Discovery of a novel series of peroxisome proliferator-activated receptor α/γ dual agonists for the treatment of type 2 diabetes and dyslipidemia

Article abstract of DOI:10.1021/jm048993p

A series of 2-aryloxy-2-methyl-propionic acid compounds and related analogues were designed, synthesized, and evaluated for their PPAR agonist activities. 2-[(5,7-Dipropyl-3-trifluoromethyl)-benzisoxazol-6-yloxy]-2- methylpropionic acid (4) was identified

Full text of DOI:10.1021/jm048993p

2
262
J. Med. Chem. 2005, 48, 2262-2265
Discovery of a Novel Series of
Peroxisome Proliferator-Activated
Receptor r/γ Dual Agonists for the
Treatment of Type 2 Diabetes and
Dyslipidemia
Although a direct link between PPARγ and insulin
sensitivity has not been fully established, one theory
suggests that the hypoglycemic effects of glitazones are,
at least in part, derived from their lipid-modulating
8
effects. Activation of PPARγ increases free fatty acid
(FFA) uptake in adipose tissue, thus reducing the FFA
inhibition of glucose disposal in skeletal muscle. While
glitazones lower glucose and improve insulin sensitivity,
they suffer from several adverse effects, including
weight gain and edema.On the other hand, PPARR-
Kun Liu,* Libo Xu, Joel P. Berger,
Karen L. MacNaul, Gauchao Zhou,
Thomas W. Doebber, Michael J. Forrest,
David E. Moller, and A. Brian Jones
Chart 1. PPAR Agonists
Merck Research Laboratories, P.O. Box 2000,
Rahway, New Jersey 07065
Received December 10, 2004
Abstract: A series of 2-aryloxy-2-methyl-propionic acid com-
pounds and related analogues were designed, synthesized, and
evaluated for their PPAR agonist activities. 2-[(5,7-Dipropyl-
3
-trifluoromethyl)-benzisoxazol-6-yloxy]-2-methylpropionic acid
(
4) was identified as a PPARR/γ dual agonist with relative
PPARR selectivity and demonstrated potent efficacy in lower-
ing both glucose and lipids in animal models without causing
body weight gain. The PPARR activity of 4 appeared to have
played a significant role in lowering glucose levels in db/db
mice.
activating fibrates have enjoyed a good safety profile in
humans over the past two decades. These hypolipi-
Type 2 diabetes is a multifactorial disease character-
ized by insulin resistance and/or abnormal insulin
secretion. This metabolic disorder, which accounts for
more than 90% of all diabetes, afflicts an estimated 6%
of the adult US population. Its worldwide frequency is
expected to grow by 6% annually, reaching a potential
9
demics are effective in lowering triglycerides and raising
HDL levels and have been demonstrated to significantly
reduce the coronary events in type 2 diabetes patients.10
The lipid modulating effect of fibrates is principally
mediated by PPARR-regulated expressions of genes
involved in lipid and lipoprotein metabolism in liver,
leading to enhanced hepatic fatty acid oxidation, which
may also be the basis of the observed independent
hypoglycemic effect of fibrates.11 In addition, fibrates
have been shown to reduce body weight in rodents, in
1
total of 200-300 million cases in 2010. Type 2 diabetes
and insulin resistance are frequently associated with
dyslipidemia and a markedly increased incidence of
2
atherosclerotic cardiovascular disease. Glycemic control
is traditionally considered the first priority in the
treatment of diabetic patients. However, results from
the UK Prospective Diabetes Study have shown that
intensive control of hyperglycemia can attenuate mi-
crovascular complications, but may not reduce mac-
rovascular disease, the main cause of morbidity and
sharp contrast to PPARγ activators, which promote
weight gain.1
1a,12
Current fibrate drugs are low affinity
PPARR ligands with only clinically marginal hypogly-
cemic effects. Further studies will be needed to establish
whether more potent PPARR agonists may treat type 2
diabetes without PPARγ-related liabilities.
