaxoSmithKline, Research Triangle
Park, North Carolina 27709
Received June 18, 2004; E-mail: kozikowa@uic.edu
The peroxisome proliferator-activated receptors (PPARs) are
members of the nuclear receptor superfamily that plays key roles
in lipid, carbohydrate, and cholesterol metabolism.1 Three subtypes,
PPARR, γ, and δ (or â), have been identified from Xenopus to
humans, each forming a functional heterodimer complex with the
9-cis-retinoic acid receptor (RXR). PPARR is predominantly
expressed in several tissues that have high lipid catabolism activity.
PPARγ is highly enriched in adipocytes, while the δ subtype is
ubiquitously expressed. PPARs are important molecular targets for
the development of drugs for the treatment of human metabolic
diseases, inflammation, and cancer,1,2 as demonstrated by the
thiazolidinediones (TZDs) and fibrates.
PPARs are known to be activated by a variety of structurally
diverse compounds, including naturally occurring fatty acids and
Figure 1. Some known PPAR agonists and the general structure.
synthetic ligands (Figure 1).1,3 Structures of the PPAR ligand
Chart 1. Novel isoxazolyl/triazolyl-serine-based PPAR agonists.
binding domains (LBDs) in the absence and presence of ligands
have been solved by X-ray crystallography.4 Each of the PPAR
subtypes possesses a large Y-shaped binding pocket (∼1300 Å3)
located within the lower half of the LBD. Because of the large
size of the ligand-binding pockets, PPARs are capable of binding
a number of structurally diverse ligands. Despite differences in their
gross chemical structure, small molecule PPAR agonists share a
common binding mode.4f The structural characteristics of PPARs
have provided molecular insights into the role ligands play in
regulating PPAR activity on one hand; they also provide key
insights for the design of new PPAR ligands with altered binding
A virtual combinatorial library based on the isoxazolyl-serine
ligand was created and docked to the LBD of PPARγ as identified
in the crystal structure of PPARγ-GW409544 complex.4b The
docking and scoring were performed using the FlexX and CScore
characteristics and modified receptor pharmacology on the other
modules available in Sybyl 6.91. Several compounds from the
hand. We present herein the identification of novel PPAR ligands
combinatorial library were identified as potential ligands. Our
using a structure-based drug design approach.
docking studies suggest that the isoxazolyl-serine-based ligands have
Our structure-based design efforts were based in part on the SAR
modes of binding similar to that seen in the crystal structure of
available for the known PPAR ligands. Most of the known current
PPARγ with GW409544 (Figure 2). The phenyl moiety of 1b
synthetic PPARR, γ, and δ ligands have several common elements
occupies the hydrophobic, phenylalanine-rich subpocket (Phe282,
(Figure 1): a polar “head” group “A” connected to an aromatic
Phe360, and Phe363).
ring “C” through a short linker “B”, and a linker “D” connecting
A common feature of many PPAR ligands is the presence of a
the aromatic ring “C” to either an aromatic or aliphatic ring system
carboxyl group. Reported crystal structures have shown that the
“E”. The linkers B and D can also contain additional substituents
carboxyl group is important for stabilizing the C-terminal helix in
F and G. The reason for these structural characteristics relates to
an active conformation and that it is coordinated by two histidines
the idea that ligands should be able to adopt a U-shaped conforma-
(His323, His449) and a tyrosine (Tyr473) in the AF2 helix.
tion in case of both PPARR and γ, and an L-shaped conformation
Superposition of the PPARγ-GW409544 crystal structure with the
in case of PPARδ.4b On the basis of this description, a novel
modeled complex formed from PPARγ and the docked ligand 1b
isoxazolyl-serine-based PPAR agonist 1 (Chart 1) was designed
shows that the carboxyl of 1b is positioned in precisely the same
manner as is the carboxyl of GW409544 in the crystal structure.