We herein report the structural design, synthesis, and
preclinical evaluation of a novel series of PPARR/γ dual
agonists. Compound 4 demonstrated potent PPARR
agonist activity in a cell-based reporter gene expression
assay, while only weakly activating PPARγ. It lowered
both plasma glucose and lipid in animal models without
causing body weight increase, an adverse effect fre-
quently associated with PPARγ agonists.
We recently reported the discovery of a new class of
O-arylmandelic acid PPAR agonists.13 Some of them,
exemplified by 3, showed PPARR/γ dual activity in vitro
and good in vivo efficacy in lowering both glucose and
lipids in rodents. On the other hand, fibrates, such as
fenofibrate, are weak PPARR agonists and have enjoyed
an excellent tolerability profile. Fenofibrate is the
prodrug of the active parent compound, fenofibric acid
3
mortality in type 2 diabetes. Accordingly, more aggres-
sive therapeutic approaches that not only lower glucose
but also simultaneously reduce cardiovascular risk
factors associated with type 2 diabetes are clearly in
high demand.
The peroxisome proliferator-activated receptors
(
PPARs) are ligand-activated transcription factors be-
4
longing to the nuclear receptor superfamily. There are
three PPAR subtypes encoded by distinct genes: PPARR
5
(
NR1C1), PPARδ (NR1C2), and PPARγ (NR1C3). These
receptors are important regulators in multiple physi-
ological pathways, such as glucose homeostasis, fatty
acid metabolism, inflammation, and cellular differentia-
6
tion. Glitazones and fibrates are two classes of PPAR
drugs currently being marketed for the treatments of
insulin resistance and dyslipidemia, respectively. Gli-
tazones (pioglitazone and rosiglitazone (1)) are insulin
sensitizers functioning through PPARγ activation.7
(2). For the structural design of novel PPAR compounds,
we combined the putative pharmacophores of these two
classes of compounds and generated several hybrid
*
To whom correspondence should be addressed. Phone: 732-594-
7
445. Fax: 732-594-9556. E-mail: kun_liu@merck.com.
1
0.1021/jm048993p CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/10/2005

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2263
Scheme 1
Table 1. Binding IC50 and TA EC50 data
TA EC50 (µM) or
SPA IC50 (µM)a
max. % at 3 µMa
compd
hR
hγ
hR
hγ
4
5
6
7
8
9
1
0.23
>10
1.51
1.03
0.10
0.48
>50
>50
35
1.58
>10
1.62
>10
>15
1.22
>15
0.25
>50
0.30
25%
na
1.10
21%
26%
1.3
na
0.02
>100
b
na
0.06
0.62
0.75
0.01
na
>10
15
0
rosiglitazone
fenofibric acid
a
Mean values are shown (n ) 3); SD ( 15%; SPA (scintillation
1
5
proximity assay); TA (chimeric GAL4-hPPAR transactivation
assay).
1
6 b
Not active at 3 µM.
The in vitro profile of compound 4 was compared with
rosiglitazone and fenofibrate, two reference compounds
used in the in vivo studies for insulin sensitizing and
lipid modulating effects, respectively. In radioligand
binding assay, compound 4 was about 150-fold more
potent as a PPARR ligand than fenofibric acid, while
6-fold less potent on PPARγ compared with rosiglita-
zone. In transiently transfected COS-1 cells, compound
4 was a potent agonist of a chimeric hPPARR/GAL4
receptor with an EC50 of 0.3 µM. It weakly activated
PPARγ with only partial activity starting to show on
the titration curve at 3 µM, the highest concentration
tested. The PPARγ agonist activity of 4 was further
evaluated in a FABP assay.17 It has been demonstrated
that PPARγ is necessary and sufficient for adipocyte
differentiation and that PPARγ agonists induce adipo-
genesis. To examine the effect of 4 on adipocyte dif-
ferentiation, the levels of mouse adipose fatty acid
binding protein (aP2) mRNA expression in differentiat-
structures.14 We hoped that further SAR studies on
these hybrid structures would lead us to novel PPARR/γ
dual agonists with improved efficacy and safety profiles.
The present paper focuses on 4, obtained by combining
the isobutyric acid headgroup of fenofibric acid 2 and
the lipophilic aryloxy moiety of the mandelate lead 3.