Several compounds containing a carbazole moiety have been
tested previously against all three PPAR subtypes, and a crystal
structure of one of these carbazole-based compounds with PPARγ
has been solved.5 The carbazole moiety of this ligand binds to a
using the de novo/rational approach in combination with our
chemical knowledge.
† University of Illinois at Chicago.
‡ The Burnham Institute.
§ Georgetown University Medical Center.
| GlaxoSmithKline.
9
16714
J. AM. CHEM. SOC. 2004, 126, 16714-16715
10.1021/ja046386l CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
step (see Supporting Information). Human PPAR binding assays
show that 1a-c and 1f possess moderate binding affinities for some
or all of the three hPPAR subtypes (see Supporting Information).
The mouse PPAR-Gal4 assays show that ligands 1a-c are PPARR
selective (inactive at 5 µM for PPARγ/δ) (Table 1). Two triazole
analogues 2a and 2b (Chart 1) were also prepared by use of click
chemistry.7 Neither shows any significant PPAR activity.
PPAR agonists regulate cardiomyocyte gene expression and
might modulate hypertrophy.8 Therefore, the novel PPAR ligands
were screened for their ability to stimulate cardiomyocyte dif-
ferentiation from murine ES cells. Ligand 1a was the most active
one tested at concentrations between 1.25 to 20 µM between days
2-6, coinciding with the period when mesodermal cells can be
recruited to become cardiomyocytes (Figure 3). Compounds 1d-f
and 2a-b were also active. Notably, the PPARR agonists fenofi-
brate and Wy-14643 were inactive in this assay, as were the PPARγ
agonists rosiglitazone and GW1929 and PPARδ agonist GW501516.
Moreover, of the compounds that induced cardiomyogenesis, only
1a and 1f bound PPARs. Thus, we conclude that the PPAR acti-
vation is not the primary mechanism for cardiomyocyte differentia-
tion.
Figure 2. Modeled binding of 1b to PPARγ.
Table 1. Transactivation of PPARR by Wy-14643 and 1a-c
ligand
EC50
(µ
M)
SEM
Wy-14643
0.47
0.67
1.05
1.65
(0.06
(0.07
(0.08
-
1a
1b
1c
In summary, we have designed and synthesized a series of
isoxazolyl-serine-based PPAR ligands with moderate affinities.
Further structural modification of these lead ligands aimed at
improving upon their binding affinities and the identification of
the primary target for our ligand-induced cardiomyocyte differentia-
tion are underway.
Acknowledgment. We thank Julie Stimmel and Lisa Leesnitzer
for the binding data. M.M. thanks the National Institutes of Health
(Grant R21HL71913) for financial support.
Supporting Information Available: Methods of biological assays,
schemes, and detailed experimental procedures with spectroscopic data.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B. R. J. Med.
Chem. 2000, 43, 527.
(2) Kopelovich, L.; Fay, J. R.; Glazer, R. I.; Crowell, J. A. Mol. Cancer Ther.
2000, 1, 357.
Figure 3. 1a stimulated cardiomyocyte differentiation. Murine CGR8 ES
cells with an RMHC-GFP transgene6 were plated onto gelatin-coated dishes
(3200 cells/cm2) under conditions permissive for differentiation (DMEM
supplemented with 10% FBS, 2 mM glutamine, 1 mM sodium pyruvate,
0.1 mM nonessential amino acids, 0.1 mM âME, 100 units/mL penicillin,
100 µg/mL streptomycin, and 0.25 µg/mL amphotericin). Cells were exposed
to compounds between days 2-6, and eGFP fluorescence was scored at
day 8. Untreated (A) or diluted vehicle (DMSO) (B) stimulated no or
minimal differentiation, whereas addition of 1a consistently stimulated
eGFP-positive cells that beat rhythmically (C). Anti-myosin antibody (MF20
co-immunostaining (red) localized with GFP; blue nuclear counterstain is
DAPI (D).
(3) Shearer, B. G.; Hoekstra, W. J. Curr. Med. Chem. 2003, 10, 267.
(4) (a) Xu, H. E.; Stanley, T. B.; Montana, V. G.; Lambert, M. H.; Shearer,
B. G.; Cobb, J. E.; McKee, D. D.; Galardi, C. M.; Plunket, K. D.; Nolte,
R. T.; Parks, D. J.; Moore, J. T.; Kliewer, S. A.; Willson, T. M.; Stimmel,
J. B. Nature 2002, 415, 813. (b) Xu, H. E.; Lambert, M. H.; Montana, V.
G.; Plunket, K. D.; Moore, L. B.; Collins, J. L.; Oplinger, J. A.; Kliewer,
S. A.; Gampe, R. T., Jr.; McKee, D. D.; Moore, J. T.; Willson, T. M.
Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 13919. (c) Gampe, R. T., Jr.;
Montana, V. G.; Lambert, M. L.; Miller, A. B.; Bledsoe, R. K.; Milburn,
M. V.; Kliewer, S. A.; Willson, T. M.; Xu, H. E. Mol. Cell 2000, 5, 545.
(d) Oberfield, J. L.; Collins, J. L.; Holmes, C. P.; Goreham, D. M.; Cooper,
J. P.; Cobb, J. E.; Lenhard, J. M.; Hull-Ryde, E. A.; Mohr, C. P.;
Blanchard, S. G.; Parks, D. J.; Moore, L. B.; Lehmann, J. M.; Plunket,
K.; Miller, A. B.; Milburn, M. V.; Kliewer, S. A.; Willson, T. M. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 6102. (e) Xu, H. E.; Lambert, M. H.;
Montana, V. G.; Parks, D. J.; Blanchard, S. G.; Brown, P. J.; Sternbach,
D. D.; Lehmann, J. M.; Wisely, G. B.; Willson, T. M.; Kliewer, S. A.;
Milburn, M. V. Mol. Cell 1999, 3, 397. (f) Nolte, R. T.; Wisely, G. B.;
Westin, S.; Cobb, J. E.; Lambert, M. H.; Kurokawa, R.; Rosenfeld, M.
G.; Willson, T. M.; Glass, C. K.; Milburn, M. V. Nature 1998, 395, 137.
(5) Sauerberg, P.; Pettersson, I.; Jeppesen, L.; Bury, P. S.; Mogensen, J. P.;
Wassermann, K.; Brand, C. L.; Sturis, J.; Wo¨ldike, H. F.; Fleckner, J.;
Andersen, A.-S. T.; Mortensen, S. B.; Svensson, L. A.; Rasmussen, H.
B.; Lehmann, S. V.; Polivka, Z.; Sindelar, K.; Panajotova, V.; Ynddal,
L.; Wulff, E. M. J. Med. Chem. 2002, 45, 789.
hydrophobic pocket formed by residues Leu270, Ile281, Phe287,
Ile341, and Met348. The lack of well-defined electron densities
indicates that the binding is not tightly constrained. In our
compounds, linker D (Figure 1) is longer than the linker of the
ligand employed in the crystal structure study.5 Therefore, the
carbazole moiety in our ligands is pushed toward the distal end of
the binding pocket. Moreover, by docking our ligands to PPARR,
we found that the carbazole moiety occupies a different binding
site, one adjacent to the hydrophobic pocket defined for PPARγ,
while the remainder of the ligand maintains the same interactions
with the protein.
(6) Takahashi, T.; Lord, B.; Schulze, P. C.; Fryer, R. M.; Sarang, S. S.;
Gullans, S. R.; Lee, R. T. Circulation 2003, 107, 1912.
(7) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001,
40, 2004.
(8) (a) Hamano, T.; Kobayashi, K.; Sakairi, T.; Hayashi, M.; Mutai, M. J.
Toxicol. Stud. 2001, 26, 275. (b) Gilde, A. J.; Van Bilsen, M. Acta Physiol.
Scand. 2003, 178, 425.
The isoxazolyl-serine-based ligands were readily prepared by use
of an intermolecular nitrile oxide cycloaddition reaction as the key
JA046386L
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16715