A series of 2-aryloxy-2-methyl-propionic acids were
prepared via simple condensations followed by hydroly-
sis as depicted in Scheme 1. Alternatively, the top
phenol piece was coupled to the monosubstituted R-bo-
mo ester, followed by alkylation and hydrolysis to
produce related analogues with different substitution
patterns at the R-position.
The individual compounds 4-10 described herein and
1
5
17b
their affinity for PPAR subtypes and agonist efficacy
in a PPAR-GAL4 chimeric transactivation assay in
ing murine 3T3-L1 preadipocytes were determined.
The aP2 mRNA displayed little increase in expression
when 3T3-L1 cells were treated with 4 at doses up to 3
µM for 5 days (data not shown). At the highest dose of
10 µM, compound 4 acted as a weak agonist of PPARγ
with only 60% maximal expression. In this assay,
rosiglitazone increased aP2 expression in a dose-de-
pendent manner with an EC50 value of 15 nM.
It has been established that PPARR agonists can
exhibit strong species-dependent transactivation pro-
files. It is desirable to identify a compound with
relatively balanced PPARR potency in species used in
preclinical animal studies and in man. Similarly con-
structed transactivation assays were conducted using
chimeric PPARR receptors derived from other animal
species. The results indicated that compound 4 exhibited
comparable agonist activity toward mouse, hamster, and
dog PPARR with EC50 values of 0.18 µM, 0.46 µM, and
0.55 µM, respectively. Furthermore, compound 4 dem-
onstrated a high degree of selectivity against other
members of the nuclear receptor family. In nuclear
receptor counterscreens, compound 4 was negative in
assays for binding to, and/or activation of, LXRR, LXRâ,
RXRR, PXR, GR, ERR, ERâ, TRR, and TRâ at concen-
trations up to 10 µM.
1
6
COS-1 cells are summarized in Table 1. Structure 4
was divided into several regions for SAR study and
optimization. The effect of substitution at the phenyl
ring was first investigated. Compared with the 2,6-
dipropyl analogue 4, which showed 6-fold selectivity for
PPARR over PPARγ receptor in the binding assay and
potent PPARR agonist activity in the transactivation
assay, the monopropyl analogue 5 was essentially
inactive, indicating the necessity of bis-substitution. The
dimethyl substitution at the R-carbon seemed to be
optimal. The corresponding monomethyl analogue 7
exhibited reduced activity. Other R,R-disubstituted ana-
logues, such as 8, failed to improve potency in the
transactivation assay. Extension of the chain length by
inserting a methylene group between the aryloxy group
and the R-carbon proved to be detrimental, as demon-
strated by the inactive analogue 10. By comparison, the
SAR requirement around the isoxazole ring was more
flexible. The methoxy analogue 6 showed nonselective
PPARR/γ dual binding affinity, and PPARR/γ dual
agonist activity in the transactivation assay with en-
hanced potency. Replacement of the isoxazole ring with
a simple propionyl functionality produced 9, which
maintained the potent dual activity. It is important to
note that all the compounds in this series demonstrated
a high degree of selectivity over PPARδ subtype. No
activity was observed for this receptor at the highest
concentration tested (10 µM) in the binding assay.
The results of in vitro experiments demonstrated that
compound 4 showed 6-fold selectivity for PPARR over
PPARγ receptor in the binding assay and relatively
PPARR selective agonist activity in the transactivation
assay. Its insulin-sensitizing and lipid-lowering effica-

2
264 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Letters
treatment with 4: 76% at 3 mpk, and 85% at 10 mpk,
more than the 45% reduction obtained from rosiglita-
zone at 10 mpk. As part of this experiment, plasma drug
levels after chronic dosing of 4 and rosiglitazone were
also measured. Similar exposure levels were obtained
for 3 mpk compound 4 and 10 mpk rosiglitazone (237
and 247 µM‚h, respectively). It is noteworthy that,
despite its weak in vitro activity on PPARγ, compound
4
was more effective in lowering glucose than rosigli-
tazone, a potent PPARγ selective agonist, suggesting
that the PPARR activity of 4 played a significant role
in its in vivo hypoglycemic efficacy.11 This notion is
further supported by the potent triglyceride lowering
effect exhibited by 4 and the observation that 4 did not
induce further increase in body weight versus vehicle
control, in sharp contrast to rosiglitazone, which caused
significant weight gain (8-11%).
The lipid altering activity of 4 was further assessed
in a well-established dog model. Fenofibrate was chosen
as a reference compound with well-characterized human
and animal efficacy. The daily oral administration of
compound 4 to dogs at 3 and 10 mpk for 14 days lowered
the average serum cholesterol levels by 10.1% and
Figure 1. Effect of oral dosing of compound 4 and rosiglita-
zone on glucose levels in db/db mice (n ) 7 mice/group). Plasma
glucose levels were measured approximately 24 h after dosing
on the preceding day.
1
6.3%, respectively, in a dose-dependent and time-
dependent fashion. Dogs treated with fenofibrate at 50
mpk in parallel showed an average decrease in total
cholesterol levels of 8.8%. Non-HDL cholesterol, the sum
of LDL- and VLDL-cholesterol, was also determined
(
data not shown). Both 4 and fenofibrate significantly
decreased non-HDL cholesterol levels, and as with total
cholesterol, compound 4 exhibited clear superiority over
fenofibrate in dose potency.
Figure 2. Percentage reduction of elevated triglycerides in
db/db mice (n ) 7 mice/group). Mice were dosed with 10 mpk
rosiglitazone or 3 and 10 mpk compound 4 for 11 days.
In summary, compound 4 has been identified as a
PPARR/γ dual agonist with relative PPARR selectivity
in a cell based transactivation assay. It demonstrated
potent efficacy in lowering both glucose and lipids in
animal models. More importantly, it did so without
causing body weight increase, an adverse effect fre-
quently associated with the glitazones. The PPARR
activity of 4 appeared to have played a significant role
in glucose lowering in db/db mice. On the basis of its in
vitro and in vivo profiles, compound 4 was selected for
further evaluation in man.
Acknowledgment. We thank Margaret Wu, Yin Li,
Neelam Sharma, and Roger Muerer for additional
technical support. Dr. Bernard Choi is acknowledged
for the high-resolution mass spectra. We are indebted
to Drs. Alan Adams, Blair Wood, and James Heck for
fruitful discussions. We also thank Dr. Peter Meinke
for assistance and helpful discussions in manuscript
preparation.
Figure 3. Percentage reduction of serum total cholesterol
levels in mature beagle dogs. Dogs were housed individually
and dosed via oral gavage daily for 14 days with fenofibrate
or compound 4.
cies in vivo were next examined. The hypoglycemic effect
of compound 4 was tested in db/db mice, an obese rodent
model of type 2 diabetes characterized by severe insulin
resistance, hypertriglyceridemia, and marked hyperg-
Supporting Information Available: Synthetic proce-
dures and characterization data for intermediates and final
compounds. This information is available free of charge via
the Internet at http://pubs.acs.org.
1
6
lycemia. The mice were treated with compound 4 or
rosiglitazone, used as a reference compound in the
study, by oral gavage at the indicated dosages for 11
days. Administration of 4 resulted in significant reduc-
tion of blood glucose levels: 64% at 3 mpk, and 71% at
References
(
1) (a) Kopelman, P. G.; Hitman, G. A. Diabetes. Exploding type II.
Lancet 1998, 352 (Suppl 4), SIV5. (b) Amos, A. F.; McCarthy,
D. J.; Zimmet, P. The rising global burden of diabetes and its
complications: estimates and projections by 2010. Diabet. Med.
1997, 14 (Suppl 5), S5-S85.
1
0 mpk. By comparison, rosiglitazone produced 34%
(
2) Haffner, S. M.; Lehto, S.; Ronnemaa, T.; Pyorala, K.; Laakso,
M. Mortality from coronary heart disease in subjects with type
correction of hyperglycemia at 10 mpk. In the same
study, plasma triglyceride levels in db/db mice were also
sharply reduced in a dose dependent fashion by the
2
diabetes and in nondiabetic subjects with or without prior
myocardial infarction. N. Eng. J. Med. 1998, 339, 229-234.

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2265
(
3) UKPDS. UK prospective diabetes study 33: intensive blood
W o´ jcik, J.; Zarnecki, A.; Blane, G. Effects of comicronised
fenofibrate on lipid and insulin sensitivity in patients with
polymetabolic syndrome X. Eur. J. Clin. Invest. 2000, 30, 871-
878. (d) Avogaro, A.; Miola, M.; Favaro, A.; Gottardo, L.; Pacini,
G.; Manzato, E.; Zambon, S.; Sacerdoti, D.; de Kreutzenberg, S.;
Piliego, T.; Tiengo, A.; Del Prato, S. Gemfibrozil improves insulin
sensitivity and flow-mediated vasodilatation in type 2 diabetic
patients. Eur. J. Clin. Invest. 2001, 31, 603-609.
glucose control with sulphonylureas or insulin compared with
conventional treatment and risk of complications with type 2
diabetes. Lancet 1998, 352, 837-853.
(
4) Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B. R. The
PPARs: from orphan receptor to drug discovery J. Med. Chem.
2
000, 43, 527-550.
(
5) Nuclear Receptors Nomenclature Committee, 1999. A unified
nomenclature system for the nuclear receptor superfamily. Cell
(12) Chaput, E.; Salafin, R.; Silvestre, M.; Edgar, A. D. Fenofibrate
and rosiglitazone lower serum triglycerides with opposing effects
on body weight. Biochem. Biophy. Res. Commun. 2000, 271,
445-450.
(13) Adams. A. D.; von Langen, D.; Dadiz, A.; Elbrecht, A.; MacNaul,
K. L.; Berger, J. P.; Zhou, G.; Doebber, T. W.; Meurer, R.; Forrest,
M.; Moller, D. E.; Jones, A. B. O-Arylmandelic acids as highly
selective human PPAR R/γ dual agonists. Bioorg. Med. Chem.
Lett. 2003, 13, 3185-3190.
(14) For a similar approach, see: Brooks, D. A.; Etgen, G. J.; Rito,
C. J.; Shuker, A. J.; Dominianni, S. J.; Warshawsky, A. M.;
Ardecky, R.; Paterniti, J. R.; Tyhonas, J.; Karanewsky, D. S.;
Kauffman, R. F.; Broderick, C. L.; Oldham, B. A.; Montrose-
Rafizadeh, C.; Winneroski, L. L.; Faul, M. M.; McCarthy, J. R.
Design and synthesis of 2-methyl-2-{4-[2-(5-methyl-2-aryloxazol-
4-yl)ethoxy]phenoxy}propionic acids: a new class of dual PPARR/γ
agonists. J. Med. Chem. 2001, 44, 2061-2064.
(15) For the binding assay procedure, see: Adams, A. D.; Yuen, W.;
Hu, Z.; Santini, C.; Jones, A. B.; MacNaul, K. L.; Berger, J. P.;
Doebber, T. W.; Moller, D. E. Amphipathic 3-phenyl-7-propyl-
benzisoxazoles; Human PPARγ, δ and R agonists. Bioorg. Med.
Chem. Lett. 2003, 13, 931-936.
(16) For the detailed procedures of the transactivation assay and db/
db mice study, see: Berger, J.; Leibowitz, M. D.; Doebber, T.
W.; Elbrecht, A.; Zhang, B.; Zhou, G.; Biswas, C.; Cullinan, C.
A.; Hayes, N. S.; Li, Y.; Tanen, M.; Ventre, J.; Wu, M. S.; Berger,
G. D.; Mosley, R.; Marquis, R.; Santini, C.; Sahoo, S. P.; Tolman,
R. L.; Smith, R. G.; Moller, D. E. Novel peorxisome proliferator-
activated receptor (PPAR) γ and PPARδ ligands produce distinct
biological effects. J. Biol. Chem. 1999, 274, 6718-6725.
(17) (a) Szalkowski, D.; White-Carrington, S.; Berger, J.; Zhang, B.
Antidiabetic thiazolidinediones block the inhibitory effect of
tumor necrosis factor-a on differentiation, insulin-stimulated
glucose uptake, and gene expression in 3T3-L1 cells. Endocrinol-
ogy 1995, 136, 1474-1481. (b) Berger, J.; Petro, A.; MacNaul,
K.; Kelly, L.; Zhang, B.; Richards, K.; Elbrecht, A.; Johnson, B.;
Zhou, G.; Doebber, T.; Biswas, C.; Parikh, M.; Sharma, N.;
Tanen, M.; Thompson, G.; Ventre, J.; Adams, A.; Mosley, R.;
Surwit, R.; Moller, D. Distinct properties and advantages of a
novel peroxisome proliferators-activated protein γ selective
modulator. Mol. Endocrinol. 2003, 17, 662-676.
1
999, 97, 161-163.
(
6) Kersten, S.; Desvergne, B.; Wahli, W. Roles of PPARs in health
and disease. Nature 2000, 405, 421-424.
(
7) (a) Berger, J.; Bailey, P.; Biswas, C.; Cullinan, C. A.; Doebber,
T. W.; Hayes, N. S.; Saperstein, R.; Smith, R. G.; Leibowitz, M.
D. Thiozolidinediones produce a conformational change in per-
oxisomal proliferator-activated receptor-γ: binding and activa-
tion correlates with antidiabetic actions in db/db mice. Endo-
crinology 1996, 137, 4189-4195. (b) Willson, T. M.; Cobb, J. E.;
Cowan, D. J.; Wiethe, R. W.; Corea, I. D.; Prakash, S. R.; Beck,
K. D.; Moore, L. B.; Kliver, S. A.; Lehman, J. M. The structure-
activity relationship between peroxisome proliferator-activatied
receptor γ agonism and the antihyperglycemic activity of thio-
zolidinediones. J. Med. Chem. 1996, 39, 665-668.
(
8) Oaks, N. D.; Kennedy, C. J.; Jenkins, A. B.; Laybutt, D. R.;
Chisholm, D. J.; Kraegen, E. W. A new antidiabtic agent, BRL
49653, reduces lipid availability and improves insulin action and
glucoregulation in the rat. Dibetes 1994, 43, 1203-1210.
9) Schoonjans, K.; Staels, B.; Auwerx, J. Role of the peroxisome
proliferators-activated receptor (PPAR) in mediating the effects
of fibrates and fatty acids on gene expression. J. Lipid Res. 1996,
(
37, 907-925.
(
10) (a) Teiner, G. Effect of fenofibrate on progression of coronary-
artery disease in type 2 diabetes: the Diabetes Atherosclerosis
Intervention Study, a randomised study. Lancet 2001, 357, 905-
910. (b) Rubins, H. B.; Robins, S. J. Conclutions from the VA-
HIT study. Am. J. Cardiol. 2000, 86, 543-544. (c) Rubins, H.
B.; Robins, S. J.; Collins, D.; Fye, C. L.; Anderson, J. W.; Elam,
M. B.; Faas, F. H.; Linares, E.; Schaefer, E. J.; Schectman, G.;
Wilt, T. J.; Wittes, J. Gemfibrozil for the secondary prevention
of coronary heart disease in men with low Levels of high-density
lipoprotein cholesterol. N. Eng. J. Med. 1999, 341, 410-418.
11) (a) Guerre-Millo; M., Gervois, P.; Raspe, E.; Madsen, L.; Poulain,
P.; Derudas, B.; Herbert, J.; Winegar, D. A.; Wilson, T. M.;
Fruchart, J.; Berge, R. K.; Staels, B. Peroxisome proliferators-
activated receptor R activators improve insulin sensitivity and
reduce adiposity. J. Biol. Chem. 2000, 275, 16638-16642. (b)
Ye, J.-M.; Doyle, P.; Iglesias, M.; Watson, D.; Cooney, G.;
Kraegen, E. Peroxosome proliferators activated receptor (PPAR)-R
activation lowers muscle lipids and improve insulin sensitivity
in high fat-fed rats. Diabetes 2001, 50, 411-417. (c) Idzior-Walus,
B.; Sieradzki, J.; Rostworowski, W.; Zdzienicka, A.; Kawalec, E.;
(
JM048993P